Split (1)

train · ~9.11M rows (showing the first 93.5k)

identifier stringlengths 11 32	pdf_url stringlengths 17 4.62k ⌀	lang stringclasses 120 values	error stringclasses 1 value	title stringlengths 2 500 ⌀	source_name stringlengths 1 435 ⌀	publication_year float64 1.9k 2.02k	license stringclasses 3 values	word_count int64 0 1.64M	text stringlengths 1 9.75M
https://openalex.org/W3003638998	https://europepmc.org/articles/pmc7141037?pdf=render	English	null	Toward “On‐Demand” Materials Synthesis and Scientific Discovery through Intelligent Robots	Advanced science	2,020	cc-by	10,799	Toward “On-Demand” Materials Synthesis and Scientific Discovery through Intelligent Robots Today, motived by the breakthrough of the AI and intelligent robots, it is right time emerging to revisit and transfer these day's advanced technology to the fundamental research areas.[3] The preciseness of computer lan- guage and robot operation benefits a lot on solving the reproducibility problem,[4] which has been proved both in molecular synthesis,[5] 2D materials assembling,[6] and nanocrystal growth.[7] Recently, the robotics technology starts to apply in an automatic chemical laboratory vigor- ously widely.[8] Big chemistry data and AI algorithms endow robots with the ability to search reaction paths and optimize experimental parameters by itself, called autonomous discovery.[9] However, these scaling-out methods by exploring ample parameter space are more focusing on working efficiency, less on understanding and development of scientific knowledge.[10] The future materials discovery needs not only the robot-assisted workforce but upgrade of methodology to make synthesis really “predictable.”[11] The “Dial a Molecule” program in the UK and the DARPA project “Make it” in the USA are typ- ical ongoing cases for “On-Demand” molecule discovery.[12] The discovery process should start from materials design, synthesis planning, automatic experiments, and characterization, to para- meter self-adjusting autonomously to achieve a closing-loop pro- cess.[13] It needs an intelligent brain that well coupled neuron networks, physical-chemistry theory and scientists-machine interactions.[14] Since yet, most of the algorithms-assisted syn- thesis system was utilized for optimizing organic molecular reaction, for example, the SNOBFIT-assisted robotic platform in Massachu-setts Institute of Technology (MIT),[15] and the cloud synthesis system based on Complex Method in Cambridge.[16] For higher demand, brains of these self-driving laboratories are expected to extract the common concepts in experimental sci- ence, having wider compatibility, and more expansion interface. Here, we report our intelligent robotics system Materials ed, with with virtual (RL) scheme and self- can work Under the heory, and d on both e distribu- of inorganic of the tedious onditions. This als synthesis y can reshape A Materials Acceleration Operation System (MAOS) is designed, with unique language and compiler architecture. MAOS integrates with virtual reality (VR), collaborative robots, and a reinforcement learning (RL) scheme for autonomous materials synthesis, properties investigations, and self- optimized quality assurance. After training through VR, MAOS can work independently for labor and intensively reduces the time cost. Toward “On-Demand” Materials Synthesis and Scientific Discovery through Intelligent Robots Under the RL framework, MAOS also inspires the improved nucleation theory, and feedback for the optimal strategy, which can satisfy the demand on both of the CdSe quantum dots (QDs) emission wavelength and size distribu- tion quality. Moreover, it can work well for extensive coverages of inorganic nanomaterials. MAOS frees the experimental researchers out of the tedious labor as well as the extensive exploration of optimal reaction conditions. This work provides a walking example for the “On-Demand” materials synthesis system, and demonstrates how artificial intelligence technology can reshape traditional materials science research in the future. www.advancedscience.com www.advancedscience.com 1. Introduction The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/advs.201901957. © 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and repro- duction in any medium, provided the original work is properly cited. Toward “On-Demand” Materials Synthesis and Scientific Discovery through Intelligent Robots Jiagen Li, Yuxiao Tu, Rulin Liu, Yihua Lu, and Xi Zhu* time, the exploration of chemical labora- tory robotics can trace backward in the 1980s.[2] The robotics technology at that time lacks broadening intelligence and only can process limited tasks. Today, motived by the breakthrough of the AI and intelligent robots, it is right time emerging to revisit and transfer these day's advanced technology to the fundamental research areas.[3] The preciseness of computer lan- guage and robot operation benefits a lot on solving the reproducibility problem,[4] which has been proved both in molecular synthesis,[5] 2D materials assembling,[6] and nanocrystal growth.[7] Recently, the robotics technology starts to apply in an automatic chemical laboratory vigor- ously widely.[8] Big chemistry data and AI algorithms endow robots with the ability to search reaction paths and optimize experimental parameters by itself, called autonomous discovery.[9] However, these scaling-out methods by exploring ample parameter space are more focusing on working efficiency, less on understanding and development of scientific knowledge.[10] The future materials discovery needs not only the robot-assisted workforce but upgrade of methodology to make synthesis really “predictable.”[11] The “Dial a Molecule” program in the UK and the DARPA project “Make it” in the USA are typ- ical ongoing cases for “On-Demand” molecule discovery.[12] The discovery process should start from materials design, synthesis planning, automatic experiments, and characterization, to para- meter self-adjusting autonomously to achieve a closing-loop pro- cess.[13] It needs an intelligent brain that well coupled neuron networks, physical-chemistry theory and scientists-machine interactions.[14] Since yet, most of the algorithms-assisted syn- thesis system was utilized for optimizing organic molecular reaction, for example, the SNOBFIT-assisted robotic platform in Massachu-setts Institute of Technology (MIT),[15] and the cloud synthesis system based on Complex Method in Cambridge.[16] For higher demand, brains of these self-driving laboratories are expected to extract the common concepts in experimental sci- ence, having wider compatibility, and more expansion interface. H t i t lli t b ti t M t i l ed, with with virtual (RL) scheme , and self- S can work Under the heory, and nd on both ze distribu- s of inorganic of the tedious onditions. This ials synthesis gy can reshape time, the exploration of chemical labora- tory robotics can trace backward in the 1980s.[2] The robotics technology at that time lacks broadening intelligence and only can process limited tasks. 1. Introduction The artificial intelligence (AI) and robotics technology are cru- cial for the future manufacturing; the excellent major feature is the new scheme can integrate the whole process from the virtual modeling equipment design to real-time modulus maintenance, which provides novel and improved customer products. The experimental laboratory, as the demo stage before the industrial assembly line, requires a more complicated degree of variables and operations to reach the localized or delocalized pattern optimizations, for the functional properties. Usually, this kind of optimization processing requires a tremendous workforce, along with the side effects like safety and reproduction. There is a long history for the machine endorsed chemistry labora- tory scheme, as early as Leonard Da Vinci's time.[1] In modern J. Li, Y. Tu, R. Liu, Dr. Y. Lu, Prof. X. Zhu Shenzhen Institute of Artificial Intelligence and Robotics for Society (AIRS) The Chinese University of Hong Kong Shenzhen, Guangdong 518172, China E-mail: zhuxi@cuhk.edu.cn J. Li, Y. Tu, R. Liu, Dr. Y. Lu, Prof. X. Zhu Shenzhen Institute of Artificial Intelligence and Robotics for Society (AIRS) The Chinese University of Hong Kong Shenzhen, Guangdong 518172, China E-mail: zhuxi@cuhk.edu.cn Shenzhen Institute of Artificial Intelligence and Robotics for Society (AIRS) The Chinese University of Hong Kong Shenzhen, Guangdong 518172, China E-mail: zhuxi@cuhk.edu.cn The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/advs.201901957. © 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and repro- duction in any medium, provided the original work is properly cited. Here, we report our intelligent robotics system Materials Acceleration Operation System (MAOS), for “On-Demand” mate- rials synthesis and scientific discovery. MAOS is of the E-com- merce typical like eBay or Alibaba, the user ordered the materials online, and then MAOS deliver it. A virtual lab was developed to provide a safe and practical way for man–machine interaction to train MAOS for the new experimental operations. Collaborative The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/advs.201901957. © 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and repro- duction in any medium, provided the original work is properly cited. © 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Sci. 2020, 7, 1901957 J. Li, Y. Tu, R. Liu, Dr. Y. Lu, Prof. X. Zhu Shenzhen Institute of Artificial Intelligence and Robotics for Society (AIRS) The Chinese University of Hong Kong Shenzhen, Guangdong 518172, China E-mail: zhuxi@cuhk.edu.cn Full Paper DOI: 10.1002/advs.201901957 1901957 (1 of 11) 1901957 (1 of 11) Adv. Sci. 2020, 7, 1901957 Adv. Sci. 2020, 7, 1901957 www.advancedscience.com www.advancedsciencenews.com Figure 1. Schematic of MAOS and integrated modules, including Human–machine interaction, hardware control interface, analyzer, and AI optimizer. Users interact with robots through the website user interface (UI) and virtual reality. The cloud server provides storage, searching, and computing service for the system. Figure 1. Schematic of MAOS and integrated modules, including Human–machine interaction, hardware control interface, analyzer, and AI optimizer. Users interact with robots through the website user interface (UI) and virtual reality. The cloud server provides storage, searching, and computing service for the system. robot and flow chemistry system are in good cooperation for higher efficiency and degree of automation. Extensive data rate communication between lab and mobile devices works under the 4G/5G network. We include typical experiments for both of the scientific discovery and quality assurance; the synthesis of broad coverage of inorganic photoluminescent (PL) quantum dots (QDs) can sufficiently satisfy the “On-Demand” request. shown in Figure 2. The yellow triangle arrow indicates the tra- ditional paradigm of materials synthesis, which works from samples to properties, it first synthesis amounts of samples, and picked up only the excellent ones after property check, the off-target ones are usually discarded. The green triangle arrow indicates the “On-Demand” paradigm, based on the AI and Robots technology; it works from properties to samples, the off-target ones seldom came into being, which saves significant resources in environments and economics. The infrastructure is constructed by the individual database including materials prop- erties database, reagent database, methods database, machine database, and results database (Details in Section S2, Supporting Information). MAOS stores long-term data through SQLAl- chemy,[17] which supports many types of database management system, such as MySQL, Postgres, Oracle, and SQLite. MySQL on the cloud server is used to store the data. In the “On-Demand” materials synthesis idea, MAOS automatically finds the target materials through the database to satisfy user's requirement on the materials property, and on the other side, MAOS screens out available ones according to the synthesis ability of local labora- tory. The online database, such as Open Quantum Materials Database[18] is linked for the target materials with property searching. This virtual screening process throughout the Request Space, Solution Space, and Resource Space to generate synthesis planning, as shown in the left part of Figure 2. DOI: 10.1002/advs.201901957 The Solution Space is essential for the automatically experimental recipe generating. It includes the database of reactant types, concentration, volume/weight, and reaction conditions. Natural language processing (NLP)[19] technology is implemented here for the construction of the method space based on the previous research by Olivetti et al.[20] Next, MAOS will check whether the laboratory has adequate synthesis modules, reagent storage, and characterization equipment in Resource Space for the required solution. For example, chemical vapor deposition, molecular beam epitaxy are not feasible for the current version of MAOS, hydrothermal method, sol–gel method, and most of the solution-based process can be well satisfied. A pricing model drew on design principles of green chemistry announced by 2. System Architecture As shown in Figure 1, the human–machine interaction module provides an input window for users to submit demands of various properties; the user also can control and monitor the system online. Currently, we support a user's user interface (UI) of web and virtual reality (VR) (the left part in Figure 1) to interact with MAOS. The Hardware control interface (right part in Figure 1) managed all hardware modules and provided a set of independent hardware instructions for AI optimizer and the system manager. The Analyze module within MAOS first receives the raw data from the characterization modules, auto- matic analyze, and optimize the experiment parameters to meet the demand on request under reinforcement learning (RL) scheme. After the precision criterion converged, it sends the feedback to the optimizer and visualizes the results to the user (The workflow of MAOS is in Section S1, Supporting Infor- mation). Cloud sever was utilized for data storage, searching, and high-performance computing if necessary. Due to the high amount data flow including data and video, the WIFI and 4G/5G communication between local lab and mobile devices are essential; we also have tested the 5G signal for the time delay in the remote control of MAOS through mobile phone. © 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.advancedscience.com www.advancedsciencenews.com www.advancedscience.com www.advancedsciencenews.com www.advancedscience.com Figure 2. Comparison on the workflow of AI-endorsed “On-Demand” materials discovery and traditional approach, legends for concept explanation is on the right side. For the AI-endorsed approach, the procedure starts by identifying target materials according to user demand properties in request space. Then, demand matching between request space and solution space was made refer to the synthesis methods database. NLP benefits a lot here in extracting keywords from journal articles and arrange them. Synthesis planning will be made by utilizing available modules in the resource space. Both request and resource space should belong to the solution space to verify the workflow keep going. Thus the virtual screening process ends, and experiments take place. In situ experimental feedback is stored in results space and flow to AI space as a learning dataset. The refined parameters will be updated for the next experiment. The closed-loop workflow constituted by resource space, result space, and AI space will take time to approach the optimal parameters and obtain the target sample. The whole process is distinguished from the traditional way for materials research, which starts from scale-up parallel synthesis and screens out very limited samples to meet the demand. Adv. Sci. 2020, 7, 1901957 2.1. Workflow Throughout Various Space The workflow of “On-Demand” materials synthesis could be described as data and matter flow throughout various Spaces, 1901957 (2 of 11) Adv. Sci. 2020, 7, 1901957 © 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim © 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 2.2. Human–Machine Interaction a) Remote users with VR glasses and gamepad operate the robot in a virtual lab. The robot in the real lab was guided refer to position information of virtual robot. b) Basic concepts in RF learning model including agent, action, environment, and reward are well coupled in materials synthesis space and played a role. st is all the information about the reaction space. at is the experimental condition determined by policy function π. rt is the reward of reaction R under at. to extract red, green and blue (RGB) information from video streams captured by the in-built camera, and then mapped it into the average particle size 〈DRGB〉 or 〈D〉 From a well-benchmarked machine learning scheme. Details of the analyzing algorithms were described in our previous work.[7] The spectrum analyzer is used to extract the demand optical information from spectrum files generated by spectrometer software. Center wavelength and intensity of peaks was first identification. Then multipeak Gaussian fitting is applied based on SciPy.[23] The emission wave- length and FWHM (Full width at half maximum) of the product are calculated from a Gaussian curve with maximum peak value. to MAOS, the administrator can remotely operate the lab through VR and train MAOS for the whole experimental operation details. The UI–VR interface is a virtual lab and isomorphic reflection of the real lab. As shown in Figure 3a, the administrator can get sight and control a virtual robot in the lab through VR equip- ment. The virtual robot communicates with reality through 5G network or transmission control protocol (TCP) based protocol (Section S5, Supporting information). The virtual robot could translate user command to controlling command for the real robot in real-time. For some hard experimental operations jobs, the commands could be executed in a virtual experiment first and then executed in a real environment. Cameras recording the real lab information enable the administrator to get a live stream of the virtual and real lab at the same time in VR glass for improved training. In UI–VR, the robot arm is forbidden to collide with other equipment. When administrators training robot arm to do jobs (such as grasping beakers) in VR, each job and all related commands would be recorded and facilitated via database. With the post-trained data, the robot can well repeat the same task in the record. 2.2. Human–Machine Interaction American Chemical Society (ACS)[21] was applied for optimal planning (Section S3, supporting information). The index indi- cates toxics, atom economy, safety, and renewability in the model. After the virtual screening, MAOS will load the history experi- ment results from Result Space. If no such database exists, the robots will start the first experiment with initial parameters gen- erated in AI Space. The in situ characterized experimental data are stored in Result Space (right top in Figure 2), and then sent to AI Space for optimization. An optimizer in AI Space will decide parameters adjustment for the next loop and finally approach the demand target (the detail of the optimization is shown in the Experimental Section; and Section S4, Supporting Information). Movie S1 in the Supporting Information shows the whole workflow of “On-Demand” material synthesis. MAOS provides two efficient ways for users to communicate with the system, through UI-Web (UI: user interface) or UI-VR devices. The UI-Web built on cloud server has two channels for the users and sellers (materials supplier or administrator), respectively. The interface for users is similar to an online shopping website like eBay or Alibaba; users can search target materials according to the request on materials properties, like the bandgap, then order after the price checking (As shown in the starting part of Movie S1, Supporting Information). MAOS will first generate the operation recipe once received the order, if the synthesis plan of the target material is in the record, MAOS then started the synthesis automatically; if the experiment is new 1901957 (3 of 11) © 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Sci. 2020, 7, 1901957 www.advancedsciencenews.com www.advancedscience.com Figure 3. Scheme of VR-based remote control of robots and RF learning model in materials synthesis. a) Remote users with VR glasses and gamepad operate the robot in a virtual lab. The robot in the real lab was guided refer to position information of virtual robot. b) Basic concepts in RF learning model including agent, action, environment, and reward are well coupled in materials synthesis space and played a role. st is all the information about the reaction space. at is the experimental condition determined by policy function π. rt is the reward of reaction R under at. www.advancedsciencenews.com www.advancedscience.com www.advancedscience.com Figure 3. Scheme of VR-based remote control of robots and RF learning model in materials synthesis. 2.2. Human–Machine Interaction The VR and collaborative robot constitute a pivotal module to replace the tedious labor works. The detail of the robot arm structure and parameters structures with VR are shown in Section S6, Supporting Information. As shown in Figure 3b, an AI optimizer acts as a decision maker to adjust reaction parameters through the whole proce- dure of synthesis. A trial and error process guided by searching algorithms endorse the system ability of autonomous discovery. A reaction could be viewed as a system taking multiple inputs (parameters generated by all hardware modules) and return multiple outputs (product property metrics, such as emission wavelength and FWHM). Optimizer needs to search the experi- mental parameter space and find the state with demanded properties, with the minimum number of steps. Here we apply the RL scheme, which focuses on how to maximize the “reward” from “environment” through “action,”[24] which proved influential in optimizing chemical reaction.[25] For mate- rials synthesis, the reaction is the “environment,” an algorithm decides the experiment parameters to try “actions” in order to find the desired product (reward). For MAOS, the RL process was guided by SNOBFIT algorithms,[26] which has been applied inorganic[27] and inorganic[28] synthesis optimization. Detailed algorithms implementations, including the language, para- meter spaces, and operators are in the Experimental Section. 2.3. Analyzer and AI Optimizer The synthesis process generates high-throughput feedback, including video frame flows and in situ spectrum by characteriza- tion modules, which occupies hundreds of million bits per second data rate. Before sent to the optimizer, these feedback data need preprocessing into certain forms by extracting demanded infor- mation to verify the learning efficiency of the optimizer. An effi- cient automatic analyzer for data processing is necessary for such a self-driving system. Analyzing models for Computer Vision (CV) and spectrometer were developed for real-time data pro- cessing. The CV analyzer based on OpenCV[22] library was applied Adv. Sci. 2020, 7, 1901957 Adv. Sci. 2020, 7, 1901957 © 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 2.4. Hardware Modules and Control Interface As shown in Figure 4a, in MAOS, hardware modules can be divided into three parts by function: for automation, synthesis, 1901957 (4 of 11) © 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Sci. 2020, 7, 1901957 www.advancedsciencenews.com www.advancedscience.com and characterization. A challenge of traditional flow chemistry time by remote administrators. The robot arm can handle the Figure 4. Available reconfigurable hardware modules in MAOS. a) Syringe pumps, collaborative robot, VR system, and a liquid handler were coupled to realize multiple types of automation. The functional modules, including vacuum or N2 atmosphere, stir, temperature gradient, and light radiation. A 3D-printer is efficient for printing specifically designed components. The PL, Abs, and CV modules collect the in-line photoluminescence, optical absorption spectrum, and computer vision data, respectively. XRD (X-ray diffraction) module work for phase identification of synthesized crystalline material. b) Photograph of assembled modules for typical materials synthesis and characterization. Various experimental conditions and operations can be realized by assembling the relevant modules. 1901957 (5 of 11) Adv. Sci. 2020, 7, 1901957 www.advancedscience.com www.advancedscience.com module (Photon-M, right middle in Figure 4b), provides variant experimental conditions for the investigation of external field perturbation on the synthesis. The cooling model controlled by a proportional integral derivative (PID) device and a K-type sensor, provides the temperature ranges from −20 to 300 °C; The photon module includes a 30W UV irradiation (365 nm). The Chembox combined the electronic weighting module, temperature meas- urement, and an intelligent interface, which displays the inside chamber view and chemical information, including solution volume, weight, reaction temperature, and stirring speed. A vac- uum-nitrogen supply system also couples with Chembox, which covered the function of traditional double row tube in chemical synthesis. A 3D printer is used to manufacture small batches of customized structures, such as sample holders and low-temper- ature reactors. For characterization, photoluminescence (PL-M, marked in yellow), absorption (Abs-M, marked in blue), and com- puter vision (CV-M, marked in white) are utilized for in situ mon- itoring the optical properties of synthesized materials. Details of hardware modules are in Section S6, Supporting Information. ratio, crystal size, and size distributions to verify target emis- sion wavelength and FWHM. These two parameters well cor- relate with the QD size[30] and the size focusing ,[31–33] thus the “property On-Demand” issue in the optoelectronic QDs system can be equally interpreted into the “size and distribution On- Demand” issue. However, it is still a challenging issue for controlling both of the two quantities in the same time because, theoretically, the diffusion-limiting growing mechanism usually comes along with the Ostwald ripening, which always broad- ening the size distribution[32] and lowering the QDs quality. Here, in this case, we apply the MAOS to further improve the QDs quality by satisfying the demand. The synthesis planning made by MAOS suggested a hot injection stratagem for CdSe nanocrystal utilizing multiple synthesis modules, as shown in Figure 5a. The available hardware modules endorsed a fully automatic synthesis workflow. Vacuum pump and N2 flow con- trolled by a solenoid valve provided an adjustable gas environ- ment for ODE-Cd precursor preparation. Uniform stirring and temperature control are integrated into this flask (in yellow) based system in Chembox, where the ODE-Cd and TOP-Se pre- cursors were well prepared, then the syringe pumps injected the liquid through the multistep reaction modules, the flowing direction is indicated by the arrow in Figure 5a (more experi- mental details are in Section S7, Supporting Information). PL \| Pc SP SP 3 4 4 ( ) ⋅ + ⋅ × × R H C PL \| Pc SP SP 3 4 4 ( ) ⋅ + ⋅ × × R H C Equations (1) and (2) interpret the Robots arm action in reac- tants preparation and PL measurements, respectively. Before optimization, we applied a parameter correlation test on all adjustable parameters in this experiment; as shown at the right top in Figure 5a. In this test, a default parameter vector was chosen as the baseline; the individual change of each parameter with 20% variation was applied. We find only two parameters [ , ] Th t i i H correlate more than 10% within the characteriza- tion results (λ and FWHM), which indicates only the heating module (H) matters here. As shown in the blocks in Figure 5a, the heat module (block H in Figure 5a) controlled the reac- tion temperature, which is parameterized as Th, with tH is the working time of H; and the reaction time (t) was well affected by the sum of injection rate τ of the pump,[33] which is param- eterized as action ai, can be represented as action [ , ] = a Th t i i i H , which are independent parameters for the optimizer. Once the experiment started, a random [ , ] 0 0 Th tH parameters are ini- tiated for step 0 as in the subscript, MAOS first collected the emission spectrum (block PL in Figure 5a) of the continues synthesized samples data from the spectrometer, checked the validation of the PL signal, then sent the spectrum data to the Analyzer block for information extraction, in this stage, the user demand properties (PL center wavelength (λt) and target FWHM (FWHMt) of synthesized materials were extracted and compared with the experimental one in the current stage; if the criterion converged, the experiment steps, or else, MAOS acti- vated the optimizer block, updating the [ , ] 0 0 Th tH to [ , ] 1 1 Th tH , iterated to the criterial convergence. www.advancedscience.com www.advancedsciencenews.com www.advancedscience.com Figure 4 Available reconfigurable hardware modules in MAOS a) Syringe pumps collaborative robot VR system and a liquid handler were coupled Figure 4. Available reconfigurable hardware modules in MAOS. a) Syringe pumps, collaborative robot, VR system, and a liquid handler were coupled to realize multiple types of automation. The functional modules, including vacuum or N2 atmosphere, stir, temperature gradient, and light radiation. A 3D-printer is efficient for printing specifically designed components. The PL, Abs, and CV modules collect the in-line photoluminescence, optical absorption spectrum, and computer vision data, respectively. XRD (X-ray diffraction) module work for phase identification of synthesized crystalline material. b) Photograph of assembled modules for typical materials synthesis and characterization. Various experimental conditions and operations can be realized by assembling the relevant modules. and characterization. A challenge of traditional flow chemistry process is that the reagents are presettled and connected with pumps, tedious human labors are inevitable for the reactants preparation work. In MAOS, a collaborative robot was intro- duced in to handle these repetitive works. At the same time, a virtual lab is constructed to train the robot or control it in real time by remote administrators. The robot arm can handle the beakers to a liquid transfer module, which is a designed liquid handler made up with stepping motor and 3D-printed com- ponents. Thus the connection between solution container and flow system links automatically. The heating (Heat-M, marked in red), cooling (Cool-M, right top in Figure 4b), and photon 1901957 (5 of 11) Adv. Sci. 2020, 7, 1901957 1901957 (5 of 11) © 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Sci. 2020, 7, 1901957 © 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.advancedscience.com The MAOS language for this experiment includes two steps pp g All robotic automation and characterization modules are controlled by a hardware interface in MAOS, which includes both high-level and low-level instructions based on JSON- RPC2.0 protocol. The high-level instruction is a formula made up of instruction symbols and operators. Each symbol indi- cates the corresponding hardware module, and each module has its specific parameter vector, which represents all adjust- able parameters this module can provide. Taking an example for the Heating module, the instruction symbol is H, and the parameter vector is [ , , ] Th Th tH  . Here, Th is the stable tempera- ture controlled by H, Th  is the temperature gradient (heating rate), tH is the working time of H. The symbol array collected by operators can describe the arrangement of modules and rea- gent in both time and space domain. Five kinds of operators are utilized: “+”,“×”,“⋅”,“\|”, and “∪” . By combining different module symbols and the operators, a language is developed for the communication between MAOS and the real experiments. Details of all instruction symbols, parameter vector (Table 2) and operators (Table 3) can be found in the Experimental Sec- tion. With the help of the language, all experimental operations can be interpreted by a formula, and then been converted into the machine code, which is defined as process of “compile” (details are shown in Section S6 and Figure S2, Supporting Information). All parameters in an experiment will be stored in vectors to construct the parameter space, which is called “parameterization.” These vectors are input data for both hard- ware control interface and AI optimizer. Pc SF SP SP 3 1 1 2 2 ( ) ( ) ( ) = ⋅ × + ⋅ + ⋅ × ∪ × ∪ ∪   R A R R A V A G O s (1) Pc SF SP SP 3 1 1 2 2 ( ) ( ) ( ) = ⋅ × + ⋅ + ⋅ × ∪ × ∪ ∪   R A R R A V A G O s (1) (1) (2) 3.1. “On-Demand” Synthesis of Colliding QDs with Target Emission Wavelength and FWHM a) Assembled modules for multistep synthesis. Two syringe pumps and the heating module was utilized to optimize the reaction condition under control of optimizer (blue arrow indicates the flow direction, solid lines, and dashed lines represent matter and information flow). The top right inset image shows the correlation of all parameters with results. b) “On-Demand” synthesized CdSe QDs with different λt mapping on the 2D parameter Th tH [ , ] 1 1 plate. Color of icons indicate λt (see legends on the right). Ellipses on this plate indicate the parameter region which achieved λt ± 0.5 nm and FWHMt < 35 nm, also described as global optimal solution Th t g gH [ , ]. The blue arrows indicate the trend of Th t g gH [ , ] along with the increase of λt. Local optimal solution Th t l l H [ , ] (marked as cubic) were discovered when approaching 530 and 540 nm demand. Optimization failed when λt = 580 nm. The red dash line is a boundary that no solution exists for FWHMt < 35 nm. c) The 3D viewers of the global and local optimal position for target λt = 540 nm. Figure 5. Synthesis schematic and optimization process in the “On-Demand” synthesis of CdSe QDs with target emission wavelength and FWHM. a) Assembled modules for multistep synthesis. Two syringe pumps and the heating module was utilized to optimize the reaction condition under control of optimizer (blue arrow indicates the flow direction, solid lines, and dashed lines represent matter and information flow). The top right inset image shows the correlation of all parameters with results. b) “On-Demand” synthesized CdSe QDs with different λt mapping on the 2D parameter Th tH [ , ] 1 1 plate. Color of icons indicate λt (see legends on the right). Ellipses on this plate indicate the parameter region which achieved λt ± 0.5 nm and FWHMt < 35 nm, also described as global optimal solution Th t g gH [ , ]. The blue arrows indicate the trend of Th t g gH [ , ] along with the increase of λt. Local optimal solution Th t l l H [ , ] (marked as cubic) were discovered when approaching 530 and 540 nm demand. Optimization failed when λt = 580 nm. Adv. Sci. 2020, 7, 1901957 3.1. “On-Demand” Synthesis of Colliding QDs with Target Emission Wavelength and FWHM Next, we introduce the whole procedure of the “On-Demand” synthesis scheme in MAOS. One of the most urgent desires for this kind of technology lies in the optoelectronic QDs, which are promising materials for application fields in the display, lighting, and sensing devices.[29] These applications often demand precise nanoscale-engineering on component 1901957 (6 of 11) © 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Sci. 2020, 7, 1901957 w.advancedsciencenews.com www.advancedscience ure 5. Synthesis schematic and optimization process in the “On-Demand” synthesis of CdSe QDs with target emission wavelength and FW Assembled modules for multistep synthesis. Two syringe pumps and the heating module was utilized to optimize the reaction condition u trol of optimizer (blue arrow indicates the flow direction, solid lines, and dashed lines represent matter and information flow). The top right age shows the correlation of all parameters with results. b) “On-Demand” synthesized CdSe QDs with different λt mapping on the 2D param tH , ] 1 plate. Color of icons indicate λt (see legends on the right). Ellipses on this plate indicate the parameter region which achieved λt ± 0.5 nm HMt < 35 nm, also described as global optimal solution Th t g gH [ , ]. The blue arrows indicate the trend of Th t g gH [ , ] along with the increase of λt. imal solution Th t l l H [ , ] (marked as cubic) were discovered when approaching 530 and 540 nm demand. Optimization failed when λt = 580 nm www.advancedsciencenews.com vancedsciencenews.com www.advancedscie Statistic results of all optimal [ , ] Th t i i H for λt ranges from 490 to 530 nm, are shown in Figure 5b, and the data are sum- marized in Table 1. We can see that multiple parameter sets [ , ] Th t i i H can satisfy the user's demand in each ellipse plane in the 2D plot. Each ellipse was established by one to three experimental results within the region (original data are in Figure S13, Supporting Information). An example of the visualization of the optimization process for a typical experi- ment (λt = 530 nm, FWHM < 35 nm) is inset as Figure 5c. RL algorithms guided the parameter optimization process. Figure 5. Synthesis schematic and optimization process in the “On-Demand” synthesis of CdSe QDs with target emission wavelength and FWHM. © 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.advancedscience.com www.advancedscience.com www.advancedsciencenews.com www.advancedsciencenews.com Table 1. A summary of partial achieved results under reinforcement learning with different target wavelength, the data are ordered by sequence in MAOS. hieved results under reinforcement learning with different target wavelength, the data are ordered by sequence in Table 1. A summary of partial achieved results under reinforcement learning with different target wavelength, the MAOS. Target λ [nm] Optimal λ [nm] Error [nm] FWHM [nm] Trials Temperature [°C] Time [s] 490 489.8 0.2 33.8 56 175 56 500 499.6 0.4 34.6 25 180 98 510 509.6 0.4 33.2 16 180 156 520 519.8 0.2 31.4 8 190 208 530 529.6 0.4 34.0 2 195 304 540 539.6 0.4 34.6 2 225 76 550 549.8 0.2 34.6 1 225 152 560 560.4 0.4 34.6 4 230 210 570 569..8 0.2 34.8 15 235 356 580 580.4 0.4 39.8 100 (Max) 245 520 Here the RL defined action is [ , ] = a Th t i i i H and the reward is ri = [Δλi,ΔFWHMi] = [λt − λi,FWHMt − FWHMi] . We define the loss function as abs( ) 20ReLu( FWHM) λ = ∆ + −∆ L , representing the difference from the target. 0.5 < L was set as converge con- dition. For RL optimization, the numerical range of Thi and tH are within(170 °C, 245 °C) and (20s, 600s) respectively, as illus- trated as the X and Y axis in Figure 5b. The z-axis illustrates the trials to reach the convergence. Since every step MAOS will redo the whole experiment again, it takes about 4 h to close the optimization. We can see that a locally optimal solution [ , ] Th t l l H (green square) was first discovered by MAOS. Then, the global searching function overcame the barrier and discov- ered a better-converged parameter set [ , ] Th t g g H in a higher tem- perature region (green ellipse), which is different from cases that λt < 530 nm. As shown in Figure 5b, the trend of [ , ] Th t g g H along with the increase of λt is to increase both Thg and tg H. Overall, the Thg changes more than the tg H during the optimiza- tion. However, when ti H approaches the red dash line around t = 470s, the emerging of Ostwald Ripening strongly limited the further optimization of FWHM, which makes the long reac- tion time (large t) a forbidden choice for FWHM optimization, MAOS will turn the temperature dimension for the numerical solution. This physics picture caused the local optimal state as shown in the green square for the case λt = 530 nm. For larger λt, the search of [ , ] Th t i i H mainly focus in the high-temperature region. Thus, for λt = 540 nm, there is no optimization solu- tion bellow 200 °C; the previous local optimal solution in the high-temperature region became the global optimal one. In this jumping process, MAOS tune the value of thi by reducing the reaction time t from 370s to near 75s, resulting in a 5 nm drop of FWHM, and the temperature increase from 195 to 225 °C, as indicated by the λt = 540 nm data in Figure 5b.With fur- ther increasing of λt, the trend of [ , ] Th t g g H approaches the red dash line; however, since the high temperature can destroy the chemical bonding in QDs, the temperature range is limited, which does not allow another jumping to the solution region. Thus there is no solution for λt = 580 nm and above with the FWHM < 35 nm constrain. D i h h l i i i MAOS d h for reaching the convergence. Moreover, based on the post- training steps from the previous task, it only takes two trials for the fifth task as 520 m and 35.0 nm target, which provides an excellent demo for the following “On-Demand” tasks. Even though the above solution is of a pure numerical style, it still reveals insightful information inside the optimization details. We now understand that it is challenging to synthesize the large size QDs with narrow size distribution due to the limi- tation of the nucleation theory as discussed above,[31,32] a more comprehensive additional 〈R〉 dependent subsection policy is essential for the solution of this issue to satisfy the demand on QDs of large radius. 3.1. “On-Demand” Synthesis of Colliding QDs with Target Emission Wavelength and FWHM The red dash line is a boundary that no solution exists for FWHMt < 35 nm. c) The 3D viewers of the global and local optimal position for target λt = 540 nm. Statistic results of all optimal [ , ] Th t i i H for λt ranges from 490 to 530 nm, are shown in Figure 5b, and the data are sum- marized in Table 1. We can see that multiple parameter sets [ , ] Th t i i H can satisfy the user's demand in each ellipse plane in the 2D plot. Each ellipse was established by one to three experimental results within the region (original data are in Figure S13, Supporting Information). An example of the visualization of the optimization process for a typical experi- ment (λt = 530 nm, FWHM < 35 nm) is inset as Figure 5c. RL algorithms guided the parameter optimization process. 1901957 (7 of 11) © 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Sci. 2020, 7, 1901957 KGaA, Weinheim www.advancedsciencenews.com www.advancedscience.com be linearly transferred as FWHM 2 2(2log2)σ = , the analytic model of the (η τ) correlated FWHM 〈〉 R dependence can be fitted by Equation (4), indicating the robust solution for the 〈R〉 dependent FWHM optimization Due to the well-integrated Figure 6. Scheme of synthesis strategy provided by assembled modules and sketch for the multi-injection mechanism. a) CdSe QDs were synthesis in Chembox and continuously extracted through Cooling and PL modules for measurement. b) R 〈〉 FWHM ~ dependence plot of the nanoparticles produced, with (η, τ) groups (2, 240s), (2, 600s), (3, 240s), and (3, 600s). The black line indicates the typical Ostwald Ripening, the red and blue line fit the data of η = 2, 3 cases, respectively. www.advancedsciencenews.com www.advancedscience.com www.advancedscience.com Figure 6. Scheme of synthesis strategy provided by assembled modules and sketch for the multi-injection mechanism. a) CdSe QDs were synthesis in Chembox and continuously extracted through Cooling and PL modules for measurement. b) R 〈〉 FWHM ~ dependence plot of the nanoparticles produced, with (η, τ) groups (2, 240s), (2, 600s), (3, 240s), and (3, 600s). The black line indicates the typical Ostwald Ripening, the red and blue line fit the data of η = 2, 3 cases, respectively. Figure 6. Scheme of synthesis strategy provided by assembled modules and sketch for the multi-injection mechanism. a) CdSe QDs were synthesis in Chembox and continuously extracted through Cooling and PL modules for measurement. b) R 〈〉 FWHM ~ dependence plot of the nanoparticles produced, with (η, τ) groups (2, 240s), (2, 600s), (3, 240s), and (3, 600s). The black line indicates the typical Ostwald Ripening, the red and blue line fit the data of η = 2, 3 cases, respectively. be linearly transferred as FWHM 2 2(2log2)σ = , the analytic model of the (η, τ) correlated ~ FWHM 〈〉 R dependence can be derived as fitted by Equation (4), indicating the robust solution for the 〈R〉 dependent FWHM optimization. Due to the well-integrated modules in MAOS, the plentiful parameter spaces cooperated hardware can afford long-time consistent work with high pre- cision and reproduction, provides more possibilities for dis- covering new physical and chemical properties than the tradi- tional experimental operation. With the sufficient high-quality experimental data, MAOS can work as an excellent platform for inspiring in-depth understanding from theoretic level, as shown in Equation (4). Adv. Sci. 2020, 7, 1901957 Previous work reported that the additional injection during the growth could significantly improve the size distribution.[7,33] Here we apply MAOS by variant injection time interval τ and the times of the injection operation η, and we keep the total injection amount ηcη constant for different η value . The experimental setup in MAOS is shown in Figure 6a, the MAOS language here is PL \| SP SP SP SP 3 3 1 1 2 2 4 ( ) ( ) ( ) ⋅ × ∪ × ∪ ∪ + ⋅ + ⋅ × ∪ ∪    ⋅ × R A V A G O R R A G O C (3) (3) PL \| PL The difference from Figure 5a is the reaction location now is Chembox, which confines all the reactions space. Since we investigate how the interval injection can affect the QDs quality through diffusion in-between the total QDs, a uniform reaction space is essential. Figure 6b shows the experimental data for the average radius (〈R〉) and FWHM with η and τ dependence which looks like chaos at a first glimpse. For the two cases (2, 600s) and (2, 240s), as shown by the empty and solid cube in Figure 6b, we can see there is an apparent radius dependent of FWHM character. The (2, 240s) has favorite 〈R〉 when 〈R〉 < 4.4 nm, for 〈R〉 > 4.4 nm, the FWHM from (2, 600s) strategy results in more narrow FWHM and QDs quality. This experimental phenomenon indicates that the injection interval τ can be an efficient operational parameter in the MAOS to optimize the size-dependent FWHM. We reset the τ value extending to the (3, 240s) and (3, 600s) cases, as shown in the empty and solid triangle in Figure 6b, we observed the same picture as the (2, 240s) and (2, 600s) cases, the critical radius enlarged to 4.7 nm. Since the transition from σ to FWHM can During the whole optimization, MAOS can record the para- meter details, for the following task, for example, 535 nm target with 35 nm FWHM, as shown in Table 1, it only takes 25 steps 1901957 (8 of 11) 1901957 (8 of 11) Adv. Sci. 2020, 7, 1901957 © 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. © 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 5. Experimental Section The AI Space: The parameters for the RL modules in AI space in this work is of the form like R as S A P ( , ,{ }, ) . Here S denotes the set of state s, which is the set of all reaction parameter and an experimental result that has been experimented. State at timestamp i is denoted by si = {a0,r0,…, ai,ri}; A denotes the set of action ai. In the context of reaction optimization. A is the combinations of reaction parameters, such as reaction time and temperature. Action at timestamp i is denoted by a a a a i i i j i n = … … [ , , , , ] ; 1 as P { } denotes the state transition probability, here pas decides what action a to make under the experimental condition s. This action decision policy is denoted by π; R denotes the reward function under state S and action A . In the environment of reaction optimization, the reward at timestamp i is denoted by r r r r i i i j i n = … … [ , , , , ] 1 , the reward function denotes the mapping ri = R(ai). At a specific experimental condition ai is mapped to get the difference between the experimental and the target then get the reward ri. Specification of Instruction Symbols in MAOS: Table 2 symbolizes and parametrizes all the modules in MAOS, with time dependence. θ is defined as the vector that concludes all adjustable parameters as T st t W F t Th Th t Tc Tc t pV t fr t pO t Ip Ia r fps et vi ti wi tf P P tr A P H C V G O i f θ η τ =   , , , , , , , , , , , , , , , , , , exit, , , , , , , , , , , , , , ,state   (5) (5) The θ vector delivers orders to all the modules in MAOS for the given project, and it also guarantees the repeatability for the typical task. The materials, like reagents, are also parametrized to make the whole system digitalized. Ri is defined the liquid reagent, as Ri = [i, vi]. 4. Conclusion To conclude, MAOS was systematically archi- tectured with language, compiler, and opera- tion modules. For the general experiments, MAOS can separate the human-being from the lab location through the VR-robot inter- action through the 4G/5G network, and for the post-trained experiments, MAOS can finish the experimental task by fully auto- matic procedures with AI assured quality control, reducing the time cost to 1/20. We also highlighted the robustness of MAOS for optimizing both the emission wave- length and FWHM quantities in CdSe QDs. A theoretical description for the wavelength dependent optimal strategy can be derived and fed back to MAOS, which makes the “On-Demand” synthesis feasible for the opto- electronic QDs more than CdSe (see Section S8, Supporting Information). MAOS provides a transparent paradigm for future materials science, which is of more safety, efficiency, and intelligence. C Cool V Vacu G Ga O Gas PL PL m Abs Absorptio CV Com SP Syri SF Solid fe CR Collab L Liqu 1 2( ),if 1 2( ),otherwise a a a a a a a a a i i i i i i i i = + − > − +    , where a a i i [ , ] is one of the largest 1 2( ),if 1 2( ),otherwise a a a a a a a a a i i i i i i i i = + − > − +    , where a a i i [ , ] is one of the largest 1 2( ),if 1 2( ),otherwise a a a a a a a a a i i i i i i i i = + − > − +    , where a a i i [ , ] is one of the largest  sub-boxes of the current partition with history action ai inside. The actions selected will finally be rounded into the nearest integral multiple of the resolution vector Δa, which is decided by the physics condition of reaction parameters (here a Th t s i i H ∆ = ∆ ∆ = ° ( , ) (5 C,3 ) , Thi and ti H are the heating temperature and heating time, respectively). A specific reaction parameter Δaj large enough can be made, then the local optimization would have little influence on this dimension because of the round procedure. 4. Conclusion For global search mentioned above, new actions would be selected in the most massive sub-boxes. Table 2. Specification of instruction symbols and parameter vectors in MAOS. Symbol Module Parameters Parameter vector A Chembox Temperature, stirring speed, time [T, st, tA] P Photonic module Power intensity, wavelength, time [W, F, tP] H Heating module Temperature, temperature gradient, time Th Th tH [ , , ]  C Cooling module Temperature, temperature gradient, time Tc Tc tC [ , , ]  V Vacuum pump Pressure, time [pV, tV] G Gas supplier Gas flow rate, time [fr, tG] O Gas treatment Pressure, time [pO, tO] PL PL measurement Excitation source, integration time [exit, Ip] Abs Absorption measurement Integration time [Ia] CV Computer vision Resolution, fps, exposure time [r, fps, et] SP Syringe pump Injection volume, time, injection rate, times of injection [vi, ti, η, τ] SF Solid feeding module Feeding weight, feeding time [wi, tf] CR Collaborative robot Initial point, final point, time [Pi, Pf,tr] L Liquid transfer State (up or down) [state] Table 2. Specification of instruction symbols and parameter vectors in MAOS. nanomaterials with which can be synthesized in MAOS, which provides a uniform para- digm toward the “On-Demand” synthesis. Like in this case, the “On-Demand” functions can be further benefited from the feedback of theo- retical endorsement, MAOS also can work for the optimiza- tion of QDs quality with large size through the time-dependent interval injection strategy. Section S8 (Supporting Information) summarizes several optoelectronic, photothermic inorganic FWHM FWHM0 0 1 1 0 1 exp 1 exp 1 exp 1 0 1 =     =     ξ ξ µ ητ µ τ µ − −     −         −         − − R R R R t t t t t (4) (4) Here ξ0 is a constant, 〈R〉t and 〈R〉0 is the average radius at time t and initiate state, µ is the distribution length defined in the reaction domain of each injection, the detail of the deri- vation is in Section S7 (Supporting Information). Figure 6b shows the main numerical results for an operational (τ, η) dependent ~ FWHM R 〈〉 plots, as mentioned in the previous part. The chaos-like discrete experimental data can be well 1901957 (9 of 11) Adv. Sci. 2020, 7, 1901957 1901957 (9 of 11) © 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Sci. 2020, 7, 1901957 Keywords [18] a) S. Kirklin, J. E. Saal, B. Meredig, A. Thompson, J. W. Doak, M. Aykol, S. Rühl, C. Wolverton, npj Comput. Mater. 2015, 1, 15010; b) J. E. Saal, S. Kirklin, M. Aykol, B. Meredig, C. Wolverton, JOM 2013, 65, 1501. artificial intelligence, lab automation, on-demand synthesis, quantum dots Received: July 29, 2019 Revised: January 8, 2020 Published online: February 3, 2020 Received: July 29, 2019 Revised: January 8, 2020 Published online: February 3, 2020 [19] J. Hirschberg, C. D. Manning, Science 2015, 349, 261 [20] E. Kim, K. Huang, A. Tomala, S. Matthews, E. Strubell, A. Saunders, A. McCallum, E. Olivetti, Sci. Data 2017, 4, 170127. [21] P. T. Anastas, J. C. Warner, Green Chemistry: Theory and Practice, Oxford University Press, New York 1998. [1] M. E. Moran, J. Endourol. 2006, 20, 986. [1] M. E. Moran, J. Endourol. 2006, 20, 986. [1] M. E. Moran, J. Endourol. 2006, 20, 986. [22] O. D. Suarez, M. d. M. Fernandez-Carrobles, N. V. Enano, G. B. Garcia, I. S. Gracia, OpenCV Essentials, Packt Publishing, 2014. [2] a) T. L. Isenhour, J. Chem. Inf. Comput. Sci. 1985, 25, 292; Anal. Chem. 1983, 55, 1100A; b) S. E. Eckert-Tilotta, T. L. Isenhour, J. C. Marshall, Anal. Chim. Acta 1991, 254, 215. [2] a) T. L. Isenhour, J. Chem. Inf. Comput. Sci. 1985, 25, 292; Anal. Chem. 1983, 55, 1100A; b) S. E. Eckert-Tilotta, T. L. Isenhour, J. C. Marshall, Anal. Chim. Acta 1991, 254, 215. [23] F. Mirri, A. W. K. Ma, T. T. Hsu, N. Behabtu, S. L. Eichmann, C. C. Young, D. E. Tsentalovich, M. Pasquali, ACS Nano 2012, 6, 9737. [3] G.-Z. Yang, J. Bellingham, P. E. Dupont, P. Fischer, L. Floridi, R. Full, N. Jacobstein, V. Kumar, M. McNutt, R. Merrifield, B. J. Nelson, B. Scassellati, M. Taddeo, R. Taylor, M. Veloso, Z. L. Wang, R. Wood, Sci. Rob. 2018, 3, eaar7650. [24] a) M. L. Littman, Nature 2015, 521, 445; b) D. Silver, A. Huang, C. J. Maddison, A. Guez, L. Sifre, G. van den Driessche, J. Schrittwieser, I. Antonoglou, V. Panneershelvam, M. Lanctot, S. Dieleman, D. Grewe, J. Nham, N. Kalchbrenner, I. Sutskever, T. Lillicrap, M. Leach, K. Kavukcuoglu, T. Graepel, D. Hassabis, Nature 2016, 529, 484. [4] a) Open-Science-Collaboration, Nature 2016, 533, 437; b) M. Baker, Nature 2016, 533, 452. [5] S. Steiner, J. Wolf, S. Glatzel, A. Andreou, J. M. Granda, G. Keenan, T. The authors declare no conflict of interest. [17] M. Bayer, The Architecture of Open Source Applications Volume II: Structure, Scale, and a Few More Fearless Hacks (Eds: G. Wilson, A. Brown), 2012. www.advancedscience.com www.advancedscience.com D. E. Fitzpatrick, E. Sliwinski, J. M. Hawkins, S. V. Ley, Angew. Chem., Int. Ed. 2015, 54, 144; d) A. Adamo, R. L. Beingessner, M. Behnam, J. Chen, T. F. Jamison, K. F. Jensen, J.-C. M. Monbaliu, A. S. Myerson, E. M. Revalor, D. R. Snead, T. Stelzer, N. Weeranoppanant, S. Y. Wong, P. Zhang, Science 2016, 352, 61; e) D. P. Tabor, L. M. Roch, S. K. Saikin, C. Kreisbeck, D. Sheberla, J. H. Montoya, S. Dwaraknath, M. Aykol, C. Ortiz, H. Tribukait, C. Amador-Bedolla, C. J. Brabec, B. Maruyama, K. A. Persson, A. Aspuru-Guzik, Nat. Rev. Mater. 2018, 3, 5; f) X. D. Xiang, X. Sun, G. Briceño, Y. Lou, K.-A. Wang, H. Chang, W. G. Wallace-Freedman, S.-W. Chen, P. G. Schultz, Science 1995, 268, 1738; g) N. J. Gesmundo, B. Sauvagnat, P. J. Curran, M. P. Richards, C. L. Andrews, P. J. Dandliker, T. Cernak, Nature 2018, 557, 228; h) D. Perera, J. W. Tucker, S. Brahmbhatt, C. J. Helal, A. Chong, W. Farrell, P. Richardson, N. W. Sach, Science 2018, 359, 429. database. MAOS conduct the experimental operation by combining the parameter spaces of modules and the chemical reagent, and an operator is essential to connect the two parameters space. As shown in Table 3, there are four well-defined operator symbols with the representation and examples, and each operator correlates the symbolic calculations within the module and chemical parameters. For example, R1 · SP1 can be interpreted as MAOS transfer the liquid chemical R1 by the first Syringe pump, as indexed SP1 from Table 2. Supporting Information Supporting Information is available from the Wiley Online Library or from the author. W. Farrell, P. Richardson, N. W. Sach, Science 2018, 359, 429 [10] D. E. Fitzpatrick, C. Battilocchio, S. V. Ley, ACS Cent. Sci. 2016, 2, 131. [11] F. Häse, L. M. Roch, A. Aspuru-Guzik, Trends Chem. 2019, 1, 282. Acknowledgements Roszak, A. Adamski, B. Sieredzin´ska, M. Mrksich, S. L. J. Trice, B. A. Grzybowski, Chem 2018, 4, 522. [15] A. C. Bédard, A. Adamo, K. C. Aroh, M. G. Russell, A. A. Bedermann, J. Torosian, B. Yue, K. F. Jensen, T. F. Jamison, Science 2018, 361, 1220. [15] A. C. Bédard, A. Adamo, K. C. Aroh, M. G. Russell, A. A. Bedermann, 5. Experimental Section Ri s the solid reagent, as R i w i s i = [ , ] . Here v is the volume of liquid reagent, w is the weight of solid reagent, i is the serial number of the reagent in p g g i For a timestamp i with the previous state si − 1, the policy function decides the action to make by ai = π(si − 1). Then, continue the experiment with ai, and get the reward ri = R(ai). For the MAOS, architecture, the state could be updated by si = si − 1 ∪ {ai, ri}. Policy function π performs local optimizations around the best conditions while searching for unexplored regions to ensure global optimality. For local optimization, the action is decided by searching from the best condition with a full quadratic model. For example, an action could be decided by approximately minimizing the local quadratic model. q a r g a a a a G a a T T = + − + − − ( ) ( ) 1 2( ) ( ) best best best best , where g and G are the gradients and the symmetric matrix estimated, respectively. [26] Table 3. Meaning and usage of operators. Table 3. Meaning and usage of operators. Operator symbol Representation Usage example · Reagent transfer R1 · SP1 (transfer R1 with SP1) + Fusion of reagent R1 + R2 (Adding R2 into R1) × React with environment R1 × A (R1 react in Chembox) \| Characterization PL\|R1 × A (Take PL measurement of R1 in Chembox) ∪ Condition combination R1 × (A ∪ V) (R1 react in Chembox with vacuum environment) q a r g a a a a G a a T T = + − + − − ( ) ( ) 1 2( ) ( ) best best best best , where g and G are the gradients and the symmetric matrix estimated, respectively. [26] dients and the symmetric matrix estimated, respectively. [26 For the global optimization, the action is decided by searching simultaneously in several promising subregions. For example, the reaction space is partitioned into sub-boxes each contains a history state.[26] The actions are chosen by 1901957 (10 of 11) Adv. Sci. 2020, 7, 1901957 Conflict of Interest [16] D. E. Fitzpatrick, T. Maujean, A. C. Evans, S. V. Ley, Angew. Chem., Int. Ed. 2018, 57, 15128. The authors declare no conflict of interest. Acknowledgements [12] P. J. Kitson, G. Marie, J.-P. Francoia, S. S. Zalesskiy [12] P. J. Kitson, G. Marie, J.-P. Francoia, S. S. Zalesskiy, R. C. Sigerson, J. S. Mathieson, L. Cronin, Science 2018, 359, 314. [12] P. J. Kitson, G. Marie, J.-P. Francoia, S. S. Zalesskiy, J. S. Mathieson, L. Cronin, Science 2018, 359, 314. J. S. Mathieson, L. Cronin, Science 2018, 359, 314. This work was supported by the Shenzhen Fundamental Research Foundation (Nos. JCYJ20170818103918295 and JCYJ20180508162801893), and the National Natural Science Foundation of China (Grant No. 21805234). Also supported by funding (No. 2019- INT018) from Shenzhen Institute of Artificial Intelligence and Robotics for Societ (AIRS). The help from CUHK-Shenzhen Library during the preparation of this work is greatly appreciated. [13] L. M. Roch, F. Häse, C. Kreisbeck, T. Tamayo-Mendoza, L. P. E. Yunker, J. E. Hein, A. Aspuru-Guzik, Sci. Rob. 2018, 3, eaat5559. [13] L. M. Roch, F. Häse, C. Kreisbeck, T. Tamayo-Mendoza [13] L. M. Roch, F. Häse, C. Kreisbeck, T. Tamayo-Mendoza, L. P. E J E H i A A G ik S i R b 2018 3 t5559 J. E. Hein, A. Aspuru-Guzik, Sci. Rob. 2018, 3, eaat555 p [14] a) V. Sans, L. Cronin, Chem. Soc. Rev. 2016, 45, 2032; b) T. Klucznik, [14] a) V. Sans, L. Cronin, Chem. Soc. Rev. 2016, 45, 2032; b) T. Klucznik, B. Mikulak-Klucznik, M. P. McCormack, H. Lima, S. Szymkuc´, [14] a) V. Sans, L. Cronin, Chem. Soc. Rev. 2016, 45, 2032; b) T. Klucznik, B. Mikulak-Klucznik, M. P. McCormack, H. Lima, S. Szymkuc´, M. Bhowmick, K. Molga, Y. Zhou, L. Rickershauser, E. P. Gajewska, A. Toutchkine, P. Dittwald, M. P. Startek, G. J. Kirkovits, R. Roszak, A. Adamski, B. Sieredzin´ska, M. Mrksich, S. L. J. Trice, B. A. Grzybowski, Chem 2018, 4, 522. [15] A C Béd d A Ad K C A h M G R ll A A B d B. Mikulak-Klucznik, M. P. McCormack, H. Lima, S. Szymkuc´, M. Bhowmick, K. Molga, Y. Zhou, L. Rickershauser, E. P. Gajewska, M. Bhowmick, K. Molga, Y. Zhou, L. Rickershauser, E. P. Gajewska, A. Toutchkine, P. Dittwald, M. P. Startek, G. J. Kirkovits, R. Roszak, A. Adamski, B. Sieredzin´ska, M. Mrksich, S. L. J. Trice, B. A. Grzybowski, Chem 2018, 4, 522. A C d d A Ad C A h G ll A A d A. Toutchkine, P. Dittwald, M. P. Startek, G. J. Kirkovits, R. Keywords Hinkley, G. Aragon-Camarasa, P. J. Kitson, D. Angelone, L. Cronin, Science 2019, 363, eaav2211. [25] Z. Zhou, X. Li, R. N. Zare, ACS Cent. Sci. 2017, 3, 1337. [26] W. Huyer, A. Neumaier, ACM Trans. Math. Software 2008, 35, 1. [27] A.-C. Bédard, A. Adamo, K. C. Aroh, M. G. Russell, A. A. Bedermann, J. Torosian, B. Yue, K. F. Jensen, T. F. Jamison, Science 2018, 361, 1220. [6] S. Masubuchi, M. Morimoto, S. Morikawa, M. Onodera, Y. Asakawa, [6] S. Masubuchi, M. Morimoto, S. Morikawa, M. Onodera, Y. Asakawa, K. Watanabe, T. Taniguchi, T. Machida, Nat. Commun. 2018, 9, 1413. [6] S. Masubuchi, M. Morimoto, S. Morikawa, M. Onodera, Y. Asakawa, K. Watanabe, T. Taniguchi, T. Machida, Nat. Commun. 2018, 9, 1413. [28] S. Krishnadasan, R. J. C. Brown, A. J. DeMello, J. C. DeMello, Lab Chip 2007, 7, 1434. [7] J. Li, Y. Lu, Y. Xu, C. Liu, Y. Tu, S. Ye, H. Liu, Y. Xie, H. Qian, X. Zhu, J. Phys. Chem. A 2018, 122, 9142. [29] M. V. Kovalenko, L. Protesescu, M. I. Bodnarchuk, Science 2017, 358, 745. [8] a) S. V. Ley, D. E. Fitzpatrick, R. J. Ingham, R. M. Myers, Angew. Chem., Int. Ed. 2015, 54, 3449; b) T. Melanie, M. D. Burke, Angew. Chem., Int. Ed. 2018, 57, 4192. [30] L. E. Brus, J. Chem. Phys. 1984, 80, 4403. [31] C. Wagner, Z. Elektrochem. 1961, 65, 581. [9] a) P. S. Gromski, A. B. Henson, J. M. Granda, L. Cronin, Nat. Rev. Chem. 2019, 3, 119; b) M. H. S. Segler, M. Preuss, M. P. Waller, Nature 2018, 555, 604; c) R. J. Ingham, C. Battilocchio, [32] I. M. Lifshitz, V. V. Slyozov, J. Phys. Chem. Solids 1961, 19, 35. [33] M. D. Clark, S. K. Kumar, J. S. Owen, E. M. Chan, Nano Lett. 2011, 11, 1976. 1901957 (11 of 11) © 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Sci. 2020, 7, 1901957 © 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
https://openalex.org/W2138441459	https://tspace.library.utoronto.ca/bitstream/1807/85857/1/13062_2007_Article_59.pdf	English	null	Extensive parallelism in protein evolution	Biology direct	2,007	cc-by	12,745	BioMed Central BioMed Central BioMed Central Extensive parallelism in protein evolution Address: 1Institute for Information Transmission Problems of the Russian Academy of Sciences (Kharkevich Institute), Bolshoi Karetny pereulok 19, Moscow, 127994, Russia, 2Department of Ecology and Evolutionary Biology, Princeton University, Princeton, NJ 08544, USA, 3Section on Ecology, Behavior and Evolution, University of California at San Diego, La Jolla, CA 92093, USA, 4Department of Computer Science and Banting & Best Department of Medical Research, University of Toronto, Toronto, ON M5S 3J4, Canada, 5Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Berkeley, CA 94720, USA, 6Department of Energy Joint Genome Institute, 2800 Mitchell Drive, Walnut Creek, CA 94598, USA and 7Life Sciences Institute and Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, MI 48109-2216, USA Email: Georgii A Bazykin - gbazykin@princeton.edu; Fyodor A Kondrashov - fkondrashov@ucsd.edu; Michael Brudno - brudno@cs.toronto.edu; Alexander Poliakov - avpoliakov@lbl.gov; Inna Dubchak - ildubchak@lbl.gov; Alexey S Kondrashov* - kondrash@umich.edu mail: Georgii A Bazykin - gbazykin@princeton.edu; Fyodor A Kondrashov - fkondrashov@ucsd.edu; g y g y p y Michael Brudno - brudno@cs.toronto.edu; Alexander Poliakov - avpoliakov@lbl.gov; Inna Dubchak - ildubchak@lbl.gov; Alexey S Kondrashov* - kondrash@umich edu ael Brudno - brudno@cs.toronto.edu; Alexander Poliakov - avpoliakov@lbl.gov; Inna Dubchak - ildubchak@lbl.go y S Kondrashov* - kondrash@umich.edu * Corresponding author Received: 25 May 2007 Accepted: 16 August 2007 Published: 16 August 2007 Biology Direct 2007, 2:20 doi:10.1186/1745-6150-2-20 This article is available from: http://www.biology-direct.com/content/2/1/20 © 2007 Bazykin et al; licensee BioMed Central Ltd. y This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creative which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cit Biology Direct Open Access Background g Although evolution is primarily divergent, parallel, con- vergent, and reversing changes in independently evolving lineages, collectively known as homoplasy, are not uncommon [1]. In particular, homoplasy should be per- vasive when evolution is considered at the level of DNA or protein sequences, because there are only 4 or 20 possible states for each site. When evolving sequences become suf- ficiently dissimilar, homoplasious changes prevent their further divergence, leading to evolutionary saturation [2] and interfering with phylogenetic reconstructions [3]. Sev- eral instances of rapid parallel evolution of similar pro- teins, apparently driven by positive selection, have been observed (e. g., [4,5]). However, contribution of parallel changes to independent evolution of similar proteins has not been investigated quantitatively at the genomic scale. Data on parallel amino acid replacements in proteins can shed light on several key aspects of their evolution. In par- ticular, because the per site rate of nonsynonymous nucle- otide substitutions dN is, on average, ~10 times smaller than the per site rate of synonymous substitutions dS, a vast majority of amino acids must be, most of the time, under substantial negative selection [2,6]. Still, neutral theory claims that most of amino acid replacements which occur in evolution are selectively neutral and, thus, must accumulate at the same rate as synonymous substi- tutions [6,7], as long as the latter are approximately assumed to be selectively neutral. The postulated rela- tively small proportion of rapidly evolving non-synony- mous sites can be investigted using data on parallel evolution of proteins. Informally, replacements along path I mark evolvable sites, and replacements, at these sites, along path II tell us how such evolvable sites evolve. Data on rates of parallel, and of coincident divergent, evolution of proteins are pre- sented in Tables 1, 2, 3. In agreement with the previous estimates [10-12], the overall rate of nonsynonymous substitutions is ~10% of the rate of synonymous substitu- tions. In contrast, the average rate P of parallel nonsynon- ymous substitutions is much higher and constitutes ~50% (yeast), ~60% (mammals) or ~80% (Drosophila) of the rate of parallel synonymous substitutions. Assuming that the fitness landscape for an amino acid site did not change in the course of evolution represented by each tree, we can draw a number of conclusions from this figure. Abstract Background: Independently evolving lineages mostly accumulate different changes, which leads to their gradual divergence. However, parallel accumulation of identical changes is also common, especially in traits with only a small number of possible states. Results: We characterize parallelism in evolution of coding sequences in three four-species sets of genomes of mammals, Drosophila, and yeasts. Each such set contains two independent evolutionary paths, which we call paths I and II. An amino acid replacement which occurred along path I also occurs along path II with the probability 50–80% of that expected under selective neutrality. Thus, the per site rate of parallel evolution of proteins is several times higher than their average rate of evolution, but still lower than the rate of evolution of neutral sequences. This deficit may be caused by changes in the fitness landscape, leading to a replacement being possible along path I but not along path II. However, constant, weak selection assumed by the nearly neutral model of evolution appears to be a more likely explanation. Then, the average coefficient of selection associated with an amino acid replacement, in the units of the effective population size, must exceed ~0.4, and the fraction of effectively neutral replacements must be below ~30%. At a majority of evolvable amino acid sites, only a relatively small number of different amino acids is permitted. Conclusion: High, but below-neutral, rates of parallel amino acid replacements suggest that a majority of amino acid replacements that occur in evolution are subject to weak, but non-trivial, selection, as predicted by Ohta's nearly-neutral theory. Reviewers: This article was reviewed by John McDonald (nominated by Laura Landweber), Sarah Teichmann and Subhajyoti De, and Chris Adami. Page 1 of 13 (page number not for citation purposes) Page 1 of 13 (page number not for citation purposes) http://www.biology-direct.com/content/2/1/20 Biology Direct 2007, 2:20 http://www.biology-direct.com/content/2/1/20 Analysis of alignments y g Phylogenetic trees used in our analysis are shown in Fig. 1. Evolutionary paths connecting human and dog, D. yakuba and D. erecta, and S. bayanus and S. mikatae are treated as "paths I", and mouse-rat, D. melanogaster-D. simulans, and S. cerevisiae-S. paradoxus paths, which are the shortest in the respective trees, are "paths II". If the two species connected by path I, and the two species con- nected by path II, display the same amino acid difference at the same site (say, amino acid A in one species from each pair, and amino acid B in the other species from each pair), parsimony implies that the same unordered amino acid replacement (A↔B) occurred along both paths I and II. In the case of Drosophila tree, with ((D. yakuba, D. erecta), (D. melanogaster, D. simulans)) topology (Fig. 1), we can be sure that replacements along both paths were parallel, i. e., occurred in the same direction (either A→B or B→A), although determining this direction is not pos- sible. In the case of mammalian or yeast trees, replace- ments along path I and path II may also occur in the opposite directions, thus constituting a reversal; for exam- ple, a site with A in dog and rat and B in human and mouse can emerge due to A→B replacement in the lineage that led to the common ancestor of human, mouse, and rat, followed by B→A replacement in the rat lineage. How- ever, the contribution of reversals should be relatively small, because the edge connecting nodes where dog and human lineages (or S. bayanus and S. mikatae lineages) branch off is rather short [8,9]. Background The minimal number of diverging sequences which makes it possible to study parallel evolution is four, because the phylogenetic tree must contain at least two nonoverlapping evolutionary paths, I and II. Here, we use sequences of three quadruplets of closely related genomes from mammals, fruit flies Drosophila, and yeasts Saccharo- myces, and consider parallel evolution in these quadru- plets at the level of all known proteins encoded by a genome. We define the rate of parallel evolution along path II as the rate with which allele replacements of a par- ticular kind accumulated along this path, provided that the same replacements did occur, at the orthologous loci, along path I, and measure this rate for various kinds of amino acid replacements. Results p p Reviewed by John McDonald (nominated by Laura Land- weber), Sarah Teichmann and Subhajyoti De, and Chris Adami. For the full reviews, please go to the Reviewers' comments section. Page 2 of 13 (page number not for citation purposes) Variability among individual sites Thus, the average value of s at sites of parallel evolution, S = ∫p'(s)sds, must be higher than f-1(P), because sites with above-average selection coefficients make smaller contri- butions to evolution than sites with below-average selec- tion coefficients (Fig. 2). Assuming any particular shape of p(s) (e. g., that p(s) is a gamma-distribution with a partic- ular shape parameter; [14]), we can calculate S which cor- responds to the observed P. y g y , g Phylogenetic trees used in our analysis, drawn to scale. For each edge, the average per site divergence at non- synonymous (red) and synonymous (blue) sites is shown. Green lines show paths I, which are used to identify evolva- ble sites, and magenta lines show paths II, which are used to measure rates of evolution at these sites. y g y g Phylogenetic trees used in our analysis, drawn to scale. For each edge, the average per site divergence at non- synonymous (red) and synonymous (blue) sites is shown. Green lines show paths I, which are used to identify evolva- ble sites, and magenta lines show paths II, which are used to measure rates of evolution at these sites. Moreover, we can estimate the maximal fraction of effec- tively neutral sites (those evolving at essentially neutral rate 1) consistent with a given P. Indeed, the contribution of a site with some v = v0 to the reduction of P is (P-v0)v0. This contribution is maximal when d [(P-v)v]/dv = 0, i. e. when v0 = 0.5P. Thus, the fraction x of neutrally evolving sites is maximal, under a given P, when there are only two kinds of sites, those evolving with rates 1 and 0.5P. The average rate of evolution with such a q(v), [x+(1-x)P2/4]/ [x+(1-x)P/2)], equals P when x = (P/(2-P))2. Thus, when P = 0.9, 0.8, 0.7, 0.6, and 0.5, the maximal possible fraction of effective neutral sites is 0.67, 0.44, 0.30, 0.18, and 0.11, respectively (Fig. 3). Of course, in reality this fraction must be lower, since p(s) and q(v) are not concentrated at just two points. switches between fixations of the two alleles, is deter- mined by s: v = f(s). When s is large enough (>>Ne -1), the site is usually occupied by the best allele, and negative selection prevents fixation of the inferior allele. Variability among individual sites Let us assume that only two alleles (amino acid variants) are possible at a locus (site). The selection coefficient asso- ciated with this pair of alleles is s = 1 - w1/w2, where w1 and w2 are constant fitnesses conferred by the two alleles (w1 < w2). Naturally, s can vary between 1 (lethality of the inferior allele) and 0 (selective neutrality). Under given parameters of mutation and the effective population size Ne, the rate of evolution at the site v, i. e. the frequency of Page 2 of 13 (page number not for citation purposes) Page 2 of 13 (page number not for citation purposes) Biology Direct 2007, 2:20 http://www.biology-direct.com/content/2/1/20 Phylogenetic trees used in our analysis, drawn to scale Figure 1 Phylogenetic trees used in our analysis, drawn to scale. For each edge, the average per site divergence at non- synonymous (red) and synonymous (blue) sites is shown. Green lines show paths I, which are used to identify evolva- ble sites, and magenta lines show paths II, which are used to measure rates of evolution at these sites. Dog Human Mouse Rat D. erecta D. yakuba D. melanogaster D. simulans S. bayanus S. mikatae S. cerevisiae S. paradoxus 0.039 0.285 0.027 0.170 0.015 0.097 0.020 0.117 0.041 0.380 0.015 0.107 0.014 0.124 0.015 0.121 0.008 0.074 0.009 0.061 0.063 0.764 0.037 0.361 0.023 0.257 0.023 0.117 0.012 0.145 tion at the site. Similarly, the distribution of v among such sites is q'(v) = vq(v)/C. Thus, the average rate of parallel evolution is Dog Human Mouse Rat D. erecta D. yakuba D. melanogaster D. simulans S. bayanus S. mikatae S. cerevisiae S. paradoxus 0.039 0.285 0.027 0.170 0.015 0.097 0.020 0.117 0.041 0.380 0.015 0.107 0.014 0.124 0.015 0.121 0.008 0.074 0.009 0.061 0.063 0.764 0.037 0.361 0.023 0.257 0.023 0.117 0.012 0.145 P = ∫p(s)f2(s)ds/C = ∫v2q(v)dv/C (2) (2) What does knowledge of P (say, 0.7; here and below P is given in the units of rate of neutral evolution M) tell us about p(s) or q(v)? The simplest option is that q'(v) and, thus, its "parent" distribution q(v), is concentrated at P = 0.7, and p(s) is concentrated at s0 = f-1(0.7) ~1.5/(4Ne) (Fig. 3). Of course, this cannot be the case, since different amino acid sites evolve at widely different rates [6]. Page 3 of 13 (page number not for citation purposes) Variability among individual sites In con- trast, when s < Ne -1, either allele can be fixed at given moment. Assuming that the rates of forward and back- ward mutation are equal, v is the highest, and equals to the rate of neutral evolution M, with s = 0, and monoto- nously approaches 0 when s increases ([6,13]; Fig. 2). Consideration of only two alleles, involved in the observed parallel replacement, is sufficient because the ratio of frequencies of any two alleles under mutation- selection-drift equilibrium does not depend on fitnesses of any other possible alleles, as at such equilibrium the reciprocal fluxes of allele replacements are always equal [13]. Discussion Patterns in parallel evolution Our data show that rate of parallel evolution of coding sequences is elevated: the probability of a change along path II is above-average at sites where the same nucleotide change also occurred along path I. This is the case even for synonymous substitutions: a synonymous substitution along path II is ~20% more probable when the same sub- stitution also occurred along path I (Tables 1, 2, 3). As far as synonymous substitutions are concerned, between- sites heterogeneity of mutation rates and/or of the strength of selection can lead to the difference observed. Both mechanisms are feasible, because mutation rate var- ies between nucleotide sites [15] and because synony- mous sites are not exactly neutral [16], and they are not mutually exclusive, but we did not attempt to determine their relative importance. The rate of parallel synonymous Let us characterize all protein sites by their distributions of s, p(s), and of v, q(v). Then, q(v) = p(f-1(v)), where f-1 is a function which relates s to the rate of evolution: s = f-1(v). The average rate of evolution at all sites is (all integrals are taken from 0 to 1) C = ∫p(s)f(s)ds = ∫vq(v)dv (1) (1) The distribution of s within sites where parallel evolution took place, i. e. Discussion Patterns in parallel evolution the same replacement occurred along both short paths I and II, is given not by p(s) per se, but by p'(s) = p(s)f(s)/C, because the probability that a replacement occurred along path I is proportional to the rate of evolu- Page 3 of 13 (page number not for citation purposes) Page 3 of 13 (page number not for citation purposes) http://www.biology-direct.com/content/2/1/20 Biology Direct 2007, 2:20 Table 1: Divergence between mouse and rat at sites of divergence between human and dog Synonymous Nonsynonymous Overall1 Parallel2 Overall3 Coincident Parallel4 Divergent5 Same site Different site Pairs of nucleotides AC 0.028 0.047 0.0028 (10.1%) 0.024 (122, 51.4%) 0.013 (280, 27.0%) 0.010 (306, 21.1%) AG 0.089 0.109 0.0102 (11.5%) 0.069 (1570, 62.9%) 0.040 (448, 36.2%) 0.037 (923, 33.9%) AT 0.021 0.024 0.0017 (8.4%) 0.028 (80, 117.4%) 0.008 (118, 32.6%) 0.006 (148, 23.5%) CG 0.029 0.036 0.0033 (11.5%) 0.021 (156, 57.3%) 0.014 (302, 38.1%) 0.011 (332, 29.5%) CT 0.076 0.090 0.0060 (7.9%) 0.051 (492, 56.5%) 0.027 (239, 30.3%) 0.020 (685, 22.6%) GT 0.027 0.040 0.0019 (7.0%) 0.023 (69, 56.6%) 0.010 (175, 25.7%) 0.006 (168, 15.8%) Average 0.045 0.058 0.0043 (9.6%) 0.036 (2489, 62.1%) 0.019 (1562, 32.2%) 0.015 (2562, 26.0%) Genes6 Low dN 0.043 0.059 0.0019 (4.3%) 0.028 (315, 47.1%) 0.012 (145, 21.1%) 0.009 (195, 14.8%) Intermediate dN 0.046 0.058 0.0045 (9.9%) 0.036 (895, 61.7%) 0.017 (526, 29.6%) 0.013 (785, 21.8%) High dN 0.047 0.057 0.0086 (18.5%) 0.038 (1279, 67.6%) 0.021 (891, 37.6%) 0.019 (1582, 32.7%) Chemical distance between amino acids7 1 0.045 0.058 0.0043 (9.6%) 0.038 (516, 65.0%) 0.012 (344, 20.1%) 0.016 (662, 27.3%) 1.5 or 2 0.045 0.058 0.0043 (9.6%) 0.033 (1537, 57.8%) 0.025 (834, 42.5%) 0.014 (1268, 24.7%) 2.5 or 3 0.045 0.058 0.0043 (9.6%) 0.031 (436, 53.7%) 0.023 (384, 40.4%) 0.018 (632, 31.0%) Table 1: Divergence between mouse and rat at sites of divergence between human and dog 1Mouse-rat divergence at all 4-fold synonymous nucleotide sites. 1Mouse-rat divergence at all 4-fold synonymous nucleotide sites. 2Mouse-rat divergence only at those 4-fold synonymous sites where human and dog also underwent divergence in the same unordered pair of nucleotides. 3Mouse-rat divergence at all nondegenerate nonsynonymous sites. The magnitude of this divergence relative to overall synonymous mouse-rat divergence is presented in parentheses. 4Mouse-rat divergence at nondegenerate nonsynonymous sites where human and dog also underwent divergence in the same unordered pair of nucleotides and of amino acids. Discussion Patterns in parallel evolution The number of sites of such mouse-rat divergence and the magnitude of this divergence relative to parallel synonymous mouse-rat divergence are presented in parentheses. 5Mouse-rat divergence at nondegenerate nonsynonymous nucleotide sites which belong to amino acid sites where human and dog also underwent nonsynonymous divergence, either at the same or at a different nucleotide site, but in a different unordered pair of nucleotides and of amino acids. The number of sites of such mouse-rat divergence and the magnitude of this divergence relative to parallel synonymous mouse-rat divergence are presented in parentheses. 6Genes were subdivided into three bins of equal sizes, according to their rates of nonsynonymous evolution along the path between mouse and rat. 7Rank of the Miyata distance between the human and dog amino acids among the distances between all pairs of amino acids that can arise due to a substitution at the same nucleotide site substitutions was used as a proxy for the rate of selectively neutral parallel nonsynonymous evolution. 1Mouse-rat divergence at all 4-fold synonymous nucleotide sites. substitutions was used as a proxy for the rate of selectively neutral parallel nonsynonymous evolution. substitutions was used as a proxy for the rate of selectively neutral parallel nonsynonymous evolution. mous evolution, but still higher than the average rate of all nonsynonymous substitutions (Tables 1, 2, 3). mous evolution, but still higher than the average rate of all nonsynonymous substitutions (Tables 1, 2, 3). The rate of parallel nonsynonymous substitutions P is ele- vated to a much larger extent than that of synonymous substitutions. Indeed, in the units of the rate of neutral evolution M, the average rate of all nonsynonymous sub- stitutions is only ~0.1, but P is 0.5–0.8 (Tables 1, 2, 3). The sign of this difference is as expected: because rates of nonsynonymous substitutions are strongly heterogeneous across sites [6,17], a replacement observed along path I must be a good predictor of the probability of a replace- ment, at the same site, along path II. P is higher for rapidly evolving proteins or when a replacement involves two chemically similar amino acids. The rate of coincident divergent evolution, such that an amino acid replacement A↔C occurred along path II at an amino acid site where an amino acid replacement A↔B occurred along path I, is always much lower than the rate of parallel nonsynony- Patterns in parallel evolution of proteins specifically reveal properties of only evolvable nonsynonymous sites. Indeed, the low overall rate of protein evolution implies that ~90% of all new nonsynonymous mutations are rejected by negative selection, but tells us little about sites where nonsynonymous substitutions do occur. In princi- ple, if we assume that selection is constant, distributions p'(s) and q'(v), which characterize evolvable sites, can be derived from p(s), because f(s) is known theoretically (Fig. 2). However, this is currently impossible, because our data on p(s) are too crude. In fact, p'(s) depends only on the left tail of p(s), roughly corresponding to 4Nes < 10 (as sites under stronger selection do not evolve), and we know that this tail contains 10–20% of the distribution [18,19], but nothing definite about its shape. It is some- times assumed that p(s) [14], as well as q(v) [20], are Page 4 of 13 (page number not for citation purposes) Page 4 of 13 (page number not for citation purposes) http://www.biology-direct.com/content/2/1/20 Biology Direct 2007, 2:20 Table 2: Divergence between D. melanogaster and D. simulans at sites of divergence between D. yakuba and D. substitutions was used as a proxy for the rate of selectively neutral parallel nonsynonymous evolution. erecta.1 Synonymous Nonsynonymous Overall1 Parallel2 Overall3 Coincident Parallel4 Divergent5 Same site Different site Pairs of nucleotides AC 0.028 0.040 0.0025 (9.2%) 0.034 (110, 86.5%) 0.011 (99, 28.7%) 0.009 (119, 23.7%) AG 0.055 0.070 0.0056 (10.3%) 0.053 (431, 75.3%) 0.023 (141, 32.7%) 0.020 (207, 28.8%) AT 0.035 0.047 0.0024 (6.7%) 0.041 (119, 86.4%) 0.014 (93, 28.9%) 0.007 (74, 15.3%) CG 0.022 0.031 0.0033 (15.0%) 0.036 (113, 114.9%) 0.011 (88, 35.0%) 0.010 (91, 31.2%) CT 0.060 0.068 0.0037 (6.2%) 0.045 (152, 65.8%) 0.020 (103, 29.6%) 0.010 (98, 15.3%) GT 0.027 0.036 0.0020 (7.6%) 0.032 (58, 88.1%) 0.010 (83, 27.9%) 0.010 (67, 27.2%) Average 0.038 0.049 0.0033 (8.6%) 0.040 (983, 82.2%) 0.015 (607, 30.5%) 0.011 (656, 22.8%) Genes6 Low dN 0.035 0.048 0.0008 (2.3%) 0.026 (97, 53.8%) 0.006 (41, 13.6%) 0.004 (33, 7.9%) Intermediate dN 0.038 0.048 0.0029 (7.4%) 0.037 (319, 77.1%) 0.012 (175, 25.4%) 0.009 (182, 18.5%) High dN 0.041 0.051 0.0072 (17.5%) 0.046 (567, 90.9%) 0.019 (391, 37.5%) 0.015 (441, 29.1%) Chemical distance between amino acids7 1 0.038 0.049 0.0033 (8.6%) 0.048 (349, 98.2%) 0.008 (143, 16.7%) 0.010 (160, 19.8%) 1.5 or 2 0.038 0.049 0.0033 (8.6%) 0.037 (461, 75.7%) 0.016 (300, 32.0%) 0.011 (338, 21.8%) 2.5 or 3 0.038 0.049 0.0033 (8.6%) 0.032 (173, 65.4%) 0.019 (164, 39.3%) 0.015 (158, 30.6%) 1This Table is analogous to Table 1. ergence between D. melanogaster and D. simulans at sites of divergence between D. yakuba and D. erecta.1 observed, sites under such selection are rather rare, at least in mammalian genomes [22]. observed, sites under such selection are rather rare, at least in mammalian genomes [22]. gamma-distributions, but these two assumptions cannot be simultaneously correct, and the issue remains obscure. Thus, P < 1 is not surprising, and can be due to both vari- able and constant selection. If fitness landscapes are dif- ferent along paths I and II, some replacements permitted along path I are forbidden along path II. In the extreme case of independent landscapes along the two paths, the rate of parallel evolution would not be elevated at all. In contrast to the case of P > 1, a single change of the fitness landscape, somewhere between paths I and II, is perfectly sufficient to explain P < 1. Selection at evolvable sites The best way to make sense of the data presented in Tables 1, 2, 3 is to consider possible reasons for a relatively small deviation of the average rate of parallel protein evolution P from the rate of neutral evolution M. Indeed, according to the neutral theory [6,7], most of accepted amino acid replacements are neutral, and in this case the two rates should coincide. The difference between P and M could be due to either variable or constant selection. Page 5 of 13 (page number not for citation purposes) Variable fitness landscape A permitted replacement along path I can be either effec- tively neutral or driven by substantial positive selection. Two observations make the first option unlikely. Tempo- rarily available opportunities for neutral evolution have been described by the covarion model [23], which assumes that selection at an amino acid site switches on and off as a result of amino acid substitutions elsewhere in the protein. Because the environment of an amino acid site is more stable within a slowly-evolving protein, cov- arion provides the least opportunity for path-specific neu- trality and should lead to the smallest reduction of P in such proteins. However, we observed exactly the opposite: P was the lowest in proteins with low dN (Tables 1, 2, 3). Further, the evolutionary distance between paths I and II The only feasible scenario which could lead to P > 1 is pos- itive selection caused by multiple changes in the fitness landscape. Only in this case can positive selection drive parallel replacements along both paths I and II. Indeed, two selection-driven parallel replacements can hardly be caused by a single change in the fitness landscape in the common ancestor of all the 4 species (Fig. 1), because a replacement must follow such a change with only a rela- tively short delay [6]. Therefore, at least two independent changes in the fitness landscape are necessary for parallel- ism – one in each lineage. Although rapidly fluctuating selection, resulting in dN > dS [21] and in parallel evolu- tion ([4] and references therein) has been repeatedly Page 5 of 13 (page number not for citation purposes) http://www.biology-direct.com/content/2/1/20 Biology Direct 2007, 2:20 Table 3: Divergence between S. cerevisiae and S. paradoxus at sites of divergence between S. mikatae and S. Variable fitness landscape bayanus.1 Synonymous Nonsynonymous Overall1 Parallel2 Overall3 Coincident Parallel4 Divergent5 Same site Different site Pairs of nucleotides AC 0.055 0.061 0.0037 (6.8%) 0.035 (132, 57.2%) 0.018 (217, 29.6%) 0.012 (223, 20.0%) AG 0.198 0.218 0.0189 (9.5%) 0.110 (1941, 50.5%) 0.070 (459, 31.9%) 0.054 (851, 24.6%) AT 0.044 0.064 0.0029 (6.6%) 0.038 (116, 59.2%) 0.015 (158, 24.0%) 0.006 (107, 9.2%) CG 0.061 0.074 0.0053 (8.6%) 0.039 (117, 52.7%) 0.021 (214, 28.3%) 0.014 (129, 18.6%) CT 0.169 0.208 0.0090 (5.3%) 0.095 (476, 45.7%) 0.044 (249, 21.1%) 0.025 (310, 11.9%) GT 0.044 0.062 0.0025 (5.6%) 0.047 (76, 76.1%) 0.012 (143, 19.4%) 0.007 (64, 11.1%) Average 0.095 0.115 0.0070 (7.4%) 0.061 (2858, 53.0%) 0.030 (1440, 26.2%) 0.020 (1684, 17.1%) Genes6 Low dN 0.085 0.110 0.0029 (3.4%) 0.049 (403, 45.0%) 0.015 (135, 13.7%) 0.011 (161, 10.1%) Intermediate dN 0.101 0.117 0.0070 (6.9%) 0.062 (976, 52.6%) 0.027 (435, 23.0%) 0.018 (527, 15.2%) High dN 0.103 0.117 0.0128 (12.5%) 0.064 (1479, 54.3%) 0.038 (870, 32.1%) 0.024 (996, 20.3%) Chemical distance between amino acids7 1 0.095 0.115 0.0070 (7.4%) 0.060 (969, 52.7%) 0.016 (308, 13.7%) 0.019 (434, 16.2%) 1.5 or 2 0.095 0.115 0.0070 (7.4%) 0.061 (1482, 53.0%) 0.032 (729, 27.9%) 0.018 (892, 15.4%) 2.5 or 3 0.095 0.115 0.0070 (7.4%) 0.053 (407, 46.7%) 0.040 (403, 34.6%) 0.032 (358, 28.0%) 1This Table is analogous to Table 1. Divergence between S. cerevisiae and S. paradoxus at sites of divergence between S. mikatae and S. bayanus. 1This Table is analogous to Table 1. observed. P < 1 is always expected if the two amino acids involved in the substitution are under constant selection (Fig. 2), as long as mutation is symmetric (see [16]). P = 0.7 is consistent with the average selection coefficient associated with an accepted amino acid replacement being 1.5/(4Ne) = 0.375Ne -1 or more, and with the frac- tion of strictly neutral replacements being ~30% or less. In fact, selection on accepted replacements is probably even stronger, because we underestimated the rate of neutral evolution, which in mammals is ~10% higher, at non- CpG-prone sites, than the rate of synonymous evolution [16]. is much larger in the mammalian phylogeny than in the other two phylogenies (Fig. 1), which should lead to more on and off switches of selective constraint between the paths and, therefore, to stronger reduction in P in mam- mals. However, P is the lowest in yeasts, and not in mam- mals. Variable fitness landscape In contrast, the deepest reduction of P in slowly-evolving proteins is consistent with the hypothesis that positive selection plays a larger role in the evolution of slowly- evolving sites [24]. Still, it is hard to imagine that > 50% of all amino-acid replacements accepted by slowly-evolv- ing proteins were adaptive, and this assumption is neces- sary to explain P < 0.5 in such proteins (Tables 1, 2, 3) through path-specific positive selection. Also, it seems implausible that changes of the fitness landscape favor radical replacements more frequently than conservative replacements, and the reduction in P is the strongest for substitutions which radically change the amino acid (Tables 1, 2, 3). Finally, the lowest rate of coincident divergent evolution, relative to P, in slowly-evolving pro- teins (Tables 1, 2, 3) implies that in such proteins the same pair of amino acids most often confers the two high- est fitnesses along the whole phylogenetic tree, which appears to be inconsistent with frequent changes of the fitness landscape, unless such changes usually do not affect which two amino acids are the best. Thus, our data on parallel evolution of proteins suggest that a majority of amino acid replacements occur at sites which are not effectively neutral, but experience weak selection, in agreement with the nearly-neutral theory [25,26]. The highest P for replacements which involve the most chemically similar pairs of amino acids (Tables 1, 2, 3) is also consistent with this explanation, because such replacements must be under weaker selection than radical replacements. If evolving lineages are at mutation-selection-drift equi- librium, the overall numbers of slightly deleterious and slightly beneficial replacements must be equal, although at a given moment a site is more often occupied by a (slightly) superior allele and, thus, is more often under negative selection. Still, at any moment, the fraction of amino acid sites occupied by slightly inferior amino acids must be well above ~10% of all evolvable protein sites Constant fitness landscape Methods Orthologs of human, dog, mouse, and rat, as well as of four species of Saccharomyces, were identified by the bidi- rectional best protein BLAST [32] hit approach [33] using the Entrez retrieval system [34] of annotated protein cod- ing genes from complete yeast and mammalian genomes, available at NCBI [35]. Alignments of amino acid sequences for each quadruplet were made using ClustalW [36] and reverse transcribed to obtain alignments of DNA sequences. Rate of evo tions of 4N Figure 2 Rate of evolution and frequency of the inferior allele as func tions of 4Nes Figure 2 Rate of evolution and frequency of the inferior allele as functions of 4Nes. Rate of evolution v is in the units of rate of neutral evolution M [13]. To align the whole genome assemblies of D. melanogaster, D. simulans, D. yakuba and D. erecta, we used whole genome multiple alignment algorithm implemented in the VISTA Genome Pipeline (Brudno et al. in prep.). This algorithm consists of two major modules – Pairwise Alignment of Sister Taxa and Progressive Multiple Align- ment. First module uses a glocal (hybrid global/local) approach based on a reimplementation of the original Shuffle-LAGAN (S-LAGAN) chaining algorithm [37,38] combined with a post-processing stage called SuperMap. The S-LAGAN chaining takes as input a set of local align- ments between the two sequences and returns the maxi- mal scoring subset of these under certain gap criteria. In order to allow our alignments to incorporate duplications in both genomes, SuperMap algorithm takes two S- LAGAN outputs, for each sequence as the base. We then classified all local alignments as belonging to both chains, and consequently orthologous (best bidirectional hits), or being in only one chain, and hence a duplication. After the two pairs of sister taxa (melanogaster/simulans and yakuba/erecta) were aligned, we used a progressive general- ization of the pairwise SuperMap algorithm to align the two alignments to each other, and get a 4-way alignment. Our algorithm is based on finding a maximum weighted matching in a graph, with the weights specified by the out- group genomes, to order the individual alignment blocks in the likely order of the ancestors of melanogaster/simulans and yakuba/erecta. After that we align the resulting ances- trally-ordered alignments to each other using LAGAN [39]. Methods To restrict our analysis to one-to-one orthologs, all cases in which an ORF in one species was aligned to more than one ORF in another species were excluded from the analysis. Since a substantial fraction of Drosophila coding regions had stretches of ambiguities and internal stop codons, we assumed that such stretches could have erro- neous length due to sequencing errors. Therefore, if it allowed us to reduce the numbers of internal stop codons, (Fig. 2) and, thus, well above 1% of all ~107 protein sites of an organism. Selection with s ~10-5 acting against > 105 slightly deleterious amino acids must cause a high genetic load [27]. Constant fitness landscape p The simpler assumption of constant selection seems to provide a more plausible explanation for the patterns Page 6 of 13 (page number not for citation purposes) Page 6 of 13 (page number not for citation purposes) http://www.biology-direct.com/content/2/1/20 Biology Direct 2007, 2:20 more likely to be permitted as well. The number of per- mitted amino acids is known to vary widely across amino acid sites [28,29], and different sets of amino acids are permitted at different sites [30,31]. Data on parallel evo- lution, preferably along many independent paths, can be used to further investigate such sets. Rate of evolution and frequency of the inferior allele as func- tions of 4Nes Figure 2 Rate of evolution and frequency of the inferior allele as functions of 4Nes. Rate of evolution v is in the units of rate of neutral evolution M [13]. Page 7 of 13 (page number not for citation purposes) http://www.biology-direct.com/content/2/1/20 http://www.biology-direct.com/content/2/1/20 http://www.biology-direct.com/content/2/1/20 Distributions of selection coefficients with extremal proper- ties Figure 3 Distributions of selection coefficients with extremal properties. Distributions p(s) that correspond to the mini- mal average selection coefficient (blue line) and to the maxi- mal fraction of selectively neutral sites (red lines), as long as the average rate of parallel evolution P constitutes 0.7 of the rate of neutral evolution. Vertical lines denote delta func- tions. Synonymous divergence was estimated from alignments of fourfold degenerate sites, flanked from each side by a nucleotide conserved between all four species. We defined synonymous divergence between two species for an unor- dered pair of nucleotides (A, B) as the ratio of the number of sites at which one of species carries A and the other B, over the total number of sites at which they both carry A, both carry B, or one carries A and the other B. Nonsynonymous divergence was estimated from nonde- generate nucleotide sites only, i. e. from sites at first or sec- ond nucleotide positions within an amino acid site at which each of the four nucleotides corresponds to a differ- ent encoded amino acid. In the analysis of divergence at the same nucleotide site, amino acid sites with divergence at more than one nucleotide site between any two of the four species were excluded from the analysis. In analysis of divergence at different nucleotide sites between paths I and II (i. e., species diverged along path I in first nucle- otide site, and along path II at the second nucleotide site of the amino acid site, or vice versa) we required that only a single nucleotide be divergent between amino acid sites for each two species. Nonsynonymous divergence for a pair of nucleotides (A, B) was defined analogously to syn- onymous divergence. Distributio ties Figure 3 p p g Distributions of selection coefficients with extremal properties. Distributions p(s) that correspond to the mini- mal average selection coefficient (blue line) and to the maxi- mal fraction of selectively neutral sites (red lines), as long as the average rate of parallel evolution P constitutes 0.7 of the rate of neutral evolution. Vertical lines denote delta func- tions. p p g Distributions of selection coefficients with extremal properties. Sets of permitted amino acids at a site Rates of coincident divergent evolution of proteins are ~3 times higher than their average rates of evolution, but still much lower than rates of parallel evolution (Tables 1, 2, 3). Indeed, rates of replacements, at an amino acid site, which involve different pairs of amino acids can be very different and must be analyzed separately [21]. Obvi- ously, not every amino acid is permitted even at an evolv- able amino acid site. The average selection coefficient > 1.5/(4Ne) associated with parallel replacements implies that the ratio of equi- librium frequencies of the two preferred amino acids at an amino acid site is ~4:1 (Fig. 2), assuming that the two amino acids involved in a parallel replacement usually confer the two highest fitnesses. Because for coincident divergent evolution P ~0.3 (Tables 1, 2, 3), the corre- sponding average selection coefficient should be > 2.5/ (4Ne) and the ratio of equilibrium frequencies of the pre- ferred over unpreferred amino acid is ~20:1 (Fig. 2). Thus, rather roughly, a typical evolvable amino acid site should be occupied by the favored amino acid, the second best amino acid, and the other possible amino acids with probabilities ~75%, ~15%, and ~10%. Of course, the sets of permitted amino acids vary greatly between sites. The rate of coincident divergence along path II is higher at amino acid sites where divergence along path I involves a pair of chemically dissimilar amino acids. Therefore, if two dissimilar amino acids are permitted at an amino acid site, other amino acids are Page 7 of 13 (page number not for citation purposes) Biology Direct 2007, 2:20 Page 8 of 13 (page number not for citation purposes) http://www.biology-direct.com/content/2/1/20 Distributions p(s) that correspond to the mini- mal average selection coefficient (blue line) and to the maxi- mal fraction of selectively neutral sites (red lines), as long as the average rate of parallel evolution P constitutes 0.7 of the rate of neutral evolution. Vertical lines denote delta func- tions. we changed the lengths of stretches of ambiguities so to minimize the number of internal stop codons in the cod- ing region. Chemical distance between a pair of amino acids was taken as the corresponding term from the Miyata matrix [41]. The distance rank of a single-nucleotide nonsynony- mous substitution between codons c1 and c2 took on val- ues (1, 1.5, 2, 2.5, 3) and was calculated as the mean of the two values: 1) rank of the amino acid distance d(c1, c2) among all distances d(c1, cn), where cn are all one-point non-stop neighboring codons of c1; 2) rank of the amino acid distance d(c1, c2) among all distances d(cm, c2), where cm are all one-point non-stop neighboring codons of c2. dS and dN were estimated using pairwise nucleotide align- ments taken from the four-species alignments for each pair of species using the codeml program of the PAML package [40]. To eliminate erroneous and nonortholo- gous gene alignments, those alignments in which pairwise dS and/or dN exceeded a pre-specified threshold between any pair of species were excluded from analysis. The thresholds were chosen manually to exclude the outlying alignments. The dS and dN values from PAML averaged over all the remaining genes were used as distances between species to produce the neighbor-joining phyloge- netic trees in Fig. 1. Genes were split into three bins of equal size (low, intermediate and high dN) according to dN value between species of path II. The total numbers of quadruplets of orthologous genes analyzed were 11,105 for mammals, 3,735 for Drosophila, and 3,040 for yeasts. All the alignments and the Perl scripts used for analysis are available upon request. Reviewers' comments Review 1 (John McDonald U For the fourfold neutral case, 0.80 of the sites will have A or C in the human-dog ancestor. The proportion of sites that will be AC in human-dog is the proportion of sites with A or C in the human-dog ancestor, times the proba- bility of a substitution to A or C in human or dog, times the probability that a substitution to G or T did not hap- pen in the other lineage, or 0.80(20.1-0.1^2)(1- 20.1) = 0.1216. Since the denominator is the proportion of sites that don't have G or T in either human or dog, it is the proportion of sites with A or C in the human-dog ancestor, times the probability that a substitution to G or T did not happen in either human or dog, or 0.80(1- 0.2)(1-0.2) = 0.512. Thus the proportion of AC sites in mouse-rat that show parallel AC differences in human- dog, divided by the proportion of sites with AA, AC, or CC in human-dog, is 0.1216/0.512 = 0.2375. For the same mutation rates, 0.2375 of the fourfold neutral sites show parallel evolution, but only 0.19 of the twofold neutral sites do. Consider the first line in table 1, which concerns sites that have A in mouse and C in rat, or vice versa. Of these AC sites in mouse-rat, 0.047 are also AC in human-dog. How- ever, this 0.047 is only the fraction of human-dog sites "where human and dog also underwent divergence in the same unordered pair of nucleotides." Thus the denomina- tor of the fraction yielding 0.047 is not the number of AC sites in mouse-rat, but the number of AC sites in mouse- rat, minus all sites with one or two non-AC bases in human-dog. Twofold neutral sites will all have A or C in human-dog, so the denominator will be larger (and thus the fraction of twofold neutral sites showing parallel evo- lution will be smaller). Author's response: There is a misunderstanding here. Table 1 describes the sites divergent between human and dog, as stated in the table heading. Contrary to what Dr. McDonald assumes, the first line in this table is concerned with sites that have A in dog and C in human, or vice versa. Reviewers' comments Review 1 (John McDonald U Reviewers' comments Review 1 (John McDonald, University of Delaware, Dept. of Biological Sciences, Newark, DE, USA; nominated by Laura Landweber, Dept. of Ecology and Evolutionary Biology, Princeton University, Princeton, NJ, USA) This manuscript compares three four-species sets of DNA sequences for coding regions. Each set of four species con- sists of two independent pairs. The main result is that there are fewer cases of parallel evolution for nonsynony- mous sites than expected, based on the proportion of fourfold synonymous sites with parallel evolution. This is offered as evidence of mildly deleterious selection. Reviewers' comments Review 1 (John McDonald, University of Delaware, Dept. of Biological Sciences, Newark, DE, USA; nominated by Laura Landweber, Dept. of Ecology and Evolutionary Biology, Princeton University, Princeton, NJ, USA) Reviewers' comments Review 1 (John McDonald, University of Delaware, Dept. of Biological Sciences, Newark, DE, USA; nominated by Laura Landweber, Dept. of Ecology and Evolutionary Biology, Princeton University, Princeton, NJ, USA) gy y J ) This manuscript compares three four-species sets of DNA sequences for coding regions. Each set of four species con- sists of two independent pairs. The main result is that there are fewer cases of parallel evolution for nonsynony- mous sites than expected, based on the proportion of fourfold synonymous sites with parallel evolution. This is offered as evidence of mildly deleterious selection. To eliminate erroneous regions of alignments, which may originate from errors in genome assemblies or annota- tions, we only analyzed those codons that were flanked by gapless alignments of length ten codons or more from each side. To avoid the effect of hypermutability of CpG dinucleotide in mammals, we only included in the analy- sis of mammalian genomes nonCpG-prone sites, not pre- ceded by C and not followed by G. I think it is about time that homoplasy was treated as a source of evidence about evolutionary processes, not just an annoyance for systematists. However, I think the defi- cit of parallel evolution at nonsynonymous sites reported here may be an artifact, and the pattern may actually be Page 8 of 13 (page number not for citation purposes) Page 8 of 13 (page number not for citation purposes) http://www.biology-direct.com/content/2/1/20 http://www.biology-direct.com/content/2/1/20 http://www.biology-direct.com/content/2/1/20 Biology Direct 2007, 2:20 explained by a "pure" neutral model, where the fitness of each mutation is either 0 or 1. explained by a "pure" neutral model, where the fitness of each mutation is either 0 or 1. Reviewers' comments Review 1 (John McDonald U The phrase in the footnote that parallel sites are those "where human and dog also under- went divergence in the same unordered pair of nucleotides" is simply a restatement of this. 0.047 is the fraction of the sites with A-C synonymous divergence between mouse and rat, among all sites with A-C synonymous divergence between human and dog and either A or C in the common ancestor of mouse and rat. As described in Methods, the denominator of the fraction that yields 0.047 is the number of sites which have A and C, or C and A, in human and dog AND have A and A, A and C, C and A, or C and C in mouse and rat. We attempted to clarify this in the revised version of the manuscript. Author's response: We do not see a problem here. In terms of Dr. McDonald's example, we did the following: 1) chose sites that experienced A-C divergence along path I (i. e., A in dog and C in human or vice versa), 1) chose sites that experienced A-C divergence along path I (i. e., A in dog and C in human or vice versa), 2) among such sites, chose those sites that had either A or C in the common ancestor for path II (assuming parsymony), 3) for these sites, calculated the rate of A-C divergence along path II. Assuming that only 1 substitution can occur alone path II at a site, we do not see how this analysis can be affected by the total number of permitted nucleotides at the site. As a first approximation, we can assume that fourfold syn- onymous sites are fourfold neutral. For nonsynonymous sites, however, it seems likely that many sites with AC in mouse-rat will be only twofold neutral; the amino acids resulting from A or C at that position may have equal fit- nesses, but the G or T amino acids may be strongly selected against. If that is the case, a smaller proportion of the twofold AC nonsynonymous neutral sites will show parallel evolution than fourfold AC synonymous sites, if the denominator excludes non-AC sites in human-dog. I suspect that this artifact may not be large enough to explain all of the results in this manuscript, but the authors need to consider it. Review 2 (Sarah Teichmann and Subhajyoti De, MRC Laboratory of Molecular Biology, Cambridge, UK) The manuscript "Extensive parallelism in protein evolu- tion" by Bazykin et al. provides an interesting insight into the parallel evolution at orthologus sites, and its effects on the observed rate of protein evolution. The work discusses how parallel accumulation of identical changes at equiva- lent sites contributes towards the effective rate of protein divergence, both under constant and variable fitness land- scape. The authors conclude that the data favours a con- stant fitness landscape on the proteins they investigate. 3. Along the same lines, are there examples of parallel evo- lution of sites that are very chemically different, and do they coincide with high dN/dS in a protein, indicating positive selection? Is there a general correlation of high dN/dS in a protein and chemical dissimilarity at coinci- dent parallel sites? One could investigate this by consider- ing individual proteins or groups of proteins rather than the entire data set. The idea and results are an interesting molecular evolu- tionary study. However, we have some technical and con- ceptual comments, which need to be addressed to improve the manuscript. Conceptual comments 1. The authors state that non-neutrality of synonymous sites and heterogeneity of mutation rates can influence the results on parallel evolution, but did not attempt to deter- mine their relative importance. Can the authors provide a rough estimate of how much those factors can affect the final conclusions? Author's response: This is indeed an interesting question. It appears from our data that proteins with LOWER dN/dS values on average allow for more radical parallel substitutions. How- ever, there is insufficient data for this analysis to be included in the main text. 4. The authors mention that the number of slightly dele- terious and slightly beneficial replacements must be equal. When selection is non-zero at mutation-selection- drift equilibrium, it is not clear why this should be the case. Author's response: Because we use the rate of only parallel syn- onymous evolution as a proxy for mutation rate, heterogeneity of the rate of evolution across synonymous sites, due to hetero- geneous mutation and/or selection, should not be a problem. However, uniform deviation of the rate of synonymous evolu- tion from the mutation rate, caused by selection, cannot be taken into account in this way. Reviewers' comments Review 1 (John McDonald U If the lower than expected pro- portion of parallel evolution for nonsynonymous sites holds up, this manuscript will be additional evidence that mildly deleterious evolution of nonsynonymous muta- tions is widespread. One minor point – since the main result of this manu- script is that there is less parallelism in protein evolution than expected, the title "Extensive parallelism in protein evolution" is misleading; something like "Low parallelism in protein evolution as support for the nearly neutral model" would be better. Here's a numerical example to illustrate this. Consider a site with A in mouse and C in rat. The substitution proba- bility is 0.1 to any other base between the mouse-rat ancestor and the human-dog ancestor (this is an unrooted tree), and 0.1 between the human-dog ancestor and either human or dog. If the site is twofold neutral, 0.19 of these sites will be AC in human-dog (2*0.1-0.1^2), ignoring multiple hits in a single lineage. So the proportion of two- fold neutral AC sites in mouse-rat showing parallel evolu- tion would be 0.19. Author's response: Indeed, our rate of parallel nonsynonymous evolution is ~30% below that of synonymous evolution, but it is also ~6 times higher than the average rate of nonsynonymous Page 9 of 13 (page number not for citation purposes) http://www.biology-direct.com/content/2/1/20 Biology Direct 2007, 2:20 http://www.biology-direct.com/content/2/1/20 evolution. Thus, if one word is used to characterize this rate, "extensive" sounds appropriate. evolution. Thus, if one word is used to characterize this rate, "extensive" sounds appropriate. Author's response: We think that the explanation may be sim- pler. If, at a site, all the possible amino acids confer different fitnesses, substitutions that involve the best and the next best amino acid would be more common in evolution than any other, thus leading to the rate of parallel evolution being higher than the rate of coincident divergent evolution. The two "most preferred" amino acids do not necessarily represent the most similar pair at a site, although, on average, their Miyata dis- tance tends to be short. As to the last point, very similar results where obtained using Grantham's similarity matrix (data not shown). Review 2 (Sarah Teichmann and Subhajyoti De, MRC Laboratory of Molecular Biology, Cambridge, UK) Because we excluded hypermut- able CpG sites (in mammals), it is likely that weak selection at synonymous sites could only decrease their rate of evolution, perhaps by ~10% (Kondrashov et al. J. Theor. Biol. 240, 616– 626, 2006). If so, using the rate of synonymous evolution as a proxy for the rate of neutral evolution should cause a slight underestimation of the strength of negative selection at evolva- ble non-synonymous sites. Author's response: Assuming constant selection, the rates of reciprocal beneficial and deleterious replacements at mutation- selection-drift equilibrium must be the same, as long as we con- sider only 2 alleles – essentially, the definition of equilibrium. Less trivially, this fact holds even when we consider an arbitrary number of alleles (still under constant selection), because mutation-selection-drift equilibrium is a detailed equilibrium, with identical reciprocal fluxes of substitutions, and no cycling possible. 5. Providing examples of parallel evolution of sites in a protein in the context of constant and variable fitness landscapes would be very helpful. 2. The authors mentioned that coincident divergent evo- lution is always much lower than the rate of parallel non- synonymous evolution. Presumably this is linked to the fact that evolvable sites will tend to switch to the most chemically similar amino acids. This is indicated by the Miyata distances in the tables, and the discussion section on permitted amino acids. However, the meaning of the percentages in the tables on the Miyata distances are not obvious to us, and are not explained in the table legend. To what extent is the ratio of parallel to coincident diver- gent sites explained by chemical similarity? What are the results using different chemical similarity measures? Author's response: We would love to do this – but how can we be sure that fitness landscape was constant – or variable – for a particular protein? Technical comments 1. Although the biological aspects of equation (1) were clear, the mathematical parts could be elucidated more clearly: dv = d(f(s)) = f'(s)d(s), yet f'(s) is not present. Page 10 of 13 (page number not for citation purposes) Page 10 of 13 (page number not for citation purposes) Biology Direct 2007, 2:20 http://www.biology-direct.com/content/2/1/20 and simply assume that the authors have assembled and analyzed the data with all due diligence. 2. The authors talk about the conditions where P > 1 or P < 1 under variable and constant fitness landscape, and mention that the distribution q(v) is concentrated at P = 0.7. However it was unclear how the value is obtained. Is it simply derived from their results that P is between 50 and 80%? And what is the expected range of P? The analysis of the rates of parallel evolution turns out to be rather tricky. The authors are mostly interested in deter- mining the average selection coefficient at these sites. Clearly, if all sites were just evolving neutrally, then the rate of synonymous parallel evolution should be equal to the rate of nonsynonymous parallel evolution, i.e., P = 1. The question then is: how do you explain the reduced rate? Is the rate reduced because the two lineages are see- ing different fitness landscapes (so that one residue is pre- ferred in one lineage but selected against in the other), or is it because many sites actually have somewhat non-opti- mal residues to begin with, which are replaced by optimal ones in one, but not the other lineage? 3. The clarity of the manuscript needs to be improved to make it readily comprehensible to the broad audience, including aspects of English grammar. Subsections within in the 'Methods' section would improve the organisation of the manuscript. 4. Figure legends need to be more elaborate to improve clarity. Figure 3 can be omitted and instead described in a table or in the text. Instead, a figure of a protein with par- allel mutations can be provided as a concrete biological example of the phenomenon. If you assume that a variable landscape explains the sup- pression (i.e., P < 1), then you must assume that the change happened in the landscape of only one of the lin- eages, because if it happened in parallel for both, you'd see P > 1. Author's response: So far, we agree with Dr. Adami. If landscapes are not changing, then the authors argue hat P < 1 implies slightly deleterious mutations at parallel sites. It is here where I have serious reservations. The rea- soning is based on assuming that the average rate of par- allel evolution reflects what is going on at single sites, that is, that the probability distribution of deleterious effects is narrowly distributed around the mean. But the authors themselves acknowledge that this is not the case (often a Gamma distribution is assumed), but then go on to say that the average effect of mutations at sites that undergo parallel evolution can nevertheless be calculated from knowing P. However, this reasoning also assumes that the distribution of P's across sites is narrowly centered around an average P, which again assumes a peaked distribution of deleterious effects. In other words, the conclusion that P < 1 implies a small mean effect of deleterious mutations is obtained by replacing distributions by their means, something that is exceedingly dangerous for distributions with an exponential tail. Review 3 (Chris Adami, Keck Graduate Institute, California Institute of Technology, Pasadena, USA) Review 3 (Chris Adami, Keck Graduate Institute, California Institute of Technology, Pasadena, USA) Technical comments If only one of the lineages sees a change in land- scape, is the change neutral or driven by positive selec- tion? The authors argue that it cannot be neutral because in this case slowly evolving proteins (for whom the land- scape of "other mutations" changes only very slowly) should have a higher P, whereas a lower P is observed. Thus, the authors conclude that positive selection must be responsible for P < 1 if landscapes are changing. 5. How many parallel sites are there per protein on aver- age in the three groups of four organisms? We cannot deduce this from any of the numbers in the manuscript. Author's response: We are thankful for these comments, some of which have been accomodated. The modified manuscript will definitely provide good insight into the parallel evolution leading to similar changes at identical sites and the evidence for weak nega- tive selection on these sites in proteins. Once the above issues are carefully addressed, the improved manuscript is expected to meet the standard for publication in the Biol- ogy Direct. Author's response: So far, we agree with Dr. Adami. Page 11 of 13 (page number not for citation purposes) References Nature 2004, 428:493-521. It is also surprising that the authors do not discuss the evi- dence for parallel evolution accumulated in experimental evolution, where extensive parallelism is explained by overwhelmingly positive selection (Wichman et al, Sci- ence 285:422–424, 1999). J y q y son J, Zhao S, Fraser CM, Shetty J, Shatsman S, Geer K, Chen Y, Ab S Ni WC H l k PH Ch R D bi KJ E A g y y y AJ, D'Souza LM, Martin K, Wu JQ, Gonzalez-Garay ML, Jackson AR, Kalafus KJ, McLeod MP, Milosavljevic A, Virk D, Volkov A, Wheeler Author's response: In the first paragraph of the paper, we cite several plausible instances of parallel evolution that is driven by positive selection. Generally, we agree with Dr. Adami that our arguments against variable fitness landscapes being the main reason for P < 1 are not water-tight. Still, weak, constant selec- tion appears to provide the most plausible explanation for the data. Review 3 (Chris Adami, Keck Graduate Institute, California Institute of Technology, Pasadena, USA) In this paper, the authors study the parallel (and diver- gent) replacement of residues of proteins in different evo- lutionary paths, obtained by aligning sequences of three quartets of organism. The main observations are the fol- lowing: the rate of parallel (non-synonymous) replace- ments is between 50% and 80% of the rate of parallel synonymous replacement, while the rate of divergent replacement (a residue replaced by another in one line, and replaced by a different one in the other line) is about between 25% and 30%. The authors conclude from this observation that most of the sites that undergo parallel evolution in these organisms (yeast, Drosophila, and mam- mal quartets) are under weak selection, as suggested by Ohta's "nearly neutral" theory. There are clearly two parts to this paper: one is the assem- bly of data and their analysis to determine the rates of par- allel evolution, the other is the interpretation of the data. I will limit my comments to the interpretation of the rates, Author's response: Here, we cannot agree: we do not assume that the distribution of rate of evolution across sites, q(v), is narrow. We only claim that the average selection coefficient S Page 11 of 13 (page number not for citation purposes) Page 11 of 13 (page number not for citation purposes) http://www.biology-direct.com/content/2/1/20 http://www.biology-direct.com/content/2/1/20 Biology Direct 2007, 2:20 is minimal when this distribution is narrow, so that any non- zero variance of q(v) leads to S > f-1(P) (see text). Moreover, we calculate the maximal fraction of selectively neutral sites that is consistent with a given P, under an arbitrary q(v). Authors' contributions y p g Brudno M, Sidow A, Stone EA, Venter JC, Payseur BA, Bourque G, GAB designed the study and performed sequence analy- ses. FAK created mammalian and yeast sequence align- ments. MB, AP and ID created Drosophila sequence alignments. ASK conceived and designed the study, per- formed mathematical analysis, and wrote the manuscript. All authors participated in writing, read and approved the final manuscript. 11. 11. Kellis M, Patterson N, Endrizzi M, Birren B, Lander ES: Sequencing and comparison of yeast species to identify genes and regu- latory elements. Nature 2003, 423:241-54. References 1. Futuyma DJ: Evolution Sunderland: Sinauer Associates; 2005. 2. Li WH: Molecular Evolution Sunderland: Sinauer Associates; 1997. 2. Li WH: Molecular Evolution Sunderland: Sinauer Associates; 1997. 3. Delsuc F, Brinkmann H, Philippe H: Phylogenomics and the reconstruction of the tree of life. Nat Rev Genet 2005, 6:361-75. 4. Zhang J: Parallel adaptive origins of digestive RNases in Asian and African leaf monkeys. Nat Genet 2006, 38:819-23. 3. Delsuc F, Brinkmann H, Philippe H: Phylogenomics and the reconstruction of the tree of life. Nat Rev Genet 2005, 6:361-75. 4. Zhang J: Parallel adaptive origins of digestive RNases in Asian and African leaf monkeys. Nat Genet 2006, 38:819-23. While I cannot argue that the authors conclusion is neces- sarily wrong, it seems to me that in the absence of detailed knowledge about the distribution of deleterious effects, no conclusion can be drawn about how neutral or nearly neutral sites are that undergo parallel evolution. Indeed, it seems to me that positive selection is not out of the ques- tion, as a single change at one site can influence whether a change at another site is beneficial or deleterious (epista- sis between mutations). Thus, rather than invoking envi- ronments being constant or changing for both lines, it is sufficient for one lineage to acquire a second mutation somewhere else in the protein in order for the change that is beneficial in the first lineage to be detrimental in the second. All else being equal, the ubiquity of epistatic inter- actions between mutations should favor an epistatic explanation of reduced parallelism. y 5. Wolf YI, Viboud C, Holmes EC, Koonin EV, Lipman DJ: Long inter- vals of stasis punctuated by bursts of positive selection in the seasonal evolution of influenza A virus. Biol Direct 2006, 1:34. 6. Kimura M: The Neutral Theory of Molecular Evolution Cambridge: Cam- bridge University Press; 1983. g y 7. Kimura M: Evolutionary rate at the molecular level. Nature 1968, 217:624-6. 8. Murphy WJ, Pevzner PA, O'Brien SJ: Mammalian phylogenomics comes of age. Trends Genet 2004, 20:631-639. 9. Galagan JE, Henn MR, Ma LJ, Cuomo CA, Birren B: Genomics of the fungal kingdom: insights into eukaryotic biology. Genome Res 2005, 15:1620-1631. 2005, 15:1620 1631. 10. References Gibbs RA, Weinstock GM, Metzker ML, Muzny DM, Sodergren EJ, Scherer S, Scott G, Steffen D, Worley KC, Burch PE, Okwuonu G, Hines S, Lewis L, DeRamo C, Delgado O, Dugan-Rocha S, Miner G, Morgan M, Hawes A, Gill R, Celera , Holt RA, Adams MD, Amanati- des PG, Baden-Tillson H, Barnstead M, Chin S, Evans CA, Ferriera S, Fosler C, Glodek A, Gu Z, Jennings D, Kraft CL, Nguyen T, Pfannkoch CM, Sitter C, Sutton GG, Venter JC, Woodage T, Smith D, Lee HM, Gustafson E, Cahill P, Kana A, Doucette-Stamm L, Weinstock K, Fechtel K, Weiss RB, Dunn DM, Green ED, Blakesley RW, Bouffard GG, De Jong PJ, Osoegawa K, Zhu B, Marra M, Schein J, Bosdet I, Fjell C, Jones S, Krzywinski M, Mathewson C, Siddiqui A, Wye N, McPher- son J, Zhao S, Fraser CM, Shetty J, Shatsman S, Geer K, Chen Y, Abramzon S, Nierman WC, Havlak PH, Chen R, Durbin KJ, Egan A, Ren Y, Song XZ, Li B, Liu Y, Qin X, Cawley S, Worley KC, Cooney AJ, D'Souza LM, Martin K, Wu JQ, Gonzalez-Garay ML, Jackson AR, Kalafus KJ, McLeod MP, Milosavljevic A, Virk D, Volkov A, Wheeler DA, Zhang Z, Bailey JA, Eichler EE, Tuzun E, Birney E, Mongin E, Ureta-Vidal A, Woodwark C, Zdobnov E, Bork P, Suyama M, Tor- rents D, Alexandersson M, Trask BJ, Young JM, Huang H, Wang H, Xing H, Daniels S, Gietzen D, Schmidt J, Stevens K, Vitt U, Wingrove J, Camara F, Mar Alba M, Abril JF, Guigo R, Smit A, Dubchak I, Rubin EM, Couronne O, Poliakov A, Hubner N, Ganten D, Goesele C, Hummel O, Kreitler T, Lee YA, Monti J, Schulz H, Zimdahl H, Him- melbauer H, Lehrach H, Jacob HJ, Bromberg S, Gullings-Handley J, Jensen-Seaman MI, Kwitek AE, Lazar J, Pasko D, Tonellato PJ, Twigger S, Ponting CP, Duarte JM, Rice S, Goodstadt L, Beatson SA, Emes RD, Winter EE, Webber C, Brandt P, Nyakatura G, Adetobi M, Chiarom- onte F, Elnitski L, Eswara P, Hardison RC, Hou M, Kolbe D, Makova K, Miller W, Nekrutenko A, Riemer C, Schwartz S, Taylor J, Yang S, Zhang Y, Lindpaintner K, Andrews TD, Caccamo M, Clamp M, Clarke L, Curwen V, Durbin R, Eyras E, Searle SM, Cooper GM, Batzoglou S, Brudno M, Sidow A, Stone EA, Venter JC, Payseur BA, Bourque G, Lopez-Otin C, Puente XS, Chakrabarti K, Chatterji S, Dewey C, Pachter L, Bray N, Yap VB, Caspi A, Tesler G, Pevzner PA, Haussler D, Roskin KM, Baertsch R, Clawson H, Furey TS, Hinrichs AS, Karol- chik D, Kent WJ, Rosenbloom KR, Trumbower H, Weirauch M, Cooper DN, Stenson PD, Ma B, Brent M, Arumugam M, Shteynberg D, Copley RR, Taylor MS, Riethman H, Mudunuri U, Peterson J, Guyer M, Felsenfeld A, Old S, Mockrin S, Collins F: Genome sequence of the Brown Norway rat yields insights into mam- malian evolution. References Nature 2004, 428:493-521. 11. Kellis M, Patterson N, Endrizzi M, Birren B, Lander ES: Sequencing and comparison of yeast species to identify genes and regu- l l N 2003 423 241 54 , 10. References Gibbs RA, Weinstock GM, Metzker ML, Muzny DM, Sodergren EJ, Scherer S, Scott G, Steffen D, Worley KC, Burch PE, Okwuonu G, Hines S, Lewis L, DeRamo C, Delgado O, Dugan-Rocha S, Miner G, Morgan M, Hawes A, Gill R, Celera , Holt RA, Adams MD, Amanati- des PG, Baden-Tillson H, Barnstead M, Chin S, Evans CA, Ferriera S, Fosler C, Glodek A, Gu Z, Jennings D, Kraft CL, Nguyen T, Pfannkoch CM, Sitter C, Sutton GG, Venter JC, Woodage T, Smith D, Lee HM, Gustafson E, Cahill P, Kana A, Doucette-Stamm L, Weinstock K, Fechtel K, Weiss RB, Dunn DM, Green ED, Blakesley RW, Bouffard GG, De Jong PJ, Osoegawa K, Zhu B, Marra M, Schein J, Bosdet I, Fjell C, Jones S, Krzywinski M, Mathewson C, Siddiqui A, Wye N, McPher- son J, Zhao S, Fraser CM, Shetty J, Shatsman S, Geer K, Chen Y, Abramzon S, Nierman WC, Havlak PH, Chen R, Durbin KJ, Egan A, Ren Y, Song XZ, Li B, Liu Y, Qin X, Cawley S, Worley KC, Cooney AJ, D'Souza LM, Martin K, Wu JQ, Gonzalez-Garay ML, Jackson AR, Kalafus KJ, McLeod MP, Milosavljevic A, Virk D, Volkov A, Wheeler DA, Zhang Z, Bailey JA, Eichler EE, Tuzun E, Birney E, Mongin E, Ureta-Vidal A, Woodwark C, Zdobnov E, Bork P, Suyama M, Tor- rents D, Alexandersson M, Trask BJ, Young JM, Huang H, Wang H, Xing H, Daniels S, Gietzen D, Schmidt J, Stevens K, Vitt U, Wingrove J, Camara F, Mar Alba M, Abril JF, Guigo R, Smit A, Dubchak I, Rubin EM, Couronne O, Poliakov A, Hubner N, Ganten D, Goesele C, Hummel O, Kreitler T, Lee YA, Monti J, Schulz H, Zimdahl H, Him- melbauer H, Lehrach H, Jacob HJ, Bromberg S, Gullings-Handley J, Jensen-Seaman MI, Kwitek AE, Lazar J, Pasko D, Tonellato PJ, Twigger S, Ponting CP, Duarte JM, Rice S, Goodstadt L, Beatson SA, Emes RD, Winter EE, Webber C, Brandt P, Nyakatura G, Adetobi M, Chiarom- onte F, Elnitski L, Eswara P, Hardison RC, Hou M, Kolbe D, Makova K, Miller W, Nekrutenko A, Riemer C, Schwartz S, Taylor J, Yang S, Zhang Y, Lindpaintner K, Andrews TD, Caccamo M, Clamp M, Clarke L, Curwen V, Durbin R, Eyras E, Searle SM, Cooper GM, Batzoglou S, Brudno M, Sidow A, Stone EA, Venter JC, Payseur BA, Bourque G, Lopez-Otin C, Puente XS, Chakrabarti K, Chatterji S, Dewey C, Pachter L, Bray N, Yap VB, Caspi A, Tesler G, Pevzner PA, Haussler D, Roskin KM, Baertsch R, Clawson H, Furey TS, Hinrichs AS, Karol- chik D, Kent WJ, Rosenbloom KR, Trumbower H, Weirauch M, Cooper DN, Stenson PD, Ma B, Brent M, Arumugam M, Shteynberg D, Copley RR, Taylor MS, Riethman H, Mudunuri U, Peterson J, Guyer M, Felsenfeld A, Old S, Mockrin S, Collins F: Genome sequence of the Brown Norway rat yields insights into mam- malian evolution. Acknowledgements Comput Appl Biosci 1997, 13:555-6. 40. Yang Z: PAML: a program package for phylogenetic analysis by maximum likelihood. Comput Appl Biosci 1997, 13:555-6. 41. Miyata T, Miyazawa S, Yasunaga T: Two types of amino acid sub- stitutions in protein evolution. J Mol Evol 1979, 12:219-36. y p pp 41. Miyata T, Miyazawa S, Yasunaga T: Two types of amino acid sub- stitutions in protein evolution. J Mol Evol 1979, 12:219-36. p 18. Yampolsky LY, Kondrashov FA, Kondrashov AS: Distribution of the strength of selection against amino acid replacements in human proteins. Hum Mol Genet 2005, 14:3191-201. p 19. Eyre-Walker A, Woolfit M, Phelps T: The distribution of fitness effects of new deleterious amino acid mutations in humans. Genetics 2006, 173:891-900. 20. Ota T, Nei M: Estimation of the number of amino acid substi- tutions per site when the substitution rate varies among sites. J Mol Evol 1994, 38:642-643. 21. Chen L, Perlina A, Lee CJ: Positive selection detection in 40,000 human immunodeficiency virus (HIV) type 1 sequences automatically identifies drug resistance and positive fitness mutations in HIV protease and reverse transcriptase. J Virol 2004, 78:3722-32. 22. Subramanian S, Kumar S: Higher intensity of purifying selection on > 90% of the human genes revealed by the intrinsic replacement mutation rates. Mol Biol Evol 2006, 23:2283-7. 23. Fitch WM, Markowitz E: An improved method for determining codon variability in a gene and its application to the rate of fixation of mutations in evolution. Biochem Genet 1970, 4:579-93. 24. Bazykin GA, Dushoff J, Levin SA, Kondrashov AS: Bursts of nonsyn- onymous substitutions in HIV-1 evolution reveal instances of positive selection at conservative protein sites. Proc Natl Acad Sci USA 2006, 103:19396-401. 25. Ohta T: Slightly deleterious mutant substitutions in evolu- tion. Nature 1973, 246:96-8. 26. Ohta T: The nearly neutral theory of molecular evolution. Annu Rev Ecol Syst 1992, 23:263-286. 27. Kondrashov AS: Contamination of the genome by very slightly deleterious mutations: why have we not died 100 times over? J Theor Biol 1995, 175:583-94. J 28. Sjolander K, Karplus K, Brown M, Hughey R, Krogh A, Mian IS, Haus- sler D: Dirichlet mixtures: a method for improved detection of weak but significant protein sequence homology. Comput Appl Biosci 1996, 12:327-45. pp 29. Lartillot N, Philippe H: A Bayesian mixture model for across- site heterogeneities in the amino-acid replacement process. Mol Biol Evol 2004, 21:1095-109. 30. Acknowledgements 12. 12. Bergman CM, Pfeiffer BD, Rincon-Limas DE, Hoskins RA, Gnirke A, Mungall CJ, Wang AM, Kronmiller B, Pacleb J, Park S, Stapleton M, Wan K, George RA, de Jong PJ, Botas J, Rubin GM, Celniker SE: Assessing the impact of comparative genomic sequence data on the functional annotation of the Drosophila genome. Genome Biol 2002, 3:RESEARCH0086. g G.A.B. gratefully acknowledges fellowships from the Pew Charitable Trusts award 2000-002558 and the Burroughs Wellcome Fund award 1001782, both to Princeton University. F.A.K. is a National Science Foundation Grad- uate Fellow. M.B.'s work is partially supported by the NSERC Discovery grant. I.D. and A.P. were partially supported by grant HL066681 (L.A.P., I.D. and S.M.), Berkeley-PGA, under the Programs for Genomic Applications, funded by National Heart, Lung, & Blood Institute and Department of Energy Contract DE-AC02-05CH11231, University of California. This work was partially supported through the Molecular and Cellular Biology Program of the Russian Academy of Sciences. 13. Bulmer M: The selection-mutation-drift theory of synony- mous codon usage. Genetics 1991, 129:897-907. g 14. Kimura M: Model of effectively neutral mutations in which selective constraint is incorporated. Proc Natl Acad Sci USA 1979, 76:3440-3444. 15. Hwang DG, Green P: Bayesian Markov chain Monte Carlo sequence analysis reveals varying neutral substitution pat- 15. Hwang DG, Green P: Bayesian Markov chain Monte Carlo sequence analysis reveals varying neutral substitution pat- Page 12 of 13 (page number not for citation purposes) Biology Direct 2007, 2:20 http://www.biology-direct.com/content/2/1/20 terns in mammalian evolution. Proc Natl Acad Sci USA 2004, 101:13994-4001. 38. Sundararajan M, Brudno M, Small K, Sidow A, Batzoglow S: Chaining algorithms for alignment of draft sequence. Proceedings of the fourth Workshop on Algorithms in Bioinformatics (WABI 2004) . terns in mammalian evolution. Proc Natl Acad Sci USA 2004, 101:13994-4001. 16. Kondrashov FA, Ogurtsov AY, Kondrashov AS: Selection in favor of nucleotides G and C diversifies evolution rates and levels of polymorphism at mammalian synonymous sites. J Theor Biol 2006, 240:616-26. p g ( ) 39. Brudno M, Do CB, Cooper GM, Kim MF, Davydov E, Green ED, Sidow A, Batzoglou S: LAGAN and Multi-LAGAN: efficient tools for large-scale multiple alignment of genomic DNA. Genome Res 2003, 13:721-31. 17. Grishin NV, Wolf YI, Koonin EV: From complete genomes to measures of substitution rate variability within and between proteins. Genome Res 2000, 10:991-1000. 40. Yang Z: PAML: a program package for phylogenetic analysis by maximum likelihood. http://www.biology-direct.com/content/2/1/20 Acknowledgements Shenkin PS, Erman B, Mastrandrea LD: Information-theoretical entropy as a measure of sequence variability. Proteins 1991, 11:297-313. 31. Pei J, Grishin NV: AL2CO: calculation of positional conserva- tion in a protein sequence alignment. Bioinformatics 2001, 17:700-12. 32. Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, Lip- man DJ: Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 1997, 25:3389-402. 33. Tatusov RL, Koonin EV, Lipman DJ: A genomic perspective on protein families. Science 1997, 278:631-7. Publish with BioMed Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp BioMedcentral Publish with BioMed Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp BioMedcentral Publish with BioMed Central and every scientist can read your work free of charge 34. Wheeler DL, Church DM, Edgar R, Federhen S, Helmberg W, Mad- den TL, Pontius JU, Schuler GD, Schriml LM, Sequeira E, Suzek TO, Tatusova TA, Wagner L: Database resources of the National Center for Biotechnology Information: update. Nucleic Acids Res 2004, 32:D35-40. 35. Benson DA, Karsch-Mizrachi I, Lipman DJ, Ostell J, Wheeler DL: GenBank. Nucleic Acids Res 2006, 34:D16-20. 36. Thompson JD, Higgins DG, Gibson TJ: CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 1994, 22:4673-80. g 37. Brudno M, Malde S, Poliakov A, Do CB, Couronne O, Dubchak I, Bat- zoglou S: Glocal alignment: finding rearrangements during alignment. Bioinformatics 2003, 19(Suppl 1):i54-62. Page 13 of 13 (page number not for citation purposes)
https://openalex.org/W2432774594	https://link.springer.com/content/pdf/10.1007%2Fs40097-016-0198-x.pdf	English	null	Antioxidant, antimicrobial and cytotoxic activities of silver and gold nanoparticles synthesized using Plumbago zeylanica bark	Journal of nanostructure in chemistry	2,016	cc-by	7,626	Antioxidant, antimicrobial and cytotoxic activities of silver and gold nanoparticles synthesized using Plumbago zeylanica bark S. Priya Velammal1 • T. Akkini Devi1 • T. Peter Amaladhas1 Received: 13 April 2016 / Accepted: 1 June 2016 / Published online: 13 June 2016 The Author(s) 2016. This article is published with open access at Springerlink.com Abstract Utilization of bioresources for the synthesis of metal nanoparticles is the latest ﬁeld in green chemistry. The present work reports the utilization of the aqueous bark extract of Plumbago zeylanica for the biosynthesis of Ag and Au NPs. The Ag and Au NPs thus obtained were characterized by UV–Vis, FT-IR, TEM, XRD, and EDAX analysis. The water-soluble components of the extract were responsible for the reduction of Ag? and Au3? ions. FT-IR spectra revealed that the –OH and[C=O groups present in the biomolecules were responsible for reduction and stabi- lization of nanoparticles. TEM images showed the existence of spherical Ag and Au NPs with average size of 28.47 and 16.89 nm, respectively, which was further substantiated by XRD analysis. The presence of elemental Ag and Au along with C and O from the attached biomolecules was proved by EDAX analysis. The antimicrobial, antioxidant, and in vitro cytotoxic activities of the synthesized nanoparticles were studied by disc diffusion, DPPH, and MTT assay methods, respectively. The free radical inhibition was found to be 78.17 and 87.34 % for Ag and Au nanoparticles, respec- tively. The Ag and Au NPs showed 61.56 and 65.61 % toxicity against DLA cell line, respectively. The DNA binding ability of Ag and Au NPs were investigated using CT-DNA. The hyperchromism shift inferred the groove binding of nanoparticles with CT-DNA. toxicity against DLA cell line, respectively. The DNA binding ability of Ag and Au NPs were investigated using CT-DNA. The hyperchromism shift inferred the groove binding of nanoparticles with CT-DNA. Keywords Biosynthesis Silver nanoparticles Gold nanoparticles Antioxidant Cytotoxicity DLA cell line J Nanostruct Chem (2016) 6:247–260 DOI 10.1007/s40097-016-0198-x J Nanostruct Chem (2016) 6:247–260 DOI 10.1007/s40097-016-0198-x ORIGINAL RESEARCH 1 PG and Research Department of Chemistry, V.O. Chidambaram College, Tuticorin 628008, Tamil Nadu, India Introduction Nanoscience and nanotechnology, the two current ﬁelds of research have strong and beautiful footprints in ancient times. Nano mercury and its sulﬁde of sizes ﬁner than 10 nm have been used as medicines in India since long back [1]. Silver and copper nanoparticles have been used to produce iridescent metallic effects on ancient ceramic objects [2]. The ancient Vedic rite of Agnihotra has used the nanoparticles to cleanse the environment. In Agnihotra, wood of trees like Butea frondosa (palasa), Mangifera indica (mango), Ficus religiosa (pipal), medicinal herbs, ghee, cloves, car- damom, and camphor are offered to the ﬁre god by throwing them into the ﬁre. This has been reported to ﬁll the atmosphere with nutrients in nano size and medicinal properties, which in turn eliminate pathogenic bacteria or germs [3]. In Ayurvedha, herbo-mineral, and herbo-metallic drugs have been prepared by prolonged heating which signiﬁcantly reduces the particle size. The medicinal use of colloidal gold (Swarnabhasma) and mercury remain popular till this day and the Swarnabhasmas (gold ash) has been characterized as globular particles of gold with the average size between 56 and 57 nm. It has been used to treat arthiritis, asthma, diabetes, and diseases of nervous system [4]. Electronic supplementary material The online version of this article (doi:10.1007/s40097-016-0198-x) contains supplementary material, which is available to authorized users. & T. Peter Amaladhas peteramaldhast@yahoo.co.in; peteramaladhas@gmail.com S. Priya Velammal subraja1950@gmail.com T. Akkini Devi t.deviagni@gmail.com 1 PG and Research Department of Chemistry, V.O. Chidambaram College, Tuticorin 628008, Tamil Nadu, India & T. Peter Amaladhas peteramaldhast@yahoo.co.in; peteramaladhas@gmail.com 12 3 3 248 J Nanostruct Chem (2016) 6:247–260 Metal and metal oxide nanoparticles have been synthe- sized using many methods in the past, such as chemical reduction [5], electrochemical reaction [6], microwave method [7], reverse micelles [8], sonochemical method [9], and biosynthesis. Biosynthesis of nanoparticles has received great attention recently as the synthesized nanoparticles are non-toxic and can be used for biomedical applications. Different fungi like Fusarium oxysporum [10], Cladosporium cladosporioides [11], Fusarium semi- tactum [12], plants like Syzygium aromaticum [13], Med- icago sativa [14], Azadirachta indica [15], Terminalia cuneata [16], and Trignellafoenum graecum seeds [17] have been employed to synthesize metal nanoparticles. Recently, sunlight-induced rapid synthesis of silver nanoparticles has been reported by our group [18, 19]. Characterization of Ag and Au NPs UV–Vis spectral analysis and DNA binding studies were performed on a JASCO, V-650 spectrophotometer. FT-IR spectra for the dry powder of bark extract and nanoparticles were recorded in the range 4000–400 cm-1 with Thermo scientiﬁc, Nicolet iS5 spectrometer. The size of the nanoparticles was assessed using PHILIPS, CM 200, TEM microscope operated at 200 kV with a resolution of 2.4 A˚ . The diffraction pattern of nanoparticles was recorded using a PANalytical X’Pert Powder X’Celerator diffractometer, with CuKa monochromatic ﬁlter. To ascertain the ele- mental composition of Ag and Au NPs, energy dispersive X-ray (EDAX) analysis was performed on a JEOL, Model JED-2300 microscope. Ultrasonic probe sonicator, Ener- tech make, model: ENUP-500A was used to disperse CT- DNA in Tris–HCl buffer solution. Biosynthesis of Ag and Au NPs Analar grade silver nitrate (AgNO3) with 99.9 % purity was purchased from Spectrochem, India and 99.999 % pure hydrogen tetrachloroaurate(III) trihydrate (HAuCl4- 3H2O) was purchased from Sigma-Aldrich, USA. To pre- pare AgNPs, about 1 mL of 20-mM AgNO3 was diluted to 20 mL with double-distilled water and treated with 10 mL of bark extract, and the solution was exposed to bright sunlight. There was a visible color change from pale yel- low to brown in 2 min, and this reaction was monitored with time using UV–Vis spectrophotometer. In the case of AuNPs, 5 mL of 1-mM HAuCl4 was mixed with 5 mL of the extract, and the intensity of pink color was monitored with time using UV–Vis spectrophotometer. In both the cases, 1 mL of the aliquot was diluted to 10 mL with double-distilled water before measuring the absorbance. Plumbago zeylanica is a medicinal herb belonging to the family of Plumbaginaceae. It has been used as a remedy for skin diseases, ringworm, leprosy, sores, and ulcers. The root or leaves are used as counter-irritant and vesicant. The powdered bark is used to treat leprosy, spleen and liver diseases and also for plaque [28]. It contains a variety of important chemical compounds, such as naphthaquinones, alkaloids, glycosides, steroids, triterpenoids, tannins, phe- nolic compounds, ﬂavanoids, saponins, coumarins, and carbohydrates [29, 30]. Plumbagin, 5-hydroxy-2-methyl- 1,4-naphthoquinone is the principal active compound of P. zeylanica [31]. In this work, biosynthesis of silver (Ag) and gold (Au) nanoparticles (NPs) was achieved using the bark extract of P. zeylanica. The water-soluble biocomponents in the extract not only act as reductants to reduce Ag? and Au3? to Ago and Auo, respectively, but also stabilize them by attaching onto the nanoparticles. The nanoparticles stabilized by active biomolecules may have the potential to treat various diseases. The results of antimicrobial, antioxidant, in vitro cytotoxicity of Ag and Au NPs and their ability to bind DNA are reported here. Introduction Biosynthesis of silver nanoparticles using bark of Cinna- mon zeylanicum [20], Piper nigrum [21], Breynia rham- noides [22], Avicennia marina [23], Callicarpa maingayi [24], Cissus quadrangularis [25], Artocarpus elasticus [26], and Saraca indica [27] has also been reported. The use of plant barks for the synthesis of nanoparticles avoids the problems that arise due to seasonal variation of bio- chemicals in the leaves. impurities and air-dried. About 2 g of bark was weighed and boiled for 10 min with 100 mL distilled water in Erlenmeyer ﬂask. After cooling to room temperature, it was ﬁltered through Whatman no. 41 ﬁlter paper. The light yellow colored extract (pH 6) was utilized for the synthesis of Ag and Au NPs. Antimicrobial activity Kirby–Bauer disc diffusion method was employed to assess the antimicrobial activity of biosynthesized Ag and Au NPs. The bacteria Escherichia coli, Pseudomonas aerugi- nosa, Bacillus subtilis, and Staphylococcus aureus, and the fungi Candida tropicalis and Aspergillus ﬂaves were cho- sen for study. For disc diffusion method, the bacterial inoculums were standardized to a density equivalent to barium sulfate standard of 0.5 McFarland units and 123 In vitro cytotoxicity of Ag and Au NPs Cytotoxic effect was determined by tetrazolium dye (3- (4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bro- mide)-based microtitration assay (MTT assay) against DLA (Dalton Lymphoma Ascites) cell lines. Cells were maintained in Dulbecco’s modiﬁed Eagle’s medium (DMEM) supplemented with 10 % fetal bovine serum (FBS) at 37 C in humidiﬁed atmosphere with 5 % CO2. The cells were seeded in 96-well ﬂat-bottom tissue culture plates at a density of approximately 1.2 9 104 cells per well and were allowed to attach overnight at 37 C. Cells were exposed to plant extract, Ag and Au NPs over a range of concentration for 24 h. After the incubation, medium was discarded; the MTT (10 lL of 5 mg/mL) was added to 100 lL of fresh medium and was incubated for additional 4 h. MTT assay was based on the measurement of the mitochondrial activity of viable cells by the reduction of the tetrazolium salt (MTT) to form a blue water-insoluble product, formazan. After 4 h of incubation, the medium was discarded, and 100 lL of DMSO was added to dis- solve the formazan crystals, which were formed by the reduction of tetrazolium salt only by metabolically active cells [32]. The absorbance assay plates were measured at 570 nm in a micro-titer plate reader. Since the absorbance DNA binding studies Antioxidant activity of the synthesized Ag and Au NPs was assessed using DPPH assay. From the stock solution of the nanoparticles (1 mg/mL), samples containing 50, 100, 150, and 200 lg were prepared and mixed with 0.1 % DPPH. The reaction mixture was incubated for 30 min at room temperature. When DPPH reacts with any antioxidant, it gets reduced, and the intensity of color decreases; and the decrease in absorbance at 517 nm was recorded, and butylated hydroxytoluene (BHT) was used as a standard control. The experiment was performed in triplicate, and the scavenging activity was calculated as % inhibition according to the following formula, where, A is absorbance. In every organism, deoxyribonucleic acid (DNA) is an important genetic substance. The study of interaction of drug and DNA has acquired great signiﬁcance for design- ing and synthesizing new drugs targeted at DNA. Pre- suming biomolecules capped nanoparticles as drug, the ability to bind DNA was investigated. UV–Visible spectral studies were employed to study the binding of Ag and Au NPs with Calf-thymus DNA (CT-DNA). DNA solutions were prepared using the buffer 5 mM Tris–HCl/50 mM NaCl at pH = 7.2 in water and sonicated for 30 cycles, where each cycle consisted of 1 min with 30 s interval; which gave ratio of absorbance at 260 and 280 nm, A260/ A280 as 1.87. This ratio indicates that the DNA was suf- ﬁciently free of protein. Absorption spectra were recorded at ﬁxed-concentration Ag and Au NPs for varying con- centration of the CT-DNA (2–10 lg/mL) or vice versa. Stock solutions of DNA were stored at 4 C and used in 4 days. % Inhibition ¼ 100 ðA of standard control A of sampleÞ A of standard control Preparation of bark extract: the reducing agent P. zeylanica bark was purchased from the local market from Madurai, Tamil Nadu, India. The bark was ﬁnely cut, washed with double-distilled water to remove any J Nanostruct Chem (2016) 6:247–260 249 was directly correlated with the number of viable cells, cell survival was calculated by the following formula: swabbed into petridish of 4 mm depth. About 2 mm loopful of extract, Ag and Au NPs were taken and lowered carefully onto the paper disc. The moistened disc was then placed on the surface of inoculated plate and incubated for 16–18 h at 35–37 C. Netilmicin and ﬂuconazole were used as controls for bacteria and fungi, respectively. % viability ¼ At Ac 100 where, At is the absorbance of sample and Ac is the absorbance of control. Percentage cytotoxicity is calculated by the formula: cytotoxicity % = 100 - % viability. where, At is the absorbance of sample and Ac is the absorbance of control. Percentage cytotoxicity is calculated by the formula: cytotoxicity % = 100 - % viability. Parameters inﬂuencing the formation of AgNPs The effect of time, volume of extract, concentration of metal precursors and pH, which inﬂuence the formation of AgNPs were studied in detail, and the results are presented here. Inﬂuence of concentration of silver nitrate Inﬂuence of concentration of silver nitrate Fig. 1 a UV–Vis spectra of the mixtures of the extract and AgNO3 with time b Variation of absorbance with time To optimize the concentration of AgNO3 required for the synthesis of stable AgNPs, it was varied from 1 to 10 mM, and after exposing the solutions to sunlight for 10 min, the SPR band was monitored. With increase in concentration of AgNO3, there was a red shift in SPR band from 418 to 442 nm (Fig. 3a) indicating the increase of particle size. At further high concentrations (5–8 mM), the metallic silver got deposited on the sides of the ﬂask as the silver mirror and the absorbance of the reaction mixture had decreased consequently (Fig. 3b). On further increase of Ag? ion concentration (9 and 10 mM), a black precipitate was obtained, maybe due to formation of Ag2O (results not shown). Fig. 1 a UV–Vis spectra of the mixtures of the extract and AgNO3 with time b Variation of absorbance with time Inﬂuence of extract volume The AgNPs were synthesized over a wide pH range (pH 2–11) using 0.1-M sulphuric acid or 0.1-M sodium hydroxide. In acidic pH, large-sized nanoparticles were formed, while in alkaline pH, small-sized nanoparticles were produced, as indicated by blue shift in kmax from 420 to 413 nm (Fig. 4a). In alkaline pH, many functional groups are ionized and are available for reduction and facilitate the formation of small size nanoparticles. The pH of the extract was 6, and stable nanoparticles were formed in this condition, and hence, all experiments were carried out at this pH unless otherwise mentioned. To assess the effect of concentration of extract on the formation of AgNPs, the volume was varied from 1 to 15 mL by maintaining the total volume of the reaction mixture as constant (20 mL). The solutions were then exposed to sunlight for 10 min, and the UV–Visible spectra were recorded. The intensity of the SPR band around 449 nm increased with increase in the volume of extract up to 10 mL, accompanied by a blue shift in kmax from 449 to 417 nm (Fig. 2a). This infers that size of nanoparticles decreases with increase in concentration of extract (re- ducing agent). When the volume of extract exceeds 10 mL, decrease in the absorbance was noticed (Fig. 2b), maybe due to the aggregation of nanoparticles. Other researchers [34] have also reported this trend in biosynthesis of AgNPs. Narrow peaks were obtained for higher volume of extract implying the formation of smaller sized AgNPs. Effect of time The mixture of the extract and silver nitrate solution was exposed to sunlight, and the SPR band was monitored in regular intervals using UV–Vis spectrophotometer. The color of the solution changed from yellow to dark brown in 2 min, the intensity of the color increased with time and ﬁnally attained a constant value in 40 min implying that the formation of AgNPs is almost complete (Fig. 1a). The decrease in peak width and the shift in kmax from 426 to 416 nm with time infer that the formation of monodis- persed and small-sized nanoparticles, respectively [33]. 12 3 J Nanostruct Chem (2016) 6:247–260 250 UV–Visible spectral analysis of AuNPs The AuNPs synthesized using P. zeylanica were analyzed by UV–Visible spectrophotometer, where SPR band was observed at 542 nm. The formation of AuNPs was so quick; the color of solution turns red with pink tinge within Fig. 2 a UV–Visible spectra of reaction mixture at various volume of extract b Change in absorbance with volume of extract 0 0.8 0.2 0.4 0.6 350 800 400 500 600 700 Absorbance Wavelength [nm] a b 4mM 3 mM 2mM 1 mM Fig. 3 a UV–Visible spectra of AgNPs at varying concentration of AgNO3 b Variation of absorbance with concentration of silver nitrate Fig. 2 a UV–Visible spectra of reaction mixture at various volume of extract b Change in absorbance with volume of extract 0 0.8 0.2 0.4 0.6 350 800 400 500 600 700 Absorbance Wavelength [nm] a b 4mM 3 mM 2mM 1 mM Fig. 3 a UV–Visible spectra of AgNPs at varying concentration of AgNO3 b Variation of absorbance with concentration of silver nitrate 0 0.8 0.2 0.4 0.6 350 800 400 500 600 700 Absorbance Wavelength [nm] a b 4mM 3 mM 2mM 1 mM Fig. 3 a UV–Visible spectra of AgNPs at varying concentration of AgNO3 b Variation of absorbance with concentration of silver nitrate Fig. 2 a UV–Visible spectra of reaction mixture at various volume of extract b Change in absorbance with volume of extract J Nanostruct Chem (2016) 6:247–260 251 NPs inferring that the attachment of biomolecules which are present in the extract to nanoparticles. 0 0.8 0.2 0.4 0.6 200 800 400 600 Absorbance Wavelength [nm] pH 11 pH 10 pH 9 pH 8 pH 7 pH 6 pH 5 pH 4 pH 3 pH 2 a b Fig. 4 a UV–Visible spectra of AgNPs synthesized at various pH b Variation of absorbance with pH 0 0.8 0.2 0.4 0.6 200 800 400 600 Absorbance Wavelength [nm] pH 11 pH 10 pH 9 pH 8 pH 7 pH 6 pH 5 pH 4 pH 3 pH 2 a b a Different plant parts of P. zeylanica contain naph- thaquinones, alkaloids, glycosides, steroids, triterpenoids, tannins, phenolic compounds, ﬂavanoids, coumarins, car- bohydrates, saponins, proteins, oils, and fats [29]. Of all the chemical constituents, plumbagin is the principle active compound. Plumbagin (5-hydroxy-2-methyl-1,4-naphtho- quinone –C11H8O3) is primarily present in roots in higher amounts. The structures of plumbagin, glycoside and derivatives of glycoside are represented below [35]. UV–Visible spectral analysis of AuNPs The major water-soluble components present in P. zeylanica extract are glycosides [36]. O O CH3 OH Plumbagin O O CH3 H3C O O OH OH 3,3 - diplumbagin OH CH3 O OH O HO HO OH Hydroplumbagin glucoside HO O O CH3 OH Plumbagin O O CH3 H3C O O OH OH 3,3 - diplumbagin Fig. 4 a UV–Visible spectra of AgNPs synthesized at various pH b Variation of absorbance with pH O O CH3 OH Plumbagin O O CH3 H3C O O OH OH 3,3 - diplumbagin OH CH3 O OH O HO HO OH Hydroplumbagin glucoside HO O O CH OH Plumbagin OH O Fig. 5 a UV–Visible spectra of the synthesized AuNPs as a function of time b Variation of absorbance with time OH Hydroplumbagin glucoside Hydroplumbagin glucoside Plumbagin 3,3 - diplumbagin The peak at 1631 cm-1 in extract gets shifted to 1615 and 1602 cm-1 in the spectra of Ag and Au NPs, respec- tively. This peak corresponds to chelated[C=O group [37], inferring that the binding of biomolecules with the nanoparticles through this group. The peaks at 3424, 3420, and 3384 cm-1 in the spectra of the extract, Ag and Au NPs are due to –OH groups in the biocomponents, and they may be responsible for bioreduction. The peaks near 2912 cm-1 in all the three spectra are due to –C–H stretching of aro- matic ring. Further, the peaks at 1033 cm-1 and 1032 cm-1 in the spectra of Ag and Au NPs, respectively, which appears as shoulder in extract, correspond to C–O stretching frequency of biocomponents. These data indicate the involvement of [C=O and –OH functional groups in the reduction and stabilization of nanoparticles. Fig. 5 a UV–Visible spectra of the synthesized AuNPs as a function of time b Variation of absorbance with time the time of mixing, and the absorbance became constant within 10 min indicating the completion of the reaction (Fig. 5b). No appreciable shift in the SPR band was observed with time, and the narrow SPR band indicates the formation of monodispersed AuNPs. Mechanism of formation and stabilization of AgNPs The phytoconstituents like plumbagin and hydroplumbagin glucoside contain many hydroxyl groups. On addition of bark extracts to the AgNO3 solution, the Ag?—phytocon- stituent complex is formed as an intermediate. The nascent hydrogens produced during this process reduce Ag? to Ago followed by coalescence, cluster formation, and growth of clusters, ﬁnally yielding AgNPs [38]. The phenolic groups present in the bark extract subsequently undergo oxidation and get converted to their quinone forms. The electro- chemical potential difference between Ag? and phyto- constituents drives the reaction. The formed AgNPs are stabilized through the lone pair of electrons and p electrons of quinone structures (Scheme 1). FT-IR studies FT-IR spectra of Ag and Au NPs were recorded to identify the possible biomolecules responsible for the reduction and capping. The Ag and Au nanosolutions were centrifuged at 13000 rpm for 15 min and air-dried. This dried powder was made as pellets with KBr, and the spectra were recorded. Figure 6 shows the FT-IR spectra of the bark extract, Ag and Au NPs. The spectrum of extract (powder) has peaks at 1063, 1384, and 1631 cm-1. The AgNPs have peaks at 1033, 1384, and 1615 cm-1, while AuNPs exhi- bits peaks at 1032, 1383, and 1602 cm-1. The main peaks present in the extract are also present in both Ag and Au 12 3 J Nanostruct Chem (2016) 6:247–260 252 Fig. 6 FT-IR spectra of a Plumbago zeylanica bark extract, b AgNPs and c AuNPs Fig. 6 FT-IR spectra of a Plumbago zeylanica bark extract, b AgNPs and c AuNPs conﬁrms the crystalline face-centered cubic (fcc) nature, as evidenced by the peaks at 2h values of 37.99, 64.52, and 76.69, which can be indexed to 111, 220, and 311 planes, respectively. AuNPs show similar characteristic peaks of metallic fcc gold (JCPDS: 4-0784) [27]. The particle sizes of Ag and Au NPs were computed by Debye–Scherrer equation from which the average size was found to be 17.26 and 12.84 nm, respectively. D ¼ kk=b cos h D ¼ kk=b cos h where, k is the Scherrer constant, k is the wavelength of the X-ray, b and h are the half-width of the peak and half of the Bragg angle, respectively. TEM studies TEM analysis reveals the size and morphology of biosyn- thesized NPs. TEM images of the Ag and Au NPs at dif- ferent magniﬁcation are depicted in Figs. 7 and 8, respectively. The TEM images clearly show that the nanoparticles are predominantly spherical in shape. The size is in the range of 2.76–51.57 nm (average size 28.47 nm) for AgNPs and 10–25 nm (average size 16.89 nm) for AuNPs. Figures 7d and 8d show the SAED patterns (circular rings) of Ag and Au NPs, that reveal the crystalline nature of nanoparticles. The diffraction rings arise due to reﬂections of planes of fcc crystals which is evinced from broad Bragg’s reﬂection observed in the XRD spectra. D ¼ kk=b cos h ð1Þ ð1Þ XRD studies EDAX (Energy Dispersive X-ray) analysis was per- formed to conﬁrm the elemental form of silver and gold. Figure 10a represents the EDAX proﬁle of AgNPs. The EDAX spectrum shows a strong signal at 3 keV, which corresponds to metallic nanosilver [40]; peak X-ray diffraction patterns of the Ag and Au NPs are shown in Fig. 9a, b, respectively. A comparison of XRD pattern of AgNPs with the standard JCPDS ﬁle no. 89-3722 [39] 123 253 J Nanostruct Chem (2016) 6:247–260 corresponding to Cl- impurity is also present in the AgNPs. Figure 10b represents EDAX proﬁle of AuNPs, OAg OAg O O + 2H+ + 2e- + 2Ag0 OH OH n n-1 + OAg OAg Scheme 1 Mechanism of reduction of Ag? to Ago Fig. 7 TEM images (a– c) AgNPs (d) SAED pattern J Nanostruct Chem (2016) 6:247 260 253 OAg OAg O O + 2H+ + 2e- + 2Ag0 OH OH n n-1 + OAg OAg Scheme 1 Mechanism of reduction of Ag? to Ago OAg OAg O O + 2H+ + 2e- + 2Ag0 OH OH n n-1 + OAg OAg Scheme 1 Mechanism of reduction of Ag? to Ago n-1 + OAg OAg OAg OAg OH O O + 2H+ + 2e- + 2Ag0 Scheme 1 Mechanism of reduction of Ag? to Ago corresponding to Cl- impurity is also present in the spectrum. There are also weak signals for carbon and oxygen, possibly due to the biocomponents attached to AgNPs. Figure 10b represents EDAX proﬁle of AuNPs, which has strong signals around 2 and 10 keV, which correspond to metallic nanogold [41]. There are also Fig. 7 TEM images (a– c) AgNPs (d) SAED pattern Fig. 7 TEM images (a– c) AgNPs (d) SAED pattern Fig. 7 TEM images (a– c) AgNPs (d) SAED pattern corresponding to Cl- impurity is also present in the spectrum. There are also weak signals for carbon and oxygen, possibly due to the biocomponents attached to AgNPs. Figure 10b represents EDAX proﬁle of AuNPs, which has strong signals around 2 and 10 keV, which correspond to metallic nanogold [41]. There are also 12 254 J Nanostruct Chem (2016) 6:247–260 TEM images of (a– NPs d SAED pattern Fig. 8 TEM images of (a– c) AuNPs d SAED pattern weak signals for carbon and oxygen, indicating capping of AuNPs by biocomponents from the plant extract. XRD studies hydrazyl) assay method was adopted to study the antioxidant activity of the synthesized Ag and Au NPs. Substances which are capable of reducing DPPH free radical are radical scavengers and considered to be antioxidants [42]. DPPH is a stable nitrogen-centered free radical with a purple color. When it reacts with an antioxidant, which can donate an electron to DPPH rad- ical, it changes to a stable compound (DPPH-H) with yellow color [43]. As a consequence, purple color decays can be monitored through UV spectrophotometer at 517 nm. Antimicrobial activity The AgNPs have desirable activity against Bacillus subtilis (12 mm) and Pseudomonas aeruginosa (11 mm), but the zone of inhibition for Staphylococcus aureus, Candida tropicalis, E.Coli was 8 mm each; for Aspergillus ﬂaves no inhibition was observed (Fig. 11). The extract and AuNPs have no effect on any of the species. The percentage inhibition of free radicals is presented in Fig. 12. Free radical scavenging activity increases with increasing the concentration of extract, Ag and Au NPs. At 200 lg/mL concentration, extract inhibits 71.16 % of free radicals, whereas, AuNPs, AgNPs, and standard BHT exhibit 87.34, 78.17, and 74.88 % inhibition, respectively. At this concentration, the antioxidant activity decreases in the order: AuNPs [ AgNPs [ extract [ standard BHT. The results are appreciable when compared with the pre- vious report [44]. The IC50 values for Au and Ag NPs are higher than standard BHT (Table S1). 123 In vitro cytotoxicity against DLA cell line In vitro cytotoxicity against DLA cell line itself shows considerable toxicity against DLA cell line. This may be due to the presence of plumbagin, which is reported to be a potent anticancer drug [47]. itself shows considerable toxicity against DLA cell line. This may be due to the presence of plumbagin, which is reported to be a potent anticancer drug [47]. Only live cells reduce the tetrazolium dye but not the dead cells. When MTT is reduced, the quaternary amine is converted into a tertiary amine by opening the tetrazolium ring. A second tertiary amine, which is formed, binds to a hydrogen atom of the quarternary amine. The absorption spectrum of the molecule changes as a result of the dis- placement of this hydrogen atom at high pH. Subsequently, the cell numbers can be estimated. Thus, absorbance is directly correlated to number of viable cells [45]. Antioxidant activity using DPPH assay Antioxidants protect cells against damaging effects of reactive oxygen species, which can neutralize free radi- cals before they can do harm and undo some damage already caused to speciﬁc cells. Medicinal plants contain many free radical scavenging molecules like phenolic compounds and vitamins. P. zeylanica contains a variety of important chemical compounds and, hence, expected to be a potential antioxidant. DPPH (2,2-diphenyl-1-picryl- 123 255 J Nanostruct Chem (2016) 6:247–260 In vitro cytotoxicity against DLA cell line itself shows considerable toxicity against DLA cell line. Thi b d t th f l b i hi h i Fig. 9 XRD pattern of a AgNPs b AuNPs DNA binding studies Nucleic acids show a strong absorbance in the region of 240–275 nm. It originates from the p ? p* transitions of the pyrimidine and purine ring systems of the nucleobases. The bases can be protonated, and therefore, the spectra of DNA and RNA are sensitive to pH. At neutral pH, the absorption maxima range from 253 nm (for guanosine) to 271 nm (for cytidine); as a consequence, polymeric DNA and RNA show a broad and strong absorbance near 260 nm. The percentage cell toxicity of extract, Ag and Au NPs, against DLA cell line is given in the Fig. 13. The data reveal that the increase in concentration of extract increases the toxicity. While comparing the results with other stud- ies, it is obvious that the higher toxicity is observed in aqueous extract, whereas in previous reports, other solvent extracts were found to have higher toxicity [46]. At 150 lg/mL concentration, the extract, Au and Ag NPs exhibit 58.35, 61.56, and 65.61 % toxicity, respectively (Table S2). One interesting observation is that the extract Electronic spectroscopy is one of the useful techniques to study the binding of DNA with small molecules. Small molecules can react with DNA via covalent or non-cova- lent interactions. For non-covalent interactions, generally there are three modes for binding viz. electrostatic attrac- tions with the anionic sugar–phosphate backbone of DNA, 12 3 3 256 J Nanostruct Chem (2016) 6:247–260 interactions with the DNA groove, and intercalation between base pairs through DNA groove [48]. The absorption spectra of CT-DNA with the increase in con- [49]. The values of the binding constant, Kapp are obtained from absorbance values as reported in the literature [50]. For weak binding afﬁnities the data are examined by linear Fig. 10 EDAX spectrum of a AgNPs b AuNPs Fig. 10 EDAX spectrum of a AgNPs b AuNPs Fig. 10 EDAX spectrum of a AgNPs b AuNPs [49]. The values of the binding constant, Kapp are obtained from absorbance values as reported in the literature [50]. For weak binding afﬁnities, the data are examined by linear reciprocal plots based on the following equation: [49]. The values of the binding constant, Kapp are obtained from absorbance values as reported in the literature [50]. For weak binding afﬁnities, the data are examined by linear reciprocal plots based on the following equation: interactions with the DNA groove, and intercalation between base pairs through DNA groove [48]. 123 DNA binding studies The absorption spectra of CT-DNA with the increase in con- centration of Ag and Au NPs are given in Figs. 14 and 15. The increase of absorption intensity (hyperchromism) at 258 nm (due to p ? p* transition of sugar phosphate in DNA) with increase in concentration of Ag and Au NPs indicates the interaction of the nanoparticles with DNA. In general, DNA exhibits both hyper and hypochromism in the electronic spectrum. Hypochromism suggests the interaction of small molecules with DNA through interca- lation, and hyperchromism suggests that the interaction is through groove binding; moreover, hyperchromism is an indication of the breakage of secondary structure of DNA 1 Aobs Ao ¼ 1 Ac Ao þ 1 KappðAc A0Þ½AgNPs] where Ao is the absorbance of CT-DNA in the absence of AgNPs, and Ac is the recorded absorbance of CT-DNA for different concentration of AgNPs. The double reciprocal plot of 1/Aobs - A0 vs 1/[AgNPs] is linear (Fig. 14a), and the binding constant (Kapp) can be calculated from the ratio of intercept to slope. Kapp is found to be 7.03 9 107 and 1.15 9 107 for Ag and Au NPs using the above equation. where Ao is the absorbance of CT-DNA in the absence of AgNPs, and Ac is the recorded absorbance of CT-DNA for different concentration of AgNPs. The double reciprocal plot of 1/Aobs - A0 vs 1/[AgNPs] is linear (Fig. 14a), and the binding constant (Kapp) can be calculated from the ratio of intercept to slope. Kapp is found to be 7.03 9 107 and 1.15 9 107 for Ag and Au NPs using the above equation. 257 J Nanostruct Chem (2016) 6:247–260 Fig. 11 Antibacterial activity of Plumbago zeylanica extract (VVE), AgNPs (VVAG), and AuNPs (VVAU) Fig. 11 Antibacterial activity of Plumbago zeylanica extract (VVE), AgNPs (VVAG), and AuNPs (VVAU) Fig. 13 In vitro cytotoxicity of extract, Ag and Au NPs Fi 13 I i i i f A d A NP Fig. 12 Antioxidant activity of plant extract and that of Ag and Au NPs Fig. 13 In vitro cytotoxicity of extract, Ag and Au NPs Fig. 12 Antioxidant activity of plant extract and that of Ag and Au NPs Fig. 12 Antioxidant activity of plant extract and that of Ag and Au NPs Fig. DNA binding studies 13 In vitro cytotoxicity of extract, Ag and Au NPs The binding constant values are high when compared to normal intercalators [51], which further support groove binding. The intensity of the SPR band also increased with the increase in concentration of Ag and Au NPs. References 1. Padhi, P., Sahoo, G., Gosh, S., Das, K., Panigrahi, S.C.: Synthesis of black and red mercury sulﬁde nano-powder by traditional indian method for biomedical application. AIP Conf. Proc. 1063, 431–438 (2008) Fig. 14 Absorption spectra of (a) CT-DNA with varying concentra- tion of (b) AgNPs; Insert: Plot of 1/[Aobs - A0] vs 1/[AgNP] 2. Fermo, P., Padeleeti, G.: The use of nanoparticles to produce iridescent metallic effects on ancient ceramic objects. J. Nanosci. Nanotechnol. 12, 8764–8769 (2012) 0 2.5 1 2 0 0 9 5 3 2 400 600 800 Abs Wavelength [nm] 200 µL 180 µL 160 µL 140 µL 120 µL 100 µL 80 µL 60 µL 40 µL 20 µL Blank a b Fig. 15 Absorption spectra of (a) CT-DNA with varying concentra- tion of (b) AuNPs; Insert: Plot of 1/[A - A0] vs 1/[AuNPs] 3. Satish Kapoor, K.: Nanoconcept in Indian tradition. Dav’s Ayurveda Holist. Health Mag. 1, 22 (2012) 4. Arvizo, R.R., Bhattacharyya, S., Kudgus, R.A., Giri, K., Bhat- tacharya, R., Mukherjee, P.: Intrinsic therapeutic applications of noble metal nanoparticles: past, present and future. Chem. Soc. Rev. 41, 2943–2970 (2012) 5. Nikhil, R.J., Latha, G., Catherine, J.M.: Wet chemical synthesis of high aspect ratio cylindrical gold nanorods. J. Phys. Chem. B. 105, 4065–4067 (2001) 6. Rodriguez, S.L., Blanco, M.C., Lopez Quintela, M.A.: Electro- chemical synthesis of silver nanoparticles. J. Phys. Chem. B 104, 9683–9686 (2009) 7. Pastoriza, S., Liz Marzan, L.M.: Preparation of PVP-protected metal nanoparticles in DMF. Langmuir 18, 2888–2895 (2002) 8. Noritomi, H., Umezawa, Y., Miyagawa, S., Kato, S.: Preparation of highly concentrated silver nanoparticles in reverse micelles of sucrose fatty acid esters through solid-liquid extraction method. Adv Chem Eng Sci 1, 299–304 (2011) Fig. 15 Absorption spectra of (a) CT-DNA with varying concentra- tion of (b) AuNPs; Insert: Plot of 1/[A - A0] vs 1/[AuNPs] 9. Zhu, J.J., Liu, S.W., Palchik, O., Koltypin, Y., Gedanken, A.: Shape-controlled synthesis of silver nanoparticles by pulse sonoelectrochemical methods. Langmuir 16, 6396–6399 (2000) have potential antioxidant activity, and the percentage of inhibition of free radicals is found to be 78 and 87 %, respectively. In vitro cytotoxicity studies showed that Ag and Au NPs have 65 and 61 % of toxicity against DLA cell lines. The aqueous extract shows enhanced cytotoxicity in comparison with previous reports. Binding of Ag and Au NPs with CT-DNA is demonstrated by UV–Visible spec- trophotometry. Conclusions Bioreductive synthesis of Ag and Au NPs using a medic- inally important plant Plumbago zeylanica, yielded spher- ical nanoparticles with average size of 28.47 and 16.89 nm, respectively. AgNPs have desirable antibacterial activity against Bacillus subtilis and Pseudomonas aeruginosa in comparison to plant extract and AuNPs. Ag and Au NPs Electronic spectra were also recorded for different concentration of CT-DNA at ﬁxed concentration of Ag and Au NPs (Figs. S3 and S4). No appreciable increase in absorbance was noticed with varying concentration of CT- DNA. 12 3 258 J Nanostruct Chem (2016) 6:247–260 Fig. 14 Absorption spectra of (a) CT-DNA with varying concentra- tion of (b) AgNPs; Insert: Plot of 1/[Aobs - A0] vs 1/[AgNP] antioxidant and cytotoxic studies. SAIF, IIT Bombay, India is acknowledged for TEM analysis. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://crea tivecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. References Kasi, G., Shanmugam, G., Ayyakannu, A.: Phytosynthesis of silver nanoparticles using Pterocarpus santalinus leaf exract and their antibacterial properties. J. Nanostruct. Chem. 3, 68 (2013) 40. Nyoman Rupiasih, N., Avinash, A., Suresh, G., Vidyasagar, P.B.: Green synthesis of silver nanoparticles using latex extract of Thevetia peruviana: a novel approach towards poisonous plant utilization. J. Phys. Conf. Ser. 423, 12032–12039 (2013). doi:10. 1088/1742-6596/423/1/012032 23. Gnanadesigan, M., Anand, M., Ravikumar, S., Maruthupandy, M., Syed Ali, M., Vijayakumar, V., Kumaraguru, K.: Antibac- terial potential of biosynthesised silver nanoparticles using Avi- cennia marina mangrove plant. Appl. Nanosci. 2, 143–147 (2012) 41. Narayanan, K.B., Sakthivel, N.: Phytosynthesis of gold nanoparticles using leaf extract of Coleus amboinicus Lour. Mater. Charact. 61, 1232–1238 (2010) 24. Kamyar, S., Mansor Bin, A., Emad, A.J., Ibrahim, N.A., Par- vaneh, S., Abdolhossein, R., Yadollah, A., Samira, B., Sanaz, A., Muhammad, S.U., Mohammed, Z.: Green biosynthesis of silver nanoparticles using Callicarpa maingayi stem bark extraction. Molecules 17, 8506–8517 (2012) 42. Sharad, M., Prachi, B., Man Mohan, S.: Enhanced antioxidant activity of gold nanoparticle embedded 3,6-dihydroxyﬂavone: a combinational study. Appl. Nanosci. 4, 153–161 (2014) 25. Santhoshkumar, T., Rahuman, A.A., Bagavan, A., Marimuthu, S., Jayaseelan, C., Kirthi, A.V., Kamaraj, C., Rajakumar, G., Zahir, A.A., Elango, G., Velayutham, K., Iyappan, M., Siva, C., Kar- thick, L.: Bhaskara Rao, K.V.: Evaluation of stem aqueous extract and synthesized silver nanoparticles using Cissus quad- rangularis against Hippobosca maculate and Rhipicephalus (Boophilus) microplus. Exp. Parasitol. 132, 156–165 (2012) 43. Elena, H.N., Miron, C.T., Gabriela, P., Mihaela, H., Titus, C., Alexandru, B.T.: Reaction of 2,2-diphenyl-1-picrylhydrazyl with HO˙ , O2 -, HO- and HOO- radicals and anions. Int. J. Mol. Sci. 7, 130–143 (2006) 44. Veerapandian, S., Sawant, S.N., Dobie, M.: Antibacterial and Antioxidant activity of protein capped silver and gold nanopar- ticles synthesized with Escherichia coli. J. Biomed. Nanotechnol. 8, 140–148 (2012) 26. Abdullah, N.I.S.B., Ahmad, M.B., Shameli, K.: Biosynthesis of silver nanoparticles using Artocarpus elasticus stem bark extract. Chem. Cent. J. 9, 61 (2015) 45. Berridge, M.V., Herst, P.M., Tan, A.S.: Tetrazolium dyes as tools in cell biology: new insights into their cellular reduction. Biotechnol. Annu. Rev. 11, 127–152 (2005) 27. Shib Shankar, D., Rakhi, M., Arun Kanti, S., Braja Gopal, B., Biplab Kumar, P.: Saraca indica bark extract mediated green synthesis of polyshaped gold nanoparticles and its application in catalytic reduction. Appl. Nanosci. 4, 485–490 (2014) 46. References The hyperchromism in absorbance suggests the groove binding of these NPs with CT-DNA. The data presented in this study may contribute to the development of nanomaterials as an alternative drug to treat cancer. 10. Ahmad, A., Mukherjee, P., Senapati, S., Mandal, D., Islam Khan, M., Kumar, R., Sastry, M.: Extracellular biosynthesis of silver nanoparticles using the fungus Fusarium oxysporum. Colloids Surf. B 28, 313–318 (2003) 11. Balaji, D.S., Basavaraja, S., Raghunandan, D., Mahesh, B., Prabhakar, B.K., Venkataraman, A.: Extracellular biosynthesis of functionalized silver nanoparticles by strains of Cladosporium cladosporioides fungus. Colloids Surf. B 68, 88–92 (2009) 12. Basavaraja, S., Balaji, D.S., Lagashetty, A., Rajasab, A.H., Venkataraman, A.: Extracellular biosynthesis of silver nanopar- ticles using the fungus Fusarium semitectum. Mater. Res. Bull. 43, 1164–1170 (2008) 13. Raghunandan, D., Basavaraja, S., Mahesh, B., Balaji, S., Man- junath, S.Y., Venkataraman, A.: Rapid biosynthesis of irregular shaped gold nanoparticles from macerated aqueous extracellular dried clove buds (Syzygium aromaticum) Solution. Colloids Surf. B 79, 235–240 (2010) Acknowledgments This work is Sponsored by University Grants Commission (UGC), New Delhi. The authors, S. Priya @ Velammal thanks UGC for Minor Research Project and T. Akkini Devi thanks UGC for Project Fellowship. Dr. M.A. Neelakantan, Dean of National Engineering College, Kovilpatti, Tamil Nadu is acknowledged for his help in DNA binding studies. The authors thank Dr. Samasun fertility lab services, Tirupur, Tamil Nadu and Dr. V. R. Mohan, PG and Research Department of Botany, V.O.C. College for their help in 14. Gardea Torresdey, L.J., Gomez, E., Peralta Videa, R.J., Parsons, J.G., Troiani, H., Jose Yacaman, M.: Alfalfa sprouts: a natural source for the synthesis of silver nanoparticles. Langmuir 19, 1357–1361 (2003) 123 3 12 J Nanostruct Chem (2016) 6:247–260 259 32. Sarah Brown, D., Paola, N., Jo-Ann, S., David, S., Paul, E.R., Balaji, V., David Flint, J., Jane Plumb, A., Duncan, G., NialWheate, J.: Gold nanoparticles for the improved anticancer drug delivery of the active component of oxaplatin. J. Am. Chem. Soc. 132, 4678–4684 (2010) 15. Shankar, S.S., Rai, A., Ahmad, A., Sastry, M.: Rapid synthesis of Au, Ag, and bimetallic Au Core–Ag shell nanoparticles using neem (Azadirachta Indica) leaf broth. J. Colloid Interface Sci. 275, 496–502 (2004) 16. Jebakumar, T., Lee, Y.R., Sethuraman, M.G.: Green synthesis of silver nanoparticles using Terminalia cuneata and its catalytic action in reduction of direct yellow-12 dye. Spectrochim. Acta Part A Mol. Biomol. Spectrosc. 161, 122–129 (2016) 33. References Sheny, D.S., Mathew, J., Philip, D.: Phytosynthesis of Au, Ag and Au-Ag bimetallic nanoparticles using aqueous extract and dried leaf of Anacardium occidentale. Spectrochemica Acta Part A 79, 254–262 (2011) Part A Mol. Biomol. Spectrosc. 161, 122–129 (2016) 17. Vidhu, V.K., Philip, D.: Catalytic degradation of organic dyes using biosynthesized silvernanoparticles. Micron 56, 54–62 (2014) 34. Muhammad, A., Farooq, A., Muhammad Ramzan, S.A.J., Muhammad, A.I., Umer, R.: Green synthesis of silver nanopar- ticles through reduction with Solanum Xanthocarpum L. berry extract: characterization, antimicrobial and urease inhibitory activities against Helicobacter pylori. Int. J. Mol. Sci. 13, 9923–9941 (2012) 18. Peter Amaladhas, T., Usha, M., Naveen, S.: Sunlight induced rapid synthesis and kinetics of silver nanoparticles using leaf extract of Achyranthes aspera L. and their antimicrobial appli- cation. Adv. Mat. Lett. 4, 779–785 (2013) 19. Akkini Devi, T., Ananthi, N., Peter Amaladhas, T.: Photobio- logical synthesis of noble metal nanoparticles using Hydrocotyle asiatica and application as catalyst for the photodegradation of cationic dyes. J. Nanostruct. Chem. 6, 75–92 (2016) 35. Zhang, O.R., Mei, Z.N., Yang, G.Z., Xiao, Y.X.: Chemical constituents from aerial parts of Plumbago zeylanica Linn. J. Chin. Med. Mater. 30, 558–560 (2007) 36. Navneet, K., Bhuwan Mishra, B., Vinod Tiwari, K., Vyasji, T.: An account of phytochemicals from Plumbago zeylanica (Fam- ily: Plumbaginaceae): A natural gift to human being. Chron. Young Sci. 3, 178 (2012) 20. Sathishkumar, M., Sneha, K., Won, S.W., Cho, C.W., Kim, S., Yun, Y.S.: Cinnamon zeylanica bark extract and powder medi- ated green synthesis of nano-crystalline silver particles and its bactericidal activity. Colloids Surf B Biointerfaces 73, 332–338 (2009) 37. Lie Chwen, L., Ling Lang, Y., Cheng Jen, C.: Cytotoxic naph- thaquinones and plumbagic acid glucosides from Plumbago zeylanica. Phytochemistry 62, 619–622 (2003) 21. Paulkumar, K., Gnanajobitha, G., Vanaja, M., Rajeshkumar, S., Malarkodi, C., Pandian, K., Annadurai, G.: Piper nigrum leaf and stem assisted green synthesis of silver nanoparticles and evalu- ation of its antibacterial activity against agricultural plant pathogens. Sci World J 2014, 1–9 (2014) 38. Emrah, B., Mahmut, O.: Rapid, facile synthesis of silver nanos- tructure using hydrolysable tannin. Ind. Eng. Chem. Res. 48, 5686–5690 (2009) 22. Gangula, A., Podila, R., Ramakrishna, M., Karanam, L., Janardhana, C., Rao, A.M.: Catalytic reduction of 4-Nitrophenol using biogenic gold and silver nanoparticles derived from Brey- nia rhamnoides. Langmuir 27, 15268–15274 (2011) 39. 51. Pin-xian, X., Zhi-hong, X., Feng-juan, C., Zheng-zhi, Z., Xiao- wen, Z.: Study on synthesis, structure and DNA binding of Ni, Zn complexes with 2-phenylquinoline-4 carboylhydrazide. J. Inorg. Biochem. 103, 210–218 (2009) 49. Nahid, S., Saba, H.: Spectroscopic studies on the interaction of calf thymus DNA with the drug levetiracetam. Spectrochim. Acta Part A Mol. Biomol. Spectrosc. 96, 278–283 (2012) 50. Swarup, R., Ratan, S., Utpal, G., Tapan, K.D.: Interaction studies between biosynthesized silver nanoparticle with calf thymus DNA and cytotoxicity of silver nanoparticles. Spectrochim. Acta Part A Mol. Biomol. Spectrosc. 141, 176–184 (2015) J Nanostruct Chem (2016) 6:247–260 References Krishnaraj, C., Muthukumaran, P., Ramachandran, R., Balaku- maran, M.D., Kalaichelvan, P.T.: Acalypa indica Linn: biogenic synthesis of silver and gold nanoparticles and their cytotoxic effects against MDA-MB-231, human breast cancer cells. Biotechnol. Rep. 4, 42–49 (2014) 28. Tilak, J.C., Adhikari, S., Devasagayam, T.P.A.: Antioxidant properties of Plumbago zeylanica, an Indian medicinal plant and its active ingredient, plumbagin. Redox Rep. 9, 219–227 (2004) 47. Jordan Sand, M., BilalBin, H., Mohammad, S.J., Olya, W., Emily, M.S., Joseph, F., Ajit, K.V.: Plumbagin (5-hydroxy-2- methyl-1,4-naphthoquinone), isolated from Plumbago zeylanica, inhibits ultraviolet radiation-induced development of squamous cell carcinomas. Carcinogenesis 33, 184–190 (2011) 29. Ming, Y., Wang, J., Yang, J., Liu, W.: Chemical constituents of Plumbago zeylanica. Adv. Mater. Res. 308, 1662–1664 (2011) 30. Kapoor, L.D.: Handbook of Ayurvedic medicinal plants. CRC Press, London (1990) 31. Krishnaswamy, M., Purusothaman, K.K.: Plumbagin: a study of its anticancer, antibacterial and antifungal properties. Indian J. Exp. Biol. 18, 876–877 (1980) 48. Yunhua, W., Guangzhong, Y.: Interaction between Garcigenrin and DNA by spectrophotometry and ﬂuorescence spectroscopy. Spectrosc. Lett. 43, 28–35 (2010) 12 12 3 J Nanostruct Chem (2016) 6:247–260 260 123 123
https://openalex.org/W3183836190	https://www.researchsquare.com/article/rs-321054/latest.pdf	English	null	Extraction and Characterization of Antibiotic Compounds produced by Streptomyces spp. VITGV01 against Selected Human Pathogens	Research Square (Research Square)	2,021	cc-by	8,442	Extraction and Characterization of Antibiotic Compounds produced by Streptomyces spp. VITGV01 against Selected Human Pathogens Vellore Institute of Technology: VIT University Godwin Christopher J (  godwinj@vit.ac.in ) Godwin Christopher J (  godwinj@vit.ac.in ) Vellore Institute of Technology: VIT University https://orcid.org/0000-0002-9328-983X Extraction and Characterization of Antibiotic Compounds produced by Streptomyces spp. VITGV01 against Selected Human Pathogens Research Article Keywords: Streptomyces, secondary metabolite, Sequencing, erythromycin, human pathogen DOI: https://doi.org/10.21203/rs.3.rs-321054/v1 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Page 1/17 Abstract Streptomyces are well known to produce a large number of diverse antibiotic compounds. New environments are being explored and some new Streptomyces strains have been discovered; the potential of the plant endophytic environments has been highly explored during the last decade. In the present study, an endophytic actinomycete, VITGV01, was isolated from a farm tomato plant collected from an agricultural field. These actinomycetes are cultured in ISP2 medium supplemented with nystatin (15µg/ml) and nalidixic acid (50µg/ml). Based on the 16S rRNA analysis and physiological tests, this isolate was found to be a member of the known genus Streptomyces spp. and suggesting that it might be a new spp. The isolated strain showed morphological and chemical characteristics of the genus Streptomyces. The strain VITGV01 produced different antibiotics on different media such as yeast extract, malt extract, glucose medium, which were active against some Gram-positive and negative bacteria [Bacillus subtilis (MTCC2756), Staphylococcus aureus (MTCC737), Escherichia coli – (MTCC1687), and Klebsiella pneumoniae (MTCC109)]. The primary and secondary screening was performed against these human pathogens. Maximum zone of inhibition was observed against K. pneumoniae (22 mm) and minimal zone of inhibition against B. subtilis (8 mm). Maximum biomass production was recorded in media containing 1% sucrose and 1% yeast extract cultured at pH 7.0 for 7 days at 30◦C. FTIR analysis revealed amine, alkane (C = C) of the aromatic ring, secondary alcohol, and alkyl halide groups. The GC–MS data showed twenty compounds, of which fifteen are antimicrobial compounds. 1. Introduction Antibiotics are chemical compounds used to prevent and treat bacterial infections. Antibiotic resistance occurs when bacteria change in response to the use of these drugs. Today, drug-resistant bacteria are a serious threat to people’s health. Currently, at least 7,00,000 people die each year due to drug- resistant diseases, including 2,30,000 people who die from multidrug-resistant tuberculosis (WHO 2019). The discovery of new antibiotics that reduce bacterial drug resistance is essential for modern medicine. The World Health Organization (WHO) claims that Antimicrobial Resistance (AMR) presents a significant challenge to public health and the ecosystem. (Allcock et al. 2017) . Among the antibiotic-producing microorganisms, the class actinomycetes represent a broad range of valuable and prominent sources of pharmaceutically active metabolites (Subramani and Sipkema 2019). Over 40 genera of filamentous Actinobacteria have been detected as endophytes in plants. Among the studied endophytes, Streptomyces species, which are isolated from a diverse plant species, are one greater group of interest because of their ability to produce antibiotics as well as other bioactive secondary metabolites (Yan et al. 2010). About 23,000 bioactive secondary metabolites were discovered from endophytic microorganisms, approximately 10,000 different substances were produced by these endophytic actinomycetes and 7,600 metabolites were found in the Streptomyces genus (Hong-Thao et al. 2016). Actinomycetes produce approximately 75% of the commercially used antibiotics (Sanghvi et al. 2014). The actinomycetes are one of the most attractive sources of new bioactive metabolites. However, the rate of discovery of new compounds has decreased since the available species have been studied extensively (Badji et al. 2007). Recently, rare actinomycetes are an important source of novel and useful antibiotics (Takahashi and Nakashima 2018). Several studies have been oriented towards the isolation of new uncommon actinomycetes from unexplored environments. Although numerous actinomycetes have been isolated and used as a producer of key drugs and biomedical agents, it is increasingly difficult to find novel compounds to identify a new organism. To achieve this objective, we have isolate actinomycetes from tomato plants and the following parameters were studied such as morphological, antimicrobial, biochemical, phylogenetic characters, and optimization of growth for Streptomyces spp. VITGV01. Antimicrobial properties (human pathogens) of the secondary metabolites production and characterized on FTIR and GCMS. 2.1 Tomato plant sample collection: Healthy tomato plants (Solanum lycopersicum) were collected from five different locations (Tirunelveli − 8.7815° N, 77.3942° E; Tuticorin 9.1727° N, 77.8715° E; Madurai − 9.9420° N, 77.9724° E; Dindigul − 10.4489° N, 77.9360° E and Theni − 10.0015° N, 77.6164° E) from the southern part of Tamil Nadu in India during August and September 2017. Plants were collected from the field at the fruiting stage. Three tomato plants were randomly collected from the corners and center of the field. Thus, a total of fifteen plants were collected in sterile polythene bags and passed directly to the laboratory for microbiological processing 2.5 Metabolite production Pure and active cultures of microbial strains selected from the secondary screening experiments were grown in ISP2 broth and incubated at 30°C for 7 days and cellular growth was confirmed by visible pellets, clumps, and aggregates in the flask. The culture broth was centrifuged and the supernatant was used to evaluate the antimicrobial activity against the above-mentioned test microorganisms. Antibiotic activities of the strains were compared with that of known commercially available erythromycin standard. 2.2 Surface sterilization Plant samples were washed in running tap water to remove all adherents. The plant samples were cut to 4–5 cm in length using a sterile surgical cutter. The cut segments were washed with sterile distilled water. Then with 70% ethanol (5 minutes), 90% ethanol (2 minutes), 1 % sodium hypochlorite (15 minutes) and 10 % sodium hydrogen carbonate (10 minutes). Sterile distilled water wash was done for every change in solution (5 minutes). Stems were cut in a cross direction and streaked on the ISP2 medium supplemented with 50 mg l− 1 nalidixic acid and nystatin 50 mg l− 1, incubated at 280C for 7–10 days (Qin et al. 2009). Thus, actinomycetes alone were allowed to grow and separated. The strains were purified and maintained in glycerol suspensions (30 % v/v) at -800 C. Isolates were labeled serially as VITGV. Different strains were identified based on the morphological characteristics of colonies on the plate, areal hyphae on an agar plate, the morphology of spores, and pigment production (Williams et al. 1983). Page 2/17 Page 2/17 2.7 Screening of antibacterial activity Concentrated crude ethyl acetate extract obtained from Streptomyces spp. VITGV01 was screened for their inhibitory activity against 4 known human pathogenic bacterial strains by agar Well Diffusion Assay (WDA) (Magaldi et al. 2004). The wells were prepared in the plate (already spread with the test organism) by using a sterile cork borer (6 mm in diameter) at 0.5 McFarland turbidity (1x108cells/mL). A volume of 25, 50, 75, and 100 µL of 1 mg/mL of crude extracts was carefully dispensed into each well and allowed to diffuse for 2 hrs and incubated at 37°C for 24 hrs. To validate the experiment, negative control (ethyl acetate) and positive control (1 mg/ml erythromycin) were also maintained, and each experiment was performed in triplicate. Diameters of inhibition zones were measured after 24 hrs. Tested pathogens such as B. subtilis (MTCC2756), S. aureus (MTCC737), E. coli – (MTCC1687), and K. pneumoniae (MTCC109). 2.3 Screening for effective strain The preliminary antimicrobial screening was done by spot inoculation method (Shomura et al. 1979) using four test microorganisms. Gram-positive (B. subtilis (MTCC2756), S. aureus (MTCC737), and two Gram-negative strains (E. coli – (MTCC1687) and K. pneumoniae (MTCC109). The diameter of inhibition zones was determined after 24 hrs. The experiment was performed in triplicate. The actinobacteria that showed promising results in the preliminary antimicrobial screening were subjected to secondary screening by the agar well diffusion method (Bauer et al. 1966). The above-specified microorganisms were used. Actinomycete antimicrobial extracts were recovered from the culture filtrate by solvent extraction using ethyl acetate (1:1, v/v). The diameters of the zone of inhibition were determined after 24 hours. Samples were assayed in triplicate. 2.8 FTIR and GCMS analysis of secondary metabolites FTIR spectra were recorded on SHIMADZU FT– IR spectrophotometer, Model: IR Affinity in the range of 4000–400 cm− 1(Sanghvi et al. 2014). GC-MS (GC trace ultra-version) was used to find out the chemical compounds. The peaks were identified with the library. 2.6 Extraction of the bioactive metabolites The culture was grown in 1000 ml Erlenmeyer flasks containing 200 ml of ISP2 broth for 7 days as specified in metabolite production. The culture was centrifuged at 10,000 rpm for 20 min to separate the biomass. The secondary metabolite was recovered from the fermented ISP2 broth using a two- phase solvent extraction system with organic solvent (ethyl acetate). (Shetty et al. 2014). The mixtures of separating flask crude culture filtrate: solvent (1:1 ratio) were vigorously shaken for 10 min and in kept shaker from 18 to 24 hrs at 200 rpm. After 24 hrs this was left to settle for 30 minutes for separating flask kept stationary from 15 to 30 min. until the separation of aqueous and organic phases. Organic phases were collected and concentrated in a rotary evaporator (model RE100‒Pro; SCILOGEX, LLC, CT, USA) at 54°C and 80 rpm. The dried crude extracts were then weighed, dissolved in 200µL of methanol, and finally stored at -20°C. 2.4 Characterization of actinomycetes Growth and morphology of VIT GV01 were observed on different media such as ISP2, Actinomycetes isolation agar, potato dextrose agar, Yeast peptone mannitol agar, Czapek’s Dox agar (Nguyen and Kim 2015), Zobell marine agar (Subramani and Sipkema 2019), Yeast Peptone Dextrose agar, soybean casein digest medium, nutrient agar, Luria Bertani agar, Tryptophan agar, Christensen agar medium (Hirsch and Christensen 1983), starch agar, Soy flour Mannitol Agar. Morphology of the strain was observed in a phase-contrast microscope (MT5210/MEIJI) and scanning electron microscopy was performed using Carl Zeiss – Evo 18). The biochemical characterization was performed for temperatures (25, 30, 35, 40, and 450C), utilization of different sugars as sole carbon sources such as sucrose, glucose, fructose, maltose and lactose, and pH (2, 4, 6, 7, 8 and 10) to find the growth optimum. Biochemical tests were performed as described by Bergey’s manual (“Bergey, D.H. and Holt, J.G. (1994) for strain identification. Analysis of the 16S rRNA was done with the primers forward (5’-GAGTTTGATCC TGGCTCA-3’) and reversed (5’-ACGGCTACCTTGTTACGACTT-3’). Amplified PCR product was sequenced and the nucleotide sequence was matched using the BLAST program. The phylogenetic tree was constructed using the neighbor-joining method (Fatima et al. 2019). The sequence of the isolate was submitted to GenBank (Accession ID: MN641011). 2 5 Metabolite production 3. Result The present study focuses on the isolation of potent antibiotic-producing endophytic actinomycetes. From two hundred and forty actinomycetes strains were isolated in five districts of Tamilnadu (Fig. 1), India. Among the isolated pure strains, VITGV01 was found to produce a wide spectrum of Page 3/17 Page 3/17 antimicrobial activities against selected two Gram-positive and two Gram-negative bacteria. The isolated VITGV01 showed a clear zone of inhibition (Fig. 2) due to antimicrobial activity. 3.2 Physiological Characters Optimization parameters for growth of this Strain VITGV01 require pH 7, temperature 300C and 1% yeast extract, 1% sucrose incubation duration 7 days were best suited (Fig. 3). Optimization parameters for growth of this Strain VITGV01 require pH 7, temperature 300C and 1% yeast extract, 1% sucrose incubation duration 7 days were best suited (Fig. 3). Optimization parameters for growth of this Strain VITGV01 require pH 7, temperature 300C and 1% yeast ex were best suited (Fig. 3). The selected strains were subjected to various biochemical tests and the results are tabulated in Table 1, which supports that it belongs to actinomycetes. The selected strains were subjected to various biochemical tests and the results are tabulated in Table 1, which supports that it belongs to actinomycetes. The selected strains were subjected to various biochemical tests and the results are tabulated in Table 1, which supports that it belongs to actinomycetes. Table 1 Table 1 Various Biochemical Characteristics of VITGV01 S. No Character Result 1 Gram staining Positive 2 Motility Non-motile 3 Indole Negative 4 Methyl Red (MR) Negative 5 Voges Proskauer (VP) Positive 6 Citrate utilization Positive 6 Catalase Positive 7 Lactose Positive 8 Glucose Positive 8. Fructose Positive 9. Sucrose Positive 10. Galactose Positive 11. Starch Positive 12 Mannitol Positive 13 Spore chain Positive 14 Shape and growth Mycelial Spore chain 15 Temperature optimum 300 C 16 Temperature range for growth 250 C − 450 C 17 Range of pH for Growth 6–12 18 Urease Positive 19 Oxidase Negative 21 Aerial Mass color Slight brown color 22 Soluble pigments Negative 21 Casein Positive 22 Gelatine Negative 23 Alkaline phosphate Positive 25 Anaerobic growth Positive 26 Chitin Positive 27 Ammonium chloride utilization Positive 28 Nitrogen sulfate utilization Positive 29 Aerial Hyphae Positive 30 Diffusible pigment Negative 31 Cellulose Positive 32 Pectinase Negative Page 4/17 Page 4/17 3.3 Morphological observation of VITGV63 The cultural characteristics and colony morphology were summarized in Table 2 and Fig. 4. These morphological observations revealed that VITGV01 was similar to Streptomyces. It showed well developed aerial and substrate mycelium and mycelial cluster showed continuous fragmented long and short chain-like structure (Fig. 5B). The scanning electron microscope image of (Fig. 5C) sample exhibited continuous filamentous spiral chains in a tubular shape. The continuous aerial hyphae differentiated into a small and smooth-surfaced spore. The center of the spore showed a cavity on the upper side of each spore (Fig. 5D). Table 2 Colony Morphology of VITGV01 on Different Agar medium S. No Different Media Size of the colony Colour of the colony Surface of the colony Consistency / Texture of the colony Overall Growth A ISP 2 Agar Small 1–2 mm Whitish Gray Rough opaque Brittle Moderate B Actinomycetes Isolation Agar Small 1–2 mm Greyish white Rough opaque Viscid Moderate C Potato Dextrose Agar Pinpoint Gray spot Smooth Viscid Meagre D Yeast peptone Mannitol agar Pinpoint White ash gray Rough & Transparent Viscid Meagre E Czapek Dox Agar Large 2–3 mm Gray Smooth opaque Dry Strong growth F Zobell Marine Agar Pinpoint Yellowish white Rough / Transparent Friable Meagre G Soybean casein digest agar Large 2–3 mm White Rough opaque Mucoid Strong growth H Yeast Peptone Dextrose agar Small 1–2 mm Gray Smooth opaque Brittle Moderate I Nutrient agar Large 2–3 mm White Smooth opaque Brittle Strong growth J Luria Bertani Agar Small 1–2 mm White Smooth opaque Viscid Moderate K Tryptophan agar Large 2–3 mm Yellowish Butryous opaque Viscid Strong growth L Christensen agar Small 1–2 mm Ash Greyish Smooth opaque Friable Moderate M Starch agar Large 2–3 mm White Smooth opaque Friable Strong growth N Soy flour Mannitol Agar Small 1–2 mm Whitish ash Gray Smooth Jelly Meagre 3.6 FTIR analysis The FTIR spectrum of the antimicrobial compounds were compared with Library. It also showed characteristic band corresponding to 21 peaks 3406.29, 3298.28, 2958.80, 2926.01, 2854.65, 1707.00, 1647.21, 1537.27, 1454.33, 1413.82, 1334.74, 1274.95, 1244.09, 1091.71, 1008.77, 894.97, 767.67, 740.67, 607.58, 489.92, 422.41 (Fig. 8). The FTIR spectrum of the antimicrobial compounds were compared with Library. It also showed characteristic band corresponding to 21 peaks 3406.29, 3298.28, 2958.80, 2926.01, 2854.65, 1707.00, 1647.21, 1537.27, 1454.33, 1413.82, 1334.74, 1274.95, 1244.09, 1091.71, 1008.77, 894.97, 767.67, 740.67, 607.58, 489.92, 422.41 (Fig. 8). 3.4 Genetic analysis of VITGV63 The 16S rRNA sequence of the strain VITGV01 (1040 nucleotide), to identify the 16S rRNA gene sequence with high similarity strain VITGV01 shared with 16S rRNA high degree of similarity with other types of genus Streptomyces. The phylogenetic tree of the strain VITGV01 forms a single separate Page 5/17 branch from other Streptomyces neighbors (Fig. 6). It shares 16S rRNA sequence to nearest neighbor is Streptomyces atacamensis (99.42 %), Phylogenetic tree builder uses sequences aligned with method System Software aligner. 3.5 Extraction of bioactive compounds The crude extract exhibited antimicrobial properties proved by the Well diffusion method (Table 3) for secondary screening. Effective zone of inhibition was also recorded against B. subtilis S. aureus, K. pneumoniae, and E. coli. (Fig. 7). The zone of inhibition was observed between the range from 8 to 22 mm. No zone of inhibition was observed in negative control with ethyl acetate. A maximum inhibition zone of 22 mm was observed against K. pneumoniae, which was also to positive control as erythromycin (22 mm). Table 3 Table 3 Zone of inhibition (mm) in secondary screening of crude extracts produced from VITGV01 by using well diffusion VITGV01 Zone of inhibition (µg/ml) Gram Positive Gram Negative B. subtilis S. aureus K. pneumoniae E. coli 25 50 75 100 25 50 75 100 25 50 75 100 25 50 75 100 Crude Extract 8 11 12 8 16 18 19 22 15 18 20 22 15 17 18 20 Positive control (Erythromycin) 10 11 13 15 14 15 17 18 10 14 16 24 10 11 12 13 Negative control (Ethyl Acetate) - - - - - - - - - - - - - - - - 3.7 GCMS analysis Streptomyces VITGV01 extract was characterized with GCMS (Fig. 9). Based on the retention time, peak area, molecular weight, and molecular formula, twenty major chemical compounds were identified by comparing mass spectra with the library as shown in Table 4. The major identified compounds were 1. Isophorone; 2. 1H-Indene, 1-methylene-; 3. 2-Trifluoromethylbenzoic acid; 4. Acetic acid, 2-phenyl ethyl ester; 5. 2-Methoxy-4-vinyl phenol; 6. Phenol, 2,4-bis(1,1-dimethyl ethyl)-; 7. Semustine; 8. Diethyl Phthalate; 9. Hexadecanoic acid, ethyl ester; 10. n-Hexadecanoic acid; 11. 2,4-Diamino-6- nitroquinazoline; 12. 3,5-dimethoxy phenol; 13. Tetradecyl trifluoroacetate; 14. 2,5-Dimethoxyaniline; 15. Octadecanoic acid, ethyl ester; 16. 4-Nitro-3- picoline-N-oxide; 17. Pyrrolo[1 2-a]pyrazine-1 4-dione hexahydro-3-(2-methyl propyl)-; 18. 2-hydroxy-3 5 5-trimethyl-2-cyclohexenone; 19. Pyrrolo[1,2-a] pyrazine-1,4-dione, hexahydro-3-(phenylmethyl)-; and 20. 2(3h)-oxazolethione 4 5-diphenyl-2-cyclohexenone are present in the sample. Their predicted chemical structures were reported in Fig. 10. Among all the twenty compounds 4-Nitro-3-picoline-N-oxide and Phenol, 2,4-bis(1,1-dimethyl ethyl)- showed the highest peak area compared to other chemical compounds in the extract of Streptomyces spp. VITGV01. Page 6/17 Page 6/17 Table 4 Major Chemical Compounds identified by GC–MS for Streptomyces spp. VITGV01. S. No Chemical Compound RT Molecular weight Molecular Formula Area % Nature of The Compound Activity Spectrum of action Reference 1 Isophorone 8.585 138.21 C9H14O 1.68 Cyclohexane Antimicrobial Broad (Kiran et al., 2013) 2 1H-Indene, 1- methylene- 9.558 128.17 C10H8 4.85 Methylene Antimicrobial Broad (Osuntokun and Cristina, 2019) 3 2- Trifluoromethylbenzoic acid, 10.313 190.12 C8H5F3O2 0.89 Ester Antimicrobial Broad (Krátký and Vinšová, 2012) 4 Acetic acid, 2- phenylethyl ester 10.464 164.20 C10H12O2 0.67 Ester Antimicrobial Broad (Al-Dhabi et al., 2016) 5 2-Methoxy-4- vinylphenol 11.487 150.17 C9H10O2 2.09 Phenolic Antimicrobial Broad (Rubab et al., 2020) 6 Phenol, 2,4-bis(1,1- dimethylethyl)- 13.165 278.5 C17H30OSi 22.35 Phenolic Antimicrobial Narrow (Padmavathi et al., 2014) 7 Semustine 13.727 247.72 C10H18ClN3O2 0.78 Methyl Anticancer Antitumor (Agarwal et al., 2015) 8 Diethyl Phthalate 15.220 222.24 C12H14O4 1.92 Ethyl ester antimicrobial Broad (Premjanu N and Jaynthy C, 2014) 9 Hexadecanoic acid, ethyl ester 17.821 284.4 C18H36O2 0.82 Stearic acid antioxidant - (Kim et al., 2020) 10 n-Hexadecanoic acid 17.913 256.4 C16H32O2 12.92 carboxylic antibacterial Narrow (Abubakar and Majinda, 2016) 11 2,4-Diamino-6- nitroquinazoline 18.349 205.1 C8H7N5O2 1.27 amine antimicrobial Broad (Mallikarjuna Reddy et al., 2015) 12 3,5-dimethoxyphenol 18.937 154.1 C8H10O3 1.61 phenol Antimicrobial Broad (Laddha and Biyani, 2019) 13 Tetradecyl trifluoroacetate 19.130 310.1 C16H29F3O2 0.66 Ester Antimycobacterial Narrow spectrum (Nimbeshaho et al., 2020) 14 2,5-Dimethoxyaniline 19.297 153.18 C8H11NO2 3.02 Benzene Antimicrobial Broad (Nordin et al., 2017) 15 Octadecanoic acid, ethyl ester 19.675 312.5 C20H40O2 0.63 Ester Antioxidant - (Kim et al., 2020) 16 4-Nitro-3-picoline-N- oxide 19.817 289.8 C14H24ClNO3 26.16 pyridine Antibacterial Broad (Karayannis et al., 1974) 17 Pyrrolo[1 2-a]pyrazine- 1 4-dione hexahydro-3- (2-methylpropyl)- 19.943 210.2 C11H18N2O2 10.36 pyrazine Antibacterial and Antioxidant Broad (Putra and Karim, 2020) 18 2-hydroxy-3 5 5- trimethyl-2- cyclohexenone 20.153 154.21 C9H14O2 0.85 Cyclohexane Antimicrobial Broad (Nguyen et al., 2019) 19 Pyrrolo[1,2-a] pyrazine- 1,4-dione, hexahydro-3- (phenylmethyl)- 24.557 244.2 C14H16N2O2 1.72 pyrazine Antioxidant - (Ser et al., 2015) 20 4,5-Diphenyloxazole- 2(3H)-thione 26.444 253.3 C15H11NOS 4.75 Oxazole antimicrobial Broad (Kakkar and Narasimhan, 2019) 4. Discussion This study support that according to carbon sources, growth in Streptomyces spp. is strongly influenced. Nitrogen source is required for the synthesis of cells and energy sources. Also, biomass yield and cell morphology are strongly influenced by the nitrogen source. Similar results have been reported by (Al-Ansari et al. 2020) in their studies associated with Streptomyces greseoplanus. In another study, by Wang et al., (2010) nitrogen sources from yeast extract supported the greatest mycelium biosynthesis in Saccharothrix yanglingensis sp. nov. Microorganisms are capable of growing over a wide pH range. Moreover, pH is the third important factor that influences microbial growth. In this way, responses to environmental signals are transmitted through global transcription factors that mediate environmental signals like carbon (CreA) (Janus et al. 2008), nitrogen (AreA) (Tudzynski et al. 1999), and pH (PacC) (Tilburn et al. 1995), translating environmental cues into Secondary Metabolites and concomitant physiological responses. The enhancement or inhibition of the secondary metabolite is influenced by a change in pH and can also affect the growth of microorganisms (Kamjam et al. 2019). In this experiment, we report the optimum pH for maximum growth is pH 7, and the lowest growth was observed in pH 2. This is very similar to the study by Streptomyces polaris sp by Kamjam et al., (2019). An important criterion for the potency assessment is the time course for growth plateau formation and continuous growth was observed till the 14th day. A similar report was documented by Yagüe et al., (2012) in Streptomyces coelicolor for growth optimization. Another factor is the incubation temperature, which affects the microbial growth, metabolite production, rate of the chemical reaction, which in turn affects the rate of metabolite production. In the present study, the maximum activity was recorded at 30°C. By increasing the incubation temperature, a gradual decrease in growth was observed. This is a negative impact on this organism; this could be due to the thermolabile nature of extracellular enzymes. Studies have suggested that to identify actinomycetes details of biochemical, morphological, and phylogenetical characterization are necessary. The biochemical characterization proved that it belongs to actinomycetes (Table 2). 4. Discussion Actinomycetes play a major role in combating various diseases. Earlier exploration has shown actinomycetes having excellent potential to control the growth of pathogens (Ayoubi et al. 2018). In our study, a total of 240 endophytic actinomycetes were isolated and screened for their antimicrobial Page 7/17 Page 7/17 activities, out of which 77 actinomycetes showed the antimicrobial property. A similar report of recognizing 619 endophytic actinomycetes was reported by Tan et al., (2006) out of this 8 alone exhibited antagonistic property. activities, out of which 77 actinomycetes showed the antimicrobial property. A similar report of recognizing 619 endophytic actinomycetes was reported by Tan et al., (2006) out of this 8 alone exhibited antagonistic property. The size of the inhibition zone is the first criterion for screening antimicrobial activity on pathogens, in the present study, there was a good number of microbes (77/240) which shows a good zone of inhibition (Fig. 3). A total of 44 (18%) isolates inhibited the growth of both Gram-positive and Gram- negative pathogen, while 18 isolates (8%) inhibited Gram-negative and 15 isolates (6%) inhibited Gram-positive bacteria. Thus seventy-seven (32%) isolates were selected to explore their potentials. Strain VITGV01 showed the highest antimicrobial activity against all four pathogens. Morphological characteristics such as aerobic, powdery, flat appearance, aerial and substrate mycelia with different colors in different media and colony size suggested that the isolates are affiliated to genus Streptomyces. The above comparison with the description given in Bergey’s manual of systematic bacteriology indicates a major resemblance of this strain with Streptomyces species. This study coincides with reports of Al-Ansari et al., (2019) SEM results showed hyphae’s were compartmentalized and spores chain-like and continuation (Fig. 5). The mature spore’s surface was smooth and unique with a small concave depression; this morphology was the same as that of actinobacteria AFD7 strains (Fatima et al. 2019) The nature and concentration of sugars such as sucrose, maltose, fructose, and glucose are crucial for the growth, nutrient consumption, and production of secondary metabolites. And are important for signal transduction, gene regulation, and development. The utilization rate of carbon and nitrogen sources by VITGV01 seems to be unique. Out of five different carbon sources, the highest growth was obtained with 1 % sucrose. On the other hand, glucose recorded the lowest growth rate. Similarly, El-Hadi et al., (2019) reported lactose for maximum and mannitol for minimal growth in Streptomyces greseoplanus. 4. Discussion Page 8/17 Infrared spectra showed a primary amine (3406.06 cm− 1, 3298.28 cm − 1), alkane groups (2958.80–2854.65cm − 1), (C = C) of aromatic ring (1537.27- 1413.82 cm− 1), Amine (C-N) (1334.74–1091.71) and secondary alcohol (1008.77, 894.97- 489.92) Alkyl Halide (C-Cl) (740.67, 607.58). A similar result was reported by Mathur et al (2015) for Streptomyces spp. and Badji et al (2007) in Nonomuraea sp. NM94. Some differences were also noted with imine functional groups as per the report of Shangvi et al (2014) in Streptomyces werranesis. (Badji et al. 2007; Kumar et al. 2014a; Mathur et al. 2015; Jung et al. 2018). Infrared spectra showed a primary amine (3406.06 cm− 1, 3298.28 cm − 1), alkane groups (2958.80–2854.65cm − 1), (C = C) of aromatic ring (1537.27- 1413.82 cm− 1), Amine (C-N) (1334.74–1091.71) and secondary alcohol (1008.77, 894.97- 489.92) Alkyl Halide (C-Cl) (740.67, 607.58). A similar result was reported by Mathur et al (2015) for Streptomyces spp. and Badji et al (2007) in Nonomuraea sp. NM94. Some differences were also noted with imine functional groups as per the report of Shangvi et al (2014) in Streptomyces werranesis. (Badji et al. 2007; Kumar et al. 2014a; Mathur et al. 2015; Jung et al. 2018). GC-MS analysis has contributed significantly to the bioprospecting of natural products isolated from Streptomyces bacteria (Schöller et al. 2002; Selvakumar et al. 2015; Ser et al. 2015a). There are many reports for the GC-MS-based chemical analysis of actinobacterial secondary metabolite. In this study, GC-MS analysis shows twenty major chemical compounds with different retention times. The identified compounds were in the class of methylene, methyl, ethyl ester, stearic acid, carboxylic, amine, phenol, benzene, ester, pyridine, pyrazine, cyclohexane, pyrazine, and oxazole in nature. The two major highest peaks (area %) of 26.16 and 22.35 were obtained for 4-Nitro-3-picoline-N-oxide and Phenol, 2,4-bis(1,1-dimethyl ethyl)- with the retention time of 26.16 and 13.165 min respectively, and the lowest peak (area %) of 0.63 was obtained for Octadecanoic acid, ethyl ester with the retention time of 19.675 min. Among the detected compounds by GC MS analysis, compound 4-Nitro-3-picoline-N-oxide is pyridine in nature (PubChem) which is most frequently detected in microbial extract of Streptomyces avermitilis NIC B62, demonstrating antibacterial and antifungal activities (Lysenkova et al. 2010). The compound, Phenol, 2,4-bis(1,1-dimethyl ethyl)- is a phenolic compound which is previously one study reported by Tan et al., (2019) from Streptomyces sp. 4. Discussion MUM265 is a potential source of the antioxidant and anti colon-cancer agent. Most of the other compounds detected by GC-MS in a few Streptomyces sp. VITGV01 extract has been reported previously in Streptomyces spp. extracts, such as n-Hexadecenoic acid (El- Naggar et al. 2017), Pyrrolo [1 2-a]pyrazine-1 4-dione hexahydro-3-(2-methyl propyl)- (Ser et al. 2015b), 1H-Indene, 1-methylene- group (Wang et al. 2013), 2-Methoxy-4-vinyl phenol (Naureen et al. 2017), 2,5-Dimethoxyaniline (Nain-Perez et al. 2016; Nepali et al. 2016), Pyrrolo [1,2-a] pyrazine-1,4- dione, hexahydro-3-(phenylmethyl)- (Sanjenbam and Kannabiran 2016), 3,5-dimethoxy phenol (Murray et al. 2020), Diethyl Phthalate (Kapanen et al. 2007), Acetic acid, 2-phenyl ethyl ester (Al-Dhabi et al. 2019). These compounds could be the key contributors to the potential antimicrobial activity of Streptomyces sp. VITGV01. These known compounds detected in GCMS are well-known ability to inhibit the growth of the four tested pathogens. However, to date, there is no report on Isophorone, 2-Trifluoromethylbenzoic acid, Semustine, 2,4-Diamino-6-nitro quinazoline, and 4,5-Diphenyloxazole- 2(3H)-thione, from any other Streptomyces spp. These major compounds 4-Nitro-3-picoline-N-oxide, Phenol, 2,4-bis(1,1-dimethyl ethyl)-, n-Hexadecanoic acid, and Pyrrolo[1 2-a]pyrazine-1 4-dione hexahydro-3-(2-methyl propyl)- comes under phenolic, carboxylic, pyridine, and pyrazine group. These compounds together constitute 71.79% of the total compounds present in the extract of Streptomyces sp. VITGV01. Kumar et al., (2014b) reported strong antimicrobial activity of phenolic compounds (Phenol, 2,4-bis(1,1-dimethyl ethyl)), from Streptomyces sp. SCA7 strain strongly inhibited the growth of Gram-positive and negative organisms. Kumari et al., (2019) reported that the antimicrobial effect of different carboxylic, pyridine, and pyrrolidine group of compounds on antibacterial activity. The above report suggests the inhibitory effect on both Gram-positive and negative properties of Streptomyces sp. VITGV01 extract. Conclusion Plants are a rich source of endophytes. The identification of novel Streptomyces strains may lead to new drug discovery. In this study, 240 endophytic actinomycetes were primarily screened and among these 77 isolates showed antimicrobial activity against selected human pathogens. Based on the biochemical characterization, and 16s rRNA studies, it can be a new strain. Our study encourages the exploration of diverse ecosystems for the isolation of new species for novel bioactive compounds. As there is an alarming increase in multi-drug resistance in pathogens, exploration of new bioactive compounds helps to combat the multi-drug resistance. Declarations Funding: Vellore Institute of Technology provided ‘VIT SEED GRANT’ for carrying out this research work, is acknowledged. Authors' contributions: All the authors contributed equally Informed consent statement: Not applicable Ethical statement: Not applicable Ethical statement: Not applicable Data availability statement: Data supporting the results are given in tables. 4. Discussion The 16s rRNA study and phylogenetic tree result again proved that this strain clustered within the genus Streptomyces and exhibited close phylogenetic affinity and high sequence similarity to Streptomyces atacamensis (99.42 %), Streptomyces mangrovi (99.33 %), Streptomyces chitinivorans (99.23 %), and Streptomyces atacamensis (99.04%), This is the first study to identify this Streptomyces spp. (Gen Bank Accession - MN641011). A similar type of conformation for Streptomyces was reported by Duddu & Guntuku, 2016). Isolation of antibacterial compounds from the endophytic environment is in advance interest in recent times to isolate novel bioactive actinomycetes. Musa et al.,(2020) isolated fifty-four endophytic actinomycetes, which showed antagonistic activities. During secondary screening with crude extracts, a wide range of inhibition zone against tested bacterial strains were observed. Similar findings were reported earlier by Gebreyohannes et al., (2013) in actinomyces spp. In general, the antibacterial activity of crude extracts has fluctuated widely. The results demonstrate that a Gram-positive and Gram- negative bacterium both were highly susceptible to the tested crude extracts. This result was in agreement with the report of Sanghvi et al., (2014) who reported it with Streptomyces werraensis. VITGV01 crude extract showed excellent activity on Gram-positive and Gram-negative bacteria, suggest that it could inhibit DNA synthesis, RNA synthesis, cell wall synthesis, or protein synthesis (Kohanski et al. 2010). There was a marked difference between the crude extracts and pure antibacterial drug (Erythromycin). Some of the crude extracts showed a higher or equivalent zone of inhibition compared to the standard drug erythromycin, which shows excellent activity against Gram-positive organisms and modest activity against Gram-negative bacteria. Rex et al. (2001) reported that a significant difference was normally presented in crude (unpurified) extracts compared with a pure drug that was already in clinical use.. The MIC varied among the tested isolates against all the tested pathogen. These results were similar to the report of Núñez-Montero et al., (2019). Therefore, crude extracts could be a potent source for antibiotic production, which leads to the development of novel drugs for the treatment of infectious diseases. The report of antibacterial activities of actinomycetes isolated from Tomato endophytes of agrofield Madurai in Tamilnadu was first-hand information as per our knowledge. References Al-Ansari M, Alkubaisi N, Vijayaragavan P, Murugan K (2019) Antimicrobial potential of Streptomyces sp. to the Gram positive and Gram negative pathogens. J Infect Public Health 12:861–866. https://doi.org/10.1016/j.jiph.2019.05.016 Al-Ansari M, Kalaiyarasi M, Almalki MA, Vijayaraghavan P (2020) Optimization of medium components for the production of antimicrobial and Al-Ansari M, Alkubaisi N, Vijayaragavan P, Murugan K (2019) Antimicrobial potential of Streptomyces sp. to the Gram positive and Gram negative pathogens. J Infect Public Health 12:861–866. https://doi.org/10.1016/j.jiph.2019.05.016 Al-Ansari M, Kalaiyarasi M, Almalki MA, Vijayaraghavan P (2020) Optimization of medium components for the production of antimicrobial and anticancer secondary metabolites from Streptomyces sp. AS11 isolated from the marine environment. J King Saud Univ - Sci 32:1993–1998. https://doi.org/10.1016/j.jksus.2020.02.005 Page 9/17 Al-Dhabi NA, Esmail GA, Duraipandiyan V, Arasu MV (2019) Chemical profiling of Streptomyces sp. Al-Dhabi-2 recovered from an extreme environment in Saudi Arabia as a novel drug source for medical and industrial applications. Saudi J Biol Sci 26:758–766. https://doi.org/10.1016/j.sjbs.2019.03.009 Allcock S, Young EH, Holmes M, et al (2017) Antimicrobial resistance in human populations: Challenges and opportunities. Glob. Heal. Epidemiol. Genomics 2:1–7. http://doi: 10.1017/gheg.2017.4 Ayoubi H, Mouslim A, Moujabbir S, et al (2018) Isolation and phenotypic characterization of actinomycetes from Rabat neighborhood soil and their potential to produce bioactive compounds. African J Microbiol Res 12:186–191. https://doi.org/10.5897/ajmr2017.8761 Badji B, Mostefaoui A, Sabaou N, et al (2007) Isolation and partial characterization of antimicrobial compounds from a new strain Nonomuraea sp. NM94. J Ind Microbiol Biotechnol 34:403–412. https://doi.org/10.1007/s10295-007-0210-z Bauer AW, Kirby WMM, Sherris JC, Turck M (1966) Antibiotic Susceptibility Testing by a Standardized Single Disk Method. Am J Clin Pathol 45:493– 496. https://doi.org/10.1093/ajcp/45.4_ts.493 Duddu MK, Guntuku G (2016) Isolation, screening and characterization of antibiotic producing actinomycetes from kapuluppada plastic waste dumping yard, visakhapatnam. Int J Pharm Pharm Sci 8:221–229. https://doi.org/10.22159/ijpps.2016v8i11.10110 El-Hadi A, Ahmed H, Hamzawy R (2019) Optimization and characterization of l-asparaginase production by a novel isolated streptomyces spp. strain. Egypt Pharm J 2:111-122. https://doi.org/10.4103/epj.epj_23_18 El-Naggar NE-A, El-Bindary AA-A, Abdel-Mogib M, Nour NS (2017) In vitro activity, extraction, separation and structure elucidation of antibiotic produced by Streptomyces anulatus NEAE-94 active against multidrug-resistant Staphylococcus aureus. Biotechnol Biotechnol Equip 31:418–430. https://doi.org/10.1080/13102818.2016.1276412 Fatima A, Aftab U, Shaaban KA, et al (2019) Spore forming Actinobacterial diversity of Cholistan Desert Pakistan: Polyphasic taxonomy, antimicrobial potential and chemical profiling. BMC Microbiol 19:49. References https://doi.org/10.1186/s12866-019-1414-x Gebreyohannes G, Moges F, Sahile S, Raja N (2013) Isolation and characterization of potential antibiotic producing actinomycetes from water and sediments of Lake Tana, Ethiopia. Asian Pac J Trop Biomed 3:426–435. https://doi.org/10.1016/S2221-1691(13)60092-1 Hirsch CF, Christensen DL (1983) Novel method for selective isolation of actinomycetes. Appl Environ Microbiol 46:925–929. https://doi.org/10.1128/aem.46.4.925-929.1983 Hong-Thao PT, Mai-Linh NV, Hong-Lien NT, Van Hieu N (2016) Biological Characteristics and Antimicrobial Activity of Endophytic Streptomyces sp. TQR12-4 Isolated from Elite Citrus nobilis Cultivar Ham Yen of Vietnam. Int J Microbiol 2016. https://doi.org/10.1155/2016/7207818 Janus D, Hortschansky P, Kück U (2008) Identification of a minimal cre1 promoter sequence promoting glucose-dependent gene expression in the β- lactam producer Acremonium chrysogenum. Curr Genet 53:35–48. https://doi.org/10.1007/s00294-007-0164-8 Janus D, Hortschansky P, Kück U (2008) Identification of a minimal cre1 promoter sequence promoting glucose-dependent gene expression in the β- lactam producer Acremonium chrysogenum. Curr Genet 53:35–48. https://doi.org/10.1007/s00294-007-0164-8 Jung MR, David Horgen F, Orski S V, et al (2018) Validation of ATR FT-IR to identify polymers of plastic marine debris, including those ingested by marine organisms. Mar Pollut Bull. 127:704-716. https://doi.org/10.1016/j.marpolbul.2017.12.061 Jung MR, David Horgen F, Orski S V, et al (2018) Validation of ATR FT-IR to identify polymers of plastic marine debris, including those ingested by marine organisms. Mar Pollut Bull. 127:704-716. https://doi.org/10.1016/j.marpolbul.2017.12.061 Kamjam M, Nopnakorn P, Zhang L, et al (2019) Streptomyces polaris sp. nov. and Streptomyces septentrionalis sp. nov., isolated from frozen soil. Antonie van Leeuwenhoek, Int J Gen Mol Microbiol 112:375–387. https://doi.org/10.1007/s10482-018-1166-x Kamjam M, Nopnakorn P, Zhang L, et al (2019) Streptomyces polaris sp. nov. and Streptomyces septentrionalis sp. nov., isolated from frozen soil. Antonie van Leeuwenhoek, Int J Gen Mol Microbiol 112:375–387. https://doi.org/10.1007/s10482-018-1166-x Kapanen A, Stephen JR, Brüggemann J, et al (2007) Diethyl phthalate in compost: Ecotoxicological effects and response of the microbial community. Chemosphere 67:2201–2209. https://doi.org/10.1016/j.chemosphere.2006.12.023 Kapanen A, Stephen JR, Brüggemann J, et al (2007) Diethyl phthalate in compost: Ecotoxicological effects and response of the microbial community. Chemosphere 67:2201–2209. https://doi.org/10.1016/j.chemosphere.2006.12.023 Kohanski MA, Dwyer DJ, Collins JJ (2010) How antibiotics kill bacteria: From targets to networks. Nat. Rev. Microbiol. 8:423–435. https://doi: 10.1038/nrmicro2333 Kohanski MA, Dwyer DJ, Collins JJ (2010) How antibiotics kill bacteria: From targets to networks. Nat. Rev. Microbiol. 8:423–435. https://doi: 10.1038/nrmicro2333 Kumar A, Paul S, Kumari P, et al (2014a) Antibacterial and phytochemical assessment on various extracts of ipomoea pes-caprae (l.) R. Brthrough ftirand gc- ms spectroscopic analysis. Asian Journal of Pharmaceutical and Clinical Research 7:3. References Kumar PS, Duraipandiyan V, Ignacimuthu S (2014b) Isolation, screening and partial purification of antimicrobial antibiotics from soil Streptomyces sp. SCA 7. Kaohsiung J Med Sci 30:435–446. https://doi.org/10.1016/j.kjms.2014.05.006 Kumar A, Paul S, Kumari P, et al (2014a) Antibacterial and phytochemical assessment on various extracts of ipomoea pes-caprae (l.) R. Brthrough ftirand gc- ms spectroscopic analysis. Asian Journal of Pharmaceutical and Clinical Research 7:3. Kumar PS, Duraipandiyan V, Ignacimuthu S (2014b) Isolation, screening and partial purification of antimicrobial antibiotics from soil Streptomyces sp. SCA 7. Kaohsiung J Med Sci 30:435–446. https://doi.org/10.1016/j.kjms.2014.05.006 Kumari N, Menghani E, Mithal R (2019) GCMS analysis of compounds extracted from actinomycetes AIA6 isolates and study of its antimicrobial efficacy. Indian Journal of Chemical Technology . 26: 362-370. Lysenkova LN, Turchin KF, Danilenko VN, et al (2010) The first examples of chemical modification of oligomycin A. J Antibiot (Tokyo) 63:17–22. https://doi.org/10.1038/ja.2009.112 Kumari N, Menghani E, Mithal R (2019) GCMS analysis of compounds extracted from actinomycetes AIA6 isolates and study of its antimicrobial efficacy. Indian Journal of Chemical Technology . 26: 362-370. Lysenkova LN, Turchin KF, Danilenko VN, et al (2010) The first examples of chemical modification of oligomycin A. J Antibiot (Tokyo) 63:17–22. https://doi.org/10.1038/ja.2009.112 Magaldi S, Mata-Essayag S, Hartung De Capriles C, et al (2004) Well diffusion for antifungal susceptibility testing. Int J Infect Dis 8:39–45. https://doi.org/10.1016/j.ijid.2003.03.002 artung De Capriles C, et al (2004) Well diffusion for antifungal susceptibility testing. Int J Infect Dis 8:39–45. 003 03 002 Page 10/17 Page 10/17 Mathur N, Paliwal A, Mathur N, et al (2015) Characterization of antimicrobial compounds from Streptomyces isolates. J Chem Pharm Res 7(4):1–10. Murray LAM, McKinnie SMK, Moore BS, George JH (2020) Meroterpenoid natural products from: Streptomyces bacteria-the evolution of chemoenzymatic syntheses. Nat. Prod. Rep. 37:1334–1366 Musa Z, Ma J, Egamberdieva D, et al (2020) Diversity and Antimicrobial Potential of Cultivable Endophytic Actinobacteria Associated With the Medicinal Plant Thymus roseus. Front Microbiol 11:191. https://doi.org/10.3389/fmicb.2020.00191 Nain-Perez A, Barbosa LCA, Maltha CRA, Forlani G (2016) First total synthesis and phytotoxic activity of Streptomyces sp. metabolites abenquines. Tetrahedron Lett 57:1811–1814. https://doi.org/10.1016/j.tetlet.2016.03.038 Naureen H, Tala MF, Shaaban KA, et al (2017) A new furan carboxamide and two potential precursors from a terrestrial streptomycete. Z. Naturforsch- Sect B J Chem Sci 72:175–182. https://doi.org/10.1515/znb-2016-0202 Nepali K, Kumar S, Huang HL, et al (2016) 2-Aroylquinoline-5,8-diones as potent anticancer agents displaying tubulin and heat shock protein 90 (HSP90) inhibition. Org Biomol Chem 14:716–723. References https://doi.org/10.1039/c5ob02100f Nguyen TM, Kim J (2015) Antifungal and antibacterial activities of Streptomyces polymachus sp. nov. isolated from soil. Int J Syst Evol Microbiol 65:2385–2390. https://doi.org/10.1099/ijs.0.000268 Núñez-Montero K, Lamilla C, Abanto M, et al (2019) Antarctic Streptomyces fildesensis So13.3 strain as a promising source for antimicrobials discovery. Sci Rep 9:1–13. https://doi.org/10.1038/s41598-019-43960-7 Qin S, Li J, Chen HH, et al (2009) Isolation, diversity, and antimicrobial activity of rare actinobacteria from medicinal plants of tropical rain forests in Xishuangbanna China. Appl Environ Microbiol 75:6176–6186. https://doi.org/10.1128/AEM.01034-09 Qin S, Li J, Chen HH, et al (2009) Isolation, diversity, and antimicrobial activity of rare actinobacteria from medicinal plants of tropical rain forests in Xishuangbanna China. Appl Environ Microbiol 75:6176–6186. https://doi.org/10.1128/AEM.01034-09 Rex MA, Stuart CT, Etter RJ (2001) Do deep-sea nematodes show a positive latitudinal gradient of species diversity? The potential role of depth. Mar. Ecol. Prog. Ser. 210:297–298 Rex MA, Stuart CT, Etter RJ (2001) Do deep-sea nematodes show a positive latitudinal gradient of species diversity? The potential role of depth. Mar. Ecol. Prog. Ser. 210:297–298 Sanghvi G V., Ghevariya D, Gosai S, et al (2014) Isolation and partial purification of erythromycin from alkaliphilic Streptomyces werraensis isolated from Rajkot, India. Biotechnol Reports 1–2:2–7. https://doi.org/10.1016/j.btre.2014.05.003 Sanghvi G V., Ghevariya D, Gosai S, et al (2014) Isolation and partial purification of erythromycin from alkaliphilic Streptomyces werraensis isolated from Rajkot, India. Biotechnol Reports 1–2:2–7. https://doi.org/10.1016/j.btre.2014.05.003 Sanjenbam P, Kannabiran K (2016) Bioactivity of Pyrrolo[1,2-a] pyrazine-1,4-dione, hexahydro- Sanjenbam P, Kannabiran K (2016) Bioactivity of Pyrrolo[1,2-a] pyrazine-1,4-dione, hexahydro- 3-(phenylmethyl)-Extracted from Streptomyces sp. VITPK9 Isolated from the Salt Spring Habitat of Manipur, India. Asian Journal of Pharmaceutics. 10: 4-12. http://dx.doi.org/10.22377/ajp.v10i04.865. Schöller CEG, Gürtler H, Pedersen R, et al (2002) Volatile metabolites from actinomycetes. J Agric Food Chem 50:2615–2621. https://doi.org/10.1021/jf0116754 3-(phenylmethyl)-Extracted from Streptomyces sp. VITPK9 Isolated from the Salt Spring Habitat of Manipur, India. Asian Journal of Pharmaceutics. 10: 4-12. http://dx.doi.org/10.22377/ajp.v10i04.865. Schöller CEG, Gürtler H, Pedersen R, et al (2002) Volatile metabolites from actinomycetes. J Agric Food Chem 50:2615–2621. https://doi.org/10.1021/jf0116754 Selvakumar J, Chandrasekaran S, Vaithilingam M (2015) Bio prospecting of marine-derived Streptomyces spectabilis VITJS10 and exploring its cytotoxicity against human liver cancer cell lines. Pharmacogn Mag 11:469. https://doi.org/10.4103/0973-1296.168974 Selvakumar J, Chandrasekaran S, Vaithilingam M (2015) Bio prospecting of marine-derived Streptomyces spectabilis VITJS10 and exploring its cytotoxicity against human liver cancer cell lines. Pharmacogn Mag 11:469. References https://doi.org/10.4103/0973-1296.168974 Ser HL, Mutalib NSA, Yin WF, et al (2015a) Evaluation of antioxidative and cytotoxic activities of Streptomyces pluripotens MUSC 137 isolated from mangrove soil in Malaysia. Front Microbiol 6:. https://doi.org/10.3389/fmicb.2015.01398 Ser HL, Mutalib NSA, Yin WF, et al (2015a) Evaluation of antioxidative and cytotoxic activities of Streptomyces pluripotens MUSC 137 isolated from mangrove soil in Malaysia. Front Microbiol 6:. https://doi.org/10.3389/fmicb.2015.01398 Ser HL, Palanisamy UD, Yin WF, et al (2015b) Presence of antioxidative agent, Pyrrolo[1,2-a]pyrazine-1,4-dione, hexahydro- in newly isolated Streptomyces mangrovisoli sp. nov. Front Microbiol 6:854. https://doi.org/10.3389/fmicb.2015.00854 Ser HL, Palanisamy UD, Yin WF, et al (2015b) Presence of antioxidative agent, Pyrrolo[1,2-a]pyrazine-1,4-dione, hexahydro- in newly isolated Streptomyces mangrovisoli sp. nov. Front Microbiol 6:854. https://doi.org/10.3389/fmicb.2015.00854 Shetty PR, Buddana SK, Tatipamula VB, et al (2014) Production of polypeptide antibiotic from Streptomyces parvulus and its antibacterial activity. Brazilian J Microbiol 45:303–312. https://doi.org/10.1590/S1517-83822014005000022 Shomura T, Yoshida J, Amano S, et al (1979) Studies on Actinomycetales producing antibiotics only on agar culture. I. Screening, taxonomy and morphology-productivity relationship of Streptomyces halstedii, strain SF-1993. J Antibiot (Tokyo) 32:427–435. https://doi.org/10.7164/antibiotics.32.427 Subramani, Sipkema (2019) Marine Rare Actinomycetes: A Promising Source of Structurally Diverse and Unique Novel Natural Products. Mar Drugs 17:249. https://doi.org/10.3390/md17050249 Tan HM, Cao LX, He ZF, et al (2006) Isolation of endophytic actinomycetes from different cultivars of tomato and their activities against Ralstonia solanacearum in vitro. World J Microbiol Biotechnol 22:1275–1280. https://doi.org/10.1007/s11274-006-9172-y Tan LTH, Chan KG, Pusparajah P, et al (2019) Mangrove derived Streptomyces sp. MUM265 as a potential source of antioxidant and anticolon-cancer agents. BMC Microbiol 19:38. https://doi.org/10.1186/s12866-019-1409-7 Page 11/17 Page 11/17 Tilburn J, Sarkar S, Widdick DA, et al (1995) The Aspergillus PacC zinc finger transcription factor mediates regulation of both acid- and alkaline- expressed genes by ambient pH. EMBO J 14:779–790. https://doi.org/10.1002/j.1460-2075.1995.tb07056.x Tilburn J, Sarkar S, Widdick DA, et al (1995) The Aspergillus PacC zinc finger transcription factor mediates regulation of both acid- and alkaline- expressed genes by ambient pH. EMBO J 14:779–790. https://doi.org/10.1002/j.1460-2075.1995.tb07056.x Tilburn J, Sarkar S, Widdick DA, et al (1995) The Aspergillus PacC zinc finger transcription factor mediates regulation of both acid- and alkaline- expressed genes by ambient pH. EMBO J 14:779–790. https://doi.org/10.1002/j.1460-2075.1995.tb07056.x Tudzynski B, Homann V, Feng B, Marzluf GA (1999) Isolation, characterization and disruption of the areA nitrogen regulatory gene of Gibberella fujikuroi. Mol Gen Genet 261:106–114. https://doi.org/10.1007/s004380050947 Wang C, Wang Z, Qiao X, et al (2013) Antifungal activity of volatile organic compounds from Streptomyces alboflavus TD-1. References FEMS Microbiol Lett 341:45–51. https://doi.org/10.1111/1574-6968.12088 Wang X, Huang L, Kang Z, et al (2010) Optimization of the fermentation process of actinomycete strain Hhs.015(T). J Biomed Biotechnol 2010:. https://doi.org/10.1155/2010/141876 Williams ST, Goodfellow M, Wellington EMH, et al (1983) A probability matrix for identification of some streptomycetes. J Gen Microbiol 129:1815– 1830. https://doi.org/10.1099/00221287-129-6-1815 Yagüe P, Lopez-Garcia MT, Rioseras B, et al (2012) New insights on the development of Streptomyces and their relationships with secondary metabolite production. Curr trends Microbiol 8:65–73 Yan LL, Han NN, Zhang YQ, et al (2010) Antimycin A 18 produced by an endophytic Streptomyces albidoflavus isolated from a mangrove plant. J Antibiot (Tokyo) 63:259–261. https://doi.org/10.1038/ja.2010.21 Yan LL, Han NN, Zhang YQ, et al (2010) Antimycin A 18 produced by an endophytic Streptomyces albidoflavus isolated from a mangrove plant. J Antibiot (Tokyo) 63:259–261. https://doi.org/10.1038/ja.2010.21 Yan LL, Han NN, Zhang YQ, et al (2010) Antimycin A 18 produced by an endophytic Streptomyces albidoflavus isolated from a mangrove plant. J Antibiot (Tokyo) 63:259–261. https://doi.org/10.1038/ja.2010.21 (2019) WHO \| No Time to Wait: Securing the future from drug-resistant infections. WHO (2019) WHO \| No Time to Wait: Securing the future from drug-resistant infections. WHO Bergey, D.H. and Holt, J.G. (1994) Bergey’s Manual of Determinative Bacteriology. 9th Edition, Williams & Wilkins, Baltimore, Maryland. - References - Scientific Research Publishing. https://www.scirp.org/(S(lz5mqp453edsnp55rrgjct55))/reference/ReferencesPapers.aspx?ReferenceID=2024734. Accessed 17 Feb 2020 Bergey, D.H. and Holt, J.G. (1994) Bergey’s Manual of Determinative Bacteriology. 9th Edition, Williams & Wilkins, Baltimore, Maryland. - References - Scientific Research Publishing. https://www.scirp.org/(S(lz5mqp453edsnp55rrgjct55))/reference/ReferencesPapers.aspx?ReferenceID=2024734. Accessed 17 Feb 2020 Bergey, D.H. and Holt, J.G. (1994) Bergey’s Manual of Determinative Bacteriology. 9th Edition, Williams & Wilkins, Baltimore, Maryland. - References - Scientific Research Publishing. https://www.scirp.org/(S(lz5mqp453edsnp55rrgjct55))/reference/ReferencesPapers.aspx?ReferenceID=2024734. Accessed 17 Feb 2020 Bergey, D.H. and Holt, J.G. (1994) Bergey’s Manual of Determinative Bacteriology. 9th Edition, Williams & Wilkins, Baltimore, Maryland. - References - Scientific Research Publishing. https://www.scirp.org/(S(lz5mqp453edsnp55rrgjct55))/reference/ReferencesPapers.aspx?ReferenceID=2024734. Accessed 17 Feb 2020 Figures Figure 1 Map showing the different districts of samples were a collection. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors. Figures Figure 1 Map showing the different districts of samples were a collection. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors. Page 12/17 Page 12/17 Page 13/17 Figure 2 Dual plate assays showing antimicrobial activity by VITGV01 Figure 3 Figure 2 Dual plate assays showing antimicrobial activity by VITGV01 Dual plate assays showing antimicrobial activity by VITGV01 Figure 3 Page 13/17 Showing optimization of growth parameters of VITGV01. Showing optimization of growth parameters of VITGV01. Figure 4 Colony Morphology of VITGV63 on Different agar medium: A- ISP2, B - Actinomycetes isolation agar, C - Potato Dextrose agar, D - Yeast Peptone Mannitol, E- Czapek Dox agar, F- Zobell Marine agar G - Soybean Casein digest agar, H – Yeast Peptone Dextrose agar I - Nutrient agar, J - Luria Bertani agar, K - Tryptophan agar, L - Christenson agar, M - Starch agar and N- Soy flour Mannitol. Colony Morphology of VITGV63 on Different agar medium: A- ISP2, B - Actinomycetes isolation agar, C - Potato Dextrose agar, D - Yeast Peptone Mannitol, E- Czapek Dox agar, F- Zobell Marine agar G - Soybean Casein digest agar, H – Yeast Peptone Dextrose agar I - Nutrient agar, J - Luria Bertani agar, K - Tryptophan agar, L - Christenson agar, M - Starch agar and N- Soy flour Mannitol. Figure 6 Neighbour- Joining tree based on 16S rRNA of Streptomyces spp. VITGV01 strain (The scale bar corresponds to 0.003 substitutions per nucleotide ) position) position) Figure 7 A. Performance of crude extract for primary screening appearance of clear zone showing antimicrobial activity. 7B. Secondary screening of crude extract showing clear zone in well diffusion method. 7C. Minimal inhibitory concentration zone forming at 10ug/ml shown with a pointer position) (2019) WHO \| No Time to Wait: Securing the future from drug-resistant infections. WHO Figure 4 Colony Morphology of VITGV63 on Different agar medium: A- ISP2, B - Actinomycetes isolation agar, C - Potato Dextrose agar, D - Yeast Peptone Mannitol, E- Czapek Dox agar, F- Zobell Marine agar G - Soybean Casein digest agar, H – Yeast Peptone Dextrose agar I - Nutrient agar, J - Luria Bertani agar, K - Tryptophan agar, L - Christenson agar, M - Starch agar and N- Soy flour Mannitol. g yp p g g Figure 5 Figure 5 Page 14/17 Page 14/17 Figure 7 A. Performance of crude extract for primary screening appearance of clear zone showing antimicrobial activity. 7B. Secondary screening of crude extract showing clear zone in well diffusion method. 7C. Minimal inhibitory concentration zone forming at 10ug/ml shown with a pointer A. Performance of crude extract for primary screening appearance of clear zone showing antimicrobial activity. 7B. Secondary screening of crude extract showing clear zone in well diffusion method. 7C. Minimal inhibitory concentration zone forming at 10ug/ml shown with a pointer Page 15/17 Figure 8 FT-IR Spectra of the partially purified compound showing structural Figure 9 GC MS Chromatogram of Streptomyces spp. VITGV01 –extract showing the presence of a secondary metabolite. FT-IR Spectra of the partially purified compound showing structural FT-IR Spectra of the partially purified compound showing structural FT-IR Spectra of the partially purified compound showing structura Figure 9 C MS Chromatogram of Streptomyces spp. VITGV01 –extract showing the presence of a secondary metabol GC MS Chromatogram of Streptomyces spp. VITGV01 –extract showing the presence of a secondary metabolite. Page 16/17 Page 16/17 Figure 10 Active compounds from the GC MS library of Streptomyces spp. VITGV01. 1. Isophorone; 2. 1H-Indene, 1-methylene-; 3. 2-Trifluoromethylbenzoic acid; 4. Acetic acid, 2-phenylethyl ester; 5. 2-Methoxy-4-vinylphenol; 6. Phenol, 2,4-bis(1,1-dimethylethyl)-; 7. Semustine; 8. Diethyl Phthalate; 9. Hexadecanoic acid, ethyl ester; 10. n-Hexadecanoic acid; 11. 2,4-Diamino-6-nitroquinazoline; 12. 3,5-dimethoxyphenol; 13. Tetradecyl trifluoroacetate; 14. 2,5- Dimethoxyaniline; 15. Octadecanoic acid, ethyl ester; 16. 4-Nitro-3-picoline-N-oxide; 17. Pyrrolo[1 2-a]pyrazine-1 4-dione hexahydro-3-(2-methylpropyl)-; 18. 2-hydroxy-3 5 5-trimethyl-2-cyclohexenone; 19. Pyrrolo[1,2-a] pyrazine-1,4-dione, hexahydro-3-(phenylmethyl)-; and 20. 2(3h)-oxazolethione 4 5- diphenyl-2-cyclohexenone Figure 10 Figure 10 Active compounds from the GC MS library of Streptomyces spp. VITGV01. 1. Isophorone; 2. 1H-Indene, 1-methylene-; 3. 2-Trifluoromethylbenzoic acid; 4. Acetic acid, 2-phenylethyl ester; 5. 2-Methoxy-4-vinylphenol; 6. Phenol, 2,4-bis(1,1-dimethylethyl)-; 7. Semustine; 8. Diethyl Phthalate; 9. Hexadecanoic acid, ethyl ester; 10. n-Hexadecanoic acid; 11. 2,4-Diamino-6-nitroquinazoline; 12. 3,5-dimethoxyphenol; 13. Tetradecyl trifluoroacetate; 14. 2,5- Dimethoxyaniline; 15. Octadecanoic acid, ethyl ester; 16. 4-Nitro-3-picoline-N-oxide; 17. Pyrrolo[1 2-a]pyrazine-1 4-dione hexahydro-3-(2-methylpropyl)-; 18. 2-hydroxy-3 5 5-trimethyl-2-cyclohexenone; 19. Pyrrolo[1,2-a] pyrazine-1,4-dione, hexahydro-3-(phenylmethyl)-; and 20. 2(3h)-oxazolethione 4 5- diphenyl-2-cyclohexenone Active compounds from the GC MS library of Streptomyces spp. VITGV01. 1. Isophorone; 2. 1H-Indene, 1-methylene-; 3. 2-Trifluoromethylbenzoic acid; 4. Acetic acid, 2-phenylethyl ester; 5. 2-Methoxy-4-vinylphenol; 6. Phenol, 2,4-bis(1,1-dimethylethyl)-; 7. Semustine; 8. Diethyl Phthalate; 9. Hexadecanoic acid, ethyl ester; 10. n-Hexadecanoic acid; 11. 2,4-Diamino-6-nitroquinazoline; 12. 3,5-dimethoxyphenol; 13. Tetradecyl trifluoroacetate; 14. 2,5- Dimethoxyaniline; 15. Octadecanoic acid, ethyl ester; 16. 4-Nitro-3-picoline-N-oxide; 17. Pyrrolo[1 2-a]pyrazine-1 4-dione hexahydro-3-(2-methylpropyl)-; 18. 2-hydroxy-3 5 5-trimethyl-2-cyclohexenone; 19. Pyrrolo[1,2-a] pyrazine-1,4-dione, hexahydro-3-(phenylmethyl)-; and 20. 2(3h)-oxazolethione 4 5- diphenyl-2-cyclohexenone Page 17/17 Page 17/17
https://openalex.org/W2125022888	http://irep.iium.edu.my/40043/1/Post-crisis_political_economy.pdf	English	null	Post-Crisis Political Economy: Neoliberalism or Islamic Alternative?	Research in applied economics	2,014	cc-by	6,335	Post-Crisis Political Economy: Neoliberalism or Islamic Alternative? Moniruzzaman1,* 1Department of Political Science, International Islamic University Malaysia, Jalan Gombak, Kuala Lumpur 53100, Malaysia Correspondence: Tel: 60-3-6196-6056. E-mail: mmzaman@iium.edu.my Moniruzzaman1, 1Department of Political Science, International Islamic University Malaysia, Jalan Gombak, Kuala Lumpur 53100, Malaysia *Correspondence: Tel: 60-3-6196-6056. E-mail: mmzaman@iium.edu.my Received: April 19, 2014 Accepted: May 16, 2014 Published: June 9, 2014 doi:10.5296/rae.v6i2.5503 URL: http://dx.doi.org/10.5296/rae.v6i2.5503 Research in Applied Economics ISSN 1948-5433 2014, Vol. 6, No. 2 Keywords: global financial crisis, neoliberalism, New Financial Architecture, Islamic financial system Abstract The global financial crisis that unfolded since the end of 2006 has shaken the very foundation of the Western free-market economic system. Its fundamental regulatory principle of the separation of state and market is now seriously discredited. The objective of this article is to argue that the economic principles of Islam can provide a better alternative to the free-market system. Firstly, an Islamic economic system is highly integrative where the purpose and interest of the state and the individual citizen overlap and are complimentary to each other. So, market and state are rather inseperable in Islamic political economy. Secondly, the free-market system is fundamentally oriented to the philosophy of unlimited consumption, that is – greatest number produces greatest happiness, which demands production and appropriation of resources beyond needs. But the Islamic economic philosophy puts restrictions on unnecessary consumption, thereby capping competition over resources. These two essential principles in Islamic political economy are highly interdependent on state and individual agency of a human being. Therefore, once the economic needs and purpose of the state and the individual citizen are properly enmeshed, it produces a balanced market system. Islamic political economy has moral regimes and instrumental structure for economic behavior that reinforce each other. Keywords: global financial crisis, neoliberalism, New Financial Architecture, Islamic financial system 157 www.macrothink.org/rae www.macrothink.org/rae Research in Applied Economics ISSN 1948-5433 2014, Vol. 6, No. 2 1. Introduction The recent US and global financial crisis has jolted the very foundation of the New Financial Architecture (NFA). The NFA is built on the philosophy of liberalism which originated in the UK in the eighteenth century, and became institutionalized in the post-World War II international economic system. The basic philosophy of liberalism is separation of the state and the market where the market is left alone on its natural forces of demand, supply and price that work as regulatory regimes. Any imbalance created in their relationship is to be counter-balanced by the “invisible hand” of the human nature of cooperation. After the end of the Cold War, ‘market economy’ was formalized through “Washington Consensus” which advocated policies to reduce inflation and fiscal deficits, privatization, deregulation, and open market economy. This meant the market was to be separated completely from the state. The system that emerged came to be known as the New Financial Architecture (NFA). However, the recent financial crisis has forced many staunch advocates of the free market economy to heavily intervene into the market abandoning the non-interventionist policy of NFA. Does this mean an end of the free market economy and the NFA? Does it indicate a need for fundamental rethinking of the Western free market economic ideology? This article argues that utilitarianism as the basic and fundamental philosophy of the Western market mechanism is faulty. Utilitarianism frees individuals from the state control which produces incompatibility between individual, market and state- the three fundamental institutions of political economy. The paper argues that the philosophy of Islamic political economy is far better than utilitarian liberalism because it creates a balance between these institutions which can ensure stability in political economy. 2. Theoretical Approach Institutionalism is a popular theory in social sciences. Its application is across the board including political science and economics. Ushering a path braking approach against behavioralism and rational choice, March and Olsen (1984) presented the theory of institutionalism arguing that individuals are always parts of greater institutional frameworks and they behave according to institutional rules and prescriptions. This fundamental argument has been subsequently elaborated by many in political science (March & Olsen 1989; 1994; Brunsson & Olsen, 1993; Olsen & Peters, 1996) and economics (North, 1990; Alston, Eggerston & North, 1996; Khalil, 1995). Since its origin, the theory has taken various shapes according to the approaches it has been looked at. Peters (1999) has identified at least seven approaches to institutionalism which are normative (March & Olsen, 1984), sociological (DiMaggio & Powell, 1991; Scott, 1995; Zucker, 1987), international regimes (Rittenberger, 1993), interest groups (Knoke, Pappi & Tsujinaka, 1996; Kickert, Klijn & Koopenjaan, 1997), historical (Steinmo, Thelen & Longstreth, 1992), rational choice, and empirical. This variation is indicative of the fact that the term means differently to different scholars. 158 www.macrothink.org/rae www.macrothink.org/rae Research in Applied Economics ISSN 1948-5433 2014, Vol. 6, No. 2 Research in Applied Economics ISSN 1948-5433 2014, Vol. 6, No. 2 Research in Applied Economics ISSN 1948-5433 2014, Vol. 6, No. 2 The most famous of these approaches is normative institutionalism advanced by March and Olsen in a series of studies (1984; 1989; 1996). They argue that the best way to understand political behavior is through a “logic of appropriateness” that individuals acquire through their membership in institutions. They contrast this normative logic with the “logic of consequentiality that is central to rational choice theories” (Peters 2000: 2). According to March and Olsen individuals’ behavior is shaped by normative standards of the institutions with which they are attached and tied to instead of human desire to maximize individual utilities. In such an institutional setting human behavior is reflected in collective form rather than individual unit. The institutions are social repositories of norms, rules and values which the individuals acquire through membership and inculcate through social interactions and engagements. According to March and Olsen, human behaviors take shape of ‘appropriateness’ to the overall institutional environment rather than the concern of utilitarian ‘consequentiality’ of individuals’ actions. Another approach that is relevant to analysis in the article is historical institutionalism. The argument of this approach is “that the policy and structural choices made at the inception of the institution will have a persistent influence over its behavior for the remainder of its existence” (Steinmo, Thelen & Longstreth, 1992). This approach is akin to “path dependency” theory in political science and international political economy which explains the stability and persistence of rules and policies. This theory of institutionalism can explain the fault lines in free-market capitalism displaying how the separation between state, individual and market distorts market mechanism leading to market failure. The theory of institutionalism can be used to explain both conventional and Islamic approaches to international political economy. www.macrothink.org/rae Research in Applied Economics ISSN 1948-5433 2014, Vol. 6, No. 2 February 2007, the US operation of the HSBC sacked top chiefs due to mushrooming of the sub-prime debts, and on April 2, the New Century Financial, the second largest sub-prime lender, filed for bankruptcy. By mid-2007, financial institutions began to foreclose their debt related losses causing panic for the lenders. On July 19, 2007, the U.S. Federal Reserve chairman declared that the sub-prime crisis could cost $100 billion, sending a shockwave into the money markets around the world. Giant financial institutions such as Bear Stearns, Citigroup, and Goldman Sachs began revealing their losses alerting the market about consequences of further lending. Since mid-2007, financial institutions virtually stopped lending as they were registering billions of dollars in loan defaults. This led to a situation in which there was a lack of funds available in the credit market, or the creditors were unwilling to lend anymore, making it difficult for borrowers to obtain financing. This situation was called credit crunch (Timeline, 2008; Krugman, 2009; Roubini & Stephen, 2010). 3. Genesis of the Crisis: Sub-prime Default to Credit Crunch The immediate source of the crisis is thought to have originated in the housing industry in the USA (Sowell, 2010). The US financial institutions such as commercial and investment banks, insurance companies and credit organizations lent billions of dollars in sub-prime lending to the so-called Ninja mortgages – no income, no job or assets, or the borrowers having no or little creditworthiness (Shiller, 2008). With interest rates at historic low in 2004, loans against the Ninja mortgages skyrocketed as individuals and institutions were eager to purchase overvalued assets. Forced by competitive pressures to generate higher profits, many banks heavily underpriced risks in lending. Many of these mortgages were bought and subsequently used by the ingenious financial institutions to create new complex instruments like Collateralized Debt Obligations (CDOs), which helped them hide many of the risks away from their balance sheets and from rating agencies and regulators. Financial institutions also issued Credit Default Swaps (CDS’s) to insure these CDOs (Morris, 2010). But in the same year, interest rates began to rise pushing the real estate prices down. Consequently, towards the end of 2006 the Ninja borrowers began to default on the mortgage payment, creating a cascading chain of losses for financial institutions that held CDOs and/or issued CDS’s leading to sub-prime default crisis (Foster and Fred, 2009). The lending institutions registered billions of dollars in loss. In www.macrothink.org/rae 159 4. Global Extent of the Crisis UK Northern Rock, Barclays, Alliance & Leicester, Royal Bank of Scotland Germany Hypo Real-Estate Iceland Kaupthing, Landsbanki, Glitnir Switzerland United Bank of Switzerland (UBS) Netherlands Fortis France BNP Paribas, Societe General Source: Prepared by the author from various sources. Table 1. Major countries and companies affected by the crisis 4. Global Extent of the Crisis In today’s world, financial institutions are highly interdependent. They invest in other institutions, buy up shares and own stakes in other corporations, make portfolio investment in the share and stock exchanges around the world. Therefore, a crisis in one institution has a contagious effect on others. This is how the US credit crunch affected other large economies especially in Europe and other developed countries. In the US, by November 2008, as many as 24 banks and a number of giant multinational corporations such as Merrill Lynch and Lehman Brothers in the US were liquidated, and some other institutions have been bailed out by the government. The second hard hit region is the European economies. The UK, Germany, Norway, Denmark, the Netherlands, Iceland, France, Spain, Italy and many other countries have injected billions of dollars into the market, nationalized private banks and institutions and ensured private depositors. Table 1 lists the top 23 institutions in the American and European economies that have been seriously affected. Many of the developed countries including the US, the UK, Germany, Spain, Japan and Singapore have officially entered recession. Non-Western economies have not been directly hurt by the crisis, though major stock exchanges in India, South Korea and Mexico have tumbled seriously. The aftermath of the crisis loomed a long shadow on the growth performance of the developing countries in the subsequent years (Alcorta & Nixson, 2011; Terazi & Secil, 2012). 160 www.macrothink.org/rae www.macrothink.org/rae Research in Applied Economics ISSN 1948-5433 2014, Vol. 6, No. 2 Table 1. Major countries and companies affected by the crisis Countries Companies USA HSBC (US), New Century Financial, Bear Stearns, American Home Mortgage Corp., Countrywide Financial, Merrill Lynch, Citigroup, Carlyle Capital, Wachovia, Freddie Mae and Fannie Mac. UK Northern Rock, Barclays, Alliance & Leicester, Royal Bank of Scotland Germany Hypo Real-Estate Iceland Kaupthing, Landsbanki, Glitnir Switzerland United Bank of Switzerland (UBS) Netherlands Fortis France BNP Paribas, Societe General Source: Prepared by the author from various sources. Research in Applied Economics ISSN 1948-5433 2014, Vol. 6, No. 2 ISSN 1948-5433 2014, Vol. 6, No. 2 Table 1. Major countries and companies affected by the crisis Countries Companies USA HSBC (US), New Century Financial, Bear Stearns, American Home Mortgage Corp., Countrywide Financial, Merrill Lynch, Citigroup, Carlyle Capital, Wachovia, Freddie Mae and Fannie Mac. 5. Response of the Free Market Economies to the Crisis The most interesting aspect of the crisis is the response of the Western free market economies to overcome the shock. The neoliberalist West has consistently forced the Third World countries to adopt market economy to qualify for financial support from the Western countries and international institutions such as the World Bank and IMF. Even during the devastating Asian financial crisis in the late 1990s, the Western countries, the World Bank and IMF severely criticized Malaysia’s “Mahathirism,” or its interventionist policies in the market. They made democratization and market economy as strict pre-conditions for non-Western countries to enter the World Trade Organization (WTO). However, despite their strict position on the principle of market economy, the Western governments have acted contrarily, and in the fashion of “Mahathirism” during the recent crisis (Pesek, 2008). Elliott (2011) has described the response in the follwoing terms: The winter of 2008-09 saw co-ordinated action by the newly formed G20 group of developed and developing nations in an attempt to prevent recession turning into a slump. Interest rates were cut to the bone, fiscal stimulus packages of varying sizes announced, and electronic money created through quantitative easing. At the London G20 summit on 2 April 2009, world leaders committed themselves to a $5tn (£3tn) fiscal expansion, an extra $1.1tn of resources to help the International Monetary Fund and other global institutions boost jobs and growth, and to reform of the banks. From this point, when the global economy was on the turn, international co-operation started to disintegrate as individual countries pursued their own agendas. 9 May 2010 marked the point at which the focus of concern switched from the private sector to the public sector. By the time the IMF and the European Union announced they would provide financial help to Greece, the issue was no longer the solvency of banks but the solvency of governments.” The US, the UK, Germany, Iceland and many other OECD member countries have come forward with billions of dollars in rescue packages to bail out private companies. The US www.macrothink.org/rae www.macrothink.org/rae 161 Research in Applied Economics ISSN 1948-5433 2014, Vol. 6, No. 2 Research in Applied Economics ISSN 1948-5433 2014, Vol. 6, No. 2 offered $700 and $800 billion in separate rescue packages, respectively, in September and November 2008. This is in addition to bailing out some of the largest companies such as the AIG, and Freddie Mac and Fannie Mei. In the European economies, the UK has nationalized the Royal Bank of Scotland and injected $850 billion to partially nationalize seven other banks (Perry, 2008). Iceland has completely nationalized four of its largest banks, while Germany and Benelux countries have nationalized the Hypo-Real Estate and Fortis, respectively (Capell, 2008; Wilson, 2008). The non-Western developed economies also acted similarly. For example, Japan injected more than $100 billion into the economy; Australia $2.4 billion; Hong Kong $200 million, and Taiwan $3.6 billion (Zachariahs, 2008). According to the Government Accountability Office (GAO) of the United States, the estimated total loss incurred from the crisis is $22 trillion (Financial crisis…2013). Serious debate over such a protectionist or socialist-like policy did not gain any momentum because of the fear of a bigger worldwide economic catastrophe. Even leading liberal economists such as Paul Krugman and Joseph Stiglitz have endorsed such part-nationalization path through “equity injection” saying: act now, think later. Joseph Stiglitz observed, “[W]e had become accustomed to the hypocrisy. The banks reject any suggestion they should face regulation, rebuff any move towards anti-trust measures – yet when trouble strikes, all of a sudden they demand state intervention: they must be bailed out; they are too big, too important to be allowed to fail” (Stiglitz, 2008). This appears to be a devastating blow to the free market economic system (Sorkin, 2009). Research in Applied Economics ISSN 1948-5433 2014, Vol. 6, No. 2 sophisticated and systematic with carefully craved rules and regulations. Yet despite the presence of such developed institutions and systemic mechanism, the market failed dramatically to the extent of historic Great Depression. The question is why? Observers believe that there are a number of reasons that have corrupted the market mechanism. Some believe that the entire NFA is deeply flowed as a system. Crotty (2008: 3) has argued that there are five fundamental flows in the system, which are: “1) the theoretical foundation of the NFA – the theory of efficient capital markets – is very weak and the celebratory narrative of the NFA accepted by regulators is seriously misleading; 2) widespread perverse incentives embedded in the NFA generated excessive risk-taking throughout financial markets; 3) mortgage-backed securities central to the boom were so complex and nontransparent that they could not possibly be priced correctly; their prices were bound to collapse once the excessive optimism of the boom faded; 4) contrary to the narrative, excessive risk built up in giant banks during the boom; and 5) the NFA generated high leverage and high systemic risk, with channels of contagion that transmitted problems in the US subprime mortgage market around the world.” There are a number of points made very clear in these reasons. First, the current financial crisis is a slap of the separation of state and market where regulatory independence of the market has created a greedy profiteering tendency in the financial institutions (‘dealers’ in Adam Smith’s term) that wanted to make unscrupulous profits (Gorton 2010). These institutions manipulated the market-play through unsound speculation, overvaluation of properties, and untraceable hedge funds in the absence of proper regulatory regimes. These are driven by the utilitarian ethics of unlimited accumulation of wealth and profit to insatiable demand. It is in the sense that the interest of the institutions overtaken that of the market or public (Friedman, 2012; Allison, 2012; FCIC 2011). Second, the natural human greed for unlimited wealth is intoxicated by the opportunity of easy credit and the prospect for infatuating profit. The fundamental market ethics being utilitarianism, the individuals and market believed in the philosophy of unlimited consumption and spending. The real estate boom in the USA was a direct consequence of such individual greed of consumption. 6. What is Wrong with the Market Mechanism? The modern neoliberal economics is based on the classical liberalism of Adam Smith, David Ricardo and others. However, the neoliberal philosophy is very much contaminated by the moral danger that the classical liberalists warned against. For example, Adam Smith (1999[1776]:88) cautioned that: The interest of the dealers… in any particular branch of trade or manufactures, is always in some respects different from, and even opposite to, that of the public. To widen the market and to narrow the competition is always the interest of the dealers. To widen the market may frequently be agreeable enough to the interest of the public; but to narrow the competition must always be against it, and can serve only to enable the dealers, by raising their profits above what they naturally would be, to levy, for their own benefit, an absurd tax upon the rest of their fellow-citizens… It comes from an order of men whose interest is never exactly the same with that of the public, who have generally an interest to deceive and even to oppress the public, and who accordingly have, upon many occasions, both deceived and oppressed it. This warning refers to the moral dimension of the market economy in the absence of strong regulatory authority. The economic and financial systems are much more advanced today than any period in the past. In the same way state and financial institutions are much well-ordered, 162 www.macrothink.org/rae www.macrothink.org/rae Research in Applied Economics ISSN 1948-5433 2014, Vol. 6, No. 2 Research in Applied Economics ISSN 1948-5433 2014, Vol. 6, No. 2 players and hold them accountable for adverse effects of competition on the public played an important role in the market failure. The liberty extended to the market to play on its own separated it from state control. The financial institutions wanted to make bigger score at low transaction cost and regulatory risk of being held legally accountable for any default or breakdown. In other words, the infatuation of greed made the institutions and individual players immune from moral and legal consequences (Foster & Magdoff, 2009; Davies, 2010). In the wake of the current crisis, many suggest a reshuffling of the operating mechanism of the international economy, especially the World Bank and IMF, and look for a new paradigm of financial system (Soros, 2009). This appears to be unavoidable at this juncture. Experts have suggested various solutions to the problem emphasizing the need for greater and tighter control and surveillance mechanism. The former head of the IMF Horst Kohler suggested that: The only good thing about this crisis is that it has made clear to any thinking, responsible person in the sector that international financial markets have developed into a monster that must be put back in its place. We need more severe and efficient regulation, higher capital requirements to underpin financial trades, more transparency and a global institution to independently oversee the stability of the international financial system. (Financial Times, “Kohler attacks markets ‘monster.’” May 4 2008). Similarly, James Crotty has suggested that “radical regulatory reform adequate to its difficult task requires two substantial changes: 1) a qualitative change in the vector of political forces impinging on the ‘what should we do about reforming financial markets’ political debate that help end the domination of the lords of finance on this issue; and 2) rejection of the theory of efficient capital markets that helped create and sustain the NFA, and reliance instead on the time-tested, realistic theory of financial markets developed by Keynes and extended by Minsky. We are not likely to achieve the first unless we achieve the second” (2008: 56). Clearly, three things are suggested here – need for the political forces or states, rejection of the neorealist NFA, and bringing back the Keynesian hegemonic stability system. It was again made possible by the near no-interest credit supply from the banks and financial institutions. The greed of consumption was thus fueled by easy access to credit (Krager, 2012; Schiller, 2012, Jenkins 2012). Thirdly, the transaction of capital, properties and portfolio investments were in most cases completed against either virtual or speculative or non-existence of the real properties. Individuals and institutions traded properties on speculations while the ownership remained fluid and untraceable. This was aided, strengthened and institutionalized by sub-prime lending, CDOs and CDS. The result was that transactions were taking place, monies were circulated and consumed, yet the real existence of the properties remained non-existent (Friedman & Kraus, 2011; Kolbe 2010). And, finally, the lack of a regulatory regime that underscores the responsibility of the market 163 www.macrothink.org/rae www.macrothink.org/rae However, the fundamental questions remain that can reforms within the aura of utilitarian neoliberalism guarantee financial system’s stability? Do we need an alternative philosophy to utilitarianism in the first place for the system to function more efficiently? Research in Applied Economics ISSN 1948-5433 2014, Vol. 6, No. 2 Research in Applied Economics ISSN 1948-5433 2014, Vol. 6, No. 2 Islamic financial system is ideologically and philosophically more ethical, pragmatic and balanced; institutionally more rigorous; and socially and legally more accountable. First of all, ideologically speaking Islamic economic system does not prescribe unlimited utilitarianism and consumerism (Khan, 1994; Ahmed, 2002; Ahmed & Hoque, 2011). Trading and transaction are natural and undeniable. Profits generated through natural trading and transactions are legitimate (Wilson & Iqbal, 2005). However, Islam places ethical regulations on consumption and utility of goods and their values. In terms of consumption, its principles are – not beyond capacity (NBC), not beyond means (NBM), and not beyond needs (NBN). Such principles impose self-regulations on the economic behaviors of the individuals. These principles negate the market practice of sub-prime or NINJA lending and borrowing. Secondly, Islamic economic and financial system is systematically more rigorous. Its basic principle of trade and finance is that goods cannot be traded without having its real existence, and equally, financing for goods without existence or real and instant ownership is illegal (Clement & Wilson 2004). This places restrictions on both trading partners in terms of selling and buying goods, thereby, eliminating speculative transactions. These principles also eliminate collateral mechanisms of CDOs and CDS. Thirdly, Islamic system prohibits manipulative and unscrupulous transactions in which the sellers intentionally make profit and inflict loss on the buyers. This principle reduces the risk of default on the part of the clients (Ahmed & Khan, 2001). It also controls the behavior of short term profit-making transactions; thereby, both sides in transaction remain socially and ethically responsible to each other (Ahmed & Chapra, 2002). These fundamental mechanisms are heavily dependent on state-market-individual interdependency. None on these agents are independent of each other. Rather, all these agents have mutual controlling power on each other. Because all these agents act under the same rules and principles, they are mutually reinforcing. Firstly, the market regulatory rules are part of the same Shariah that the state is under obligation to enforce. Therefore, market is in no way separate from the state. Secondly, the individuals are to follow the same Shariah rules at personal transactions; therefore, they are bound by the same market mechanism. In this ways, state, market and individual agents are bound by the same regulatory principles. 7. An Islamic Alternative? Can Islamic finance, a system of banking rooted in the principles of Shariah (Islamic socio-legal system), be an alternative to the free-market utilitarianism? Without venturing into the details of Islamic finance here, a clear affirmative answer to the question is ideologically substantive. However, the instrumentality for an Islamic financial system to be a ready and a complete alternative to the existing system is apparently premature (Moore, 1997; Vogel, 1998). Yet, progressively the Islamic finance has been gaining ground worldwide, and developing into a viable alternative. There are a number of trends that substantiate the rise of Islamic finance as an alternative. 164 www.macrothink.org/rae All these agents act according to institutional prescriptions of the ‘logic of appropriateness’ of the Shariah rather than individual self-benefit at the cost of others. Above these principles, there is also an extra-human accountability factor that has the strongest controlling power on market mechanism. This factor is accountability to God on the part of individual as well as the state. That the individual and state authorities are separately accountable to the same God makes all the agents harmoniously follow the same principles. This ‘invisible hand’ of God is in fact more powerful than the ‘invisible hand’ of Adam Smith as a balancing force in market mechanism. Definitely, the moral and individual self-regulatory controlled economic behaviors with extra-institutional accountability factor at work, the Islamic economic and financial system has its superiority in controlling the market. The three-way accountability to God, society and state is the most powerful and balancing ‘invisible hand’ in Islamic political economy that makes it 165 www.macrothink.org/rae www.macrothink.org/rae Research in Applied Economics ISSN 1948-5433 2014, Vol. 6, No. 2 an alternative to the Western liberal or neoliberal economic ideology. It is precisely because of this logically balanced system that many of its principles are being incorporated in the conventional economic system today, especially after the devastating years of the current global financial crisis. A number of such trends can be referred to here. First, apart from its steady growth in the Muslim world, Islamic finance has been adopted by many giant institutions such as the HSBC, Standard Chartered, the UBS, Citibank, Moody’s, Goldman Sachs, ABN AMRO and Kleinwort Benson (Global Note, 1999). Second, at the government level, some of the Western countries have officially incorporated principles of Islamic finance in their public financial policies. For instance, the US Treasury Department recently announced its intention to teach “Islamic finance” to US banking regulatory agencies, Congress and other parts of the executive branch, in collaboration with Harvard University Islamic Finance Project. On November 6, 2008 the US Department of Treasury had itself offered a seminar on “Islamic Finance 101” to institutionalize the move (Schlling, 2008). In the UK, where more than 26 banks offer Islamic financial products, the British government is attempting to transform London into the Western world’s centre for Islamic finance (Quinn, 2008). And finally, an Islamic index at the Wall Street and Dow Jones Islamic Market (DJIM), with more than 75 licensees and over $800 billion in assets, indicates an alternative market operating and competing progressively at the global level. These all are indicative of a viable alternative mechanism of finance and market that can potentially create a moral economy safeguarding the public goods. However, a much more systematic and innovative development in Islamic finance is needed for it to become a global alternative. 8. Conclusion The philosophy of the free market economy lies in its separation from government intervention. The fundamental argument behind the philosophy is that the market forces can face and fix their own problem naturally. However, the credibility of free-market capitalism or the so called New Financial Architecture (NFA) of the post-Cold War has been seriously eroded by heavy government interventions all over the world, and especially in the highly industrialist free market economies, into the market in the wake of the ongoing financial crisis since 2007. This contradictory behavior indicates that the existing free market global financial and economic system has serious shortcomings. This article has found that numerous researches since 2007 have generated a consensus that uncontrolled and unlimited utilitarian philosophy of consumptionism and profiteering tendency propel the market forces to behave competitively in pursuit of unscrupulous profit. Furthermore, lack of legal, social and moral accountability puts the system at all-time risk. Therefore, the system requires a fundamental shift in its basic philosophy. Can the system be revised remaining within the philosophy? Can the systemic revision remaining with the liberal philosophy prevent such financial catastrophe in future? Skeptics are of the opinion that an entire philosophical reorientation is needed. But what that could be? 166 www.macrothink.org/rae Research in Applied Economics ISSN 1948-5433 2014, Vol. 6, No. 2 Research in Applied Economics ISSN 1948-5433 2014, Vol. 6, No. 2 Many scholars argue that the Islamic financial system can offer a solid alternative in this regard. They justify this by providing the facts that Islamic economic and financial regulatory mechanism offers a more balanced and accountable market environment in which the individual, market and state share similar goals; follow the same regulatory principles; and remain responsible to the same extra-human ‘invisible hand’ of God. A clear triple helix of legal bindings, moral imperatives and social liabilities control the economic behaviors of the agent. In this way, Islamic economics is controlled by mutual and interdependent institutional checks and balance. Furthermore, the three behavioral patterns of the Islamic economic system namely- not beyond needs, not beyond capacity and not beyond means- work as strong regulatory regimes for individual behaviors. Therefore, this article has argued that with such a balanced state-market-society interdependent mechanism, Islamic economic and financial system can make a substantial contribution to the future paradigm of international political economy. References “Global Note: Principles of Islamic Finance.” (1999). New York: Tannenbaum Helpern Syracuse & Hirschtritt LLP. Ahmed, H. (2002). Theoretical Foundations of Islamic Economics. Book of Readings No. 3. Jeddah: Islamic Research and Training Institute, Islamic Development Bank. Ahmed, H., & Chapra, M. U. (2002). Corporate Governance in Islamic Financial Institutions. Jeddah: Islamic Research and Training Institute, Islamic Development Bank. Ahmed, H., & Khan, T. (2001). Risk Management: An Analysis of Issues in Islamic Financial Industry. Jeddah: Islamic Research and Training Institute, Islamic Development Bank. Ahmed, H., & Muhammad S. H. (2011). Handbook of Islamic Economics, Vol I, Exploring the Essence of Islamic Economics. Jeddah: Islamic Research & Training Institute, Islamic Development Bank. Alcorta, L., & Nixson F. (2011). The Global Financial Crisis and the Developing World: Impact on and Implications for the Manufacturing Sector. Vienna: United Nations Industrial Development Organization. Allison, J. A. (2012). The Financial Crisis and the Free Market Cure: Why Pure Capitalism is the World Economy’s Only Hope. New York: McGraw-Hill. Alston, L. J., Eggerston T., & North D. C. (1996). Empirical Studies of Organizational Change. Cambridge: Cambridge University Press. http://dx.doi.org/10.1017/CBO9781139174633 Brunsson, N., & Olsen J. P. (1993). The Reforming Organization. London: Routledge. Capell, K. (2008). The Stunning Collapse of Iceland. Businessweek 9 October. Clement, H., & Wilson, R. (2004). The Politics of Islamic Finance. Edinburgh and New York: Edinburgh University Press and Columbia University Press. www.macrothink.org/rae 167 www.macrothink.org/rae Research in Applied Economics ISSN 1948-5433 2014, Vol. 6, No. 2 Crotty, J. (2008). “Structural Causes of the Global Financial Crisis: A Critical Assessment of the ‘New Financial Architecture.’” Working Paper, University of Massachusetts, Department of Economics, No. 2008-14. Available at http://hdl.handle.net/10419/64158. Davies, H. (2010). The Financial Crisis: Who is to Blame. Cambridge: Polity. Dimaggio, P., & Powell W. (1991). The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields. American Sociological Review, 48, 147–60. http://dx.doi.org/10.2307/2095101 Elliott, L. (2011). Global financial crisis: five key stages 2007-2011. The Guardian, August 7. Retrieved on 25 September 2013 from http://www.theguardian.com/business/2011/aug/07/global-financial-crisis-key-stages FCIC (Financial Crisis Inquiry Commission). (2011). The Financial Crisis Inquiry Report, Authorized Edition: Final Report of the National Commission on the Causes of the Financial and Economic Crisis in the United States. New York: Public Affairs. Financial Crisis Cost Tops $22 Trillion, GAO Says. (2013). Retrieved from http://www.huffingtonpost.com/2013/02/14/financial-crisis-cost-gao_n_2687553.html Foster, J. B., & Fred M. (2009). The Great Financial Crisis: Causes and Consequences. New York: Monthly Review Press. Friedman, J. (ed.). (2010). What Caused the Financial Crisis. Pennsylvania: University of Pennsylvania Press. Friedman, J., & Wladimir K. (2011). Engineering the Financial Crisis: Systemic Risk and the Failure of Regulation. Pennsylvania: University of Pennsylvania Press. Gorton, G. B. (2010). Slapped by the Invisible Hand: The Panic of 2007. USA: Oxford University Press. Jenkins, D. A., & Marlin I. C. (2012). Shadow Banking and Its Role in the Financial Crisis: Global Recession - Causes, Impacts and Remedies. New York: Nova Science Publishers. Khalil, E. L. (1995). Organizations versus Institutions. Journal of Institutional and Theoretical Economics, 151, 445–66. Khan, M. A. (1994). An Introduction to Islamic Economics. Islamabad: The International Institute of Islamic Thought. Kickert, W. J. M., Klijn E. H., & Koopenjaan J. F. M. (1997). Managing Complex Networks: Strategies for the Public Sector. London: Sage. Knoke, D., Pappi F. U., & Tsujinaka Y. (1996). Comparing Policy Networks: Labor Politics in the U.S., Germany and Japan. Cambridge: Cambridge University Press. http://dx.doi.org/10.1017/CBO9781139174497 Kolb, R. (2010). Lessons from the Financial Crisis: Causes, Consequences, and Our Economic Future. UK: Wiley. http://dx.doi.org/10.1002/9781118266588 www.macrothink.org/rae 168 Research in Applied Economics ISSN 1948-5433 2014, Vol. 6, No. 2 Krager, T. (2012). Skullduggery: The True Causes of the Financial Crisis. New York: Author House. Krugman, P. (2009). The Return of Depression Economics and the Crisis of 2008. NY: W. W. Norton & Company. March, J. G., & Olsen J. P. (1984). The New Institutionalism: Organizational Factors in Political Life. American Political Science Review, 78, 738–49. http://dx.doi.org/10.2307/1961840 March, J. G., & Olsen J. P. (1989). Rediscovering Institutions. New York: Free Press. Moore, P. (1997). Islamic Finance: A Partnership for Growth. London: Euromoney Publications. Morris, C. R. (2009). The Two Trillion Dollar Meltdown: Easy Money, High Rollers, and the Great Credit Crash. NY: PublicAffairs. North, D. C. (1990). Institutions, Institutional Change, and Economic Performance. Cambridge: Cambridge University Press. http://dx.doi.org/10.1017/CBO9780511808678 Olsen, J. P., & Peters B. G. (1996). Introduction: Learning from Experience? In Olsen, J. P., and Peters, B. G. (Eds.), Lessons from Experience: Experiential Learning from Administrative Reform in Eight Democracies. Oslo: Scandinavian University Press. Perry, B. (2012). Britain unveils $850 billion bank rescue amid market panic. Retrieved October 15, 2013 from http://uk.biz.yahoo.com/08102008/ Pesek, W. (2008). Mahathirism and the Asian Economies. The Star (Malaysia). 27 November. Peters, B. G. (1999). Institutional Theory: The “New Institutionalism” in Political Science. London: Cassells. Quinn, Ben. (2008). London Warms to Islamic Finance. Christian Science Monitor. Retrieved from www.csmonitor.com/2008 /1128/p06s02-wogn.html Rittberger, V. (1993). Regime Theory and International Relations. Oxford: Clarendon Press. Roubini, N., & Stephen M. (2010). Crisis Economics: A Crash Course in the Future of Finance. NY: Penguin Press. Schilling, C. (2008). US Treasury teaches ‘Islamic Finance 101’: advisers, scholars to promote controversial Shariah funding.’ [Online] Available: www.worldnetdaily.com/index.php?fa=PAGE.view&pageId=80003 (5 November). Scott, W. R. (1995). Institutions and Organizations. California: Sage. Shiller, R. J. (2012). The Subprime Solution: How Today's Global Financial Crisis Happened, and What to Do about It. Princeton: Princeton University Press. Smith, A. (1999 [1776]). The Wealth of Nations. Book I. London: Penguin Books. www.macrothink.org/rae 169 www.macrothink.org/rae Sowell, T. (2010). The Housing Boom and Bust. NY: Basic Books. Sowell, T. (2010). The Housing Boom and Bust. NY: Basic Books. Steinmo, S., Thelen K., & Longstreth F. (1992). Structuring Politics: Historical Institutionalism in Comparative Analysis. Cambridge: Cambridge University Press. http://dx.doi.org/10.1017/CBO9780511528125 Stiglitz, J. E. (2008). The fruit of hypocrisy; Dishonesty in the finance sector dragged us here, and Washington looks ill-equipped to guide us out. The Guardian, September 16, 2008. Stiglitz, J. E. (2010). Freefall: America, Free Markets, and the Sinking of the World Economy. NY: W. W. Norton & Company. Terazi, E., & Secil S. (2012). The Effects of the Global Financial Crisis on the Central and Eastern European Union Countries. International Journal of Business and Social Science, 2(17), 186-192. Timeline: Global Credit Crunch. (2008). Retrieved from http://news.bbc.co.uk/2/hi/business/751250.stm Vogel, F. E., & Samuel L. H. (1998). Islamic Law and Finance: Religion, Risk and Return. The Hague: Kluwer Law International. Wilson, A. (2012). Financial crisis: Benelux bank Fortis nationalized to stop collapse. Retrieved on 29 September from www.telegraph.co.uk Wilson, R., & Iqbal, M. (2005). Islamic Perspectives on Wealth Creation. New York: Edinburgh University Press and Columbia University Press. Zachariahs, C., & David Y. (2008). Asian Money Market. Retrieved on 18 September 2012 from www.bloomberg.com Zucker. L. (1987). Institutional Theories of Organizations. Annual Review of Sociology, 13, 443–64. http://dx.doi.org/10.1146/annurev.so.13.080187.002303 Copyright Disclaimer Copyright reserved by the author(s). Research in Applied Economics ISSN 1948-5433 2014, Vol. 6, No. 2 Sorkin, A. R. (2009). Too Big to Fail: The Inside Story of How Wall Street and Washington Fought to Save the Financial System. NY: Viking Adult. Soros, G. (2009). The Crash of 2008 and What it Means: The New Paradigm for Financial Markets. NY: PublicAffairs. Copyright reserved by the author(s). This article is an open-access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/). 170 www.macrothink.org/rae
https://openalex.org/W2230463286	https://europepmc.org/articles/pmc4742389?pdf=render	English	null	Parkin promotes proteasomal degradation of p62: implication of selective vulnerability of neuronal cells in the pathogenesis of Parkinson’s disease	Protein & Cell	2,016	cc-by	15,755	ABSTRACT tyrosine hydroxylase positive neurons and have worse performance in motor test when treated with 6-hydrox- ydopamine hydrochloride in aged mice. These results suggest that, in addition to their critical role in regulating autophagy, p62 are subjected to parkin mediated pro- teasomal degradation and implicate that the dysregula- tion of parkin/p62 axis may involve in the selective vulnerability of neuronal cells during the onset of PD pathogenesis. Mutations or inactivation of parkin, an E3 ubiquitin ligase, are associated with familial form or sporadic Parkinson’s disease (PD), respectively, which mani- fested with the selective vulnerability of neuronal cells in substantia nigra (SN) and striatum (STR) regions. How- ever, the underlying molecular mechanism linking par- kin with the etiology of PD remains elusive. Here we report that p62, a critical regulator for protein quality control, inclusion body formation, selective autophagy and diverse signaling pathways, is a new substrate of parkin. P62 levels were increased in the SN and STR regions, but not in other brain regions in parkin knock- out mice. Parkin directly interacts with and ubiquitinates p62 at the K13 to promote proteasomal degradation of p62 even in the absence of ATG5. Pathogenic mutations, knockdown of parkin or mutation of p62 at K13 pre- vented the degradation of p62. We further showed that parkin deﬁciency mice have pronounced loss of p y s a t t v p K s I P r o n to o B a n Electronic supplementary material The online version of this article (doi:10.1007/s13238-015-0230-9) contains supplementary material, which is available to authorized users. © The Author(s) 2015. This article is publish Mutations or inactivation of parkin, an E3 ubiquitin ligase, are associated with familial form or sporadic Parkinson’s disease (PD), respectively, which mani- fested with the selective vulnerability of neuronal cells in substantia nigra (SN) and striatum (STR) regions. How- ever, the underlying molecular mechanism linking par- kin with the etiology of PD remains elusive. Here we report that p62, a critical regulator for protein quality control, inclusion body formation, selective autophagy and diverse signaling pathways, is a new substrate of parkin. P62 levels were increased in the SN and STR regions, but not in other brain regions in parkin knock- out mice. Parkin directly interacts with and ubiquitinates p62 at the K13 to promote proteasomal degradation of p62 even in the absence of ATG5. DOI 10.1007/s13238-015-0230-9 Protein Cell 2016, 7(2): – 114 129 DOI 10.1007/s13238-015-0230-9 Protein Cell 2016, 7(2): – 114 129 Protein&Cell Protein&Cell KEYWORDS parkin, sequestosome1/p62, ubiquitin, substantia nigra KEYWORDS parkin, sequestosome1/p62, ubiquitin, substantia nigra Parkin promotes proteasomal degradation of p62: implication of selective vulnerability of neuronal cells in the pathogenesis of Parkinson’s disease Protein & Cell Pingping Song1, Shanshan Li1, Hao Wu2, Ruize Gao1, Guanhua Rao1, Dongmei Wang3, Ziheng Chen2, Biao Ma1, Hongxia Wang1, Nan Sui3, Haiteng Deng4, Zhuohua Zhang5, Tieshan Tang2, Zheng Tan2, Zehan Han6, Tieyuan Lu6&, Yushan Zhu1&, Quan Chen1,2& Protein & Cell ABSTRACT Pathogenic mutations, knockdown of parkin or mutation of p62 at K13 pre- vented the degradation of p62. We further showed that parkin deﬁciency mice have pronounced loss of KEYWORDS parkin, sequestosome1/p62, ubiquitin, substantia nigra KEYWORDS parkin, sequestosome1/p62, ubiquitin, substantia nigra 1 State Key Laboratory of Medicinal Chemical Biology, Tianjin Key Laboratory of Protein Science, College of Life Sciences, Nankai University, Tianjin 300071, China 2 State Key Laboratory of Biomembrane and Membrane Biotechnology, Institute of Zoology, Chinese Academy of Sciences, Beijing 100101, China 3 Institute of Psychology, Chinese Academy of Sciences, Beijing 100101, China 4 College of Life Sciences, Tsinghua University, Beijing 100084, China 5 State Key Laboratory of Medical Genetics, Xiangya Medical School, Central South University, Changsha 410078, China 6 Department of Health and Sports Science, Tianjin University of Sport, Tianjin 300381, China & Correspondence: tjlutieyuan@aliyun.com (T. Lu), zhuys@nankai.edu.cn (Y. Zhu), chenq@ioz.ac.cn (Q. Chen) Received October 8, 2015 Accepted October 31, 2015 INTRODUCTION In particular, mutations in parkin represent one of the major causes of early-onset familial PD (Biskup et al., 2008; Kitada et al., 1998; Lesage and Brice, 2009); It was thus proposed that mutations in parkin, an E3 ubiquitin ligase which ubiquitinates and degrades a diverse array of substrates, would cause accu- mulation and aggregation of these substrates due to B A D C STR SN 0 0.2 0.4 0.6 0.8 1.0 Relative p62 level parkin P = 0.001 P = 0.0001 +/+ +/+ -/- -/- β-Actin p62 parkin Mr (K) 62 45 +/+ +/+ -/- -/- STR SN - - 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 SN STR HIP CTX CB Relative mRNA level P = 0.02 P = 0.12 P = 0.01 P = 0.67 P = 0.05 parkin+/+ parkin-/- β-Actin TIMM 23 VDAC Mfn2 Mfn1 Drp1 p62 Hypoxia 0 7 0 7 0 7 0 7 0 7 0 7 0 7 0 7 0 7 0 7 d parkin +/+ +/+ -/- -/- +/+ +/+ -/- -/- +/+ +/+ -/- -/- +/+ +/+ -/- -/- +/+ +/+ -/- -/- STR SN COX4 HIP CTX CB Mr (K) 62 84 90 90 35 23 17 45 Figure 1. P62 level is negatively correlated with parkin activity in vivo. (A) P62 is reduced in STR and SN of parkin+/+, but not parkin−/−mice under hypoxic stress. parkin+/+and parkin−/−mice, 18-month-old male C57Bl/6, were treated with 8% oxygen conditions for 0 (control) or 7 days. The striatum (STR), substantia nigra (SN), Hippocampus (HIP), frontal cortex (CTX) and cerebellum (CB) regions were isolated and homogenized in lysis buffer. Western blotting was performed to examine the level of indicated proteins. (B) P62 level increased in parkin−/−mice in STR and SN regions. The STR and SN regions from 8-week-old male C57Bl/6 mice brain were further isolated and homogenized in lysis buffer, and Western blotting was performed to examine the level of p62 (A). (C) Relative protein levels of p62 of individual mice in (1B) were quantiﬁed according to the results of ten independent blots and normalized to β-actin. (D) The mRNA levels detected by qPCR in mice tissues were described in Fig. 1B and 1D. The intensity of bands was measured with Image J software in B, mean ± SEM, from 3 independent experiments, one-way ANOVA, the P-value were indicated ﬁgures. INTRODUCTION Parkinson’s disease (PD) is one of the most common neu- rodegenerative diseases affecting over 2% of the population over 65 years of age. The selective loss of dopaminergic neurons that project from the midbrain substantia nigra (SN) to the striatum (STR) could account for the movement dis- order symptom in PD (Ishikawa and Tsuji, 1996; Thomas and Beal, 2007). Sporadic PD or “classical parkinsonism” accounts for the majority of the disease and is multisystem neurodegenerative disorders, morphologically characterized Electronic supplementary material The online version of this article (doi:10.1007/s13238-015-0230-9) contains supplementary material, which is available to authorized users. © The Author(s) 2015. This article is published with open access at Springerlink.com and journal.hep.com.cn RESEARCH ARTICLE Parkin promotes proteasomal degradation of p62 by Lewy bodies (LBs) formation. The formation of LBs includes the stepwise condensation to ubiquitinated dense ﬁlamentous inclusions with incorporation of alpha-synuclein (Singleton et al., 2003; Spillantini et al., 1997) or p62 (Nakaso et al., 2004), which ultimately invoke the death and disappearance of the involved neurons, and ubiquitination seems to increase aggregation and neurotoxicity of alpha- synuclein in cultured human dopaminergic cells (Lee et al., 2008; Rott et al., 2008). A number of genes have been identiﬁed, and investigation of the underlying mechanisms of how these genes function has provided tremendous insights into the pathogenesis of both familial and sporadic PD (Bossy-Wetzel et al., 2004; Dawson, 2007; Dawson and Dawson, 2003; Farrer, 2006). INTRODUCTION insufﬁcient E3 ligase activity for ubiquitin-proteasomal dependent protein turnover (Kahle and Haass, 2004; Li and Guo, 2009; Sriram et al., 2005). However, a few of known substrates were found to be accumulated in parkin deﬁcient mice brain or in disease phenotypes, and the molecular link of how mutation of parkin leads to the etiology of PD remains elusive. Recent studies have revealed that parkin plays general role for mitochondrial motility and mitochondrial quality (Bingol et al., 2014; Gegg et al., 2010; Matsuda et al., 2010; Narendra et al., 2008; Tanaka et al., 2010) through modulating the stability of Miro (Wang et al., 2011b) and many other mitochondrial proteins (Chen and Dorn, 2013; Gegg and Schapira, 2011). It is thus proposed that mutation of parkin could results in mitochondrial dysfunction, which may causally link with the pathogenesis of PD. However, knockout of parkin in mice could not faithfully recapitulate the PD phenotype, raising the question of the physiological function and the pathologic role of parkin in PD (Dawson and Dawson, 2010; Johnson et al., 2012; Shin et al., 2011). P62, also known as sequestosome 1, is a shuttle protein transporting polyubiquitinated proteins for both the protea- somal and autophagy/lysosomal dependent degradation (Komatsu et al., 2007; Pankiv et al., 2007; Seibenhener et al., 2004; Wooten et al., 2008). P62 and ubiquitinated proteins are conserved markers of neuronal aging, aggre- gate formation and progressive autophagic defects (Bartlett et al., 2011). In particular, p62 was commonly detected in ubiquitinated protein aggregates in neuronal diseases including LBs in PD, neuroﬁbrillary tangles in Alzheimer’s disease, Huntington aggregates in Huntington’s disease, and skein-like inclusions in amyotrophic lateral sclerosis (Lowe et al., 1988; Rue et al., 2013; Seibenhener et al., Protein & Cell Protein & Cell Vector control parkin-Myc DMSO parkin-Myc MG132 Myc p62 Merge 1 0.7 0.5 1 1.6 1.9 β-Actin parkin p62 Scramble No. 1 No. 2 parkin KD Mr (K) -62 -55 -45 Mr (K) 1 1.1 0.4 2.1 2.0 1.4 1.9 1.0 Mock GFP DMSO MG132 BA1 3-MA Chloro. INTRODUCTION PMSF β-Actin GFP p62 GFP-parkin -62 -77 -45 -26 β-Actin Mr (K) 1 0.5 1.1 1.0 1.0 0.9 GFP parkin K161N T240R R275W C431F p62 GFP -62 -77 -45 -26 A B C D E F - - - - - - - - - - - + + + + + + + Mr (K) β-Actin GFP-parkin 1 1.1 0.5 1.4 1.4 1.3 p62 GFP MG132 -62 -77 -45 -26 GFP Vector control parkin-Myc DMSO parkin-Myc MG132 Cells with low level of p62 (%) 0 20 40 60 80 100 120 P < 0.0001 C - - - - - - - - - - - + + + + + + + Mr (K) β-Actin GFP-parkin 1 1.1 0.5 1.4 1.4 1.3 p62 GFP MG132 -62 -77 -45 -26 GFP Myc p62 Mr (K) 1 1.1 0.4 2.1 2.0 1.4 1.9 1.0 Mock GFP DMSO MG132 BA1 3-MA Chloro. PMSF β-Actin GFP p62 GFP-parkin -62 -77 -45 -26 F B E 1 0.7 0.5 1 1.6 1.9 β-Actin parkin p62 Scramble No. 1 No. 2 parkin KD Mr (K) -62 -55 -45 β A C B B β-Actin Mr (K) 1 0.5 1.1 1.0 1.0 0.9 GFP parkin K161N T240R R275W C431F p62 GFP -62 -77 -45 -26 D F Vector control parkin-Myc DMSO parkin-Myc MG132 Cells with low level of p62 (%) 0 20 40 60 80 100 120 P < 0.0001 F ) E D E Figure 2. Parkin mediates the degradation of p62 by both proteasomal and autophagic pathway. (A) Knockdown of parkin increases p62 levels. Immunoblot analysis of parkin and p62 protein levels in SH-SY5Y cells transfected with shRNA speciﬁcally targeting two different regions of parkin mRNA. (B) Parkin-mediated p62 degradation can be inhibited by both proteasomal and autophagic inhibitors. SH-SY5Y cells were transfected with GFP or GFP-parkin for 24 h before treatment with inhibitors: MG132 (5 μmol/L), Baﬁlomycin A1 (20 nmol/L), 3-MA (10 mmol/L), Chloroquine (100 μmol/L), and PMSF (100 μmol/L) using DMSO as vehicle control. Cells were then harvested and immunoblotted with anti-p62 and anti-GFP antibodies. β-Actin was used as a loading control in the Western blotting analysis. (C) MG132 inhibited parkin induced p62 degradation. Cells were treated and analyzed as that in Fig. 2B in the presence or absence of 5 μmol/L MG132. (D) Wild-type, but not disease causing parkin mutants, reduces p62 levels. INTRODUCTION B A D β-Actin TIMM 23 VDAC Mfn2 Mfn1 Drp1 p62 Hypoxia 0 7 0 7 0 7 0 7 0 7 0 7 0 7 0 7 0 7 0 7 d parkin +/+ +/+ -/- -/- +/+ +/+ -/- -/- +/+ +/+ -/- -/- +/+ +/+ -/- -/- +/+ +/+ -/- -/- STR SN COX4 HIP CTX CB Mr (K) 62 84 90 90 35 23 17 45 A Protein & Cell Protein & Cell D B C STR SN 0 0.2 0.4 0.6 0.8 1.0 Relative p62 level parkin P = 0.001 P = 0.0001 +/+ +/+ -/- -/- β-Actin p62 parkin Mr (K) 62 45 +/+ +/+ -/- -/- STR SN - - D 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 SN STR HIP CTX CB Relative mRNA level P = 0.02 P = 0.12 P = 0.01 P = 0.67 P = 0.05 parkin+/+ parkin-/- C STR SN 0 0.2 0.4 0.6 0.8 1.0 Relative p62 level parkin P = 0.001 P = 0.0001 +/+ +/+ -/- -/- Figure 1. P62 level is negatively correlated with parkin activity in vivo. (A) P62 is reduced in STR and SN of parkin+/+, but not parkin−/−mice under hypoxic stress. parkin+/+and parkin−/−mice, 18-month-old male C57Bl/6, were treated with 8% oxygen conditions for 0 (control) or 7 days. The striatum (STR), substantia nigra (SN), Hippocampus (HIP), frontal cortex (CTX) and cerebellum (CB) regions were isolated and homogenized in lysis buffer. Western blotting was performed to examine the level of indicated proteins. (B) P62 level increased in parkin−/−mice in STR and SN regions. The STR and SN regions from 8-week-old male C57Bl/6 mice brain were further isolated and homogenized in lysis buffer, and Western blotting was performed to examine the level of p62 (A). (C) Relative protein levels of p62 of individual mice in (1B) were quantiﬁed according to the results of ten independent blots and normalized to β-actin. (D) The mRNA levels detected by qPCR in mice tissues were described in Fig. 1B and 1D. The intensity of bands was measured with Image J software in B, mean ± SEM, from 3 independent experiments, one-way ANOVA, the P-value were indicated ﬁgures. © The Author(s) 2015. This article is published with open access at Springerlink.com and journal.hep.com.cn 115 RESEARCH ARTICLE Pingping Song et al. INTRODUCTION PMSF β-Actin GFP p62 GFP-parkin -62 -77 -45 -26 β-Actin Mr (K) 1 0.5 1.1 1.0 1.0 0.9 GFP parkin K161N T240R R275W C431F p62 GFP -62 -77 -45 -26 A B C D E F - - - - - - - - - - - + + + + + + + Mr (K) β-Actin GFP-parkin 1 1.1 0.5 1.4 1.4 1.3 p62 GFP MG132 -62 -77 -45 -26 GFP Vector control parkin-Myc DMSO parkin-Myc MG132 Cells with low level of p62 (%) 0 20 40 60 80 100 120 P < 0.0001 Figure 2. Parkin mediates the degradation of p62 by both proteasomal and autophagic pathway. (A) Knockdown of parkin increases p62 levels. Immunoblot analysis of parkin and p62 protein levels in SH-SY5Y cells transfected with shRNA speciﬁcally targeting two different regions of parkin mRNA. (B) Parkin-mediated p62 degradation can be inhibited by both proteasomal and autophagic inhibitors. SH-SY5Y cells were transfected with GFP or GFP-parkin for 24 h before treatment with inhibitors: MG132 (5 μmol/L), Baﬁlomycin A1 (20 nmol/L), 3-MA (10 mmol/L), Chloroquine (100 μmol/L), and PMSF (100 μmol/L) using DMSO as vehicle control. Cells were then harvested and immunoblotted with anti-p62 and anti-GFP antibodies. β-Actin was used as a loading control in the Western blotting analysis. (C) MG132 inhibited parkin induced p62 degradation. Cells were treated and analyzed as that in Fig. 2B in the presence or absence of 5 μmol/L MG132. (D) Wild-type, but not disease causing parkin mutants, reduces p62 levels. Cells were transfected with indicated plasmids for 24 h, then cell lysates were subjected to Western blotting by the anti-p62 and anti- GFP antibodies. (E) Immunostaining of Myc or parkin-Myc (red) and p62 (green) in SH-SY5Y cells transfected with Myc or parkin-Myc for 24 h. MG132 were added 8 h before ﬁxed for assay the parkin and p62 protein levels. (F) Quantiﬁcation of the cells with low level of p62 from Myc or parkin-Myc expression cells as shown in Fig. 2E. Mean ± SEM; n = 100 cells from 3 independent experiments, two-way ANOVA. Vector control parkin-Myc DMSO parkin-Myc MG132 Myc p62 Merge 1 0.7 0.5 1 1.6 1.9 β-Actin parkin p62 Scramble No. 1 No. 2 parkin KD Mr (K) -62 -55 -45 Mr (K) 1 1.1 0.4 2.1 2.0 1.4 1.9 1.0 Mock GFP DMSO MG132 BA1 3-MA Chloro. INTRODUCTION Cells were transfected with indicated plasmids for 24 h, then cell lysates were subjected to Western blotting by the anti-p62 and anti- GFP antibodies. (E) Immunostaining of Myc or parkin-Myc (red) and p62 (green) in SH-SY5Y cells transfected with Myc or parkin-Myc for 24 h. MG132 were added 8 h before ﬁxed for assay the parkin and p62 protein levels. (F) Quantiﬁcation of the cells with low level of p62 from Myc or parkin-Myc expression cells as shown in Fig. 2E. Mean ± SEM; n = 100 cells from 3 independent experiments, two-way ANOVA. © The Author(s) 2015. This article is published with open access at Springerlink.com and journal.hep.com.cn 116 RESEARCH ARTICLE Parkin promotes proteasomal degradation of p62 2004; Zatloukal et al., 2002). P62 shuttles misfolded proteins to the aggresome and autophagosome (Bjorkoy et al., 2006; Kirkin et al., 2009; Pankiv et al., 2010). Mutations in p62 have been linked with the occurrence of familial and sporadic amyotrophic lateral sclerosis (Fecto et al., 2011; Rubino et al., 2012). Furthermore, knockout of the p62 protein alone leads to neuropathological lesions including the accumula- tion of hyperphosphorylated tau and neuroﬁbrillary tangles, synaptic deﬁciencies with loss of working memory and neuronal apoptosis (Babu et al., 2005; Wooten et al., 2008). It is thus crucial to maintain a homeostatic level of p62 for normal cellular functions. Dysregulation of p62 could result in the perturbation of cell signaling and accumulation of dam- aging protein aggregates, leading to neuronal loss and pathogenesis of neurodegenerative diseases. In an effort to understand whether and how parkin deﬁciency leads to the dysregulated mitochondrial dynamics and mitochondrial quality, we were interested to ﬁnd that p62 is a new substrate of parkin and p62 is selectively accumulated in dopamingeric neuronal cells in parkin deﬁcient mice. Our results showed that parkin plays a critical role for regulating p62 stability and implied that dysregulation of parkin/p62 axis could account for the selective vulnerability during pathogenesis of PD. Figure 3. Parkin mediated the degradation of p62 in Atg5−/−MEF cells. (A) Cycloheximide (CHX)-chase assay for the half-life of p62 in SH-SY5Y cells. Top panel, SH- SY5Y cells were transfected with GFP or GFP-parkin for 24 h, and then treated with CHX (100 μg/mL) for the indicated time, and Western blotting detected the indicated antibodies. Bottom panel, the level of remaining p62 at different time points was normalized to β-actin from 3 separate experiments. INTRODUCTION (B) Parkin decreases the steady- state levels of p62 in CHX-chase experiments in SH-SY5Y cells can be blocked by MG132. Top panel, SH-SY5Y cells were transfected with GFP or GFP-parkin for 24 h, and then treated with CHX (100 μg/mL) for the indicated time and 5 μmol/L MG132 for 8 h, and Western blotting detected the indicated antibodies. Bottom panel, the level of remaining p62 at different time points was normalized to β-actin and/or p62 levels at time 0 from 3 separate experiments. (C) P62 and its LIR mutant can be degraded in Atg5−/−MEF cells. Atg5−/−MEF cells were co-trans- fected with GFP or GFP-parkin and FLAG-p62 or FLAG- p62 LIR deletion mutant for 24 h. Cells were then harvested and Western blotted with anti-FLAG or anti- GFP antibodies. (D) P62 and its LIR mutant can be degraded in p62−/−MEF cells. P62−/−MEF cells were co- transfected with GFP or GFP-parkin and FLAG-p62 or FLAG-p62 LIR deletion mutant for 24 h. Cells were then harvested and Western blotted with anti-FLAG or anti-GFP antibodies. (E) CHX-chase assay for the half-life of p62 in Atg5−/−MEF cells. Top panel, Atg5−/−MEF cells were transfected with GFP or GFP-parkin for 24 h. Cells were then treated with CHX (100 μg/mL) for the indicated time, and Western blotting was performed with anti-p62 or anti- GFP antibodies. Bottom panel, the level of remaining p62 at different time points was normalized to p62 levels at time 0 from 3 separate experiments. (F) The decrease in the steady-state levels of p62 by parkin in CHX-chase exper- iments in Atg5−/−MEF cells can be blocked by MG132. Top panel, Cells was then treated with CHX (100 μg/mL) for the indicated time and 5 μmol/L MG132 for 8 h, and Western blotting was performed with anti-p62 or anti-GFP antibod- ies. Bottom panel, the level of remaining p62 at different time course was normalized to β-actin from 3 separate experiments. (The intensity of bands was measured with Image J software. mean ± SEM, from 3 independent experiments, one-way ANOVA, P < 0.05 compared with control group). Protein & Cell Protein & Cell © The Author(s) 2015. This article is published with open access at Springerlink.com and journal.hep.com.cn RESEARCH ARTICLE RESEARCH ARTICLE Pingping Song et al. P62 level is negatively correlated with parkin activity Neuronal cells in the brain are highly sensitive to oxygen for energy production and neuronal activity. We thus were interested to measure the mitochondrial protein levels could change in response to hypoxic treatments in vivo. We ﬁrst treated the mice for 7 days in 8% oxygen chamber and isolated different regions of brain, including the striatum (STR), substantia nigra (SN), hippocampus (HIP), frontal cortex (CTX) and cerebellum (CB), and then compared the protein levels of a number of mitochondrial and autophagy markers before and after hypoxic treatment. Mitochondrial proteins such as Mfn1/2, Drp1 in STR, HIP and SN were signiﬁcantly reduced in wild-type mice, while other mito- chondrial proteins such as VDAC1, TIMM23, COX4 were maintained. However, no changes of Drp1 and Mfn1/2 levels were observed in other regions including the cerebellum and the frontal cortex (Fig. 1A). Given Mfn1/2 and Drp1 are known substrates of parkin, we wondered that parkin may be involved in selectively degradation of Mfn1/2 and Drp1 in STR and SN regions and thus further compared the protein levels of a number of mitochondrial and autophagy markers in wild-type and germline parkin exon 3 knockout mice (parkin−/−) (Goldberg et al., 2003). The reduction of Mfn1/2, Drp1 protein levels in STR and SN regions were largely blocked in parkin deﬁcient mice (Fig. 1A). Interestingly, we found that p62 levels were also reduced in STR, SN and HIP regions in response to hypoxia in wild-type mice, similar to the known mitochondrial substrates of parkin, while its pro- tein levels were maintained in other brain regions and in parkin deﬁcient mice (Fig. 1A). Careful examination of p62 levels revealed that there was an increase of p62 levels in the STR and SN of parkin−/−mice brain compared to controls (Fig. 1B and 1C), while the other regions of the brain including the HIP, the CB and CTX exhibited no such increase (Fig. S1). Real-time PCR analysis showed that mRNA levels of p62 are reduced in the STR, CB and CTX, and there were no signiﬁcant changes in the SN and HIP regions (Fig. 1D). These data demonstrate that parkin is © The Author(s) 2015. This article is published with open access at Springerlink.com and journal.hep.com.cn article is published with open access at Springerlink.com and journal.hep.com.cn 117 RESEARCH ARTICLE - - - - - - - - - - - - + + + + + + + + + + + + Mr (K) β-Actin GFP-parkin GFP p62 CHX 10 8 6 4 2 0 0 2 4 6 8 10 h GFP -62 -77 -45 -26 Mr (K) - - - - - - - - - - - - + + + + + + + + + + + + 10 8 6 4 2 0 0 2 4 6 8 10 h MG132 -62 -77 -45 -26 β-Actin GFP-parkin GFP p62 CHX GFP - - - + + + - - - - - - - - + + + + - - - + + + Mr (K) 1 1 0.6 0.9 0.7 0.4 Atg5 -/- -72 -77 -45 -26 β-Actin GFP-parkin FLAG GFP FLAG-p62-ΔLIR FLAG-p62 GFP Mr (K) - - - + + + - - - + - - - - + + + + - - - + + + p62 -/- 1 0.9 0.3 0.3 0.7 0.8 -72 -77 -45 -26 β-Actin GFP-parkin FLAG-p62-ΔLIR FLAG-p62 GFP FLAG GFP Mr (K) - - - - - - - - - - - - + + + + + + + + + + + + Atg5 -/- p62 10 8 6 4 2 0 0 2 4 6 8 10 h -62 -77 -45 -26 β-Actin GFP-parkin GFP CHX GFP Mr (K) - - - - - - - - - - - - + + + + + + + + + + + + Atg5 -/- + MG132 10 8 6 4 2 0 0 2 4 6 8 10 h -62 -77 -45 -26 β-Actin GFP-parkin GFP p62 CHX GFP A B C D E F 0 0.2 0.4 0.6 0.8 1 1.2 0 2 4 6 8 10 h * Relative p62 level P = 0.037 P = 0.026 GFP GFP-parkin 0 0.2 0.4 0.6 0.8 1 1.2 1.4 0 2 4 6 8 10 h Relative p62 level * * * P = 0.0497 P = 0.0353 P = 0.0375 GFP GFP-parkin 0 0.2 0.4 0.6 0.8 1 1.2 0 2 4 6 8 10 h Relative p62 level GFP GFP-parkin Relative p62 level 0 0.2 0.4 0.6 0.8 1 1.2 1.4 0 2 4 6 8 10 h GFP GFP-parkin - - - - - - - - - - - - + + + + + + + + + + + + Mr (K) β-Actin GFP-parkin GFP p62 CHX 10 8 6 4 2 0 0 2 4 6 8 10 h GFP -62 -77 -45 -26 A C 0 0.2 0.4 0.6 0.8 1 1.2 0 2 4 6 8 10 h * * Relative p62 level P = 0.037 P = 0.026 GFP GFP-parkin Mr (K) - - - - - - - - - - - - + + + + + + + + + + + + 10 8 6 4 2 0 0 2 4 6 8 10 h MG132 -62 -77 -45 -26 β-Actin GFP-parkin GFP p62 CHX GFP B Relative p62 level 0 0.2 0.4 0.6 0.8 1 1.2 1.4 0 2 4 6 8 10 h GFP GFP-parkin B Protein & Cell Protein & Cell D C - - - + + + - - - - - - - - + + + + - - - + + + Mr (K) 1 1 0.6 0.9 0.7 0.4 Atg5 -/- -72 -77 -45 -26 β-Actin GFP-parkin FLAG GFP FLAG-p62-ΔLIR FLAG-p62 GFP C Mr (K) - - - + + + - - - + - - - - + + + + - - - + + + p62 -/- 1 0.9 0.3 0.3 0.7 0.8 -72 -77 -45 -26 β-Actin GFP-parkin FLAG-p62-ΔLIR FLAG-p62 GFP FLAG GFP D Mr (K) - - - - - - - - - - - - + + + + + + + + + + + + Atg5 -/- p62 10 8 6 4 2 0 0 2 4 6 8 10 h -62 -77 -45 -26 β-Actin GFP-parkin GFP CHX GFP E 0 0.2 0.4 0.6 0.8 1 1.2 1.4 0 2 4 6 8 10 h Relative p62 level * * * P = 0.0497 P = 0.0353 P = 0.0375 GFP GFP-parkin Mr (K) - - - - - - - - - - - - + + + + + + + + + + + + Atg5 -/- + MG132 10 8 6 4 2 0 0 2 4 6 8 10 h -62 -77 -45 -26 β-Actin GFP-parkin GFP p62 CHX GFP F 0 0.2 0.4 0.6 0.8 1 1.2 0 2 4 6 8 10 h Relative p62 level GFP GFP-parkin F E © The Author(s) 2015. Parkin regulates p62 levels via a proteasomal- dependent pathway Previous reports including ours have shown that parkin is a potent E3 ligase that mediates ubiquitination and proteaso- mal-dependent degradation of its substrates (Burchell et al., 2013; Ko et al., 2006; Sarraf et al., 2013; Wang et al., 2011a). We were thus prompted to understand if parkin promoted the proteasomal degradation of p62 in addition to its well-docu- mented autophagic degradation. We ﬁrst knocked down parkin by speciﬁc shRNA in SH-SY5Y cells, a neuroblas- toma cell line that expresses endogenous parkin, and found that there was an accumulation of p62 when parkin was knocked down (Figs. 2A and S2). Conversely, overexpres- sion of wild-type parkin signiﬁcantly reduced the levels of p62, but not in those cells that expressed the vector alone (Fig. 2B–F). The disease-causing mutations in parkin with impaired E3 ligase activity (Sriram et al., 2005) failed to induce the reduction of p62 levels (Fig. 2D), indicating that the level of p62 is dependent on the E3 ligase activity of parkin. Consistent with previous reports (Ichimura and Komatsu, 2010; Komatsu and Ichimura, 2010), we found that the autophagic inhibitors Baﬁlomycin A1 (BA1), 3-MA and Chloroquine could also inhibit the reduction of p62 when wild-type parkin was expressed. MG132, a proteasomal inhibitor, could also potently inhibit p62 reduction when parkin is ectopically expressed (Fig. 2B and 2C). Immunoﬂuorescent image analysis further conﬁrmed that overexpression of parkin could reduce the levels of p62, which can be prevented by MG132 (Fig. 2E and 2F). To further substantiate this ﬁnding, we performed a cyclohex- imide (CHX)-chase assay and found a striking decrease of the p62 half-life in cells overexpressing GFP-parkin in SH- SY5Y cells, which can be inhibited by MG132 (Fig. 3A and 3B). Collectively, these data suggest that p62 levels can be down-regulated by both the autophagic and the proteaso- mal-dependent pathway. Protein & Cell Protein & Cell parkin in Atg5−/−MEF cells, which can be inhibited by MG132 (Fig. 3E and 3F). Taken together, we conclude that parkin functions as an E3 ligase to mediate the proteasomal degradation of p62, or in other words, p62 is a novel sub- strate of parkin. To further ascertain the proteasomal-dependent degra- dation of p62 by parkin, the Agt5−/−MEF cells (Fig. S3), in which autophagic activity is abrogated, were employed to detect the p62 protein. Parkin regulates p62 levels via a proteasomal- dependent pathway We transfected Agt5−/−MEF cells with wild-type p62 or a LIR deletion mutant, which was reported to mediate its interaction with LC3 for autophagic degradation, and found that the protein level of p62 or its LIR deletion mutant were signiﬁcantly reduced when parkin is expressed in these cells (Fig. 3C). The degradation of p62 and its LIR deletion mutant was also evident when p62 and parkin were expressed in p62−/−MEF cells (Figs. 3D and S3). The CHX-chase assay further revealed a signiﬁcant decrease of the p62 half-life in cells overexpressing GFP- RESEARCH ARTICLE Parkin promotes proteasomal degradation of p62 involved in regulating mitochondrial protein levels in response to hypoxia in dopamingeric neuronal cells, and the protein levels of p62 are negatively correlated with the expression of parkin in the SN and STR regions that have selective vulnerability in PD. Figure 4. Parkin interacts with p62 in vivo and in vitro. (A) SH-SY5Y cells were harvested and the cell lysates were subjected to immunoprecipitation with an anti-p62 antibody or an IgG control, and the immunoprecipitates were examined by Western blotting using an anti-parkin antibody. (B) P62+/+ or p62−/−MEF cells were transfected with GFP-parkin for 24 h, and cell lysates were subjected to immunoprecipitation with an anti-p62 antibody or an IgG control, and the immunoprecipitates were examined by Western blotting using anti-GFP or anti-p62 antibodies. (C) In vitro translated p62 was incubated with bacterially puriﬁed MBP-parkin or MBP immobilized on MBP beads, and Western blotting was performed to detect the p62 protein by using the anti-p62 antibodies. (D) 293T cells were transfected with GFP-parkin or GFP (control), and immunoprecipitation and Western blotting were performed to examine the interaction between GFP-parkin and endogenous p62. (E) 293T cells were co-transfected with FLAG-p62 and GFP-parkin or parkin mutants, and immunoprecipitation and Western blotting were performed to examine the interaction between p62 and parkin or parkin mutants. (F) Top panel, schematic representation of various deletion mutants of GFP-parkin, including the Linker domain deletion mutant, ΔLinker; RING1 deletion mutant, ΔRING1; RING2 deletion mutant, ΔRING2; RING1 and RING2 double deletion mutant, ΔRING; Linker and double RING ﬁnger deletion mutant, ΔL-R. Bottom panel, 293T cells were co-transfected with FLAG-p62 and GFP- parkin or parkin mutants, and immunoprecipitation and Western blotting were performed to examine the interac- tion between p62 and parkin. (G) Top panel, schematic representation of various deletion forms of FLAG-p62, including PB1 domain deletion mutant, ΔPB1; LIR deletion mutant, ΔLIR; UBA deletion mutant, ΔUBA. Bottom panel, 293T cells were co-transfected with GFP-parkin and FLAG-p62 or p62 deletion mutants, and immunoprecipita- tion and Western blotting were performed to examine the interaction between p62 and parkin. RESEARCH ARTICLE This article is published with open access at Springerlink.com and journal.hep.com.cn © The Author(s) 2015. This article is published with open access at Springerlink.com and journal.hep.com.cn 118 RESEARCH ARTICLE Parkin interacts with and ubiquitinates p62 for its degradation To understand the mechanisms of parkin mediated protea- somal degradation of p62, we ﬁrst checked if these two molecules interact with each other. Co-immunoprecipitation analysis showed that endogenous p62 interacts with endogenous parkin (Fig. 4A and 4B). Pull-down analysis of recombinant MBP-parkin with recombinant p62 further showed that their interaction was direct (Fig. 4C). Ectopically 119 article is published with open access at Springerlink.com and journal.hep.com.cn © The Author(s) 2015. This article is published with open access at Springerlink.com and journal.hep.com.cn RESEARCH ARTICLE Pingping Song et al. Parkin interacts with and ubiquitinates p62 for its degradation A B C D E F G Ubl Ubl Ubl Ubl Ubl Ubl Linker Linker Linker Linker R1 R1 R1 IBR IBR IBR IBR IBR IBR R2 R2 R2 REP REP REP REP REP R0 R0 R0 R0 R0 ΔRING1 ΔRING ΔRING2 WT ΔLinker ΔL-R parkin p62 PB1 PB1 PB1 ZZ ZZ ZZ ZZ TB TB TB TB LIR LIR LIR WT ΔPB1 ΔLIR ΔUBA UBA UBA UBA p62 p62 IgG p62 1/20 Input -55 -62 -55 -62 IP: p62 parkin parkin Mr (K) IP: GFP GFP p62 GFP p62 -62 -77 -26 -62 -77 -26 GFP 1/20 Input - - + + Mr (K) GFP-parkin GFP GFP GFP ΔLinker ΔRING1 ΔRING2 ΔRING ΔL-R -55 -77 -26 -77 -62 -62 - - * * * ** Mr (K) GFP-parkin FLAG-p62 FLAG-p62 IP: FLAG 1/20 Input FLAG FLAG IP: FLAG FLAG ΔPB1 ΔLIR ΔUBA -77 -55 -62 -77 Mr (K) GFP-parkin GFP-parkin FLAG-p62 1/20 Input FLAG-p62 FLAG-p62 GFP GFP IP: FLAG GFP R42P K161N T240R R275W Q311X T415N G430D C431F -55 -77 -26 -77 -26 -72 1/20 Input Mr (K) GFP-parkin p62 p62 MBP -62 -97 + + + + + + - - - - - - Mr (K) MBP-parkin MBP-parkin p62 1/10 Input p62 IP: IgG p62 -77 -62 -77 +/+ +/+ -/-Mr (K) GFP-parkin GFP-parkin C E G FLAG-p62 FLAG-p62 GFP GFP IP: FLAG GFP R42P K161N T240R R275W Q311X T415N G430D C431F -55 -77 -26 -77 -26 -72 1/20 Input Mr (K) GFP-parkin p62 p62 MBP -62 -97 + + + + + + - - - - - - Mr (K) MBP-parkin MBP-parkin p62 1/10 Input p62 IP: IgG p62 -77 -62 -77 +/+ +/+ -/-Mr (K) FP-parkin FP-parkin B p62 1/10 Input p62 IP: IgG p62 -77 -62 -77 +/+ +/+ -/-Mr (K) GFP-parkin GFP-parkin A p62 p62 IgG p62 1/20 Input -55 -62 -55 -62 IP: p62 parkin parkin Mr (K) C p62 p62 MBP -62 -97 + + + + + + - - - - - - Mr (K) MBP-parkin MBP-parkin C D IP: GFP GFP p62 GFP p62 -62 -77 -26 -62 -77 -26 GFP 1/20 Input - - + + Mr (K) GFP-parkin E D Protein & Cell Protein & Cell 1/20 Input 1/20 Input G F Ubl Ubl Ubl Ubl Ubl Ubl Linker Linker Linker Linker R1 R1 R1 IBR IBR IBR IBR IBR IBR R2 R2 R2 REP REP REP REP REP R0 R0 R0 R0 R0 ΔRING1 ΔRING ΔRING2 WT ΔLinker ΔL-R parkin GFP GFP GFP ΔLinker ΔRING1 ΔRING2 ΔRING ΔL-R -55 -77 -26 -77 -62 -62 - - * * * ** Mr (K) GFP-parkin FLAG-p62 FLAG-p62 IP: FLAG 1/20 Input G p62 PB1 PB1 PB1 ZZ ZZ ZZ ZZ TB TB TB TB LIR LIR LIR WT ΔPB1 ΔLIR ΔUBA UBA UBA UBA FLAG FLAG IP: FLAG FLAG ΔPB1 ΔLIR ΔUBA -77 -55 -62 -77 Mr (K) GFP-parkin GFP-parkin FLAG-p62 1/20 Input Mr (K) IP: FLAG 1/20 Input 1/20 Input 120 © The Author(s) 2015. © The Author(s) 2015. This article is published with open access at Springerlink.com and journal.hep.com.cn Parkin interacts with and ubiquitinates p62 for its degradation This article is published with open access at Springerlink.com and journal.hep.com.cn RESEARCH ARTICLE Parkin promotes proteasomal degradation of p62 A HA-UB parkin HA No. 2 No. 1 Scramble -170 -55 -97 -62 -45 IP: p62 1/20 Input - - - - - - + + + + + + Mr (K) β-Actin A C B D E F G H I MBP-parkin p62 T240R K161N MBP-parkin MBP p62 UB -170 -97 -62 - - - - - - - - - - - - - - - - - - - - - - - - - + + + + + + + + + + + + + + + + + Mr (K) HA-UB parkin HA No. 2 No. Parkin interacts with and ubiquitinates p62 for its degradation 1 Scramble -170 -55 -97 -62 -45 IP: p62 1/20 Input - - - - - - + + + + + + Mr (K) β-Actin p62 p62 UB parkin -130 -62 -62 -62 -45 IP: p62 1/10 Input +/+ -/- Mr (K) β-Actin GFP HA HA-UB GFP p62 p62 ΔL-R -170 -77 -62 -62 -26 -62 IP: p62 1/20 Input - - - - - - + + + + + + Mr (K) GFP-parkin HA-UB GFP p62 HA GFP T240R R275W C431F -170 -77 -62 -62 -26 IP: p62 1/20 Input - - - - - - - - - - - - - - - - - - - - - - - - - + + + + + + + + + + + Mr (K) GFP-parkin HA-UB HA-UB-K29 HA-UB-K48 HA-UB-K63 GFP p62 HA GFP -170 -77 -62 -62 -26 IP: p62 1/20 Input - - - - - - - - - - - - - - - - - - - - - - - - - + + + + + + + + + + + Mr (K) GFP-parkin HA-UB ΔPB1 GFP 1/20 Input FLAG FLAG HA IP: FLAG ΔLIR ΔUBA FLAG-p62 -170 -77 -62 -62 -62 - - - - - - - - - - - - + + + + + + + + + + + + Mr (K) GFP-parkin FLAG-p62 K420R GFP K13R FLAG GFP 1 0.5 1.1 0.8 -62 -77 -45 -26 - - - + - - - + - - - + - + + + + - + - Mr (K) β-Actin GFP-parkin HA-UB IP: FLAG FLAG-p62 -62 HA FLAG HA GFP GFP K13R K420R -170 -77 -62 -62 -26 -170 -62 FLAG 1/20 Input - - - + + + + + - - - + - - - + - + + + + - + - Mr (K) GFP-parkin B p62 p62 UB parkin -130 -62 -62 -62 -45 IP: p62 1/10 Input +/+ -/- Mr (K) β-Actin C GFP HA HA-UB GFP p62 p62 ΔL-R -170 -77 -62 -62 -26 -62 IP: p62 1/20 Input - - - - - - + + + + + + Mr (K) GFP-parkin B 1/10 Input 1/20 Input D HA-UB GFP p62 HA GFP T240R R275W C431F -170 -77 -62 -62 -26 IP: p62 1/20 Input - - - - - - - - - - - - - - - - - - - - - - - - - + + + + + + + + + + + Mr (K) GFP-parkin F MBP-parkin p62 T240R K161N MBP-parkin MBP p62 UB -170 -97 -62 - - - - - - - - - - - - - - - - - - - - - - - - - + + + + + + + + + + + + + + + + + Mr (K) E p p HA-UB ΔPB1 GFP 1/20 Input FLAG FLAG HA IP: FLAG ΔLIR ΔUBA FLAG-p62 -170 -77 -62 -62 -62 - - - - - - - - - - - - + + + + + + + + + + + + Mr (K) GFP-parkin F T240R K161N MBP-parkin MBP p62 UB -170 - - - - - - - - - - - - - - - - - - - - - - - - - + + + + + + + + + + + + + + + + + Mr (K) F F E D Protein & Cell HA-UB GFP p62 HA -170 -77 -62 -62 -26 IP: p62 1/20 Input - + + + + + Mr (K) 1/20 Input G 1 HA-UB HA-UB-K29 HA-UB-K48 HA-UB-K63 GFP p62 HA GFP -170 -77 -62 -62 -26 IP: p62 1/20 Input - - - - - - - - - - - - - - - - - - - - - - - - - + + + + + + + + + + + Mr (K) GFP-parkin I FLAG-p62 K420R GFP K13R FLAG GFP 1 0.5 1.1 0.8 -62 -77 -45 -26 - - - + - - - + - - - + - + + + + - + - Mr (K) β-Actin GFP-parkin H G IP: FLAG © The Author(s) 2015. © The Author(s) 2015. This article is published with open access at Springerlink.com and journal.hep.com.cn RESEARCH ARTICLE RESEARCH ARTICLE Pingping Song et al. Figure 5. Wild-type parkin, but not disease causing mutants, ubiquitinates p62. (A) SH-SY5Y cells were trans- fected with HA-UB together with scramble or parkin siRNAs as indicated for 36 h. Cells were treated with 5 μmol/L MG132 for 8 h before harvest. Cell lysates were then subjected to immunoprecipitation and Western blotting to detect the ubiqui- tination of p62. (B) The midbrain of parkin+/+ or parkin−/−mice, 8-week-old male C57Bl/6, were isolated and homogenized in lysis buffer. Cell lysates were then subjected to immunoprecip- itation with anti-p62 antibody and Western blotting to detect the ubiquitination levels of p62. (C) SH-SY5Y cells were transfected with HA-UB together with GFP, GFP-parkin or ΔL-R for 36 h, and were treated with 5 μmol/L MG132 for 8 h before harvest. Cell lysates were then subjected to immunoprecipitation and Western blotting for the ubiquitination levels of p62. (D) SH- SY5Y cells were transfected with HA-UB together with GFP, GFP-parkin or parkin mutants (T240R, R275 W and C431F). Cells were treated with 5 μmol/L MG132 for 8 h before harvest. Cell lysates were then subjected to immunoprecipitation and Western blotting the ubiquitination levels of p62. (E) SH-SY5Y cells were transfected with HA-UB together with GFP-parkin and FLAG-p62 or p62 deletion mutants. Cells were treated with 5 μmol/L MG132 for 8 h before harvest. Cell lysates were then subjected to immunoprecipitation and Western blotting the ubiquitination levels of p62. (F) In vitro translated p62 protein was incubated with commercial puriﬁed ubiquitin, E1, E2 (UbcH7), and bacterially puriﬁed parkin proteins. The reaction products were analyzed by Western blotting with anti-p62 antibodies. (G) 293T cells were transfected with various HA-UB constructs (K29, K48 and K63) together with GFP-parkin or GFP vector. Cells were treated with 5 μmol/L MG132 for 8 h before harvest. Cell lysates were then subjected to immuno- precipitation and Western blotting the ubiquitination levels of p62. (H) 293T cells were transfected with HA-UB together with GFP-parkin and FLAG-p62 or mutants (K13R, K420R). Cells were treated with 5 μmol/L MG132 for 8 h before harvest. Cell lysates were then subjected to immunoprecipitation and Western blotting the ubiquitination levels of p62. (I) 293T cells were co-transfected with GFP or GFP-parkin and FLAG-p62 or K13R/K420R mutants for 24 h. Cells were then harvested and immunoblotted with anti-FLAG or anti-GFP antibodies. β-Actin was Western blotted as a loading control. RESEARCH ARTICLE b ubiquitinated by wild-type parkin, but not by known disease- causing mutants (Fig. 5C and 5D). Furthermore, a Linker- RING ﬁnger deletion mutant parkin that fails to interact with p62 is also unable to mediate the ubiquitination of p62 (Fig. 5C). Similarly, the PB1 deletion mutant of p62, which does not interact with parkin, is not ubiquitinated, but not found in either UBA or LIR domain deletion (Fig. 5E). These data suggest that p62 is an authentic substrate of parkin in both cell and animal model systems. Parkin interacts with and ubiquitinates p62 for its degradation This article is published with open access at Springerlink.com and journal.hep.com.cn 121 P62 is ubiquitinated at K13 site for proteasomal degradation Protein & Cell To directly conﬁrm that p62 is directly ubiquitinated by parkin, we carried out in vitro ubiquitination assay and found that in vitro puriﬁed parkin ubiquitinates p62 in the presence of E1, E2, ubiquitin and ATP. These data demonstrate that the ubiquitination of p62 is speciﬁcally mediated by parkin, while the disease causing mutants that have impaired E3 ligase activity fail to ubiquiniate p62 for its subsequent degradation (Fig. 5F). Immunoprecipitation analysis revealed that parkin was able to induce the poly-ubiquitination of p62 in the presence of wild-type or K48 ubiquitin, but signiﬁcantly reduced in the presence of K29 or K63 ubiquitin (Fig. 5G). This experimental result suggests that parkin mediates the poly-ubiuqitination of p62 mainly via K48-linked ubiquitin chains for proteasomal degradation, while K63 ubiquitin modiﬁcation occurs to a lesser extent (Fig. 5G). Protein & Cell Protein & Cell To further demonstrate that parkin ubiquitinates p62, we sought to determine the unique site of ubiquitination of p62 by parkin. We transfected 293T cells with HA-Ubiquitin (HA- UB), GFP-parkin and FLAG-p62 and immunoprecipitated with anti-FLAG antibody, and immunoprecipitates were fur- ther analyzed by mass spectrometry, the mass results showed that both K13 and K420 are ubiquitinated. To con- ﬁrm that these K13 and K420 residues were the sites of ubiquitination by parkin, we mutated K13 or K420 to arginine and co-transfected these mutants with GFP-parkin in 293T cells. We found that both wild-type p62 and the p62 K420R mutant, but not the p62 K13R mutant, are ubiquitinated by parkin (Fig. 5H). Importantly, we showed that the protein levels of the p62 K13R mutant, but not wild-type p62 or the p62 K420R mutant, were not reduced in the presence of wild-type parkin (Fig. 5I). expressed GFP-parkin, but not GFP itself, could interact with endogenous p62 (Fig. 4D). Although pathological mutations of parkin failed to induce the reduction of p62 levels, they still interact with p62 in cells (Fig. 4E). Domain mapping indi- cates that both the RING1 and RING2 domains, which mediate the E3 ligase activity, and the Linker domain of parkin were required for binding to p62 via its PB1 domain (Fig. 4F and 4G). Parkin regulates p62 degradation in response to 6-OHDA Parkin regulates p62 degradation in response to 6-OHDA Parkin regulates p62 degradation in response to 6-OHDA Parkin regulates p62 degradation in response to 6-OHDA And we also found the increase of p62 protein levels in parkin−/−MEF cells compared with the wild type MEF cells (Data not shown). insoluble that diminishes its activity. As 6-OHDA is widely used to induce parkinsonal phenotypes in mice, we tested the functional implication of parkin for PD after 6-OHDA treatments. Consistent with previous reports (Perez and Palmiter, 2005), rotation and slip/step analysis do not reveal PD-like phenotypes in younger mice (6 months) (Fig. S5). However, such analysis showed that parkin−/−aged mice (18 months) performed worse than that of wild-type controls upon injection of 6-OHDA (Fig. 6A). We also observed that there was pronounced loss of TH positive neurons in parkin deﬁcient mice compared to its wild-type control after injec- tions (Fig. 6B and 6C). These data suggest that parkin is of functional importance for selective loss of dopamingeric neurons and the onset of PD in aged mice. We further tested the effects of 6-OHDA on p62 and mitochondrial protein degradation in vivo. We directly injected 6-OHDA into SN region and examined the p62 protein levels and found that p62 levels are increased in wild-type mice, while the levels were maintained after the treatment with 6-OHDA (Fig. 6D and 6E). As only small amount of sample can be obtained from the SN regions of treated mice, we turned to analyze p62 degradation in response to 6-OHDA treatments in cul- tured cells. Indeed, treatment of 6-OHDA reduced the protein levels of parkin (likely due to its self-degradation) and such treatment signiﬁcantly increased the p62 levels in the insol- uble fractions in wild-type cells (Fig. 6F). And we also found the increase of p62 protein levels in parkin−/−MEF cells compared with the wild type MEF cells (Data not shown). p62 degradation in parkin dependent manner (Fig. 1A), while 6-OHDA induced parkin inactivation, p62 accumulation and aggregation and autophagic degradation (Fig. 6). The exact detail of how hypoxia activates parkin dependent p62 degra- dation requires further investigation and we observed the enhanced ubiquitination of p62 when treated with hypoxia, which is absent in parkin−/−neuronal cells. Collectively, our results suggest a new link between these two important molecules that play essential roles for protein quality control, mitochondrial dynamics and cell signaling. Both parkin and p62 were found to play role in mitophagy and/or general autophagy. Parkin regulates p62 degradation in response to 6-OHDA Consistent with previous reports, parkin deﬁcient mice did not exhibit degeneration of dopaminergic neurons (Goldberg et al., 2003; Itier et al., 2003; Perez et al., 2005; Perez and Palmiter, 2005), likely due to the lack of aging related stresses. Dopamine can covalently modify and inactivate parkin through its conjugation with cysteine (431) (Lazarou et al., 2013) at its reactive center or making it becoming We next examined whether parkin is able to ubiquitinate p62 for its subsequent degradation. Knockdown of parkin in SH-SY5Y cells by speciﬁc shRNA could signiﬁcantly reduce the ubiquitination of p62 (Fig. 5A). Also, the level of ubiqui- tinated p62 was signiﬁcantly higher in the midbrain of wild- type mice than that in parkin−/−mice (Fig. 5B). P62 is 122 © The Author(s) 2015. This article is published with open access at Springerlink.com and journal.hep.com.cn RESEARCH ARTICLE Parkin promotes proteasomal degradation of p62 insoluble that diminishes its activity. As 6-OHDA is widely used to induce parkinsonal phenotypes in mice, we tested the functional implication of parkin for PD after 6-OHDA treatments. Consistent with previous reports (Perez and Palmiter, 2005), rotation and slip/step analysis do not reveal PD-like phenotypes in younger mice (6 months) (Fig. S5). However, such analysis showed that parkin−/−aged mice (18 months) performed worse than that of wild-type controls upon injection of 6-OHDA (Fig. 6A). We also observed that there was pronounced loss of TH positive neurons in parkin deﬁcient mice compared to its wild-type control after injec- tions (Fig. 6B and 6C). These data suggest that parkin is of functional importance for selective loss of dopamingeric neurons and the onset of PD in aged mice. We further tested the effects of 6-OHDA on p62 and mitochondrial protein degradation in vivo. We directly injected 6-OHDA into SN region and examined the p62 protein levels and found that p62 levels are increased in wild-type mice, while the levels were maintained after the treatment with 6-OHDA (Fig. 6D and 6E). As only small amount of sample can be obtained from the SN regions of treated mice, we turned to analyze p62 degradation in response to 6-OHDA treatments in cul- tured cells. Indeed, treatment of 6-OHDA reduced the protein levels of parkin (likely due to its self-degradation) and such treatment signiﬁcantly increased the p62 levels in the insol- uble fractions in wild-type cells (Fig. 6F). Parkin regulates p62 degradation in response to 6-OHDA In response to the loss of mito- chondrial membrane potential, PINK1 becomes stabilized at the outer mitochondrial membrane where it can interact and phosphorylate parkin for its recruitments and activation at the site of mitochondria (Kane et al., 2014; Kazlauskaite et al., 2014; Koyano et al., 2014; Lazarou et al., 2015; Matsuda et al., 2010; Pickrell and Youle, 2015). Parkin then ubiquiti- nates a number of mitochondrial membrane proteins, both p62 and ULK1 are recruited toward mitochondria for selective cargo recognition (Li et al., 2015). For example, it is sug- gested that p62 can interact with LC3 for selective cargo recognition for mitophagy, although the exact role of p62 in parkin mediated mitophagy remains controversial. Much of the studies are carried out in the cultured cell system and physiological relevance of this critical mitophagy pathway with Parkinson’s diseases needs to be critically evaluated. Our results suggest that one of the physiological functions of parkin is to monitor the protein quality of its substrates through proteasomal degradation, while its mitophagic or autophagic role occurs under stress conditions such as pro- tein aggregation or complete loss of mitochondrial membrane potential (Sterky et al., 2011). Previous studies have shown that parkin may not translocate onto mitochondria in response to the loss of mitochondrial membrane potential in primary neuronal cells (Van Laar et al., 2011) and is dis- pensable for the progressive mitochondrial respiration deﬁ- ciency and loss of dopamine neuron caused by the loss of mtDNA (Sterky et al., 2011). It is possible that, during normal physiological situations, the parkin/p62 axis is able to keep cellular p62 levels in check for the well-being of the cell. As both parkin and p62 are sensitive to oxidative stress and the perturbation of redox signaling (Jain et al., 2010; LaVoie et al., 2007), it is possible that, when parkin is mutated or its E3 ligase activity is inhibited upon oxidative stress, there will be an increase in the level of p62. This increased level of p62 will initially be protective to the cells for the removal of mis- folded proteins and protein aggregate through enhanced protein turnover or selective autophagy. However, drastic and persistent perturbation of the parkin/p62 axis, would redeﬁne a threshold where proteins fail to degrade, neuronal signaling is impaired, and the hallmarks of PD are manifested. P62 is a multifunctional protein involved in multiple cel Protein & Cell Protein & Cell © The Author(s) 2015. This article is published with open access at Springerlink.com and journal.hep.com.cn article is published with open access at Springerlink.com and journal.hep.com.cn DISCUSSION The major ﬁndings in our current study are the identiﬁcation of p62 as a new substrate of parkin and inactivation of parkin (either by genetic manipulations or enhanced degradation) may lead to the accumulation of p62 in dopamingeric neuronal cells and the selective vulnerability of these cells during the onset of the diseases. It is known that parkin is normally kept in the inactive state and can be either activated or inactivated through post-translational modiﬁcation in response to mito- chondrial or oxidative stresses. Once activated, parkin inter- acts with and subsequently ubiquitinates p62 at the K13 residue, resulting in the degradation of p62 via the proteaso- mal-dependent pathway. We found that the degradation of p62 can be blocked by the proteasomal inhibitor-MG132 even in Atg5−/−MEF cells (Fig. 3). We clearly showed that p62 is required to be ubiquitinated by wild-type parkin at the K13 residue and mutation at this site blocked its degradation (Fig. 5). Recent work by Lee and colleagues suggested that p62 degradation could be inhibited by the proteasomal inhi- bitor MG132, which may also block the autophagic activities (Lee et al., 2012). Our results suggest that both proteasomal degradation and autophagic degradation are involved in p62 stability since BA1 and other lysosomal inhibitors also pre- vented its degradation (Fig. 2B). Their relative importance for p62 stability is likely cellular context and stress type depen- dent. For instance, mild stress such as hypoxia may promote P62 is a multifunctional protein involved in multiple cel- lular functions such as signal transduction and the degra- dation of both proteins and organelles. Accumulating protein aggregates, dysfunctional mitochondria and DNA damage 123 RESEARCH ARTICLE Pingping Song et al. B E D parkin+/+ parkin-/- Total TH+ positive Cell 0 10000 20000 30000 40000 50000 60000 CTR 6-OHDA F A C β-Actin p62 Parkin Saline (left) parkin+/+ mice parkin-/- mice 6-OHDA (right) Saline (left) 6-OHDA (right) -62 -55 -45 No. DISCUSSION (D) The mice described in A were sacriﬁced and the SN regions were isolated, then tissue lysate D C E parkin+/+ parkin-/- Total TH+ positive Cell 0 10000 20000 30000 40000 50000 60000 CTR 6-OHDA C E D parkin+/+ parkin-/- Total TH+ positive Cell 0 10000 20000 30000 40000 50000 60000 CTR 6-OHDA F C β-Actin p62 Parkin Saline (left) parkin+/+ mice parkin-/- mice 6-OHDA (right) Saline (left) 6-OHDA (right) -62 -55 -45 No. 1 2 3 1 2 3 1 2 3 1 2 3 Mr (K) p62 parkin 6-OHDA 0 6 12 24 0 6 12 24 0 6 12 24 0 6 12 24 h Soluble Insoluble Soluble Insoluble -62 -55 -45 β-Actin parkin+/+ MEFs parkin-/- MEFs Mr (K) 0 0 0.5 1.0 1.5 Relative p62 level P = 0.045 P = 0.420 E D parkin+/+ parkin-/- Total TH+ positive Cell 0 10000 20000 30000 40000 50000 60000 CTR 6-OHDA F C β-Actin p62 Parkin Saline (left) parkin+/+ mice parkin-/- mice 6-OHDA (right) Saline (left) 6-OHDA (right) -62 -55 -45 No. 1 2 3 1 2 3 1 2 3 1 2 3 Mr (K) 6-OHDA 0 6 12 24 0 6 12 24 0 6 12 24 0 6 12 24 h Soluble Insoluble Soluble Insoluble parkin+/+ MEFs parkin-/- MEFs M (K) 1 5 P = 0.045 F E parkin+/+ parkin-/- p62 parkin 6-OHDA 0 6 12 24 0 6 12 24 0 6 12 24 0 6 12 24 h Soluble Insoluble Soluble Insoluble -62 -55 -45 β-Actin Mr (K) +/+ +/+ + + + + - - - - -/- -/- 0.0 0.5 1.0 1.5 parkin Saline 6-OHDA Relative p62 level P = 0.045 P = 0.420 E parkin parkin pa 6-O β-A +/+ +/+ + + + + - - - - -/- -/- 0.0 0.5 1.0 1.5 parkin Saline 6-OHDA Relative p62 level P = 0.045 P = 0.420 E Figure 6. Parkin affects p62 degradation in response to 6-OHDA treatments. (A) The parkin+/+and pa Figure 6. Parkin affects p62 degradation in response to 6-OHDA treatments. (A) The parkin+/+and parkin−/−mice, 18-month-old male C57Bl/6 mice were injured by striatal stereotactic injections with 5.4 μg of 6-OHDA or saline after two weeks, behavioral tests were carried as described in MATERIALS AND METHODS. DISCUSSION (C) Analysis of the SN cell number by Image-J soft was described in (B). (D) The mice described in A were sacriﬁced and the SN regions were isolated, then tissue lysate was subjected to Western blotting for the indicated protein levels. (E) The relative p62 protein levels were described in (D), which were normalized to β-actin from 3 separate experiments. (F) The parkin+/+ and parkin−/−MEF cells were treated with 100 μmol/L 6-OHDA for indicated times, both the soluble and insoluble lysate were subjected to Western blotting for the indicated protein levels. All data are from three independent experiments (n = 7–9 mice). Mean ± SEM, one-way ANOVA, P < 0.05, P < 0.01 compared with control group. * Control 6-OHDA parkin-/- parkin+/+ 15 10 5 0 Rotation (n) Protein & Cell E D parkin+/+ parkin-/- Total TH+ positive Cell 0 10000 20000 30000 40000 50000 60000 CTR 6-OHDA F C β-Actin p62 Parkin Saline (left) parkin+/+ mice parkin-/- mice 6-OHDA (right) Saline (left) 6-OHDA (right) -62 -55 -45 No. 1 2 3 1 2 3 1 2 3 1 2 3 Mr (K) L p62 parkin 6-OHDA 0 6 12 24 0 6 12 24 0 6 12 24 0 6 12 24 h Soluble Insoluble Soluble Insoluble -62 -55 -45 β-Actin parkin+/+ MEFs parkin-/- MEFs Mr (K) +/+ +/+ + + + + - - - - -/- -/- 0.0 0.5 1.0 1.5 parkin Saline 6-OHDA Relative p62 level P = 0.045 P = 0.420 Figure 6. Parkin affects p62 degradation in response to 6-OHDA treatments. (A) The parkin+/+and parkin−/−mice, 18-month-old male C57Bl/6 mice were injured by striatal stereotactic injections with 5.4 μg of 6-OHDA or saline after two weeks, behavioral tests were carried as described in MATERIALS AND METHODS. (B) The parkin+/+ and parkin−/−mice, 18-month-old male C57Bl/6 mice were injured by striatal stereotactic injections from one side in brain with 5.4 μg of 6-OHDA or saline after two weeks, and the SN cells number was detected by immunohistochemical staining using anti-TH antibody antibodies. (C) Analysis of the SN cell number by Image-J soft was described in (B). DISCUSSION 1 2 3 1 2 3 1 2 3 1 2 3 Mr (K) * * Control 6-OHDA Control 6-OHDA Control 6-OHDA Control 6-OHDA Control 6-OHDA Control 6-OHDA 6000 4000 2000 0 Distance moved (cm) parkin-/- parkin+/+ parkin-/- parkin+/+ parkin-/- parkin+/+ parkin-/- parkin+/+ parkin-/- parkin+/+ parkin-/- parkin+/+ 6 4 2 0 60 40 20 0 Velocity (cm/s) Steps (n) 15 10 5 0 Rotation (n) Legs off the pole frequency Ratio of legs off the pole 40 30 20 10 0 0.8 0.6 0.4 0.2 0.0 parkin+/+ mice parkin-/- mice Left/Control Right/6-OHDA Right/6-OHDA Left/Control p62 parkin 6-OHDA 0 6 12 24 0 6 12 24 0 6 12 24 0 6 12 24 h Soluble Insoluble Soluble Insoluble -62 -55 -45 β-Actin parkin+/+ MEFs parkin-/- MEFs Mr (K) +/+ +/+ + + + + - - - - -/- -/- 0.0 0.5 1.0 1.5 parkin Saline 6-OHDA Relative p62 level P = 0.045 P = 0.420 Figure 6. Parkin affects p62 degradation in response to 6-OHDA treatments. (A) The parkin+/+and parkin−/−mice, 18-month-old male C57Bl/6 mice were injured by striatal stereotactic injections with 5.4 μg of 6-OHDA or saline after two weeks, behavioral tests were carried as described in MATERIALS AND METHODS. (B) The parkin+/+ and parkin−/−mice, 18-month-old male C57Bl/6 mice were injured by striatal stereotactic injections from one side in brain with 5.4 μg of 6-OHDA or saline after two weeks, and the SN cells number was detected by immunohistochemical staining using anti-TH antibody antibodies. (C) Analysis of the SN cell number by Image-J soft was described in (B). (D) The mice described in A were sacriﬁced and the SN regions were isolated, then tissue lysate was subjected to Western blotting for the indicated protein levels. DISCUSSION (E) The relative p62 protein levels were described in (D), which B A ** * * Control 6-OHDA Control 6-OHDA Control 6-OHDA Control 6-OHDA Control 6-OHDA Control 6-OHDA 6000 4000 2000 0 Distance moved (cm) parkin-/- parkin+/+ parkin-/- parkin+/+ parkin-/- parkin+/+ parkin-/- parkin+/+ parkin-/- parkin+/+ parkin-/- parkin+/+ 6 4 2 0 60 40 20 0 Velocity (cm/s) Steps (n) 15 10 5 0 Rotation (n) Legs off the pole frequency Ratio of legs off the pole 40 30 20 10 0 0.8 0.6 0.4 0.2 0.0 parkin+/+ mice parkin-/- mice Left/Control Right/6-OHDA Right/6-OHDA Left/Control A Control 6-OHDA 6000 4000 2000 0 Distance moved (cm) parkin-/- parkin+/+ B - parkin+/+ mice parkin-/- mice Left/Control Right/6-OHDA Right/6-OHDA Left/Control Control 6-OHDA - parkin-/- parkin+/+ 60 40 20 0 Steps (n) B parkin+/+ mice Right/6-OHDA parkin-/- mice Right/6-OHDA Left/Control E D parkin+/+ parkin-/- Total TH+ positive Cell 0 10000 20000 30000 40000 50000 60000 CTR 6-OHDA F C β-Actin p62 Parkin Saline (left) parkin+/+ mice parkin-/- mice 6-OHDA (right) Saline (left) 6-OHDA (right) -62 -55 -45 No. 1 2 3 1 2 3 1 2 3 1 2 3 Mr (K) * * ** Control 6-OHDA Control 6-OHDA Control 6-OHDA D p p p p p p parkin-/- parkin+/+ parkin-/- parkin+/+ parkin-/- parkin+/+ 15 10 5 0 Rotation (n) Legs off the pole frequency Ratio of legs off the pole 40 30 20 10 0 0.8 0.6 0.4 0.2 0.0 parkin-/- mice Right/6-OHDA Left/Control p62 parkin 6-OHDA 0 6 12 24 0 6 12 24 0 6 12 24 0 6 12 24 h Soluble Insoluble Soluble Insoluble -62 -55 -45 β-Actin parkin+/+ MEFs parkin-/- MEFs Mr (K) +/+ +/+ + + + + - - - - -/- -/- 0.0 0.5 1.0 1.5 parkin Saline 6-OHDA Relative p62 level P = 0.045 P = 0.420 Figure 6. Parkin affects p62 degradation in response to 6-OHDA treatments. (A) The parkin+/+and parkin−/−mice, 18-month-old male C57Bl/6 mice were injured by striatal stereotactic injections with 5.4 μg of 6-OHDA or saline after two weeks, behavioral tests were carried as described in MATERIALS AND METHODS. (B) The parkin+/+ and parkin−/−mice, 18-month-old male C57Bl/6 mice were injured by striatal stereotactic injections from one side in brain with 5.4 μg of 6-OHDA or saline after two weeks, and the SN cells number was detected by immunohistochemical staining using anti-TH antibody antibodies. DISCUSSION This article is published with open access at Springerlink.com and journal.hep.com.cn 124 RESEARCH ARTICLE Parkin promotes proteasomal degradation of p62 contribute to the aging process and the onset of age-related conditions. Alzheimer’s (AD), Parkinson’s (PD), Huntington’s (HD) and other neurodegenerative diseases are character- ized by the accumulation of protein aggregates in the brain. A different aggregation-prone protein characterizes the pathology of each of these diseases, but virtually all these protein aggregates associate with p62/SQSTM1. Our results thus uncover a potential role of parkin-p62 axis for the selective vulnerability of dopamingeric neuronal cells during the onset of the diseases. It is interesting that p62 levels are increased in STR and SN regions in parkin−/−mice, while p62 in the other regions are not (Fig. 1). It was reported that there is a selective inactivation of parkin in the SN and STR in sporadic PD through nitrosative and dopaminergic stress in aged mice (Chung et al., 2004; LaVoie et al., 2005, 2007). It is thus conceivable that the selective inactivation of parkin in dopamingeric neurons may result in the accumulation and aggregation of p62 leading to the inclusion body formation (Komatsu et al., 2007) and toxic protein aggregation, critical step for Lewy body formation. We also noticed that aged mice showed the characteristic phenotype of movement disorder (Fig. S4), which younger mice have less obvious phenotypes, suggesting the aged related cellular events, such as protein oxidation and aggregation, or altered dopamine metabolisms may be involved (Goldberg et al., 2003), and our data directly showed that parkin mediated the aggregated p62 degrada- tion in cells (Fig. S5). During aging, the inactivation of parkin in dopaminogeric neurons may promote the aggregation of p62 and neurotoxic proteins for the loss neuronal cells. Indeed, we observed the increase of insoluble p62 in the brain of PD patients (data not shown). Thus, our results are consistent with previous suggestions that inactivation of parkin is closely associated with the sporadic and progres- sive nature of PD. Further dissection of how the dysregulated parkin/p62 axis in dopamigeric neuronal cells will offer new insights of the molecular pathogenesis of PD and possible new intervention strategies for ﬁghting PD. Signaling 2132, Millipore AB9244), FLAG (Sigma F1804) were from the indicated sources. contribute to the aging process and the onset of age-related conditions. Cell cultures and plasmids Cell cultures and plasmids SH-SY5Y, 293T cells were cultured at 37°C (5% CO2) in DMEM (GIBCO) supplemented with 10% FBS (HyClone). The mammalian expression plasmids for GFP-parkin, HA-UB were generated as described previously (Wang et al., 2011a). The site mutants and deletion mutants of parkin were generated by PCR with different primers using pEGFPC1-parkin as template. Full length p62 cDNA was cloned into the pCMV-tag-2B vector. The deletion mutants of p62 were generated by PCR with different primers using pCMV-tag- 2B-p62 as a template. The soluble and insoluble cell lysate were prepared in buffer 1% Trion-100 and 1% SDS buffer described in reference (Kawahara et al., 2008). Transfections and shRNAs DNA transfections were performed using PEI according to the manufacturer’s instructions. The target sequences of parkin shRNA 1 and parkin shRNA 2 were 5′-gatccGTGATTTGCTTAGACTGT TTTTCAAGAGAAAACAGTCTAAGCAAATCATTTTTTg-3′, 5′-aattc AAAAAATGATTTGCTTAGACTGTTTTCTCTTGAAAAACAGTCTAA GCAAATCACg-3′; and 5′-gatccGCTTGGCTACTCCCTGCCTTTTC AAGAGAAAGGCAGGGAGTAGCCAAGTTTTTTg-3′, 5′-aattcAAAA AACTTGGCTACTCCCTGCCTTTCTCTTGAAAAGGCAGGGAGTA GCCAAGCg-3′. Protein & Cell The qPCR primers were as follows: p62-1, CAGAGAATACCT TTGCCTCCCA; p62-2, AATCTTGGAGCTCCCCATGTC; parkin-1, ATTCAGAAGCAGCCAGAGGTC; parkin-2, CTGGCACTCACCAC TCATCC. Protein & Cell Immunoﬂuorescence analysis SH-SY5Y cells transfected with parkin-Myc or Myc-vector grown on coverslips were washed with phosphate-buffered saline (PBS) and ﬁxed in 4% formaldehyde in DMEM for 30 min at 37°C. Fixed cells were permeabilized with 0.2% Triton X-100 in PBS for 5 min at 4°C, and blocked with 3% BSA in PBS for 1 h. Then cells were stained with primary antibody (mouse anti-Myc, diluted 1:200; rabbit anti- p62, diluted 1:700) overnight at 4°C. After washing, cells were incubated with ﬂuorescein isothiocyanate (FITC)-conjugated anti- rabbit IgG and Cytm3-linked anti-mouse IgG for 1 h. Unbound anti- body was removed with PBS, and cells were imaged using a Zeiss ﬂuorescence microscope. Western blotting, immunoprecipitation and MBP pull-down were per- formed, as described previously (Wang et al., 2011a). Brieﬂy, cells were transfected with parkin or parkin mutant constructs for 36 h. Cells were washed with ice-cold PBS and lysed with lysis buffer (pNAS buffer: 50 mmol/L Tris–HCl [pH 7.5], 150 mmol/L NaCl, 1 mmol/L EDTA, and 1% Nonidet P-40). The soluble fractions were resolved by SDS-PAGE and transferred onto nitrocellulose ﬁlter membrane. Blots were probed with antibodies against p62, GFP, and β-actin. For immunoprecipitation, cell lysate was incubated with anti-FLAG anti- body and then protein A/G-agarose beads (Pierce Biotechnology). The beads were washed extensively and boiled in SDS loading buffer, and the precipitated proteins were detected by Western blotting. For MBP pull-down, MBP or MBP-parkin fusion protein immobilized on amylose magnetic beads was incubated with in vitro-translated p62. The beads were washed and boiled in the SDS loading buffer, and the precipitated proteins were detected by Western blotting. DISCUSSION Alzheimer’s (AD), Parkinson’s (PD), Huntington’s (HD) and other neurodegenerative diseases are character- ized by the accumulation of protein aggregates in the brain. A different aggregation-prone protein characterizes the pathology of each of these diseases, but virtually all these protein aggregates associate with p62/SQSTM1. Our results thus uncover a potential role of parkin-p62 axis for the selective vulnerability of dopamingeric neuronal cells during the onset of the diseases. It is interesting that p62 levels are increased in STR and SN regions in parkin−/−mice, while p62 in the other regions are not (Fig. 1). It was reported that there is a selective inactivation of parkin in the SN and STR in sporadic PD through nitrosative and dopaminergic stress in aged mice (Chung et al., 2004; LaVoie et al., 2005, 2007). It is thus conceivable that the selective inactivation of parkin in dopamingeric neurons may result in the accumulation and aggregation of p62 leading to the inclusion body formation (Komatsu et al., 2007) and toxic protein aggregation, critical step for Lewy body formation. We also noticed that aged mice showed the characteristic phenotype of movement disorder (Fig. S4), which younger mice have less obvious phenotypes, suggesting the aged related cellular events, such as protein oxidation and aggregation, or altered dopamine metabolisms may be involved (Goldberg et al., 2003), and our data directly showed that parkin mediated the aggregated p62 degrada- tion in cells (Fig. S5). During aging, the inactivation of parkin in dopaminogeric neurons may promote the aggregation of p62 and neurotoxic proteins for the loss neuronal cells. Indeed, we observed the increase of insoluble p62 in the brain of PD patients (data not shown). Thus, our results are consistent with previous suggestions that inactivation of parkin is closely associated with the sporadic and progres- sive nature of PD. Further dissection of how the dysregulated parkin/p62 axis in dopamigeric neuronal cells will offer new insights of the molecular pathogenesis of PD and possible new intervention strategies for ﬁghting PD. DISCUSSION (B) The parkin+/+ and parkin−/−mice, 18-month-old male C57Bl/6 mice were injured by striatal stereotactic injections from one side in brain with 5.4 μg of 6-OHDA or saline after two weeks, and the SN cells number was detected by immunohistochemical staining using anti-TH antibody antibodies. (C) Analysis of the SN cell number by Image-J soft was described in (B). (D) The mice described in A were sacriﬁced and the SN regions were isolated, then tissue lysate was subjected to Western blotting for the indicated protein levels. (E) The relative p62 protein levels were described in (D), which were normalized to β-actin from 3 separate experiments. (F) The parkin+/+ and parkin−/−MEF cells were treated with 100 μmol/L 6-OHDA for indicated times, both the soluble and insoluble lysate were subjected to Western blotting for the indicated protein levels. All data are from three independent experiments (n = 7–9 mice). Mean ± SEM, one-way ANOVA, P < 0.05, P < 0.01 compared with control group. Figure 6. Parkin affects p62 degradation in response to 6-OHDA treatments. (A) The parkin+/+and parkin−/−mice, 18-month-old male C57Bl/6 mice were injured by striatal stereotactic injections with 5.4 μg of 6-OHDA or saline after two weeks, behavioral tests were carried as described in MATERIALS AND METHODS. (B) The parkin+/+ and parkin−/−mice, 18-month-old male C57Bl/6 mice were injured by striatal stereotactic injections from one side in brain with 5.4 μg of 6-OHDA or saline after two weeks, and the SN cells number was detected by immunohistochemical staining using anti-TH antibody antibodies. (C) Analysis of the SN cell number by Image-J soft was described in (B). (D) The mice described in A were sacriﬁced and the SN regions were isolated, then tissue lysate was subjected to Western blotting for the indicated protein levels. (E) The relative p62 protein levels were described in (D), which were normalized to β-actin from 3 separate experiments. (F) The parkin+/+ and parkin−/−MEF cells were treated with 100 μmol/L 6-OHDA for indicated times, both the soluble and insoluble lysate were subjected to Western blotting for the indicated protein levels. All data are from three independent experiments (n = 7–9 mice). Mean ± SEM, one-way ANOVA, P < 0.05, **P < 0.01 compared with control group. © The Author(s) 2015. 6-OHDA-lesion model of Parkinson’s disease in mice with 5 μmol/L MG132 for 8 h before harvesting. The cells were lysed for 30 min at 4°C in either pNAS buffer: 50 mmol/L Tris–HCl [pH 7.5], 150 mmol/L NaCl, 1 mmol/L EDTA, and 1% Nonidet P-40 (to detect noncovalent interaction) or 50 mmol/L Tris [pH 8.0], 150 mmol/L NaCl, 1% Triton, 0.5% sodium deoxycholate, and 0.1% sodium dodecyl sulfate (SDS) (to detect covalent interaction), both con- taining protease inhibitors. Cell lysate was incubated with anti-p62 antibody. The precipitates were subjected to Western blotting with anti-HA or anti-UB antibodies. Mice were anaesthetized using Chloral hydrate and placed into a stereotactic frame with nose and ear bars specially adapted for mice. 6-OHDA was dissolved at a concentration of 3 μg/μL saline in 0.1% ascorbic acid and injected at ﬁnal dosages 5.4 μg. The lesion was performed using a Hamilton syringe at the following coordinates: AP: −2.9 mm; ML: +1.3 mm; DV: −4.6 mm. The injection was conducted at a rate of 0.2 μL/min and the needle was left in place for another 5 min after the injection before it was slowly drawn back. Wound healing and recovery were monitored after the injection was done. An in vitro ubiquitination assay was performed, as described pre- viously (Wang et al., 2011a). Brieﬂy, 2 μg MBP, MBP-parkin or MBP- parkin mutants, expressed and puriﬁed in a E. coli expression system, was incubated with in vitro translated p62 (2 μg) in 50 μL ubiquitintion reaction buffer, containing 50 mmol/L Tris–HCl [pH 7.5], 5 mmol/L MgCl2, 2 mmol/L DTT, 2 mmol/L ATP, 10 μg ubiquitin, 100 ng E1, and 200 ng E2 (UbcH7). Reaction was performed for 2 h at 25°C and terminated by addition of the SDS loading buffer. The reaction prod- ucts were then subjected to Western blotting with anti-p62 antibodies. ABBREVIATIONS 293Tcells were co-transfected with GFP-parkin and FLAG-p62. Cells were treated with 5 μmol/L MG132 for 8 h before harvest. Cell lysates were then subjected to immunoprecipitation with anti-FLAG antibody. The immunoprecipitates were resolved by SDS-PAGE and visualized by Coomassie blue staining. Protein samples were reduced, alky- lated and digested with trypsin. The digests were subsequently analyzed by liquid chromatography tandem mass spectrometry. 6-OHDA, 6-Hydroxydopamine hydrochloride; BA, Baﬁlomycin A1; CB, cerebellum; CHX, cycloheximide; CTX, frontal cortex; PD, Parkinson’s disease; PMSF, Pheylmethylsulfonyl ﬂuoride; SDS, sodium dodecyl sulfate; SN, substantia nigra; STR, striatum; TH, tyrosine hydroxylase; UB, ubiquitin. Statistical analysis Statistical analysis between groups was performed by unpaired two- tailed Student’s t test. Data are presented as mean ± SEM. ACKNOWLEDGMENTS We are grateful to Drs. Ted Dawson and Jian Feng for generously providing the plasmids. We are also grateful to Professor Mark Bartlam from Nankai University, Tianjin, China for a critical reading of the manuscript. The research was supported by the National Basic Research Program (973 Program) (No. 2011CB910903) from MOST and project (Grant Nos. 81130045, 31471300, 31271529, 301520103904) from the National Natural Science Foundation of China. Mass spectrometry analysis of ubiquitination sites Animal behavior tests Protein & Cell The methods for the behavioral tests were described previously (Goldberg et al., 2003). All animal tests were carried out between 9:00 and 15:00 and they were scored by the same rater in an observation sound-atten- uated room under low-intensity light (12 l×), where the mice had been habituated for at least 1–2 h before the beginning of the tests. Behavior was monitored through a video camera positioned above the apparatuses and the videos were later analyzed. The apparatus were cleaned with 10% ethanol between animals to avoid odor cues. Brieﬂy a Plexiglass beam consisting of four Sects. (25 cm each, 1 m total length) of varying width (3.5, 2.5, 1.5, and 0.5 cm) was used. Individual parkin+/+ or parkin−/−mice were tested after being trained twice, and their performance was videotaped. The numbers of steps and slips were counted by viewing the videotapes in slow motion. Protein & Cell RESEARCH ARTICLE Pingping Song et al. 6-OHDA-lesion model of Parkinson’s disease in mice Immunocytochemical and histochemicalanalysis Mice brains were removed and washed with ice-cold PBS. The brains then were post-ﬁxed with 4% paraformaldehyde for 12 h and cryoprotected in 30% sucrose. Coronal sections were cut throughout the midbrain and sections were reacted with rabbit polyclonal anti- p62 and mouse monoclonal anti-Tyrosine hydroxylase (TH) and visualized with ﬂuorescein isothiocyanate (FITC)-conjugated anti- rabbit IgG and cytm3-linked anti-mouse IgG. Four different brain regions from wild-type or parkin knockout mice were homogenized in lysis buffer containing 10 mmol/L Tris– HCl [pH 7.4], 150 mmol/L NaCl, 5 mmol/L EDTA, 0.5% Nonidet P-40, Phosphate Inhibitor Cocktail I and II (Sigma), and Complete Pro- tease Inhibitor Mixture (Roche), using homogenizer. After homoge- nization, samples were rotated at 4°C for 30 min for complete lysis, then the homogenate was centrifuged at 14,000 rpm for 20 min, and the resulting fractions were collected and analyzed by immunoblot. Immunoblotting was performed with an antibody of interest and was performed with chemiluminescence (Pierce). The densitometric analyses of the bands were performed using Image-J. Data are expressed as mean ± SEM. The results were evaluated for statistical signiﬁcance by applying the unpaired two-tailed Student’s t test. Ubiquitination assays Antibodies against Myc (Sc-40), GFP (Sc-9996) and HA (Sc-7392) were purchased from Santa Cruz. Antibodies against p62 (MBL PM045), β-actin (Sigma A5441), ATG5 (Sigma, A0856), parkin (Cell 293T or SH-SY5Y cells were transfected with indicated tagged constructs in each experiment employing PEI. Cells were treated article is published with open access at Springerlink.com and journal.hep.com.cn uthor(s) 2015. This article is published with open access at Springerlink.com and journal.hep.com.cn © The Author(s) 2015. This article is published with open access at Springerlink.com and journal.hep.com.cn 125 RESEARCH ARTICLE COMPLIANCE WITH ETHICS GUIDELINES Chronic hypoxia treatments of mice REFERENCES Ichimura Y, Komatsu M (2010) Selective degradation of p62 by autophagy. Semin Immunopathol 32:431–436 Babu JR, Geetha T, Wooten MW (2005) Sequestosome 1/p62 shuttles polyubiquitinated tau for proteasomal degradation. J Neurochem 94:192–203 Ishikawa A, Tsuji S (1996) Clinical analysis of 17 patients in 12 Japanese families with autosomal-recessive type juvenile parkin- sonism. Neurology 47:160–166 Bartlett BJ, Isakson P, Lewerenz J, Sanchez H, Kotzebue RW, Cumming RC, Harris GL, Nezis IP, Schubert DR, Simonsen A et al (2011) p62, Ref(2)P and ubiquitinated proteins are con- served markers of neuronal aging, aggregate formation and progressive autophagic defects. Autophagy 7:572–583 Itier JM, Ibanez P, Mena MA, Abbas N, Cohen-Salmon C, Bohme GA, Laville M, Pratt J, Corti O, Pradier L et al (2003) Parkin gene inactivation alters behaviour and dopamine neurotransmission in the mouse. Hum Mol Genet 12:2277–2291 Bingol B, Tea JS, Phu L, Reichelt M, Bakalarski CE, Song Q, Foreman O, Kirkpatrick DS, Sheng M (2014) The mitochondrial deubiquitinase USP30 opposes parkin-mediated mitophagy. Nature 510:370–375 Jain A, Lamark T, Sjottem E, Larsen KB, Awuh JA, Overvatn A, McMahon M, Hayes JD, Johansen T (2010) p62/SQSTM1 is a target gene for transcription factor NRF2 and creates a positive feedback loop by inducing antioxidant response element-driven gene transcription. J Biol Chem 285:22576–22591 Biskup S, Gerlach M, Kupsch A, Reichmann H, Riederer P, Vieregge P, Wullner U, Gasser T (2008) Genes associated with Parkinson syndrome. J Neurol 255(Suppl 5):8–17 Johnson BN, Berger AK, Cortese GP, Lavoie MJ (2012) The ubiquitin E3 ligase parkin regulates the proapoptotic function of Bax. Proc Natl Acad Sci USA 109:6283–6288 Bjorkoy G, Lamark T, Johansen T (2006) p62/SQSTM1: a missing link between protein aggregates and the autophagy machinery. Autophagy 2:138–139 Kahle PJ, Haass C (2004) How does parkin ligate ubiquitin to Parkinson’s disease? EMBO Rep 5:681–685 Bossy-Wetzel E, Schwarzenbacher R, Lipton SA (2004) Molecular pathways to neurodegeneration. Nat Med 10(Suppl):S2–S9 Kane LA, Lazarou M, Fogel AI, Li Y, Yamano K, Sarraf SA, Banerjee S, Youle RJ (2014) PINK1 phosphorylates ubiquitin to activate Parkin E3 ubiquitin ligase activity. J Cell Biol 205:143–153 Burchell VS, Nelson DE, Sanchez-Martinez A, Delgado-Camprubi M, Ivatt RM, Pogson JH, Randle SJ, Wray S, Lewis PA, Houlden H et al (2013) The Parkinson’s disease-linked proteins Fbxo7 and Parkin interact to mediate mitophagy. OPEN ACCESS Protein & Cell Gegg ME, Cooper JM, Chau KY, Rojo M, Schapira AH, Taanman JW (2010) Mitofusin 1 and mitofusin 2 are ubiquitinated in a PINK1/parkin-dependent manner upon induction of mitophagy. Hum Mol Genet 19:4861–4870 This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/ licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Goldberg MS, Fleming SM, Palacino JJ, Cepeda C, Lam HA, Bhatnagar A, Meloni EG, Wu N, Ackerson LC, Klapstein GJ et al (2003) Parkin-deﬁcient mice exhibit nigrostriatal deﬁcits but not loss of dopaminergic neurons. J Biol Chem 278:43628–43635 AUTHOR CONTRIBUTIONS Dawson TM, Dawson VL (2003) Molecular pathways of neurode- generation in Parkinson’s disease. Science 302:819–822 P.S, Y.Z and Q.C designed the study and wrote the paper. P.S performed most of the experiments and data analysis, P.S, S.L, D.W and Z.C performed the animal experiments and tissues samples assay, S.L and G.R performed the qPCR for tissues. Y.Z, S.L, N.S and D.W performed the animal behavior test analysis. Z.T and T.T performed the animal behavior data analysis and text critical revision. H.D performed the mass analysis. Z.Z provide the parkin+/−mice. All authors analyzed the results and approved the ﬁnal version of the manuscript. Dawson TM, Dawson VL (2010) The role of parkin in familial and sporadic Parkinson’s disease. Mov Disord 25(Suppl 1):S32–S39 Farrer MJ (2006) Genetics of Parkinson disease: paradigm shifts and future prospects. Nat Rev Genet 7:306–318 Fecto F, Yan J, Vemula SP, Liu E, Yang Y, Chen W, Zheng JG, Shi Y, Siddique N, Arrat H et al (2011) SQSTM1 mutations in familial and sporadic amyotrophic lateral sclerosis. Arch Neurol 68:1440– 1446 Protein & Cell Gegg ME, Schapira AH (2011) PINK1-parkin-dependent mitophagy involves ubiquitination of mitofusins 1 and 2: implications for Parkinson disease pathogenesis. Autophagy 7:243–245 Chronic hypoxia treatments of mice Pingping Song, Shanshan Li, Hao Wu, Ruize Gao, Guanhua Rao, Dongmei Wang, Ziheng Chen, Biao Ma, Nan Sui, Haiteng Deng, Zhuohua Zhang, Tieshan Tang, Zheng Tan, Zehan Han, Tieyuan Lu, Yushan Zhu and Quan Chen declare that they have no conﬂict of parkin+/+ or parkin−/−mice (provided by Prof. Zhuohua Zhang from Central South University) were subjected to a mice hypoxic chamber for the indicated duration, with an atmosphere of 8% O2 in nitrogen and free access to food and water. 126 © The Author(s) 2015. This article is published with open access at Springerlink.com and journal.hep.com.cn RESEARCH ARTICLE Parkin promotes proteasomal degradation of p62 interest with the contents of this article. All institutional and national guidelines for the care and use of laboratory animals were followed. interest with the contents of this article. All institutional and national guidelines for the care and use of laboratory animals were followed. regulates ubiquitination and compromises parkin’s protective function. Science 304:1328–1331 regulates ubiquitination and compromises parkin’s protective function. Science 304:1328–1331 Dawson TM (2007) Unraveling the role of defective genes in Parkinson’s disease. Parkinsonism Relat Disord 13(Suppl 3): S248–S249 RESEARCH ARTICLE BMC neuroscience 6:71 LaVoie MJ, Cortese GP, Ostaszewski BL, Schlossmacher MG (2007) The effects of oxidative stress on parkin and other E3 ligases. J Neurochem 103:2354–2368 Pickrell AM, Youle RJ (2015) The roles of PINK1, Parkin, and mitochondrial ﬁdelity in Parkinson’s disease. Neuron 85:257–273 Rott R, Szargel R, Haskin J, Shani V, Shainskaya A, Manov I, Liani E, Avraham E, Engelender S (2008) Monoubiquitylation of alpha- synuclein by seven in absentia homolog (SIAH) promotes its aggregation in dopaminergic cells. J Biol Chem 283:3316–3328 Lazarou M, Narendra DP, Jin SM, Tekle E, Banerjee S, Youle RJ (2013) PINK1 drives Parkin self-association and HECT-like E3 activity upstream of mitochondrial binding. J Cell Biol 200:163–172 Rubino E, Rainero I, Chio A, Rogaeva E, Galimberti D, Fenoglio P, Grinberg Y, Isaia G, Calvo A, Gentile S et al (2012) SQSTM1 mutations in frontotemporal lobar degeneration and amyotrophic lateral sclerosis. Neurology 79:1556–1562 Lazarou M, Sliter DA, Kane LA, Sarraf SA, Wang CX, Burman JL, Sideris DP, Fogel AI, Youle RJ (2015) The ubiquitin kinase PINK1 recruits autophagy receptors to induce mitophagy. Nature 524:309–314 Lee JT, Wheeler TC, Li L, Chin LS (2008) Ubiquitination of alpha- synuclein by Siah-1 promotes alpha-synuclein aggregation and apoptotic cell death. Hum Mol Genet 17:906–917 Rue L, Lopez-Soop G, Gelpi E, Martinez-Vicente M, Alberch J, Perez-Navarro E (2013) Brain region- and age-dependent dys- regulation of p62 and NBR1 in a mouse model of Huntington’s disease. Neurobiol Dis 52:219–228 Lee J, Kim HR, Quinley C, Kim J, Gonzalez-Navajas J, Xavier R, Raz E (2012) Autophagy suppresses interleukin-1beta (IL-1beta) signaling by activation of p62 degradation via lysosomal and proteasomal pathways. J Biol Chem 287:4033–4040 Sarraf SA, Raman M, Guarani-Pereira V, Sowa ME, Huttlin EL, Gygi SP, Harper JW (2013) Landscape of the PARKIN-dependent ubiquitylome in response to mitochondrial depolarization. Nature 496:372–376 Lesage S, Brice A (2009) Parkinson’s disease: from monogenic forms to genetic susceptibility factors. Hum Mol Genet 18:R48–R59 Li H, Guo M (2009) Protein degradation in Parkinson disease revisited: it’s complex. J Clin Investig 119:442–445 Seibenhener ML, Babu JR, Geetha T, Wong HC, Krishna NR, Wooten MW (2004) Sequestosome 1/p62 is a polyubiquitin chain binding protein involved in ubiquitin proteasome degradation. Mol Cell Biol 24:8055–8068 Li J, Qi W, Chen G, Feng D, Liu JH, Ma B, Zhou CQ, Mu CL, Zhang WL, Chen Q et al (2015) Mitochondrial outer-membrane E3 ligase MUL1 ubiquitinates ULK1 and regulates selenite-induced mitophagy. RESEARCH ARTICLE RESEARCH ARTICLE Pingping Song et al. Kirkin V, McEwan DG, Novak I, Dikic I (2009) A role for ubiquitin in selective autophagy. Mol Cell 34:259–269 Nakaso K, Yoshimoto Y, Nakano T, Takeshima T, Fukuhara Y, Yasui K, Araga S, Yanagawa T, Ishii T, Nakashima K (2004) Transcrip- tional activation of p62/A170/ZIP during the formation of the aggregates: possible mechanisms and the role in Lewy body formation in Parkinson’s disease. Brain Res 1012:42–51 Kitada T, Asakawa S, Hattori N, Matsumine H, Yamamura Y, Minoshima S, Yokochi M, Mizuno Y, Shimizu N (1998) Mutations in the parkin gene cause autosomal recessive juvenile parkin- sonism. Nature 392:605–608 Narendra D, Tanaka A, Suen DF, Youle RJ (2008) Parkin is recruited selectively to impaired mitochondria and promotes their autop- hagy. J Cell Biol 183:795–803 Ko HS, Kim SW, Sriram SR, Dawson VL, Dawson TM (2006) Identiﬁcation of far upstream element-binding protein-1 as an authentic Parkin substrate. J Biol Chem 281:16193–16196 Pankiv S, Clausen TH, Lamark T, Brech A, Bruun JA, Outzen H, Overvatn A, Bjorkoy G, Johansen T (2007) p62/SQSTM1 binds directly to Atg8/LC3 to facilitate degradation of ubiquitinated protein aggregates by autophagy. J Biol Chem 282:24131–24145 Komatsu M, Ichimura Y (2010) Physiological signiﬁcance of selec- tive degradation of p62 by autophagy. FEBS Lett 584:1374–1378 Komatsu M, Waguri S, Koike M, Sou YS, Ueno T, Hara T, Mizushima N, Iwata J, Ezaki J, Murata S et al (2007) Homeostatic levels of p62 control cytoplasmic inclusion body formation in autophagy- deﬁcient mice. Cell 131:1149–1163 Pankiv S, Lamark T, Bruun JA, Overvatn A, Bjorkoy G, Johansen T (2010) Nucleocytoplasmic shuttling of p62/SQSTM1 and its role in recruitment of nuclear polyubiquitinated proteins to promyelo- cytic leukemia bodies. J Biol Chem 285:5941–5953 Protein & Cell Koyano F, Okatsu K, Kosako H, Tamura Y, Go E, Kimura M, Kimura Y, Tsuchiya H, Yoshihara H, Hirokawa T et al (2014) Ubiquitin is phosphorylated by PINK1 to activate parkin. Nature 510:162–166 Perez FA, Palmiter RD (2005) Parkin-deﬁcient mice are not a robust model of parkinsonism. Proc Natl Acad Sci USA 102:2174–2179 Protein & Cell LaVoie MJ, Ostaszewski BL, Weihofen A, Schlossmacher MG, Selkoe DJ (2005) Dopamine covalently modiﬁes and functionally inactivates parkin. Nat Med 11:1214–1221 Perez FA, Curtis WR, Palmiter RD (2005) Parkin-deﬁcient mice are not more sensitive to 6-hydroxydopamine or methamphetamine neurotoxicity. REFERENCES Nat Neurosci 16:1257– 1265 Kawahara K, Hashimoto M, Bar-On P, Ho GJ, Crews L, Mizuno H, Rockenstein E, Imam SZ, Masliah E (2008) alpha-Synuclein aggregates interfere with Parkin solubility and distribution: role in the pathogenesis of Parkinson disease. J Biol Chem 283:6979– 6987 Chen Y, Dorn GW 2nd (2013) PINK1-phosphorylated mitofusin 2 is a Parkin receptor for culling damaged mitochondria. Science 340:471–475 Kazlauskaite A, Kondapalli C, Gourlay R, Campbell DG, Ritorto MS, Hofmann K, Alessi DR, Knebel A, Trost M, Muqit MM (2014) Parkin is activated by PINK1-dependent phosphorylation of ubiquitin at Ser65. Biochem J 460:127–139 Chung KK, Thomas B, Li X, Pletnikova O, Troncoso JC, Marsh L, Dawson VL, Dawson TM (2004) S-nitrosylation of parkin 127 © The Author(s) 2015. This article is published with open access at Springerlink.com and journal.hep.com.cn RESEARCH ARTICLE Autophagy 11:1216–1229 Shin JH, Ko HS, Kang H, Lee Y, Lee YI, Pletinkova O, Troconso JC, Dawson VL, Dawson TM (2011) PARIS (ZNF746) repression of PGC-1alpha contributes to neurodegeneration in Parkinson’s disease. Cell 144:689–702 Lowe J, Blanchard A, Morrell K, Lennox G, Reynolds L, Billett M, Landon M, Mayer RJ (1988) Ubiquitin is a common factor in intermediate ﬁlament inclusion bodies of diverse type in man, including those of Parkinson’s disease, Pick’s disease, and Alzheimer’s disease, as well as Rosenthal ﬁbres in cerebellar astrocytomas, cytoplasmic bodies in muscle, and mallory bodies in alcoholic liver disease. J Pathol 155:9–15 Singleton AB, Farrer M, Johnson J, Singleton A, Hague S, Kachergus J, Hulihan M, Peuralinna T, Dutra A, Nussbaum R et al (2003) alpha-Synuclein locus triplication causes Parkinson’s disease. Science 302:841 Spillantini MG, Schmidt ML, Lee VM, Trojanowski JQ, Jakes R, Goedert M (1997) Alpha-synuclein in Lewy bodies. Nature 388:839–840 Matsuda N, Sato S, Shiba K, Okatsu K, Saisho K, Gautier CA, Sou YS, Saiki S, Kawajiri S, Sato F et al (2010) PINK1 stabilized by mitochondrialdepolarizationrecruitsParkintodamagedmitochondria and activates latent Parkin for mitophagy. J Cell Biol 189:211–221 Sriram SR, Li X, Ko HS, Chung KK, Wong E, Lim KL, Dawson VL, Dawson TM (2005) Familial-associated mutations differentially 128 © The Author(s) 2015. This article is published with open access at Springerlink.com and journal.hep.com.cn RESEARCH ARTICLE Parkin promotes proteasomal degradation of p62 disrupt the solubility, localization, binding and ubiquitination properties of parkin. Hum Mol Genet 14:2571–2586 disrupt the solubility, localization, binding and ubiquitination properties of parkin. Hum Mol Genet 14:2571–2586 Wang H, Song P, Du L, Tian W, Yue W, Liu M, Li D, Wang B, Zhu Y, Cao C et al (2011a) Parkin ubiquitinates Drp1 for proteasome- dependent degradation: implication of dysregulated mitochondrial dynamics in Parkinson disease. J Biol Chem 286:11649–11658 Sterky FH, Lee S, Wibom R, Olson L, Larsson NG (2011) Impaired mitochondrial transport and Parkin-independent degeneration of respiratory chain-deﬁcient dopamine neurons in vivo. Proc Natl Acad Sci USA 108:12937–12942 Wang XN, Winter D, AshraﬁG, Schlehe J, Wong YL, Selkoe D, Rice S, Steen J, LaVoie MJ, Schwarz TL (2011b) PINK1 and Parkin target miro for phosphorylation and degradation to arrest mito- chondrial motility. Cell 147:893–906 Tanaka A, Cleland MM, Xu S, Narendra DP, Suen DF, Karbowski M, Youle RJ (2010) Proteasome and p97 mediate mitophagy and degradation of mitofusins induced by Parkin. © The Author(s) 2015. This article is published with open access at Springerlink.com and journal.hep.com.cn RESEARCH ARTICLE J Cell Biol 191:1367–1380 Wooten MW, Geetha T, Babu JR, Seibenhener ML, Peng J, Cox N, Diaz-Meco MT, Moscat J (2008) Essential role of sequestosome 1/p62 in regulating accumulation of Lys63-ubiquitinated proteins. J Biol Chem 283:6783–6789 Thomas, B., and Beal, M.F. (2007). Parkinson’s disease. Human molecular genetics 16 Spec No. 2, R183-194 Van Laar VS, Arnold B, Cassady SJ, Chu CT, Burton EA, Berman SB (2011) Bioenergetics of neurons inhibit the translocation response of Parkin following rapid mitochondrial depolarization. Hum Mol Genet 20:927–940 Zatloukal K, Stumptner C, Fuchsbichler A, Heid H, Schnoelzer M, Kenner L, Kleinert R, Prinz M, Aguzzi A, Denk H (2002) p62 Is a common component of cytoplasmic inclusions in protein aggre- gation diseases. Am J Pathol 160:255–263 Protein & Cell Protein & Cell 129
https://openalex.org/W2051652496	https://zenodo.org/records/1623497/files/article.pdf	German	null	Die Röntgenstrahlen in der Anatomie	Deutsche medizinische Wochenschrift/Deutsche Medizinische Wochenschrift	1,905	public-domain	1,124	DETTSCiiE M]DTZ1NIOHE WOC}IENSIDI[RIFP. DETTSCiiE M]DTZ1NIOHE WOC}IENSIDI[RIFP. No. 17 Die Röntgenstrahlen in der Anatomie. Von Karl y. Bardeleben in Jena. Sofort nach Bekanntwerden der Entdeckung Röntgens bemächtigte sich auch die Anatomie derX-Strahlen; im Jahre 1896 erschienen bereits über 40 Arbeiten auf diesem Gebiete. In der Wiener klinischen Wochenschrift 1896, No. 4, reproduzierten E. Haschek und O. Th. Lindenthal ein Photogramm eines mit Teichmannscher Masse injizierten Präparates, und fast gleichzeitig veröffentlichte H. B raus, damals Assistent in Jena, ein Photogramm (Lichtdruck) einer Quecksilberinjektion der Harngefäße im Anatomischen Anzeiger 1896, Bd. 11, No. 21, März. Diesen Aufnahmen von Präparaten folgten bald solche von lebenden Körpern oder zunächst Körperteilen, vor allem von Hand und Fuß, sowie von kleinen Tieren. Die Erfahrung lehrte bald, daß wir mit den Röntgenstrahlen nicht nur Knochen und Gelenke, sondern auch weiche Teile, vor allem die Gefäße, und andere Blut, also Eisen enthaltende Organe gut sehen können. Das Gebiet, auf dem die Röntgenstrahlen in der nor- malen Anatomie und Entwicklungsgeschichte des Menschen, in der vergleichenden Anatomie (Paläontologie), in der Anthro- DEUTSCHE MEDIZENESCHE WOCHENSCHRIF'T. 675 27. ApriL taren, ihre typische Abweichung von denen des jetzt lebenden Menschen! An diese Möglichkeit hätte vor zehn Jahren noch niemand gedacht! pologie brauchbar sind, ja zum Teil zu neuen wichtigen Ent- deckungen geführt haben, ist seitdem von Jahr zu Jahr ge- wachsen und vermutlich noch nicht in seiner ganzen Aus- dehnung erforscht. pologie brauchbar sind, ja zum Teil zu neuen wichtigen Ent- deckungen geführt haben, ist seitdem von Jahr zu Jahr ge- wachsen und vermutlich noch nicht in seiner ganzen Aus- dehnung erforscht. g Aber wir sind ja noch mitten in der Entwicklung der Dinge. Leider ist es bisher nur wenigen Anatomen vergönnt gewesen, die Methode zu benutzen. Dazu gehören bekanntlich kostspielige Apparate, viel Geduld und Uebung. g In der systematischen und topographischen Ana- tomie des Menschen wandte sich das Interesse zunächst dem Skelett zu, das man ja beim Lebenden in allen Einzelheiten sehen kann. Ueberzählige Finger und Fingerglieder wurden gefunden, die Architektur der Knochen und Zähne beim Leben- den studiert, vor allem wurde das Studium der Anatomie und Mechanik der Gelenke, besonders der so vielfache Probleme darbietenden Handgelenke auf eine unerwartete, früher ausgeschlossene Weise vervollkommnet, für viele Fragen überhaupt erst ermöglicht. Außer den Handgelenken waren es die Fufigelenke, ferner vor allem das Kniegelenk, dessen höchst komplizierte Mechanik seit Jahrzehnten eine Streitfrage war, die direkter Beobachtung beim Lebenden und während der verschiedenen Stellungen und Aenderungen zugängig wur- den. DETTSCiiE M]DTZ1NIOHE WOC}IENSIDI[RIFP. Diese Fragen dürfen jetzt als entschieden betrachtet werden. p g pp g Auf manchen Gebieten der Anatomie ist noch viel zu tun übrig, vor allem in der Gelenkmechanik. Hier ist weiterer Ausbau nötig - und in Sicht. F ß M h L b g Ferner müßten Massenuntersuchungen von Leben- den, wie sie an Leichen unmöglich sind, besonders für die Entwicklung des Skeletts, des Herzens, des Kehikopfes und verschiedenes andere angestellt werden. g Auch die ziemlich häufigen Varietäten der großen Gefäße, besonders am Arm, ließen sieh jetzt vor der Operation fest- stellen. Dieses Dokument wurde zum persönlichen Gebrauch heruntergeladen. Vervielfältigung nur mit Zustimmung des Verlages. Viel ist in den neun Jahren seit Röritgens Entdeckung auf dem Gebiete der Anatomie geleistet worden - wer ahei kann wissen oder ahnen, was uns die Zukunft noch bringen wird'? Nächst den Gelenken waren es besonders Herz und große Gefäile und die allerdings schon in. die Physiologie hineinspielenden Bewegungen des herzens, dann aber auch die viel umstrittenenBewegungen des Zwerchfelles in seinem relativ fixierten Centrum tendineum - wiederum in Zusammen- hang mit den Totalbewegungen, den Lageverschiebungen des Herzens, deren Feststellungen große Erfolge aufwiesen. Die Anatomen hatten in den letzten Jahrzehnten das Centrum tendineum als fast ganz fixiert betrachtet und sehr beschränkte Bewegungen, höchstens bis zu 2 cm, zugegeben. Es hat sich gezeigt, daß die Exkursionen des Zentrums und die Lageveränderungen des Herzens erheblich größer sind oder doch sein können. Sehr wichtig wurden die Röntgenstrahlen für die Erken- nung der Form und Lage des Magens in den verschiedenen Füllungszuständen und sonstigen Varietäten. Diesen Unter- suchungen schlossen sich die über die Lage der Leber und der anderen Bauchorgane, ferner die der Brustorgane, der Lungen, an. Das Heller- und Dunklerwerden des Lungen. schattens bei der Ein- und Ausatmung ist jetzt für den Geübten leicht erkennbar. Neuerdings ist sogar der Verlauf des Ureters am Lebenden festgestellt worden. Die bekanntlich regelmäßig mit dem Alter auftretenden Verknöcherungen des Kehlkopfes, die aber, wie alle solche phylogenetischen, in diesem Falle prospektiven Erscheinungen, individuell sehr variabel sind, lassen sich jetzt sehr leicht nachweisen. di i kl hi h i Für die Entwicklungsgeschichte ist besonders das Studium der Knochenkerne des embryonalen (schon intra uterum erkennbar!), kindlichen und heranwachsenden Skeletts wichtig geworden. Auch können jetzt diese Dinge unter Schonung des Materials direkt beobachtet werden. Di l i h d A i h h i A Die vergleichende Anatomie hat noch wenig An- wendung von den X-Strahlen gemacht. DETTSCiiE M]DTZ1NIOHE WOC}IENSIDI[RIFP. Die Paläontologie verdankt der Methode den Nachweis von Knochen und deren Struktur im Gestein. Sehr wichtig ist die Röntgenphotographie für die Anthro- pologie, besonders für die Urgeschichte des Menschen geworden. Wenn der Wert der Methode für das Studium der Architektur der Spongiosa insofern nur ein relativer ist, als man durch das Studium von Schnitten oder Schliffen doch mehr in das Innere der Natur dringt als durch Photographie des ganzen Knochens, so ist die Röntgenmethode geradezu unübertrefflich und nicht genug zu schätzen, ja die einzige, zurzeit mögliche, wenn es sich darum handelt, die Strukturen in dem unersetzlichen, meist nur als Unicum vorhandenen Ma- terial unserer antecjjluvianjschen oder antediluvialen Vorfahren und näheren oder entfernteren Verwandten sichtbar zu machen. Die Knochen von Neanderthal, von Spy, den Höhlen in Belgien, am Niederrhein, in Südösterreich und anderes mehr, ebenso die spärlichen Reste des Pithecanthropus erectus (Java) sind unersetzlich und dürfen nicht zersägt werden. Die X-Strahlen zeigen uns nun auf den von Walkhoff neuerdings veroffentlichten prachtvollen Tafeln die ganzen inneren Struk- Sehr wichtig ist die Röntgenphotographie für die Anthro- pologie, besonders für die Urgeschichte des Menschen geworden. Wenn der Wert der Methode für das Studium der Architektur der Spongiosa insofern nur ein relativer ist, als man durch das Studium von Schnitten oder Schliffen doch mehr in das Innere der Natur dringt als durch Photographie des ganzen Knochens, so ist die Röntgenmethode geradezu unübertrefflich und nicht genug zu schätzen, ja die einzige, zurzeit mögliche, wenn es sich darum handelt, die Strukturen in dem unersetzlichen, meist nur als Unicum vorhandenen Ma- terial unserer antecjjluvianjschen oder antediluvialen Vorfahren und näheren oder entfernteren Verwandten sichtbar zu machen. Die Knochen von Neanderthal, von Spy, den Höhlen in Belgien, am Niederrhein, in Südösterreich und anderes mehr, ebenso die spärlichen Reste des Pithecanthropus erectus (Java) sind unersetzlich und dürfen nicht zersägt werden. Die X-Strahlen zeigen uns nun auf den von Walkhoff neuerdings veroffentlichten prachtvollen Tafeln die ganzen inneren Struk-
https://openalex.org/W4291370967	https://zenodo.org/records/3445801/files/Evaluation%20of%20tourism%20resources%20in%20developing%20the%20urban%20economy%20(case.pdf	English	null	Evaluation of tourism resources in developing the urban economy (case study: Zahak)	Zenodo (CERN European Organization for Nuclear Research)	2,014	cc-by	6,240	Introduction Today, tourism is the most important economic activity of one of the most important scientific issues of the day. (Movahad, 2008, 1) World Tourism Organization, one of the most powerful glances at the international level for to develop tourism, the tourism economy knows. Effects tremendous tourism in reducing unemployment, increasing income levels, providing energy resources, and thriving crafts market, according to statistics, 10 million people since feeding are just some of the special advantages of addressing this industry. Exchange for kidnapping tourists from 2 to 6 jobs will directly create jobs in sectors including manufacturing and a service for tourists, creating jobs is for 9 to 15 people. (Navrozi and etal, 2012). g , g j p p ( , ) "Exploring Sirvan in different lands, a part of human life and the needs of the booming in this century, extensive transport facilities, limitations of human communication between different communities destroyed and therefore tourism as a major issue in human life has been raised. Industry to develop economic, social, cultural and political nation, and has created countless jobs, provide friendship and peace between nations and the promotion of human knowledge is "(Mansouri, 2003). Tourism in the process of becoming a tourist destination, the quality of life, value systems, family relationships, attitudes, traditions, customs, behavior patterns, and many other components of economic, social and cultural influences is These effects can be overpopulation, crowding, use of infrastructure, employment status, impact on earnings and poverty in the community pointed out (Kazemi, 2004) The impact of such a change to be able to establish that the precipitous growth in tourism over the past decade and the future based on its With emphasis economists, technicians and technologists and support for developing countries to strengthen the industry towards sustainable development goals in these countries are allowed With emphasis economists, technicians and technologists and support for developing countries to strengthen the industry towards sustainable development goals in these countries are allowed This strategy not only in developing countries but also includes the industrial world. International Journal of Management Sciences and Business Research, 2013 ISSN (2226-8235) Vol-3, Issue 1 ernational Journal of Management Sciences and Business Research, 2013 ISSN (2226-8235) V Abstract bst act Tourism in today's world economy is an integral part of a serious role in income and Gross National Income (GDP )clean industry is headed Zahak of ancient and historical monuments and especially in terms of strategic location, bordering Afghanistan, regional and international transit, tourism has a privileged situation. The purpose of this study was to identify introducing potential to attract tourists in the border town Zahak (bordering Afghanistan) to tourists aimed to economic development of the city, and provide solutions executive is required. Fusion method in this study document - library and field is. The software, used to test «spearman correlation coefficient», «Somers D test», «Kendall tau be test and Gis, Excel, Spss «t-test for 2 independent means», «Analysis Of Variance » and « Kolmogorov -Smirnov test» is. Preliminary results show there are natural attractions of the region to attract tourists can Zahak to play an important role in the economic development of the city. The most consistent way to achieve economic development in Zahak, identify potential tourism areas and using local communities to engage in this way an effective step in strengthening economic level of economic development is to be removed. f p Keywords: Tourism, Economic Development, Empowerment and Attractions, Zahak Evaluation of tourism resources in developing the urban economy (case study: Zahak) Author’s Details : Hossien Manuchehrian 1*, Dr. Gholam Reza Miri2, Dr. Ali Asghar Kadiuar3 1 MSc of Geography & Tourism Planning, Islamic Azad University of Zahedan (Corresponding Author) 2 Assistant professors of Geography and Urban Planning, Islamic Azad University, Branch of Zahedan, Iran 3 Assistant professors of Geography and Urban Planning, Islamic Azad University, Branch of Mashhad, Iran http://www.ijmsbr.com Theoretical Principles Research Theoretical Principles Research "Tourism and the Greek origin of the word is getting around the Tour means Greek to Spanish and French and has finally found its way to England the following definitions are used in the Oxford Dictionary: 1) The place or places where the short travel a lot to see. "Tourism and the Greek origin of the word is getting around the Tour means Greek to Spanish and French and has finally found its way to England the following definitions are used in the Oxford Dictionary: "Tourism and the Greek origin of the word is getting around the Tour means Greek to Spanish and French and has finally found its way to England the following definitions are used in the Oxford Dictionary: 1) The place or places where the short trave 2) Pass a short trip somewhere. 3) Means that large amounts of money through the tourism industry, with foreign tourists who come to earn some countries are. 3) Means that large amounts of money through the tourism industry, with foreign tourists who come to earn some countries are. In general, the word tour is derived from the word tourism means a trip back to the spot and follow a specific itinerary. (Ali Asgari, 1997)« Tourism (Tourism) is the activity or activities that people can relax, work or other reasons outside of their usual travel and at least one night stay for a maximum of one year in succession where the chosen "(Sanai, 1996). "Tourism is a phenomenon of the distant past, societies have been considered According to the different needs of economic, social, historical, and ... The momentum has continued This phenomenon since the Industrial Revolution, particularly during the last century, has expanded dramatically, Underlying science and technology in international relations and economic development of nations "(Ali Asgari, 1997). When "Thomas Cook" of London launched his first tour, about 150 years ago, this action over a century and a half of ups and downs and changes, the largest social movement in human history became here goes the next century, the greatest economic, social and cultural wellbeing ". International Journal of Management Sciences and Business Research, 2013 ISSN (2226-8235) Vol-3, Issue 1 (absolute champion, 1997). And of course, the best use possible of the effects caused by the Thus the social and economic structure of the different functions that can be important is tourism the study in this area will need. As the tourism industries speak, while providing comfort and recreational industry, causing economic development income urban communities are also included. The world tourism industry is one of the most successful because of the significant achievements of the lining can be noted the gain of foreign currency revenues, direct and indirect job creation to demonstration of stability and security in the country. Our active and effective for the tourism industry by geographical location , historical and cultural attractions of the potential for various reasons , including lack of facilities to attract tourists then welfare facilities , Iran's tourism industry is not your real position and set apart (Yosfipour, 2002, 45 ) therefore, tourism can Zahak economic development in the northern city of Sistan and Baluchistan province bordering Afghanistan and Iran and Afghanistan on the transit route is provided. http://www.ijmsbr.com Introduction Due to the wide variety of tourist community, a group of researchers, the impact of the industry in three categories: economic, environmental and social - are classified cultural At the same time they pointed out that some of the items specified for each class are the same as in other classes overlap (Kazemi, 2004) Due to the influence and impact of tourism on the social and economic structure of society, can be noted The proper facilities and services for economic and social policy can be expanded in a certain period of booming tourism demand will be more Conversely, they cannot demand a more restrictive and more restrictive leverage returns Page 45 Page 45 http://www.ijmsbr.com International Journal of Management Sciences and Business Research, 2013 ISSN (2226-8235) Vol-3, Issue 1 International Journal of Management Sciences and Business Research, 2013 ISSN (2226-8235) Vol-3, Issue 1 Materials and Methods This prospective study was conducted using a comprehensive system approach to research - descriptive analysis. Mostly through literature review and resources related theoretical research has been developed. Then the status of questionnaires and statistical analysis methods were included in the study. Complete field data from the questionnaire were used. This respect between tourists scattered 400 questionnaires, of which 350 questionnaires were returned and analyzed. Cochran formula is used to decide the sample size. Research suggests that these problems are rarely the main reason for this is that tourism but can they provide developments "(World Tourism Organization 1994, 82). Research suggests that these problems are rarely the main reason for this is that tourism but can they provide developments "(World Tourism Organization 1994, 82). "Cultural environment - social media activity is tourism. Thus, culture and tourism community interact with each other. This discussion is related to two aspects: - Impact and influence on each other s culture and society, culture and tourism community. - The impact of tourism on culture and society. "(Ibid. 55). p y ( ) "The benefits of tourism to the local community and cultural patterns and also bring problems. Social work - cultural tourism, including tourism development in each region are the most important considerations, however, measure the impact of the economic and environmental impact is far more difficult. In countries that are still socially and economically very traditional, very critical of these works are important. Although the incidence of social tourism - cultural, but it should be remembered that in principle lead to the development of any new changes are sought. Tourism is the only source of change in society "(World Tourism Organization 1994, 78-79). Economic consequences usually measurable and manifest contrast, only some of the social costs of tourism are obvious, but in other cases not. In the past, the Dutch capital of Amsterdam was the center of attraction hippie and drug addicts. The city's reputation was tarnished, which is a social cost (Lund Berg and others 1995, 83). "Social problems caused by drug addiction, alcoholism, crime, and prostitution, as well as the tourists are likely to intensify. Theoretical Principles Research In total, a staggering increase in tourism due to improved communications, increased wealth and prosperity and more leisure is (Bonnie Fee 1995, 15) Tourism or tourism industry since 1950 with great changes that some have called it a revolution tourism this led to the development and maturation of "mass tourism" and in the near future "quality tourism" will become. We have a strong and professional tourism activity worldwide is great So that before the first oil, then placed in a row and then above it is growing (Masumi 2007, 9). p g g ( ) "According to the World Tourism Organization, which is affiliated to the United Nations, tourism is the largest industry in the world" (Lund Berg and others, 1995, 3) The first travel in most countries and regions to exploit its economic interests" (World Tourism Organization 2001, 66) The World Tourism Organization statistics that show offers an important impact on tourism in the global economic system, the third largest industry after oil and automobile industries, dynamic and growing industry is the only industry where the introduction "(Nasser M. 2004). "As a force for social change, tourism has been the impact of the Industrial Revolution. In less than three decades, tourism has transformed the world's view of business ". Today tourism has become one of the greatest levers of economic social change in human history So that the number of international tourists in 1950, which did not exceed 25 million in 1992 have been to over 594 million people . According to the World Tourism Organization predicts the number of tourists in 2010 to one billion people in 2020 to 6/1 billion people will increase (Sacrifice 2004). The magnitude of tourism to create jobs and income is not limited. If the thought of planning and development, direct and indirect benefits of tourism to the economic, social, cultural and political environment and make a significant contribution to play in national development In other words, to promote social justice indicators, improved living standards, prosperity and regional balance lead. In some cases, regional economists, tourism as the only resource deployment and development of less developed regions have seen "(Masumi 2007, 10). Page 46 Page 46 http://www.ijmsbr.com International Journal of Management Sciences and Business Research, 2013 ISSN (2226-8235) Vol-3, Issue 1 International Journal of Management Sciences and Business Research, 2013 ISSN (2226-8235) Vol-3, Issue 1 "Cultural environment - social media activity is tourism. Thus, culture and tourism community interact with each other. This discussion is related to two aspects: "Cultural environment - social media activity is tourism. Thus, culture and tourism community interact with each other. This discussion is related to two aspects: Analysis y Tourist attractions in the city Zahak 38 were identified, of which about 10 attractions and 28 Natural attractions, places of historical, cultural and religious needs. Given the natural attractions are low, but most tourists are attracted. Attractions, historical, cultural and religious are too low given that attract tourists. Table (1): Composition of tourist attractions and city Zahak Type The number Percent Total 38 100 Natural 10 26.31 Cultural, historical, religious 28 73.69 References: Administration of Cultural Heritage and Tourism Zahak, 2013 Natural and humans making tourism city Zahak The map below about 10 natural point of tourism in the city Zahak identified include Chahnimeh first and fourth integrated Baqiyatallah Azam along well in the first half, built a wiki of points that most attracts tourists throughout the year. Dams include (dam Jarykah - Kahak - Zahak and Sistan), the Helmand River, forests (Jazink, Niatak, Tiflak and Shirdal). Page 47 http://www.ijmsbr.com http://www.ijmsbr.com International Journal of Management Sciences and Business Research, 2013 ISSN (2226-8235) Vol-3, Issue 1 International Journal of Management Sciences and Business Research, 2013 ISSN (2226-8235) Vol-3, Issue 1 International Journal of Management Sciences and Business Research, 2013 ISSN (2226-8235) Vol-3, Issue 1 Figure (1): natural attractions tourist city Zahak, Source: Authors, 2013. Spatial structure of historical attractions - cultural Zahak religious city According to the map, including 28 historical attractions, cultural and religious Zahak were identified in the city. Figure (1): natural attractions tourist city Zahak, Source: Authors, 2013. Figure (1): natural attractions tourist city Zahak, Source: Authors, 2013. Spatial structure of historical attractions - cultural Zahak religious city According to the map, including 28 historical attractions, cultural and reli the city. Spatial structure of historical attractions - cultural Zahak religious city Spatial structure of historical attractions - cultural Zahak religious city According to the map, including 28 historical attractions, cultural and religious Zahak were identified in the city. Spatial structure of historical attractions - cultural Zahak religious city According to the map, including 28 historical attractions, cultural and religious Zahak were identified in the city. Figure (2): Natural attractions tourist city Zahak, Source: Authors, 2013 Figure (2): Natural attractions tourist city Zahak, Source: Authors, 2013 Examine connection between some variables Examine connection between some variables Evaluation of connection satisfaction collision and another trip to town residents Zahak To study the connection between two variables, the satisfaction rate of people treated Zahak and another trip to this city «Somers D test» was used. The results of these tests show that most people are very satisfied with the deal people are traveling back to the city. Most people who have low satisfaction and stop from further travel. This result is due to the significance level of statistical tests for general population and significant relation between sig = 0.000 D. Summers, less than (0.05) of two variables satisfying meet townspeople Zahak and another trip to the city to verify. Evaluation of connection satisfaction collision and another trip to town residents Zahak To study the connection between two variables, the satisfaction rate of people treated Zahak and another trip to this city «Somers D test» was used. To study the connection between two variables, the satisfaction rate of people treated Zahak and another trip to this city «Somers D test» was used. The results of these tests show that most people are very satisfied with the deal people are traveling back to the city. Most people who have low satisfaction and stop from further travel. This result is due to the significance level of statistical tests for general population and significant relation between sig = 0.000 D. Summers, less than (0.05) of two variables satisfying meet townspeople Zahak and another trip to the city to verify. Examine connection between some variables http://www.ijmsbr.com Page 48 Table (2): Evaluation of connection satisfaction collision and another trip to town residents Zahak Table (2): Evaluation of connection satisfaction collision and another trip to town residents Zahak Page 48 http://www.ijmsbr.com http://www.ijmsbr.com International Journal of Management Sciences and Business Research, 2013 ISSN (2226-8235) Vol-3, Issue 1 Total Very much high Average Low Very little 4.0 100.0 Very little trip to town residents Zahak 6.1 14.3 100.0 Low 32.3 76.5 85.7 Average 40.4 65.3 23.5 high 17.2 34.7 Very much 100.0 100.0 100.0 100.0 100.0 100.0 Total Amount statistic: 23.635 Level of significance: 0.000 Source: Authors, 2013 nal of Management Sciences and Business Research, 2013 ISSN (2226-8235) Vol-3, Issue 1 Find out the Connection between gender and the impact of tourism on economic development To test these two variables test «t-test for 2 independent means» is used, the results of this test show that men have a more positive opinion about the potential impact of tourism on economic development. So the mean male perspective (5) higher than the mean score of women view of (4.33) is. It can therefore be concluded sig = 0.000, given the test is less than the significance level (0.05) and the mean differences between the views of men and women about the impact of the tourism potentials of the economic development of the statistically significant is confirmed. Table (3): Find out the Connection between gender and the impact of tourism on economic development Significance level Amount statistic t Standard error of the mean Standard deviation Average gender 0.000 9.527 .00000 .00000 5.0000 man .06997 .54202 4.3333 Woman Source: Authors, 2013 Source: Authors, 2013 Find out the connection between age and abilities of the impact of tourism on economical development T t t th ti b t d th i t f t i i l d l t d t t biliti Find out the connection between age and abilities of the impact of tourism on economical development To test the connection between age and the impact of tourism on economical development and test abilities. «Kendall tau b test» was used. The test results show the higher the age, the fewer faith in the abilities of the impact of tourism on economical development. Evaluate the potential impact on tourism related jobs and economic developmen This result is due to the significance level of the test is less than 0.05sig = 0.000 statistically generalize able to the entire population and therefore we can say that a significant connection between the two variables is confirmed. Table (6): Evaluate the potential impact on tourism related jobs and economic development Level of significance Amount statistic f Average Job 0.000 455.891 5.0000 Employee 5.0000 Self-employed 5.0000 pensionary 5.0000 Currently studying 4.8889 Worker 4.0000 Other Source: Authors, 2013 Explore the connection between income and potential impact of tourism on economical development To test the connection between these variables and test «t-test for 2 independent means» is used, the results of this test show that people who earn over one million dollars are less positive about the impact of tourism potentials economic development and people earning less favorable than the impact potentials of tourism in economic development, the conclusion on the significance level of the test) is generalize able to the target population and significant relation between the two variables, sig = 0.000 less 0.05 will be verified. Table (6): Evaluate the potential impact on tourism related jobs and economic development Level of significance Amount statistic f Average Job 0.000 455.891 5.0000 Employee 5.0000 Self-employed 5.0000 pensionary 5.0000 Currently studying 4.8889 Worker 4.0000 Other Source: Authors, 2013 Table (6): Evaluate the potential impact on tourism related jobs and economic development Level of significance Amount statistic f Average Job 0.000 455.891 5.0000 Employee 5.0000 Self-employed 5.0000 pensionary 5.0000 Currently studying 4.8889 Worker 4.0000 Other Source: Authors, 2013 Source: Authors, 2013 tion between income and potential impact of tourism on economical development b h i bl d f i d d i d h l Explore the connection between income and potential impact of tourism on economical development To test the connection between these variables and test «t-test for 2 independent means» is used, the results of this test show that people who earn over one million dollars are less positive about the impact of tourism potentials economic development and people earning less favorable than the impact potentials of tourism in economic development, the conclusion on the significance level of the test) is generalize able to the target population and significant relation between the two variables, sig = 0.000 less 0.05 will be verified. Examine connection between some variables And contrary to what people are younger, their belief in the economic development of the tourism potential impact is greater, given that the result is less than the significance level of the test statistic is the generalizations of indicates a significant relation between the two variables confirm (sig = 0.000) is 0.05. http://www.ijmsbr.com Page 49 Table (4): Find out the connection between age and abilities of the impact of tourism on economical development Age Total 0-14 15-34 35-59 +60 Tourism impact on economic development Average 9.5 1.3 high 42.5 90.5 24.0 Very much 100.0 100.0 57.5 74.7 Total 100.0 100.0 100.0 100.0 100.0 Amount statistic: -10.216 Level of significance: 0.000 Source: Authors, 2013 Find out the connection between age and abilities of the impact of tourism on economical development A T International Journal of Management Sciences and Business Research, 2013 ISSN (2226-8235) Vol-3, Issue 1 Find out the connection between education and tourism potential impact on economical development To test the connection between Kendall tests is useless.The results of these tests show the more educated people have higher ability impact of tourism on economical development than they know. Conversely, a decline in tourism and education powers of people believes that they have a very powerful impact on economic development. This result is following the test significance level of less than 0.01, sig = 0.000 generalize able to the target population and significant relation between the two variables is confirmed. Table (5): Find out the connection between education and tourism potential impact on economical development Total education Masters Bachelor of science Technician Diplom a Under diploma Illiterat e 22.2 100.0 100.0 3.8 high Tourism impact on economic development 77.8 96.2 100.0 100.0 100.0 Very much 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Total Amount statistic:-8962 Level of significance: 0.000 Source: Authors, 2013 Table (5): Find out the connection between education and tourism potential impact on economical dev Total education Source: Authors, 2013 Source: Authors, 2013 aluate the potential impact on tourism related jobs and economic development Evaluate the potential impact on tourism related jobs and economic development First theory: y the optimal use of the abilities and potential of tourism to economical development will Zahak. To test the hypothesis that the normal state of the variables examined. Test to check the normality of the variables «Kolmogorov-Smirnov test» was used. The results are given in the following table. the optimal use of the abilities and potential of tourism to economical development will Zahak. To test the hypothesis that the normal state of the variables examined. Test to check the normality of the variables «Kolmogorov-Smirnov test» was used. The results are given in the following table. Table (8): Evaluation of the normal status variables Kolmogorov- Smirnov Standard deviation Average Variables Row 0.000 0.44 3.375 Tourism activities 1 0.002 0.76 2.144 Economic Development 2 Source: Authors 2013 Source: Authors, 2013 , The results of the test «Kolmogorov-Smirnov test» in connection with the distribution of variable features and capabilities of both tourism and economic development of the show. These two variables have a normal distribution, so nonparametric tests should be used to assess the connection between them. Evaluate the potential impact on tourism related jobs and economic developmen Explore the connection between income and potential impact of tourism on economical development To test the connection between these variables and test «t-test for 2 independent means» is used, the results of this test show that people who earn over one million dollars are less positive about the impact of tourism potentials economic development and people earning less favorable than the impact potentials of tourism in economic development, the conclusion on the significance level of the test) is generalize able to the target population and significant relation between the two variables, sig = 0.000 less 0.05 will be verified. Explore the connection between income and potential impact of tourism on econom Page 50 http://www.ijmsbr.com International Journal of Management Sciences and Business Research, 2013 ISSN (2226-8235) Vol-3, Issue 1 Table (7): Explore the connection between income and potential impact of tourism on economical development Level of significance Amount statistic f Average Job 0.000 573.500 5.0000 Less than 3000 thousand Riyals 5.0000 Between 3000 and 6000 thousand Riyals 5.0000 Between 6000 thousand to 10 million Rials 4.0588 Above 10 million Rials Source: Authors, 2013 nternational Journal of Management Sciences and Business Research, 2013 ISSN (2226-8235 International Journal of Management Sciences and Business Research, 2013 ISSN (2226-8235) Vol-3, Issue 1 Examining the connection between usability and capabilities of Tourism and Economic Development Test was used to assess the connection between these two variables is the spearman correlation coefficient. The results of the test «spearman correlation coefficient» show highly variable connection between tourism and economic development of the abilities and abilities of 0.945 and the equation is positive. Which shows a strong positive connection exists between these two variables. The functionality and abilities of the tourist town Zahak greater economic development in the city is higher and vice versa. This result is due to a less than significant level test) is generalize able to the target population. The researcher’s theory is confirmed between the two variables sig = 0.000, 0.01 (features and capabilities, there is a positive significant connection between tourism and economic development. Table (9): Examining the connection between usability and capabilities of Tourism and Economic Development Value Statistics 0.945 The Spearman correlation coefficient 0.000 Level of significance Source: Authors, 2013 Source: Authors, 2013 According to the correlation coefficient of two variables 0.945 the following diagram shows a strong positive connection exists between these two variables. Page 51 http://www.ijmsbr.com http://www.ijmsbr.com International Journal of Management Sciences and Business Research, 2013 ISSN (2226-8235) Vol-3, Issue 1 Figure (3): investigate the relationship between features and capabilities of Tourism and Economic Development Source: Authors, 2013 International Journal of Management Sciences and Business Research, 2013 ISSN (2226-8235) Vol-3, Issue 1 Figure (3): investigate the relationship between features and capabilities of Tourism and Economic Development Source: Authors, 2013 e relationship between features and capabilities of Tourism and Economic Development Source: Authors, Figure (3): investigate the relationship between features and capabilities of Tourism and Economic Development Source: Authors, 2013 y the level of employment and tourism activities Zahak people were in town there is a significant connection. the level of employment and tourism activities Zahak people were in town there is a significant connection. To check the status of normal variables tourism activity and higher levels of employment and income test «Kolmogorov-Smirnov test» was used. The results of these tests show that the assumption of normality is rejected by these two variables, and so they do not have a normal distribution. Therefore, nonparametric tests were used to test the connection between the two. Here is a test to examine the connection between these two variables has been used is the Spearman correlation test. Table (10): Evaluation of the normal status variables Kolmogorov- Smirnov Standard deviation Average Variables Row 0.000 0.67 1.820 Tourism activities 1 0.000 0.57 1.262 Level of employment and income 2 Source: Authors, 2013 The connection between tourism activity and increased levels of employment and income The results of the test «spearman correlation coefficient» show highly variable connection betwe y p y The results of the test «spearman correlation coefficient» show highly variable connection between tourism activity and an increase in income and employment levels 0.769 and the equation is positive. So we can say that there is a strong positive connection between these two variables. The higher the level of tourism activity in the area of employment and higher income residents, and vice versa. This result is due to the significance level of the test is less) is generalize able to the target population and the researcher’s hypothesis is confirmed. Among the sig = 0.000, 0.01 (Zahak and increased tourism activity in the employment and income, there is a significant connection. Table (11): The connection between tourism activity and increased levels of employment and income Values Statistics 0.769 The Spearman correlation coefficient 0.000 Level of significance Source: Authors, 2013 Table (11): The connection between tourism activity and increased levels of employment and income Values Statistics 0.769 The Spearman correlation coefficient 0.000 Level of significance Source: Authors, 2013 Source: Authors, 2013 http://www.ijmsbr.com Page 52 According to the correlation coefficient of two variables 0.769 the graph below shows that the connection is positive. According to the correlation coefficient of two variables 0.769 the graph below shows that the connection is positive. According to the correlation coefficient of two variables 0.769 the graph below shows that the connection is positive. Conclusions Zahak city due to historical, cultural and natural heritage that is to be hoped that the development of the tourism industry and discovered that: Zahak city due to historical, cultural and natural heritage that is to be hoped that the development of the tourism industry and discovered that: The first problem is to develop tourism in this city tourist attractions remain unknown. As the first step of this work must be identified and its attraction to tourists frequently presented with advertising. Another point that should be addressed in the planning principles for the development of urban economy with tourism. If appropriate facilities are provided, and a logical approach to analyze of various aspects of the future of this region can be transformed into a tourist hub. Given the high unemployment rate in the city's tourism industry can create jobs for the unemployed. Tubs are engaged in work related services to tourists. And with lead to the economic development of the region and provided Zahak. Which will provide the background to the economic development of the country? Others who should be serious consideration on the development of motels, hotels, recreational facilities and services to provide greater urban area Zahak attraction to tourists in Zahak more time to stop. Thereby providing improved hotel facilities and local attractions, tourists will provide long life. Because more tourists are longer lasting, more Zahak will result in the economic development of the city. Note the development of crafts. The development of crafts can be fertile ground for job creation and revenue. Both the copacetic this city a positive impact. And, as creativity can be a symbol of lifestyle, attitudes and manners of the natives. Development and crafts, a good advertisement for the city will introduce the customs. The results of the first research hypothesis based on the optimal use of the capabilities and potential of tourism to economic development will Zahak. The intensity of the relationship between the two variables is 0.945 usability and capabilities of tourism and economic development. And the relationship is positive. Which indicates a strong positive relationship exists between these two variables. The functionality and capabilities of the tourist town Zahak further economic development in the city is higher and vice versa. The results of the second hypothesis test between tourism activities employment and income levels in the city Zahak significant connection exists. y the level of employment and tourism activities Zahak people were in town there is a significant connection. http://www.ijmsbr.com International Journal of Management Sciences and Business Research, 2013 ISSN (2226-8235) Vol-3, Issue 1 Figure (4): investigate the relationship between tourism activities and the level of employment and income Source: Authors, 2013 International Journal of Management Sciences and Business Research, 2013 ISSN (2226-8235) Vol-3, Issue 1 Figure (4): investigate the relationship between tourism activities and the level of employment and income Source: Authors, 2013 Conclusions Shows the connection between two variables, severity, tourism activities and increasing income and employment levels 0.769 and the equation is positive. So we can say that there is a strong positive connection between these two variables. The higher the level of tourism activity in the area of employment and higher income residents, and vice versa. Suggestions gg development seminars and meetings with other agencies and officials associated with the investment in the tourism industry and entrepreneurs develop and strengthen the publicity and information programs to introduce the province's natural attractions, historical and cultural features of the hospitality of the people of the City Zahak for employment and income for residents of The main focus of the activities on sampling develop and strengthen the publicity and information programs to introduce the province's natural attractions, historical and cultural features of the hospitality of the people of the City Zahak for employment and income for residents of The main focus of the activities on sampling p g also review the types and how to take advantage of people's contributions to the development, preparation and implementation of urban plans and equipment beyond the city terms of tourist facilities and services also review the types and how to take advantage of people's contributions to the development, preparation and implementation of urban plans and equipment beyond the city terms of tourist facilities and services Page 53 http://www.ijmsbr.com International Journal of Management Sciences and Business Research, 2013 ISSN (2226-8235) Vol-3, Issue 1 necessary Specialists from various institutions in the areas of rehabilitation and development measures are necessary. inviting local and foreign investors to invest in construction of facilities and special rates for hotels, residential complexes, resorts, forest, creating sports fields, and ... tourist resources and tourist attractions such as the free use of the potential of cultural and historical Zahak economic use of resources, tourism and the tourism industry categories - commercial plant and related workshops http://www.ijmsbr.com references  Absolute hero, Hassan. 1997. Geographical impacts of tourism in the city of Hamedan. Master's thesis, martyr Beheshti University.  Absolute hero, Hassan. 1997. Geographical impacts of tourism in the city of Hamedan. Master's thesis, martyr Beheshti University.  Ali Asgari, Abdul Ali. 1997. Organization and management of the tourism sector. MS Thesis, Tehran University.  Ali Asgari, Abdul Ali. 1997. Organization and management of the tourism sector. MS Thesis, Tehran University.  Anonymous, Vahid. 1996. Sustainable development and tourism. Journal of Political Information Economics No. 95-96.  Anonymous, Vahid. 1996. Sustainable development and tourism. Journal of Political Information Economics No. 95-96.  Bonnie Face, Priscilla. , 1995. Management of cultural tourism. Mahmoud Translations Abdolahzadeh 0.2002. Tehran: Office for Cultural Research.  Bonnie Face, Priscilla. , 1995. Management of cultural tourism. Mahmoud Translations Abdolahzadeh 0.2002. Tehran: Office for Cultural Research.  Dakhil Kohnamoui, J. 2004. Of improving tourism in East Azerbaijan. Master's thesis, University of Tabriz.  Dakhil Kohnamoui, J. 2004. Of improving tourism in East Azerbaijan. Master's thesis, University of Tabriz.  Joker, M. 2006. Nature tourism. Tehran: Peek Kosar.  Kazemi, Mehdi. 2004. Economic and cultural effects of tourism development from the perspective of residents. Journal of Tourism Studies. Number 1:3-16.  Kazemi, Mehdi. 2004. Economic and cultural effects of tourism development from the perspective of residents. Journal of Tourism Studies. Number 1:3-16.  Landbrg, Donald, Filomena M. Krishna, H. Mink. , 1995. Tourism economy. Translated by Mohammad Reza Farzin. 2005. Tehran: commercial publishing company.  Landbrg, Donald, Filomena M. Krishna, H. Mink. , 1995. Tourism economy. Translated by Mohammad Reza Farzin. 2005. Tehran: commercial publishing company.  Mansoori, Ali. 2003. Tourism and sustainable development. Journal of Geography Education. Number 41:36-63.  Mansoori, Ali. 2003. Tourism and sustainable development. Journal of Geography Education. Number 41:36-63.  Movahed, A., 2008, Urban Tourism, University of Shahydchmran  Movahed, A., 2008, Urban Tourism, University of Shahydchmran  Naser M, Nader. 2004. Review and explain the barriers to tourism development in the province of Lorestan and determination in this area. Master's thesis, University of Mazandaran. Movahed, A., 2008, Urban Tourism, University of Shahydchmran  Naser M, Nader. 2004. Review and explain the barriers to tourism development in the province of Lorestan and determination in this area. Master's thesis, University of Mazandaran.  Naser M, Nader. 2004. Review and explain the barriers to tourism developmen Lorestan and determination in this area. references Master's thesis, University of Mazandara orouzi, A. and et al, 2011, The role of tourism in urban economics, case study Sar rovince.  The victim, the messenger. 2004. Tourism and dynamics of urban space. Journal of Faculty of Humanities and Social Sciences University. Ninth year. No. 3. World Tourism Organization. 1994. National and regional tourism planning. B Ranjbarian, Mohammad Zahedi. 2001. Isfahan: Jahad University, Isfahan unit.  Yousefi Pour, Gh. R, 2001, The role of tourism in Iran, and ways to expand its economic development  Yousefi Pour, Gh. R, 2001, The role of tourism in Iran, and ways to expand its economic development http://www.ijmsbr.com Page 54
https://openalex.org/W2083379765	https://zenodo.org/records/1570346/files/article.pdf	English	null	XXIV.—The formation of heterocyclic compounds	Journal of the Chemical Society. Transactions	1,900	public-domain	5,815	View Article Online / Journal Homepage / Table of Contents for this issue View Article Online / Journal Homepage / Table of Contents for this issue 239 THE FORMATION OF HETEROCYCLIC COMPOUNDS, Downloaded by OHIO STATE UNIVERSITY on 19 March 2013 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 Downloaded by OHIO STATE UNIVERSITY on 19 March 2 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039 XX1V.-The Formation of Heterocyclic Compoufids, By SIEGFRIED RUHEMANN and H. E. STAPLETON, Scholar of St. John's College, Oxford. THE research on the formation of a-pyrone compounds (Ruhemann, Trans., 1899,75,245, and 41 I), by the action of ethyl phenylpropiolate on the sodium derivatives of P-diketones and the ethyl esters of P-ketonic acids, induced one of us and A. V. Cunnington to examine the be- haviour of organic bases towards the ethyl esters of the acids of the acetylene series, with the view of ascertaining whether in those cases also additive substances were formed which subsequently lost alcohol, and condensed to heterocyclic compounds. A few months ago (Trans., 1899,75, 954) we published the first part of o w investigations in this direction, and were able to show that by the interaction of benz- amidine and ethyl phenylpropiolate a cyclic compound was produced. The constitution of this substance we represented by the symbol, A closer examination of the reaction has led the authors of the present paper to the discovery of the following interesting fact. At the ordinary temperature, benxalphenylglyoxalidone is formed almost exclusively, but on digesting the mixture of the ethyl ester and the amidine on the water-bath, besides this substance another is pro- duced which has been identified as diphenyJpyrimidone, 240 RUHEMANN AND STAPLETON : View Article Online This result can readily be explained on the assumption that benz- amidine and ethyl phenylpropiolate first react with formation of the benzamidide of phenylpropiolic acid, Downloaded by OHIO STATE UNIVERSITY on 19 March 2013 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 and that subsequently, according to the conditions which exist, the hydrogen atom ( x ) unites either to the p- or the a-carbon atom, in the first case a 5-membered, and in the second a 6-membered ring being pro- duced. The view expressed above as to the intermediate phase of the reaction between benzamidine and ethyl phenylpropiolate is also sup- ported by the results of the study of the behaviour of other bases towards ethyl phenylpropiolate. Downloaded by OHIO STATE UNIVERSITY on 19 March 2013 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 241 View Article Online 241 Online From isoxazole, or the dihydroxyisoxazoles, there may be derived three isoxazolones, which are represented by the following symbols : From isoxazole, or the dihydroxyisoxazoles, there may be derived three isoxazolones, which are represented by the following symbols : p - p 2 EH-70 FO-EH \o./ \O’ \O’ N co, N CH,, NH CH. I. 11. IIJ. p - p 2 EH-70 FO-EH \o./ \O’ \O’ N co, N CH,, NH CH. I. 11. IIJ. ded by OHIO STATE UNIVERSITY on 19 March 2013 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 The first isoxazolones which were obtained belonged to type I, and have been studied especially by Claisen (Ber., 1891, 24, lal), whilst some derivatives of 4-isoxazolone (11) were prepared by Abenius and Soderbaum (Ber., 1892, 25, 3468). The compound which is formed by the interaction of hydroxylamine and ethyl phenylpropiolate is therefore to be regarded as a 3-substitution product of type 111. The conclusion arrived at with regard to the mode of the inter- action of benzamidine and hydroxylamine with ethyl phenylpropiolate induced us also to re-examine the product which is formed on allowing a mixture of alcoholic solutions of sodium ethoxide, urea, and ethyl phenylpropiolate to stand at the ordinary temperature. This sub- stance, which had been regarded before by one of us and A. V. Cunning- ton (Zoc. cit.) as the ureide of the formula NH,. CO NH COCiCC,H5, we find, however, to be a cyclic compound. Its constitution is un- doubtedly to be expressed by the symbol, which characterises it as benza2hydantoi.n. This view is supported by the following facts. The substance, as mentioned previously, has acidic properties, dissolving freely in caustic potash. On boiling the alkaline solution, the odour of benzaldehyde is first perceptible, then am- monia is evolved, and finally an acid results, which we have found to be identical with Pliichl’s phenylglycidic acid ” (Bey., 1883, 16, 2817). This acid, which has since been obtained by Wislicenus from ethyl phenyloxaloacetate (Bey., 1887,20, 591), has undoubtedly either the formula C6H,* CH,* COCO,H, or C,H, CH:C(OH)CO,H. Although E. Erlenmeyer, jun. (Annulen, 1892, 271, 137), assigned to the acid the first formula, yet the second, which characterises it as a-hydroxy- cinnamic acid, would also agree with the behaviour of the acid as described by him. The mode of its formation, and its relation to the A which characterises it as benza2hydantoi.n. Downloaded by OHIO STATE UNIVERSITY on 19 March 2013 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 It had been previously assumed that addition of the base to the ester would precede the ring-formation, which would be effected by subsequent removal of alcohol, and we were led to this assumption by the fact that ethyl phenylpropiolate forms additive products, not only with the sodium derivatives of ethy malonate and ketonic compounds, but also with secondary bases such as diethylamine and piperidine, Since the action of hydroxylamine on ethyl phenylpropiolate does not yield 3-phenyl-5-isoxazolone according to the equation, and that subsequently, according to the conditions which exist, the hydrogen atom ( x ) unites either to the p- or the a-carbon atom, in the first case a 5-membered, and in the second a 6-membered ring being pro- duced. The view expressed above as to the intermediate phase of the reaction between benzamidine and ethyl phenylpropiolate is also sup- ported by the results of the study of the behaviour of other bases towards ethyl phenylpropiolate. It had been previously assumed that addition of the base to the ester would precede the ring-formation, which would be effected by subsequent removal of alcohol, and we were led to this assumption by the fact that ethyl phenylpropiolate forms additive products, not only with the sodium derivatives of ethy malonate and ketonic compounds, but also with secondary bases such as diethylamine and piperidine, Since the action of hydroxylamine on ethyl phenylpropiolate does not yield 3-phenyl-5-isoxazolone according to the equation, C6H5 3 ; i y H 2 C6H,* cicco2c2H5 + NH2 OH =I N 5co +C$5OH, ‘0’ it was supposed that the product of the reaction was phenylpropiolo- hydroxamic acid, C6H5 CICC<koH. A closer investigation of this it was supposed that the product of the reaction was phenylpropiolo- hydroxamic acid, C6H5 CICC<koH. A closer investigation of this substance has, however, proved it to be a cyclic compound isomeric with 3-phenyl-5-isoxazolone. Its formation can only be interpreted by assuming that the action takes place in the manner expressed by the following equation : g q The compound appears, therefore, as 5-p~nyZ-3-isoxazolone, or as its tautomeride, 5-phenyl-3-hydroxyisoxazo~e, THE FORMATION OF HETEROCYCLIC COMPOUNDS. Downloaded by OHIO STATE UNIVERSITY on 19 March 2013 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 This view is supported by the following facts. The substance, as mentioned previously, has acidic properties, dissolving freely in caustic potash. On boiling the alkaline solution, the odour of benzaldehyde is first perceptible, then am- monia is evolved, and finally an acid results, which we have found to be identical with Pliichl’s phenylglycidic acid ” (Bey., 1883, 16, 2817). This acid, which has since been obtained by Wislicenus from ethyl phenyloxaloacetate (Bey., 1887,20, 591), has undoubtedly either the formula C6H, CH,* COCO,H, or C,H, CH:C(OH)CO,H. Although E. Erlenmeyer, jun. (Annulen, 1892, 271, 137), assigned to the acid the first formula, yet the second, which characterises it as a-hydroxy- cinnamic acid, would also agree with the behaviour of the acid as described by him. The mode of its formation, and its relation to the A A isomeride, phenyIglycidic acid, C,H, CH-CHI CO,H, would thus find a ready explanation. This formula indicates the probable ex- istence of two stereoisomeric forms, B YH BH C,H,C :CCO,H and C,H,Y:CCO,H, B YH BH C,H,C :CCO,H and C,H,Y:CCO,H, H B YH BH C,H,C :CCO,H and C,H,Y:CCO,H, H 242 RUHEMANN AND STAPLETON : View Article Online The first grouping could readily change into pheny lglycidic acid, whilst the second might be expected to be the more stable. Moreover, the formation of phenylglycidic acid takes place by the action of cold caustic potash on a-c hlorophenyllactic acid, that of a-hydroxy cinna mic acid, on the other hand, under the influence of the boiling reagent from compounds of the cinnamyl series. These facts are in full agreement with the researches of Fittig and his pupils on the trans- formation of unsaturated acids into their stereoisomerides, and suggest that boiling caustic potash would partially change phenylglycidic acid into a-hydroxycinnamic acid. Downloaded by OHIO STATE UNIVERSITY on 19 March 2013 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 Downloaded by OHIO STATE UNIVERSITY on 19 March 20 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/C Downloaded by OHIO STATE UNIVERSITY on 19 March 20 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/C The fact that ferric chloride gives a green colour with a-hydroxy- cinnamic acid (p. 246) whilst it gives a violet with esters of /3-ketonic acids, such as ethyl acetoacetate, seems to indicate that the structure of the former is similar to that of catechol, just as the structure of acetoacetic acid is similar to that of resorcinol. This similarity is shown by the formulae : CH CH COH COH Catechol. a-Hydroxycinnamic Eesorcinol. Ace toacetic acid. acid. Thiourea and guanidine react with ethyl phenylpropiolate in a manner analogous to urea and yield respectively benxalthiohydantoin and benxalimidohydantoirt. These substances are similarly decomposed by caustic potash, and yield a-hydroxycinnamic acid. The destruction of the ring takes place in these cases with the same readiness, and is accompanied by the production of benzaldehyde. Benzalphenylgly- oxalidone, on the other hand, is more stable, only partial decomposition being effected by boiling with the alkali, but a small quantity of the hydroxy-acid is certainly formed. Besides the record of our work in the direction indicated above, this communication contains the results at which we have arrived, up to the present, by using ethyl acetylenedicarboxylate instead of ethyl phenylpropiolate. The study of the action of organic bases on this salt appeared to us to be of especial interest, since it had been found (Ruhemann and Cunnington, Trans., 1899, 75, 784) that a-pyrone derivatives do not result from its union with the ethyl esters of &ketonic acids. This fact led to the conclusion that both hydrogen atoms of the CH2 group of the ketonic compound play a part in the reaction. Of the two expressions (R denoting CH, or C6H5), I. 11. CO,C,H,* ;C;* CH2* C02C2H5 C02C2H5* FIC GO&&&, CO,C,H,* C’C0.R and C02C2H5* CH >c<co.R 11. 11. 11. C02C2H5* FIC GO&&&, C02C2H5* CH >c<co.R I. C* CO,C,H,* ;C;* CH2* C02C2H5 CO,C,H,* C’C0.R THE FORMATION OF HETEROCYCLIC COMPOUNDS. Downloaded by OHIO STATE UNIVERSITY on 19 March 2013 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 View Artic 243 nline which thus have to be considered, preference had been given to the second, since ammonia does not act on these additive compound8 with formation of pyridine derivatives." On the other hand, the behaviour of o-phenylenediamine towards ethyl acetylenedicarboxylate points to the first formula, as we find that the diamine reacts with this salt, forming ethyl puinoxalidoneacetate, according to the equation Downloaded by OHIO STATE UNIVERSITY on 19 March 2013 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 ded by OHIO STATE UNIVERSITY on 19 March 2013 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 Downloaded by OHIO STATE UNIVERSITY on 19 March 2013 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 - - OC,H,* 70 Ci C* C02C2H, C H < N H 2 -+- NH, NHCO C2H5 OH + C 6 H 4 < ~ = 6. CH,, co c H s 2 2 5 This result would correspond to the view expressed in formula I regarding the constitution of the products formed by the union of ethyl acetoacetate and ethyl benzoylacetate with ethyl acetylene- dicarboxylate. Moreover, this formula is supported by the following considerations. It may be assumed that ethyl malonate would re- semble the ester of the B-ketonic acids in its reaction with ethyl acetylenedicarboxylate. If symbol I1 were assigned to the former additive products, the conclusion may be drawn that ethyl malonate would also yield a trimethylene derivative with ethyl acetylenedi- carboxylate. Michael (J. p . Chew., 1894, [ii], 49, 20) however, has shown that the additive compound is the ester of an unsaturated acid, since, on hydrolysis, i h yields aconitic acid. These facts, whilst indi- cating that the substances formed by the union of the esters of P-ketonic acids with ethyl acetylenedicarboxylate are to be represented by formula I, would at the same time lead to the assumption that the constitution of the additive product of ethyl malonate with ethyl acetylenedicarboxylate is not EH* C02C2H, 7H2* C02C2H, CO,C,HI; but 8C02C2H5 H(C02c2H5)2 C(CO2C2HJ2 The latter formula would not be at variance with the formation of aconitic acid from the substance which it represents. It further agrees with the result of the action of ammonia on the ethyl ester of the tetracarboxylic acid. As shown below, there is thus formed ethyl 2 6-dihpdroxypyridine-3 : 4-dicacrboxyZctte, CC02C2H, CO,C,K,C(\CH OH.C,~COH ' N CC02C2H, CO,C,K,C(\CH OH.C,~COH ' N As will be shown in a later communication, these esters are identical with those obtained by the action of ethyl chlorofumarate on the sodium derivatives of ethyl acetoacetate and ethyl benzoylacetate (Trans., 1896, 69, 630, 1383 ; 1897, 71, 323). 244 RUHEMANN AND STAPLETON View Article Online which, on hydrolysis with hydrochloric aoid, loses carbon dioxide and yields citraainic acid, Nt 11.32, 0.2262 gave 22 C.C. moist nitrogen at 18O and 767 mm. Cl,Hl,ON, requires N = 11.29 per cent, Action of Bennxamidine on Ethyl Phenyi'propiotate. As mentioned in the previous paper (Ruhsmann and Cunnington, 2oc. cit.), benzamidine hydrochloride and ethyl phenylpropiolate react in the presence of sodium ethoxide to form benzalphenylgly oxalidone. We now find that for the preparation of this compound 1 mol. of the ethoxide is sufficient, but that the reaction takes a different course, if, instead of allowing the mixture to stand at the ordinary tempera- ture, it is heated on the water-bath for 2-3 hours. On adding water and a little acetic acid to the contents of the flask, a yellow solid is precipitated which dissolves in boiling glacial acetic acid, and, on cooling, crystallises out in yellowish needles melting at 284", whilst the melting point of benzalphenylglyoxalidone is 274". The percentage composition of this substance, as ascertained by the following analysis, is, however, identical with that of the compound of lower melting point. 0.2010 gave 0.5695 GO, and 0.09 10 H,O. 0.2871 ,, 28 C.C. moist nitrogen at 2 3 O and 767 mm. N= 11.08 C,,H,,ON, requires C = 77.41 ; H = 4-83 ; N = 11.29 per cent. C = 77-15 ; H = 5-02. The two compounds differ, not only in their melting points, but also in their other properties. Whilst benzalphenylglyoxalidone, when treated with concentrated hydrochloric acid, at first dissolves although the solution immediately sets to a solid, the second substance, which is also soluble in the concentrated acid, remains in solution. This fact, as well as the readiness with which the latter dissolves in caustic potash, agrees with the statements of the behaviour of diphnyl- pyrimidone. The compound differs from the pyrimidone as it has a yellow colour and evolves benzaldehyde when boiled with caustic potash, h closer examination has shown, however, that these properties are due to benzalphenylglyoxalidone contained in the above product, as the solution of the latter in concentrated hydro- chloric acid, on standing for a day, gives a separation which, after crystallisation from glacial acetic acid, melts at 274' and has all the properties of this compound. The presence of diphenylpyrimidone in the substance melting at 284" is established by its behaviour to- wards bromine. Action of Bennxamidine on Ethyl Phenyi'propiotate. 00 adding the halogen to a solution of the product in concentrated hydrochloric acid, a white precipitate is at once formed which crystallises from glacial acetic acid in colourless needlecl melting at 297-298", and has the composition C16Hl10N2Br, THE FORMATION OF HETEROCYCLIC COMPOUNDS, 245 View Article Online 245 Online 0.2498 gave 0.1420 AgBr. Br = 24.20. 0.2310 ,, 16.8 C.C. moist nitrogen at 16O and 765 mm, N = 8-53, C,,H,,ON,Br requires Br = 24-46 ; N = 8.56 per cent. 0.2498 gave 0.1420 AgBr. Br = 24.20. 0.2310 ,, 16.8 C.C. moist nitrogen at 16O and 765 mm, N = 8-53, C,,H,,ON,Br requires Br = 24-46 ; N = 8.56 per cent. Diphenylpgrimidone, which we have prepared according to Pinner's directions (see '' Die Imidozther und ihre Derivate," 1892, p. 254 ; we find the melting point to be 287-288", compared with 284" as found by Pinner), yields, under similar conditions, a bromo-derivative which in every respect is identical with our compound. Its composition, C16Hl10N,Br, has been verified by the following nitrogen determina- tion : Downloaded by OHIO STATE UNIVERSITY on 19 March 2013 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 0.2081 gave 23.5 C.C. moist nitrogen at 15' and 764 mm. N = 13-30 0.2740 ), 31.5 )) 9 9 1'7' ,, 764 mm. N=1340. C,,H,ON,S requires N = 13 72. RUHEMANN AHD STAPLETON : RUHEMANN AHD STAPLETON : View Article Online Action of Urea, Z'hiouma, and Quanidine on Ethyl Phenylpropiobttt. Benxalhydantoin. We have already mentioned in the introduction to this paper that the compound which is formed by the action of sodium ethoxide on mixture of urea and ethyl phenylpropiolate has not the structure C,H,CNCO NH* CO NH,, as has been stated previously (Zoc. cit.), bub Downloaded by OHIO STATE UNIVERSITY on 19 March 2013 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 0.2303 gave 16.6 C.C. moist nitrogen at 16O and 765 mm. N = 8-40. After having thus proved that benzamidine reacts with ethyl phenylpropiolate at a higher temperature to form a mixture of benzal- pheny lglyoxalidone and diphenylpyrimidone, we were anxious to effect their complete separation. Although the solution of the mixture in concentrated hydrochloric acid deposits the glyoxalidone derivative on standing, yet a certain amount remains in the filtrate, and, on the addi- tion of water, is precipitated along with the diphenylpyrimidone. Even on boiling the mixture with caustic potash for some hours, only partial decomposition of the glyoxalidone takes place. The product obtained on acidifying the alkaline solution and recrystallising the precipitate from glacial acetic acid still has a yellow colour, and even after prolonged boiling with alkali the odour of benzaldehyde is still perceptible. Diphenylpyrimidone can, however, be isolated from the mixture in a pure state by making use of the fact that it is very stable, whilst benzalphenylglyoxalidone, although only partially decomposed by alkalis and acids, is readily: de.st,royed by oxidising agents. Fuming nitric acid is especially convenient for this purpose, for, as we have ascertained, it does not give rise to nitro- compounds when added to the solution of diphenylpyrimidone in glacial acetic acid, whilst the glyoxalidone is thus transformed into readily soluble substances. I n order to isolate the pyrimidone from the mixed product which is formed by digesting a mixture of benz- amidine and ethyl phenylpropiolate on the water-bath, it is sufficient to treat its solution in glacial acetic acid with fuming nitric acid, and then to add water. By this means, a solid is precipitated which crystallises from glacial acetic acid in colourless prisms, and has all the properties characteristic of diphenylpyrimidone. Its composi- tion has been verified by the following nitrogen determination : 246 When alcoholic solutions of sodium ethoxide, guaaidine thiocyanate and ethyl phenylpropiolate (I mol. of each) are mixed a yellowish- red coloration is produced, and, on slightly warming on the water- bath, the contents of the flask set to a semi-solid mass. The product is almost insoluble in water or alcohol, but dissolves in boiling glacial acetic acid, from which, on carefully adding water, it separates in yellowish, microscopic needles. These darken at about 285O and melt at 295' to a black liquid. On analysis : 0.2030 gave 0.4785 CO, and 0.0900 H,O. 0.2570 ,, 47.5 C.C. moist nitrogen at 10' and 768 mm. N = 22-36, C,,H90N, requires C = 64.17 ; H = 4.78 ; N = 22.46 per cent. C = 64.18 ; H = 4.92. 0.2030 gave 0.4785 CO, and 0.0900 H,O. 0.2570 ,, 47.5 C.C. moist nitrogen at 10' and 768 mm. N = 22-36 C,,H90N, requires C = 64.17 ; H = 4.78 ; N = 22.46 per cent. C = 64.18 ; H = 4.92. This compound possesses both acidic and basic properties. It freely dissolves in hot concentrated hydrochloric acid, and the solution, on cooling, deposits yellowish needles of the hydrochloride ; this, how- ever, is unstable, and on drying loses hydrogen chloride. Its constitution is proved, as in the former cases, by the decom- position which the substance suffers under the influence of caustic potash; when boiled with it, the odour of benzsldehgde is developed, an evolution of ammonia takes place, and an acid is formed which we have proved to be a-hydroxycinnamic acid. Downloaded by OHIO STATE UNIVERSITY on 19 March 2013 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 Downloaded by OHIO STATE UNIVERSITY on 19 March 2 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/ Downloaded by OHIO STATE UNIVERSITY on 19 March 2013 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 This formula, which represents the compound as benzalhydantoin, follows from its behaviour towards alkali. Although it is freely soluble in dilute caustic potash and is precipitated unchanged by hydrochloric acid, yet a concentrated solution of the alkali only gradually dissolves it on boiling, and at the same time, decomposition takes place. At first, the odour of benzaldehgde is perceptible, and, later, an evolution of ammonia sets in, which ceases after the heating bas been continued for three hours. The alkaline liquor, on being acidified with hydrochloric acid, yields a solid which readily dissolves in ether and drystallises from chloroform in colourless plates. This compound is characterised as a-hydroxycinnnmic acid (phenylpyruvio acid) by its melting point, 154-155', the green coloration it gives with ferric chloride, and by the following analysis : 0.2206 gave 0,5339 CO, and 0.0969 H,O. C = 66-00 ; H = 4.88, C,H,O, requires C = 65.85 j H = 4.87 per cent. CO-YH Benxalthiohydatntoin, C,H, CH:C<NH-cs . This compound is obtained in a similar manner to benzalhydantoin by mixing alcoholic solutions of sodium ethoxide and thiourea with ethyl phenylpropiolate. After standing for several hours, the alcohol is removed by evaporation, and then water and a little dilute sulphuric acid are added. A yellowish solid is precipitated, which is insoluble in water, and only sparingly soluble in alcohol or boiling glacial acetic acid. From the solution in the latter solvent, it gradually separates, on adding water, as a yellowish powder, which is washed with alcohd and dried a t 100'. On slowly heating, it decomposes about 2SOo, but, when the temperature is rapidly raised, decomposition takes place at 300O. On analysis : 0.2081 gave 23.5 C.C. moist nitrogen at 15' and 764 mm. N = 13-30 0.2740 ), 31.5 )) 9 9 1'7' ,, 764 mm. N=1340. C,,H,ON,S requires N = 13 72. 0.2081 gave 23.5 C.C. moist nitrogen at 15' and 764 mm. N = 13-30 0.2740 ), 31.5 )) 9 9 1'7' ,, 764 mm. N=1340. C,,H,ON,S requires N = 13 72. THE FORMATION OF HETEROCYCLIC COMPOUNDS. 24 View Article Onli 247 Online THE FORMATION OF HETEROCYCLIC COMPOUNDS. View Article O On boiling this compound with concentrated caustic potash for three hours in a flask fitted with a reflux condenser, it suffers a decomposition analogous to that of benzalhydantoin and yields a-hydroxycinnamic acid, This is isolated by acidifying the alkaline liquor with hydrochloric acid, extracting the solution with ether, and removing the solvent by evaporation. The residue, recrystallised from chloroform, shows all the properties characteristic of a-hydroxy- cinnamic acid. Downloaded by OHIO STATE UNIVERSITY on 19 March 2013 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 To the description of the properties of the isoxazolone contained the previous paper, we add the following cryoscopic determination the molecular weight, glacial acetic acid being used as the solvent. in of Weight of Weight of substance. solvent. Depression. Mol. wt. 0.625 21.62 0.73" 161 C,H70,N requires mol. wt. = 169 Downloaded by OHIO STATE UNIVERSITY on 19 March 20 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/ Further experiments, which we intend to perform, are necessary in order to ascertain whether in the methyl derivative, C9H6O2NCH3 the alkyl group is united with the nitrogen or with the ketonic oxygen of the isoxazolone. Aetion of Hydroxylamine on Ethyl Phenylpropiolate 5-Phinyl- 3-isoxazolone. The results of our experiments, which are described above, induced us to re-examine the product of the action of hydroxylamine on ethyl phenylpropiolate. A closer study of tbis substance has shown that it is not a hydroxamic acid, but a cyclic compound of the formula This constitution follows from the remarkable stability of the This constitution follows from the remarkable stability of the 248 RUHEMANN ANb STAPLETON : View Article Online substance towards acids and alkalis. Even on boiling it with these reagents for some hours, hardly any decomposition takes place, although a subgtance of the formula C,H,C i CC(NOH)OH, might be expected to readily break up into phenylpropiolic acid and hydroxylamine. substance towards acids and alkalis. Even on boiling it with these reagents for some hours, hardly any decomposition takes place, although a subgtance of the formula C,H,C i CC(NOH)OH, might be expected to readily break up into phenylpropiolic acid and hydroxylamine. ded by OHIO STATE UNIVERSITY on 19 March 2013 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 Downloaded by OHIO STATE UNIVERSITY on 19 March 2013 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 Downloaded by OHIO STATE UNIVERSITY on 19 March 2013 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 Downloaded by OHIO STATE UNIVERSITY on 19 March 2013 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 Downloaded by OHIO STATE UNIVERSITY on 19 March 20 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/C previously prepared by Hinsberg (Annalcn, 1896, 292, 249). composition of the substance was verified by the following analyses : The previously prepared by Hinsberg (Annalcn, 1896, 292, 249). composition of the substance was verified by the following analyses : The 0.2184 gave 0.5395 CO, and 0.0985 H,O. 0,2303 ,, 34.4 C.C. moist nitrogen at 15' and 759 mm. N = 17.47. C = 67.37 ; H = 5.01. C,H,ON, requires C = 67.50 ; H = 5.00 ; N = 17.50 per cent. 0.2184 gave 0.5395 CO, and 0.0985 H,O. 0,2303 ,, 34.4 C.C. moist nitrogen at 15' and 759 mm. N = 17.47. C = 67.37 ; H = 5.01. C,H,ON, requires C = 67.50 ; H = 5.00 ; N = 17.50 per cent. Xwrnation qf Ethyl 2 : 6-Dihydroxypyridine-3 : 4-dicarboxyZc&~ Action 03' o- Phenylenediamine on Ethyl Acetylenedicarboxylate. We have pointed out already that ethyl acetylenedicarboxylate and ethyl phenglpropiolate seem to show differences in their behaviour with organic bases similar to those with esters of P-ketonic acids. Of special interest in this respect is the fact that ethyl phenylpro- piolate does not condense with o-phenylenediamine whilst ethyl acetylenedicarboxylate, on treatment with the base, yields a com- pound which i8 to be regarded as ethyl puinoxalidoneacetate. The formation of this substance takes place on mixing an alcoholic solution of sodium ethoxide (2 mols.) with o-phenylenediamine hydro chloride (1 mol.) and ethyl acetylenedicarboxylate (1 mol.), and heat- ing the mixture on the water-bath for 2-3 hours, After removing the alcohol by evaporation, water and a little hydrochloric acid are added, and the resulting yellow solid is washed with dilute spirit. It dissolves in boiling alcohol with great difficulty, and, on cooling, crystallises in yellow needles, which soften at 205' and melt at 210' to a red liquid. q On analysis, the following numbers were obtained : 0,2074 gave 0.4714 CO, and 0.0968 H20. C = 61.98 ; H = 5.18. 0.2713 ,, 28.5 c.c.moist nitrogen at 17Oand 763 mm. N = 1223, C,,H,,0,N2 requires C-62.04; Ht5.17; N-12.06 per cent. The ethyl ester has both basic and acidic properties, It dissolves THE FORMATION OF HETEROCYCLIC COMPOUNDS. 249 View Article Online 249 Online in concentrated hydrochloric acid, but the hydrochloride thus formed ig decomposed by water. Whilst the solution of ethyl quinoxalidone- acetate in hydrochloric acid is yellow, its solution in dilute caustic potash is colourless ; the compound, therefore, has properties charac- teristic of the hydroxyquinoxalines. On boiling the ester with con- centrated caustic potash for 3 hours and then acidifying the alkaline liquor, carbon dioxide is evolved and a solid is precipitated. This crystallises from boiling water in almost colourless needles melting a t 245O, and was found to be identical with hydroxymethylquinoxaline, Downloaded by OHIO STATE UNIVERSITY on 19 March 2013 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 Downloaded by OHIO STATE UNIVERSITY on 19 March 2013 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 We find that the ethyl ester of the remaining dicarboxylic acid, C(oH)>N, CH C02C2H5* CeC(C02C,H,)* C(0H) is readily prepared by placing concentrated aqueous ammonia with ethyl propenetetracarboxglate in a bottle, and agitating the mixture from time to time. I n the course of 4-5 days, the oil disappears and the vessel contains a reddish solution, together with a yellowish solid. The latter we have not examined, since its amount is very small. The filtrate is concentrated by heating under diminished pressure on the water-bath, and then acidified with hydrochloric acid. This precipitates a solid, which dissolves readily in hot alcohol, but is only sparingly soluble in boiling water, and separates from the hot aqueous solution in colourless prisms, melting at 161-162'. Ferric chloride gives a reddish-viole t coloration with aqueous or alcoholic solutions of the substance. On analysis : 0.2100 gave 0-3975 CO, and 0,0993 H20. 0.2340 ,, 11.5 C.C. moist nitrogen at 14" and 745 mm. N = 5.66. C,,H1,O,N requires C = 51.76 ; H = 5.10 ; N = 5.49 per cent. C=5162 ; H=525. 0.2100 gave 0-3975 CO, and 0,0993 H20. 0.2340 ,, 11.5 C.C. moist nitrogen at 14" and 745 mm. N = 5.66. C,,H1,O,N requires C = 51.76 ; H = 5.10 ; N = 5.49 per cent. C=5162 ; H=525. Ethyl 2 : 6-dihydroxypyridine-3 : 4-dicarboxylate, besides acidic pro- perties, has also those of a base, since it dissolves in Concentrated hydrochloric acid with the greatest ease. On boiling this solution for 3 hours, hydrolysis takes place, accompanied by the removal of 1 mol. of carbon dioxide, and there is formed a monocarboxylic acid, which is precipitated by adding water to the acid solution. This acid has been identified as citrazinic acid by direct comparison of its pro- perties and by the following analyses : 0.1465 gave 0.2485 CO, and 0.0405 H,O. 0.2480 ,, 18.5 C.C. moist nitrogen at 9" and 769 mm. N= 9.08. C,H50,N requires C = 46.45 ; H = 3.22 ; N = 9.03 per cent. C = 46.26 ; H = 3.03. 0.1465 gave 0.2485 CO, and 0.0405 H,O. 0.2480 ,, 18.5 C.C. moist nitrogen at 9" and 769 mm. N= 9.08. C,H50,N requires C = 46.45 ; H = 3.22 ; N = 9.03 per cent. Xwrnation qf Ethyl 2 : 6-Dihydroxypyridine-3 : 4-dicarboxyZc&~ The considerations brought forward in the introduction to this paper point to the formula CO,C,H,* CH,* C(CO,C,H,):C( CO,C,H,), for ethyl propenetetracarboxylate, the product which Michael (Zoc. cit. ; see also Ruhemann and Cunnington, Trans., 1899,75, 954) obtained by the addition of ethyl malonate to ethyl acetylenedicarboxylate. We have subjected this compound to the action of ammonia with the view of preparing ethyl 2 : 6-dihydroxypyridine-3 : 4-dicarboxylate, and thus completing the series of mono- and di-carboxylic acids of 2 : 6-dihydroxypyridine. From 2 : 6-dihydroxypyridine, two mono- and two di-carboxylic acids can be derived. The monocarboxylic acid containing the CO,H group in the 4-position, namely, is citrazinic acid, which was discovered by Behrmann and Hofmann (Ber., 1884, 17, 2687), and subsequently obtained by one of us (Ruhe- maun, Ber., 1887, 20, 3367) in the course of his researches on the action of ammonia on the ethyl esters of unsaturated acids. The ethyl ester of the second or 3-monocarboxylic acid has lately been isolated by Errera (Ber., 1898, 31, 1245). is citrazinic acid, which was discovered by Behrmann and Hofmann (Ber., 1884, 17, 2687), and subsequently obtained by one of us (Ruhe- maun, Ber., 1887, 20, 3367) in the course of his researches on the action of ammonia on the ethyl esters of unsaturated acids. The ethyl ester of the second or 3-monocarboxylic acid has lately been isolated by Errera (Ber., 1898, 31, 1245). Of the dicarboxylic acids, only ethyl dihydroxynicotinate, VOL. LXXVII. T THE FORMATION OF HETEROCYCLIC COMPOUNDS. View Arti 250 has, up to the present, been recorded (see Guthzeit, Ber., 1893, 26, 2795; Ruhemann and Sedgwick, Ber., 1895, 28, 825; Errera, Gccxeetta, 1897, 27, ii, 393 ; Ruhemann and Browning, Trans., 1898, '73, 280). On hydrolysis with hydrochloric acid, it loses two mols. of carbon dioxide, and yields 2 : 6-dihydroxypyridine (Ruhemann, Trans, , 1898, '73, 350; see also Errera, Ber., 1898, 31, 1246). Downloaded by OHIO STATE UNIVERSITY on 19 March 2013 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 C = 46.26 ; H = 3.03. The decomposition which ethyl 2 : 6-dihydroxypyridine-3 : 4-dicarb- oxylate suffers on heating with hydrochloric acid is in accordance with the behaviour of the corresponding ester of the 3 : 5-substituted di- carboxylic acid which, as mentioned before, under the influence of boiling hydrochloric acid, loses 2 mols. of carbon dioxide. These facts show that 2 : 6-dihydroxypyridinecarboxylic acids with C0,H groups PART I. 251 View Article Online FORSTER: STUDIES IN THE CAMPHANE SERIES. in the 3-position readiIy lose them with elimination of carbon dioxide. In conclusion it may be stated that further experiments on the in the 3-position readiIy lose them with elimination of carbon dioxi in the 3-position readiIy lose them with elimination of carbon dioxide. In conclusion, it may be stated that further experiments on the lines indicated in this paper are in progress. Downloaded by OHIO STATE UNIVERSITY on 19 March 2013 Published on 01 January 1900 on http://pubs.rsc.org \| doi:10.1039/CT9007700239 GONVILLE AND CAWS COLLEGE, CAMBRIDQE.
https://openalex.org/W4324132771	http://capaciousjournal.com/cms/wp-content/uploads/2023/06/katz-a-subway-story.pdf	English	null	A Subway Story: Boundaries, Transmission, and Covid-19	null	2,023	cc-by	4,762	INTERSTICE INTERSTICE 2023  Capacious: Journal for Emerging Affect Inquiry Vol. 3, No. 1 A SUBWAY STORY: BOUNDARIES, TRANSMISSION, AND COVID-19 WESLEYAN UNIVERSITY Margalit Katz urban, affect, anthropology, subway, Covid-19 capaciousjournal.com  DOI: Creative Commons Attribution 4.0 (CC BY 4.0) https://doi.org/10.22387/CAP2022.71 capaciousjournal.com  DOI: Creative Commons Attribution 4.0 (CC BY 4.0) https://doi.org/10.22387/CAP2022.71 capaciousjournal.com  DOI: Creative Commons Attribution 4.0 (CC BY 4.0) https://doi.org/10.22387/CAP2022.71 ABSTRACT In a brief autoethnographic vignette, I explore manifestations of the transmission of affect in the New York City subway system amid the rise of the Delta variant of Covid-19 in the summer of 2021. I relate the transmissibility of affect to the contagion of Covid-19, particularly as both are approached in the subway, a place marked by proximity among strangers. Finally, I investigate the social construc- tion of disgust and its involvement in organizing social proximity through the movement of affect in the subway. KEYWORDS urban, affect, anthropology, subway, Covid-19 capaciousjournal.com  DOI: Creative Commons Attribution 4.0 (CC BY 4.0) APACIOUS ban, affect, anthropology, subway, Covid-19 EYWORDS https://doi.org/10.22387/CAP2022.71 KEYWORDS July 6th, 2021, 2:00 pm On my way into Manhattan on the C train, a young Black man gets on at Utica Avenue. He wears new yellow Jordan’s, jeans, a belt, a heather grey t-shirt, sunglasses, and a fresh single-use blue surgical mask. As he gets on the train, I hear him repeatedly mutter, “Ohhh man, oh shit, bro! Don’t do it!” He snickers to himself in a low, gravelly tone. He sits down on the bench diagonal from me. As the train barrels through the darkness, he continues to speak, and directs his gaze toward the window across from him, almost as if to talk to someone on the other side of the glass. Suddenly, he takes his t-shirt off. He flexes his bicep muscles and lets out a deep growl. He pulls his shirt back over his head, but only partly, leaving one sleeve flapping on top of his bare right shoulder. His back and chiseled abdomen remain exposed. The young white woman closest to him looks up at him out of the corner of her eye. I look down for a moment to take notes when suddenly I hear a loud smack- ing sound, as if someone just slapped another person. I look up, and once again, smacking sounds, only this time I see the source—the half-shirtless man jolts and pounds his fist into his own abdomen. The sound of skin slapping against skin ricochets down the car. A white man in a suit looks up from his phone and over at the half-shirtless man. At the next stop a white man in dirty, tattered clothing gets on and launches into his panhandling spiel. His CAPACIOUS 174 A Subway Story Figure 1. MTA Stop the Spread Campaign (Margalit Katz, 2021). Figure 1. MTA Stop the Spread Campaign (Margalit Katz, 2021). CAPACIOUS CAPACIOUS 175 Margalit Katz speech remains stable as he tells his story, but I am wary of the potential for his tone to flip, to curse us out for not helping him. Although in many cases, people are moved to give by pity or guilt or some other affect, on this ride everyone inspects their shoes. No one donates a cent, so we avoid the pang of envy of other passengers more generous than ourselves. The panhandler passes the half-shirtless man, and I hold my breath. However, to my surprise, the half-shirtless man does not engage. The presumably homeless panhan- dler moves on to the next car. The half-shirtless guy continues to mutter and sporadically punch himself in various parts of his body. A middle-aged white man who sits close to the end of the car lets out a cough that seems to dispel the tension. The half-shirtless man hangs his head and remains silent for several stops. Then, for no apparent reason, he jerks his head up and bangs his palm on the metal pole, hard. The pole vibrates and rings for what feels like minutes. No one else so much as stirs. At Fulton Street in Manhattan, he stands up and bolts off the train, his sleeve still hanging off his naked back. In these sounds, sights, smells, movements, and stillnesses, serendipitous feelings of tension, connection, and inexplicability arise, subside, and re-emerge in the subway car. The experience of physical proximity in the subway makes it a potent environment for observing the movement of affect and its social meaning.1 As the primary mode of transportation in New York City, the subway mobilizes New Yorkers from every corner of the city. It is one of the only places that individuals from disparate backgrounds regularly come into close contact with each other. As diverse as New York is, boundaries are pre-constructed by neighborhood, socio- economic class, ethnicity, gender, and race, and are only reified by the perceived threat of proximity to the ‘other.’ Despite socio-cultural and physical boundaries in places of close proximity, like the subway, the borders between individuals become blurred. The social interpenetrability that results is an inherently affec- tive phenomenon. In order to mitigate anxieties surrounding this transmission of affect in enclosed space, passengers have employed social distancing since long before Covid-19 (Brennan 2015). July 6th, 2021, 2:04 pm The coughing man continues to cough. He’s wearing a striped t-shirt and no mask. He looks like a disheveled version of Jerry Seinfeld. A young Brown guy on the other end of the car wearing an army green t-shirt and a camou- flage-print backpack stands up with purpose and tears down the car. He takes a seat directly across from me, which puts me on edge. He appears erratic and flicks his MetroCard over and over in his hand rapidly before he sprints to the other end of the train, his shoes thundering down the linoleum aisle. As the doors open, he cuts in front of the white man in the suit, who grimaces. They both get off. The coughing man wheezes again. It’s getting ridiculous, and I want to ask if anyone has an extra mask for him, but I don’t. A rail-thin Black woman wearing a wifebeater, sweatpants, and flip flops barges through the gangway door, then turns back and yells something I can’t make out to someone in the other car. She rushes to the end of the car and sits in the corner. She closes her eyes. Suddenly she opens her eyes and yells something I can’t catch, then jumps off at the next stop. When the doors open at West 4th Street, the cough- ing man coughs up gelatinous globs of green phlegm and spits them onto the platform three times. He notices me looking at him and says something I can’t hear. He hangs his head and appears distressed. The erratic guy from long before jumps back on and sits down right next to the coughing dude. I guess he never actually got off the train, he just switched cars. After about one minute he springs up again and jets off. I feel antsy, like there’s tension building, like a bubble that eventually needs to pop. Affect or “intensity” as Brian Massumi (2002) defines it, “is…a nonconscious… autonomic remainder” (Massumi 2002, 25). In other words, for Massumi (2002), affect is an other-than-conscious free-floating “potential” that escapes confine- ment in an individual being in the form of emotion (35). However, as Sara Ahmed (2004) notes, these intensities exist neither inside nor outside an individual—rath- er, “they ‘affect’ the very distinction of inside and outside in the first place” (28). July 6th, 2021, 2:04 pm During my train ride, I became aware of the unclaimed affect within the car, prior to the moment it was converted into emotion. Emotion, according to Massumi CAPACIOUS To avoid viral and affective contamination, passengers tend to disperse within subway cars, maximizing the distance between themselves and strangers. The pursuit of personal space surpasses a mere desire for physical comfort and a range of movement. In the subway, “personal space may be defined as that area surrounding, or belonging to, a person, which is cathected consciously or unconsciously to the self” (Fried and DeFazio 1974, 49). In other words, we imagine our affective selves as projected beyond our skin. To reassert our boundaries, despite our uncontained bodies, we distance ourselves from others. Other ways to distance are to avoid meeting another’s gaze via what Erving Goff- man (1980) calls civil inattention, and to adopt what Georg Simmel (1950) refers to as a “blasé attitude,” or nonchalance (Boy 2021). These performances reinforce the boundary between the individual and any affective “impressions” or potential 176 A Subway Story A Subway Story that they encounter (Brennan 2015). During rush hours, when physical contact is impossible to avoid, tension and discomfort can build as passengers struggle to maintain their physical and affective integrity. During the pandemic, the entanglement of physical and non-physical forms of interpersonal permeability could be observed in a heightened way. CAPACIOUS 177 Margalit Katz (2002), is affect contextualized, “owned and recognized” (28). The “remainder” of affect, however, moves between individual bodies and evolves, prior to being owned or “captured” (Massumi 2002). These remnants of intensity develop as “unactualized, inseparable from but unassimilable to any particular, functional- ly anchored perspective” until they are captured (Massumi 2002, 35). Ahmed, however, would argue that, despite this process of transformation and conversion, affect can never, even momentarily, exist completely unclaimed or captured. She flags this partial capture as an example of the “stickiness” of affect and the ways in which it permeates the boundaries of individuals and connects them to each other through shared affect (Ahmed 2004). The uncaptured portion of intensity, despite constant movement and development, is often undetectable until its moment of capture (Massumi 2002). Thus, people often describe the capture of intensity as sudden, as I do in the scene above when I use language that conveys an abrupt turn or flareup, such as my description of a passenger’s apparent “eruption.” At the same time, since I observed the same setting for an extended period, I was able to discern a growing tension, or the transformation of the uncaptured “remainder” as it underwent various evolutions and micro-captures that justified the outburst as a mere continuity in the arc of the affect’s trajectory. The affect that I felt accumulate and explode can also be understood not just as movement and capture, but also as transmission between individual bodies that have captured affect. The free-floating, unbounded remainder that Massumi (2002) describes is not captured intentionally—it is transmitted. Affect is social and can arise from within and outside of individuals, as “the emotions or affects of one person, and the enhancing or depressing energies these affects entail, can enter into another” (Brennan 2015, 3). Transmission can manifest emotionally, physically, and biologically. As the transmission of affect connects humans to each other, it highlights the boundaries, or lack thereof, through which affect permeates (Ahmed 2004). Much as our physical borders are undetectable until the moment of contact with another body, our boundaries are reified by our exchange of affect (Ahmed 2004). In response to this contagion of affect and the interpersonal connection it causes, we develop methods of self-containment to defend our boundaries via physical and emotional means. Again, it is only in relation to a defined ‘other’ that we can identify our own borders. CAPACIOUS Thus, processes of self-containment require a foil, or “other” who can transmit and/or capture affect (Brennan 2015). However, Bren- nan (2015) argues that Western society does not legitimize the transmission of affect as real—the idea that emotions and bodies are permeable by invisible forces 178 A Subway Story A Subway Story A Subway Story counters the entire individualized belief system surrounding affect. As a result, we urgently reproduce the constructed illusion of the bounded body as well as norm-based boundaries—such as social distancing—in an attempt to prevent and dismiss the possibility of transmission. (Brennan 2015). July 6th, 2021, 2:11 pm Another splitting cough cuts through the silent car. Finally, someone cracks. A person I read as Latinx and gender non-conforming sitting across from the coughing man erupts, “YOU WANNA FUCKING COUGH UP A STORM BUT NOT WEAR A MASK?!? IT’S FUCKING DISGUSTING!” They get up and dart toward the opposite end of the car. Their booming voice startles me, and I’m impressed by their ability to project. As they pass me, I nod my head at them, but they ignore me and sit almost as far from the coughing man as possible. He replies with a weak, “oh fUck yOU.” No one else even looks up. The person continues their tirade: “COPS GIVING OUT TICKETS BUT NOTHING? ANOTHER STRAIN IS GOING AROUND!” I shoot them a look to convey my gratitude, but they seem too riled up to notice. The guy across from me listens to music through his headphones and drums a beat. The coughing guy looks at his combatant and mutters, “Fuck you, fa—.” I can’t tell if he says ‘fat- something’ or ‘faggot.’ He rises to get off at 59th and once again yells, “have a nice day fa—” but again I can’t hear what he says because he’s so soft-spoken. His rival bellows, “YO GET YOURSELF CHECKED OUT! YOU HAVE THAT COUGH BUT IT MIGHT HAVE GONE TO YOUR BRAIN! HOW ARE YOU GONNA YELL AT ME BECAUSE YOU DONT HAVE A MASK ON?!?” The closing doors jingle plays as the coughing man walks up the steps on the platform. The person who spoke up shakes their head, and we exchange a glance. I shake my head and smirk. The bubble has popped. CAPACIOUS In the summer of 2021, due the ubiquity of information on the transmission of Covid-19, the public experienced a heightened awareness of the permeability of the body via the breath.2 The transmission of affect happened to follow the same trajectory as the potential transmission of Covid-19. The presence of Covid-19 highlights our interconnectedness, as well as our separation, from other humans and threatens to violate our already partial boundaries. Perhaps the coughing man’s intrusion and refusal to maintain physical boundaries reminded his angry fellow-rider of the impossibility of sealed borders. Covid-19 exposes the illusion of enclosed physical bodies, sometimes with fatal consequences, only amplifying the existing anxiety surrounding proximity in the subway. Margalit Katz Margalit Katz The implicit impetus for the conflict was the other passenger’s fear of catching Covid-19 from the coughing man. However, the focus placed on this concern conceals the looming contagion of affect. Would this interaction have happened if not for the intrusions of the shirtless man, the screaming woman, the running man, and the panhandler just a few minutes prior? To what extent had their presence agitated the affect in the car through their transgressions, building the tension, the intensity that I also experienced, such that it needed to be transformed and stabilized via capture and transmission? I wondered if the anger the passenger conveyed was an expression of discomfort with the transgressions and their sub- sequent transmission of affect. Similarly, I wondered to what extent their anger stemmed from the larger context of death, loss, and precarity outside the subway, their own accumulated ‘potential’ they brought into the subway to transmit. To what extent was Covid-19 and the (absence of the) mask, as the available social symbol of the virus, a catalyst for the passenger to make sense of and express their discomfort, their pain? There is no way to know for sure. I only know the tension and release I felt as affects accumulated and dispersed. Although the social distancing practiced during Covid-19 is distinct from that solely brought on by the transmission of affect in situations of physical proximity, it is still surprisingly affective. If we did not develop intense affective attachments and reservations surrounding Covid-19, we might not so vehemently abide by or disobey the medical knowledge that informs our behavior and the science-based precautions deemed necessary to avoid transmission. Consequently, the trans- mission of affect about Covid-19 brought about by news coverage and political polarization is one of the very forces determining whether people follow Cov- id-19 precautions and, thereby, also impacts the transmission of Covid-19 in the subway. Affectively charged opinions and the lengths that people are willing to go to abide by new norms compel us to ‘other’ those with opposing views. We must separate ourselves from those ‘others’ whose breath threatens to permeate us. In Covid-times, these ‘others’ either penetrate us through an infringement on our personal liberties (such as to go without a mask) or through a potential or imagined transmission of Covid-19. CAPACIOUS 179 Margalit Katz Therefore, we can view the structure of feeling of Covid-affect not as a novel form of sociality in the subway but as a routine—if exacerbated—manipulation of affect to negotiate boundary formation. The involvement of disgust, an embodied sensation, in the conflicts surrounding Covid-19 further illustrates the relation between the transmission of affect and the transmission of Covid-19. During my affectively charged ride on the C train, I felt the visceral sensation of disgust as the man spat mucus onto the platform. Everything indicated that the person who yelled at him also acted out of disgust, 180 A Subway Story A Subway Story along with other emotions (anger, perhaps fear). The production and expression of the feeling of disgust depends upon our immersion within hierarchical social structures. In the West, disgust is often described as corporeal in ways that other emotions are not. It is a sensation frequently felt in the stomach and, in cases of extreme disgust, can produce nausea. The physicality of disgust, however, is often conflated with instinct (Durham 2011). Due to this conflation, Deborah Durham (2011) argues that the visceral nature of disgust is precisely the source of its power. She claims that disgust is naturalized and unchallenged because of its physical nature. However, like Mary Douglas’ (1966) definition of dirt, as “matter out of place,” disgust is relational (Durham 2011). Nothing is inherently, universally, or naturally dirty or disgusting. Georges Bataille (1986) explains, “We imagine that it is the stink of excrement that makes us feel sick. But would it stink if we had not thought it was disgusting in the first place? We do not take long to forget what trouble we go to pass on to our children the aversions that make us…human beings” (38). Here Bataille (1986) upends the claim of visceral essentialism, arguing that disgust, due to its dependency on taboo, humanizes, but in doing so, animalizes the object or ‘other’ and, by extension, organizes social hierarchies. Durham (2011) takes the argument for the social construction of disgust one step further. She adopts William Miller’s (1997) cultural argument in The Anatomy of Disgust that something becomes disgusting only when it threatens to contaminate the self through proximity. Margalit Katz Accordingly, if we go by developments in the West, “The unwashed poor are pathetic, admirable or disapproved, and different while living in distant fields; when they come into the cities in masses with industrial- ization, they become disgusting, as people [the emergent bourgeoisie] draw up new moral and aesthetic boundaries” (Durham 2011, 148). The historical European imagination of certain ethnic groups as malodorous similarly works to construct moral boundaries by evoking disgust and motivating social distance between ethnic groups. Durham (2011) argues that, as this classist, racist, xenophobic mentality suggests, disgust requires imagining the self in proximity to or as the object of disgust. Disgust creates relation between the object of disgust and the “sphere of intimacy of the disgusted bourgeoisie,” which threatens the elite (Durham 2011, 149). In bridging subject and object and defining them as oppositional, disgust creates and reifies identities and borders, much like affect more broadly (Ahmed 2004). This argument speaks to the ways bourgeois values became naturalized as the norm, assisted by the physicality of disgust (Durham 2011). Margalit Katz Margalit Katz This physical proximity among strangers in the subway creates a sense of inti- macy with the ‘other.’ Passengers choose to fill their commute time in a variety of ways: reading, listening to music, closing their eyes and resting. Many people, however, engage in the solitary contemplation of their surroundings. During this time, I become keenly aware of the ephemeral nature of the setting. I know that I might never see the people in my company again. I do not know them or any- thing about them. This empathetic contemplation amid physical proximity elicits curiosity and an artificially augmented sense of intimacy with other passengers. Yet this imagination of proximity, despite efforts to maintain social distance, establishes the prime conditions for disgust to incubate. If someone in tattered clothing, dirty socks, and no shoes stumbles down the car, I may instinctively recoil. I do not feel disgusted because the clothing or the person are inherently disgusting, but because I imagine myself as that person or in physical contact with them and my mind and body have been trained to reject this idea. This visible new form of radical social difference further promotes disgust because disgust involves a relation to an ‘other.’ Our fear of contagion deepens in the company of someone, like the coughing man, who improperly wears or fails to wear a mask, because with this behavior, the likelihood for transmission rises. But with the emergence of Covid-19 our sensation of disgust goes beyond this knowl- edge. The presence or absence of a face mask has been commonly interpreted as indicating political affiliation. Non-compliance with the mask mandate in spaces like the subway pointed to the position assumed by the far right and by politicians toward whom many liberal New Yorkers feel disgust. Conversely, anti-maskers might feel disgust as they are surrounded by mandate-abiding liberals whose presence galvanizes anti-maskers’ revulsion at the idea of wearing a mask them- selves. The coughing man’s disgust toward masked people only escalated when he was verbally attacked by someone in a violation of civil inattention, transfer- ring affect to him. Hence, the disgust directed at the coughing man converted, probably alongside anger and shame, into emotion, which he expressed through his use of profanity. CAPACIOUS 181 Margalit Katz Perhaps a subconscious discomfort with the transmissibility of affect was the reason the coughing man also reaffirmed boundaries by labeling the passenger as a member of a stigmatized identity, an ‘other.’ Yet this socially constructed experience of disgust, in which we imagine or expe- rience proximity to the ‘other,’ does not only urge us to maintain distance. Rather, it incites a conflicting push-pull tension, also inspiring shared spectatorship and fascination among those who ostracize the ‘other’ because of his behavior. In Ugly Feelings, Sianne Ngai (2004) remarks on the pleasure of the experience of disgust, which not only stems from the process of social exclusion disgust entails, but also from the human fascination with transgression. While taboo dictates limits, or so- 182 A Subway Story A Subway Story cial norms, they are “only there to be overreached. Fear and horror are not the real and final reaction; on the contrary, they are a temptation to overstep the bounds” (Bataille 1986, 144). Those who animate disgust by transgressing taboo in public, like the man coughing up mucus, are not merely dismissed as disgusting. They attract attention—we often may feel disgusted, but we cannot peel our gaze away. Ngai underscores the way that disgust not only warns us to assert our boundaries via the fantasy of proximity, which creates a tension between ourselves and an ‘other’—it is itself an act of othering in which we locate ourselves within a group. She reminds us of Miller’s observation, that, because disgust is learned, as Bataille (1986) demonstrates, “the avowal of disgust expects concurrence” (Miller 1997, 194). The sensation of disgust, Ngai (2004) writes, “seeks to include or draw others into its exclusion of its object, enabling a strange kind of sociability” (336). Through shared experience of disgust, social groups communicate shared taste or refinement. Like disgust, the shame that the coughing man may or may not have felt highlights the separation (or lack thereof) between the self and others. Calling on Silvan Tom- kins, Eve Sedgwick and Adam Frank (1995) note how shame marks the individual as other, or strange, to other individuals. They go on to argue that “shame, like disgust and contempt, is activated by drawing a boundary line or barrier” (520). Berlant adds that certain marginalized groups, perhaps like conservatives or an- ti-maskers in New York City, experience shaming and, as a result, adopt a “shamed” subjectivity. Margalit Katz The performance of shamelessness, on the other hand, can be a form of political refusal, or a way to claim freedom (Berlant 2008). One might argue that the coughing man was shameless in his choice not to wear a mask, openly refusing to conform to the social norm of masking. Berlant (2008) argues that when someone performs shamelessness, others often react negatively, “not having skills for maintaining composure amidst the deflation of their fantasy about how their world is organized” (209). Perhaps the other person yelled because they felt attacked by the coughing man’s refusal to maintain the norm of boundary forma- tion through masking. During the height of the Covid-19 pandemic, as the illumination of our phys- ical permeability spread through the growing knowledge of how the airborne virus circulates in crowds, the subway lost millions of customers due to anxiety of contagion. For those who continued to ride the subway, this anxiety solidified pre-existing boundaries and enforced new ones, such as wearing facial masks and Endnotes 1. In this vignette I have intentionally identified the racial presentation of the individuals involved, with conscious consideration of my positionality as a white person and the im- pact of my words, their ability to perpetuate racial bias, and the value of such details in the context of this story. In the subway, where I cannot know someone’s background or how they identify, I have made assumptions based on phenotype, behavior, general appear- ance, etc. At the same time, recognizing that racial and ethnic biases play a large role, both implicitly and explicitly, in social interaction and in the affects that emerge from proximity, I have flagged racial indicators in this story to highlight tensions and connections between individuals that may have been racially motivated, or that require racial markers to better illustrate the affective tension created by boundaries, proximity, and otherness. 2. While a novel realization for many, such awareness is commonplace for individuals with disabilities and chronic illness, and has long been studied in Disability Studies. However, the emergence of Covid-19 marked widespread consciousness across all identity and cultural groups in the city for the first time. CAPACIOUS 183 Margalit Katz social distancing. In this way, the material transmission of Covid-19 mirrors the transmission of affect. We construct social boundaries to contain both. Methods of virus avoidance, such as social distancing, are informed by the pre-existing norma- tive techniques of affect management in the subway. Consequently, conflicts that arise surrounding the enforcement of boundaries invented to mitigate the spread of Covid-19 are also affectively charged. Thus, the tension surrounding Covid-19 has become a means of communicating anxieties about transmissions of affect at large. References Ahmed, S. (2004). Collective Feelings: Or, the Impressions Left by Others. Theo- ry, Culture & Society, 21(2), pp. 25–42. Available at: https://journals.sagepub. com/doi/pdf/10.1177/0263276404042133. Bataille, G. (1986) Erotism: Death & Sensuality. San Francisco: City Lights Books Boy, J. D. (2021). The Metropolis and the Life of Spirit’ by Georg Simmel: A New Translation. Journal of Classical Sociology, 21(2), pp.188–202. Available at: https://journals.sagepub.com/doi/full/10.1177/1468795X20980638 Brennan, T. (2015). The Transmission of Affect. Ithaca and London: Cornell Uni- versity Press. Douglas, M. (1966). Purity and Danger: An Analysis of Concepts of Pollution and Taboo. London: Routledge & Kegan Paul, 1966. 184 A Subway Story Durham, D. (2011). Disgust and the Anthropological Imagination. Ethnos, 76 (2), pp. 131–56. https://www.tandfonline.com/doi/full/10.1080/00141844 .2010.547941 Fried, M. L. and DeFazio, V.J. (1974). Territoriality and Boundary Conflicts in the Subway. Psychiatry 37(1), pp. 47–59. https://www.proquest.com/ docview/1301437116?&imgSeq=1 Goffman, E. (1980). Behavior in Public Places: Notes on the Social Organization of Gatherings. Westport, Conn: Greenwood Press. Massumi, B. (2002). Parables for the Virtual: Movement, Affect, Sensation. Durham, NC: Duke University Press. Miller, W. I. (1997). The Anatomy of Disgust. Cambridge, MA: Harvard Uni- versity Press. Najafi, S & Serlin, D and Berlant, L. (2008). The Broken Circuit: An Interview with Lauren Berlant. Cabinet Magazine. Available at: https://www.cabinet- magazine.org/issues/31/najafi_serlin_berlant.php Ngai, S. Ugly Feelings. Cambridge, MA: Harvard University Press, 2004. Sedgwick, E. K., and Frank, A. (1995). Shame in the Cybernetic Fold: Reading Silvan Tomkins. Critical Inquiry, 21 (2), pp. 496–522.
https://openalex.org/W4295309618	https://www.research-collection.ethz.ch/bitstream/20.500.11850/572570/3/s41598-022-19294-2.pdf	English	null	Effects of sediment flushing operations versus natural floods on Chinook salmon survival	Scientific reports	2,022	cc-by	15,370	ETH Library Effects of sediment flushing operations versus natural floods on Chinook salmon survival Manisha Panthi 1,2, Aaron A. Lee 3, Sudesh Dahal 4, Amgad Omer 5, Mário J. Franca 6,7 & Alessandra Crosato 1,7 Flushing is a common measure to manage and reduce the amount of sediment stored in reservoirs. However, the sudden release of large volumes of sediment abruptly increases the suspended solids concentration and alters the riverbed composition. Similar effects can be produced also by natural flood events. Do flushing operations have more detrimental impacts than natural floods? To answer this question, we investigated the impact of flushing on the survival of the Chinook salmon (Oncorhynchus tshawytscha) in the Sandy River (OR, USA), assuming that sediment is flushed from hypothetical bottom gates of the, now decommissioned, Marmot Dam. The effects of several flushing scenarios are analyzed with a 2D morphodynamic model, together with habitat suitability curves and stress indicators. The results show that attention has to be paid to duration: the shorter the flushing operation, the lesser the stresses on fish survival and spawning habitats. Flushing causes high stress to salmon eggs and larvae, due to unbearable levels of suspended sediment concentrations. It also decreases the areas usable for spawning due to fine-sediment deposition, with up to 95% loss at peak flow. Without the dam, the corresponding natural flood event would produce similar effects, with up to 93% loss. The study shows that well-planned flushing operations could mimic a natural impact, but only partly. In the long-term, larger losses of spawning grounds can be expected, since the removal of fine sediment with the release of clear water from the reservoir is a lengthy process that may be undesirable due to water storage reduction. Reservoirs trap sediment transported by rivers, with coarser material settling first, near the reservoir entrance, and finer sediment reaching further areas closer to the dam1. Progressive deposition of sediment reduces the water storage capacity and impacts reservoir and dam operations2–4. Flushing operations are a way to remove part of the deposited sediment, particularly the finest component that accumulates near the dam5. However, the rapid release of large quantities of water and sediment produces sudden alterations in flow velocity, water depth6 and temperature7, together with an abrupt increase of the concentration of suspended solids2,8–12 in the downstream river. Journal Article This page was generated automatically upon download from the ETH Zurich Research Collection. For more information, please consult the Terms of use. This page was generated automatically upon download from the ETH Zurich Research Collection. For more information, please consult the Terms of use. www.nature.com/scientificreports Effects of sediment flushing operations versus natural floods on Chinook salmon survival These sudden alterations highly impact the river ecology13–16, also because they often produce persisting changes in bed composition and morphology17–19.hifi p g g p p gy The release of fine sediment affects fish and macroinvertebrates in several ways. High concentrations of sus- pended sediment have acute/short-term effects on respiratory organs, which can be lethal20, whereas the attenu- ation of light penetration produces negative effects on fishes and other forms of life, and hence their growth and development21. The release of reservoir water often results in a change in water temperature7 and this can lead to a behavioral drift and alter the distribution of benthos, as well as increase the mortality of invertebrates22. In addition, the deposition of fine sediment on the riverbed may: (i) bury and suffocate benthos, laid eggs, larvae and fry (acute/short-term effect)23,24; (ii) deteriorate spawning grounds (mid-to-long-term effect)25–27; and (iii) clog the hyporheic region of the bed hindering fluxes of oxygen and metabolic waste, as well as groundwater- surface water exchanges (mid-to-long-term effect)28. 1Department of Water Resources and Ecosystem, IHE Delft Institute for Water Education, Westvest 7, 2611 AX Delft, The Netherlands. 2Utah Water Research Laboratory, Department of Civil and Environmental Engineering, Utah State University, 1600 Canyon Rd, Logan, UT 84321, USA. 3Natural Systems Design, 127 E 1st. St., Port Angeles, WA 98362, USA. 4Laboratory of Hydraulics, Hydrology and Glaciology, ETH Zürich, Hönggerbergring 26, 8093 Zurich, Switzerland. 5Rivers‑Reservoirs Dynamics and Morphology, Deltares, Boussineqweg 1, 2629 HV Delft, The Netherlands. 6Karlsruhe Institute of Technology, Engesserstraße 22, Geb. 10.83 ‑ Raum 108, 76131 Karlsruhe, Germany. 7Department of Hydraulic Engineering, Faculty of Civil Engineering and Geosciences, Delft University of Technology, Delft, The Netherlands. email: manisha.panthi@usu.edu \| https://doi.org/10.1038/s41598-022-19294-2 Scientific Reports \| (2022) 12:15354 www.nature.com/scientificreports/ www.nature.com/scientificreports/ Figure 1. Location of Marmot Dam and study area showing the tributaries and gauging stations (Adapted from Major et al.9, Credit: Department of Interior/ USGS) created using Adobe Illustrator 2021: https://www.adobe. com/products/illustrator.html, U/S indicates “upstream” and D/S “downstream”. Figure 1. Location of Marmot Dam and study area showing the tributaries and gauging stations (Adapted from Major et al.9, Credit: Department of Interior/ USGS) created using Adobe Illustrator 2021: https://www.adobe. com/products/illustrator.html, U/S indicates “upstream” and D/S “downstream”. Quantifying the effects of flushing operations on biota requires the assessment of both acute and long-term impacts29. However, most studies focus either on the effects of high suspended solids concentrations10,30,31 or on riverine habitat deterioration32,33. Effects of sediment flushing operations versus natural floods on Chinook salmon survival Empirical relations have been developed to quantify the effects of high levels of suspended sediment concentration on fish20,34,35, whereas habitat suitability curves can be used to quantify the impact of fine sediment on spawning grounds25,29,36,37.lf pi p g g Natural floods might have similar acute and long-term effects, because they also present a sudden increase in flow velocity and sediment transport, particularly in mountain streams38–41 and near the vicinity of urban areas, due to heavy rainfall42,43. However, in general, instream biota have adapted to the specific hydrology and sedi- ment transport regimes of their river system, so natural floods may not threaten the long-term survival of fish and invertebrate species38,40,41,44–47. What is then the difference between natural floods and flushing operations? Can reservoir-flushing operations be designed with reduced acute and long-term effects on fish?fl lfi To answer these questions, this research evaluates and compares the effects of both flushing operations and the corresponding natural flood on the survival and spawning habitats of Chinook salmon (Oncorhynchus tshaw- ytscha) in the Sandy River (OR, USA). The study considers both the severity of stress caused by high sediment concentrations (acute short-term effects) and the decrease of spawning habitat due to fine sediment deposition (medium to long-term effects). A key assumption is that the Marmot Dam, built on the Sandy River in 1913 and decommissioned in 2007, is operational and provided by bottom gates for sediment flushing. The releases of water and sediment volumes from the reservoir are derived by simulating the opening of the hypothetical bottom gates of the dam using a two-dimensional (2D) morphodynamic model, here called the Reservoir model. The Marmot Dam is assumed absent for the quantification of the flow discharges and the sediment transport rates during the natural flood. The changes in hydraulic and sediment characteristics in the downstream river reach are computed with another 2D model, here called the River model. The results are post-processed to estimate the potential impacts of the considered events on Chinook salmon. Case study area. In 1913, Portland General Electric (PGE) constructed the Marmot Dam across the Sandy River, 48 km upstream of the confluence with the Columbia River, as a wooden crib dam for hydropower gen- eration (Fig. 1). The dam was later upgraded to a 15 m high concrete structure with an overflow spillway crest48 without bottom gates, which restricted bedload transport downstream. www.nature.com/scientificreports/ where, as low-gradient sand-bed river, it forms an inland delta. The river is fed by rainfalls and spring melts from snowpack50. Three-fourths of the annual precipitation falls in the period October–March, with the highest amounts occurring in November, December and January51.h where, as low-gradient sand-bed river, it forms an inland delta. The river is fed by rainfalls and spring melts from snowpack50. Three-fourths of the annual precipitation falls in the period October–March, with the highest amounts occurring in November, December and January51.h g J y The study area extends from 3.5 km upstream to 18 km downstream of the (ex) Marmot Dam where the Sandy River meets the Bull Run River. At the Marmot Dam location, the river has a mild slope and presents a wider floodplain but 5 km downstream of the dam the river is confined and steep. Here the river presents rock outcrops and its bed is only partly alluvial. This river reach, known as the Sandy River Gorge, extends for 6.5 km. At the exit of the gorge, the river becomes unconfined with a mild slope, forming meanders for another 9.4 km. These reaches of river have different characteristics altering the flow depth, velocity and bed material composi- tion, hence the use of these areas by different life stages of Chinook salmon differs with time of the year altering the consequences spatially and temporally.tl q p y p y In 1964, after a flood with a discharge peak of 2400 m3/s, the highest since 1910, parts of the Sandy River and its tributaries were artificially straightened, their banks protected by rock berms and large obstructions caused by boulders and woody debris were removed52. After the removal of the Marmot Dam (2007) the river runs freely without any discharge regulation, providing extended habitat for existing biotas.ht y g g g g The abundance of hydrodynamic and morphological data before and after the removal of the Marmot Dam is the reason behind the choice of this case study, since it allowed the construction of two modeling tools simulating the erosion of the reservoir deposit during flushing operations5 and the transport and deposition of the released sediment along the Sandy River32,53. Chinook salmon (Oncorhynchus tshawytscha). Two Pacific salmon species find their spawning habi- tats in the Sandy River52: Chinook salmon, both spring-run and fall-run, and coho salmon. www.nature.com/scientificreports/ Chinook is the larg- est Pacific salmon species, reaching a weight between 6 and 23 kg54. As most of Pacific salmon, the adults return to their natal gravel-bed streams from the ocean to spawn. The alevins emerge from the eggs and live within the gravel until they are large enough (fry) and start their migration towards deeper waters to finally reach the ocean where they grow into adults55. With a water temperature of 11 °C, Chinook salmon eggs hatch in roughly 47 days, whereas the alevins need 84 days to absorb their yolk sacs and become fry56,57.h y y y y The spawning periods of fall and spring Chinook salmon in the Sandy River are September-December and September–October, respectively58. This means that fry appear in December-March, approximately four months after eggs are laid.h t gg The spawning grounds are especially found upstream of the ex-Marmot Dam and downstream of the Dodge Park (Fig. 1)59. The requirements regarding the physical properties (flow, velocity, depth) and water quality (temperature, dissolved oxygen, etc.60,61) must be met for successful salmon spawning37,62. Other important requirements are: 1. Adult fish must arrive healthy at spawning grounds. This requires river continuity and that sub-lethal or lethal stress conditions are not met during migration and spawning.ih 1. Adult fish must arrive healthy at spawning grounds. This requires river continuity and that sub-lethal or lethal stress conditions are not met during migration and spawning.ih g g p g . Spawning fish must be able to construct a nest. This requires appropriate gravel size at spawning areas. Th b d ff d d b Th l l d g g p g 2. Spawning fish must be able to construct a nest. This requires appropri 2. Spawning fish must be able to construct a nest. This requires appropriate gravel size at spawning areas. 3 Th t t t b d ff t d d i i b ti Thi i t l li it d . The nest must not be scoured or suffocated during egg incubation. This requires a strongly limited presence of fine sediment among gravel. Materials and methods General approach. Our investigation focuses on Chinook salmon, considering two main effects of reser- voir flushing and natural flood events: the effects of acute stress conditions imposed during these extreme occur- rences and the effects of fine sediment deposition in the spawning areas.i fi Two distinct two-dimensional morphodynamic models are used: the first one, the Reservoir Model, is used to route the water and sediment through the reservoir, generating water and sediment fluxes to the river downstream of the dam. It is an extended version of the model of Dahal et al.5. The second one, the River Model, represents the Sandy River from the downstream of Marmot Dam location to the confluence with the Bull Run River. It is an extension of the model developed by Lee32.ht p y The two models were developed using the open-source Delft3D code version 4.03 (https://www.deltares. nl/en/software/delft3d-4-suite/) considering sediment ranging from sand to cobbles, neglecting thus the finest components of the deposited material that are transported in suspension: fine sand, silt, and clay. Considering that this study focuses on fine sediment processes, both models had to be extended.hi i The extended Reservoir Model is calibrated and validated, whereas lack of data on fine sediment deposition and transport rates along the Sandy River make calibration and validation of the extended River model impos- sible. For this, the runs include a sensitivity analysis on the effects of changing the size of suspended sediment, represented by its fall velocity, with the idea of covering a reasonable range of plausible scenarios. The results of the River model indicating the effects on water flow and sediment of the flushing scenarios and of the natural flood are then analyzed by means of stress and severity indices and suitability curves distinguishing the different life stages of salmon and its spawning habitat. Figure 2 schematizes the approach and the work flow. Modeling tools. Delft3D is an open-source software used to simulate non-steady flow and transport phe- nomena in rivers, coasts and estuaries. It allows simulating 2D and 3D hydrodynamic and transport processes of sediment mixtures5,63–65. The use of both cohesive and non-cohesive sediment of different sizes is possible, and makes this tool useful for investigating sediment transport processes in different contexts66.h g g p pf The extended models used in this study are two-dimensional (2D) and solve depth-averaged shallow-water equations. Effects of sediment flushing operations versus natural floods on Chinook salmon survival It was finally removed in 2007, when the anticipated cost of upgrading for the licensing requirement had become higher than the benefits.i p pg g g q gi At the time of dam removal, the reservoir was filled by 750,000 m3 of sediment, ranging from silt to boulders subdivided into layers of different compositions9,49. The upper layers comprised coarser sediment, mainly gravel and cobbles. While the lower layers contain fine sediment, ranging from clay to sand, but dominated by sand. A coarser layer underneath was the original riverbed. The dam removal was facilitated by the construction of a temporary cofferdam which diverted the flow of Sandy River bypassing the concrete dam; this cofferdam was later breached in a controlled way.hl y The Sandy River flows from the western part of the Mt. Hood to the Columbia River. Its major tributaries are the Zigzag River, the Salmon River, the Little Sandy River, draining from Mt. Hood, and the Bull Run River draining from the Bull Run Lake. The combined basin area is 1300 km3, with an altitude ranging from 3428 m at Mt. Hood, to 3 m a. s. l.9 at the confluence with the Columbia River (Fig. 1). The gradient of the Sandy River gradually reduces from its steep mountain parts, where it transports cobbles and boulders, to its confluence, https://doi.org/10.1038/s41598-022-19294-2 Scientific Reports \| (2022) 12:15354 \| www.nature.com/scientificreports/ Materials and methods The transport of suspended sediment is computed by means of 2D advection–diffusion equations https://doi.org/10.1038/s41598-022-19294-2 Scientific Reports \| (2022) 12:15354 \| www.nature.com/scientificreports/ Figure 2. General approach of the study and workflow direction developed using Adobe Illustrator 2021 (https://www.adobe.com/products/illustrator.html). Figure 2. General approach of the study and workflow direction developed using Adobe Illustrator 2021 (https://www.adobe.com/products/illustrator.html). coupled with a sediment entrainment and a sediment deposition equation. The sediment entrainment rate is assumed to be proportional to the difference between the local bed shear stress and its critical value for bed erosion, following the Krone and Partheniades approach67,68, multiplied by a coefficient, here named erosion coefficient. The deposition rate is obtained by multiplying the fall velocity of the sediment particles with the local depth-averaged sediment concentration69. Transport capacity formulas are used for the computation of bedload rates. The presence of several grain sizes requires considering the hiding of the smaller particles provided by the presence of the larger particles and the exposure of the latter being surrounded by smaller particles. However, the effect of hiding and exposure is not accounted in Delft 3D in the combination bedload/suspended load. This might overestimate the transport rates of the smallest bed-load particles and underestimate the largest particles. The computation of bed level change is based on sediment balances. For fine sediment travelling in suspension, bed level changes are given by the difference between sediment deposition and sediment entrainment rates. For bedload, the sediment balances follow Exner’s approach70. The effects of transverse slope on bedload direction, important for 2D bed topography changes, are accounted for according to Ikeda71, and the effects of longitudinal slope according to Bagnold72. The domains of the models and their computational grids are shown in Fig. 3. Model 1: Reservoir model. The Reservoir model comprises a spatial domain that covers the Marmot Dam and its reservoir area (3.5 km long), with the boundaries being located 1 km upstream of the reservoir influence zone and 1.4 km downstream of the Marmot Dam, schematized with a curvilinear grid (Fig. 3).t In the Delft3D model, the strata in the reservoir sediment deposit can be distinguished by sediment composi- tion, extension, and thickness. Based on the granulometric analyses of the sediment collected from the different strata49 in the reservoir (refer Fig. Materials and methods 10 of the report from Stillwater Sciences71), the extended model includes three sediment fractions: fine sediment (several sizes from sand to clay, depending on scenario), gravel (D50 = 20 mm), and cobbles (D50 = 100 mm). The distinction in fractions is needed to correctly describe the processes of sedi- ment transport, bed erosion and deposition along the river. The composition of units5,73 are selected as, Unit 1 imposed with a ratio 25:35:40 for fine sediment, gravel and cobbles, whereas Unit 2 and of the pre-dam riverbed have ratios 85:15:0 and 25:30:45, respectively. The use of different units and composition of strata is needed to correctly simulate sediment removal from the reservoir during the flushing operation. A cofferdam and wash load layer are also added to facilitate breaching. These percentages of sediment composition are derived from a range of percentages, since the sediment composition depends on location (data from Squier Associates49). Bedload is computed using the capacity formula of Aschida and Michiue74.h p g p y The time series of the discharge measured at the Marmot Dam station (located 400 m upstream of Marmot Dam before dam removal and shifted downstream of Dam after removal) allows constructing the inflow hydro- graph, whereas the sediment input at the upstream boundary is generated using a regression analysis from the available sediment measurement data at the Brightwood station.i g To properly reproduce the fine sediment input to the downstream river, a wash load layer, is added in the last 500 m of the reservoir to represent the most recent suspended sediment deposits located in the downstream part of the reservoir, at the time of cofferdam breaching.ht pf g The extended Reservoir model is calibrated and validated for sediment erosion after dam removal to estab- lish the properties of fine sediment. During the calibration process, the fall velocity and the erosion coefficient parameters, among others are tuned, to reproduce the field data, of suspended solids concentration measured during and after the Marmot Dam removal, assuming this operation is an extreme case of sediment flushing. The calibration runs cover a period of 5 days, which involves the dynamic process of cofferdam breaching. The cofferdam breaching process is modeled as a notch that is cumulatively eroded by flowing water. Scientific Reports \| (2022) 12:15354 \| Materials and methods Detailed information on the extended model is provided by Panthi75.l p y For the flushing scenarios, a virtual Marmot Dam consisting of bottom gates is introduced at the old dam location (Fig. 4). A real-time control (RTC) mechanism is used in compliance with the model to simulate the openings of the gates5. Model 2: River model. The Sandy River model covers 18 km from the Marmot Dam to the confluence of the Bull Run River (Fig. 3). The model grid comprises 1133 × 61 curvilinear cells. gh g Lee32 derived the initial riverbed topography from the LiDAR survey 200732, but this does not comprise the submerged portions of the channel bed. An initial run (spin-up) was thus executed to compute a complete realistic 2D riverbed topography and sediment composition. The spin-up simulation started with a transversally plane-bed in the submerged portion, composed of sand, gravel, and cobbles32 with sizes of 0.3, 22 and 100 mm, respectively. The morphological evolution of the riverbed was obtained by imposing a schematic discharge time series with no sediment input, due to the presence of the dam, with a time step of 0.05 min. The sediment trans- port rate was computed with the transport formula of Meyer-Peter-Muller76, applicable for bedload transport in gravel-bedded streams. The result of the spin-up run was then compared with the morphological features that are visible from aerial imagery, such as extension and location of sediment deposits and the deep areas that are evident at low-flow conditions. Model calibration is based on this comparison, since no measured data on bed topography are available for the submerged part of the river channel. p g p y g p Information regarding the sediment composition of the Sandy River before dam removal is limited to pebble counts. Field investigations found bed armoring downstream of the Marmot Dam with an average grain size of the bed material of 100 mm8,73,77,78, whereas the percentage of fines was estimated from visual valuation32. This means that there is no way to quantitively check the model results in terms of sediment composition.h The River model32 is here extended and the spin-up run re-done to include the suspended-sediment processes. The transport of coarse material is computed with the transport formula of Meyer-Peter and Müller76, whereas the transport of fine sediment is computed by means of advection–diffusion equations coupled to sediment entrainment and deposition formulations. Materials and methods Since the outflow to the downstream river reach is a function of the cofferdam breach process, the measured water discharge record at the Marmot Dam station is used as a calibration metric. The concentration of the wash load layer is calibrated on suspended sediment concentration measurements, peaking at 49 g/m3, during dam removal9. Scientific Reports \| (2022) 12:15354 \| https://doi.org/10.1038/s41598-022-19294-2 www.nature.com/scientificreports/ Figure 3. Domain of the Reservoir and River models. The boundaries are indicated by cross-sections identified by letters. Along the Sandy River the letters are in alphabetic order from upstream to downstream. Note that the boundary B defines the junction between the two models. The inflow boundary of the Bull Run River is indicated by the letter “D”. The numbered red dots indicate the USGS gauging stations. Names and identifiers are listed in the right-below corner. The figure is developed using QGIS 3.16.16 (https://qgis.org/en/site/index. html) and edited in Adobe Illustrator 2021 (https://www.adobe.com/products/illustrator.html). Figure 3. Domain of the Reservoir and River models. The boundaries are indicated by cross-sections identified by letters. Along the Sandy River the letters are in alphabetic order from upstream to downstream. Note that the boundary B defines the junction between the two models. The inflow boundary of the Bull Run River is indicated by the letter “D”. The numbered red dots indicate the USGS gauging stations. Names and identifiers are listed in the right-below corner. The figure is developed using QGIS 3.16.16 (https://qgis.org/en/site/index. html) and edited in Adobe Illustrator 2021 (https://www.adobe.com/products/illustrator.html). Validation of the extended model is based on measured data of sediment concentration and reservoir bed erosion. These data were collected in the three months that followed dam removal, during which 45% of the reservoir sediment was eroded. Information on the erosion process is acquired from the USGS report9 (https:// pubs.usgs.gov/pp/1792/) and from the surveys conducted by USGS, David Evans and PGE. Detailed information on the extended model is provided by Panthi75.l Validation of the extended model is based on measured data of sediment concentration and reservoir bed erosion. These data were collected in the three months that followed dam removal, during which 45% of the reservoir sediment was eroded. Information on the erosion process is acquired from the USGS report9 (https:// pubs.usgs.gov/pp/1792/) and from the surveys conducted by USGS, David Evans and PGE. Materials and methods The characteristics of the sediment-related variables in the extended model are listed in Table 1. Note that the adopted sediment characteristics are those obtained from the calibration of the extended Reservoir model. The initial bed composition is derived from the characteristics of the strata https://doi.org/10.1038/s41598-022-19294-2 Scientific Reports \| (2022) 12:15354 \| www.nature.com/scientificreports/ Figure 4. Reservoir model at the location of the Marmot Dam with hypothetical gate openings, with three gates for flushing and one for water diversion use. Figure 4. Reservoir model at the location of the Marmot Dam with hypothetical gate openings, with three gates for flushing and one for water diversion use. Table 1. Initial bed composition, sediment characteristics and sediment transport formula in the spin-up run of the River model computing the initial state of the river. Where D50 is the median diameter of bedload material. M-P = Meyer-Peter. Bedload characteristics Suspended load characteristics Gravel (D50) 20 mm Fall velocity 0.036 m/s Cobble (D50) 100 mm Erosion coefficient 4.1710−3 kg/m2/s Transport Formula M-P* Muller76 Critical shear stress 1 N/m2 Table 1. Initial bed composition, sediment characteristics and sediment transport formula in the spin-up run of the River model computing the initial state of the river. Where D50 is the median diameter of bedload material. *M-P = Meyer-Peter. provided by Squier Associates49. Morphological calibration of the extended River model is not possible due to lack of measured data on bed topography, bed composition and deposition of fine sediment in the river down- stream of the Marmot Dam. This means that some scenarios have to include different sediment characteristics to cover a wide range of possibilities. provided by Squier Associates49. Morphological calibration of the extended River model is not possible due to lack of measured data on bed topography, bed composition and deposition of fine sediment in the river down- stream of the Marmot Dam. This means that some scenarios have to include different sediment characteristics to cover a wide range of possibilities. Scenarios. A natural flood described by the discharge hydrograph of December 2007, with a sharp rising limb and a gentler falling limb, typical of Sandy River floods, is selected as a reference for the flushing and the natural flood scenarios (see Supplementary Fig. A1). Based on the flow frequency statistics provided by Major et al.9, this flood has a return time of less than two years. Materials and methods Model scenarios.The River model also includes the additional natural flood scenario. ment and cause behavioral problems, physical damage and even death to fish and invertebrates20,24,30,31,34,35. The exposure to suspended solids is detrimental to salmon species too and depends on concentration and duration of exposure34,79. The severity scale developed by Newcombe and Jensen35 is here used to study the acute impact on salmonids caused by high suspended sediment concentrations during the flushing operation or by the flood event. The severity scale is given as a function of concentration of suspended sediment (SSC) and duration (D). The severity scale of Newcombe and Jensen are based on regression lines fitting literature data on various salmon species, considering suspended sediment with particle size up to 250 µm35. The scale relates to the effects of SSC on individual fish and does not account the fish preconditions nor any other factors causing unsuitable environment, e.g. rise in water temperature or decrease in dissolved oxygen. Considering the high level of uncer- ment and cause behavioral problems, physical damage and even death to fish and invertebrates20,24,30,31,34,35. The exposure to suspended solids is detrimental to salmon species too and depends on concentration and duration of exposure34,79. The severity scale developed by Newcombe and Jensen35 is here used to study the acute impact on salmonids caused by high suspended sediment concentrations during the flushing operation or by the flood event. The severity scale is given as a function of concentration of suspended sediment (SSC) and duration (D). Th i l f N b d J b d i li fi i li d i ment and cause behavioral problems, physical damage and even death to fish and invertebrates20,24,30,31,34,35. The exposure to suspended solids is detrimental to salmon species too and depends on concentration and duration of exposure34,79. The severity scale developed by Newcombe and Jensen35 is here used to study the acute impact on salmonids caused by high suspended sediment concentrations during the flushing operation or by the flood event. The severity scale is given as a function of concentration of suspended sediment (SSC) and duration (D).hi h y g p The severity scale of Newcombe and Jensen are based on regression lines fitting literature data on various salmon species, considering suspended sediment with particle size up to 250 µm35. Materials and methods The scale relates to the effects of SSC on individual fish and does not account the fish preconditions nor any other factors causing unsuitable environment, e.g. rise in water temperature or decrease in dissolved oxygen. Considering the high level of uncer- tainty, also related to the fact that the data are from differing river environments, the severity index should be used to compare situations rather than to quantitatively predict the effect of SSC on a specific fish species in a specific river. Notwithstanding this, the choice of using this severity index lies in its simplicity of application and in the scope of the work, aiming at comparing the effects of flushing operation and flood events.f p g p gfl g pl Newcombe and Jensen35 expressed severity on a scale of 0 to 14, distinguishing the effects in behavioral (0 sub-lethal (5–8) and lethal (9–14) for each life stage. Adult salmonids (1) SEV = 1.6814 + 0.4769 loge D + 0.7565 loge SSC (1) Juvenile salmonids Juvenile salmonids (2) SEV = 0.7262 + 0.7034 loge D + 0.7144 loge SSC (2) Eggs and larvae of salmonids (3) SEV = 3.7466 + 1.0946 loge D + 0.3117 loge SSC (3) where, D is the duration of exposure in hours and SSC is the suspended sediment concentration in mg/l. The severity indexes are here calculated using three different metrics: where, D is the duration of exposure in hours and SSC is the suspended sediment concentration in mg/l. The severity indexes are here calculated using three different metrics: where, D is the duration of exposure in hours and SSC is the suspended sediment concentration in mg/l. The severity indexes are here calculated using three different metrics: The severity indexes are here calculated using three different metrics: i. The concentration of suspended sediment is plotted against the duration in which it is exceeded, which creates a concentration duration curve of the flushing operation. The value of the concentration that is exceeded for 50% of the time is then multiplied by total duration of the flushing operation to calculate the representative value of severity30.hf i. The concentration of suspended sediment is plotted against the duration in which it is exceeded, which creates a concentration duration curve of the flushing operation. Materials and methods The value of the concentration that is exceeded for 50% of the time is then multiplied by total duration of the flushing operation to calculate the representative value of severity30.hf p y ii. The time series of severity is computed for the different life stages based on the duration curves of sus- pended sediment concentration for different locations along the river reach.hf ii. The time series of severity is computed for the different life stages based on the duration curves of sus- pended sediment concentration for different locations along the river reach.hf f iii. The acceptable duration of exposure is compared with the actual duration for different concentration values35. f iii. The acceptable duration of exposure is compared with the actual duration for different concentration values35. Fine sediment deposition: impacts on spawning areas. Deposition of fine sediment on spawning grounds with laid eggs can cause mortality and sub-lethal effects to eggs and emerging larvae, creating an acute impact on salmon80. Moreover, the deposition alters the composition of the riverbed, which may further result in unsuitability of the area as spawning ground, which might be a persisting, and thus long-term effect.i Fine sediment deposition: impacts on spawning areas. Deposition of fine sediment on spawning grounds with laid eggs can cause mortality and sub-lethal effects to eggs and emerging larvae, creating an acute impact on salmon80. Moreover, the deposition alters the composition of the riverbed, which may further result in unsuitability of the area as spawning ground, which might be a persisting, and thus long-term effect.i y p g g g p g gf Habitat suitability index (HSI) models are developed to categorize and quantify the ability of defined areas to meet the physical requirements of specific habitats. They relate the values of each physical variable describing the local aquatic environment81,82, e.g. flow velocity, water depth, bed composition and water temperature, to its suitability for a specific species or populations of fish or invertebrates25,28,29,36,37,83. Considering the strongly differing requirements, distinct indices are derived for eggs, larvae, juveniles and adults of the same species35. Each HSI is normally based on the statistical analysis of scarce field data84, for instance local fish counts versus the value of a specific physical variable. It may thus present a large degree of uncertainty due to the weak or even inconsistent statistical relations that are often found85. Materials and methods The boundary conditions of the Reservoir model are the measured daily discharge time series and the sediment inflow derived from the data measured at the Brightwood station. The input of suspended solids is derived using a relation between discharge and sediment load generated from a regression analysis of the available sediment concentration data from the same station.hll g y The flushing scenarios are chosen based on the timing of gate opening during the peaking of the flood wave, with only Gate 2 (width: 9.72 m; height: 3 m) in operation (Fig. 4). This pattern of gate opening is considered the most effective one5, since it allows flushing higher volumes of sediment in a shorter period.ll fl In the corresponding natural flood scenario, the Sandy River is assumed to be freely flowing without the dam. This means that in this case the Reservoir model is not used to generate the discharge and sediment inputs to the River model. The input of fine sediment to the river is computed from the regression law derived for the Brightwood station. The input of coarse sediment is computed using the selected transport formula. All scenarios are listed in Table 2.hff The effects of different suspended sediment sizes are studied separately considering three fall velocities rep- resenting the characteristics of clay, silt, and medium sand (see Supplementary Table A1). The input of water and sediment to the river for these three scenarios corresponds to reservoir flushing with gate opening at 80% of peak flow (moment of gate opening is indicated in Supplementary Fig. A1). The impact of suspended solids concentration: stress and severity on salmonids. High sus- pended sediment concentration in the water column causes acute effects on aquatic biota. These short-term effects are not accounted while defining the suitability of habitats, but they can create an unbearable environ- https://doi.org/10.1038/s41598-022-19294-2 Scientific Reports \| (2022) 12:15354 \| www.nature.com/scientificreports/ Table 2. Model scenarios.The River model also includes the additional natural flood scenario. Scenario number Reservoir model River model Input: discharge & sediment load Input = Output from reservoir model 1 RF_20 Opening at 20% of peak (Qin = 48 m3/s) RF_20 2 RF_50 Opening at 50% of peak (Qin = 120 m3/s) RF_50 3 RF_80 Opening at 80% of peak (Qin = 192 m3/s) RF_80 4 No dam Natural flood (Upstream boundary as described in Fig. A1) Table 2. Scientific Reports \| (2022) 12:15354 \| Results R i Reservoir model: calibration and validation. The Reservoir model is updated and re-calibrated based on the measured sediment transport data collected by USGS (see appendix in Major et al.9). These data are col- lected during the first five days after dam removal, during the major erosive process of the cofferdam breaching. The results are: a fall velocity of 0.036 m/s, a critical bed shear stress for erosion of 1 N/m2 and an erosion coef- ficient of 4.17 × 10−3 kg/m2/s for the sand fraction. The insertion of a wash load layer with a concentration of 0.5 kg/m2 along a distance of 500 m upstream of dam in the reservoir water surface provides the best representa- tion of the suspended sediment flux at the time of breaching. The suspended sediment flux reached the value of 0.6 m3/s, nevertheless, the computed values remain lower than the measured ones (Fig. 5a). Further details on this re-calibration are given by Panthi75. Validation is based on the cumulative sediment volumes eroded from the reservoir that is measured by USGS9 after cofferdam breaching. The computed values are in good agreement with the measured ones (Fig. 5b): after three months, 45% of the total reservoir sediment is eroded whereas the modelled erosion is 40%. Suspended sediment outputs during flushing. The sudden opening of the bottom gates results in elevated levels of SSC, but for a relatively short duration (minutes/hours). The highest flushing discharge is 513 m3/s, obtained for gate opening at 80% of the peak flow (Fig. 6). The highest sand transport rates are 4.75 m3/s, 3.68 m3/s and 2.00 m3/s for gate opening at 80%, 50% and 20% of the peak flow, respectively. The highest concentrations, however, are rather similar, with 39,000 mg/l, 40,000 mg/l and 33,000 mg/l, respectively (Fig. 6).hfi g The results of the sensitivity analysis, considering different sizes of the fine sediment fraction, show that deposition of silt and clay occurs sparsely in the study area (see Supplementary Fig. A2). This is due to the low fall velocity of these sediment types and the high velocity of the water flow. Instead, sand settles at several loca- tions, mostly between the location of the dam and the entrance of the Sandy River gorge (upper reach, Fig. 1), but also in the pools within the gorge. Materials and methods Nevertheless, habitat suitability indexes are commonly used to assess the distribution of specific species, as well as to objectively translate the physical alterations caused by external factors into effects on habitats86. Th d h hl h h h h b b l d b fl h f Habitat suitability index (HSI) models are developed to categorize and quantify the ability of defined areas to meet the physical requirements of specific habitats. They relate the values of each physical variable describing the local aquatic environment81,82, e.g. flow velocity, water depth, bed composition and water temperature, to its suitability for a specific species or populations of fish or invertebrates25,28,29,36,37,83. Considering the strongly differing requirements, distinct indices are derived for eggs, larvae, juveniles and adults of the same species35. l its suitability for a specific species or populations of fish or invertebrates25,28,29,36,37,83. Considering the strongly differing requirements, distinct indices are derived for eggs, larvae, juveniles and adults of the same species35. Each HSI is normally based on the statistical analysis of scarce field data84, for instance local fish counts versus the value of a specific physical variable. It may thus present a large degree of uncertainty due to the weak or even inconsistent statistical relations that are often found85. Nevertheless, habitat suitability indexes are commonly used to assess the distribution of specific species, as well as to objectively translate the physical alterations caused by external factors into effects on habitats86.hl yf This study aims at highlighting the changes in spawning habitat suitability caused by flushing operations or by the corresponding flood event. Being specific for Chinook salmon, the study adopts the HSI model developed by Raleigh et al.37 based on a large number of field and laboratory data. The analysis is based on three variables: water depth, flow velocity and percentage of fines in the riverbed. The index is scaled as a real number from 0 to https://doi.org/10.1038/s41598-022-19294-2 Scientific Reports \| (2022) 12:15354 \| www.nature.com/scientificreports/ Figure 5. Modelled and measured (a) suspended sediment transport rates in the Sandy River at the Marmot Dam Station during cofferdam breaching (results of model calibration and validation) and, (b) percentage of sediment eroded from the reservoir during the validation period. The blue line indicates the incoming discharge. Figure 5. Materials and methods Modelled and measured (a) suspended sediment transport rates in the Sandy River at the Marmot Dam Station during cofferdam breaching (results of model calibration and validation) and, (b) percentage of sediment eroded from the reservoir during the validation period. The blue line indicates the incoming discharge. 1, where 0 means unsuitable and 1 is an optimal condition. HSI is computed for different life stages. The HSI is then calculated as the product of the suitability indexes of the three variables as , where 0 means unsuitable and 1 is an optimal condition. HSI is computed for different life stages. The HSI is hen calculated as the product of the suitability indexes of the three variables as (4) HSI = SI1 ∗SI2 ∗SI3 (4) The total suitable area for spawning in the considered river reach is finally calculated by applying the total weighted usable area (WUA) approach29 in which the suitable area is expressed as a percentage of the total surface area (wet area) of the considered reach: (5) WUA = AiHSIi WUA = AiHSIi (5) where, Ai is area of the grid cell and HSIi is the value of the HSI value for the same cell. Results R i Reference value of severity relative to three life stages of salmon for the considered flushing scenarios and the natural flood scenario at different locations along the Sandy River downstream of the Marmot Dam. Egg refers to egg and larvae. The numbers in bold fall within the lethal range, whereas numbers in italics refer to sub-lethal range. Impact of suspended sediment concentration on Chinook salmon. The results of the three flush- ing scenarios and of the corresponding natural flood are post-processed to determine the corresponding value of severity (Eqs. 1–3) with the aim to quantify the impact of SSC on Chinook salmon, distinguishing the different life stages. The sediment concentration distribution and its duration at different locations along the Sandy River were computed by the River model. The upstream inputs were computed by the Reservoir model, but only for the flushing scenarios. The reference value of the severity index is obtained considering the value of SSC that exceeds 50% in the duration curve, multiplied by the duration of the entire flushing period (Table 3). Table 3 shows lethal conditions for eggs and larvae with values ranging from 10 to 12 indicating up to 40% mortality35 for all scenar- ios and along the entire Sandy River. Increasing order of severity is observed in the downstream direction. The severity is comparatively higher for the flushing scenario with gate opening at 20% of the incoming flood peak, whereas the lowest severity is found for the corresponding natural flood. Note that this analysis is performed for sand. The computed values of SSC for silt and clay are even higher, resulting in greater severity indexes.hh h The time series of severity for the three life stages of salmon is derived from the duration curves (Fig. 7). The results show lethal conditions for egg and larvae for all scenarios, including the corresponding natural flood. Low concentrations are present for longer periods of time and exceed their acceptable duration limit. Instead, the adult and juvenile stages of salmon do not reach the mortality level (SEV > 10).ll j g y ( ) For the natural flood, the SSC does not exceed the value of 2500 mg/l, whereas for the flushing operations he concentration is as high as 40,000 mg/l.ll For the flushing scenario starting at 20% of the flood peak, the suspended sediment concentrations have the longest duration of exposure and produce the highest values of severity. Results R i Due to deposition, sand presents a decrease in concentration in the downstream direction, whereas silt and clay, instead, present an increase due to the entrainment of particles from the riverbed and banks. These fine particles are found to only alter the suspended sediment concentration and hardly change the bed composition. Therefore, the analysis of the effects of flushing on salmon spawning habitats concentrates on sand.h The sand that immediately settles in the upper reach, between the dam location and the gorge, is later trans- ported away by the later flow. This results in persisting high sand concentrations in the downstream reaches. The river banks were eroded in the first few hours of flushing, but later bank erosion stopped as the discharge decreased hence decreasing proximal sediment source. Scientific Reports \| (2022) 12:15354 \| https://doi.org/10.1038/s41598-022-19294-2 www.nature.com/scientificreports/ Figure 6. Computed discharge and suspended sediment concentration (sand) for different flushing scenarios corresponding to gate opening at 20%, 50% and 80% of the peak flow, 200 m downstream of the Marmot Dam location. Figure 6. Computed discharge and suspended sediment concentration (sand) for different flushing scenarios corresponding to gate opening at 20%, 50% and 80% of the peak flow, 200 m downstream of the Marmot Dam location. Table 3. Reference value of severity relative to three life stages of salmon for the considered flushing scenarios and the natural flood scenario at different locations along the Sandy River downstream of the Marmot Dam. Egg refers to egg and larvae. The numbers in bold fall within the lethal range, whereas numbers in italics refer to sub-lethal range. Distance from the dam [km] Natural Flood scenario Flushing scenario Gate opening at 20% Flushing scenario Gate opening at 50% Flushing scenario Gate opening at 80% Adult Juvenile Egg Adult Juvenile Egg Adult Juvenile Egg Adult Juvenile Egg 0.2 6.79 6.96 10.85 7.48 7.61 11.13 7.52 7.64 11.14 7.78 7.89 11.25 4.7 8.08 8.18 11.38 8.78 8.84 11.67 8.48 8.55 11.54 8.74 8.80 11.65 9.2 8.37 8.45 11.50 8.96 9.01 11.74 8.63 8.69 11.60 8.73 8.79 11.64 10.5 8.52 8.59 11.56 8.99 9.04 11.75 8.85 8.90 11.69 8.90 8.95 11.72 15.7 9.23 9.26 11.85 9.32 9.34 11.89 9.35 9.38 11.90 9.35 9.37 11.90 18.0 9.27 9.30 11.87 9.37 9.39 11.91 9.40 9.43 11.92 9.36 9.39 11.91 Table 3. www.nature.com/scientificreports/ www.nature.com/scientificreports/ Figure 8. Temporal variation of the percentage of river surface that is suitable for the different life stages of Chinook salmon after the start of (a) the natural flood event and (b) the flushing operation starting at 80% of the flood peak and, the percentage of area suitable for eggs according to water depth, flow velocity and fine sediment ftl Figure 8. Temporal variation of the percentage of river surface that is suitable for the different life stages of Chinook salmon after the start of (a) the natural flood event and (b) the flushing operation starting at 80% of the flood peak and, the percentage of area suitable for eggs according to water depth, flow velocity and fine sediment and composite Habitat Suitability (HSI) at different times after the start of (c) the natural flood event and (d) the flushing operation starting at 80% of the flood peak. Figure 8. Temporal variation of the percentage of river surface that is suitable for the different life stages of Chinook salmon after the start of (a) the natural flood event and (b) the flushing operation starting at 80% of the flood peak and, the percentage of area suitable for eggs according to water depth, flow velocity and fine sediment and composite Habitat Suitability (HSI) at different times after the start of (c) the natural flood event and (d) the flushing operation starting at 80% of the flood peak. Impact of sediment deposition on Chinook salmon spawning habitat. Being the most effective flushing scenario, for the study of the impact on spawning grounds and other habitat areas, we only consider the operation that starts at 80% of the flood peak. The changes in habitat suitability caused by this operation and by the corresponding natural flood at different times are shown in Fig. 8a, b. The foregoing weighted usable areas (Eq. 5) for spawning in the Sandy River are indicated too, for sake of comparison.li Figure 8c, d shows the contribution of water depth, flow velocity and percentage of fines in the substrate through their specific suitability indices for eggs (Eq. 4). The initial condition at time t = 0 h presents a relatively low flow velocity and a pre-flushed riverbed composition. This corresponds to higher suitability for adults with optimum flow velocity. The initial riverbed composition appears less suitable, especially for juvenile fry and egg. Results R i The scenarios with gate opening at 50% and 80% of the flood peak have similar impacts. However, the scenario with the operation starting at 80% of the flood peak releases higher sediment volumes from the reservoir with similar or less acute stress to salmon and for this it is regarded as the most effective flushing operation. https://doi.org/10.1038/s41598-022-19294-2 Scientific Reports \| (2022) 12:15354 \| www.nature.com/scientificreports/ com/scientificreports/ Figure 7. Instantaneous severity for three flushing scenarios at different locations: just downstream of the dam (200 m), at the end of the river gorge (10,500 m) and just before the confluence of the Bull Run River (18,000 m) (a) during the corresponding natural flood event; (b) gate opening at 80% of incoming peak, (c) gate opening at 50% of incoming peak and (d) gate opening at 20% of incoming peak. Figure 7. Instantaneous severity for three flushing scenarios at different locations: just downstream of the dam (200 m), at the end of the river gorge (10,500 m) and just before the confluence of the Bull Run River (18,000 m) (a) during the corresponding natural flood event; (b) gate opening at 80% of incoming peak, (c) gate opening at 50% of incoming peak and (d) gate opening at 20% of incoming peak. Figure 7. Instantaneous severity for three flushing scenarios at different locations: just downstream of the dam (200 m), at the end of the river gorge (10,500 m) and just before the confluence of the Bull Run River (18,000 m) (a) during the corresponding natural flood event; (b) gate opening at 80% of incoming peak, (c) gate opening at 50% of incoming peak and (d) gate opening at 20% of incoming peak. https://doi.org/10.1038/s41598-022-19294-2 Scientific Reports \| (2022) 12:15354 \| Discussionh The concentration of suspended sediment is key to salmon survival. However, the results indicate that the magnitude of the suspended sediment release can be underestimated by the Reservoir model. This model represents well the overall phenomena, with rapid channel excavation within the reservoir deposit and retrogressive bed erosion. The widening of the excavated channel, though, is small compared to the measured one and might be the reason for the underestima- tion of sediment release. However, this might be also related to the uncertainty posed by the composition of the deposit layers and horizontal sediment sorting in the reservoir. The possible underestimation of the sediment releases from the reservoir should be always considered when analyzing the results of the model, particularly if the predicted concentra- tions fall just under the sub-lethal threshold. j About the downstream river, the availability of fine sediment in the bed is important for the computation of both SSC and deposition rates by the River model. The initial bed composition was obtained in a straightforward way through the spin-up run, but without the possibility to calibrate the morphodynamic part of the model. Lack of field data meant uncertainty in the identification of the initial spawning ground suitability, i.e. before the flush- ing operations and the natural event, affecting the assessment of their impacts. The difficulty of establishing the initial riverbed composition affected the impact assessment even more, considering that the results of the River model show that sediment entrainment from the riverbed and eroding banks87 is important for the assessment of both stress and habitat losses (Fig. 7).f g About the stress caused by the exposure to suspended solids, the results show that the differences in severity between the flushing scenarios and the natural flood are small. Flushing operations produced higher suspended sediment concentrations but for shorter periods of time compared to the natural flood and for this they resulted in similar severity levels. For instance, the conditions become lethal for eggs and larvae in all cases and the natural flood conditions result only in slightly lower severity for adults and juveniles as compared to the flushing opera- tions. The deposition of fine sediment on the riverbed caused a similar reduction of suitable spawning habitat, but the recovery was higher and faster at the end of the flood event compared to the flushing. www.nature.com/scientificreports/ Temporal variation of spawning Weighted Usable Area (WUA) for Chinook salmon (black bars) and suitability in terms of % of fines deposited (grey bars) in the Sandy River with clear water release at a rate of 30 m3/s after the flushing operation. The WUA is based on the percentage of fines in the riverbed. “Initial” refers to the conditions just before the flushing operation. base-flow of the hydrograph (see Supplementary Fig. A1). As clear water flows, fine sediment is progressively removed from the riverbed. After nine days of clear water release, however, the riverbed is still only slightly cleaned (Fig. 9). This shows that at the chosen flow rate the riverbed takes weeks to be cleaned up and indicates that flushing would cause a long-term disturbance to the spawning habitats. base-flow of the hydrograph (see Supplementary Fig. A1). As clear water flows, fine sediment is progressively removed from the riverbed. After nine days of clear water release, however, the riverbed is still only slightly cleaned (Fig. 9). This shows that at the chosen flow rate the riverbed takes weeks to be cleaned up and indicates that flushing would cause a long-term disturbance to the spawning habitats. www.nature.com/scientificreports/ However, this might be also related to the uncertainty posed by the composition of the deposit layers and horizontal sediment sorting in the reservoir. The possible underestimation of the sediment releases from the reservoir should be always considered when analyzing the results of the model, particularly if the predicted concentra- tions fall just under the sub-lethal threshold. About the downstream river, the availability of fine sediment in the bed is important for the computation of both SSC and deposition rates by the River model. The initial bed composition was obtained in a straightforward way through the spin-up run, but without the possibility to calibrate the morphodynamic part of the model. Lack of field data meant uncertainty in the identification of the initial spawning ground suitability, i.e. before the flush- ing operations and the natural event, affecting the assessment of their impacts. The difficulty of establishing the Figure 9. Temporal variation of spawning Weighted Usable Area (WUA) for Chinook salmon (black bars) and suitability in terms of % of fines deposited (grey bars) in the Sandy River with clear water release at a rate of 30 m3/s after the flushing operation. The WUA is based on the percentage of fines in the riverbed. “Initial” refers to the conditions just before the flushing operation. base-flow of the hydrograph (see Supplementary Fig. A1). As clear water flows, fine sediment is progressively removed from the riverbed. After nine days of clear water release, however, the riverbed is still only slightly Figure 9. Temporal variation of spawning Weighted Usable Area (WUA) for Chinook salmon (black bars) and suitability in terms of % of fines deposited (grey bars) in the Sandy River with clear water release at a rate of 30 m3/s after the flushing operation. The WUA is based on the percentage of fines in the riverbed. “Initial” refers to the conditions just before the flushing operation. Figure 9. Temporal variation of spawning Weighted Usable Area (WUA) for Chinook salmon (black bars) and suitability in terms of % of fines deposited (grey bars) in the Sandy River with clear water release at a rate of 30 m3/s after the flushing operation. The WUA is based on the percentage of fines in the riverbed. “Initial” refers to the conditions just before the flushing operation. Figure 9. www.nature.com/scientificreports/ After the events, the habitat suitability for salmon at the study area is reduced by both the natural flood and the flushing operation. The reduction in suitable area for adult salmon is due to flow velocity, which depends on discharge and is thus not permanent, whereas for the eggs it is due to the deposition of fines. The highest content of fines in the riverbed occurs 12 h after the discharge peak (Fig. 8c, d). Part of the deposited fine sediment is eroded when the incoming suspended sediment concentration is reduced. At the end of the simulation, when the discharge reaches to baseflow, the usable area for spawning is reduced to 8.3% and 5.6% of the total wet area by the natural flood and the flushing operation, respectively, compared to an initial 15%. This corresponds to a loss of 44% and 62%, respectively.hllt , p y The comparison between the natural flood and the flushing operation (Fig. 8c, d) shows that after the natural flood the substrate presents a slightly higher recovery with less fine sediment content in the riverbed.l li To investigate the possibility to clean the riverbed from the sand deposited during the flushing operation, an additional run is carried out in which the flushing scenario with gate opening at 80% of the flood peak is followed by bottom gate closure and release of clear water to the river at a rate of 30 m3/s, which represents the https://doi.org/10.1038/s41598-022-19294-2 Scientific Reports \| (2022) 12:15354 \| www.nature.com/scientificreports/ base-flow of the hydrograph (see Supplementary Fig. A1). As clear water flows, fine sediment is progressively removed from the riverbed. After nine days of clear water release, however, the riverbed is still only slightly cleaned (Fig. 9). This shows that at the chosen flow rate the riverbed takes weeks to be cleaned up and indicates that flushing would cause a long-term disturbance to the spawning habitats. Discussion The concentration of suspended sediment is key to salmon survival. However, the results indicate that the magnitude of the suspended sediment release can be underestimated by the Reservoir model. This model represents well the overall phenomena, with rapid channel excavation within the reservoir deposit and retrogressive bed erosion. The widening of the excavated channel, though, is small compared to the measured one and might be the reason for the underestima- tion of sediment release. Discussionh The severity was found to increase in downstream direction, due to the rise of suspended sediment concentration caused by the entrainment of fines from the riverbed, and the eroding banks.hh i g The severity for eggs and larvae mainly depends on the duration of exposure to suspended solids. The analysis is based on the severity scale of Newcombe and Jensen35, but this might not well represent the effects. This scale does not consider the stage or condition of the fish before the exposure, and the severity, calculated with one value of the concentration, does not consider the cumulative effect of exposure. Moreover, the derived severity scale is based on field data analysis, whereas the response to suspended solids is highly site-specific10. For instance, the fish that is exposed to extreme events frequently can be more tolerant or able to escape to neighboring tributar- ies with lower SSC88.i Next to stress severity, the deposition of fines on the riverbed during the events is found to produce an important loss of spawning habitats. The results of the River model indicate that up to 95% of the eggs laid in the riverbed would be buried. Lab experiments by Jensen et al.24 showed a decrease in egg and fry survival by 16.9% for every 1% increase in the content of fines (< 0.85 mm) in the substrate.l yi ( ) It is important to note that the highest natural floods of the Sandy River normally occur in November, December and January, followed by lower flow peaks in February-March50,51. In this river, salmon spawns in September-December. This means some natural floods occur during the critical spawning and egg incubation https://doi.org/10.1038/s41598-022-19294-2 Scientific Reports \| (2022) 12:15354 \| www.nature.com/scientificreports/ Figure 10. Comparison of suitable spawning areas of the riverbed with natural flood, flushing only, and flushing followed by clean water release at a constant rate equal to the river base-flow. “Suitability” assumes that gravel of the right size is present in the riverbed. Figure 10. Comparison of suitable spawning areas of the riverbed with natural flood, flushing only, and flushing followed by clean water release at a constant rate equal to the river base-flow. “Suitability” assumes that gravel of the right size is present in the riverbed. periods. Discussionh A decline in live fish and redd counts after the rain storms of October 2016 was reported58, indicating that the water was highly turbid, qualitatively confirming the results of this study.hl The comparison between spawning area suitability shows that flushing causes a more persisting disturbance to spawning habitats than the natural flood, even if followed by riverbed cleaning by the constant release of clear water at a rate equal to the river base-flow (Fig. 10). The impact is present for almost a year if the SSC of the flow is high89, slowing the river recovery process and any restoration plans90. A more irregular clear water release, presenting short but higher discharge peaks, could increase the riverbed flushing rate and reduce the habitat restoration times. In any case, releasing large amounts of clear water might reduce the water storage in the reservoir to unacceptable levels, which limits the applicability of this type of operations. This means that riverbed flushing requires a specific study leading to the optimization of the release of precious clear water from the reservoir.il It is necessary to finally consider that the analysis of habitat suitability is only based on flow velocity, water depth and percentage of fines. The study considered a specific flood event, the equivalent one, i.e. the one consid- ered also for the design of the flushing operation. Considering another flood event for both the flushing opera- tions and the natural flood would lead to different physical conditions and habitat alterations. It is important to consider that also other factors, such as temperature, dissolved oxygen, turbidity, season, and water quality, as well as multiple events, which might affect the results, are not considered in this study. f Extreme events change not only the sediment composition, but also the riverbed topography and the river course, which might alter the connectivity of the floodplains and hence the riverine habitats90,91. This study does not consider this aspect. Being based on models (severity and HSI) with high degrees of uncertainty, the results of this study should be interpreted in terms of comparison between scenarios.i y p p Finally, it should be considered that the results of this study are case-sensitive, with specific availability and type of sediment in the reservoir, flushing discharge, and reservoir characteristics, all factors that govern the quantity and quality of the sediment released from the reservoir and its impact on salmon. www.nature.com/scientificreports/ simulate the propagation and deposition of fine sediment along the river downstream of the dam. The sediment inputs generated by the flushing operations were simulated by means of another 2D model covering the Marmot Dam Reservoir and the dam, assumed to be in operation with appropriate bottom gates.hf simulate the propagation and deposition of fine sediment along the river downstream of the dam. The sediment inputs generated by the flushing operations were simulated by means of another 2D model covering the Marmot Dam Reservoir and the dam, assumed to be in operation with appropriate bottom gates.hf g The model results allowed assessing the stress caused by the exposure to suspended solids (short-term effect) through the analysis of severity indices derived for salmonids and the habitat loss (long-term effect) through the analysis of Habitat Suitability Indexe for Chinook salmon.h y y The results indicate that excessive exposure to suspended sediment concentration during either the natural or the artificial (flushing) high-flow events is lethal to salmon eggs and alevins. Spawning habitat losses for flushing and flood events are found to be very similar, with 95% and 93% of losses, respectively. The eggs already laid would have severe physiological and lethal effects during a flushing operation, as well as during a natural flood.i p y gf gl g p gl Although the short-term impacts are similar, the long-term impact on spawning grounds due to fine sedi- ment deposition is found to be higher in case of flushing. Cleaning of the riverbed by releasing sediment-free water at a rate equal to the river base-flow appears not effective in gaining suitable spawning beds and reach- ing the conditions that are present after the natural flood. The effectiveness of riverbed cleaning with different clear-water flow releases, possibly including peak discharges of short duration mimicking natural flows, should be further investigated.ll g Reservoir flushing with gate opening at 80% of an incoming flood peak is advisable, not only considering the efficiency of the operation in terms of sediment release and duration, but also because this is the flushing operation having the smallest impact, although only slightly.l In the study area the highest flows normally occur in November-January whereas the critical spawning and egg incubation period of salmon is September to December. Fry appear approximately four months after eggs are laid, i.e. in December-March. www.nature.com/scientificreports/ This means that most natural floods harm salmon eggs and alevins, as the results of this study show, especially if they occur in the last months of the year. Considering this, the best moment for sediment flushing would be after March. However, the operation should start at the peak of a flood wave and subsequent riverbed cleaning operations ideally need natural floods too. Looking at the typical yearly hydrograph of the Sandy River, having this operation in the second half of January could offer an acceptable compromise between loosing eggs and recently hatched fish and need to flush the reservoir and clean the riverbed.ll g gg yil If flushing starts at the peak of a flood wave with duration comparable to the natural event, the timing of the operation would fall in the temporal ranges of natural floods, to which the river biota has adapted. By showing that the effects of floods and flushing are comparable, at least with regard to salmon, the results of this study indicate that well-planned timing of flushing followed by effective river bed cleaning, if achievable, would sub- stantially minimize the effects of dam operation. Data availability y All data, model inputs and files are downloadable upon request to the corresponding author from: http://www. hydroshare.org/resource/7674e96fa50b436fba80ba59d566a767. Information on discharge data and sediment data can be obtained from: https://pubs.usgs.gov/pp/1792/ and USGS stations. Received: 29 March 2022; Accepted: 26 August 2022 References 1. Morris, G. L. & Fan, J. Reservoir Sedimentation Handbook: Design and Management of Dams, Reservoirs, and Watersheds for Sus- tainable Use (McGraw Hill Professional, 1998).h 2. White, R. Evacuation of Sediments from Reservoirs, HR Wallingford, http://www.thomastelford.com (Thomas Telford Publishing 2001).i 3. Kondolf, G. M. et al. Sustainable sediment management in reservoirs and regulated rivers: Experiences from five continents. Earth’s Future 2, 256–280 (2014). 4. Schleiss, A. J., Franca, M. J., Juez, C. & De Cesare, G. Reservoir sedimentation. J. Hydraul. Res. 54, 595–614 (2016). hleiss, A. J., Franca, M. J., Juez, C. & De Cesare, G. Reservoir sedim 5. Dahal, S., Crosato, A., Omer, A. Y. A. & Lee, A. A. Validation of model-based optimization of reservoir sediment releases by dam removal. J. Water Resour. Plan. Manag. 147, 04021033 (2021).f 6. Williams, G. P. & Wolman, M. G. Effects of dams and reservoirs on surface water hydrology—Changes in rivers downstream from dams. Natl. Water Summ. Hydrol. Events Surf. Water Resour. 2300, 83 (1986).fhf f dams. Natl. Water Summ. Hydrol. Events Surf. Water Resour. 2 y f 7. Toffolon, M., Siviglia, A. & Zolezzi, G. Thermal wave dynamics in rivers affected by hydropeaking. Water Resour. Res. https://doi org/10.1029/2009WR008234 (2010). g 8. Stewart, G. B. Patterns and Processes of Sediment Transport Following Sediment-Filled Dam Removal in Gravel Bed Rivers. (PhD Thesis, Oregon State University, Oregon USA, 2006). h 9. Major, J. J. et al. Geomorphic Response of the Sandy River, Oregon, to Removal of Marmot Dam. U.S. Geological Survey Professional Paper, 64p https://pubs.usgs.gov/pp/1792/ (2012).fli p p p p g g pp 0. Espa, P., Castelli, E., Crosa, G. & Gentili, G. Environmental effects of storage preservation practices: Controlled flushing of fine sediment from a small hydropower reservoir. Environ. Manag. 52, 261–276 (2013).ll y g 11. Tena, A., Vericat, D. & Batalla, R. J. Suspended sediment dynamics during flushing flows in a large impounded river (the lower River Ebro). J Soils Sediments 14, 2057–2069 (2014).i 12. Antoine, G., Camenen, B., Jodeau, M., Némery, J. & Esteves, M. Downstream erosion and deposition dynamics of fine suspended sediments due to dam flushing. J. Hydrol. 585, 124763 (2020). sediments due to dam flushing. J. Hydrol. 585, 124763 (2020) l g y 3. Power, M., Dietrich, W. & Finlay, J. Dams and downstream aquatic biodiversity: Potential food web consequences of hydrologic and geomorphic change. Environ. Manag. 20, 887–895 (1996).lf g p g g 14. Clarke, K. Conclusionsh This work investigates the short (acute stress) and the long-term effects (river habitat alterations) of sediment flushing from a reservoir on Chinook salmon. Considering that flushing operations and natural floods present many similarities, the aim of this study is to assess the difference in terms of impacts between reservoir flush- ing and natural floods. Can the negative effects be reduced by proper planning/design of flushing operations?fll lfl To meet the goal, the effects of reservoir flushing, conveniently starting during a flood event, are compared to the effects of the corresponding natural flood wave without reservoir. The Sandy River, Oregon USA, was chosen as a case study due to the availability of measured data from the period around the Marmot Dam removal in 2007, considering dam removal as an extreme case of sediment flushing. A 2D morphological model was used to https://doi.org/10.1038/s41598-022-19294-2 Scientific Reports \| (2022) 12:15354 \| www.nature.com/scientificreports/ www.nature.com/scientificreports/ www.nature.com/scientificreports/ 17. Brandt, S. A. & Swenning, J. Sedimentological and geomorphological effects of reservoir flushing: the Cachí Reservoir, Costa Rica, 1996. Geogr. Ann. Ser. B 81, 391–407 (1999).f 8. Grant, G. E., Schmidt, J. C. & Lewis, S. L. A geological framework for interpreting downstream effects of dams on rivers. In Water Science and Application (eds O’Connor, J. E. & Grant, G. E.) 203–219 (American Geophysical Union, 2003). https://doi.org/10 1029/007WS13. 1029/007WS13. 19 Petts G E & Gurnell A M Dams and geomorphology Research progress and future directions Geomorphology 71 27 47 (2005) 19. Petts, G. E. & Gurnell, A. M. Dams and geomorphology: Research progress and future directions. Geomorphology 71, 27–47 (2 20. Newcombe, C. & MacDonald, D. D. Effects of suspended sediments on aquatic ecosystems. J. N. Am. J. Fish. Manag. 11, 7 (1991). ( ) 21. Gilles, B. & Le Bail, P.-Y. Does light have an influence on fish growth?. Aquaculture 177, 129–152 (1999). li 2. Carolli, M., Bruno, M. C., Siviglia, A. & Maiolini, B. Responses of benthic invertebrates to abrupt changes of temperature in flume simulations. River Res. Appl. 28, 678–691 (2012). 3. Bennel, D. H., Connor, W. P. & Eaton, C. A. Substrate composition and emergence success of fall Chinook salmon in the Snake river. Northwest Sci. 77, 93–99 (2003).ii 4. Jensen, D. W., Steel, E. A., Fullerton, A. H. & Pess, G. R. Impact of fine sediment on egg-to-fry survival of Pacific salmon: A meta analysis of published studies. Rev. Fish. Sci. 17, 348–359 (2009). y p , ( ) 5. Bjornn, T. C. & Reiser, D. W. Habitat requirements of salmonids in streams. Am. Fish. Soc. Spec. Publ. 19, 83–138 (1991). 26. ASCE, N. Sediment and aquatic habitat in river systems. J. Hydraul. Eng. 118, 669–687 (1992). h k k h b f l l d b 27. Louhi, P., Mäki-Petäys, A. & Erkinaro, J. Spawning habitat of Atlantic salmon and brown trout: General criteria and intragravel factors. River Res. Appl. 24, 330–339 (2008). 8. Baxter, C. V. & Hauer, F. R. Geomorphology, hyporheic exchange, and selection of spawning habitat by bull trout (Salvelinus confluentus). Can. J. Fish. Aquat. Sci. 57, 1470–1481 (2000).il fl q 9. Peviani, M., Saccardo, I., Crosato, A. & Gentili, G. Natural and artificial floods connected with river habitat. in Ecohydraulics 2000 Proceedingsof the 2nd International Symposium on Habitat Hydraulics, IAHR- Que´bec,Canada, vol. www.nature.com/scientificreports/ B 175–186 (1996).fli g f y p y 0. Crosa, G., Castelli, E., Gentili, G. & Espa, P. Effects of suspended sediments from reservoir flushing on fish and macroinvertebrates in an alpine stream. Aquat. Sci. 72, 85 (2009).l q 1. Espa, P., Crosa, G., Gentili, G., Quadroni, S. & Petts, G. Downstream ecological impacts of controlled sediment flushing in an Alpine valley river: A case study. River Res. Appl. 31, 931–942 (2014).ht p y y pp ( ) 32. Lee, A. Modelling Salmon Spawning Habitat Response to Dam Removal. (MSc Thesis, IHE Delft, the Netherlands, 2017). ht 33. van Oorschot, M. et al. Impact of dam operations on the habitat suitability of Plecoglossus altivelis downstream of the Fun dam, Japan. In River Flow 2020 (eds Uijttewaal et al.) (2020 Taylor & Francis Group, CRC Press, 2020). b d d d ff f f d d f 4. Newcombe, C. P. Suspended sediments in acquatic ecosystem: III effects as a function of concentration and duration of exposure (1994).i 5. Newcombe, C. P. & Jensen, J. O. T. Channel suspended sediment and fisheries: A synthesis for quantitative assessment of risk and impact. N. Am. J. Fish. Manag. 16, 693–727 (1996).l p g 6. Hubert, W. A., Helzner, R. S., Lee, L. A. & Nelson, P. C. Habitat suitability index models and instream flow suitability curves: Arctic grayling riverine populations. Western Energy and Land Use Team, Division of Biological Services, Research and Development h d ldl f S S f l l ( ) ( ) 36. Hubert, W. A., Helzner, R. S., Lee, L. A. & Nelson, P. C. Habitat suitability index models and instream flow suitability curves: Arctic grayling riverine populations. Western Energy and Land Use Team, Division of Biological Services, Research and Development, Fish and Wildlife Service, US Department of Interior, Biological report 82 (10.110) (1985).l 36. Hubert, W. A., Helzner, R. S., Lee, L. A. & Nelson, P. C. Habitat suitability index models and instream flow suitability curves: Arctic grayling riverine populations. Western Energy and Land Use Team, Division of Biological Services, Research and Development, Fish and Wildlife Service US Department of Interior Biological report 82 (10 110) (1985) 37. Raleigh, R. F., Miller, W. J. & Nelson, P. C. Habitat suitability index models and instream flow suitability curves: Chinook salmon. Fish and Wildlife Service, US Department of the Interior, Biological Report 82(10.122) www.nwrc.usgs.gov/wdb/pub/hsi/hsi-122. pdf (1986).ll p 38. www.nature.com/scientificreports/ Fisher, S., Gray, L., Grimm, N. & Busch, D. E. Temporal succession in a desert stream ecosystem following flash flooding. Ecol. Monogr. https://doi.org/10.2307/2937346 (1982).l 9. Lapointe, M., Eaton, B., Driscoll, S. & Latulippe, C. Modelling the probability of salmonid egg pocket scour due to floods. Can. J Fish. Aquat. Sci. 57, 11 (2000).hfl 40. Baldwin, D. & Mitchell, A. M. The effects of drying and re-flooding on the sediment and soil nutrient dynamics of lowland river– floodplain systems: A synthesis. River Res. Appl. 16, 457–467 (2000).hfl aldwin, D. & Mitchell, A. M. The effects of drying and re-flooding on the sediment and soil nutrient dynamics of lowland river–l oodplain systems: A synthesis. River Res. Appl. 16, 457–467 (2000).hfl owalski, D. The effects of stream flow on the trout populations of th hfl 42. Konard, C. P. Effects of urban development on floods. U.S. Geological Survey—Water Resources Fact Sheet 076-03 https://p usgs.gov/fs/fs07603/ (2016).hl g g 3. Miller, J. D. & Hutchins, M. The impacts of urbanisation and climate change on urban flooding and urban water quality: A review of the evidence concerning the United Kingdom. J. Hydrol. Reg. Stud. 12, 345–362 (2017).fl 44. Poff, N. L. & Ward, J. V. Implications of streamflow variability and predictability for lotic community structure: A regional ana of streamflow patterns. Can. J. Fish. Aquat. Sci. 46, 1805–1818 (1989).fli l 45. George, S. D., Baldigo, B. P., Smith, A. J. & Robinson, G. R. Effects of extreme floods on trout populations and fish communities in a Catskill Mountain river. Freshw. Biol. 60, 2511–2522 (2015).fli 6. Carlson, A. K., Fincel, M. J., Longhenry, C. M. & Graeb, B. D. Effects of historic flooding on fishes and aquatic habitats in a Mis- souri river delta. J. Freshw. Ecol. 31, 271–288 (2016).hi 47. Ríos-Pulgarín, M. I., Barletta, M. & Mancera-Rodríguez, N. J. The role of the hydrological cycle on the distribution patterns of fish assemblages in an Andean stream. J. Fish Biol. 89, 102–130 (2016). g 48. United States Federal energy regulatory Commission (FERC). Application for Surrender of License, Bull Run Hydropower Pr Environmental Impact Statement. https://catalog.hathitrust.org/Record/100940309 (2003). p p g g ( ) 49. Squier Associates. Sandy river sediment study, Bull Run Hydroelectric Project. (2000). es. Sandy river sediment study, Bull Run Hydroelectric Project. (2 y y y j 50. Taylor, B. Salmon and Steelhead Runs and Related Events of the Clackamas River Basin–A Historical Perspective. References D., Pratt, T. C., Randall, R. G., Scruton, D. A. & Smokorowski, K. E. Validation of the flow management pathway: Effects of altered flow on fish habitat and fishes downstream from a hydropower dam. Can. Tech. Rep. Fish. Aquat. Sci. 2784, 111 (2008). g p g g 14. Clarke, K. D., Pratt, T. C., Randall, R. G., Scruton, D. A. & Smokorowski, K. E. Validation of the flow management pathway: Effects f l d fl fi h h bi d fi h d f h d d C T h R Fi h A S i ( ) 14. Clarke, K. D., Pratt, T. C., Randall, R. G., Scruton, D. A. & Smokorowski, K. E. Validation of the flow managemen of altered flow on fish habitat and fishes downstream from a hydropower dam. Can. Tech. Rep. Fish. Aquat. Sci. 4. Clarke, K. D., Pratt, T. C., Randall, R. G., Scruton, D. A. & Smokorowski, K. E. Validation of the flow management pathway: Effects of altered flow on fish habitat and fishes downstream from a hydropower dam. Can. Tech. Rep. Fish. Aquat. Sci. 2784, 111 (2008) lii y 15. Poff, N. L. & Zimmerman, J. K. H. Ecological responses to altered flow regimes: A literature review to inform the science management of environmental flows. Freshw. Biol. 55, 194–205 (2010).h 15. Poff, N. L. & Zimmerman, J. K. H. Ecological responses to altered flow regimes: A literature review to inform the science and management of environmental flows. Freshw. Biol. 55, 194–205 (2010). k Th f A ’ d d h l h d h b l A A gl 6. Juracek, K. E. The aging of America’s reservoirs: In-reservoir and downstream physical changes and habitat implications. JAWRA J. Am. Water Resour. Assoc. 51, 168–184 (2015). https://doi.org/10.1038/s41598-022-19294-2 Scientific Reports \| (2022) 12:15354 \| www.nature.com/scientificreports/ The shifting habitat mosaic of river ecosystems. Int. Ver. Theor. Angew. Limnol. Verh. 29, 123–136 (2005). q 1. Stanford, J. A., Lorang, M. S. & Hauer, F. R. The shifting habitat mosaic of river ecosystems. Int. Ver. Theor. Angew. Limnol. Verh 29, 123–136 (2005). www.nature.com/scientificreports/ E., Keefer, M. L. & Caudill, C. C. Elevated stream temperature, origin, and individual size influence Ch prespawn mortality across the Columbia River Basin. Fish. Res. 237, 105874 (2021). p p y 62. Beechie, T. J. et al. Process-based principles for restoring river ecosystems. Bioscience 60, 209–222 (2010).fl 3. Crosato, A. & Saleh, M. S. Numerical study on the effects of floodplain vegetation on river planform style. Earth Surf. Process Landf. 36, 711–720 (2011). f 64. Schuurman, F., Marra, W. A. & Kleinhans, M. G. Physics-based modeling of large braided sand-bed rivers: Bar pattern formation, dynamics, and sensitivity. J. Geophys. Res. Earth Surf. 118, 2509–2527 (2013). y y p y f 5. Singh, U., Crosato, A., Giri, S. & Hicks, M. Sediment heterogeneity and mobility in the morphodynamic modelling of gravel-bed braided rivers. Adv. Water Resour. 104, 127–144 (2017).ht , ( ) 66. van ledden, M. Sand-Mud Segregation in Estuaries and Tidal Basins. (PhD Thesis, University of Technology Delft, 2003). h 67. Kandiah, A. Fundamental Aspects of Surface Erosion of Cohesive Soils. (University of California, Davis, 1974). 68. Partheniades, E. Cohesive Sediments in Open Channels: Erosion, Transport and Deposition (Butterworth-Heinemann, 2009) 69. Jiang, J. An Examination of Estuarine Lutocline Dynamics. (PhD Thesis, University of Florida, USA, 1999). 70. Exner, F. M. Uber die wechselwirkung zwischen wasser und geschiebe in flussen (about the interaction between water and bedload in rivers). Akad. Wiss. Wien Math. Naturwiss. Kl. 134, 165–204 (1925). g f yh y 70. Exner, F. M. Uber die wechselwirkung zwischen wasser und geschiebe in flussen (about the interaction between water and bed in rivers) Akad Wiss Wien Math Naturwiss Kl 134 165 204 (1925) h 70. Exner, F. M. Uber die wechselwirkung zwischen wasser und geschiebe in flussen (about the interaction betw in rivers). Akad. Wiss. Wien Math. Naturwiss. Kl. 134, 165–204 (1925). l . Akad. Wiss. Wien Math. Naturwiss. Kl. 134, 165–204 (1925). 71. Ikeda, S. Incipient motion of sand particles on side slopes. J. Hydraul. Div. 108, 95–114 (1982). , p p p J y , ( ) 72. Bagnold, R. A. An approach to the sediment transport problem from general physics. Physiographic and hydraulic studies of rivers in US Geological Survey Professional Paper, vol. 422 I, 231–291 (1966). g y f p 3. Stillwater Sciences. Numerical modeling of sediment transport in the Sandy river, Oregon following removal of Marmot dam. (2000) 73. Stillwater Sciences. www.nature.com/scientificreports/ Formulation of habitat suitability models for stream fish guilds: Do the standard methods work?. Trans. Am. Fish. Soc. 130, 217–235 (2001). 86. Moir, H. J., Gibbins, C. N., Soulsby, C. & Youngson, A. F. PHABSIM modelling of Atlantic salmon spawning habitat in an upland stream: Testing the influence of habitat suitability indices on model output. River Res. Appl. 21, 1021–1034 (2005). 6. Moir, H. J., Gibbins, C. N., Soulsby, C. & Youngson, A. F. PHABSIM modelling of Atlantic salmon spawning habitat in an upland stream: Testing the influence of habitat suitability indices on model output. River Res. Appl. 21, 1021–1034 (2005). gl y 87. Hauer, C. et al. State of the art, shortcomings and future challenges for a sustainable sediment management in hydropow review. Renew. Sustain. Energy Rev. 98, 40–55 (2018).hi l 7. Hauer, C. et al. State of the art, shortcomings and future challenges for a sustainable sediment management in hydropower: A review. Renew. Sustain. Energy Rev. 98, 40–55 (2018).hi gy 88. Koizumi, I., Kanazawa, Y. & Tanaka, Y. The fishermen were right: Experimental evidence for tributary refuge hypothesis during floods. Zool. Sci. 30, 375–379 (2013).fl gy 88. Koizumi, I., Kanazawa, Y. & Tanaka, Y. The fishermen were right: Experimental evidence for tributary refuge hypothesis during floods. Zool. Sci. 30, 375–379 (2013).fl l 89. Quadroni, S. et al. Effects of sediment flushing from a small Alpine reservoir on downstream aquatic fauna. Ecohydrolo 1276–1288 (2016).il 1276–1288 (2016). 90. Bond, M. H., Nodine, T. G., Beechie, T. J. & Zabel, R. W. Estimating the benefits of widespread floodplain reconnection for Colum- bi i Chi k l C J Fi h A t S i 76 1212 1226 (2019) 1276 1288 (2016). 90. Bond, M. H., Nodine, T. G., Beechie, T. J. & Zabel, R. W. Estimating the benefits of widespread floodplain reconnection for Colum- bia river Chinook salmon. Can. J. Fish. Aquat. Sci. 76, 1212–1226 (2019). 0. Bond, M. H., Nodine, T. G., Beechie, T. J. & Zabel, R. W. Estimating the benefits of widespread floodplain reconnection for Colum bia river Chinook salmon. Can. J. Fish. Aquat. Sci. 76, 1212–1226 (2019).hth 90. Bond, M. H., Nodine, T. G., Beechie, T. J. & Zabel, R. W. Estimating the benefits of widespread floodplain reconnection for Co bia river Chinook salmon. Can. J. Fish. Aquat. Sci. 76, 1212–1226 (2019).hth J q ( ) 91. Stanford, J. A., Lorang, M. S. & Hauer, F. R. Author contributions Manuscript writing by M.P. and A.C. Initial model development by S.D and A.L. Model extension, model simula- ions and result analysis by M.P. Manuscript revision by M.P, A.C., S.D., A.L, A.O and M.F. www.nature.com/scientificreports/ Portland Ge Electric Company, 64 https://www.eaglecreekfriends.org/links-references (1999). Electric Company, 64 https://www.eaglecreekfriends.org/links references (1999). 1. Trimble, D. E. Geology of Portland, Oregon, and Adjacent Areas. Bulletin U.S. G.P.O. https://pubs.er.usgs.gov/publication/b1119 htt //d i /10 3133/b1119 (1963) 1. Trimble, D. E. Geology of Portland, Oregon, and Adjacent Areas. Bulletin U.S. G.P.O. https://pubs.er.usgs.gov/publication/b1119 https://doi.org/10.3133/b1119. (1963). g 52. Sandy River basin Working Group. Sandy River basin aquatic habitat restoration strategy: An anchor habitat-based prioritization of restoration opportunities Oregon Trout. Portland, Oregon. Preprint at https://www.fs.usda.gov/Internet/FSE_DOCUMENTS/ stelprdb5325660.pdf (2007). p p 53. Lee, A., Crosato, A., Omer, A. Y. A. & Bregoli, F. Applying a two-dimensional morphodynamic model to assess impacts to Chi salmon spawning habitat from dam removal. in AGU Fall meeting (2017).i g g 54. Healey, M. C. Life history of Chinook salmon (Oncorhynchus tshawytscha). Pacific Salmon Life Histories 311–394 (1991). d ll f f l h k l l f h h h l 55. Bourret, S. L., Caudill, C. C. & Keefer, M. L. Diversity of juvenile Chinook salmon life history pathways. Rev. Fish. Biol. Fish. 26, 375–403 (2016). 56. Alderdice, D. & Velsen, F. Relation between temperature and incubation time for eggs of Chinook salmon (Oncorhynchus tshaw- ytscha). J. Fish. Res. Board Can. 35, 69–75 (1978). y 7. Seattle Aquarium. Redd alert: Our Chinook salmon are hatching! \|. Seattle Aquarium https://www.seattleaquarium.org/blog/ redd-alert-our-chinook-salmon-are-hatching (2015). g 58. Whitman, L., Cannon, B. & Hart, S. Spring Chinook salmon in the Willamette and Sandy rivers: Sandy river basin Spring Chinook salmon spawning surveys. Oregon Department of Fish and Wildlife 4034 Fairview Industrial Drive SE Salem, Oregon 97302, 30 https://doi.org/10.1038/s41598-022-19294-2 Scientific Reports \| (2022) 12:15354 \| www.nature.com/scientificreports/ https://odfw.forestry.oregonstate.edu/willamettesalmonidrme/sites/default/files/2016_sandy_basin_spring_chinook_spawning_ survey.pdf (2016). y p ( ) 9. Cramer, S. P. Fish and habitat surveys of the lower Sandy and Bull Run rivers. Report of SP Cramer&Associates, Inc. to Portland General Electric and Portland Water bureau, Portland, Oregon (1998). g 60. Westley, P. A. Documentation of en route mortality of summer chum salmon in the Koyukuk river, Alaska and its potential linkage to the heatwave of 2019. Ecol. Evol. 10, 10296–10304 (2020).l g . Westley, P. A. Documentation of en route mortality of summer y y to the heatwave of 2019. Ecol. Evol. 10, 10296–10304 (2020). 61. Bowerman, T. E., Keefer, M. L. & Caudill, C. C. Elevated stream temperature, origin, and individual size influ prespawn mortality across the Columbia River Basin. Fish. Res. 237, 105874 (2021). 61. Bowerman, T. www.nature.com/scientificreports/ Numerical modeling of sediment transport in the Sandy river, Oregon following removal of Marmot dam. (2000). 74. Ashida, K. & Michiue, M. Study on hydraulic resistance and bed transport rate in alluvial stream. Proc. Jpn. Soc. Civ. Eng. 201, 59–69 (1972).ht ( ) 5. Panthi, M. Generation and Fate of Fine Sediment from Dam Flushing. (MSc Thesis, IHE Delft, Institute for Water Education, the Netherlands, 2020). ) 76. Meyer-Peter, E. & Müller, R. Formulas for bed-load transport. in IAHSR 2nd Meeting, Stockholm, Appendix 2 (IAHR, 1948). . Meyer-Peter, E. & Müller, R. Formulas for bed-load transport. in 7. Podolak, C. & Pittman, S. Marmot Dam Removal Geomorphic Monitoring & Modeling Project: Final Report. Sandy river basin watershed council. (2011).h 8. Keith, M. K. Reservoir Evolution Following the Removal of Marmot Dam on the Sandy River, Oregon. (MSc Thesis, Portland State University, Oregon, USA, 2012).f 9. Redding, J. M., Schreck, C. B. & Everest, F. H. Physiological effects on coho salmon and steelhead of exposure to suspended solids Trans. Am. Fish. Soc. 116, 737–744 (1987). 0. Lisle, T. E. Sediment transport and resulting deposition in spawning gravels, north coastal California. Water Resour. Res. 25 1303–1319 (1989).l ( ) 1. Pitlick, J. & Wilcock, P. Relations between streamflow, sediment transport, and aquatic habitat in regulated rivers. In Geomorphi Processes and Riverine Habitat (eds Dorava, J. M. et al.) 185–198 (American Geophysical Union, 2001). y 82. Toupin, L. Freshwater Habitats: Life in Freshwater Ecosystems. (Franklin Watts, Watts Library, 2005). . Toupin, L. Freshwater Habitats: Life in Freshwater Ecosystems. (F 3. Bovee, K. D. A guide to Stream Habitat Analysis Using the Instream Flow Incremental Methodology. IFIP No. 12. FWS/OBS https:// pubs.er.usgs.gov/publication/fwsobs82_26 (1982). p g g p . Elith, J. et al. Novel methods improve prediction of species’ dist 4. Elith, J. et al. Novel methods improve prediction of species’ distributions from occurrence data. Ecography 29, 129–151 (2006). 5. Vadas, R. L. & Orth, D. J. Formulation of habitat suitability models for stream fish guilds: Do the standard methods work?. Trans Am. Fish. Soc. 130, 217–235 (2001). , J p p p g p y , ( ) 5. Vadas, R. L. & Orth, D. J. Formulation of habitat suitability models for stream fish guilds: Do the standard methods work?. Trans Am. Fish. Soc. 130, 217–235 (2001). J p p p g p y ( ) 85. Vadas, R. L. & Orth, D. J. Competing interests h p g The authors declare no competing interests. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Additional informationh upplementary Information The online version contains supplementary material available at https://doi.org/ 0.1038/s41598-022-19294-2. Supplementary Information The online version contains supplementary material available at https://doi.org/ 10.1038/s41598-022-19294-2. Correspondence and requests for materials should be addressed to M.P. Correspondence and requests for materials should be addressed to M.P. Correspondence and requests for materials should be addressed to M.P. Reprints and permissions information is available at www.nature.com/reprints. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. https://doi.org/10.1038/s41598-022-19294-2 Scientific Reports \| (2022) 12:15354 \| www.nature.com/scientificreports/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2022 https://doi.org/10.1038/s41598-022-19294-2 Scientific Reports \| (2022) 12:15354 \|
https://openalex.org/W2131365485	https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0029588&type=printable	English	null	Baseline Predictors of Sputum Culture Conversion in Pulmonary Tuberculosis: Importance of Cavities, Smoking, Time to Detection and W-Beijing Genotype	PloS one	2,012	cc-by	6,181	Marianne E. Visser1, Michael C. Stead2,3, Gerhard Walzl2, Rob Warren2, Michael Schomaker4, Harleen M. S. Grewal5, Elizabeth C. Swart6, Gary Maartens7 1 School of Public Health, University of the Western Cape, Cape Town, South Africa, 2 MRC Centre for Molecular and Cellular Biology, University of Stellenbosch, Cape Town, South Africa, 3 Division of Medical Microbiology, Institute of Infectious Diseases and Molecular Medicine, University of Cape Town, Cape Town, South Africa, 4 School of Public Health and Family Medicine, Centre for Infectious Disease Epidemiology and Research, University of Cape Town, Cape Town, South Africa, 5 Section of Microbiology and Immunology, The Gade Institute, University of Bergen, Haukeland University Hospital, Bergen, Norway, 6 Division of Dietetics, University of the Western Cape, Cape Town, South Africa, 7 Division of Clinical Pharmacology, Department of Medicine, University of Cape Town, Cape Town, South Africa Abstract This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits tion, and reproduction in any medium, provided the original author and source are credited. Copyright: 2012 Visser et al. This is an open-access article distributed under the terms of the Creative Commons Attributi unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Copyright: 2012 Visser et al. This is an open-access article distributed under the t unrestricted use, distribution, and reproduction in any medium, provided the original aut Funding: This work was supported by the Norwegian Programme for Development, Research and Higher Education (NUFUPRO-2007/10183) www.siu.no, the Research Council of Norway (RCN)(183694/S50) www.forskningsradet.no, and the National Research Foundation (South Africa)/RCN (180353/S50) www.nrf.ac.za. GM was supported in part by Grant Number U2RTW007370 from the Fogarty International Center. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. E-mail: marianne.visser@gmail.com Baseline Predictors of Sputum Culture Conversion in Pulmonary Tuberculosis: Importance of Cavities, Smoking, Time to Detection and W-Beijing Genotype Marianne E. Visser1, Michael C. Stead2,3, Gerhard Walzl2, Rob Warren2, Michael Schomaker4, Harleen M. S. Grewal5, Elizabeth C. Swart6, Gary Maartens7 PLoS ONE \| www.plosone.org Citation: Visser ME, Stead MC, Walzl G, Warren R, Schomaker M, et al. (2012) Baseline Predictors of Sputum Culture Conversion in Pulmonary Tuberculosis: Importance of Cavities, Smoking, Time to Detection and W-Beijing Genotype. PLoS ONE 7(1): e29588. doi:10.1371/journal.pone.0029588 Abstract Background: Time to detection (TTD) on automated liquid mycobacterial cultures is an emerging biomarker of tuberculosis outcomes. The M. tuberculosis W-Beijing genotype is spreading globally, indicating a selective advantage. There is a paucity of data on the association between baseline TTD and W-Beijing genotype and tuberculosis outcomes. Aim: To assess baseline predictors of failure of sputum culture conversion, within the first 2 months of antitubercular therapy, in participants with pulmonary tuberculosis. Design: Between May 2005 and August 2008 we conducted a prospective cohort study of time to sputum culture conversion in ambulatory participants with first episodes of smear and culture positive pulmonary tuberculosis attending two primary care clinics in Cape Town, South Africa. Rifampicin resistance (diagnosed on phenotypic susceptibility testing) was an exclusion criterion. Sputum was collected weekly for 8 weeks for mycobacterial culture on liquid media (BACTEC MGIT 960). Due to missing data, multiple imputation was performed. Time to sputum culture conversion was analysed using a Cox-proportional hazards model. Bayesian model averaging determined the posterior effect probability for each variable. Results: 113 participants were enrolled (30.1% female, 10.5% HIV-infected, 44.2% W-Beijing genotype, and 89% cavities). On Kaplan Meier analysis 50.4% of participants underwent sputum culture conversion by 8 weeks. The following baseline factors were associated with slower sputum culture conversion: TTD (adjusted hazard ratio (aHR) = 1.11, 95% CI 1.02; 1.2), lung cavities (aHR = 0.13, 95% CI 0.02; 0.95), ever smoking (aHR = 0.32, 95% CI 0.1; 1.02) and the W-Beijing genotype (aHR = 0.51, 95% CI 0.25; 1.07). On Bayesian model averaging, posterior probability effects were strong for TTD, lung cavitation and smoking and moderate for W-Beijing genotype. Conclusion: We found that baseline TTD, smoking, cavities and W-Beijing genotype were associated with delayed 2 month sputum culture. Larger studies are needed to confirm the relationship between the W-Beijing genotype and sputum culture conversion. Citation: Visser ME, Stead MC, Walzl G, Warren R, Schomaker M, et al. (2012) Baseline Predictors of Sputum Culture Conversion in Pulmonary Tuberculosis: Importance of Cavities, Smoking, Time to Detection and W-Beijing Genotype. PLoS ONE 7(1): e29588. doi:10.1371/journal.pone.0029588 Editor: Pere-Joan Cardona, Fundacio´ Institut d’Investigacio´ en Cie`ncies de la Salut Germans Trias i Pujol. Universitat Auto`noma Received August 1, 2011; Accepted November 30, 2011; Published January 4, 2012 ser et al. January 2012 \| Volume 7 \| Issue 1 \| e29588 Predictors of Culture Conversion in Tuberculosis Predictors of Culture Conversion in Tuberculosis media using the BACTEC MGIT 960 system (Becton Dickinson, Sparts, Maryland). Positive cultures were confirmed as M. tuberculosis complex using an in-house PCR assay [21]. The date of culture conversion was taken as the date of the first negative culture provided there were no subsequent positive cultures. Participants whose first negative culture occurred at week 8, were regarded as converters. Routine phenotypic drug susceptibility testing for isoniazid and rifampicin of isolates was carried out using the MGIT 960 system from January 2008; isolates cultured prior to that date were tested retrospectively, but some isolates had lost viability. media using the BACTEC MGIT 960 system (Becton Dickinson, Sparts, Maryland). Positive cultures were confirmed as M. tuberculosis complex using an in-house PCR assay [21]. The date of culture conversion was taken as the date of the first negative culture provided there were no subsequent positive cultures. Participants whose first negative culture occurred at week 8, were regarded as converters. Routine phenotypic drug susceptibility testing for isoniazid and rifampicin of isolates was carried out using the MGIT 960 system from January 2008; isolates cultured prior to that date were tested retrospectively, but some isolates had lost viability. There is limited evidence on the influence of M. tuberculosis genotype on the clinical outcome of tuberculosis. It has been suggested that the W-Beijing genotype has a selective advantage over other M. tuberculosis strains, since it is emerging worldwide and is linked to multidrug resistance in many areas [11,12]. Data is conflicting regarding the role of the W-Beijing genotype on the risk of treatment failure or relapse [13–16]. We assessed baseline predictors, including W-Beijing genotype, of failure of sputum culture conversion within the first 2 months in participants with first episodes of smear positive pulmonary tuberculosis attending a primary care centre in Cape Town, South Africa. During 2009, spoligotyping was performed on the majority of serial isolates collected during the trial period [22]. Strain-specific PCR assays were performed to detect M. tuberculosis infections by W-Beijing and non-W-Beijing isolates [23]. In order to optimize the baseline data, we captured the strain type for all the available baseline isolates, as well as those within the first 2 weeks of treatment, in order to categorize participants as either W-Beijing or non-W-Beijing strain types. Study population and setting y p p g Participants with first episodes of sputum smear positive pulmonary tuberculosis were enrolled in a randomized controlled trial of a micronutrient intervention (vitamin A and zinc), which has been reported elsewhere [17]. In brief, adults who attended two primary care tuberculosis clinics in Delft, Cape Town, South Africa between May 2005 and August 2008 were recruited. Rates of poverty and unemployment are high in Delft and the tuberculosis incidence rate increased from 496 to 737 cases per 100 000 during the study period. Study participants received standard anti-tubercular therapy for 5 days a week consisting of combination tablets contributing 600 mg rifampicin, 300 mg isoniazid, 1.6 g pyrazinamide and 1.1 g ethambutol for partici- pants weighing 38 to 55 kg (Rifafour, Aventis Pharma Pty Ltd, Johannesburg). Doses were adjusted for participants weighing less than 38 kg or more than 55 kg. All participants were on directly observed therapy at the clinic or in the community. Research staff collected early morning sputum specimens weekly from partici- pants for 8 weeks. The micronutrient intervention had no effect on . time to sputum smear or culture conversion up to 8 weeks on Kaplan Meier analysis (P = 0.15 and P = 0.38 respectively; log rank test) The current analysis is restricted to trial participants with a positive baseline sputum culture and with data on susceptibility to rifampicin and isoniazid. Participants with rifampicin resistance during the 8 week follow-up period, were excluded. Predictors of Culture Conversion in Tuberculosis Ten participants who had two strain types, one of which was W-Beijing, were categorized as having the W-Beijing strain for the analysis. Statistical analysis Data was captured in duplicate in Microsoft Office Excel 2003 and validated with SAS 9.2 software. To account for missing data, multiple imputation was conducted with the Amelia II software package [24],a reliable imputation procedure implemented in the statistical software package R,and five imputed data sets were generated. This number of data sets is generally considered adequate [25]. Cox proportional hazard regression modeling with the imputed datasets was used to estimate the hazard ratio of culture conversion for the following pre-determined baseline variables: age, gender, smoking status, alcohol abuse, sputum smear grading, sputum TTD, presence and extent of cavitation, M. tuberculosis genotype, HIV-infection, body mass index, haemoglobin, total lymphocyte count, serum C-Reactive Protein (CRP) (at baseline and change in concentrations within the first 2 weeks) and albumin. Hazard ratios are reported together with 95% confidence intervals based on Rubin’s rules [25]. The proportional hazards assumption was verified by testing interaction effects of analysis time with baseline variables (a = 0.05) and graphically via log-log-plots [26]. To confirm the stability of our results we conducted a sensitivity analysis using Bayesian model averaging to each imputed data set (M = 20), and the results were combined by Rubins’ rules [25,27]. Model averaging is an alternative to model selection and combines estimates from different models, which contain different sets of variables, with the aim of achieving stable estimates that incorporate modelling uncertainty in addition to sampling uncertainty: Models that have a higher posterior probability to be correct receive a higher weight [26]. The combined weighted estimates can then be interpreted as a stabilized hazard ratio of culture conversion, which, together with adjusted standard errors, reflects the uncertainty in the modeling process. The posterior effect probabilities presented in our analysis state the probability that the hazard ratio in the Cox regression model for a variable is not one, thus providing a summary measure of how likely it is that a baseline variable has an effect and ranks the importance of the baseline variables. Study procedures and selection of isolates In keeping with national guidelines at the time, none of our HIV-infected participants received anti-retroviral therapy, but all received co-trimoxazole prophylaxis. All participants were screened for misuse of alcohol with the Cut down, Annoyance, Guilt and Eye-opener (CAGE) questionnaire [18]. Smoking status was categorized as either never or ever smoked. Body weight and height of participants was determined. The extent and size of lung cavities at baseline was assessed independently by two pulmonol- ogists experienced in the use of the Chest Radiograph Reading and Recording System (CRRS)- and disagreements on radio- graphic reading were resolved by consensus [19]. Prior to anti-tubercular therapy, trial participants supplied one spontaneous sputum specimen and thereafter one early morning unassisted sputum specimen every week, up to 8 weeks. Sputum specimens were examined by means of fluorescent microscopy (Auramine stain) and graded according to international standards [20]. All specimens were decontaminated and cultured on liquid Ethics statement The trial protocol was approved by the University of Cape Town Ethics and Research Committee and written consent was obtained from all participants. PLoS ONE \| www.plosone.org Introduction is a recognized surrogate biomarker of cure [1,2]. Time to detection (TTD) on automated liquid mycobacterial cultures reflects mycobacterial load more accurately than sputum smear or culture grading, and has been shown to predict month 2 culture conversion, tuberculosis recurrence and relapse [3,4]. Tuberculosis cure after anti-tubercular therapy is best measured by bacteriological relapse within 2 years after completion of treatment. Because relapse is uncommon after short course therapy large sample sizes are required to detect differences in relapse rate, which, together with extended follow up periods, makes clinical trials using relapse as the primary endpoint very expensive. Sputum culture conversion after 2 months of treatment Important factors associated with smear and culture non- conversion after 2 months include increasing age, chest radio- graphic features (cavitation and extent of disease), and higher sputum smear and culture grading at diagnosis [5–10]. PLoS ONE \| www.plosone.org January 2012 \| Volume 7 \| Issue 1 \| e29588 January 2012 \| Volume 7 \| Issue 1 \| e29588 1 Predictors of Culture Conversion in Tuberculosis Predictors of Culture Conversion in Tuberculosis fulfilled the inclusion criteria for the current study (Fig. 1). Four of the 113 sputum isolates were mono-resistant to isoniazid. Complete data on all the variables investigated in the current analysis was available in 72 patients. The baseline characteristics of these participants before and after multiple imputation, is shown in Table 1. There were 11 HIV-infected participants (5 males, 6 females) and HIV-status was not determined in 9 who were lost to follow-up before the test could be done. All participants underwent chest radiograph examination at baseline, but 22 could not be located at the clinic site either due to mis-filing or loss (e.g. chest radiographs sent with participants who were referred to hospitals). Strain data was unavailable for seven participants whose sputum cultures were discarded before spoligotyping could be performed. was not significant in univariate analysis, the confidence interval was suggestive of a delayed effect on conversion, even after adjustment for co-variates. INH mono-resistance was not associated with conversion in univariate analysis (HR = 0.59; 95%CI, 0.082; 4.28) and was not explored further due to the limited number of participants. The estimated hazard ratios of the variables remained similar after model averaging (estimates not shown) confirming the stability of our main results. After Bayesian model averaging ever smokers, the presence of lung cavities and TTD had a posterior effect probability of greater than 80%, thus having a rather strong effect on sputum conversion, even after taking model uncertainty into account. A moderate effect of W-Beijing genotype was also present (Table 2). Participants had a high baseline bacillary burden, as reflected by the large proportion with a high sputum smear grading and a short median duration of TTD of 7 days (Table 1). Lung cavities were common and there was a high prevalence of smoking and alcohol abuse among participants. Spoligotyping showed that almost half of our participants had the W-Beijing strain type. Discussion We found that a shorter baseline TTD, tobacco smoking, the presence of lung cavities and W-Beijing genotype were associated with delayed sputum culture conversion within the first 2 months in adults with smear-positive pulmonary tuberculosis. Bayesian model averaging showed that TTD, lung cavitation and smoking had strong effects and the W-Beijing genotype a moderate effect. The association we found between W-Beijing and sputum culture conversion has only been reported previously among a small number of participants, who participated in a multicentre trial [28]. Our finding that a shorter TTD is associated with delayed culture conversion adds to the very limited data on this potential surrogate marker [3,4]. An increase of 3 days in TTD at baseline was associated with a 40% increase in the likelihood of conversion. Cavitation is a well- known factor associated with delayed sputum culture conversion [3,5] and relapse [29]. By week 6, nine percent of participants were lost-to follow-up and by week 8, this increased to 15%. Two cultures from participants who had not undergone culture conversion up to week 7, had contaminated cultures at week 8 and were categorized as not having converted. On Kaplan Meier analysis 50.4% of participants had undergone culture conversion by 8 weeks (data not shown). Baseline characteristics A total of 154 participants with new smear-positive pulmonary tuberculosis were enrolled in the micronutrient trial, of whom 113 PLoS ONE \| www.plosone.org January 2012 \| Volume 7 \| Issue 1 \| e29588 2 Predictors of delayed sputum culture conversion Cox-proportional hazard regression analysis showed that TTD, the presence of lung cavities at diagnosis and ever smokers, were all associated with delayed culture conversion in univariate analysis. After adjustment for co-variates, TTD and lung cavities remained significant (Table 2). Although the W-Beijing genotype Figure 1. Participant selection from the parent micronutrient trial for the sputum culture conversion analysis. doi:10.1371/journal.pone.0029588.g001 Figure 1. Participant selection from the parent micronutrient trial for the sputum culture conversion analysis. doi:10.1371/journal.pone.0029588.g001 Figure 1. Participant selection from the parent micronutrient trial for the sputum culture conversion analysis. doi:10.1371/journal.pone.0029588.g001 January 2012 \| Volume 7 \| Issue 1 \| e29588 January 2012 \| Volume 7 \| Issue 1 \| e29588 PLoS ONE \| www.plosone.org PLoS ONE \| www.plosone.org 3 Predictors of Culture Conversion in Tuberculosis Table 1. Baseline characteristics of participants with drug sensitive smear-positive pulmonary tuberculosis.# ble 1. Baseline characteristics of participants with drug sensitive smear-positive pulmonary tuberculosis Characteristic Before multiple imputation After multiple imputation n Age, years 113 30.0 (22.0–43.0) NA Sex 113 Female 34 (30.1%) NA Male 79 (69.9%) HIV-status 104 Positive 11 (10.5%) 10.8% (95% CI:5.0; 16.6) Negative 93 (89.4%) 89.2% (95% CI:83.4; 95.0) Treatment group 113 Micronutrient 56 (49.6%) NA Placebo 57 (50.4%) Sputum smear grade 113 Scanty positive 1 (0.9%) NA 1+ 10 (8.9%) 2+ 13 (11.5%) 3+ 89 (78.8%) Time to culture detection, days 113 7.0 (6.0–11.0) NA M. tuberculosis genotype 106 W-Beijing 50 (44.2%) 46.5% (95% CI:37.2; 55.9) Other 56 (52.8%) 53.5% (95% CI:44.1; 62.3) Chest radiograph 91 Lung cavities 81 (89.0%) 88.1% (95% CI:82.1; 94.2) No. of lung zones affected by cavities 2.0 (1.0–2.0) 1.8 (1.0–2.6) Body Mass Index, (kg/m2) 112 18.9 (17.5–21.2) 18.9 (17.5–21.2) Cigarette smoking 113 Ever 85 (75.2%) NA Never 28 (24.8%) Alcohol misuse 113 43 (38.1%) NA Haemoglobin (g/dl) 111 12.1 (10.75–12.9) 12.03 (10.7–12.9) Albumin (g/l) 111 36.9 (33.2–38.6) 36.86 (33.18–38.6) C-Reactive protein (mg/l) 103 53.6 (36.4–75.8) 53.7 (36.6–75.8) Total Lymphocyte count (6109/l) 113 1.8 (1.35–2.24) NA #Continuous data expressed as median (IQR). NA: Not applicable. Data was complete, therefore not imputed. doi:10.1371/journal.pone.0029588.t001 #Continuous data expressed as median (IQR). NA: Not applicable. Data was complete, therefore not imputed. doi:10.1371/journal.pone.0029588.t001 #Continuous data expressed as median (IQR). NA: Not applicable. Data was complete, therefore not imputed. doi:10.1371/journal.pone.0029588.t001 There is conflicting data on the role of smoking in sputum culture conversion [33]. January 2012 \| Volume 7 \| Issue 1 \| e29588 Predictors of delayed sputum culture conversion Our unadjusted analyses showed that ever smokers had a significant longer time to culture conversion, compared with never-smokers during the first 2 months of treatment (HR = 0.45 (95% CI: 0.25–0.82)). This finding is similar to that of two other clinical trials [3,34]. Retrospective data show that smokers are more likely to have clinical symptoms of tuberculosis such as cough and dyspnoea, as well as cavitary lesions, compared with non-smokers [35,36]. When we adjusted for the presence of cavities and other confounders, the statistical significance of ever smoking was somewhat reduced, but smoking still had a strong effect in the final model, confirmed by model averaging estimates. Our data suggests that W-Beijing strains may be associated with delayed sputum culture conversion within the first 2 months, after adjusting for baseline bacillary load, the presence of lung cavities and the effect of smoking. Since the association between W-Beijing and culture conversion was largely unaffected by the adjustment for bacillary load at diagnosis and cavitatory disease, it suggests that other factors such as the host response to anti-tubercular treatment, may be important. Almost half of our participants had the W-Beijing genotype at baseline or within the first 2 weeks of treatment. The proportion of W-Beijing strains reported among adults and children with tuberculosis in Cape Town has ranged from 17 to 33% [30–32] and data from a paediatric referral hospital shows a marked increase in the prevalence of W-Beijing strains among children with tuberculosis between 2000 and 2003 [32]. One of the properties accounting for the rapid spread of the W-Beijing strains in our area may be operating by prolonging transmission in treated patients. We were not able to demonstrate any significant effect of either baseline CRP concentrations, or change in concentrations after 2 weeks on treatment, on the risk of sputum culture conversion. CRP is a non-specific acute phase reactant and a marker of PLoS ONE \| www.plosone.org 4 January 2012 \| Volume 7 \| Issue 1 \| e29588 Predictors of Culture Conversion in Tuberculosis Table 2. Cox proportional hazards regression analysis of baseline variables associated with sputum culture conversion after multiple imputation. Predictors of delayed sputum culture conversion Unadjusted Hazard ratio (95% CI)* Adjusted Hazard ratio (95%CI)* Posterior Effect Probability# Age 0.99 (0.98–1.02) 0.98 (0.94–1.02) 1.1 Male sex 1.06 (0.57–1.97) 2.38 (0.88–6.25) 14.04 HIV-positive 1.62 (0.67–3.92) 0.65 (0.14–3.12) 0.43 Time to culture detection (days) 1.09 (1.03–1.16) 1.11 (1.02–1.2) 80.08 Sputum smear grading 0.79 (0.55–1.13) 0.75 (0.48–1.18) 10.08 Presence of lung cavities 0.32 (0.12–0.81) 0.13 (0.02–0.95) 87.61 No. of lung zones affected by cavities 0.92 (0.70–1.22) 0.99 (0.54–1.83) 9.52 W-Beijing genotype 0.62 (0.34–1.10) 0.51 (0.25–1.07) 41.24 Ever smoker 0.45 (0.25–0.82) 0.32 (0.1–1.02) 91.52 Alcohol misuse 1.13 (0.64–1.99) 1.67 (0.73–3.79) 10.97 Body Mass Index (kg/m2) 1.03 (0.94–1.13) 1.11 (0.96–1.31) 0.07 Haemoglobin (g/dl) 0.86 (0.71–1.03) 0.77 (0.58–1.03) 7.89 Albumin (g/l) 0.96 (0.90–1.03) 1.01 (0.89–1.15) 1.78 C Reactive Protein (CRP) (mg/l) 1.01 (0.99–1.01) 1.01 (0.99–1.02) 5.51 Change in CRP (baseline to week 2) 1.01(0.99–1.02) 0.99 (0.97–1.02) 0.98 Total Lymphocyte count (6109/l) 0.86 (0.54–1.37) 0.81 (0.42–1.56) 5.42 Likelihood of sputum clearance per unit change in predictor variable. #Posterior effect probability after Bayesian Model averaging; this is the posterior probability that the Hazard in the Cox regression model for a variable is not one, taking model selection uncertainty into account. doi:10.1371/journal.pone.0029588.t002 Table 2. Cox proportional hazards regression analysis of baseline variables associated with sputum culture conversion after multiple imputation. Likelihood of sputum clearance per unit change in predictor variable. #Posterior effect probability after Bayesian Model averaging; this is the posterior probability that the Hazard in the Cox regression model for a variable is not one, taking model selection uncertainty into account. doi:10.1371/journal.pone.0029588.t002 Likelihood of sputum clearance per unit change in predictor variable. #Posterior effect probability after Bayesian Model averaging; this is the posterior probability that the Hazard in the Cox regression model for a variable is not one, taking model selection uncertainty into account. doi:10.1371/journal.pone.0029588.t002 region [4]. Variables associated with sputum conversion may be different in populations with lower rates of smoking and alcohol abuse. macrophage activation in active tuberculosis that is increased at diagnosis in proportion to the extent of disease and decreases during and after treatment [37–39]. Pre-treatment serum CRP concentrations have a high sensitivity for the diagnosis of active tuberculosis among smear-negative tuberculosis suspects [40] and was recently shown to be a risk factor for sputum smear conversion among participants in a micronutrient trial in Nigeria [9]. Delayed sputum culture conversion increases the risk of relapse and prolongs the period of infectiousness. Predictors of delayed sputum culture conversion Therefore our findings about factors delaying culture conversion have several implications for public health. Smoking is associated with many health hazards, including increasing the risk of developing active tuberculosis [33]. Global tobacco use is increasing [44] and remains high in most countries with a high burden of tuberculosis [45,46]. Effective tobacco control measures are needed, particularly in low-income countries, where smoking rates are highest among poorer and less educated communities [47]. Earlier detection of tuberculosis should reduce the number of patients presenting with cavitary disease and those with a higher sputum bacillary load and thus contribute to earlier sputum conversion. Finally, as mentioned above, BCG vaccination offers particularly poor coverage of W- Beijing strains. Newer tuberculosis vaccines will hopefully have greater activity against all strains, including W-Beijing. Our study has several limitations. First, there were missing observations in our data. We dealt with this by imputing missing data, which has been shown to be superior to complete case analysis (in which only subjects with no missing values are analyzed). If data is missing at random, it means that the missingness of data depends on observed co-variables or the outcome [41]. We find this assumption reasonable in our study, since the missing data related mainly to data capturing issues. Therefore, multiple imputation yields unbiased estimates that correctly reflect the uncertainty related to the missing data. Second, a large proportion of our study participants did not convert during the 2 month period (49.6%), resulting in less statistical power. Similar high rates of culture positivity have been observed at 2 months at African sites using liquid culture media [4,9,42]. One of the strengths of our analysis is that we measured the time to sputum culture conversion, instead of a binary endpoint of culture conversion. Third, we did not assess the presence of a BCG scar among our participants. The absence of a BCG scar was recently shown to be an important factor in culture non-conversion among a Tanzanian cohort [8]. On the contrary, it has been suggested that the widespread use of the BCG vaccine may aid in the spread of W-Beijing strains, since it offers poor protection against W-Beijing strains in experimental models [11]. The BCG coverage among our local infants is high (99%) [43]. Predictors of delayed sputum culture conversion Fourth, rates of smoking and alcohol misuse were high in our study, but similar to those reported in another study from our In conclusion, a shorter baseline TTD, tobacco smoking, the presence of lung cavities, and W-Beijing genotype were associated with delayed sputum culture conversion within 2 months in adults with smear-positive pulmonary tuberculosis. Larger studies in high-burden settings are needed to confirm the relationship between the W-Beijing genotype and sputum culture conversion. PLoS ONE \| www.plosone.org References 1. Mitchison DA (1993) Assessment of new sterilizing drugs for treating pulmonary tuberculosis by culture at 2 months. Am Rev Respir Dis 147: 1062–1063. 23. Mallard K, McNerney R, Crampin AC, Houben R, Ndlovu R, et al. (2010) Molecular Detection of Mixed Infections of Mycobacterium tuberculosis Strains in Sputum Samples from Patients in Karonga District, Malawi. J Clin Microbiol 48(12): 4512–4518. 2. Wallis RS, Doherty TM, Onyebujoh P, Vahedi M, Laang H, et al. (2009) Biomarkers for tuberculosis disease activity, cure, and relapse. Lancet Infect Dis 9: 162–172. 24. Honaker J, King G, Blackwell M (2011) Amelia II: A program for missing data. R Package version 1.5–5. Available: http://www.gking.harvard.edu/amelia. Accessed 2011 Nov 29. 3. Dorman SE, Johnson JL, Goldberg S, Muzanye G, Padayatchi N, et al. (2009) Substitution of moxifloxacin for isoniazid during intensive phase treatment of pulmonary tuberculosis. Am J Resp Crit Care Med 180: 273–280. 25. Little RJA, Rubin DB (2002) Statistical analysis with missing data. Second edition. New York: Wiley. 408 p. 4. Hesseling AC, Walzl G, Enarson EA, Carroll NM, Duncan K, et al. (2010) Baseline sputum time to detection predicts month 2 culture conversion and recurrent tuberculosis in HIV-uninfected patients. Int J Tuberc Lung Dis 14(5): 560–570. 26. Volinsky CT, Madigan D, Raftery AE, Kronmal RA (1997) Bayesian model averaging in proportional hazard models. Assessing the risk of a stroke. Applied Statistics- Journal of the Royal Statistical Society Series C 46(4): 433–448. 5. Telzak EE, Barkat AF, Pollard CL, Turett GS, Justman JE, et al. (1997) Factors influencing time to sputum conversion among patients with smear-positive pulmonary tuberculosis. Clin Infec Dis 25: 666–670. 27. Schomaker M, Wan ATK, Heumann C (2010) Frequentist model averaging with missing observations. Computational Statistics & Data analysis 54: 3336–3347. 6. Gu¨ler M, U¨ nsal E, Dursun B, Aydin O¨ , Capan N (2007) Factors influencing sputum smear and culture conversion time among patients with new case pulmonary tuberculosis. Int J Clin Prac 61(2): 231–235. 28. Nahid P, Bliven EE, Kim EY, Mac Kenzie WR, Stout JE, et al. (2010) Influence of M. tuberculosis lineage variability within a clinical trial for pulmonary tuberculosis. PLoS ONE 5(5): e10753. 7. Banu Rekha VV, Balasubramanian R, Swaminathan S, Ramachandran R, Rahman F, et al. (2007) Sputum conversion at the end of intensive phase of Category-1 regimen in the treatment of pulmonary tuberculosis patients with diabetes mellitus or HIV infection: An analysis of risk factors. References (2006) Mycobacterium tuberculosis Beijing Genotype and risk for treatment failure and relapse, Vietnam. Emerg Infect Dis 12: 736–743. 35. Leung CC, Yew WW, Chan CK, Tam CM, Lam CW, et al. (2003) Smoking and tuberculosis in Hong Kong. Int J Tuberc Lung Dis 7(10): 980–986. 14. Burman WJ, Bliven EE, Cowan L, Bozeman L, Nahid P, et al. (2009) Relapse associated with active disease caused by Beijing strain of Mycobacterium tuberculosis. Emerg Infect Dis 15(7): 1061–1067. 36. Altet-Gomez MN, Alcaide J, Godoy P, Romero MA, del Rey H (2005) Clinical and epidemiological aspects of smoking and tuberculosis: a study of 13,038 cases. Int J Tuberc Lung Dis 9(4): 430–436. 15. Parwati I, AlisjahbanaB, Apriani L, Soetikno RD, Ottenhoff TH, et al. (2010) Mycobacterium tuberculosis Beijing genotype is an independent risk factor for tuberculosis treatment failure in Indonesia. J Infect Dis 201: 553–557. g 37. Scott GM, Murphy PG, Gemidjioglu ME (1990) Predicting deterioration of treated tuberculosis by corticosteroid reserve and C-reactive protein. J Infect 21(1): 61–69. 16. Buu TN, Huyen MNT, van Soolingen D, Lan NTN, Quy HT, et al. (2010) The Mycobacterium tuberculosis Beijing genotype does not affect tuberculosis treatment failure in Vietnam. Clin Infect Dis 51(8): 879–886. 38. Plit ML, Theron AJ, Fickl H, van Rensburg CE, Pendel S, et al. (1998) Influence of antimicrobial chemotherapy and smoking status on the plasma concentrations of vitamin C, Vitamin E, beta-carotene, acute phase reactants, iron and lipid peroxides in patients with pulmonary tuberculosis. Int J Tuberc Lung Dis 2(7): 590–596. 17. Visser ME, Grewal HMS, Swart EC, Dhansay MA, Walzl G, et al. (2011) The effect of vitamin A and zinc supplementation on treatment outcomes in pulmonary tuberculosis: a randomised controlled trial. Am J Clin Nutr 93(1): 93–100. 39. Djoba Siawaya JF, Bapela NB, Ronacher K, Veenstra H, Kidd M, et al. (2008) Immune parameters as markers of tuberculosis extent of disease and early prediction of anti-tuberculosis chemotherapy response. J Infect 56: 340–347. 18. Schoeman JH, Parry CDH, Lombard CJ, Klopper HJ (1994) Assessment of alcohol-screening instruments in tuberculosis patients. Tuber Lung Dis 75: 371–376. 40. Wilson D, Badri M, Maartens G (2011) Performance of serum C-Reactive protein as a screening test for smear-negative tuberculosis in an ambulatory high HIV prevalence population. PLoS ONE 6(1): e15248. 19. Den Boon S, Bateman ED, Enarson DA, Borgdorff MW, Verver CJ, et al. References Indian J Med Res 126: 452–458. 29. Benator D, Bhattacharaya M, Bozeman L, Burman W, Cantazaro A, et al. (2002) Rifapentine and isoniazid once a week versus rifampicin and isoniazid twice a week for treatment of drug-susceptible pulmonary tuberculosis in HIV- negative patients: a randomized trial. Lancet 360(9332): 528–534. 8. Wang J, Lee L, Yu C, Chien Y, Yang P, et al. (2009) Factors influencing time to smear conversion in patients with smear-positive pulmonary tuberculosis. Respirology 14: 1012–1019. 30. Richardson M, van Lill SWP, van der Spuy GD, Munch Z, Booysen CN, et al. (2002) Historic and recent events contribute to the disease dynamics of Beijing- like Mycobacterium tuberculosis isolates in a high incidence region. Int J Tuberc Lung Dis 6(11): 1001–1011. 9. Jeremiah K, PrayGod G, Faurholt-Jepsen, Range N, Andersen AB, et al. (2010) BCG vaccination status may predict sputum conversion in patients with pulmonary tuberculosis: a new consideration for an old vaccine? Thorax 65(12): 1072–1076. 31. Marais BJ, Victor RC, Hesseling AC, Barnard M, Jordaan A, et al. (2006) Beijing and Haarlem genotypes are overrepresented among children with drug- resistant tuberculosis in the Western Cape Province of South Africa. J Clin Microbiol 44(10): 3539–3543. 10. Lawson L, Thacher TD, Yassin MA, Onuoha NA, Usman A, et al. (2010) Randomized controlled trial of zinc and vitamin A as co-adjuvants for the treatment of pulmonary tuberculosis. Trop Med Int Health 15(12): 1481–1490. 32. Cowley D, Govender D, February B, Wolfe M, Steyn L, et al. (2008) Recent and rapid emergence of W-Beijing strains of Mycobacterium tuberculosis in Cape Town, South Africa. Clin Infect Dis 47: 1252–1259. p y p ( ) 11. Nicol MP, Wilkinson RJ (2008) The clinical consequences of strain diversity in Mycobacterium tuberculosis. Trans R Soc Trop Med Hyg 102(10): 955–965. 33. Slama K, Chian C-Y, Enarson DA, Hassmiller K, Fanning A, et al. (2007) Tobacco and tuberculosis: a qualitative systematic review and meta-analysis. Int J Tuberc Lung Dis 11(10): 1049–1061. 12. Parwati I, van Crevel R, van Soolingen D (2010) Possible underlying mechanisms for successful emergence of the Mycobacterium tuberculosis Beijing genotype strains. Lancet Infect Dis 10(2): 103–111. 34. Durban Immunotherapy group (1999) Immunotherapy with Micobacterium vaccae in patients with newly diagnosed pulmonary tuberculosis: a randomized controlled trial. Lancet 354: 116–119. 13. Lan NTN, Lien HTK, Tung LB, Borgdorff MW, Kremer K, et al. Acknowledgments We would like to thank our field staff as well as the clinic staff at the Delft and Delft-South Community Health Centres, for their valuable contribu- tions towards the micronutrient trial. M.C.S. was responsible for strain typing under supervision of Dr. H. Segal at the Institute for Infectious Disease and Molecular Medicine, University of Cape Town and R.W. We would also like to thank the following people and institutions: Dr. A. PLoS ONE \| www.plosone.org January 2012 \| Volume 7 \| Issue 1 \| e29588 January 2012 \| Volume 7 \| Issue 1 \| e29588 5 Predictors of Culture Conversion in Tuberculosis Predictors of Culture Conversion in Tuberculosis Predictors of Culture Conversion in Tuberculosis Whitelaw, V. January and K. Mentoor from the NHLS Groote Schuur Microbiology Laboratory; M. Marais from the Nutritional Intervention Unit, MRC for the biochemical analyses; A. Menezes, Dr. N. Chegou and M.de Kock from the MRC Centre of Molecular and Cellular Biology at the University of Stellenbosch for the processing and storing of all blood samples; Dr. R. Dawson and Dr. K. Narunsky from the UCT Lung Institute for the evaluation of the chest radiographs. Author Contributions Conceived and designed the experiments: MEV HMSG GW RW ECS GM. Performed the experiments: MEV MCS. Analyzed the data: MEV MCS MS GM. Contributed reagents/materials/analysis tools: HMSG GW RW MS. Wrote the paper: MEV MCS GW RW HMSG MS ECS GM. January 2012 \| Volume 7 \| Issue 1 \| e29588 47. Hosseinpoor AR, Parker LA, d’Espaignet, Chatterji S (2011) Social determi- nants of smoking in low- and middle-income countries: results from the World Health survey. PLoS ONE 6(5): e20331. 45. World Health Organization (2011) WHO report on the global tobacco epidemic. World Health Organization, Geneva. Available: http://www.who. int/tobacco. Accessed 2011 Oct 25. 46. World Health Organization (2011) Global tuberculosis control 2011. World Health Organization, Geneva. Available: http://www.who.int/tb/publications/ global_report/en/www.who. Accessed 2011 Oct 25. 46. World Health Organization (2011) Global tuberculosis control 2011. World Health Organization, Geneva. Available: http://www.who.int/tb/publications/ global_report/en/www.who. Accessed 2011 Oct 25. 45. World Health Organization (2011) WHO report on the global tobacco epidemic. World Health Organization, Geneva. Available: http://www.who. int/tobacco. Accessed 2011 Oct 25. Predictors of Culture Conversion in Tuberculosis References (2005) Development and evaluation of a new chest radiograph reading and recording system for epidemiological surveys of tuberculosis and lung disease. Int J Tuberc Lung Dis 9(10): 1088–1096. 41. Van der Heijden GJMG, Donders ART, Stijnen T, Moons KGM (2006) Imputation of missing values is superior to complete case analysis and the missing-indicator method in multivariable diagnostic research: A clinical example. J Clin Epidemiol 59: 1102–1109. 20. World Health Organization, Regional Office for South -East Asia (2006) Chapter 17 Tuberculosis. In: Blood Safety and Clinical Technology. Guidelines on Standard Operating Procedures for Microbiology. New Delhi, India. Available: http://www.searo.who.int/EN/Section10/Section17/Section53/ Section482_1799.htm. Accessed 2009 Dec 1. 42. Rustomjee R, Lienhardt C, Kanyok T, Davies GR, Levin J, et al. (2008) A Phase II study of the sterilising activities of ofloxacin, gatifloxacin and moxifloxacin in pulmonary tuberculosis. Int J Tuberc Lung Dis 12(2): 128–138. 21. De Wit D, Steyn L, Shoemaker S, Sogin M (1990) Direct detection of Mycobacterium tuberculosis in clinical specimens by DNA amplification. J Clin Microbiol 28: 2437–2441. p y J g ( ) 43. Corrigal J (2006) Vaccination coverage of the Western Cape Province, 2005. 43. Corrigal J (2006) Vaccination coverage of the Western Cape Province, 200 Provincial Government of the Western Cape, Cape Town, South Africa. igal J (2006) Vaccination coverage of the Western Cape Province, 2 Provincial Government of the Western Cape, Cape Town, South Africa. 44. Food and Agriculture Organization of the United Nations (2003) Projections of tobacco production, consumption and trade to the year 2010. FAO, Rome, Italy. 22. Kamerbeek J, Schouls L, Kolk A, van Agterveld M, van Soolingen D, et al. (1997) Simultaneous detection and strain differentiation of Mycobacterium tuberculosis for diagnosis and epidemiology. J Clin Microbiol 35(4): 907–914. PLoS ONE \| www.plosone.org January 2012 \| Volume 7 \| Issue 1 \| e29588 6 Predictors of Culture Conversion in Tuberculosis Predictors of Culture Conversion in Tuberculosis PLoS ONE \| www.plosone.org January 2012 \| Volume 7 \| Issue 1 \| e29588 7 7
https://openalex.org/W4297832880	https://zenodo.org/records/814850/files/1475-9276-13-10.pdf	English	null	Inequities in access to health care in different health systems: a study in municipalities of central Colombia and north-eastern Brazil.	Zenodo (CERN European Organization for Nuclear Research)	2,014	cc-by	13,380	* Correspondence: igarcia@consorci.org 1Health Policy and Health Services Research Group, Health Policy Research Unit, Consortium for Health Care and Social Services of Catalonia, Avenida Tibidabo, 21, Barcelona 08022, Spain 2Ph. D. Programme in Biomedicine, Department of Experimental and Health Sciences, Universitat Pompeu Fabra, Barcelona, Spain Full list of author information is available at the end of the article Inequities in access to health care in different health systems: a study in municipalities of central Colombia and north-eastern Brazil Irene Garcia-Subirats1,2*, Ingrid Vargas1, Amparo Susana Mogollón-Pérez3, Pierre De Paepe4, Maria Rejane Ferreira da Silva5, Jean Pierre Unger4, Carme Borrell6,7,8 and Maria Luisa Vázquez1 Abstract Introduction: Health system reforms are undertaken with the aim of improving equity of access to health care. Their impact is generally analyzed based on health care utilization, without distinguishing between levels of care. This study aims to analyze inequities in access to the continuum of care in municipalities of Brazil and Colombia. Methods: A cross-sectional study was conducted based on a survey of a multistage probability sample of people who had had at least one health problem in the prior three months (2,163 in Colombia and 2,167 in Brazil). The outcome variables were dichotomous variables on the utilization of curative and preventive services. The main independent variables were income, being the holder of a private health plan and, in Colombia, type of insurance scheme of the General System of Social Security in Health (SGSSS). For each country, the prevalence of the outcome variables was calculated overall and stratified by levels of per capita income, SGSSS insurance schemes and private health plan. Prevalence ratios were computed by means of Poisson regression models with robust variance, controlling for health care need. Introduction: Health system reforms are undertaken with the aim of improving equity of access to health care. Their impact is generally analyzed based on health care utilization, without distinguishing between levels of care. This study aims to analyze inequities in access to the continuum of care in municipalities of Brazil and Colombia. Results: There are inequities in favor of individuals of a higher socioeconomic status: in Colombia, in the three different care levels (primary, outpatient secondary and emergency care) and preventive activities; and in Brazil, in the use of outpatient secondary care services and preventive activities, whilst lower-income individuals make greater use of the primary care services. In both countries, inequity in the use of outpatient secondary care is more pronounced than in the other care levels. Income in both countries, insurance scheme enrollment in Colombia and holding a private health plan in Brazil all contribute to the presence of inequities in utilization. Conclusions: Twenty years after the introduction of reforms implemented to improve equity in access to health care, inequities, defined in terms of unequal use for equal need, are still present in both countries. The design of the health systems appears to determine access to the health services: two insurance schemes in Colombia with different benefits packages and a segmented system in Brazil, with a significant private component. © 2014 Garcia-Subirats et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. RESEARCH Open Access Inequities in access to health care in different health systems: a study in municipalities of central Colombia and north-eastern Brazil Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Introduction In both countries, care is organized by levels of com- plexity, with primary care as the entry point and care coordinator for the patient and the secondary level in a supporting role [16,18]. In the private healthcare subsystem, outpatient secondary care services can be accessed directly (via a private health plan or out-of- pocket payment). The effects of these reforms on access in Brazil and Colombia have been analyzed indirectly through study- ing the design of the health system [19,20], changes in coverage in the family health program in Brazil [21], or enrollment with an insurer in Colombia [22], and in a more direct way through looking at changes in utilization and equity in utilization before and after the reforms [23,24]. There are also studies available for both countries analyzing equity in the utilization of the health services, mostly based on national surveys of the general population: the Quality of Life Survey (Encuesta de Calidad de Vida, ECV) and the Demog- raphy and Health Survey (Encuesta de Demografía y Salud, EDS) in Colombia, and the National Household Sample Survey (Pesquisa Nacional de Amostra por Domicilios, PNAD) in Brazil. These studies indicate that, irrespective of need, individuals of a higher socio- economic status – higher income [5,7,24-27], higher level of education [23,25,28,29] or more favorable working conditions (stable employment) [30] – make greater use of the health services. In Brazil, people who have private medical insurance are shown to be more likely to use the health services [31]. These analyses, however, have certain limitations. In both countries, the analysis of equity in access focuses on the utilization of ser- vices in general [5,6,26,27], without distinguishing between care levels, despite the international evidence showing that inequities vary between primary and secondary care [32-34], with the inequities being more pronounced in the latter. Only one study has been found for Colombia which Both countries reformed their health systems more than two decades ago with the common objective (among others) of improving equity of access, but they opted for different models. Colombia introduced the General System of Social Security in Health (Sistema General de Seguridad Social en Salud or SGSSS in Spanish), based on managed competition and made up of two insurance schemes: the contributory scheme, for formal sector employees and those able to pay, and the subsidized scheme for the low income population. Introduction Competition for contracts with the insurers was also introduced among public and private healthcare pro- viders (Instituciones Prestadoras de Salud or IPS in Spanish). The uninsured population, 12.1% [15], receives care in public healthcare service networks which are organized by regional and local health authorities [16]. In an egalitarian approach, based on the notion of social justice, equity is defined either as equal treatment for equal need (horizontal equity) or as different treatment for different needs (vertical equity) [1]. In health services research [2-7] access is usually analyzed based on the notion of horizontal equity, which would be considered to exist when access is dependent on need and not on other socioeconomic or demographic factors [8]. In Brazil, the health sector is made up of two subsys- tems: on one side, the Unified Health System (Sistema Único de Saúde or SUS in Portuguese), conceived as a tax-funded national health system, decentralized accord- ing to the political structure of the country (federation, states and municipalities) and free at point of delivery [17], with care provided by public or contracted private providers; and on the other side, the private system (sup- plementary system), to which the population gains ac- cess via direct payment or private insurance schemes, which have their own network provider. Colombia and Brazil are the most populous countries of Latin America. With a Gross Domestic Product (GDP) per capita of $7,752 and $11,340 (US dollars) respectively in 2012 [9,10], they are both classified as middle income countries [11]. Despite the gains in re- cent years in terms of poverty reduction and improve- ments in the Gini index [12], they still have the highest Gini coefficients in Latin America: in 2009, 0.57 [9] and 0.55 respectively [10], indicating considerable in- equalities in income distribution. The illiteracy rate in Colombia (6.4%) is lower than the Latin American average (7.8%) and that of Brazil (9.6%) [13]. Public health expenditure as a percentage of the GDP was 4.6% in Colombia and 4.1% [9] in Brazil in 2011 [10]. They have similar life expectancy at birth (74.0 in Colombia and 73.9 in Brazil) which is close to the Latin American average [13], while the infant mortality rate is slightly higher in Colombia (18.4 per 1000 live births) than in Brazil (15.3 per 1000 live births) or Latin America in general (16.3 per 1000 live births) [13]. Abstract Keywords: Access to health care, Inequities, Primary health care, Secondary care, Emergency care, Preventive health services, Colombia, Brazil © 2014 Garcia-Subirats et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. Page 2 of 15 Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Introduction Those that do not manage to enroll in either of the two schemes remain uninsured. Health insurance is managed by healthcare insurers (Empresas Promotoras de Salud or EPS in Spanish) for both the contributory scheme (EPS-C) and the subsidized scheme (EPS-S). Competi- tion was introduced between insurers for the enrollment of the population and they receive a capitation payment to cover different benefits packages in each scheme: the Obligatory Health Plan (Plan Obligatorio de Salud or POS in Spanish) for the contributory scheme and the Obligatory Health Plan - Subsidized (POS-S) for the sub- sidized scheme [14], with fewer services than the POS. The insurers are responsible for organizing their own network of health providers for their enrollees. The pro- viders are different for the two schemes. Page 3 of 15 Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Page 3 of 15 Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 differentiates utilization according to care levels [7] and there are none for Brazil. Likewise, equity in access to pre- ventive services [28,35,36] is a little explored subject matter. In Colombia, analyses also tend to focus on the differences between those covered by the SGSSS –either in general [24] or in the subsidized scheme [37,38]- and the unin- sured, whilst studies comparing the use of services in the different insurance schemes are rare and give conflicting results [27,28,39]. (http://www2.equity-la.eu/), the broader Project in which this study is framed [52]. The areas were selected for be- ing densely populated urban spaces with a high propor- tion of the population belonging to the low or medium- low socioeconomic strata and with varying geographical access to specialist care. Kennedy has approximately 1,000,000 residents and the other three study areas have about 300,000 residents. Kennedy and the two micro- regions of District 3 in Recife share the characteristic of being some of the more deprived areas in their respec- tive cities. In both countries the studies tend to analyze inequities between large geographical regions (Departments or States) [40,41], but not between smaller areas within these, although there are signs of inequities in access within regions, for example between the different local- ities in the city of Bogotá [42], a region with one of the highest percentages of health services utilization in Colombia [15]. Introduction In Brazil, the studies which analyze smaller areas mainly focus on the south of the country [43-46], and there are practically none for the northeast, despite this region having lower utilization rates than the national average and greater inequity in the use of the health services [40]. Study population and sample The study population was made up of residents in the study areas who had had at least one health problem or had used to the health services in the three months prior to the survey. The sample size was calculated for each study area based on the population size and an estimated propor- tion of 50% (maximum uncertainty principle); degree of confidence: 90% (alpha error of 0.1); precision: 2.5. The sample size was 2,163 in Colombia (1,083 in Kennedy, 1,080 in Soacha) and 2,155 in Brazil (1,076 in district 3 of Recife, 1,079 in Caruaru). Lastly, equity of access in the health systems of Colombia and Brazil has been analyzed in the context of Latin American-wide studies which compare either the design of different health systems [1,47-49] or equity in utilization [4,41,50]. These are generally based on na- tional surveys which were not designed for this purpose, so differences in the questions asked make them difficult to compare [4,41]. The only comparative study between the two countries analyzes inequities in health [51], but not in access to services. In both countries, multistage probability sampling was conducted. In the first stage, census tracts were ran- domly selected (in Soacha, from all six comunas – i.e. districts) with replacement. In the second stage, house- holds were systematically selected. The sample range was calculated according to sample size and number of households in each neighborhood; the initial home was randomly selected. The household was considered the primary sampling unit to avoid the effect of associated samples in individuals belonging to a family. Identifying the care levels in which there is inequity in utilization (and to what degree) in different health sys- tems may contribute to the design of policies aimed at reducing this problem. The purpose of this study is to conduct a comparative analysis of equity in access to the health services distinguishing between different care levels (primary, outpatient secondary, emergency and preventive care) in two areas of Colombia and Brazil, focusing particularly on the role of insurance scheme enrollment in Colombia and the possession of a private health plan in Brazil. Questionnaire A i i A questionnaire was designed to analyze access to health care. Based on the Behavioral Model of Health Services Use [53] and previous qualitative research [54-58], di- mensions and variables were identified. In addition, systematic reviews were conducted of studies on access and equity of access to health care and of available tools in order to identify additional variables. The literature search was carried out using the most relevant electronic databases (Pubmed, CINAHL, Social Science Citation Index, PsycInfo, Lilacs, IBECS, The Cochrane Library, System for Information on Grey Literature in Europe, Information system of the WHO Library, Panamerican Health Organization Library) to minimize the likelihood of excluding relevant studies. The search strategy in- cluded the combination of descriptors and keywords relating to equity in access to health care, delivery of health care and measurement instruments, utilizing the Boolean operator ‘AND’. Given the large number of Variables The outcome variables are three dichotomous variables on utilization of healthcare services in the three months prior to the survey: a) consultation of a general practi- tioner or pediatrician, b) consultation of a specialist, c) consultation of emergency care services; and three variables on the use of preventive services by adults in the last year: d) glycemic control, e) caries prevention and f) mammography in women. Need for care was measured by self-rated health (di- chotomous: i) good - very good and good, and ii) poor - fair, poor and very poor) and having at least one chronic condition [59]. The main explanatory variables were per capita income and holding a private health plan in both countries as a proxy of socio-economic status, and type of SGSSS insurance scheme in Colombia. Per capita in- come (less than half minimum wage (MW), ½ - 1 MW, 2 or more MWs) was estimated by dividing the house- hold income by the family size. Sociodemographic vari- ables - sex, age (0–17, 18–40, 41–65, 66 or over) and race/color- were used to adjust the models. Design and study area studies identified, only studies from Colombia, Brazil, United States and Spain were included. The U.S. and Spain were included because their health systems are comparable to the Colombian and Brazilian system re- spectively. A total of 41 different articles were identified up to 2010. Based on these results, the first version of the questionnaire was built and validated by means of discussions with experts, after which it was adapted to both context and languages. A pre-test and a pilot test were conducted in each country to evaluate the rhythm of the interview, interviewer burden and comprehensibility. Both the rhythm of the interview and the interviewer bur- den proved adequate. With respect to comprehensibility, some questions were removed and others modified because the terms used were unfamiliar and people had problems understanding them. Finally, since many changes had been made, another pilot was deemed necessary, in which the questionnaire was finally considered to be adequate. The final questionnaire is divided into nine sections. The first collects information on perceived health needs and related behavior in the three months prior to the survey, in other words, whether or not the individual had used the health services and the level of care at which they had been attended. The next four sections refer to their most recent experience - within the three months prior to the survey - using the different levels of health care (primary, outpatient secondary, emergency, and inpatient care) of the SGSSS and SUS services. The last three sections include a Likert scale to measure care continuity, knowledge of the health- care system and sociodemographic data. The Colombian questionnaire has an additional section related to insurance enrollment. Data collection D l Data were collected from February to June 2011 by means of face-to-face interviews conducted by specific- ally trained interviewers in both countries. Strategies to ensure the quality and consistency of data included close supervision of interviewers in the field, the review of all questionnaires and 20% of re-interviews (selected randomly). Inconsistencies during data entry were controlled through the double-entry method. In addition, stratified analyses were carried out: firstly, re- gression models were generated stratifying by income to evaluate the effect of the insurance scheme in Colombia and additional private health insurance in Brazil on health services utilization for each socioeconomic stratum (Tables 4 and 5); and secondly, regression models were made stratifying by type of insurance in both countries (contributory and subsidized in Colombia, and only SUS or SUS and private health plan in Brazil) (Table 6). Analyses were carried out with STATA statistical pack- age version 12 [61]. Data analysis A univariate analysis was performed to describe the dis- tribution of the outcome and explanatory variables for each country (Table 1). To establish the relative associ- ation between each healthcare utilization variable and socioeconomic status, prevalence ratios (PR) and their corresponding 95% confidence intervals were computed by means of Poisson regression models with robust vari- ance (Tables 2 and 3). PR was considered better suited to our study than other measures of inequality such as the Relative Index of Inequality (RII) because the socio- economic variables included in the models are not strictly hierarchical [60]. Absolute differences in utilization preva- lence at both extremes of the three socioeconomic variables were also calculated. Both absolute and relative measures were used to assess socioeconomic inequalities in the utilization of healthcare services. Design and study area A cross-sectional study was conducted by means of a population survey in central Colombia and the northeast of Brazil. The study areas were two municipalities in each country: in Colombia, Kennedy (a district of Bogotá) and Soacha; and in Brazil, two micro-regions (3.2 and 3.3) of District 3 in Recife, Pernambuco’s cap- ital, and Caruaru, in the interior of Pernambuco state. These four areas are the areas of the Equity-LA project Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Page 4 of 15 Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 studies identified, only studies from Colombia, Brazil, United States and Spain were included. The U.S. and Spain were included because their health systems are comparable to the Colombian and Brazilian system re- spectively. A total of 41 different articles were identified up to 2010. Based on these results, the first version of the questionnaire was built and validated by means of discussions with experts, after which it was adapted to both context and languages. A pre-test and a pilot test were conducted in each country to evaluate the rhythm of the interview, interviewer burden and comprehensibility. Both the rhythm of the interview and the interviewer bur- den proved adequate. With respect to comprehensibility, some questions were removed and others modified because the terms used were unfamiliar and people had problems understanding them. Finally, since many changes had been made, another pilot was deemed necessary, in which the questionnaire was finally considered to be adequate. The final questionnaire is divided into nine sections. The first collects information on perceived health needs and related behavior in the three months prior to the survey, in other words, whether or not the individual had used the health services and the level of care at which they had been attended. The next four sections refer to their most recent experience - within the three months prior to the survey - using the different levels of health care (primary, outpatient secondary, emergency, and inpatient care) of the SGSSS and SUS services. The last three sections include a Likert scale to measure care continuity, knowledge of the health- care system and sociodemographic data. The Colombian questionnaire has an additional section related to insurance enrollment. withdraw from the survey, anonymity, confidentiality and protection of data were all guaranteed. Ethical considerations Ethics approval was obtained from the ethics committees in the participating countries: the National Committee of Research Ethics in Brazil; the Research Ethics Committee of the Health Sciences School of Universidad del Rosario in Colombia; the Institutional Review Board of the Institute of Tropical Medicine in Belgium; and the Clinical Research Ethics Committee of Parc de Salut Mar in Spain. All inter- viewees participated on a voluntary basis, after signing an informed consent. The right to refuse to participate or Page 5 of 15 Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Results Colombia and Brazil are used here to refer to the study th h th lt t t l t d t Table 1 Sociodemographic characteristics and perceived health care need of the study sample, in the study areas of Colombia and Brazil (2011) Colombia (n = 2,163) Brazil (n = 2,155) n % n % Sex Male 691 31.9 626 29.0 Female 1,472 68.1 1,529 71.0 Age 0-17 300 13.9 483 22.4 18-40 667 30.8 497 23.1 41-65 909 42.0 766 35.6 >65 287 13.3 409 19.0 Level of education None 344 16.0 623 29.5 Primary 804 37.4 583 27.6 Secondary 821 38.2 798 37.8 University 180 8.4 106 5.0 Per capita income < ½ MW 951 44.0 1,121 52.0 ½ - 1 MW 775 35.8 720 33.4 > 1 MW 437 20.2 314 14.6 Private health plan Yes 42 2.0 434 20.1 No 2,024 98.0 1,721 79.9 SGSSS scheme Contributory 1,144 56.0 Subsidized 574 28.0 Special 97 4.8 Uninsured 231 11.3 Self-rated health status Good 1,346 62.3 962 44.7 Poor 816 37.7 1,192 55.3 Chronic condition Yes (at least one) 650 30.0 903 41.9 No 1,513 70.0 1,252 58.1 MW, minimum wage; SGSSS, General System of Social Security in Health. Table 1 Sociodemographic characteristics and perceived health care need of the study sample, in the study areas of Colombia and Brazil (2011) Table 1 Sociodemographic characteristics and perceived health care need of the study sample, in the study areas of Colombia and Brazil (2011) SGSSS enrollment, 56.0% are enrolled in the contribu- tory scheme, 28.0% in the subsidized scheme, and 11.3% are uninsured. Use of health care services In the Colombia sample more than half of the individuals used the first level of care (52.2%), a higher proportion than in that of Brazil (44.5%). However, for outpatient secondary and emergency care, utilization in Brazil was significantly higher than in Colombia (Figure 1). Prevent- ive activities in the last year differ between countries ac- cording to type of activity: while the proportions of people who used glycemic control and mammography services were similar, significant differences were found in caries prevention in adults (37.7% in Colombia vs. 19.9% in Brazil) (Figure 1). Results Colombia and Brazil are used here to refer to the study areas even though the results are not extrapolated at country level. Differences in socioeconomic-related inequalities in health services utilization At primary care level, per capita income does not figure as a source of inequality in Colombia but it does in Brazil, where individuals with a higher per capita income make less use of this care level (prevalence difference with respect to the poorest is −17.9, PR = 0.71, CI 95%: 0.59-0.85). In Colombia, it is the insurance scheme which is associated with the utilization of services at this level. The utilization prevalence of individuals in the contributory scheme is 58.0%, in the subsidized scheme 52.1%, and 24.7% for the uninsured (Table 2). In out- patient secondary care, per capita income is a source of inequity in both countries: higher-income individuals are more likely to use these services. In Colombia, the PR of individuals with more than one minimum wage per capita to those with the lowest income is 1.26 (CI 95%: 1.02-1.56) and in Brazil, 1.42 (CI 95%: 1.14-1.77), which is the opposite pattern of use to that detected at primary care level. Holding a private health plan is also associ- ated with a higher prevalence of utilization in both countries. Furthermore, in Colombia, insurance enroll- ment reveals the same pattern as that found in primary Ethical considerations The socioeconomic status of the sample is not very high in either country: most people have less than half of the minimum wage (MW) per month, but distribution is less equal in Brazil than in Colombia. The socioeconomic level in Colombia is closely related to en- rollment; the proportion of people with less than half MW in the previous month is higher in the subsidized scheme and the uninsured group than in the contribu- tory scheme: 72.0%, 63.2% and 36.4% respectively. In Colombia 2.0% and in Brazil 20.1% hold a private health plan (PHP) (Table 1). Self-rated health status is better in Colombia: 62.2% report good or very good health, compared to 44.7% in Brazil; 30.0% claim to have at least one chronic disease, as opposed to 41.9% in Brazil (Table 1). Sample characteristics In both samples, the majority of participants are women, and while in the Colombia sample there is a predomin- ance of the central age groups (18 to 40 and 41 to 65 years), in that of Brazil there is a more even distribution of the sample across all age groups. With regard to Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 Page 6 of 15 http://www.equityhealthj.com/content/13/1/10 Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Page 6 of 15 Table 2 Prevalence, prevalence difference and prevalence ratios (95% confidence intervals) of health services utilization in the last three months by per capita income, SGSSS insurance scheme and private health plan, in the study areas of Colombia and Brazil (2011) Colombia Brazil n Prev. PD PR (CI 95%)a n Prev. Sample characteristics PD PR (CI 95%)a Primary care Per capita income < 1/2 MW 473 49.8 1 550 49.1 1 1/2 - 1 MW 409 52.8 3.0 1.04 (0.94 - 1.14) 310 43.1 −6.0 0.89 (0.80 - 0.99) > 1 MW 246 56.4 6.6 1.01 (0.91 - 1.13) 98 31.2 −17.9 0.71 (0.59 - 0.85) SGSSS scheme Contributory + special 718 58.0 1 Subsidized 298 52.1 −5.9 0.90 (0.81 - 0.99) Uninsured 57 24.7 −33.3 0.46 (0.37 - 0.58) Private Health Plan No 1,041 51.5 1 800 46.5 1 Yes 27 64.3 12.8 1.13 (0.90 - 1.41) 158 36.4 −10.1 0.91 (0.79 - 1.05) Outpatient secondary care Per capita income < 1/2 MW 160 16.9 1 228 20.3 1 1/2 - 1 MW 150 19.4 2.5 1.02 (0.83 - 1.26) 200 27.8 7.5 1.26 (1.07 - 1.48) > 1 MW 118 27.1 10.2 1.26 (1.02 - 1.56) 96 30.6 10.3 1.42 (1.14 - 1.77) SGSSS scheme Contributory + special 300 24.2 1 Subsidized 93 16.3 −7.9 0.72 (0.57 - 0.90) Uninsured 17 7.4 −16.8 0.40 (0.25 - 0.64) Private Health Plan No 393 19.5 1 391 22.7 1 Yes 16 38.1 18.6 1.74 (1.14 - 2.66) 133 30.7 8.0 1.40 (1.18 - 1.67) Emergency care Per capita income < 1/2 MW 192 20.2 1 390 34.8 1 1/2 - 1 MW 141 18.2 −2.0 1.14 (0.92 - 1.41) 240 33.3 −1.5 1.09 (0.95 - 1.25) > 1 MW 70 16.1 −4.1 0.98 (0.74 - 1.29) 74 23.6 −11.2 0.88 (0.70 - 1.11) SGSSS scheme Contributory + special 230 18.6 1 Subsidized 123 21.5 2.9 0.94 (0.76 - 1.17) Uninsured 33 14.3 −4.3 0.63 (0.45 -0.88) Private Health Plan No 371 18.4 1 579 33.6 1 Yes 12 28.6 10.2 1.75 (1.06 - 2.86) 125 28.8 −4.8 0.93 (0.78 - 1.11) Prev, Prevalence; PD, Prevalence Difference; PR, Prevalence ratio; CI, Confidence Interval; MW, minimum wage; SGSSS, General System of Social Security in Health. aPrevalence Ratio and 95% confidence interval adjusted for sex, age, race/color, chronic condition, self-rated health status. Statistically significant results are shown in bold. Sample characteristics Table 2 Prevalence, prevalence difference and prevalence ratios (95% confidence intervals) of health services utilization in the last three months by per capita income, SGSSS insurance scheme and private health plan, in the study areas of Colombia and Brazil (2011) Prev, Prevalence; PD, Prevalence Difference; PR, Prevalence ratio; CI, Confidence Interval; MW, minimum wage; SGSSS, General System of Social Security in Health. aPrevalence Ratio and 95% confidence interval adjusted for sex, age, race/color, chronic condition, self-rated health status. Statistically significant results are shown in bold. Prev, Prevalence; PD, Prevalence Difference; PR, Prevalence ratio; CI, Confidence Interval; MW, minimum wage; SGSSS, General System of Social Security in Health. aPrevalence Ratio and 95% confidence interval adjusted for sex, age, race/color, chronic condition, self-rated health status. Statistically significant results are shown in bold. to contributory is 0.40 (CI 95%; 0.24-0.64) (Table 2). With regard to emergency care services, the higher the per capita income is, the lower the prevalence of utilization in both to contributory is 0.40 (CI 95%; 0.24-0.64) (Table 2). With regard to emergency care services, the higher the per capita income is, the lower the prevalence of utilization in both care. Individuals in the contributory scheme display the highest use of this care level: the PR of subsidized to contributory is 0.72 (CI 95%: 0.57-0.90) and of uninsured Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 Page 7 of 15 http://www.equityhealthj.com/content/13/1/10 Garcia-Subirats et al. Sample characteristics International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Page 7 of 15 Page 7 of 15 Table 3 Prevalence, prevalence difference and prevalence ratios (95% confidence intervals) of preventive care activities in the last year by per capita income, SGSSS insurance scheme and private health plan, in the study areas of Colombia and Brazil (2011) Table 3 Prevalence, prevalence difference and prevalence ratios (95% confidence intervals) of preventive care activities in the last year by per capita income, SGSSS insurance scheme and private health plan, in the study areas of Colombia and Brazil (2011) Colombia Brazil n Prev PD PR (CI 95%)a n Prev PD PR (CI 95%)a Glycemic control (adults) Per capita income < 1/2 MW 441 49.8 1 427 52.6 1 1/2 - 1 MW 275 47.5 −2.3 0.97 (0.87 - 1.08) 346 58.8 6.2 1.11 (1.01 - 1.21) > 1 MW 230 57.6 7.8 1.09 (0.98 - 1.22) 160 58.8 6.2 1.11 (0.98 - 1.26) SGSSS scheme Contributory + special 626 56.6 1 Subsidized 223 47.8 −8.8 0.83 (0.75 - 0.93) Uninsured 54 28.3 −28.3 0.59 (0.47 - 0.75) Private Health Plan No 890 50.9 1 726 53.6 1 Yes 19 54.3 3.4 0.98 (0.73 - 1.33) 207 65.1 11.5 1.26 (1.14 - 1.40) Caries prevention (adults) Per capita income < 1/2 MW 308 34.8 1 123 15.2 1 1/2 - 1 MW 232 40.1 5.3 1.13 (0.98 - 1.30) 129 21.9 6.7 1.34 (1.07 - 1.67) > 1 MW 162 40.6 5.8 1.12 (0.94 - 1.32) 81 29.8 14.6 1.41 (1.08 - 1.84) SGSSS scheme Contributory + special 437 39.5 1 Subsidized 172 36.8 −2.7 0.95 (0.81 - 1.11) Uninsured 59 30.9 −8.6 0.75 (0.60 - 0.94) Private Health Plan No 658 37.6 1 218 16.1 1 Yes 15 42.9 5.3 1.15 (0.80 - 1.65) 115 36.2 20.1 1.86 (1.52 - 2.28) Breast cancer prevention (women) Per capita income < 1/2 MW 144 21.9 1 168 26.3 1 1/2 - 1 MW 120 30.0 8.1 1.27 (1.04 - 1.57) 128 27.9 1.6 1.10 (0.90 - 1.33) > 1 MW 96 37.2 15.3 1.37 (1.10 - 1.71) 69 37.3 11.0 1.47 (1.14 - 1.90) SGSSS scheme Contributory + special 261 34.7 1 Subsidized 63 17.8 −16.9 0.59 (0.46 - 0.76) Uninsured 15 11.0 −23.7 0.43 (0.27 - 0.70) Private Health Plan No 344 27.8 1 276 26.5 1 Yes 9 36.0 8.2 0.99 (0.58 - 1.67) 89 36.8 10.3 1.38 (1.11 - 1.70) Prev, Prevalence; PD, Prevalence Difference; PR, Prevalence ratio; CI, Confidence Interval; MW, minimum wage; SGSSS, General System of Social Security in Health. Sample characteristics aPrevalence Ratio and 95% confidence interval adjusted for sex, age, race/color, chronic condition, self-rated health status. Statistically significant results are shown in bold. Prev, Prevalence; PD, Prevalence Difference; PR, Prevalence ratio; CI, Confidence Interval; MW, minimum wage; SGSSS, General System of Social Security in Health. aPrevalence Ratio and 95% confidence interval adjusted for sex, age, race/color, chronic condition, self-rated health status. Statistically significant results are shown in bold. individuals not insured under the SGSSS are unfavorably positioned with respect to those insured under the con- tributory scheme (PR: 0.63, CI 95% 0.45-0.88) (Table 2). individuals not insured under the SGSSS are unfavorably positioned with respect to those insured under the con- tributory scheme (PR: 0.63, CI 95% 0.45-0.88) (Table 2). countries, although once this is adjusted for need there are no perceivable inequities associated with income and pri- vate insurance in either country. However, in Colombia Garcia-Subirats et al. Sample characteristics International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Page 8 of 15 Page 8 of 15 Table 4 Prevalence, prevalence difference and prevalence ratios (95% confidence intervals) of health services utilization in the last three months and of preventive activities in the last year by SGSSS insurance scheme, according to per capita income, in the study areas of Colombia (2011) < ½ MW (per capita) ½ - 1 MW (per capita) > 1 MW (per capita) n Prev PD PR (CI 95%)a n Prev PD PR (CI 95%)a n Prev PD PR (CI 95%)a Primary care Contributory + special 261 57.9 1 243 58.0 1 214 58.0 1 Subsidized 210 51.0 −6.9 0.88 (0.78 - 1.01) 70 54.7 −3.3 0.92 (0.77 - 1.10) 18 56.3 −1.7 0.91(0.65 - 1.29) Uninsured 37 25.3 −32.6 0.48 (0.35 - 0.64) 14 20.9 −37.1 0.41 (0.26 - 0.65) 6 33.3 −24.7 0.59 (0.31 - 1.13) Outpatient secondary care SGSSS scheme Contributory + special 105 23.3 1 89 21.2 1 106 28.7 1 Subsidized 65 15.8 −7.5 0.70 (0.53 - 0.94) 19 14.8 −6.4 0.59 (0.38 - 0.91) 9 28.1 −0.6 1.16 (0.66 - 2.05) Uninsured 12 8.2 −15.1 0.45 (0.25 - 0.81) 5 7.5 −13.7 0.38 (0.16 - 0.88) 0 0 – – Emergency care Contributory + special 87 19.3 1 84 20.1 1 59 16.0 1 Subsidized 91 22.1 2.8 0.92 (0.70 - 1.21) 28 21.9 1.8 1.04 (0.71 - 1.54) 4 12.5 −3.7 0.60 (0.21 - 1.70) Uninsured 23 15.8 −3.5 0.67 (0.44 - 1.01) 7 10.5 −9.6 0.52 (0.26 - 1.05) 3 16.7 0.7 0.97 (0.37 - 2.57) Glycemic control (adults) Contributory + special 230 58.4 1 188 50.4 1 208 61.4 1 Subsidized 152 46.9 −11.5 0.80 (0.69 - 0.92) 57 50.0 −0.4 0.90 (0.74 - 1.10) 14 48.3 −13.1 0.83 (0.57 - 1.23) Uninsured 33 29.0 −29.4 0.61 (0.45 - 0.82) 19 31.7 −18.7 0.70 (0.48 - 1.04) 2 11.8 −49.6 0.23 (0.06 - 0.82) Caries prevention (adults) Contributory + special 149 37.8 1 149 40.0 1 139 41.0 1 Subsidized 108 33.3 −4.5 0.83 (0.67 - 1.03) 51 44.7 4.7 1.13 (0.89 - 1.44) 13 44.8 3.8 1.06 (0.68 - 1.64) Uninsured 32 28.1 −9.7 0.70 (0.51 - 0.97) 22 36.7 −3.3 0.84 (0.58 - 1.20) 5 29.4 −11.6 0.64 (0.30 - 1.33) Breast cancer prevention (women) Contributory + special 88 31.3 1 86 33.6 87 40.5 1 Subsidized 32 12.8 −18.5 0.46 (0.31 - 0.66) 25 31.3 −2.3 0.82 (0.57 - 1.16) 6 26.1 −14.4 0.68 (0.35 - 1.34) Uninsured 9 10.5 −20.8 0.48 (0.26 - 0.88) 5 12.8 −20.8 0.42 (0.18 - 1.00) 1 8.3 −32.2 0.28 (0.04 - 2.03) Prev, Prevalence; PD, Prevalence Difference; PR, Prevalence ratio; CI, Confidence Interval; MW, minimum wage; SGSSS, General System of Social Security in Health. Sample characteristics aPrevalence Ratio and 95% confidence interval adjusted for sex, age, race/color, chronic condition, self-rated health status. Statistically significant results are shown in bold. Table 4 Prevalence, prevalence difference and prevalence ratios (95% confidence intervals) of health services utilization in the last three months and of preventive activities in the last year by SGSSS insurance scheme, according to per capita income, in the study areas of Colombia (2011) Table 4 Prevalence, prevalence difference and prevalence ratios (95% confidence intervals) of health services utilization in the last three months and of preventive activities in the last year by SGSSS insurance scheme, according to per capita income, in the study areas of Colombia (2011) Table 4 Prevalence, prevalence difference and prevalence ratios (95% confidence intervals) of health services utilization in the last three months and of preventive activities in the last year by SGSSS insurance scheme, according to per capita income, in the study areas of Colombia (2011) Prev, Prevalence; PD, Prevalence Difference; PR, Prevalence ratio; CI, Confidence Interval; MW, minimum wage; SGSSS, General System of Social Security in Health. aPrevalence Ratio and 95% confidence interval adjusted for sex, age, race/color, chronic condition, self-rated health status. Statistically significant results are shown in bold. services is particularly marked between those without pri- vate insurance (16.1%) and those with (36.2%) (Table 3). There are noticeable inequities in preventive activities in both countries. The absolute differences in utilization prevalence (of all three socioeconomic variables) are al- ways in favor of those who are most socioeconomically favored. In Colombia, uninsured individuals are less likely to use the three preventive services than those of the contributory scheme. Moreover, individuals in the subsidized scheme are less likely to have glycemic con- trols and mammographies than individuals in the con- tributory scheme. The inequities are accentuated in breast cancer prevention according to per capita income: the prevalence of utilization increases as the level of in- come rises (21.9%, 30.0% and 37.2%). In Brazil, people with a lower income and without private insurance are less likely to use the three preventive services, and the difference in prevalence of using caries prevention What are the effects of type of SGSSS insurance scheme and private health plan at each level of income? This second section looks closely at the effects in each income stratum of insurance scheme in Colombia and holding a private health plan in Brazil on the utilization of healthcare services, after having observed an associ- ation between higher socioeconomic status and greater use of the health services. In Colombia, in the two lower-income strata, individuals of the contributory scheme are more likely to use the primary and out- patient secondary care services than the uninsured; and they are also more likely to use outpatient secondary care than individuals in the subsidized scheme (PR = Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Garcia-Subirats et al. What are the effects of income on each health subsystem? 0.70, CI 95% 0.53-0.94 in the lower per capita income stratum and PR = 0.59, CI 95%: 0.38-0.91 in the middle stratum). Inequalities between schemes were not found in the case of emergency care (Table 4). The effect of the insurance scheme on the use of preventive services is particularly noticeable in the lowest per capita income stratum, where belonging to the subsidized scheme or being uninsured significantly reduces the probability of using glycemic control services (prevalence difference between individuals in the contributory scheme and in the subsidized scheme is −11.5, and between individuals in the contributory scheme and the uninsured is −29.4) and mammography services (prevalence difference is −18.5 for the subsidized enrollees and −20.8 for the uninsured) (Table 4). 0.70, CI 95% 0.53-0.94 in the lower per capita income stratum and PR = 0.59, CI 95%: 0.38-0.91 in the middle stratum). Inequalities between schemes were not found in the case of emergency care (Table 4). The effect of the insurance scheme on the use of preventive services is particularly noticeable in the lowest per capita income stratum, where belonging to the subsidized scheme or being uninsured significantly reduces the probability of using glycemic control services (prevalence difference between individuals in the contributory scheme and in the subsidized scheme is −11.5, and between individuals in the contributory scheme and the uninsured is −29.4) and mammography services (prevalence difference is −18.5 for the subsidized enrollees and −20.8 for the uninsured) (Table 4). Lastly, the presence of inequities was evaluated accor- ding to insurance scheme in Colombia and type of coverage in Brazil (Table 6). In Colombia, the prevalence values for services utilization are consistently higher in the contributory scheme than in the subsidized scheme in all income strata, with the exception of emergency care and caries prevention. In the contributory scheme, level of income does not reveal inequity in health ser- vices utilization, except in the cases of outpatient sec- ondary care and breast cancer prevention, which are used more frequently among higher income individuals. What are the effects of type of SGSSS insurance scheme and private health plan at each level of income? Table 5 Prevalence, prevalence difference and prevalence ratios (95% confidence intervals) of health services utilization in the last three months and of preventive activities in the last year by private health plan (PHP), according to per capita income, in the study areas of Brazil (2011) Table 5 Prevalence, prevalence difference and prevalence ratios (95% confidence intervals) of health services utilization in the last three months and of preventive activities in the last year by private health plan (PHP), according to per capita income, in the study areas of Brazil (2011) Table 5 Prevalence, prevalence difference and prevalence ratios (95% confidence intervals) of health services utilization in the last three months and of preventive activities in the last year by private health plan (PHP), according to per capita income, in the study areas of Brazil (2011) What are the effects of type of SGSSS insurance scheme and private health plan at each level of income? International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Page 9 of 15 Page 9 of 15 Table 5 Prevalence, prevalence difference and prevalence ratios (95% confidence intervals) of health services utilization in the last three months and of preventive activities in the last year by private health plan (PHP), according to per capita income, in the study areas of Brazil (2011) < ½ MW (per capita) ½ - 1 MW (per capita) > 1 MW (per capita) n Prev PD PR (CI 95%)a n Prev PD PR (CI 95%)a n Prev PD PR (CI 95%)a Primary care Without PHP 506 49.3 1 247 45.2 1 47 32.0 1 With PHP 44 46.8 −2.5 0.99 (0.79 - 1.25) 63 36.4 −8.8 0.79 (0.64 - 0.99) 51 30.5 −1.5 0.95 (0.68 - 1.33) Outpatient secondary care Without PHP 212 20.6 1 141 25.8 1 38 25.9 1 With PHP 16 17.0 −3.6 0.98 (0.64 - 1.52) 59 34.1 8.3 1.57 (1.23 - 2.00) 58 34.7 8.8 1.40 (1.02 - 1.93) Emergency care Without PHP 348 33.9 1 194 35.5 1 37 25.2 1 With PHP 42 44.7 10.8 1.35 (1.05 - 1.72) 46 26.6 −8.9 0.66 (0.50 - 0.86) 37 22.2 −3.0 1.03 (0.68 - 1.55) Glycemic control (adults) Without PHP 391 52.1 1 268 57.0 1 67 50.0 1 With PHP 36 58.1 6.0 1.16 (0.93 - 1.45) 78 66.1 9.1 1.22 (1.06 - 1.42) 93 67.4 17.4 1.40 (1.15 - 1.71) Caries prevention (adults) Without PHP 107 14.3 1 88 18.7 1 23 17.2 1 With PHP 16 25.8 11.5 1.80 (1.15 - 2.82) 41 34.8 16.1 1.58 (1.17 - 2.12) 58 42.0 24.8 2.42 (1.59 - 3.68) Breast cancer prevention (women) Without PHP 149 25.3 1 100 27.4 1 27 31.0 1 With PHP 19 38.0 12.7 1.61 (1.09 - 2.36) 28 29.8 2.4 1.15 (0.81 - 1.63) 42 42.9 11.9 1.48 (1.02 - 2.15) Prev, Prevalence; PD, Prevalence Difference; PR, Prevalence ratio; CI, Confidence Interval; MW, minimum wage; PHP, Private Health Plan. aPrevalence Ratio and 95% confidence interval adjusted for sex, age, race/color, chronic condition, self-rated health status. Statistically significant results are shown in bold. What are the effects of income on each health subsystem? In the subsidized scheme, no inequity was detected ac- cording to level of income in use of the curative services, but it was observed in the preventive services such as caries prevention (PR comparing those in the middle in- come stratum with those in the lower is 1.45, CI 95%: 1.12 - 1.87) and breast cancer prevention (PR = 2.00, CI 95%: 1.28 - 3.11) (Table 6). This analysis could not be performed in the uninsured group due to its low level of use of the health services. In the case of Brazil, the effect of private insurance varies depending on care level and income. It is most influential in the middle-income stratum, where individuals with a pri- vate health plan are less likely to use primary and emer- gency care services and more likely to use outpatient secondary care: the PR of those with insurance to those without is 1.57 (CI 95% 1.23-2.00) (Table 5). Having a pri- vate health plan implies a greater likelihood of using the three preventive services in all income strata (Table 5). ce r me olom ) ) ) ) ) ) ) ) ) ) ce ra me a olomb 2 7 1 6 1 ) 9 2 1 5 1 1 5 1 What are the effects of income on each health subsystem? In Brazil, higher-income individuals are less likely to use the primary care level whether they hold a private e ratios (95% confidence intervals) of health services utilization in the last three months and of e according to SGSSS insurance scheme and private health plan, in the study areas of Colombia and ombia Brazil Subsidized scheme Without private health plan With private health plan n Prev PD PR (CI 95%)a n Prev PD PR (CI 95%)a n Prev PD PR (CI 95%)a 210 51.0 1 506 49.3 1 44 46.8 1 70 54.7 3.7 1.08 (0.90 - 1.29) 247 45.2 −4.1 0.91 (0.81 - 1.02) 63 36.4 −10.4 0.75 (0.56 - 1.02) 18 56.3 5.3 1.07 (0.78 - 1.48) 47 32.0 −17.3 0.69 (0.54 - 0.88) 51 30.5 −16.3 0.66 (0.48 - 0.93) 65 15.8 1 212 20.6 1 16 17.0 1 19 14.8 −1.0 0.87 (0.55 - 1.39) 141 25.8 5.2 1.17 (0.98 - 1.41) 59 34.1 17.1 1.94 (1.22 - 3.10) 9 28.1 12.3 1.72 (0.98 - 3.02) 38 25.9 5.3 1.34 (1.00 - 1.79) 58 34.7 17.7 1.98 (1.23 - 3.20) 91 22.1 1 348 33.9 1 42 44.7 1 28 21.9 −0.2 1.19 (0.82 - 1.73) 194 35.5 1.6 1.19 (1.04 - 1.38) 46 26.6 −18.1 0.63 (0.45 - 0.87) 4 12.5 −9.6 0.72 (0.29 - 1.75) 37 25.2 −8.7 0.90 (0.67 - 1.22) 37 22.2 −22.5 0.70 (0.48 - 1.01) 152 46.9 1 391 52.1 1 36 58.1 1 57 50.0 3.1 1.07 (0.87 - 1.31) 268 57.0 4.9 1.11 (1.00 - 1.23) 78 66.1 8.0 1.15 (0.90 - 1.47) 14 48.3 1.4 1.03 (0.71 - 1.51) 67 50.0 −2.1 1.05 (0.87 - 1.26) 93 67.4 9.3 1.21 (0.95 - 1.53) 108 33.3 1 107 14.3 1 16 25.8 1 51 44.7 11.4 1.45 (1.12 - 1.87) 88 18.7 4.4 1.37 (1.06 - 1.76) 41 34.8 9.0 1.34 (0.83 - 2.17) 13 44.8 11.5 1.45 (0.97 - 2.16) 23 17.2 2.9 1.11 (0.73 - 1.69) 58 42.0 16.2 1.87 (1.16 - 3.01) 32 12.8 1 149 25.3 1 19 38.0 1 25 31.3 18.5 2.00 (1.28 - 3.11) 100 27.4 2.1 1.20 (0.97 - 1.47) 28 29.8 −8.2 0.77 (0.48 - 1.23) 6 26.1 13.3 1.69 (0.82 - 3.45) 27 31.0 5.7 1.50 (1.05 - 2.14) 42 42.9 4.9 1.10 (0.73 - 1.68) e Interval; MW, minimum wage. r, chronic condition, self-rated health status. enc com Co %)a .17) .16) .29) .58) .45) .29) .05) .25) .22) .27) What are the effects of income on each health subsystem? d %)a .02) 0.93) 3.10) 3.20) 0.87) .01) .47) .53) 2.17) 3.01) .23) .68) Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 Page 11 of 15 http://www.equityhealthj.com/content/13/1/10 Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Page 11 of 15 28.4 27.4 19.9 37.7 55.8 50.8 32.7 18.7 24.3 19.8 44.5 52.2 0 20 40 60 100 Primary care Outpatient secondary care Emergency care Glycemic control (adults) Caries prevention (adults) Breast cancer prevention (women) Brazil (Recife and Caruaru) (n=2,155) 28.4 27.4 19.9 37.7 55.8 50.8 32.7 18.7 24.3 19.8 44.5 52.2 80 Colombia (Soacha and Kennedy) (n = 2,163) % Figure 1 Prevalence of health services utilization in the last three months and of preventive care activities in the last year, study areas of Colombia and Brazil (2011). Figure 1 Prevalence of health services utilization in the last three months and of preventive care activities in the last year, study areas of Colombia and Brazil (2011). of health services utilization in the last three months and of preventive care activities in the last year, study areas azil (2011). comparative articles [41]. It has therefore allowed us to compare the inequities in access in two countries with different health systems, as well as to fill a current gap in the literature with regard to inequities in the utilization of different care levels and preventive services. health plan or not. In outpatient secondary care there is an income gradient in individuals with double coverage: the higher the income, the higher the probability of utilization (PR = 1.94, CI 95%: 1.22- 3.10 and PR = 1.98, CI 95%; 1.23 - 3.20) (Table 6). The prevalence of use of preventive services are higher in the group with double coverage (SUS and PHP), and the relationship with in- come is significant in the case of caries prevention (in both groups), as well as in mammography in the group without private insurance (Table 6). What are the effects of income on each health subsystem? The inequities found would have been greater if the general population had been the object of the study: firstly, due to the fact that the study was also oriented to analyzing barriers in access to the health services of the General System of Social Security in Health (SGSSS) and the Unified Health System (SUS) [62], so areas were se- lected in which the use of these was predominant over the use of private health services, in other words, areas in which there is a concentration of the low to middle- low income population; and secondly, because the study population was made up of individuals with some kind of health problem or a perceived healthcare need in the three months prior to the interview, and the perception of need for care is greater in the groups with the lowest socioeconomic status than in the general population. Discussion T Twenty years after the reforms implemented to improve equity in access to health care, in both countries inequi- ties persist, in terms of an unequal use for equal need. Although caution should be exercised when drawing generalizations from its results, this study, conducted in two areas in the northeast of Brazil and two in central Colombia, reveals the presence of inequities according to per capita income, possession of a private health plan and SGSSS insurance scheme in the study areas ana- lyzed, where a better socioeconomic status favors utilization of the health services. (In)equity in healthcare services utilization in Colombia (In)equity in healthcare services utilization in Colombia One of the successes attributed to the Colombian re- forms is an increase in the population enrolled in the SGSSS, as this was considered to facilitate access and use of the health services [7,27,38]. Although the unin- sured population fell from 41.5% in 2000 to 12.1% in 2010 [15], there is a lack of consensus in the literature on the impact this has had on equity in utilization [39]. Our results coincide with those studies in which authors show that enrollment in the SGSSS improves access to the health services [24,27,42], to the extent that uninsured individuals have practically no access. However, this study reveals that inequities also exist between the insurance schemes. Individuals of the subsidized scheme, with equal need, make the least use of primary, secondary and In the Colombian areas inequities are found in favor of individuals of a higher socioeconomic status in the three care levels and in the use of preventive services; whereas in the Brazilian areas, the use of secondary ser- vices and preventive activities is unequal in favor of the higher-income population, and lower-income individuals make greater use of the primary care services. In terms of the strengths of the study, we should high- light that this is a comparative analysis based on primary data and with a common questionnaire in which the questions and the recall period are identical, thus avoid- ing the methodological limitations of international stud- ies based on secondary data which have arisen in other Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Page 12 of 15 At the primary care level there is inequality in favor of the poorest (i.e. it is pro-poor), meaning that lower- income individuals make greater use of this level, which is the direct opposite of the situation in secondary care. This phenomenon has been described in previous research conducted in other countries with a national health system [8,32,33,65]. (In)equity in healthcare services utilization in Colombia Inequity in health care utilization is concen- trated at the secondary level, which might be due to the fact that higher-income individuals access private secondary care services directly (via a private health plan or out- of-pocket payment) instead of accessing secondary care through the SUS [66], in order to avoid the significant ac- cess barrier of long waiting times in the SUS [62,67]. The lack of specific studies differentiating between these two levels of care in Brazil hinders the comparison of results. preventive care. This result is in contrast with those of other authors who conclude that enrollment guarantees ac- cess [37,38], basing this conclusion on a utilization com- parison between individuals of the subsidized scheme and the uninsured, without mentioning the differences between the two schemes. Furthermore these inequities are more pronounced in secondary care services, results which agree with the only study available that differentiates between the two care levels [7]. In emergency care, inequities between the two insu- rance schemes were not found, but they were found be- tween the insured and uninsured, which appears to indicate that the legal obligation of all public and private entities to offer this type of care to everyone is not ful- filled [14], because it depends on the individual’s insu- rance status. In emergency care no inequities were observed, al- though the utilization prevalence shows that use falls as per capita income rises and may signal a higher level of use among the poorest, which could be the consequence of the presence of access barriers at the primary care levels, such as the lack of doctors or low levels of health problem resolution reported in other studies [62]. It is interesting to note that the role of the insurance scheme is more evident among lower-income indivi- duals, as in these strata it acts as a source of inequity in the utilization of preventive and curative services (with the exception of emergency care), whilst in the higher- income population no differences according to insurance scheme were found in the utilization of services. More- over, no inequities according to income were found in individuals belonging to the same scheme. In other words, health services utilization by the poorest popula- tion – the one which has the greatest need for it – is determined by insurance scheme and not by need. (In)equity in preventive activities in both countries f d (In)equity in preventive activities in both countries In terms of preventive activities, our study once again reveals inequities in both countries that benefit individ- uals with a higher socioeconomic status. Only a few studies in Colombia on access to preventive activities analyze specific activities, and those assess inequalities in access to mammography [68] or cervical cytology [69], which hinders their comparison with our results. De Charry et al. pointed out that women in the contributory scheme reported higher levels of mammography services utilization [68]. In Brazil, some studies identify lower odds of having had a mammography among women with a low educational level, low family income and no pri- vate health plan [70,71]. (In)equity in healthcare services utilization in Colombia Furthermore, the results show that people with a pri- vate health plan have more access to all levels of care than those without, thus the private health plan facili- tates access to services [65], but nonetheless income in- equalities persist within this group. This may be a result of the existing inequalities in relation to the coverage of- fered in private health plans, since higher-income indi- viduals pay for more expensive plans, which offer better coverage in terms of complementary tests and medical exams [64]. This study corroborates the argument that the design of the SGSSS displays a structural inequity [1,39], i.e. in- dividuals under the subsidized scheme use the health services less than those under the contributory scheme because the benefits package offered is smaller. Thus the principle of horizontal equity is violated by the design of the health system itself, as it determines that the use of health services depends on the type of SGSSS scheme the individual is enrolled in instead of on the individual’s health needs. It will be necessary to analyze what has oc- curred since the new law came into effect to make the benefits package equal in the two schemes [63]. Author details 1 1Health Policy and Health Services Research Group, Health Policy Research Unit, Consortium for Health Care and Social Services of Catalonia, Avenida Tibidabo, 21, Barcelona 08022, Spain. 2Ph. D. Programme in Biomedicine, Department of Experimental and Health Sciences, Universitat Pompeu Fabra, Barcelona, Spain. 3Escuela de Medicina y Ciencias de la Salud, Universidad del Rosario, Calle 14, Número 6-25, Bogotá, Colombia. 4The Prince Leopold Institute of Tropical Medicine, Nationalestraat 15, Antwerpen, Belgium. 5FIOCRUZ/PE, Brazil, Universidade de Pernambuco, Av. Agamenon (In)equity in health care services utilization in Brazil (In)equity in health care services utilization in Brazil In the case of Brazil, this study is consistent with previ- ous research in terms of the existence of inequities in ac- cess to the health services in spite of the fact that it has a national health system [25,26,64], and it also contrib- utes new results: inequity in utilization varies according to care level and is more pronounced in outpatient sec- ondary care and preventive services. There are authors who point out that inequity in access has been reduced in recent years [5,6], although this statement should be treated with caution as they do not distinguish between care levels. Moreover, as our results in Brazil corroborated, in- equalities have been described in the preventive use of dental services in the south of the country [72] and other authors reveal inequalities in the use of dental ser- vices without differentiating between curative and pre- ventive visits [5]. Furthermore, studies in both countries that do not distinguish between the types of activity Page 13 of 15 Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 identify that the higher the individual’s socioeconomic level, the higher the probability that the medical visit was for preventive rather than curative reasons [7,28,35]. This cycle, in which the poor (lower-income, subsidized scheme and uninsured individuals in Colombia, and those without a private health plan in Brazil) have worse health and worse access to curative health care and preventive activities, makes them particularly vulnerable and may be related to worse health outcomes. Competing interests The authors declare that they have no competing interests. Authors’ contributions IV and MLV were responsible for the study design and the supervision of all research phases. IG, IV, ASM, MRF and MLV were in charge of the supervision of the fieldwork. IG supervised the data entry, was responsible for the data cleaning, statistical analysis and its interpretation, and drafted the text. IV, CB and MLV contributed to the data analysis and writing of the article. ASM, PdP, MRF, JPU and CB participated in data interpretation. All authors reviewed and approved the final version of the article. The authors alone are responsible for the content of this paper. Funding h g The research leading to these results received funding from the European Union’s Seventh Framework Program (FP7/2007-2013) under grant agreement n° 223123. Moreover, a grant from the Ministry of Foreign Affairs and Cooperation for Development (Spain) was awarded to Irene Garcia- Subirats for the development of the fieldwork (Beca MAEC-AECID 2010–2011 and 2011–2012). The funding sources had no involvement in study design, nor in the collection, analysis and interpretation of the data, nor in writing the article and the decision to submit it for publication. Acknowledgements h h The authors are most grateful to the people that participated in the study and generously gave their time. We thank Kate Bartlett for her help in correcting the English version of this article. This article forms part of the doctoral dissertation of Irene Garcia-Subirats at the Universitat Pompeu Fabra of Barcelona. Finally, we should bear in mind that the evaluation of equity in access based on the use of services has been widely criticized for ambiguities in its interpretation: it is difficult to differentiate between necessary use and ex- cessive or insufficient use, both in terms of quantity and quality of care received [73]. In this study, quality of care has not been assessed, although in both countries some results suggest that care quality varies according to socio- economic level. In Colombia, fewer subsidized scheme enrollees reported that their health problem was solved than contributory scheme enrollees. In Brazil, higher- income individuals reported higher levels of positive outcomes [62]. As there is insufficient data to serve as evidence, further evaluation of equity in access to health services is required, focusing not only on utilization of services but also on quality of care. Conclusions In Colombia and Brazil the health systems have not achieved the equity in utilization prescribed in their laws. Inequity in the utilization of services varies accor- ding to care level. It is most noticeable at the secondary care level; only the use of emergency care in Brazil does not show inequities by socioeconomic status. The in- equities found in both countries illustrate the close relationship between health services utilization and the design of each health system. Twenty years after the reforms, Colombia, a managed competition model with two insurance schemes, has managed to increase SGSSS enrollment but the benefits packages are still different and remain tied to the purchasing power of the population, which creates conditions of unequal access for equal need. In Brazil, despite universal coverage under the SUS there are still significant in- equities in access, especially with regard to outpatient secondary care. The existence of a health system formed by the SUS, steered by principles of equity and universal coverage, together with the private sector, driven by liberalist principles, generates a pattern of inequality in access to the continuum of care. In both countries inequity in the preventive services also points to inequalities at the primary care level. Magalhães, S/N, Recife, Brazil. 6Agència de Salut Pública de Barcelona, Plaça Lesseps 1, Barcelona 08023, Spain. 7Universitat Pompeu Fabra, Barcelona, Spain. 8Ciber of Epidemiology and Public Health (CIBERESP), Madrid, Spain. Received: 2 August 2013 Accepted: 24 January 2014 Published: 31 January 2014 Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Rev Panam Salud Publica 2005, 17(3):210–220. 25. Travassos C, Viacava F, Pinheiro R, Brito A: Utilization of health care services in Brazil: gender, family characteristics, and social status. Rev Panam Salud Publica 2002, 11(5–6):365–373. 48. Levino A, Carvalho EF: Comparative analysis of health systems on the triple border between Brazil, Colombia, and Peru. Rev Panam Salud Publica 2011, 30(5):490–500. 26. Cataife G, Courtemanche CJ: Is universal health care in Brazil really universal? 2011. http://papers.ssrn.com/sol3/papers.cfm? abstract_id=1690662. 49. Almeida CM, Travassos C, Porto S, Labra ME: Health sector reform in Brazil: a case study of inequity. Int J Health Serv 2000, 30(1):129–162. 27. Flórez C, Soto V, Acosta O, Karl C, Misas J, Forero N, Lopera C: Avances y desafíos de la equidad en el sistema de salud colombiano. Bogotá: Fundación Corona; 2007. 50. Wallace SP, Gutierrez VF: Equity of access to health care for older adults in four major Latin American cities. Rev Panam Salud Publica 2005, 17(5–6):394–409. 28. Mejía-Mejía A, Sanchez-Gandur AF, Tamayo-Ramirez JC: Equity in access to health-services in Antioquia, Colombia. Rev Salud Publica (Bogota) 2007, 9(1):26–38. 51. Esteves RJF: The quest for equity in Latin America: a comparative analysis of the health care reforms in Brazil and Colombia. Int J Equity Health 2012, 11(6). doi:10.1186/1475-9276-11-6. 29. da Costa JS D, Reis MC, Silveira Filho CV, Linhares RS, Piccinini F: Prevalence of medical visits and associated factors, Pelotas, Southern Brazil, 1999–2000. Rev Saude Publica 2008, 42(6):1074–1084. 52. Vázquez ML, da Silva MR F, Unger JP, Mogollón AS, Vargas I, De Paepe P: Impact on equity of access and efficiency of Integrated Health care Networks (IHN) in Colombia and Brazil. Equity-LA project id: 223123 (2008–2013). 2013. http://www2.equity-la.eu/publications/technical-reports?set_language=en. 30. Giatti L, Barreto SM: Labor market status and the use of healthcare services in Brazil. Cien Saude Colet 2011, 16(9):3817–3827. 31. Viacava F, Souza-Junior PR, Szwarcwald CL: Coverage of the Brazilian popu- lation 18 years and older by private health plans: an analysis of data from the World Health Survey. Cad Saude Publica 2005, 21(Suppl):119–128. 53. Aday LA, Andersen R: A framework for the study of access to medical care. Health Serv Res 1974, 9(3):208–220. 54. Vargas I, Vázquez ML, Mogollón AS, Unger JP: Barriers of access to care in a managed competition model: lessons from Colombia. BMC Health Serv Res 2010, 10:297. 32. Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 10. The World Bank Group: Data Brazil. 2013. http://data.worldbank.org/ country/brazil. 36. Díaz-Grajales C, Zapata-Bermúdez Y, Aristizábal-Grisales JC: Access to preventive services in contributive and subsidized regimes of health in a (slums) stratum two neighoborhood of the city of Cali. Rev Geren Polit Salud 2011, 12(21):153–175. 11. The World Bank Group: How we Classify Countries. 2013. http://data. worldbank.org/about/country-classifications/country-and-lending-groups. 37. Ruiz F, Amaya L, Venegas S: Progressive segmented health insurance: Colombian health reform and access to health services. Health Econ 2007, 16(1):3–18. 12. Bárcena A, Prado A, Hopenhayn M, Beccaria L, Jaspers-Faijer D, Quarless D, Pérez R: Panorama Social de América Latina. Santiago de Chile: United Nations Publications; 2012. 38. Giedion U, Díaz BY, Alfonso EA, Savedoff WD: The Impact of Subsidized Health Insurance on Health Status and on Access to and Use of Health Services. In From few to many: Ten Years of Health Insurance Expansion in Colombia. Edited by Glassman A, Giuffrida A, Escobar ML, Giedion U. Washington, D.C: Inter-American Development Bank and the Brookings Institutions; 2009. 13. Organización Panamericana de la Salud: Información y Análisis de Salud: Situación de Salud en las Américas: Indicadores Básicos 2013. Washington, D.C: OPS; 2013. 14. República de Colombia: Ley 100/1993, de 23 de diciembre, por la cual se crea el Sistema de Seguridad Social Integral y se dictan otras disposiciones. Diario Oficial n. 41. 23/12/1993. 15. Profamilia: Encuesta Nacional de Demografía y Salud-ENDS 2010. Bogotá: Profamilia; 2010. 39. Yepes FJ, Ramírez M, Sánchez LH, Ramírez ML, Jaramillo I: Luces y sombras de la reforma de la salud en Colombia. Bogotá: Centro internacional de investigaciones para el desarrollo; 2010. 16. Morales LG: El financiamiento del sistema de seguridad social en salud en Colombia. Santiago de Chile: United Nations Publication; 1997. 40. Travassos C, de Oliveira EXG, Viacava F: Geographic and social inequalities in the access to health services in Brazil: 1998 and 2003. Cien Saude Colet 2006, 11(4):975–986. 17. Presidência da República: Constituição da República Federativa do Brasil. Brasilia; 1988. 41. Dachs JN, Ferrer M, Florez CE, Barros AJ, Narvaez R, Valdivia M: Inequalities in health in Latin America and the Caribbean: descriptive and exploratory results for self-reported health problems and health care in twelve countries. Rev Panam Salud Publica 2002, 11(5–6):335–355. 18. Mesa-Lago C: O sistema de saúde brasileiro: seu impacto na pobreza e na desigualdade. Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 ABACO Revista de Cultura y Ciencias Sociales 2007, 41:115–131. 19. Hernandez M: Health reform, equity and the right to health in Colombia. Cad Saude Publica 2002, 18(4):991–1001. 42. Rubio-Mendoza ML: Equity in gaining access to health services and equity in funding being attended by them in Bogota. Rev Salud Publica (Bogota) 2008, 10(Suppl):29–43. 20. Ocke-Reis CO, Marmor TR: The Brazilian national health system: an unfulfilled promise? Int J Health Plann Manage 2010, 25(4):318–329. 43. Capilheira MF, da Silva Dos Santos I: Individual factors associated with medical consultation by adults. Rev Saude Publica 2006, 40(3):436–443. 21. Sousa MF: The family health program in Brazil: analysis of access to basic care. Rev Bras Enferm 2008, 61(2):153–158. 21. Sousa MF: The family health program in Brazil: analysis of access to basic care. Rev Bras Enferm 2008, 61(2):153–158. 22. Restrepo JH, Zambrano A, Vélez M, Ramírez M: Health insurance as a strategy for access: streamlined facts of the Colombian Health Care Reform. Bogotá: Universidad del Rosario; 2007. 22. Restrepo JH, Zambrano A, Vélez M, Ramírez M: Health insurance as a strategy for access: streamlined facts of the Colombian Health Care Reform. Bogotá: Universidad del Rosario; 2007. 44. da Costa JS D, Presser AD, Zanolla AF, Ferreira DG, Perozzo G, Freitas IB, Portolan LT, Tavares RT, Olinto MT, Pattussi MP: Use of outpatient health services by women: a population-based study in southern Brazil. Cad Saude Publica 2008, 24(12):2843–2851. 23. Paim J, Travassos C, Almeida CM, Bahia L, Macinko J: The Brazilian health system: history, advances, and challenges. Lancet 2011, 377(9779):1778–1797. 45. Bastos GA, Duca GF, Hallal PC, Santos IS: Utilization of medical services in the public health system in the Southern Brazil. Rev Saude Publica 2011, 45(3):475-54. 24. Céspedes JE, Jaramillo-Pérez I, Martínez R, Olaya S, Reynales J, Uribe C, Castaño R, Garzón E, Almeida C, Travassos C, et al: Efectos de la reforma de la Seguridad Social en Salud en Colombia sobre la equidad en el acceso y la utilización de servicios de salud. Rev Salud Publica (Bogota) 2000, 2(2):145–164. 46. Pessoto UC, Heimann LS, Boaretto RC, Castro IE, Kayano J, Ibanhes LC, Junqueira V, da Rocha JL, Barboza R, Cortizo CT, et al: Health care services utilization and access inequalities in the Sao Paulo Metropolitan Region. Cien Saude Colet 2007, 12(2):351–362. 47. Homedes N, Ugalde A: Neoliberal reforms in health services in Latin America: a critical view from two case studies. References 1. Vargas I, Vázquez ML, Jane E: Equity and health systems reform in Latin America. Cad Saude Publica 2002, 18(4):927–937. 1. Vargas I, Vázquez ML, Jane E: Equity and health systems reform in Latin America. Cad Saude Publica 2002, 18(4):927–937. 1. Vargas I, Vázquez ML, Jane E: Equity and health systems reform in Latin America. Cad Saude Publica 2002, 18(4):927–937. 2. Culyer AJ, Wagstaff A: Equity and equality in health and health care. J Health Econ 1993, 12(4):431–457. 2. Culyer AJ, Wagstaff A: Equity and equality in health and health care. J Health Econ 1993, 12(4):431–457. 3. van Doorslaer E, Wagstaff A, van der Burg H, Christiansen T, Citoni G, Di Biase R, Gerdtham UG, Gerfin M, Gross L, Häkinnen U, et al: The redistributive effect of health care finance in twelve OECD countries. J Health Econ 1999, 18(3):291–313. 4. Suárez-Berenguela RM: Health System Inequalities and Inequities in Latin America and the Caribbean: Findings and policy implications. Working document prepared for the Health and Human Development Division of the Pan American Health Organization-World Health Organization. Geneva: WHO; 2000. 5. Macinko J, Lima Costa MF: Horizontal equity in health care utilization in Brazil, 1998–2008. Int J Equity Health 2012, 11:33. 6. Almeida G, Sarti FM, Ferreira FF, Diaz MD, Campino AC: Analysis of the evolution and determinants of income-related inequalities in the Brazilian health system, 1998–2008. Rev Panam Salud Publica 2013, 33(2):90–97. 7. Ruiz Gómez F, Zapata Jaramillo T, Garavito Beltran L: Colombian health care system: results on equity for five health dimensions 2003–2008 , q y , 6. Almeida G, Sarti FM, Ferreira FF, Diaz MD, Campino AC: Analysis of the evolution and determinants of income-related inequalities in the Brazilian health system, 1998–2008. Rev Panam Salud Publica 2013, 33(2):90–97. 7. Ruiz Gómez F, Zapata Jaramillo T, Garavito Beltran L: Colombian health care system: results on equity for five health dimensions, 2003–2008. Rev Panam Salud Publica 2013, 33(2):107–115. 8. Hanratty B, Zhang T, Whitehead M: How close have universal health systems come to achieving equity in use of curative services? A systematic review. Int J Health Serv 2007, 37(1):89–109. 9. The World Bank Group: Data Colombia. 2013. http://data.worldbank.org/ country/colombia. Page 14 of 15 Page 14 of 15 Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Garrido-Cumbrera M, Borrell C, Palencia L, Espelt A, Rodriguez-Sanz M, Pasarin MI, Kunst A: Social class inequalities in the utilization of health care and preventive services in Spain, a country with a national health system. Int J Health Serv 2010, 40(3):525–542. 55. Vargas I, Unger JP, Mogollón AS, Vázquez ML: Effects of managed care mechanisms on access to healthcare: results from a qualitative study in Colombia. Int J Health Plann Manage 2013, 28(1):e13–e33. 33. Palencia L, Espelt A, Rodriguez-Sanz M, Rocha KB, Pasarin MI, Borrell C: Trends in social class inequalities in the use of health care services within the Spanish National Health System, 1993–2006. Eur J Health Econ 2013, 14(2):211–219. 56. Vanderlei LCM: Mortalidade infantil evitável e acesso à atenção à saúde no Recife, Brasil na perspectiva dos principais atores sociais, PhD thesis. Barcelona: Universitat Autònoma de Barcelona; 2010. 34. van Doorslaer E, Masseria C, Koolman X: Inequalities in access to medical care by income in developed countries. CMAJ 2006, 174(2):177–183. 57. Vargas I: Barreras en el acceso a la atención en salud en modelos de competencia gestionada: un estudio de caso en Colombia, PhD thesis. Barcelona: Universitat Autònoma de Barcelona; 2009. 57. Vargas I: Barreras en el acceso a la atención en salud en modelos de competencia gestionada: un estudio de caso en Colombia, PhD thesis. Barcelona: Universitat Autònoma de Barcelona; 2009. 35. Pinheiro RS, Viacava F, Travassos C, Brito A: Gender, morbidity, access and utilization of health services in Brazil. Cien Saude Colet 2002, 7(4):687–707. Page 15 of 15 Page 15 of 15 Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 58. Vanderlei LCM, Vázquez ML: Preventable infant mortality and barriers to access to primary care in Recife, Northeastern Brazil. Rev Saude Publica 2013, 47(2):379–389. 59. O’Halloran J, Miller GC, Britt H: Defining chronic conditions for primary care with ICPC-2. Fam Pract 2004, 21(4):381–386. care with ICPC-2. Fam Pract 2004, 21(4):381–386. 60. Kunst A: Cross-national comparisons of socio-economic differences in Mortality, PhD thesis. Rotterdam: Eramus University; 1997. 60. Kunst A: Cross-national comparisons of socio-econ PhD thesis. Rotterdam: Eramus University; 1997. PhD thesis. Rotterdam: Eramus University; 1997. 61. StataCorp: Stata Statistical Sorftware: Release 12. College Station, TX: StataCorp LP; 2011. 62. Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Garcia-Subirats I, Vargas I, Mogollón AS, De Paepe P, Silva MR, Unger JP, Vázquez ML: Barriers in access to healthcare in countries with different health systems. A study in municipalities of central Colombia and north-eastern Brazil. Soc Sci Med 2013. in press. 63. República de Colombia: Acuerdo por el cual se unifican los Planes Obligatorios de Salud de los Regímenes Contributivo y Subsidiado a nivel nacional. Acuerdo número 032: Comisión de Regulación en Salud; 2012. 64. Neri M, Soares W: Social inequality and health in Brazil. Cad Saude Publica 2002, 18(Suppl):77–87. 65. Borrell C, Palencia L, Rodriguez-Sanz M, Malmusi D, Bartoll X, Puigpinos R: Trends in social inequalities in health in Catalonia, Spain. Med Clin (Barc) 2011, 137(Suppl 2):60–65. ell C, Palencia L, Rodriguez-Sanz M, Malmusi D, Bartoll X, Puigpinos R: Trends in social inequalities in health in Catalonia, Spain. Med Clin (Barc) 2011, 137(Suppl 2):60–65. 66. Porto SM, Uga MA, Moreira RS: An analysis of use of the health services by financing system: Brazil 1998–2008. Cien Saude Colet 2011, 16(9):3795–3806. 67. Solla J, Chioro A: Atenção ambulatorial especializada. In Politicas e Sistema de Saúde no Brasil. Edited by Giovanella L, Escorel S, Costa Lobato LV, Noronha JC, de Carvalho I. Rio de Janeiro: Fiocruz Editora, CEBES; 2009:627–998. 68. Charry LC, Carrasquilla G, Roca S: Equity regarding early breast cancer screening according to health insurance status in Colombia. Rev Salud Publica (Bogota) 2008, 10(4):571–582. 69. Ivan Lucumi CD, Gomez Gutierrez LF: Accessibility to healthcare services in the recent cervical cytology performed in an urban area in Colombia. Rev Esp Salud Publica 2004, 78(3):367–377. 70. Lages RB, Oliveira GP, Simeão Filho VM, Nogueira FM, Teles JBM, Vieria SC: Inequalities associated with lack of mammography in Teresina-Piauí-Brazil, 2010–2011. Rev Bras Epidemiol 2012, 15(4):737–747. 71. de Oliveira EX, Pinheiro RS, Melo EC, Carvalho MS: Socioeconomic and geographic constraints to access mammography in Brasil, 2003–2008. Cien Saude Colet 2011, 16(9):3649–3664. 71. de Oliveira EX, Pinheiro RS, Melo EC, Carvalho MS: Socioeconomic and geographic constraints to access mammography in Brasil, 2003–2008. Cien Saude Colet 2011, 16(9):3649–3664. 72. Camargo MB, Dumith SC, Barros AJ: Regular use of dental care services by adults: patterns of utilization and types of services. Cad Saude Publica 2009, 25(9):1894–1906. 73. Hulka BS, Wheat JR: Patterns of utilization. The patient perspective. Med Care 1985, 23(5):438–460. Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 Garcia-Subirats et al. International Journal for Equity in Health 2014, 13:10 http://www.equityhealthj.com/content/13/1/10 doi:10.1186/1475-9276-13-10 Cite this article as: Garcia-Subirats et al.: Inequities in access to health care in different health systems: a study in municipalities of central Colombia and north-eastern Brazil. International Journal for Equity in Health 2014 13:10. Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission
https://openalex.org/W4388112160	https://aacr.figshare.com/articles/journal_contribution/Figure_S2_from_CBF-Beta_Mitigates_PI3K-Alpha_Specific_Inhibitor_Killing_through_PIM1_in_i_PIK3CA_i_-Mutant_Gastric_Cancer/24474148/1/files/43002745.pdf	Walloon	null	Figure S2 from CBF-Beta Mitigates PI3K-Alpha–Specific Inhibitor Killing through PIM1 in <i>PIK3CA</i>-Mutant Gastric Cancer	null	2,023	cc-by	64	B C A Difference Score Gene Rank Difference Score Gene Rank Difference Score Gene Rank 584-A2 KYSE510 CRC119 B C A Difference Score Gene Rank Difference Score Gene Rank Difference Score Gene Rank 584-A2 KYSE510 CRC119 B A Difference Score Difference Score 584-A2 KYSE510 A Difference Score KYSE510 B A KYSE510 Difference Score Difference Score Gene Rank C Difference Score Difference Score Gene Rank
https://openalex.org/W4281786774	https://www.nrel.gov/docs/fy22osti/82418.pdf	English	null	Synechocystis sp. PCC 6803 Requires the Bidirectional Hydrogenase to Metabolize Glucose and Arginine Under Oxic Conditions	Frontiers in microbiology	2,022	cc-by	11,021	Synechocystis sp. PCC 6803 Requires the Bidirectional Hydrogenase to Metabolize Glucose and Arginine Under Oxic Conditions Heinrich Burgstaller 1, Yingying Wang 1, Johanna Caliebe 1,2, Vanessa Hueren 1, Jens Appel 1,2 Marko Boehm 1,2 Sinje Leitzke 1 Marius Theune 1,2 Paul W King 3 and 1 Plant Cell Physiology and Biotechnology, Botanical Institute, University of Kiel, Kiel, Germany, 2 Molecular Plant Physiology, Bioenergetics in Photoautotrophs, University of Kassel, Kassel, Germany, 3 National Renewable Energy Laboratory, Biosciences Center, Golden, CO, United States Keywords: hydrogenase, diaphorase, photosynthetic complex I (NDH-1), photosynthesis, respiration, arginine, photomixotrophy ORIGINAL RESEARCH published: 31 May 2022 doi: 10.3389/fmicb.2022.896190 Correspondence: Kirstin Gutekunst kirstin.gutekunst@uni-kassel.de Specialty section: This article was submitted to Microbial Physiology and Metabolism, a section of the journal Frontiers in Microbiology Received: 14 March 2022 Accepted: 27 April 2022 Published: 31 May 2022 The cyanobacterium Synechocystis sp.PCC 6803 possesses a bidirectional NiFe-hydrogenase, HoxEFUYH. It functions to produce hydrogen under dark, fermentative conditions and photoproduces hydrogen when dark-adapted cells are illuminated. Unexpectedly, we found that the deletion of the large subunit of the hydrogenase (HoxH) in Synechocystis leads to an inability to grow on arginine and glucose under continuous light in the presence of oxygen. This is surprising, as the hydrogenase is an oxygen-sensitive enzyme. In wild-type (WT) cells, thylakoid membranes largely disappeared, cyanophycin accumulated, and the plastoquinone (PQ) pool was highly reduced, whereas 1hoxH cells entered a dormant-like state and neither consumed glucose nor arginine at comparable rates to the WT. Hydrogen production was not traceable in the WT under these conditions. We tested and could show that the hydrogenase does not work as an oxidase on arginine and glucose but has an impact on the redox states of photosynthetic complexes in the presence of oxygen. It acts as an electron valve as an immediate response to the supply of arginine and glucose but supports the input of electrons from arginine and glucose oxidation into the photosynthetic electron chain in the long run, possibly via the NDH-1 complex. Despite the data presented in this study, the latter scenario requires further proof. The exact role of the hydrogenase in the presence of arginine and glucose remains unresolved. In addition, a unique feature of the hydrogenase is its ability to shift electrons between NAD(H), NADP(H), ferredoxin, and ﬂavodoxin, which was recently shown in vitro and might be required for ﬁne-tuning. Taken together, our data show that Synechocystis depends on the hydrogenase to metabolize organic carbon and nitrogen in the presence of oxygen, which might be an explanation for its prevalence in aerobic cyanobacteria. Edited by: Constanze Pinske, Martin Luther University of Halle-Wittenberg, Germany Reviewed by: Tobias Goris, German Institute of Human Nutrition Potsdam-Rehbruecke (DIfE), Germany Peter Lindblad, Uppsala University Sweden Correspondence: Kirstin Gutekunst kirstin.gutekunst@uni-kassel.de INTRODUCTION of equimolar amounts of aspartic acid and arginine and can be quickly metabolized when required. Besides the accumulation of cyanophycin, cells cultivated on arginine also disassemble and restructure the thylakoid membranes, have reduced photosynthetic activity, and appear yellow due to a reduced chlorophyll content (Stephan et al., 2000). The addition of nitrate in combination with arginine partly rescues this phenotype (Stephan et al., 2000). The 1psbO mutant has a compromised water-splitting apparatus at PSII and displays reduced H2O oxidizing capacities. It grows better than the wild-type (WT) on arginine, stays green, and displays only minor morphological changes (Stephan et al., 2000). The diﬀerence between the WT and the 1psbO mutant strain is even more pronounced at higher light intensities. While nitrate needs to be reduced by ferredoxin and is an important electron sink of the photosynthetic electron chain, arginine needs to be oxidized to be utilized for biosynthetic reactions. Synechocystis encodes enzymes for two distinct arginine oxidation pathways: the arginine deiminase pathway and the arginine dehydrogenase pathway (Schriek et al., 2007, 2009). The arginine deiminase pathway yields NH+ 4 , NADH, and ATP, whereas the arginine dehydrogenase pathway yields NH+ 4 , NAD(P)H, and succinate. The arginine deiminase in Synechocystis contains predicted transmembrane helices which implies that the enzyme could be membrane attached (Schriek et al., 2007). Arginine dehydrogenase has been clearly demonstrated to attach to the thylakoid membrane (Schriek et al., 2009). Electrons from arginine oxidation can either be fed directly into the plastoquinone (PQ)-pool via the arginine dehydrogenase or indirectly via succinate and the succinate dehydrogenase, which likewise transfers electrons into the PQ- pool (refer to photosynthetic electron transfer chain of Figure 8) (Schriek et al., 2009; Mullineaux, 2014). In contrast to nitrate, which accepts electrons from the photosynthetic electron chain, arginine reduces the PQ-pool. However, nitrate has the potential to relieve surplus electrons from photosynthesis, and arginine has the potential to overload the PQ pool and cause oxidative stress. This could be an explanation for the observation that the 1psbO mutant with reduced H20 splitting capacities is superior to the WT in metabolizing arginine especially in high light (Stephan et al., 2000; Schriek et al., 2009). INTRODUCTION The addition of glucose to cells cultured on arginine should enhance the supply of electrons by the activity of respiratory dehydrogenases that reduce the PQ pool (refer to photosynthetic electron transfer chain of Figure 8) (Lea-Smith et al., 2016; Wang et al., 2022). Photomixotrophic conditions on arginine should, therefore, lead to highly reducing conditions in the cells. The observation that 1hoxH strain is unable to grow on arginine under mixotrophic dark-light conditions was interpreted as HoxEFUYH being required for electron dissipation via hydrogen production (Gutekunst et al., 2014). This interpretation appeared most obvious in cells during dark, fermentative H2 production, or during transient hydrogen production at the onset of illumination (photohydrogen production). However, surprisingly, we found in this study that the 1hoxH strain was unable to grow under photomixotrophic conditions on arginine, even under continuous illumination and O2 saturating conditions. This is especially remarkable, as HoxH k b d h f ( h The bidirectional NiFe-hydrogenase of the cyanobacterium Synechocystis is composed of a diaphorase sub-complex (HoxEFU), which reacts with its redox partners NAD(P)(H), ferredoxin, and ﬂavodoxin and a hydrogenase sub-complex (HoxYH), which catalyzes the production and consumption of hydrogen (H2) (Gutekunst et al., 2014; Artz et al., 2020). HoxEFUYH is the only hydrogenase encoded in this organism. It is localized in the cytoplasm and is found to associate dynamically with thylakoid membranes (Burroughs et al., 2014). The production of H2 occurs under fermentative growth conditions as a means for balancing redox poise. Furthermore, at the onset of photosynthesis in dark-adapted cells, HoxEFUYH catalyzes hydrogen production (photohydrogen) during the brief anaerobic phase of growth. Under these conditions, surplus electrons from the photosynthetic electron chain that are not accepted by the Calvin–Benson–Bassham (CBB) cycle are transferred to the hydrogenase and utilized for H2 production. In this situation, the enzyme works as an electron valve to protect the photosynthetic electron chain from over reduction (Appel et al., 2000; Cournac et al., 2004). Photohydrogen is subsequently oxidized by HoxEFUYH, and the electrons are most likely transferred to the photosynthetic electron chain via PQ (Dutta and Vermaas, 2016). As soon as oxygen accumulates due to H2O splitting at photosystem II (PSII), hydrogen is neither produced nor oxidized. Citation: Burgstaller H, Wang Y, Caliebe J, Hueren V, Appel J, Boehm M, Leitzke S, Theune M, King PW and Gutekunst K (2022) Synechocystis sp. PCC 6803 Requires the Bidirectional Hydrogenase to Metabolize Glucose and Arginine Under Oxic Conditions. Front. Microbiol. 13:896190. doi: 10.3389/fmicb.2022.896190 May 2022 \| Volume 13 \| Article 896190 Frontiers in Microbiology \| www.frontiersin.org 1 Cyanobacterial NiFe-Hydrogenase Under Oxic Conditions Burgstaller et al. Frontiers in Microbiology \| www.frontiersin.org Oxygen and Hydrogen Measurements yg y g To measure the concentration of dissolved oxygen and hydrogen in the cultures, oxygen and hydrogen sensors from Unisense (Unisense, Aarhus, Denmark) were used according to the manufacturer’s instructions. After a two-point calibration of the sensors, they were placed in the respective culture in glass tubes, and the measurement was started. Photosynthetic and respiratory activities were monitored via the oxygen evolution rate in the light and oxygen uptake rate in darkness. For this, calibrated sensors were put in a 25 ml culture (one sensor per culture, 3 repeats), which was illuminated for 15 min at 100 µmol photons m−2 s−1 followed by another 15 min in the darkness. For monitoring of oxygen and hydrogen concentrations during growth experiments, one hydrogen and one oxygen sensor were placed into the glass tubes with a diameter of 3.5 cm (Makowka et al., 2020). For the experiments shown in Supplementary Figure 1, one oxygen and one hydrogen sensor were placed into the glass tubes with cultures. The cultures with the sensors were incubated in dark and anoxic conditions from 0 to 0.73 h in order to prove their ability to produce fermentative hydrogen. As soon as the cells were illuminated, fermentative hydrogen was consumed. The cultures were purged with ambient air until they were saturated with O2. The aeration was turned oﬀ, and the cultures were left under continuous light. Construction of Mutants and Utilized Strains All mutants that were constructed or utilized in this study are listed in Supplementary Table 1. All the primers used in this study are listed in Supplementary Table 2. All mutants were constructed in the non-motile GT WT of Synechocystis sp. PCC 6803 (Trautmann et al., 2012). Constructs for the deletion of the genes were generated by Gibson cloning (Gibson et al., 2009) assembling three fragments into the pBluescript SK(+) in a single step. After examination by sequencing, the plasmids were transformed into Synechocystis sp. PCC 6803 cells as described (Williams, 1988). In short, genes were replaced with antibiotic resistance cassettes via homologous recombination. The transformation eﬃciency is especially high in the exponential growth phase. Therefore, 250 ml Synechocystis cultures were inoculated in glass tubes (diameter of 3.5 cm) from a preculture with an OD750 of 0,15 on the day prior to transformation. On the day of transformation, the cells were harvested and resuspended in 600 µl BG11. A volume of 300 µl of the cell suspension was mixed with 6–18 µg plasmid DNA and incubated for 6 h at 30◦C in darkness. Cells were plated on agar plates without antibiotics and kept in a climate chamber at 28◦C and 50 µE m2s1. On the third day, antibiotics were added for selection pressure. After 2 weeks, single colonies appeared that were streaked on new BG11 agar plates with antibiotics for segregation six to eight times. Resulting transformants were either checked by PCR or Southern hybridization (Supplementary Figures 3–7). Expression of HoxH mRNA p After puriﬁcation, 1 µg of RNA was subjected to a DNase digest. To check if the digest was complete, a test PCR with primers speciﬁc for rnpB was performed. If no PCR product was found, the RNA was reverse transcribed using the High Capacity RNA-to-cDNA Kit with the MuLV reverse transcriptase (Applied Biosystems, Warrington, UK). Subsequently, equal amounts of cDNA were used for the real-time PCR with the Power SYBR R⃝Green PCR Master Mix (Applied Biosystems, Warrington, UK). In this case, the temperature program was step 1 95◦C 10 min, step 2 95◦C 15 s, step 3 60 ◦C 60 s, and additional 39 cycles between steps 2 and 3 in the PCR Cycler Rotor- Gene Q (Qiagen, Hilden, Germany). For the quantiﬁcation, the 2−11CT method was used to analyze relative changes in transcript abundance (Livak and Schmittgen, 2001). The CT value was determined for each sample. DNA was diluted 1:30, 1:300, and 1:3,000. The threshold was set for all samples to the normalized ﬂuorescence 10−1. A typical serial dilution is shown in Supplementary Figure 8. Data were analyzed using Rotor-Gene Q Software (version 2.0.2). INTRODUCTION The current model assumes that exposure to oxygen blocks the active site of the NiFe-hydrogenase via a hydroxy (OH−) group that bridges Ni and Fe in the active site of HoxH and thereby prevents H2-turnover (Pandelia et al., 2010; McIntosh et al., 2011; Caserta et al., 2020). Due to its susceptibility to inactivation, the enzyme is considered oxygen sensitive. Therefore, it is remarkable that the enzyme is widespread in cyanobacteria, which seldomly encounter anoxic conditions in their natural environment. It is also worthy to note that the enzyme is constitutively expressed in Synechocystis under standard laboratory conditions, i.e., continuous light and constant O2 supersaturation. Distinct subcomplexes of the enzyme (HoxEFU, HoxFUYH, and HoxFU) have been detected in addition to the complete HoxEFUYH pentameric enzyme in vivo (Eckert et al., 2012). However, it is still unknown whether these subcomplexes fulﬁll speciﬁc functions. Synechocystis can grow on a variety of diﬀerent nitrogen sources. For example, cells can consume either inorganic nitrogen in the form of nitrate and ammonium or organic nitrogen such as urea or arginine and glutamine (Flores and Herrero, 1994). Nitrate is reduced intracellularly to nitrite and ammonium, which is subsequently incorporated into biomass (Forchhammer and Selim, 2020). It has previously been shown that the hydrogenase is essential for growth in a dark-light cycle under mixotrophic conditions when nitrate is replaced by arginine as the nitrogen source (Gutekunst et al., 2014). The replacement of nitrate with arginine has a dramatic eﬀect on the metabolism and phenotype of Synechocystis. Arginine can readily be taken up by the cells and metabolized as a cellular nitrogen source and in the production of cyanophycin (Stephan et al., 2000; Schriek et al., 2007). Cyanophycin consists May 2022 \| Volume 13 \| Article 896190 2 Cyanobacterial NiFe-Hydrogenase Under Oxic Conditions Burgstaller et al. diluted with BG11 medium by 1:10 when samples showed an OD750 value above 0.5. et al., 2011). To address this conundrum, investigations were undertaken to ensure the growth phenotype arose from the absence of HoxH and to elucidate the physiological function of the enzyme under these growth conditions. Our results provide strong evidence that under oxic conditions, the hydrogenase has a physiological function beyond hydrogen cycling and is required for photomixotrophic growth on organic carbon and nitrogen. Hydrogenase Activity Measurements via Methyl Viologen Hydrogenase activity was determined via methyl viologen (MV) as described before (Appel et al., 2020). Frontiers in Microbiology \| www.frontiersin.org Photochemical Quenching g Photochemical quenching (qP) measurements were conducted using a Multi-Color-PAM (Multiple Excitation Wavelength, Chlorophyll Fluorescence Analyzer, Heinz Walz, Eﬀeltrich, DE). All cultures were brought to a chlorophyll concentration of 2.5 µg/ml and independently measured at least three times each day. A volume of 2 ml of each culture were measured in a cuvette. The actinic light was adjusted to 58 µE/m2/s, and there were ﬁve repetitions of a sequence of 120 s of actinic light, a saturation ﬂash (SAT), and then a dark phase of 30 s in one measurement. qP was calculated from FV/(FM-F0) (van Kooten and Snel, 1990). and is given in the diﬀerent ﬁgures. and is given in the diﬀerent ﬁgures. To measure the electron transfer around PSI (from plastocyanin to P700 to ferredoxin), cell suspensions were adjusted to an OD750 of 5.7 and illuminated for 600 ms with 1,350 µE/m2/s. This illumination was repeated 20 times with an intermittent dark period of 30 s, and the average of the 20 measurements was calculated. Deconvolution of PC, P700+, and ferredoxin from the original traces of the Dual-KLAS/NIR was performed as described (Theune et al., 2021). The Dual-KLAS/NIR measures at six diﬀerent wavelengths in the near-infrared. From these wavelengths, four diﬀerent signals are calculated. Since the absorption of the three components (i.e., plastocyanin, P700+, and ferredoxin) contributes to a diﬀerent extent to the four signals, the proportions of oxidized plastocyanin, P700+, and reduced ferredoxin can be calculated by deconvolution (Klughammer and Schreiber, 2016). Determination of Total Protein Amount Based on the OD750 750 To determine if Synechocystis cells can be disrupted with similar eﬃciency, the protein concentration was measured in relation to the OD750. Therefore, liquid cultures were cultivated in glass tubes for 4 days. Then, the OD750 of the cultures was measured, and the culture volume that corresponded to an OD750 of 0.1 and 0.3 was calculated, respectively. After centrifugation for 5 min at room temperature (RT), the pellet was resuspended in 500 µl ACA buﬀer (750 mM ε-aminocaproic acid, 50 mM Tris-HCl pH 7.5, and 0.5 mM EDTA), and the suspension was transferred to a new 2 ml reaction tube. In the next step, precisely 0.25 g of glass beads (0.17–0.18 mm, Sigma Aldrich, Germany) was added to the cell suspension. The mixture was vortexed at 4◦C for 2 min using the following program: 12 cycles: 10 s “ON” followed by 10 s “OFF.” The speed was set to maximum. Later, the glass beads were pelleted at 5,000 × g and 4◦C for ∼1 min. To pellet the membrane fraction, the supernatant was transferred to a new 1.5 ml reaction tube and was centrifuged again at 12,000 × g and 4◦C for 20 min. Then, the protein concentration was determined using the Bradford assay. To calculate the number of cells that were not broken up during the method described, the samples were examined under a light microscope. To simplify the counting, a Neubauer counting chamber was used. These chambers are divided into nine large squares, each with a 1 mm2 area. The large squares are divided into 25 medium squares, and these are again divided into 16 small squares. Each of the smallest squares has an area of 0.0025 mm2. Four of the medium squares were counted, and the total number of intact cells was Chl a content (µg/ml) = (maximum of OD665 to OD666)−OD750 0.0809 dilution Growth Conditions Strains were either cultivated in BG11 medium which contains 17.6 mM nitrate or alternatively in BG110 medium without nitrate that was supplemented with 5 mM arginine. Notably, 10 mM glucose was added as indicated. DCMU was added at a concentration of 10 mM. For precultures, 50 ml of BG11 medium were inoculated with cells and antibiotics in the case of mutants in 100 ml Erlenmeyer ﬂasks on a rotary at 28◦C, 50 µE/m2/s, and 100 rpm. After several days of growth, cultures were pelleted and washed twice in the medium of choice without antibiotics for growth experiments. Cells were inoculated into 200 ml BG-11 at an OD750 of 0.05 and placed into glass tubes with a diameter of 3.5 cm bubbled with air at 50 µE/m2/s at 28◦C, and growth was monitored every 24 h by measuring the optical density at 750 nm as described earlier (Makowka et al., 2020). The optical density (OD) of the culture was determined by photometrical analysis (UV 2501 PC Photometer, Shimadzu, Kyoto, Japan) at 750 nm, and its data were recorded and analyzed using the aﬃliated software UVProbe 2.33 (Shimadzu, Kyoto, Japan). Samples were The CT value was determined and normalized to the reference gene 16S rRNA under two diﬀerent conditions (in BG11 and BG110 with arginine and glucose). From the CT values, the Frontiers in Microbiology \| www.frontiersin.org May 2022 \| Volume 13 \| Article 896190 3 Cyanobacterial NiFe-Hydrogenase Under Oxic Conditions Burgstaller et al. following ratio was determined: following ratio was determined: extrapolated. The counting was repeated in triplicates, and the mean value was calculated. We found that it was possible to break cells quantitatively at low OD750 between 0.1 and 0.3. Therefore, total protein contents were determined at these cell densities. ratio = 2CT(gene of intest condition 1)−CT(gene of interest condition 2) 2CT(refernence gene condition 1)−CT(reference gene condition 2) ratio = 2CT(gene of intest condition 1)−CT(gene of interest condition 2) 2CT(refernence gene condition 1)−CT(reference gene condition 2) Frontiers in Microbiology \| www.frontiersin.org Protein Preparation, Protein Analysis, and Immunoblotting g Soluble protein extracts of various Synechocystis sp. PCC 6803 strains were generated by glass bead breakage and diﬀerential centrifugation as described (Appel et al., 2020). Protein concentrations were determined by Bradford assay (Carl Roth). Soluble protein samples were either separated on 10% (w/v) denaturing 1-D SDS-PAGE Bis/Tris gels using an MES running buﬀer or on 0.75 mm thick 12% (w/v) native 1-D BN-PAGE gels (Boehm et al., 2009). Prior to 1-D BN- PAGE, ß-dodecyl-D-maltoside (ß-DM) was added to a ﬁnal concentration of 0.5% (w/v) from a 10% (w/v) stock to solubilize any remaining thylakoid membrane fragments and as a last step, 5% (v/v) Coomassie loading solution [750 mM ε-aminocaproic acid, 5% (w/v) Coomassie-G] was added. 2-D BN/SDS-PAGE and immunoblotting were performed as described by Appel et al. (2020). Primary antibodies used in this study were kindly provided by Prof. Peter Nixon (Imperial College, UK; puriﬁed, polyclonal antisera from rabbit against HoxE, HoxF, HoxU, HoxY, and HoxH). Chlorophyll Content To measure the chlorophyll a content, 3 × 100 µl of culture was centrifuged at 14,000 rpm (Centrifuge 5424, Eppendorf ) for 5 min at RT. The cell pellet was resuspended in 1 ml of 100% methanol, vortexed for 3 min, and centrifuged again at 14,000 rpm (Centrifuge 5424, Eppendorf ) for 5 min at RT to spin down cell debris. Using a spectrophotometer, the wavelengths 665, 666, and 750 nm were measured since chlorophyll a absorbs at the wavelengths 440–450 nm and 650–700 nm. According to the following formula (Lichtenthaler, 1987), the chlorophyll a content was calculated as follows: Arginine or Cyanophycin Extraction and Quantiﬁcation When WT and 1hoxH were cultivated on arginine and glucose under continuous light, 1hoxH displayed no growth, whereas the growth rate was similar to WT when cultured on nitrate, on arginine, or on glucose plus nitrate (Figures 1A–D). One molecule of cyanophycin contains one molecule of arginine. Therefore, the concentration of cyanophycin equals the amount of arginine. Cyanophycin quantiﬁcation was modiﬁed according to the study by Elbahloul et al. (2005) (Messineo, 1966). Step 1: cyanophycin extraction: 25 ml culture (OD750:1.0) was collected by centrifugation (10 min, 4,000 g, RT), and 1 ml acetone was added to the pellet and incubated in a shaker at 1,400 g for 30 min. The sample was centrifuged at 13,000 g for 10 min, the supernatant was discarded, and the pellet was resuspended in 1.2 ml of 0.1 M HCl and incubated for 1 h at 60◦C at 1,400 g. The sample was centrifuged at 13,000 g for 10 min at 4◦C, the supernatant was transferred to a new reaction cup, and 300 µl of 0.1 M Tris-HCl, pH 9.0 was added. The sample was incubated at 4◦C for 40 min, centrifuged at 18,000 g for 15 min, and the supernatant was discarded. The pellet was resuspended in 1 ml of 0.1 M HCl and used for the quantiﬁcation. Arginine standards with concentrations of 0, 10, 20, 30, 40, 50, 60, and 70 µg were prepared. Step 2: cyanophycin quantiﬁcation using the Sakaguchi reaction: 166 µl reagent A [300 mg KI (potassium iodide) in 100 ml distilled H2O] was added to 166 µl sample/standard containing between 10 and 100 µg arginine (0.057–0.5 µM, 0.1 M HCl). A volume of 500 µl of reagent B (100 ml of 5 M KOH, 2 g potassium sodium tartrate, 0.1 g 2,4-dichloro-1-naphthol, 180 ml absolute ethanol, and 0.2 ml NaClO) was added, and the reaction was incubated for 1 h at RT. A volume of 166 µl of reagent C [5%(v/v) NaClO with distilled H2O] was added, and the sample was incubated for exactly 10 min. The absorption at 520 nm was read immediately against a blank without arginine/cyanophycin. g Oxygen and hydrogen concentrations and OD750 were monitored in WT cells cultured on arginine and glucose for 96 h. Oxygen concentrations were close to saturation throughout the experiment, whereas, no hydrogen was detected (Figure 1E). Transmission Electron Microscopy py A volume of 30 ml of each Synechocystis culture were harvested at 8,000 g for 3 min and resuspended in 2 ml of the corresponding growth medium. The concentrated cells were ﬁxed by adding 2 ml of a solution containing 5% glutaraldehyde (GA) and 2% paraformaldehyde (FA) in 0.2 M cacodylate buﬀer (pH 7.3). Samples were either stored at 4◦C or directly observed with a transmission electron microscope (TEM, Tecnai G2 Spirit BioTWIN, Fei Company, Thermo Fisher Scientiﬁc, MS, USA). Glucose Quantiﬁcation Glucose was quantiﬁed enzymatically by measuring the evolution of NADPH photometrically as a change of absorption at 340 nm in a sample volume of 5 µl (containing <5 mM glucose) in the presence of 0.5 units hexokinase, 0.5 units glucose-6-phosphate dehydrogenase, 2 mM ATP, 2 mM NADP+, and 3 mM MgCl2 in 91.3 mM Tris-HCl buﬀer in a total volume of 50 µl. Absorption May 2022 \| Volume 13 \| Article 896190 Frontiers in Microbiology \| www.frontiersin.org Frontiers in Microbiology \| www.frontiersin.org 4 Cyanobacterial NiFe-Hydrogenase Under Oxic Conditions Burgstaller et al. RESULTS measurements were performed using a 96-well plate reader (Plate Reader Inﬁnite M200Pro, Tecan, Austria). measurements were performed using a 96-well plate reader (Plate Reader Inﬁnite M200Pro, Tecan, Austria). Arginine or Cyanophycin Extraction and Quantiﬁcation As the cultures were purged with ambient air, in this setting, hydrogen might have been gassed out from the cultures. Therefore, hydrogen and oxygen concentrations in WT cultures on arginine and glucose were monitored during growth under continuous light without purging (Supplementary Figure 1). Again, no hydrogen was detected, whereas oxygen accumulated. These results show that oxygen is present, but no hydrogen is detected under conditions where HoxH is required for growth. Morphology and Metabolic State of WT and 1hoxH on Arginine and Glucose Wild-type and 1hoxH reduced their chlorophyll content relative to OD750 on arginine and glucose. WT cultures appeared yellow, and 1hoxH cultures appeared colorless due to their low optical densities (Figure 2). Morphology and Metabolic State of WT and 1hoxH on Arginine and Glucose Wild-type and 1hoxH reduced their chlorophyll content relative to OD750 on arginine and glucose. WT cultures appeared yellow, and 1hoxH cultures appeared colorless due to their low optical densities (Figure 2). ( g ) Quantiﬁcation of glucose and arginine in the medium of WT and 1hoxH cultures revealed that WT cells consumed both compounds, whereas consumption by 1hoxH was negligible (Figures 3A,B). In agreement with this, the WT produced and stored large amounts of cyanophycin, which is composed of arginine and aspartic acid, whereas 1hoxH did not (Figure 3C). Microscopic images conﬁrmed that cyanophycin granules accumulated in WT cells only. Furthermore, they revealed that WT cells underwent dramatic morphological changes as they degraded their thylakoid membranes in contrast to 1hoxH (Figure 3D). The morphological changes observed in WT cells are well in line with earlier observations that were made on WT cells cultivated on arginine (Stephan et al., 2000). In contrast, 1hoxH cells cultured on arginine and glucose looked similar to 1hoxH cells cultured on nitrate. The photosynthetic activities of WT and 1hoxH cells were similar on nitrate and were strongly reduced in both when cultivated on arginine and glucose (Figures 3E,F). Dark respiration in 1hoxH was decreased compared with the WT, which is in line with its lower glucose and arginine consumption rates under the growth in continuous light (Figure 3G). It is remarkable that 1hoxH reduces its photosynthesis and respiratory activity even though the cells look normal and healthy with thylakoid membranes and Photosynthesis and Respiration Rate Measurements The respiration and photosynthesis rates of Synechocystis cultures were determined as the oxygen uptake rate in darkness and oxygen evolution rate in the light, which was determined using a Unisense oxygen microsensor (Unisense, Denmark) according to the manufacturer’s instruction. May 2022 \| Volume 13 \| Article 896190 Frontiers in Microbiology \| www.frontiersin.org 5 Cyanobacterial NiFe-Hydrogenase Under Oxic Conditions Burgstaller et al. FIGURE 1 \| Growth of wild type (WT) and 1hoxH under different conditions and oxygen and hydrogen concentrations in cultures on arginine and glucose in continuous light (60 µE/m2/s). (A) Growth in arginine and glucose (AG), (B) growth in nitrate (N), (C) growth in arginine (A), (D) growth in nitrate and glucose (NG), (E) O2 and H2 concentration in the cultures cultivated on arginine and glucose. FIGURE 1 \| Growth of wild type (WT) and 1hoxH under different conditions and oxygen and hydrogen concentrations in cultures on arginine and glucose in continuous light (60 µE/m2/s). (A) Growth in arginine and glucose (AG), (B) growth in nitrate (N), (C) growth in arginine (A), (D) growth in nitrate and glucose (NG), (E) O2 and H2 concentration in the cultures cultivated on arginine and glucose. FIGURE 2 \| (A) Chlorophyll content of WT and 1hoxH during growth experiments under photoautotrophic conditions and on arginine and glucose. (B) The appearance of strains on day 3 after inoculation. FIGURE 2 \| (A) Chlorophyll content of WT and 1hoxH during growth experiments under photoautotrophic conditions and on arginine and glucose. (B) The appearance of strains on day 3 after inoculation. Frontiers in Microbiology \| www.frontiersin.org May 2022 \| Volume 13 \| Article 896190 Cyanobacterial NiFe-Hydrogenase Under Oxic Conditions Burgstaller et al. Burgstaller et al. Cyanobacterial NiFe-Hydrogenase Under Oxic Conditions FIGURE 3 \| Metabolism and morphology of WT and 1hoxH on arginine and glucose. (A) Growth (solid line) and glucose consumption (dotted line). (B) Arginine consumption, (C) Cyanophycin content of cells displayed as arginine in the cells, (D) Transmission electron microscopy (TEM) images of WT and 1hoxH that were cultivated on arginine and glucose. Cyanophycin granules are marked with a red arrow in the WT. (E) Growth experiment in which (F) dark respiration and (G) photosynthesis were determined by O2 uptake and evolution on days 2–4. Stars on the column diagram indicate differences between mutant and WT by Tukey’s HSD test (P < 0.05, P < 0.01). Throughout the ﬁgure ‘NS’ indicates ‘not signiﬁcant’. Photosynthesis and Respiration Rate Measurements All error bars indicate standard deviation (s.d.) of four daily independent experiments. FIGURE 3 \| Metabolism and morphology of WT and 1hoxH on arginine and glucose. (A) Growth (solid line) and glucose consumption (dotted line). (B) Arginine consumption, (C) Cyanophycin content of cells displayed as arginine in the cells, (D) Transmission electron microscopy (TEM) images of WT and 1hoxH that were cultivated on arginine and glucose. Cyanophycin granules are marked with a red arrow in the WT. (E) Growth experiment in which (F) dark respiration and (G) photosynthesis were determined by O2 uptake and evolution on days 2–4. Stars on the column diagram indicate differences between mutant and WT by Tukey’s HSD test (P < 0.05, **P < 0.01). Throughout the ﬁgure ‘NS’ indicates ‘not signiﬁcant’. All error bars indicate standard deviation (s.d.) of four daily independent experiments. only small amounts of cyanophycin granules. In view of the fact that 1hoxH cells neither consumed glucose nor arginine in large quantities, were unable to grow and showed no morphological changes, it is assumed that 1hoxH cells entered a dormant- like state and were unable to adapt in the same manner as WT cells. May 2022 \| Volume 13 \| Article 896190 Frontiers in Microbiology \| www.frontiersin.org 7 Cyanobacterial NiFe-Hydrogenase Under Oxic Conditions Burgstaller et al. FIGURE 4 \| Signiﬁcance of the diaphorase and the hydrogenase on arginine and glucose. (A) SDS page and immunoblots of WT, 1hoxH, and 1hoxW with antibodies against all hox subunits (B) 2D Blue Native PAGE of WT, 1hoxH, and 1hoxW with antibodies against all hox subunits. The second dimension allows to check for the assembly of complexes. (C) Growth of WT, 1hoxH, and 1hoxW on arginine and glucose. (D) Complementation of 1hox with hoxYH (1hox/hoxYH). Growth of WT, 1hox/hoxYH, and 1hox on arginine and glucose. FIGURE 4 \| Signiﬁcance of the diaphorase and the hydrogenase on arginine and glucose. (A) SDS antibodies against all hox subunits (B) 2D Blue Native PAGE of WT, 1hoxH, and 1hoxW with antib check for the assembly of complexes. (C) Growth of WT, 1hoxH, and 1hoxW on arginine and gluc G th f WT 1h /h YH d 1h i i d l FIGURE 4 \| Signiﬁcance of the diaphorase and the hydrogenase on arginine and glucose. Photosynthesis and Respiration Rate Measurements (A) SDS page and immunoblots of WT, 1hoxH, and 1hoxW with antibodies against all hox subunits (B) 2D Blue Native PAGE of WT, 1hoxH, and 1hoxW with antibodies against all hox subunits. The second dimension allows to check for the assembly of complexes. (C) Growth of WT, 1hoxH, and 1hoxW on arginine and glucose. (D) Complementation of 1hox with hoxYH (1hox/hoxYH). Growth of WT, 1hox/hoxYH, and 1hox on arginine and glucose. Frontiers in Microbiology \| www.frontiersin.org Importance of Diaphorase (HoxEFU) and Hydrogenase (HoxYH) Moiety on Arginine and Glucose complexes that are visible in the second dimension are HoxEFUYH and HoxFUYH (Figure 4B). These complexes are missing in hoxH and hoxW deletion strains. However, complexes that contain the diaphorase subunits only, namely HoxEFU and HoxFU, are present in the WT and in both mutants. As no hydrogen was detected and hydrogenase is known to be oxygen-sensitive, one explanation for the function of Hox may be that the diaphorase is critical to growth under oxic conditions on arginine and glucose. To test this, levels of HoxEFU in 1hoxH cell extracts were monitored by immunoblot. We also constructed a mutant strain 1hoxW that is defective in the maturase HoxW that activates HoxH (Eckert et al., 2012). This mutant should in principle contain a fully assembled and functional HoxEFU diaphorase but lack a functional HoxH. As the diaphorase was apparently present and assembled in 1hoxW, we tested the growth of this mutant on arginine and glucose and found it to be similar but slightly better than the growth of 1hoxH cells (Figure 4C). In addition, when a 1hox mutant, in which hoxEFUYH was deleted, was complemented with hoxYH, growth improved considerably, but not to WT levels (Figure 4D). These results indicate that the hydrogenase (HoxYH) is the critical component and that the diaphorase (HoxEFU) alone cannot support the growth of arginine and glucose. However, we cannot rule out that the holoenzyme of HoxEFUYH is required for the proper functioning of the diaphorase in vivo. The fact that the diminished growth of 1hox (1hoxEFUYH) could be partly rescued by the introduction of hoxYH, furthermore, indicates that the hydrogenase indeed fulﬁlls an important function Immunoblots and 2D Blue Native PAGE revealed the presence and assembly of the diaphorase HoxEFU in both the 1hoxH and 1hoxW cells (Figures 4A,B). HoxE levels were slightly reduced in the 1hoxH background but were fully present in 1hoxW cells. HoxY levels were very low in both mutants, whereas HoxH was absent in 1hoxH and only present in its unprocessed form in 1hoxW cells (Figure 4A). In the WT the most prominent May 2022 \| Volume 13 \| Article 896190 Frontiers in Microbiology \| www.frontiersin.org 8 Cyanobacterial NiFe-Hydrogenase Under Oxic Conditions Burgstaller et al. FIGURE 5 \| Growth of WT and 1hoxH on arginine and glucose. (A) Growth at high light (200 µE/m2/s) and low light (20 µE/m2/s). Importance of Diaphorase (HoxEFU) and Hydrogenase (HoxYH) Moiety on Arginine and Glucose (B) Growth at medium light (60 µE/m2/s) in the absence and presence of the inhibitor DCMU, which blocks electron transfer from PSII to the plastoquinone (PQ) pool. FIGURE 5 \| Growth of WT and 1hoxH on arginine and glucose. (A) Growth at high light (200 µE/m2/s) and low light (20 µE/m2/s). (B) Growth at medium light (60 µE/m2/s) in the absence and presence of the inhibitor DCMU, which blocks electron transfer from PSII to the plastoquinone (PQ) pool. valve under these conditions. To test this, WT and 1hoxH cells were cultivated on arginine and glucose under high light (200 µE/m2/s) and under low light (20 µE/m2/s) in order to determine if low light relieves the reductive stress to improve the growth of 1hoxH. The WT strain grew well under both light intensities, whereas 1hoxH which did not grow at all at 200 µE/m2/s but was able to grow under low light at 20 µE/m2/s (Figure 5A). without the diaphorase under these conditions but that, in contrast, the holoenzyme HoxEFUYH is required for WT-like growth on arginine and glucose. Transcription and Expression of the Hydrogenase Under Mixotrophic Conditions on Arginine The addition of DCMU, which blocks electron transfer from PSII to the PQ pool, enhanced the growth of WT under medium light intensities (60 µE/m2/s), and the growth of 1hoxH was completely rescued to WT levels (Figure 6B). These data show that the requirement for the hydrogenase on arginine and glucose is related to electron transport during photosynthesis. As the growth of the WT and 1hoxH strains deviates from each other within the ﬁrst day of cultivation, mRNA was isolated from cells cultured on nitrate (BG11) as control and on arginine and glucose (BG110AG) at 12, 24, 36, 48, and 72 h after inoculation. In cells cultured on arginine and glucose, the transcription of hoxH was upregulated 2-fold after 24 h and 5-fold after 36 h from inoculation (Supplementary Figure 2A). The expression levels of the hoxEFUYH operon 24 h after inoculation on arginine and glucose revealed that the entire hox operon was upregulated (Supplementary Figure 2B). Despite increased transcription, protein levels of Hox subunits (HoxE, F, U, Y, H) appeared to be unaﬀected (Supplementary Figures 2C,D). To test if the hydrogenase has a direct and immediate impact on the photosynthetic electron transfer chain, the reduction and oxidation of plastocyanin (PC), P700, and ferredoxin were monitored upon the addition of arginine and glucose in WT and 1hoxH via Dual-KLAS/NIR (Figure 6). PC and P700 are oxidized in WT and 1hoxH at the onset of illumination (Figures 6A–D). Under photoautotrophic conditions, the ﬁrst oxidation peak is reached ∼100 ms into the light pulse and is followed by a transient phase of stronger reduction of PC and P700 (Figure 6A). The reduction is caused by electrons coming in from PSII due to the reduced electron acceptor pool (Fdx and NADP+) of PSI. After another 200 ms, this reduction phase reverts, and oxidation predominates again before the light is turned oﬀafter 600 ms (Theune et al., 2021). This second oxidation phase is governed by the activity of the ﬂavodiiron proteins (mainly Flv1/3) that catalyze oxygen reduction at the acceptor side of PSI. In the absence of the ﬂavodiiron proteins, this phase is missing (refer to Figure 7B) (SÈtif et al., 2020). In 1hoxH, the reduction phase is faster and deeper for PC and Does the Hydrogenase Work as an Electron Valve for Photosynthesis on Arginine and Glucose? Taken together, these data show that the hydrogenase works as a transient electron valve in an immediate response to the addition of organic carbon and nitrogen in the presence of oxygen. acceptor side limitation at PSI was more severe in the presence of the hydrogenase (Figures 7A–D). Thus, our measurements show that the hydrogenase does not function as an oxidase. We then examined the PSII ﬂuorescence properties of WT and 1hoxH cells cultured over a 5-day period on nitrate, or on arginine plus glucose, in order to monitor the redox state of the PQ pool. Both strains exhibited a similar PQ pool redox poise when cultivated on nitrate. When cultured on arginine plus glucose, the PQ pools of both strains were more reduced than in cells cultured on nitrate, becoming more pronounced in WT cells later in growth (Figure 7G). These data are well in line with the Dual-KLAS/NIR measurements during the growth on organic carbon and nitrogen in WT, 1ﬂv241ﬂv31cox1cyd, and 1ﬂv241ﬂv31cox1cyd1hox (Figures 7A–F). The fact that components of the photosynthetic electron transfer chain were more reduced in the WT in comparison to 1hoxH on arginine and glucose (Figure 7G) is comprehensible considering that 1hoxH did neither metabolize arginine nor glucose in large amounts and furthermore paused growth (Figure 3). Taken together, the results indicate that in the immediate response to the growth on arginine and glucose, the hydrogenase functions to dissipate electrons (Figure 6), however, in the long run for hours and days, hydrogenase supports feeding electrons into the PQ pool under these conditions, suggesting it has a dual function. yg However, as no hydrogen was traceable, the hypothesis was put forward that the hydrogenase might release electrons from the photosynthetic electron transport chain by reducing oxygen instead of protons. According to the current understanding, Ni and Fe in the active site of the hydrogenase are inactivated in the presence of oxygen by the formation of a bridging hydroxy group which prevents H2-turnover (Pandelia et al., 2010; McIntosh et al., 2011; Caserta et al., 2020). The oxygen tolerance of the soluble NiFe-hydrogenase of Ralstonia eutropha is based on its oxidase activity which reactivates the enzyme by reducing the bound hydroxyl group to water (Lauterbach and Lenz, 2013). Does the Hydrogenase Work as an Electron Valve for Photosynthesis on Arginine and Glucose? Based on this ﬁnding, the hypothesis was put forward that the oxygen-sensitive NiFe-hydrogenase of Synechocystis might also function as an oxidase and switch between H2 production under anaerobic conditions and oxygen reduction under aerobic conditions. To test this hypothesis, two mutants were constructed. The ﬁrst one lacked all of the terminal oxidases that are associated with the thylakoid membrane, namely, the respiratory terminal oxidase cytochrome C oxidase (cyd) and quinol oxidase (cox) but retained the alternative respiratory oxidase (arto; ctaII). In addition, ﬂavodiiron proteins, namely, Flv2, Flv4, and Flv3, which have the potential to reduce oxygen, were deleted (Shimakawa et al., 2014; Brown et al., 2019). This resulted in the mutant strain 1ﬂv241ﬂv31cox1cyd. The second mutant was the same as the ﬁrst but also lacked the hox operon (hoxEFUYH), yielding the mutant strain 1ﬂv241ﬂv31cox1cyd1hox (note that in this strain, the entire hox operon instead of hoxH was deleted). The redox status of photosynthetic components [plastocyanin (PC), P700 in PSI, and ferredoxin] were measured in these mutants and compared with the WT strain cultured on arginine and glucose on the second day after inoculation in the absence and in the presence of oxygen via Dual-KLAS/NIR (Figure 7). If the hydrogenase is functioning as an oxidase and transferring surplus electrons to oxygen, it was expected that the deletion of hoxEFUYH would result in more reduced photosynthetic components in the presence of oxygen. However, the ferredoxin pool and P700 were more reduced in 1ﬂv241ﬂv31cox1cyd in comparison to 1ﬂv241ﬂv31cox1cyd1hox indicating that Does the Hydrogenase Work as an Electron Valve for Photosynthesis on Arginine and Glucose? Three main sources feed electrons into the PQ pool of the photosynthetic electron transport chain in cultures that are cultivated on arginine and glucose as follows: water splitting at PSII, the oxidation of the carbon skeleton of arginine, and the oxidation of glucose. This might lead to an over-reduced photosynthetic electron transfer chain with a highly reduced PQ pool. Therefore, the hydrogenase might work as an electron May 2022 \| Volume 13 \| Article 896190 Frontiers in Microbiology \| www.frontiersin.org 9 Cyanobacterial NiFe-Hydrogenase Under Oxic Conditions Burgstaller et al. FIGURE 6 \| Oxidation of PC and P700 and reduction of ferredoxin and NADPH of WT and 1hoxH cultures as measured by the Dual-KLAS/NIR. On the left (A, C, E, G), curves of cultures grown photoautotrophically are shown and on the right (B, D, F, H), the same cultures are shown that were measured a few minutes after the addition of arginine and glucose. \| \| \| FIGURE 6 \| Oxidation of PC and P700 and reduction of ferredoxin and NADPH of WT and 1hoxH cultures as measured by the Dual-KLAS/NIR. On the left (A, C, E, G), curves of cultures grown photoautotrophically are shown and on the right (B, D, F, H), the same cultures are shown that were measured a few minutes after the addition of arginine and glucose. May 2022 \| Volume 13 \| Article 896190 Frontiers in Microbiology \| www.frontiersin.org 10 Cyanobacterial NiFe-Hydrogenase Under Oxic Conditions Burgstaller et al. P700 in contrast to the WT. This eﬀect is more pronounced in the presence of arginine and glucose (Figures 6B,D). The ﬁrst peak of oxidation becomes narrower and is reached after only 35 ms before reduction becomes stronger. 1hoxH shows a deeper reduction trough in the light phase and a faster reduction in the dark phase. The mutant obviously suﬀers from a stronger acceptor side limitation. This is noticeable when comparing the reduction curves of NAD(P)H (Figures 6G,H). Under autotrophic conditions, NAD(P)H ﬂuorescence shows a large increase in illumination, but in the presence of arginine and glucose, this peak is negligible. This indicates that the NAD(P)H pool is already mainly reduced on arginine and glucose. The same was observed for the ferredoxin pool, which was instantly fully reduced upon illumination on arginine and glucose (Figures 6E,F). Frontiers in Microbiology \| www.frontiersin.org Signiﬁcance of Components of the Respiratory Electron Chain for Growth on Arginine and Glucose To evaluate the signiﬁcance of components of the respiratory electron transfer chain for the growth on arginine and glucose, additional deletion mutants were constructed and tested. Synechocystis possesses three respiratory terminal oxidases: the quinol oxidase (cox) is localized in thylakoid membranes, cytochrome C oxidase (cyd) is localized in the thylakoid and cytoplasmic membranes, and the alternative respiratory oxidase (arto) is restricted to the cytoplasmic membranes (Figure 8A). Deletion of both thylakoid-associated oxidases (1cyd1cox) or all three of the terminal oxidases (1cyd1cox1arto) resulted in mutants that were still able to grow on arginine and glucose albeit at impaired rates (Figure 8B). Three main respiratory dehydrogenases feed electrons from glucose oxidation into the PQ pool in Synechocystis: photosynthetic complex I (NDH-1) accepts electrons from reduced ferredoxin (Schuller et al., 2019), NDH-2 accepts electrons from NADH, and the succinate dehydrogenase (SDH) catalyzes the oxidation from succinate to fumarate. In the 1ndh2 mutant strain, all three homologs ndbA/ndbB/ndbC of the NADH dehydrogenase-like complex 2 (NDH2; slr0851, slr1743, and sll1484) were deleted. In the 1sdh mutant, the two subunits sdh1 and sdh2 of the succinate dehydrogenase (SDH; sll1625, and sll0823)) were deleted. Additionally, in the 1ndh1 mutant, the subunits ndh-D1 and ndh-D2 of complex 1 (NDH1; slr0331, and slr1291) were deleted. Growth of all three mutant strains was impaired on arginine and glucose (Figure 8C), where the severest impairment was observed for 1ndh1 followed by May 2022 \| Volume 13 \| Article 896190 11 Cyanobacterial NiFe-Hydrogenase Under Oxic Conditions Burgstaller et al. GURE 7 \| (A–F) Redox states of components of the photosynthetic electron chain on arginine and glucose in the presence and absence of oxygen in WT, ﬂv241ﬂv31cox1cyd, and 1ﬂv241ﬂv31cox1cyd1hox. The oxidation state was measured as absorption change (1I/I × 10−3). Positive values show oxidation and gative values show a reduction of components. (G) Photochemical quenching (qP) in WT and 1hoxH on nitrate (WT and 1hoxH) and on arginine and glucose (WT G and 1hoxH AG). tiers in Microbiology \| www.frontiersin.org 12 May 2022 \| Volume 13 \| Article 89619 FIGURE 7 \| (A–F) Redox states of components of the photosynthetic electron chain on arginine and glucose in the presence and absence of oxygen in WT, 1ﬂv241ﬂv31cox1cyd, and 1ﬂv241ﬂv31cox1cyd1hox. The oxidation state was measured as absorption change (1I/I × 10−3). Positive values show oxidation and negative values show a reduction of components. Signiﬁcance of Components of the Respiratory Electron Chain for Growth on Arginine and Glucose (G) Photochemical quenching (qP) in WT and 1hoxH on nitrate (WT and 1hoxH) and on arginine and glucose (WT AG and 1hoxH AG). Frontiers in Microbiology \| www.frontiersin.org May 2022 \| Volume 13 \| Article 896190 12 Cyanobacterial NiFe-Hydrogenase Under Oxic Conditions Burgstaller et al. FIGURE 8 \| Signiﬁcance of the respiratory electron chain for growth on arginine and glucose. (A) Overview of photosynthetic and respiratory electrons chains in the thylakoid and cytoplasmic membrane. (B) Growth of WT and mutants in which terminal oxidases were deleted on arginine and glucose. (C) Growth of WT and dehydrogenases, that feed electrons into the PQ pool on arginine and glucose. FIGURE 8 \| Signiﬁcance of the respiratory electron chain for growth on arginine and glucose. (A) Overview of photosynthetic and respiratory electrons chains in the thylakoid and cytoplasmic membrane. (B) Growth of WT and mutants in which terminal oxidases were deleted on arginine and glucose. (C) Growth of WT and dehydrogenases, that feed electrons into the PQ pool on arginine and glucose. 1sdh. Thus, the terminal oxidases appeared to be dispensable for growth on arginine and glucose, whereas SDH has a signiﬁcant role, and hydrogenase and NDH-1 are essential. in fact, encounter anoxic conditions at night due to intense respiration within the microbial mat community and produce fermentative hydrogen in darkness and photohydrogen on illumination (Burow et al., 2012a,b; Bolhuis et al., 2014; Nielsen et al., 2015). The function of the cyanobacterial bidirectional NiFe-hydrogenase is thus well-understood in microbial mats. However, the enzyme is also widespread in cyanobacteria from surface waters that stay oxygenated also at night (Barz et al., 2010; Beimgraben et al., 2014; Greening et al., 2015). Remarkably, it is absent from oligotrophic habitats as the open ocean but is found in freshwater and coastal waters that have a higher load of nutrients and occasionally encounter phytoplankton blooms (Barz et al., 2010). Excretion and lysis of organisms in senescent phytoplankton blooms and the bacterial breakdown of organic carbon to shorter sugars create mixotrophic conditions in these habitats that are typically accompanied by increased levels of dissolved organic nitrogen as arginine (Wetz and Wheeler, 2003, 2004; Teeling et al., 2012). The replacement of nitrate with arginine and addition of glucose in this study thus Frontiers in Microbiology \| www.frontiersin.org CONCLUSION The cyanobacterial bidirectional NiFe-hydrogenase of Synechocystis is required for growth on organic carbon and nitrogen in the presence of oxygen and fulﬁlls a function apart from hydrogen turnover. Its presence has an impact on the redox states of photosynthetic components in the presence of oxygen. It acts as an electron valve as an immediate response to the supply of arginine and glucose. The enzyme does not work as an oxidase and it does not have the role to dissipate electrons during long-term growth. Rather, hydrogenase supports the input of electrons into the photosynthetic electron transport chain. In the presence of the hydrogenase, Synechocystis performs a metabolic switch, which allows cells to grow on arginine and glucose but enters a dormant-like state in its absence. The exact manner by which hydrogenase inﬂuences the thylakoid membrane’s electron transport chain remains unresolved. Our observations support a model for the more global occurrence of a bidirectional NiFe-hydrogenase in cyanobacteria that were found predominantly in oxygenated surface waters of nutrient- rich habitats. Its unique ability to shuttle electrons between NAD(H), NADP(H), ferredoxins, and ﬂavodoxin might help to mediate an input of electrons via the NDH-1 complex into the photosynthetic electron chain. However, this view requires further experimental evidence. q y y g In-vitro experiments have revealed that the hydrogenase HoxEFU diaphorase sub-complex is able to exchange electrons between NAD(H), NAD(P)H, several ferredoxins and ﬂavodoxin (Artz et al., 2020). This is a unique property of HoxEFU in having reactivity with more than four redox carriers. In contrast, FNR shuttles electrons between NADPH and ferredoxin, and the transhydrogenase (PntAB) transfers electrons from NADH to NADP+ (Kämäräinen et al., 2016). Although the hydrogenase catalytic subunits, HoxYH, were not required for the HoxEFU activity with redox carriers in vitro, it cannot be ruled out that this function requires the presence of the HoxYH subunits in vivo. The data presented in this study add another dimension to Hox function by demonstrating that the hydrogenase subunit HoxH, but not the hydrogen reactivity of HoxH, is required for electron exchange with the PQ pool in cells grown on arginine and glucose. As the photosynthetic complex I was likewise required for growth on organic carbon and nitrogen, it is possible that the hydrogenase transfers electrons via NDH-1 into the PQ pool. DISCUSSION An obvious question that suggests itself is why the bidirectional NiFe-hydrogenase, which is an oxygen-sensitive enzyme, is widespread in organisms that produce oxygen and rarely encounter anoxic conditions. Its presence has been regarded as a leftover from ancient times. However, the ancestors of cyanobacteria possessed several NiFe- and FeFe-hydrogenases (types 2a, 3b, 3c, 3d, and 4), whereas cyanobacteria kept only two NiFe-hydrogenases: the uptake NiFe-hydrogenase (type 2a) in association with N2 ﬁxation and the bidirectional NiFe-hydrogenase (type 3d) (Vignais and Billoud, 2007). The advantage for the uptake of hydrogenase, which consumes H2, which is a by-product of N2 ﬁxation, is apparent. But why was the bidirectional enzyme kept? Cyanobacteria in microbial mats, May 2022 \| Volume 13 \| Article 896190 Frontiers in Microbiology \| www.frontiersin.org 13 Cyanobacterial NiFe-Hydrogenase Under Oxic Conditions Burgstaller et al. somewhat mimics the situation of a senescent phytoplankton bloom in habitats where cyanobacterial NiFe-hydrogenases are found. The presented data indicate strongly that the hydrogenase has a function under oxic conditions which are most likely not based on hydrogen turnover but linked to photosynthesis and respiration to support a metabolic switch for growth in the presence of organic carbon and nitrogen. Please note that the concentrations of glucose and arginine that were utilized in this study are above those levels found in nature. We thus created artiﬁcial laboratory conditions so that the comparison with natural conditions must be performed cautiously. We did not ﬁnd any evidence for the hypothesis that the hydrogenase works as an oxidase or functions as an electron valve during long- term growth. Rather, our results indicate that the hydrogenase supports to feed electrons into the photosynthetic electron transport chain. Since the 1hoxH mutant does not consume signiﬁcant amounts of glucose and arginine, it is not clear whether the hydrogenase is actively involved in feeding electrons into the photosynthetic electron transport chain in the WT, or if its absence in 1hoxH rather prevents a metabolic switch and thereby results in less reduced photosynthetic components. The functional signiﬁcance of hydrogenase is most likely linked to the tuning of the light reactions of photosynthesis (Figure 5). For example, Synechocystis possesses various complexes that ﬁne-tune the intersystem electron transport chain including ﬂavodiiron proteins, respiratory dehydrogenases, and terminal oxidases. Our results in this study demonstrate that most of these were dispensable for the growth on arginine and glucose and raise the question as to why hydrogenase is essential. CONCLUSION This idea is further supported by the fact that respiration was diminished in 1hoxH cells in comparison to the WT cells (Figure 3F). It has been shown that the cyanobacterial NDH-1 complex accepts electrons exclusively from reduced ferredoxin (Schuller et al., 2019). It is conceivable that the Hox hydrogenase complex might mediate an input of electrons via the NDH-1 complex into the photosynthetic electron transport chain and thereby tune photosynthesis and respiration in the presence of DISCUSSION organic carbon and nitrogen. However, this idea requires further experimental support. DATA AVAILABILITY STATEMENT The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s. AUTHOR CONTRIBUTIONS HB, YW, JC, VH, JA, MB, SL, and MT performed experiments. HB, YW, JA, and KG contributed to conception and design of the study. HB, YW, JC, VH, JA, MB, SL, MT, PK, and KG analyzed and interpreted data. HB, YW, VH, and KG wrote the ﬁrst draft of the manuscript. All authors contributed to manuscript revision, read, and approved the submitted version. Frontiers in Microbiology \| www.frontiersin.org REFERENCES 148, 350–382. doi: 10.1016/0076-6879(87)48036-1 Bolhuis, H., Cretoiu, M. S., and Stal, L. J. (2014). Molecular ecology of microbial mats. FEMS Microbiol. Ecol. 90, 335. doi: 10.1111/1574-6941.12408 Kämäräinen, J., Huokko, T., Kreula, S., Jones, P. R., Aro, E.-M., and Kallio, P. (2016). Pyridine nucleotide transhydrogenase PntAB is essential for optimal growth and photosynthetic integrity under low-light mixotrophic conditions in Synechocystis sp. PCC 6803. New Phytol. 214. doi: 10.1111/nph. 14353 Brown, K. A., Guo, Z., Tokmina-Lukaszewska, M., Scott, L. W., Lubner, C. E., Smolinski, S., et al. (2019). The oxygen reduction reaction catalyzed by Synechocystis sp. PCC 6803 ﬂavodiiron proteins. Sustain. Energy Fuels 3, 3191–3200. doi: 10.1039/C9SE00523D Klughammer, C., and Schreiber, U. (2016). Deconvolution of ferredoxin, plastocyanin, and P700 transmittance changes in intact leaves with a new type of kinetic LED array spectrophotometer. Photosyn. Res. 128, 195–214. doi: 10.1007/s11120-016-0219-0 Burow, L. C., Woebken, D., Bebout, B. M., McMurdie, P. J., Singer, S. W., Pett- Ridge, J., et al. (2012a). Hydrogen production in photosynthetic microbial mats in the Elkhorn Slough estuary, Monterey Bay. ISME J. 6, 863–874. doi: 10.1038/ismej.2011.142 Lauterbach, L., and Lenz, O. (2013). Catalytic production of hydrogen peroxide and water by oxygen-tolerant [NiFe]-hydrogenase during H2 cycling in the presence of O2. J. Am. Chem. Soc. 135, 17897–17905. doi: 10.1021/ja408420d Burow, L. C., Woebken, D., Marshall, I. P. G., Lindquist, E. A., Bebout, B. M., Prufert-Bebout, L., et al. (2012b). Anoxic carbon ﬂux in photosynthetic microbial mats as revealed by metatranscriptomics. ISME J. 7, 1–13. doi: 10.1038/ismej.2012.150 Lea-Smith, D. J., Bombelli, P., Vasudevan, R., and Howe, C. J. (2016). Photosynthetic, respiratory and extracellular electron transport pathways in cyanobacteria. Biochim. Biophys. Acta Bioenerget. 1857, 247–255. doi: 10.1016/j.bbabio.2015.10.007 Burroughs, N. J., Boehm, M., Eckert, C., Mastroianni, G., Spence, E. M., Yu, J., et al. (2014). Solar powered biohydrogen production requires speciﬁc localization of the hydrogenase. Energy Environ. Sci. 7, 3791–3800. doi: 10.1039/C4EE02502D Livak, K. J., and Schmittgen, T. D. (2001). Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25, 402–408. doi: 10.1006/meth.2001.1262 Caserta, G., Pelmenschikov, V., Lorent, C., Tadjoung Waﬀo, A. F., Katz, S., Lauterbach, L., et al. (2020). Hydroxy-bridged resting states of a [NiFe]- hydrogenase unraveled by cryogenic vibrational spectroscopy and DFT computations. Chem. Sci. 12, 2189–2197. doi: 10.1039/D0SC05022A Makowka, A., Nichelmann, L., Schulze, D., Spengler, K., Wittmann, C., Forchhammer, K., et al. (2020). REFERENCES Acinetobacter calcoaceticus strain ADP1. Appl. Environ. Microbiol. 71, 858–866. doi: 10.1128/AEM.71.2.858-866.2005 Appel, J., Hueren, V., Boehm, M., and Gutekunst, K. (2020). Cyanobacterial in vivo solar hydrogen production using a photosystem I–hydrogenase (PsaD-HoxYH) fusion complex. Nat. Energy 5, 458–467. doi: 10.1038/s41560-020-0609-6 Flores, E., and Herrero, A. (1994). “Assimilatory nitrogen metabolism and its regulation,” in The Molecular Biology of Cyanobacteria, ed D. A. Bryant (Dordrecht: Springer Netherlands), 487–517. fusion complex. Nat. Energy 5, 458–467. doi: 10.1038/s41560-02 Forchhammer, K., and Selim, K. A. (2020). Carbon/nitrogen homeostasis control Appel, J., Phunpruch, S., Steinmüller, K., and Schulz, R. (2000). The bidirectional hydrogenase of Synechocystis sp. PCC 6803 works as an electron valve during photosynthesis. Arch. Microbiol. 173, 333–338. doi: 10.1007/s002030000139 in cyanobacteria. FEMS Microbiol. Rev. 44, 33–53. doi: 10.1093/femsre/ fuz025 Gibson, D. G., Young, L., Chuang, R.-Y., Venter, J. C., Hutchison, C. A., and Smith, H. O. (2009). Enzymatic assembly of DNA molecules up to several hundred kilobases. Nat. Methods 6, 343–345. doi: 10.1038/nmeth.1318 Artz, J. H., Tokmina-Lukaszewska, M., Mulder, D. W., Lubner, C. E., Gutekunst, K., Appel, J., et al. (2020). The structure and reactivity of the HoxEFU complex from the cyanobacterium Synechocystis sp. PCC 6803. J. Biol. Chem. 295, 9445–9454. doi: 10.1074/jbc.RA120.013136 Greening, C., Biswas, A., Carere, C. R., Jackson, C. J., Taylor, M. C., Stott, M. B., et al. (2015). Genomic and metagenomic surveys of hydrogenase distribution indicate H2 is a widely utilised energy source for microbial growth and survival. ISME J. 10, 1–17. doi: 10.1038/ismej.2015.153 Barz, M., Beimgraben, C., Staller, T., Germer, F., Opitz, F., Marquardt, C., et al. (2010). Distribution analysis of hydrogenases in surface waters of marine and freshwater environments. PLoS ONE 5, e13846. doi: 10.1371/journal.pone.0013846 Gutekunst, K., Chen, X., Schreiber, K., Kaspar, U., Makam, S., and Appel, J. (2014). The Bidirectional NiFe-hydrogenase in Synechocystis sp. PCC 6803 is reduced by ﬂavodoxin and ferredoxin and is essential under mixotrophic, nitrate-limiting conditions. J. Biol. Chem. 289, 1930–1937. doi: 10.1074/jbc.M113.526376 Beimgraben, C., Gutekunst, K., Opitz, F., and Appel, J. (2014). hypD as a marker for [NiFe]-hydrogenases in microbial communities of surface waters. Appl. Environ. Microbiol. 80, 3776. doi: 10.1128/AEM.00690-14 Boehm, M., Nield, J., Zhang, P., Aro, E.-M., Komenda, J., and Nixon, P. J. (2009). Structural and mutational analysis of band 7 proteins in the cyanobacterium Synechocystis sp. strain PCC 6803. J. Bacteriol. 191, 6425–6435. doi: 10.1128/JB.00644-09 H. K. Lichtenthaler (1987). Chlorophylls and carotenoids: Pigments of photosynthetic biomembranes. Methods in Enzymol. ACKNOWLEDGMENTS The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmicb. 2022.896190/full#supplementary-material We thank Cay Kruse for assistance with microscopy and the lab of Karl Forchhammer for help with the quantiﬁcation of cyanophycin. FUNDING This study was supported by grants from the China Scholarship Council (CSC) (Grant # 201406320187), the German Ministry of Science and Education (BMBF FP309), and the German Science Foundation (DFG GU1522/2-1, GU1522/5-1, and FOR2816). May 2022 \| Volume 13 \| Article 896190 14 Cyanobacterial NiFe-Hydrogenase Under Oxic Conditions Burgstaller et al. REFERENCES Glycolytic shunts replenish the calvin–benson– bassham cycle as anaplerotic reactions in cyanobacteria. Mol. Plant. 13, 471–482. doi: 10.1016/j.molp.2020.02.002 Cournac, L., Guedeney, G., Peltier, G., and Vignais, P. M. (2004). Sustained photoevolution of molecular hydrogen in a mutant of Synechocystis sp. strain PCC 6803 deﬁcient in the Type I NADPH-dehydrogenase complex. J Bacteriol. 186, 1737–1746. doi: 10.1128/JB.186.6.1737-1746.2003 McIntosh, C. L., Germer, F., Schulz, R., Appel, J., and Jones, A. K. (2011). The [NiFe]-hydrogenase of the cyanobacterium Synechocystis sp. PCC 6803 works bidirectionally with a bias to H2 production. J. Am. Chem. Soc. 133, 11308–11319. doi: 10.1021/ja203376y Dutta, I., and Vermaas, W. F. J. (2016). The electron transfer pathway upon H2 oxidation by the NiFe bidirectional hydrogenase of Synechocystis sp. PCC 6803 in the light shares components with the photosynthetic electron transfer chain in thylakoid membranes. Int. J. Hydrogen Energy 41, 11949–11959. doi: 10.1016/j.ijhydene.2016.01.172 Messineo, L. (1966). Modiﬁcation of the Sakaguchi reaction: Spectrophotometric determination of arginine in proteins without previous hydrolysis. Arch. Biochem. Biophys. 117, 534–540. doi: 10.1016/0003-9861(66)90094-4 Eckert, C., Boehm, M., Carrieri, D., Yu, J., Dubini, A., Nixon, P. J., et al. (2012). Genetic analysis of the hox hydrogenase in the cyanobacterium Synechocystis sp. PCC 6803 reveals subunit roles in association, assembly, maturation, and function. J. Biol. Chem. 287, 43502–43515. doi: 10.1074/jbc.M112.392407 Mullineaux, C. W. (2014). Co-existence of photosynthetic and respiratory activities in cyanobacterial thylakoid membranes. Biochim. Biophys. Acta Bioenerget. 1837, 503–511. doi: 10.1016/j.bbabio.2013.11.017 Nielsen, M., Revsbech, N. P., and Kühl, M. (2015). Microsensor measurements of hydrogen gas dynamics in cyanobacterial microbial mats. Front. Microbiol. 6, 726. doi: 10.3389/fmicb.2015.00726 Elbahloul, Y., Krehenbrink, M., Reichelt, R., and Steinbüchel, A. (2005). Physiological conditions conducive to high cyanophycin content in biomass of Frontiers in Microbiology \| www.frontiersin.org May 2022 \| Volume 13 \| Article 896190 15 Cyanobacterial NiFe-Hydrogenase Under Oxic Conditions Burgstaller et al. Pandelia, M.-E., Ogata, H., and Lubitz, W. (2010). Intermediates in the catalytic cycle of [NiFe] hydrogenase: functional spectroscopy of the active site. Chemphyschem 11, 1127–1140. doi: 10.1002/cphc.200900950 van Kooten, O., and Snel, J. F. H. (1990). The use of chlorophyll ﬂuorescence nomenclature in plant stress physiology. Photosyn. Res. 25, 147–150. doi: 10.1007/BF00033156 Vignais, P. M., and Billoud, B. (2007). Occurrence, classiﬁcation, and biological function of hydrogenases: an overview. Chem. Rev. 107, 4206–4272. doi: 10.1021/cr050196r Schriek, S., Kahmann, U., Staiger, D., Pistorius, E. K., and Michel, K.-P. (2009). Detection of an L-amino acid dehydrogenase activity in Synechocystis sp. PCC 6803. J. REFERENCES Exp. Bot. 60, 1035–1046. doi: 10.1093/jxb/ern352 Wang, Y., Chen, X., Spengler, K., Terberger, K., Boehm, M., Appel, J., et al. (2022). Pyruvate:ferredoxin oxidoreductase and low abundant ferredoxins support aerobic photomixotrophic growth in cyanobacteria. Elife 11, e71339. doi: 10.7554/eLife.71339 Schriek, S., Rückert, C., Staiger, D., Pistorius, E. K., and Michel, K.-P. (2007). Bioinformatic evaluation of L-arginine catabolic pathways in 24 cyanobacteria and transcriptional analysis of genes encoding enzymes of L-arginine catabolism in the cyanobacterium Synechocystis sp. PCC 6803. BMC Genom. 8, 437. doi: 10.1186/1471-2164-8-437 Wetz, M. S., and Wheeler, P. A. (2003). Production and partitioning of organic matter during simulated phytoplankton blooms. Limnol. Oceanogr. 48, 1808–1817. doi: 10.4319/lo.2003.48.5. 1808 Schuller, J. M., Birrell, J. A., Tanaka, H., Konuma, T., Wulfhorst, H., Cox, N., et al. (2019). Structural adaptations of photosynthetic complex I enable ferredoxin-dependent electron transfer. Science 363, 257–260. doi: 10.1126/science.aau3613 Wetz, M. S., and Wheeler, P. A. (2004). Response of bacteria to simulated upwelling phytoplankton blooms. Mar. Ecol. Prog. Ser. 272, 49–57. doi: 10.3354/meps272049 SÈtif, P., Shimakawa, G., Krieger-Liszkay, A., and Miyake, C. (2020). Identiﬁcation of the electron donor to ﬂavodiiron proteins in Synechocystis sp. PCC 6803 by in vivo spectroscopy. Biochim. Biophys. Acta Bioenerget. 1861:148256. doi: 10.1016/j.bbabio.2020.148256 Williams, J. G. K. (1988). “Construction of speciﬁc mutations in photosystem II photosynthetic reaction center by genetic engineering methods in Synechocystis 6803,” in Methods in Enzymology (Academic Press), 167, 766–778. doi: 10.1016/0076-6879(88)67088-1 Williams, J. G. K. (1988). “Construction of speciﬁc mutations in photosystem II photosynthetic reaction center by genetic engineering methods in Synechocystis 6803,” in Methods in Enzymology (Academic Press), 167, 766–778. doi: 10.1016/0076-6879(88)67088-1 Shimakawa, G., Shaku, K., Nishi, A., Hayashi, R., Yamamoto, H., Sakamoto, K., et al. (2014). Flavodiiron 2 and 4 proteins mediate an oxygen- dependent alternative electron ﬂow in Synechocystis sp. PCC 6803 under CO2-limited conditions. Plant Physiol. 167, 472–480. doi: 10.1104/pp.114. 249987 Conﬂict of Interest: The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Stephan, D. P., Ruppel, H. G., and Pistorius, E. K. (2000). Interrelation between cyanophycin synthesis, l-arginine catabolism and photosynthesis in the cyanobacterium Synechocystis Sp. strain PCC 6803. Zeitschrift Naturforschung 55, 927–942. Frontiers in Microbiology \| www.frontiersin.org May 2022 \| Volume 13 \| Article 896190 REFERENCES doi: 10.1515/znc-2000-11-1214 Publisher’s Note: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their aﬃliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Teeling, H., Fuchs, B. M., Becher, D., Klockow, C., Gardebrecht, A., Bennke, C. M., et al. (2012). Substrate-controlled succession of marine bacterioplankton populations induced by a phytoplankton bloom. Science 336, 608–611. doi: 10.1126/science.1218344 Theune, M. L., Hildebrandt, S., Steﬀen-Heins, A., Bilger, W., Gutekunst, K., and Appel, J. (2021). In-vivo quantiﬁcation of electron ﬂow through photosystem I – Cyclic electron transport makes up about 35% in a cyanobacterium. Biochim. Biophys. Acta Bioenerget. 1862, 148353. doi: 10.1016/j.bbabio.2020. 148353 Copyright © 2022 Burgstaller, Wang, Caliebe, Hueren, Appel, Boehm, Leitzke, Theune, King and Gutekunst. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Trautmann, D., Voss, B., Wilde, A., Al-Babili, S., and Hess, W. R. (2012). Microevolution in cyanobacteria: re-sequencing a motile substrain of Synechocystis sp. PCC 6803. DNA Res. 19, 435–448. doi: 10.1093/dnares/dss024 May 2022 \| Volume 13 \| Article 896190 Frontiers in Microbiology \| www.frontiersin.org 16
https://openalex.org/W2981285264	https://europepmc.org/articles/pmc6843952?pdf=render	English	null	Evaluation of the Effectiveness of Comprehensive Smoke-Free Legislation in Indoor Public Places in Shanghai, China	International journal of environmental research and public health/International journal of environmental research and public health	2,019	cc-by	7,584	Received: 23 August 2019; Accepted: 18 October 2019; Published: 21 October 2019 Abstract: This study evaluated compliance with the comprehensive smoke-free law in public indoor places introduced in Shanghai in March 2017. Observations and PM2.5 monitoring over 30 min intervals in 8 types of the venue were conducted three times: within a month before implementation and 3- and 12-months post implementation. Observations of evidence of smoking decreased from 66.2% before legislation to 52.8% three months after (p = 0.002) and 49.7% one year after (p < 0.001). The density of lit cigarettes also reduced signiﬁcantly after implementation (p < 0.001). When adjusting for outdoor, indoor PM2.5 levels were signiﬁcantly lower after the legislation, but only by a small amount (three months later: −0.27, p = 0.08; one year later: −0.12; p = 0.03). Evidence of compliance was weakest in farmer’s markets and bars, and smoking in male toilets did not change signiﬁcantly. The reduction in smoking was aﬀected by the management performance of their obligations. The comprehensive smoke-free law led to modest reductions in smoking and PM2.5 levels as a result, but from levels suggesting quite high levels of pre-compliance. However, compliance was limited in some areas, suggesting more eﬀort is required on management to gain better compliance in some places like farmer’s markets, bars, and toilets. Evaluation of the Eﬀectiveness of Comprehensive Smoke-Free Legislation in Indoor Public Places in Shanghai, China Yanxia Wei 1, Ron Borland 2, Pinpin Zheng 1,, Hua Fu 1, Fan Wang 3, Jingyi He 1 and Yitian Yanxia Wei 1, Ron Borland 2, Pinpin Zheng 1,, Hua Fu 1, Fan Wang 3, Jingyi He 1 and Yitian Feng 1 1 Department of Preventive Medicine and Health Education, School of Public Health, Fudan University, Shanghai 200032, China; 17111020015@fudan.edu.cn (Y.W.); hfu@fudan.edu.cn (H.F.); 16211020052@fudan.edu.cn (J.H.); 15211020019@fudan.edu.cn (Y.F.) 2 Yanxia Wei 1, Ron Borland 2, Pinpin Zheng 1,, Hua Fu 1, Fan Wang 3, Jingyi He 1 and Yitian Feng 1 1 Department of Preventive Medicine and Health Education, School of Public Health, Fudan University, Shanghai 200032, China; 17111020015@fudan.edu.cn (Y.W.); hfu@fudan.edu.cn (H.F.); 16211020052@fudan.edu.cn (J.H.); 15211020019@fudan.edu.cn (Y.F.) , , p g , , g , J gy g 1 Department of Preventive Medicine and Health Education, School of Public Health, Fudan University, Shanghai 200032, China; 17111020015@fudan.edu.cn (Y.W.); hfu@fudan.edu.cn (H.F.); 16211020052@fudan.edu.cn (J.H.); 15211020019@fudan.edu.cn (Y.F.) 2 Melbourne School of Psychological Sciences, The University of Melbourne, Melbourne 3010, Australia; rborland@unimelb.edu.au 3 Department of Politics, East China Normal University, Shanghai 200241, China; wangfan512@126.com Correspondence: zpinpin@shmu.edu.cn * Correspondence: zpinpin@shmu.edu.cn International Journal of Environmental Research and Public Health International Journal of Environmental Research and Public Health Int. J. Environ. Res. Public Health 2019, 16, 4019; doi:10.3390/ijerph16204019 Keywords: comprehensive smoke-free law; SHS; PM2.5; indoor public places; complianc 1. Introduction Secondhand smoke (SHS) remains a major public health issue in the world [1]. SHS causes more than 890,000 premature deaths globally every year [2]. Exposure to SHS can lead to cardiovascular and respiratory diseases and some adverse reproductive outcomes such as infertility, low birth weight, infant death and so on [3,4]. China is the largest producer and consumer of tobacco in the world [5] with more than half (52.1%) of males smoking regularly [6]. Smoking in indoor public places is also very common. In China, 72.4% of nonsmokers reported being exposed to SHS and more than 100,000 people die from it annually [7]. According to the 2015 national adult tobacco survey: 54.3% of respondents saw someone smoking at the workplace, 76.3% reported having seen smoking in restaurants, and even more attendees reported smoking in bars, 93.1% [8]. The only intervention shown to fully protect people from the health dangers of SHS is establishing environments that are completely smoke-free [9,10]. On 29 March 2004, the Republic of Ireland became the ﬁrst country in the world to implement 100% smoke-free legislation [11], and as at 2016 nationally Int. J. Environ. Res. Public Health 2019, 16, 4019; doi:10.3390/ijerph16204019 www.mdpi.com/journal/ijerph www.mdpi.com/journal/ijerph 2 of 11 Int. J. Environ. Res. Public Health 2019, 16, 4019 comprehensive smoke-free legislation was in place for almost 1.5 billion people in 55 countries [12]. However, there are still many countries that have not realized smoke-free. China ratiﬁed the WHO Framework Convention on Tobacco Control (FCTC) in 2005 [13], Article 8 of the FCTC calls for the adoption of the eﬀective legislative measures to achieve a 100% smoke-free compliance in all indoor public places, workplaces and public transport [14], which have not been met yet. Between 2009 and 2013, nearly half of all provincial capitals had passed tobacco control legislation [13], but all of them did not realize 100% smoke-free. In 2015, Peking enacted the most restricted tobacco control law in China and realize 100% smoke-free, then other cities started to follow it. On 1st March 2017, Shanghai implemented a comprehensive law following Peking. Article 6 stipulates that smoking is prohibited in indoor public places, indoor workplaces, and public transportation, and article 9 require managers of the above smoke-free venues to bear the obligation to post no-smoking signs, remove ashtray and ashtray equivalents and so on. 1. Introduction We believed that in China, with the largest smokers and populations in the world, national smoke-free legislation will make a great contribution to tobacco control in the world. However, until now, there are only a few pieces of evidence to demonstrate the impact of the smoke-free policy, which is essential to advocate for national law. We searched the related articles in PubMed in July 2019, there were only three articles that evaluated the 100% smoke-free law in China. Two of them from the same population survey, which showed the reported SHS exposure rates in public places declined [15] and the smoking prevalence in adults also decreased [16] in Beijing after the one year implementation. In the other article also from Beijing, less smoking was observed (from 40.3% to 14.8%) and posting of no-smoking signage increased (from 52.6% to 82.4%) in restaurants after the law came into eﬀect [17]. These evidences are too weak to understand law enforcement in China because there are only a few simple indexes in these studies. Our study assessed compliance with the Shanghai law using multiple methods, direct observation, monitoring of air quality and observation of eﬀorts to enhance compliance, to know more details of law enforcement and provide evidence for national law. We hope this study can provide sound evidence to promote a smoke-free China and provide a reference for policymaking in tobacco control in other countries. 2.2. Data Collection Evidence of smoking is a binary variable based on observations of any observed smoking, presence of cigarette butts and the smell of smoke. The number of lit cigarettes was recorded using the average of four observation points (0 min, 10 min, 20 min, 30 min) within 30 min. We also recorded the number of patrons at each point and reported mean levels. The density of lit cigarettes equals the number of lit cigarettes divided by the number of patrons multiplied by 100 to give percentages. Observations of smoking were also made in the male toilets. The venue manager’s performance of compliance obligations was assessed as a binary variable: doing all of posting no-smoking signs, not having a smoking section and no ashtrays and ashtray equivalents present versus all others. g g y y q p PM2.5 levels were recorded using a TSI SidePak Model AM510 Personal Aerosol Monitor) ﬁtted with a 2.5 µm impactor. The airﬂow rate was set at 1.7 L/min, and the logging interval was 1 min. A calibration factor of 0.295 was applied to all raw measurements to correct for the properties of SHS particles [18]. The monitor was calibrated to zero each time before use. The investigators carried the monitor in a small bag with a short length of tubing attached to the inlet and left protruding to the outside. When entering an indoor public place, the investigator sat in the central area or stood in the corridor and staircase and detected PM2.5 level for about 30 min. The PM2.5 concentration in toilets was also measured for 30 min. The monitor was placed as close as possible to the breathing zone, away from any ventilation opening and not within 1 m of any smoking people. Outdoor PM2.5 measurement was carried out for 5 min before entering each venue. The outdoor inspection point was located away from any obvious concentrated source of PM2.5 as practically possible (e.g., venting outlets from kitchens) or sources of ﬁre. 2.1. Procedure Observations and PM2.5 monitoring were conducted within a month before implementation (February 2017) and 3 months post implementation in May 2017 and again 12 months post implementation. The same methods were used in all three observation periods. Indoor public places were included if they met the following two criteria: smoking was allowed before new law; entry was allowed. Multistage sampling was adopted to recruit venues. Eight types of venues were targeted: oﬃce buildings, hotels (four-star hotel and a few low-level hotels), restaurants (Chinese restaurant), bars, Karaoke (KTV), gaming rooms, farmer’s markets, buses and railway stations. First-stage: all eight urban regions (Huangpu, Pudong, Xuhui, Changning, Jing’an, Putuo, Hongkou, and Yangpu) and four of eight suburbs (Minhang, Baoshan, Songjiang, Jiading) in Shanghai were selected, except for bus and railway stations, which are limited in number, so we chose the eight main stations. We chose 20 venues from each of the other 7 venue types (2 each from the urban regions and one from each suburb). We chose staircases as the inspection point in oﬃce buildings because we could not enter oﬃce areas. Moreover, according to our previous observation, smoking in oﬃce buildings is also likely to occur at staircases. For KTVs and hotels, patrons mainly stay in their private room. Considering this limitation, we chose the corridor of KTVs and the lobby in hotels as the inspection point. For other venues, the inspection point is the main business area. Due to the high smoking prevalence among males, where possible, we also made observations in the male toilets. 3 of 11 Int. J. Environ. Res. Public Health 2019, 16, 4019 The three trained investigators formed a group and surveyed in one district each day during the busy hours of each venue. To ensure validity, investigators entered each venue as guests, investigated for 30 min. One observed and ﬁlled the questionnaire, and two investigators were responsible for the PM2.5 concentration detection at the corresponding indoor inspection points. Survey time ranged from 8 am at the farmer’s markets to 11 pm in bars, depending on the peak business hours of diﬀerent venues. Oﬃce buildings were only observed on weekdays. We followed the same schedule in the three-wave surveys. 3.1. Sample Characteristics Sampling details are shown in Table 1. In the ﬁrst survey wave, data from 1 bar, 1 restaurant were lost because of a machine fault, and 1 farmer’s market and 1 oﬃce building’s toilet could not be located, so we did not try to resurvey them. Besides, we choose not to survey male toilets in gaming rooms, hotels, and farmer’s markets, because most of these venues did not have their toilet. In total there were 145 indoor public places with 85 male toilets that were monitored three times. There was a small amount of missing data, and none of them exceeded 20%. Table 1. Sample details. Site Inspection Point Number Time Oﬃce building Staircase/male toilet 20/20 10:00–11:30 Restaurant Main hall/male toilet 19/19 11:30–14:00 17:00–19:00 Bar Main hall/male toilet 19/19 21:00–23:00 Gaming room Main hall/male toilet 20/0 15:30–17:00 Hotel Lobby/male toilet 20/0 14:00–15:30 KTV Corridor/male toilet 20/20 19:00–21:00 Farmer’s market Main hall/male toilet 19/0 8:00–10:00 Bus and railway station Waiting hall/male toilet 8/8 10:00–17:00 3.2. Comparation of Observational Items before and after Legislation Table 1. Sample details. 3.2. Comparation of Observational Items before and after Legislation 3.2. Comparation of Observational Items before and after Legislation Notably, the level of observed smoking-related objects was limited before the ban, suggesting some degree of either enforcement or voluntary changes in behavior in anticipation (see Table 2). Compared with before the legislation, tobacco smell, cigarette butts, lit cigarettes and total evidence of smoking all decreased signiﬁcantly at three months and stayed at the lower level one year later. In all, evidence of smoking signiﬁcantly decreased from 66.2% (before legislation) to 52.8% (three months later) to 49.7% (one year later), the latter small change not being signiﬁcant. Similarly, the fewer density of cigarettes was observed after legislation. In venues where smoking occurred at all three times, the density of cigarettes observed also declined (N = 28, p 1 < 0.001, p 2 = 0.005). The number of total people in venues was similar (p > 0.05) across the three survey periods. Table 2. The comparation of observational items before and after legislation. Observational Items. 2.3. Statistical Analysis The numbers of people in venues and the density of lit cigarettes were not normally distributed, so mean, median, and interquartile range is reported. We chose not to record the number of people and cigarettes in railway and bus stations because there were too many people to count, therefore all the analysis concluding the number of people and cigarettes did not include the railway and bus stations. The comparison of dichotomous items of diﬀerent times using Generalized Linear Mixed Model (GLMM) with each place as random eﬀects. The package is “mlmRev” in R. Binomial family with logit link were used. Continuous variables compared using nonparametric test for related samples in SPSS for they do not meet the requirement of GLMM. Time-weighted average PM2.5 concentrations were calculated for each visit. Because the PM2.5 concentrations were log normally distributed geometric means (GM) and 95% CI are reported. We compared values between indoor and outdoor, indoor and toilet using Paired Sampled T-Tests. Three times of PM2.5 value was compared using the “nlme” package in R. Gaussian family with logit link were used. All analyses were conducted in SPSS 25.0 (SPSS, IBM, Armonk, NY USA) and R 3.5.2 (R Foundation for Statistical Computing, Vienna, Austria) (Manufacture, city, state abbrev., country). 4 of 11 Int. J. Environ. Res. Public Health 2019, 16, 4019 3.1. Sample Characteristics Smoking at doorways did not change signiﬁcantly three months later, but it became signiﬁcantly higher one year after legislation (Table 2). As indicated in Table 3, in venues where management complied better with the law (i.e., moved from failure to meeting obligations), evidence of smoking and the density of lit cigarettes both decreased signiﬁcantly at both 3 months and one year later. Where obligations were met on both occasions, evidence of smoking dropped signiﬁcantly at 3 months (but not signiﬁcantly at one year), while the density of lit cigarettes reduced both after 3 months and one year. Similarly, where obligations were not met at either time the density of lit cigarettes became signiﬁcant less one year later, even though overall compliance did not change. In the small number of venues where performance of obligations declined there were no signiﬁcant changes in either outcome. Not surprisingly, overall the level of evidence of smoking and the density of lit cigarettes in venues was lower where management complied with their obligations compared with those not fully compliant. Table 3. Evidence of smoking and density of lit cigarettes following the change of venue manager’s performance of obligation. Table 3. Evidence of smoking and density of lit cigarettes following the change of venue manager’s performance of obligation. Table 3. Evidence of smoking and density of lit cigarettes following the change of venue manager’s performance of obligation. 3.1. Sample Characteristics Time p 1 p 2 Before Legislation Three Months Later One Year Later Evidence of smoking (% (N)) Tobacco smell 53.8 (78) 29.2 (42) 29.7 (43) <0.001 <0.001 Cigarette butts 43.7 (62) 29.7 (43) 26.6 (38) 0.003 <0.001 Lit cigarettes 52.4 (76) 40.7 (59) 31.0 (45) * 0.020 <0.001 Any 66.2 (94) 52.8 (76) 49.7 (71) 0.002 <0.001 Density of lit cigarettes (mean/median (IR)) N = 122 5.99/1.23 2.71/0.00 1.50/0.00 0.005 <0.001 (0.00~8.37) (0.00~2.49) (0.00~0. 93) Number of people (mean/median (IR)) N = 127 34.35/18.00 32.48/19.00 30.32/17.25 0.474 0.115 (7.00~43.75) (7.25~32.00) (6.25~37.0) Performance of obligation (% (N)) Posting no-smoking signs 73.8 (107) 88.3 (128) 91.0 (132) <0.001 <0.001 Have no smoking section 95.8 (138) 98.6 (142) 98.6 (143) 0.126 0.124 Have no ashtray and ashtray equivalents 66.9 (95) 84.1 (116) 89.0 (129) <0.001 <0.001 All 51.1 (72) 73.6(103) 80.0(116) <0.001 <0.001 Smoking in toilet (% (N)) 48.2 (40) 35.4 (29) 46.3 (38) 0.075 0.800 Smoking at doorway (% (N)) 21.5 (31) 29.6 (42) 39.3 (57) * 0.096 <0.001 Note: p 1 is the outcome of comparation between before legislation and three months later, p 2 is between before legislation and one year later. Signiﬁcant codes indicate between three months later and one year later: ‘*’ < 0.01, ‘’ < 0.05. Table 2. The comparation of observational items before and after legislation. Note: p 1 is the outcome of comparation between before legislation and three months later, p 2 is between before legislation and one year later. Signiﬁcant codes indicate between three months later and one year later: ‘*’ < 0.01, ‘’ < 0.05. As for venue management’s compliance with the law, there were signiﬁcantly more no-smoking signs and less ashtray and ashtray equivalents after legislation and the presence of smoking section Int. J. Environ. Res. Public Health 2019, 16, 4019 5 of 11 did not change signiﬁcantly. There were already a high proportion of venues that posted no-smoking signs, had no smoking sections and ashtrays before legislation. The number of venues that performed all of the above three obligations raised signiﬁcantly after law. For all of the above items except for lit cigarettes, the diﬀerences between three months later and one year later were not signiﬁcant. Smoking in male toilets did not signiﬁcantly change after legislation and in nearly half (47.4%) of the venues, someone was observed smoking in the toilet one year after the new law was introduced. Note: p 1 is the outcome of comparation between before legislation and three months later, p 2 is between before legislation and one year later. 3.1. Sample Characteristics Evidence of Smoking (%) Density of Lit Cigarettes (Mean/Median (IR)) Perform of Obligation Both Comply all Neither Comply all Increase from Baseline Decrease from Baseline Both Comply all Neither Comply all Increase from Baseline Decrease from Baseline Before legislation 59.6 85.2 70.0 40.0 3.84/0.00 14.50/10.73 6.91/1.27 1.32/0.00 (0.00~2.94) (3.95~18.75) (0.00~9.50) (0.00~2.17) Three months later 42.1 81.5 45.0 60.0 0.67/0.00 7.75/7.41 2.34/0.00 1.82/0.00 (0.00~0.77) (1.59~14.29) (0.00~0.92) (0.00~2.77) p 1 0.031 1.000 0.013 0.500 0.002 0.136 0.003 0.593 (N) (N = 57) (N = 27) (N = 40) (N = 10) (N = 50) (N = 23) (N = 38) (N = 9) Before legislation 55.7 89.5 70.8 55.6 1.54/0.00 17.44/13.03 7.01/3.85 2.79/0.00 (0.00~2.61) (6.41~24.11) (0.00~9.86) (0.00~6.25) One year later 47.5 78.9 43.8 55.6 0.80/0.00 3.83/1.03 1.44/0.00 1.37/0.00 (0.00~0.94) (0.00~8.40) (0.00~0.00) (0.00~1.36) p 2 0.332 0.500 0.004 1.000 0.031 0.001 <0.001 0.225 (N) (N = 61) (N = 19) (N = 48) (N = 9) (N = 53) (N = 18) (N = 47) (N = 9) Note: p 1 is the outcome of comparation between before legislation and three months later, p 2 is between before legislation and one year later. Note: p 1 is the outcome of comparation between before legislation and three months later, p 2 is between before legislation and one year later. 3.3. Change of Evidence of Smoking and Density of Cigarettes after Legislation in Diﬀerent Venues and Districts 3.3. Change of Evidence of Smoking and Density of Cigarettes after Legislation in Diﬀerent Venues and Districts 3.3. Change of Evidence of Smoking and Density of Cigarettes after Legislation in Diﬀerent Venues and Districts. Table 4 presents data on trends by individual venues. Due to the small samples, there is very limited power especially for the assessment of signs of smoking. That said, one-year levels were lower in all seven venues, which had not been subject to prior total bans, but increased marginally (non-signiﬁcantly) in stations where it was already prohibited. Similarly, there were numerical declines in the density of lit cigarettes in all seven venues, and this was signiﬁcant in two cases at one year (farmers markets and oﬃce buildings) and clear trends in others. It is notable that at one year, no lit cigarettes were observed in more than half of all venues. There was most serious evidence of smoking in the following venues both before and after legislation: farmer’s markets, oﬃce buildings, and bars. The median of density of lit cigarettes in farmer’s markets and bars were highest compared to all the other places, although they all had signiﬁcantly and near signiﬁcant dropped compared with before legislation. After legislation, evidence of smoking and the density of lit cigarettes both dropped signiﬁcantly in both urban and suburban venues. Levels were comparable between these two districts types one year after legislation as seen in Table 4. 6 of 11 Int. J. Environ. Res. Public Health 2019, 16, 4019 Table 4. Changes of evidence of smoking and density of lit cigarettes in diﬀerent venues and districts. Table 4. Changes of evidence of smoking and density of lit cigarettes in diﬀerent venues and districts. 3.4. PM2.5 Concentrations 3.4. PM2.5 Concentrations PM2 5 l i l Change of PM2.5 value after adjusting for outdoor background value. Table 5. Change of PM2.5 value after adjusting for outdoor background value. Indoor PM2.5 Value Toilet PM2.5 Value B 1 S.E. 2 p-value B 1 S.E. 2 p-value Intercept −2.30 0.14 < 0.001 −1.52 0.22 <0.001 Legislation (reference = before legislation) Three months later −0.27 0.08 < 0.001 0.01 0.18 0.954 One year later −0.12 0.03 < 0.001 0.45 0.12 <0.001 Outdoor PM2.5 value 5.67 0.74 < 0.001 2.29 1.41 0.107 B 1: regression coefficient. S.E. 2: standard error. Table 5. Change of PM2.5 value after adjusting for outdoor background value. Indoor PM2.5 Value Toilet PM2.5 Value B 1 S.E. 2 p-value B 1 S.E. 2 p-value Intercept −2.30 0.14 < 0.001 −1.52 0.22 <0.001 Legislation (reference = before legislation) Three months later −0.27 0.08 < 0.001 0.01 0.18 0.954 One year later −0.12 0.03 < 0.001 0.45 0.12 <0.001 Outdoor PM2.5 value 5.67 0.74 < 0.001 2.29 1.41 0.107 B 1: regression coeﬃcient. S.E. 2: standard error. 3.3. Change of Evidence of Smoking and Density of Cigarettes after Legislation in Diﬀerent Venues and Districts Venue No Lit Cigarettes in all of Three Times (% (N)) Before Legislation Three Months Later One Year Later p 1 p 2 Farmer’s market Evidence of smoking (% (N)) 94.4 (17) 94.7 (18) 88.9 (16) 0.969 0.554 Density of lit cigarettes (mean/median (IR)) N = 18 0.0 (0) 3.02/2.69 (1.28~4.03) 1.12/0.71 (0.51~1.45) 1.55/1.31 (0.00~2.55) 0.002 0.008 Hotel Evidence of smoking (% (N)) 40.0 (8) 21.1 (4) 10.0 (2) 0.149 0.065 Density of lit cigarettes (mean/median (IR)) N = 18 66.7 (12) 1.47/0.00 (0.00~0.96) 1.19/0.00 (0.00~0.00) 0.04/0.00 (0.00~0.00) 0.934 0.560 Bar Evidence of smoking (% (N)) 83.3 (15) 78.9 (15) 63.2 (12) 0.802 0.163 Density of lit cigarettes (mean/median (IR)) N = 16 0.6(1) 17.46/14.53 (4.79~23.52) 4.79/3.24 (0.00~8.10) 5.44/0.44 (0.00~11.77) 0.077 0.052 Oﬃce building Evidence of smoking (% (N)) 90.0(18) 50.0(10) 70.0(14) 0.014 0.116 Density of lit cigarettes (mean/median (IR)) N = 16 43.8(7) 8.14/2.58 (0.00~8.82) 2.19/0.00 (0.00~3.14) 0.62/0.00 (0.00~0.00) 0.289 0.034 Restaurant Evidence of smoking (% (N)) 52.6 (10) 21.1 (4) 27.8 (5) 0.035 0.099 Density of lit cigarettes (mean/median (IR)) N = 17 58.9(10) 3.54/0.00 (0.00~5.15) 0.25/0.00 (0.00~0.00) 0.64/0.00 (0.00~0.00) 0.123 0.198 KTV Evidence of smoking (% (N)) 60.0 (12) 60.0 (12) 55.0 (11) 1.000 0.699 Density of lit cigarettes (mean/median (IR)) N = 17 47.1(8) 1.81/0.00 (0.00~1.92) 5.34/0.00 (0.00~6.20) 0.28/0.00 (0.00~0.33) 0.391 0.932 Gaming room Evidence of smoking (% (N)) 65.0 (13) 55.0 (11) 36.8 (7) 0.273 0.055 Density of lit cigarettes (mean/median (IR)) N = 20 35.0 (7) 7.43/7.42 (0.00~14.89) 4.10/0.39 (0.00~8.33) 2.07/0.00 (0.00~1.93) 0.385 0.082 Railway and bus station Evidence of smoking (% (N)) 14.3 (1) 25.0 (2) 25.0 (2) 0.609 0.609 District Urban Evidence of smoking (% (N)) 60.5 (69) 51.3 (59) 45.6 (52) 0.061 0.005 Density of lit cigarettes (mean/median (IR)) N = 95 37.9(36) 5.68/1.00 (0.00~0.080) 2.50/0.00 (0.00~2.00) 1.34/0.00 (0.00~1.00) <0.001 0.007 Suburb Evidence of smoking (% (N)) 89.3 (25) 58.6 (17) 62.1 (18) 0.008 0.013 Density of lit cigarettes (mean/median (IR)) N = 27 29.7(8) 7.22/3.00 (0.00~10.00) 3.52/1.00 (0.00~6.00) 2.15/0.00 (0.00~3.00) 0.276 0.048 Note: p 1 is the outcome of comparation between before legislation and three months later, p 2 is between before legislation and one year later. 7 of 11 7 of 11 Int. J. Environ. Res. Public Health 2019, 16, 4019 Int. J. Environ. Res. Public Health 2019, 16, x 7 of 11 7 of 11 3.4. PM2.5 Concentrations 3.4. PM2.5 Concentrations PM2 5 l i l PM2.5 values in male toilets were signiﬁcantly higher than in the main indoor areas, and the indoor value was signiﬁcantly higher than outdoors in all three surveys. There was a signiﬁcant diﬀerence among the outdoor values in three waves (p < 0.001), with the lowest level detected before the legislation and the highest detected one year after the legislation (Figure 1). Generalized mixed model analysis indicated that outdoor PM2.5 concentrations signiﬁcantly inﬂuenced the indoor value (positive relationship as would be expected), but had no signiﬁcant eﬀect on the toilet levels. Levels were slightly lower overall after the ban suggesting a small eﬀect on PM2.5 level in the venue. When adjusting for outdoor values, indoor PM2.5 value three months later and one year later all signiﬁcantly decreased from before legislation, but compared with concentrations after three months, PM2.5 values after one year increased signiﬁcantly (B = 0.15, p = 0.011) even though still signiﬁcantly lower than the baseline. All these diﬀerences were quite small. However, PM2.5 values recorded in toilets were signiﬁcantly increased one year after legislation. (Table 5) PM2.5 values in male toilets were significantly higher than in the main indoor areas, and the indoor value was significantly higher than outdoors in all three surveys. There was a significant difference among the outdoor values in three waves (p < 0.001), with the lowest level detected before the legislation and the highest detected one year after the legislation (Figure 1). Generalized mixed model analysis indicated that outdoor PM2.5 concentrations significantly influenced the indoor value (positive relationship as would be expected), but had no significant effect on the toilet levels. Levels were slightly lower overall after the ban suggesting a small effect on PM2.5 level in the venue. When adjusting for outdoor values, indoor PM2.5 value three months later and one year later all significantly decreased from before legislation, but compared with concentrations after three months, PM2.5 values after one year increased significantly (B = 0.15, p = 0.011) even though still significantly lower than the baseline. All these differences were quite small. However, PM2.5 values recorded in toilets were significantly increased one year after legislation. (Table 5) Figure 1. Description of PM2.5 value. Figure 1. Description of PM2.5 value. Figure 1. Description of PM2.5 value. Figure 1. Description of PM2.5 value. Table 5. Change of PM2.5 value after adjusting for outdoor background value. Table 5. 4. Discussion One year 4. Discussion One year after implementation, Shanghai s comprehensive smoke-free law has achieved significant reductions in smoking and parallel improvements in indoor air quality. However, the latter effects are small, and only apparent when controlling for outdoor air quality. Evidence of smoking dropped from 66.2% before legislation to 52.8% after three months and to 49.7% after one year, and totally reduced by 25%. The density of cigarettes in the venue also decreased, even in some venues where smoking occurred and in all three survey periods this decrease remained significant. Notably, PM2.5 levels in all indoor public places slightly but significantly dropped after the law was implemented. One year after implementation, Shanghai’s comprehensive smoke-free law has achieved signiﬁcant reductions in smoking and parallel improvements in indoor air quality. However, the latter eﬀects are small, and only apparent when controlling for outdoor air quality. Evidence of smoking dropped from 66.2% before legislation to 52.8% after three months and to 49.7% after one year, and totally reduced by 25%. The density of cigarettes in the venue also decreased, even in some venues where smoking occurred and in all three survey periods this decrease remained signiﬁcant. Notably, PM2.5 levels in all indoor public places slightly but signiﬁcantly dropped after the law was implemented. 8 of 11 Int. J. Environ. Res. Public Health 2019, 16, 4019 The venue managers showed good compliance with the legislation. Less than one month before the new law, there was already a high proportion of venues posting no-smoking signs (73.8%), having no smoking section places (95.8%) and ashtray and its equivalents (66.9%). It is reported that in three main railway stations [19] and some hospitality places [20], smoking sections were abolished just before the legislation. After legislation, more venues posted no-smoking signs and canceled the provision of ashtrays and its equivalents. When we combined all three items, the total performance of obligation also increased. It is clear from the data, that smokers’ behavior is aﬀected by management, with smoker’s compliance higher when managers perform their obligations better when controlled for legislation. Even in venues where the compliance of management didn’t improve after the new law, smoking in indoor places signiﬁcantly improved, which suggests that the legislation itself can change smoker’s behavior to some extent independent of management behavior. Further, our results demonstrate that the ﬂow of people did not reduce after legislation, which is the same as other similar researches [21–23]. 4. Discussion One year 4. Discussion Some studies even found smoke-free environments had positive impacts on venue proﬁt and value [24,25], which should encourage operators to enhance enforcement. Smoking in male toilets did not change at all, with smoking detected in almost half of the venues (47.4%) one-year post ban. Toilets are places that can be easily ignored by enforcers and management. Another interesting phenomenon is the increasing number of observed smokers smoking in doorways when smoking was prohibited in the venue. Surada et al. found SHS in main entrances (outdoors) were very similar to those in adjacent indoors and SHS from outdoors settings drifts to indoors [26]. Laws need to include prohibitions on smoking in doorways as patrons need to enter or leave through them, and there is a risk of the smoke contaminating the indoor areas. In this study smoking and the density of cigarettes decreased three months later and one year after the ban was formally implemented, but overall there was no change between three months and one year later. In contrast, Fong’s study showed that smoke-free policies in France lead to the near total elimination of smoking after one year and continuously reduced ﬁve years later [27]. Other studies also certiﬁed the constant and stronger eﬀect of comprehensive legislation [28,29]. There is strong evidence that 100% smoke-free law had a sustained eﬀect and can totally stop observed smoking. Shanghai’s comprehensive smoke-free law may be less successful compared with other countries at this point, so more eﬀorts should be put into law enforcement. There was some variation in the extent of compliance with obligations and compliance by smokers across venues with both poor in some venues especially farmer’s markets and bars, but even here, the density of smoking declined. This supports the view that Shanghai’s legislation is partially successful with occasional violations, common in some places, but less smoking overall. The legislation has been more successful in reducing the number of people smoking rather than creating environments where there is no smoking. One year after the new law, over half of the farmer’s markets (88.9%), oﬃce buildings (70.09%), and bars (63.16%) had evidence of smoking. Bars and farmer’s markets had both the highest evidence of smoking and the density of cigarettes. 4. Discussion One year 4. Discussion International studies have shown that there are more smoking in bars than in other places [30–32], which is consistence with the ﬁndings in our study as the highest PM2.5 levels and number of cigarettes occurred in bars. However, almost no study has focused on smoking in farmer’s markets where we also found high levels of non-compliance demonstrating that such places should be a focus for improving compliance. If compliance is to be improved it is likely to involve further public education about the importance of protecting non-smokers from passive smoking, and eﬀorts targeting management of areas where compliance with their obligations was particularly poor. As the results show, the outdoor PM2.5 values were much higher than most of the other similar research, and the three-time waves of outdoor levels were signiﬁcantly diﬀerent, so the indoor PM2.5 concentrations did not decrease directly as other researches [33,34]. When adjusting outdoor background value, the PM2.5 level was slightly dropped three months later then slightly increased one year later but still lower than before legislation. The SHS in indoor public places declined following the reduction of smoking. Protano et al. also found the PM concentrations closely related with smoking in Int. J. Environ. Res. Public Health 2019, 16, 4019 9 of 11 indoor places [35]. Male toilets had higher PM2.5 concentrates than indoor and its PM2.5 levels slightly but signiﬁcantly increased one year later. The less but signiﬁcant change of PM2.5 concentration is relative to our very low baseline number of cigarettes (0.25 (0.00~1.50)). The legislation had some eﬀect on SHS in the venue, but it did not remove it completely and the SHS in male toilets even increased. When combined with our observational outcome that smokers in male toilets did not drop after the new laws were introduced, it is clear that smoking in toilets needs further attention by compliance enforcement. p One limitation of our study is that our pre-ban survey was too close in time to when the law came into force. A lot of changes appear to have happened, presumably because of advocacy, and/or preparation for the formal implementation, so what we measured at baseline almost certainly reﬂects a higher level of control over indoor smoking than we would have observed before the law was announced. This means that the eﬀects we measured may be somewhat conservative and the actual improvements are likely to be greater. 5. Conclusions The comprehensive smoke-free law can eﬀectively reduce smoking in indoor public places and protect nonsmokers from SHS. Our study detailed the areas that need to be enhanced in law enforcement and the new challenges that arise after legislation is introduced, which are the potential problems that many other countries in the world may encounter in the enforcement of tobacco control legislation since it is diﬃcult to totally eliminate smoking in indoor public places even in countries with national comprehensive law, like France [27], Italy [33], Ireland [39], Uruguay [40] and so on. Importantly, we also found that management participation is an important factor in the elimination of smoking in public places and this needs to be the target of compliance enforcement. Author Contributions: Conceptualization, P.Z.; methodology, P.Z., F.W., H.F. and R.B.; software, Y.W.; validation, P.Z., H.F. and R.B.; formal analysis, Y.W. and R.B.; investigation, Y.W., J.H. and Y.F.; resources, P.Z.; data curation, P.Z.; writing—original draft preparation, Y.W.; writing—review and editing, R.B. and P.Z.; supervision, P.Z. and R.B.; project administration, P.Z.; funding acquisition, P.Z. Funding: This study is supported by the National Natural Science Foundation of China [71573047]; and Bloomberg Initiative to Reduce Tobacco Use [China-RI-16-02]. Acknowledgments: The authors would like to thank all of investigators for their role in data collection. They are also grateful to Michael Le Grande for English language edition. Conﬂicts of Interest: The authors declare no conﬂict of interest. 4. Discussion One year 4. Discussion Secondly, we only chose the lobby of hotels and the corridor of KTVs because of fund limitations. Some studies have found that there are serious SHS in guest rooms of hotels and KTVs [36,37], so our ﬁndings may not generalize areas where we did not survey. Thirdly, PM2.5 value is not a sensitive index for SHS when there is a small number of cigarettes and when the outside level is relatively high, as it was in our study. Although we adjusted the outdoor value and found an eﬀect, there is some doubt over the magnitude and thus the likely beneﬁts. That said, the levels we detected are consistent with indoor exposure typically being quite low post-ban, and thus of limited concern. Fourthly, This study adopted the before-and-after studies with no control area for reference, which were unable to control for possible confounders and changes in secular trends over time [38]. These unsolvable changes may be small within one year as we did not observe obvious changes from venues. But for much longer investigation periods, ﬁnd another area for control is better. p [ ] 2. Fact Sheet on Tobacco. Available online: https://www.who.int/en/news-room/fact-sheets/detail/tobacco (accessed on 20 May 2019). . Protano, C.; Vitali, M. The new danger of thirdhand smoke: Why passive smoking does not sto secondhand smoke. Environ. Health Perspect. 2011, 119, a422. [CrossRef] References 10 of 11 Int. J. Environ. Res. Public Health 2019, 16, 4019 3. Gong, X.; Luo, X.; Ling, L. Prevalence and associated factors of secondhand smoke exposure among internal Chinese migrant women of reproductive age: Evidence from China’s labor-force dynamic survey. Int. J. Environ. Res. Public Health 2016, 13, 371. [CrossRef] 4. Llewellyn, D.J.; Lang, I.A.; Langa, K.M.; Naughton, F.; Matthews, F.E. Exposure to secondhand smoke and cognitive impairment in non-smokers: National cross sectional study with cotinine measurement. BMJ 2009, 338, b462. [CrossRef] 5. Li, S.; Meng, L.; Chiolero, A.; Ma, C.; Xi, B. Trends in smoking prevalence and attributable mortality in China, 1991–2011. Prev. Med. 2016, 93, 82–87. [CrossRef] 6. Zhao, J.; Li, X.; Stewart, S.L.; Gao, W.; Qi, F.; Zhang, L.; Pang, Z.; Qiao, Q.; Ning, F.; Tong, E. Cigarette smoking and secondhand smoke exposure before and after a tobacco-free Olympic policy period: Qingdao, China. Nicotine Tob. Res. 2018. [CrossRef] 7. Chinese Center for Disease Control and Prevention. Global Adult Tobacco Survey (GATS) China 2010 Country Report; Chinese Center for Disease Control and Prevention: Beijing, China, 2010. 8. Chinese Center for Disease Control and Prevention. 2015 National Adults Tobacco Survey; Chinese Center for Disease Control and Prevention: Beijing, China, 2015. 9. Gan, Q.; Hammond, S.K.; Jiang, Y.; Yang, Y.; Hu, T.W. Eﬀectiveness of a smoke-free policy in lowering secondhand smoke concentrations in oﬃces in China. J. Occup. Environ. Med. 2008, 50, 570–575. [CrossRef] 10. Cains, T.; Cannata, S.; Poulos, R.; Ferson, M.J.; Stewart, B.W. Designated “no smoking” areas provide from partial to no protection from environmental tobacco smoke. Tob. Control 2004, 13, 17–22. [CrossRef] 11. Allwright, S.; Paul, G.; Greiner, B.; Mullally, B.J.; Pursell, L.; Kelly, A.; Bonner, B.; D’Eath, M.; McConnell, B.; McLaughlin, J.P.; et al. Legislation for smoke-free workplaces and health of bar workers in Ireland: Before and after study. BMJ 2005, 331, 1117. [CrossRef] y 12. World Health Organization. WHO Report on the GlobalL Tobacco Epidemic, 2017; World Health Organization: Geneva, Switzerland, 2017. 13. Yang, G.; Wang, Y.; Wu, Y.; Yang, J.; Wan, X. The road to eﬀective tobacco control in China. Lancet 2015, 385, 1019–1028. [CrossRef] 14. Guidelines for Implementation of Article 8. Available online: https://www.who.int/fctc/guidelines/adopted/ article_8/en/ (accessed on 20 May 2019). 15. Li, Y.Q.; Shi, J.H.; Cao, Y.; Qi, L.; Xu, L.T.; Qian, Y.L.; Liu, X.R. References Survey of exposure to second-hand smoke in residents aged 15 years and over one year after implementation of tobacco control regulation in public places in Beijing. Zhonghua Liu Xing Bing Xue Za Zhi 2019, 40, 327–330. [CrossRef] 16. Li, Y.Q.; Shi, J.H.; Cao, Y.; Qi, L.; Liu, X.R. One year after the implementation of 2015 Tobacco Control Regulation on persons aged 15 years and over tobacco use in Beijing. Zhonghua Liu Xing Bing Xue Za Zhi 2018, 39, 1188–1192. [CrossRef] 17. Xiao, L.; Jiang, Y.; Liu, X.; Li, Y.; Gan, Q.; Liu, F. Smoking reduced in urban restaurants: The eﬀect of Beijing Smoking Control Regulation. Tob. Control 2017, 26, e75–e78. [CrossRef] 18. Semple, S.; Creely, K.S.; Naji, A.; Miller, B.G.; Ayres, J.G. Secondhand smoke levels in Scottish pubs: The eﬀect of smoke-free legislation. Tob. Control 2007, 16, 127–132. [CrossRef] 19. Liu, S. Shanghai’s Three Major Stations will Be Removed from the Indoor Smoking Room Today. The Two Major Airport Terminals will Also Be Smoke-Free. Available online: https://www.jiemian.com/article/920142. html (accessed on 20 May 2019). 20. Shanghai Public Place Smoke Control Regulation will Be Implemented on March 1, and Indoor Public Places will Be 100% Smoke-Free. Available online: http://www.shanghai.gov.cn/nw2/nw2314/nw2315/nw17239/ nw18078/u21aw1202933.html (accessed on 20 May 2019). 21. Cornelsen, L.; McGowan, Y.; Currie-Murphy, L.M.; Normand, C. Systematic review and meta-analysis of the economic impact of smoking bans in restaurants and bars. Addiction 2014, 109, 720–727. [CrossRef] 22. Luk, R.; Ferrence, R.; Gmel, G. The economic impact of a smoke-free bylaw on restaurant and bar sales in Ottawa, Canada. Addiction 2006, 101, 738–745. [CrossRef] 23. Melberg, H.O.; Lund, K.E. Do smoke-free laws aﬀect revenues in pubs and restaurants? Eur. J. Health Econ. 2012, 13, 93–99. [CrossRef] 23. Melberg, H.O.; Lund, K.E. Do smoke-free laws aﬀect revenues in pubs and restaurants? Eur. J. Health Econ. 2012, 13, 93–99. [CrossRef] 24. Alamar, B.C.; Glantz, S.A. Smoke-free ordinances increase restaurant proﬁt and value. Contemp. Econ. Policy 2004, 22, 520–525. [CrossRef] 24. Alamar, B.C.; Glantz, S.A. Smoke-free ordinances increase restaurant proﬁt and value. Contemp. Econ. Policy 2004, 22, 520–525. [CrossRef] 11 of 11 Int. J. Environ. Res. Public Health 2019, 16, 4019 25. Scollo, M.; Lal, A.; Hyland, A.; Glantz, S. Review of the quality of studies on the economic eﬀects of smoke-free policies on the hospitality industry. Tob. Control 2003, 12, 13–20. [CrossRef] 26. References Sureda, X.; Martinez-Sanchez, J.M.; Lopez, M.J.; Fu, M.; Aguero, F.; Salto, E.; Nebot, M.; Fernandez, E. Secondhand smoke levels in public building main entrances: Outdoor and indoor PM2.5 assessment. Tob. Control 2012, 21, 543–548. [CrossRef] 27. Fong, G.T.; Craig, L.V.; Guignard, R.; Nagelhout, G.E.; Tait, M.K.; Driezen, P.; Kennedy, R.D.; Boudreau, C.; Wilquin, J.L.; Deutsch, A.; et al. Evaluating the eﬀectiveness of France’s indoor smoke-free law 1 year and 5 years after implementation: Findings from the ITC France Survey. PLoS ONE 2013, 8, e66692. [CrossRef] 28. Farrelly, M.C.; Nonnemaker, J.M.; Chou, R.; Hyland, A.; Peterson, K.K.; Bauer, U.E. Changes in hospitality workers’ exposure to secondhand smoke following the implementation of New York’s smoke-free law. Tob. Control 2005, 14, 236–241. [CrossRef] 29. Heloma, A.; Jaakkola, M.S. Four-year follow-up of smoke exposure, attitudes and smoking behaviour following enactment of Finland’s national smoke-free work-place law. Addiction 2003, 98, 1111–1117. [CrossRef] 30. Hahn, E.J.; Rayens, M.K.; York, N.; Okoli, C.T.; Zhang, M.; Dignan, M.; Al-Delaimy, W.K. Eﬀects of a smoke-free law on hair nicotine and respiratory symptoms of restaurant and bar workers. J. Occup. Environ. Med. 2006, 48, 906–913. [CrossRef] 31. Thomson, G.; Wilson, N. One year of smokefree bars and restaurants in New Zealand: Impacts and responses. BMC Public Health 2006, 6, 64. [CrossRef] 32. Barnoya, J.; Arvizu, M.; Jones, M.R.; Hernandez, J.C.; Breysse, P.N.; Navas-Acien, A. Secondhand smoke exposure in bars and restaurants in Guatemala City: Before and after smoking ban evaluation. Cancer Causes Control 2011, 22, 151–156. [CrossRef] 33. Valente, P.; Forastiere, F.; Bacosi, A.; Cattani, G.; di Carlo, S.; Ferri, M.; Figa-Talamanca, I.; Marconi, A.; Paoletti, L.; Perucci, C.; et al. Exposure to ﬁne and ultraﬁne particles from secondhand smoke in public places before and after the smoking ban, Italy 2005. Tob. Control 2007, 16, 312–317. [CrossRef] 34. Lee, K.; Hahn, E.J.; Robertson, H.E.; Lee, S.; Vogel, S.L.; Travers, M.J. Strength of smoke-free air laws and indoor air quality. Nicotine Tob. Res. 2009, 11, 381–386. [CrossRef] 5. Protano, C.; Cattaruzza, M.S.; Osborn, J.F.; Vitali, M. Indoor particulate matter and secondhand sm Simulation of an exposure scenario. Ann. Ig. 2014, 26, 186–189. [CrossRef] p g 36. Yu, J. Study on the Correlation Between PM2.5 and Tobacco Control in Indoor Public Places in Gongshu District. Master’s Thesis, Zhejiang University, Hangzhou, China, 2015. 37. Matt, G.E.; Quintana, P.J.; Fortmann, A.L.; Zakarian, J.M.; Galaviz, V.E.; Chatﬁeld, D.A.; Hoh, E.; Hovell, M.F.; Winston, C. References Thirdhand smoke and exposure in California hotels: Non-smoking rooms fail to protect non-smoking hotel guests from tobacco smoke exposure. Tob. Control 2014, 23, 264–272. [CrossRef] 38. Frazer, K.; Callinan, J.E.; McHugh, J.; van Baarsel, S.; Clarke, A.; Doherty, K.; Kelleher, C. Legislative smoking bans for reducing harms from secondhand smoke exposure, smoking prevalence and tobacco consumption. Cochrane Database Syst. Rev. 2016, 2, CD005992. [CrossRef] 39. Fong, G.T.; Hyland, A.; Borland, R.; Hammond, D.; Hastings, G.; McNeill, A.; Anderson, S.; Cummings, K.M.; Allwright, S.; Mulcahy, M.; et al. Reductions in tobacco smoke pollution and increases in support for smoke-free public places following the implementation of comprehensive smoke-free workplace legislation in the Republic of Ireland: Findings from the ITC Ireland/UK Survey. Tob. Control 2006, 15 (Suppl. 3), iii51–iii58. [CrossRef] 39. Fong, G.T.; Hyland, A.; Borland, R.; Hammond, D.; Hastings, G.; McNeill, A.; Anderson, S.; Cummings, K.M.; Allwright, S.; Mulcahy, M.; et al. Reductions in tobacco smoke pollution and increases in support for smoke-free public places following the implementation of comprehensive smoke-free workplace legislation in the Republic of Ireland: Findings from the ITC Ireland/UK Survey. Tob. Control 2006, 15 (Suppl. 3), iii51–iii58. [CrossRef] 40. Blanco-Marquizo, A.; Goja, B.; Peruga, A.; Jones, M.R.; Yuan, J.; Samet, J.M.; Breysse, P.N.; Navas-Acien, A. Reduction of secondhand tobacco smoke in public places following national smoke-free legislation in Uruguay. Tob. Control 2010, 19, 231–234. [CrossRef] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
https://openalex.org/W4312205508	https://www.mdpi.com/2076-2615/13/1/89/pdf?version=1672060124	English	null	Molecular Marker-Assisted Selection of ABCG2, CD44, SPP1 Genes Contribute to Milk Production Traits of Chinese Holstein	Animals	2,022	cc-by	13,000	Article Molecular Marker-Assisted Selection of ABCG2, CD44, SPP1 Genes Contribute to Milk Production Traits of Chinese Holstein 1 Joint International Research Laboratory of Agriculture and Agri-Product Safety, The Ministry of Education of China, Institutes of Agricultural Science and Technology Development, Yangzhou University, Yangzhou 225009, China 2 Key Laboratory of Animal Genetics & Breeding and Molecular Design of Jiangsu Province, Yangzhou University, Yangzhou 225009, China * C d @ d T l 86 0514 87979230 1 Joint International Research Laboratory of Agriculture and Agri-Product Safety, The Ministry of Education of China, Institutes of Agricultural Science and Technology Development, Yangzhou University, Yangzhou 225009, China 2 Key Laboratory of Animal Genetics & Breeding and Molecular Design of Jiangsu Province, Yangzhou University, Yangzhou 225009, China 1 Joint International Research Laboratory of Agriculture and Agri-Product Safety, The Ministry of Education of China, Institutes of Agricultural Science and Technology Development, Yangzhou University, China, Institutes of Agricultural Science and Technology Development, Yangzhou University, Yangzhou 225009, China 2 Key Laboratory of Animal Genetics & Breeding and Molecular Design of Jiangsu Province, Yangzhou University, Yangzhou 225009, China * C d @ d T l 86 0514 87979230 * Correspondence: yzp@yzu.edu.cn; Tel.: +86-0514-87979230 Simple Summary: Chinese Holstein cattle are the main breed of dairy cows in China. This work performed an association analysis between polymorphisms of three candidate genes (ABCG2, CD44, SPP1) and milk production performance on Chinese Holstein cattle. The results showed that the different genotypes of these 10 SNPs from the ABCG2, CD44, and SPP1 genes signiﬁcantly affected the milk production performance of Chinese Holstein cattle in terms of milk yield, milk fat percentage, milk protein percentage, somatic cell score, and urea nitrogen content. The ABCG2, CD44, and SPP1 genes could be selected for marker-assisted selection, which is of great signiﬁcance for future precise molecular breeding. Abstract: Based on our results of genome-wide association analysis, we performed gene ontology (GO) annotation and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis; three candidate genes (ABCG2, CD44, SPP1) were screened in this study for SNPs association analysis with production traits in 999 Holstein cattle. In this research, ﬂight mass spectrometry genotyping was used to detect the polymorphism of SNP seats. Citation: Sun, Y.; Wu, X.; Ma, Y.; Liu, D.; Lu, X.; Zhao, T.; Yang, Z. Molecular Marker-Assisted Selection of ABCG2, CD44, SPP1 Genes Contribute to Milk Production Traits of Chinese Holstein. Animals 2023, 13, 89. https://doi.org/10.3390/ ani13010089 Keywords: ABCG2; CD44; SPP1; association analysis; SNP loci; production performance; Chinese Holstein cattle Received: 19 November 2022 Revised: 18 December 2022 Accepted: 23 December 2022 Published: 26 December 2022 Article Molecular Marker-Assisted Selection of ABCG2, CD44, SPP1 Genes Contribute to Milk Production Traits of Chinese Holstein It was shown that four, four, and two single nucleotide polymorphisms (SNP) loci were detected for the ABCG2, CD44, and SPP1 genes, respectively, and the different genotypes of these 10 SNPs signiﬁcantly affected the milk production performance of Chinese Holstein cattle in terms of milk yield, milk fat percentage, milk protein percentage, somatic cell score, and urea nitrogen content. Among them, ABCG2-G.80952G > T locus, ABCG2-G.120017G > A locus and CD44-G.2294G > C locus had signiﬁcant effects on somatic cell score (p < 0.01). Cows with GG genotypes at ABCG2-G.80952G > T locus, AA and GG genotypes at ABCG2-G.120017G > A locus, and GG genotypes at CD44-G.2294G > C locus had lower somatic cell scores. The present study elucidated that ABCG2, CD44, and SPP1 could be selected for marker-assisted selection and will beneﬁt for future precise molecular breeding. animals animals animals animals Based on the results of GWAS, we performed gene ontology (GO) annotation and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis; three candidate genes (ABCG2, CD44, SPP1) were selected in this study for SNPs association analysis with production traits in Holstein cattle. The present study elucidated that ABCG2, CD44, and SPP1 closely related to dairy cattle performance were selected for the SNPs screening, which contributed to the marker-assisted selection and is of great signiﬁcance for future precise molecular breeding. In our previous work, we used the recorded data of 86,281 test days of 8580 Holstein cows in Jiangsu, and we used the random regression measurement day model to estimate the genetic parameters of Holstein somatic cell scores (SCS). At the same time, the adjusted phenotype was used to perform a genome-wide association analysis (GWAS) of dairy cow SCS using the FarmCPU method to screen for signiﬁcant SNPs and candidate genes [7,8]. g g g Based on the results of GWAS, we performed gene ontology (GO) annotation and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis; three candidate genes (ABCG2, CD44, SPP1) were selected in this study for SNPs association analysis with production traits in Holstein cattle. The present study elucidated that ABCG2, CD44, and SPP1 closely related to dairy cattle performance were selected for the SNPs screening, which contributed to the marker-assisted selection and is of great signiﬁcance for future precise molecular breeding. 2. Materials and Methods 2.1. Sample Collection The experimental population of this study was from four large-scale dairy farms in Jiangsu, China. In this study, SNP microarray samples were collected from cows for genetic parameter evaluation and association analysis. SNP chips were used to collect samples of hair follicles from cows, and hair follicle samples from 999 cows in four farms were collected from 8580 cows that passed the quality control, these experimental cows were of the same age (24 months) and the same parity (parity 2). The collected samples were sent to Neogen Biotechnology Co. Ltd (Lansing, MI, USA), and DNA was extracted and genotyped from the hair follicles of 999 cows. 2.2. Enrichment Analysis and DNA Detection The database (DAVID) online software (https://david.ncifcrf.gov, accessed on 4 November 2022) and the bovine reference genome ARS-UCD1.2(ftp://hgdownload.soe.ucsc.edu/golden path/bosTau9/, accessed on 10 September 2022) in the UCSC database were used, and other online tools have carried out the GO (https://david.ncifcrf.gov, accessed on 4 November 2022) annotation and KEGG (https://www.kegg.jp/, accessed on 4 November 2022) pathway enrichment analysis of candidate gene functions. DNA purity and concentration were tested as follows: using NanoDroP1000 spectrophotometer can directly detect the ratio of DNA OD260/OD280, OD260/OD230 and the concentration of DNA, using the following formula to calculate the concentration of DNA: DNA concentration (ug/mL) = 50 × OD260, when the ratio of OD260/OD280 is between 1.7–1.9, it indicates that the concentration of DNA is relatively ideal. The performance of dairy cows mainly refers to milk production traits. Milk-producing traits have always been the more concerned quantitative economic traits in dairy farming, which are jointly controlled by micro-efﬁciency polygenes and are susceptible to environ- mental inﬂuences [1]. Due to the long generation interval of dairy cows, the selection of milk producing traits of dairy cows is slow in conventional breeding work. Therefore, molecular marker-assisted breeding is a new situation in the current breeding work. Sin- gle nucleotide polymorphisms (SNP) have been widely used in DNA molecular genetic markers, and the prospects are very promising [2]. Copyright: © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). https://www.mdpi.com/journal/animals Animals 2023, 13, 89. https://doi.org/10.3390/ani13010089 Animals 2023, 13, 89 2 of 15 SNP is the third generation of molecular markers after restriction fragment length polymorphism (RFLP) and microsatellite polymorphism (MPP) [3]. SNP refers to DNA sequence polymorphism caused by single nucleotide variation in the genome, including the insertion, deletion, transversion, and transition of single bases, with an allele variation frequency above 1%. SNP mutations may cause changes in genetic codons, and different genetic codons encode different proteins, and protein changes directly affect biological diversity [4,5]. Therefore, the change in the SNP mutation site may cause a change in individual traits, such as growth, reproduction, milk production, and other traits. Through the continuous reproduction of the population, the good characters are preserved, and the unsuitable characters are eliminated. If it is applied to the breeding of livestock and poultry, the selection and breeding of excellent breeds can be accelerated. More and more studies have found that SNPs are important to the growth, production, reproduction, disease, and other traits of cattle [6]. In our previous work, we used the recorded data of 86,281 test days of 8580 Holstein cows in Jiangsu, and we used the random regression measurement day model to estimate the genetic parameters of Holstein somatic cell scores (SCS). At the same time, the adjusted phenotype was used to perform a genome-wide association analysis (GWAS) of dairy cow SCS using the FarmCPU method to screen for signiﬁcant SNPs and candidate genes [7,8]. 2.5. Fractal Detection of SNPs The screened mutation loci were determined by matrix-assisted laser desorption ionization time-of-ﬂight mass spectrometry (MALDI-TOF-MS) for typing detection. The genotyping detection system adopts the time-of-ﬂight mass spectrometry biochip system (MassARRAY®MALDI-TOF System) developed by Sequenom, Inc. (San Diego, CA, USA). This system has been widely promoted and used in the Human Genome Hapmap Project. 2.4. Screening of SNP Mutation Sites 20 DNA samples were randomly selected for each pair of primers for PCR ampliﬁ- cation, and the successfully ampliﬁed products were sent to Qingke Biological Company (Nanjing, China) for sequencing. The sequencing results were compared with Vector NTI Advance 11 software to ﬁnd out the mutation sites and their speciﬁc positions. The am- pliﬁcation procedure was as follows: pre-denaturation at 94 ◦C for 5 min, denaturation at 94 ◦C for 30 s, renaturation at 68 ◦C for 30 s, and extension at 72 ◦C for 2 min 10 s, 17 cycles. Denaturation at 94 ◦C for 30 s, renaturation at 51 ◦C for 30 s, and extension at 72 ◦C for 2 min 10 s, 20 cycles. The ﬁnal extension was at 72 ◦C for 10 min and stored at 4 ◦C. 2.3. PCR Ampliﬁcation, and Commercial Sequences Based on the sequences of the bovine ABCG2, SPP1, and CD44 genes published on NCBI, primers were designed by software in the CDS region and part of the intron region of Animals 2023, 13, 89 3 of 15 the ABCG2, SPP1, and CD44 genes, respectively. A total of 7 pairs of primers were involved in this test, and the details of the primers were shown in Table 1. the ABCG2, SPP1, and CD44 genes, respectively. A total of 7 pairs of primers were involved in this test, and the details of the primers were shown in Table 1. the ABCG2, SPP1, and CD44 genes, respectively. A total of 7 pairs of primers were involved in this test, and the details of the primers were shown in Table 1. Table 1. Primer sequence and ampliﬁed fragment length of gene PCR ampliﬁcation. Table 1. Primer sequence and ampliﬁed fragment length of gene PCR ampliﬁcation. Primer Sequences of Primer Size of Production (bp) Position of Production P1 (ABCG2) F: AAGGAGGAAAGGAGCCAGAG 460 57,031 R: TGCTACCAGACACGAAATCG P2 (ABCG2) F: TTGGATGATGATGACTTTGG 766 80,828 R: GAACTTTCTCTCTGGCTACTG P3 (ABCG2) F: TGCTTTCAACTTCTCTGCTC 392 94,583 R: GTCCTTTTTCTTTCTCCTCC P4 (ABCG2) F: TGGTTATATTGGGTGGTTGG 606 119,653 R: ACTATGGGATGAGGTTCGTG P5 (CD44) F: GCTTTGCTTCTGAGGATTCTG 590 1985 R: TCGCTTCACTGCTCTTTACC P6 (CD44) F: CCCGCTCCTCGAGTTTTCTG 324 86,737 R: ATTGAGTCCGCTGGGCTTTC P7 (SPP1) F: AATAAACCCTTTTCCCTCCC 495 50,063 R: CCTTACAAATTGACCTTCCC Sequences of Primer Size of Production (bp) Position of Production 2.6. Statistical Analysis The Hardy-Weinberg equilibrium test (HWE), genotype and allele frequency analysis, and linkage disequilibrium (LD, as measured by D′ and r2 ) analysis were performed for SNP sites after ABCG2, SPP1, and CD44 genotyping by SHEsis software (http://analysis2 .bio-x.cn/myAnalysis.php, accessed on 4 November 2022) [9]. Multi-factor ANOVA with SPSS (Ver. 18.0) software and general linear models were used to perform the association analysis between the genotype and production performance of ABCG2, SPP1, and CD44 gene SNPs loci. The reduced linear model excluded ﬁxed effects of age and paternity. The speciﬁc models were as follows: Yijklmnop = µ + Ci + Mj + Pk + Jl + Nm + Gn + Fo + eijklmnop where Yijklmnop is the observed value of the trait, µ is the population mean, and Ci is the ﬁxed effect of calving season. Mj is the ﬁxed effect of the lactation stage, Pk is the ﬁxed effect of parity, Jl is the ﬁxed effect of measured season, Nm is the ﬁxed effect of the test year, Gn is the ﬁxed effect of genotype, Fo is the cattle ﬁeld effect, and eijklmnop is the random error. Multiple comparisons between genotypes were obtained using Duncan’s method. 4 of 15 Animals 2023, 13, 89 3.2. Screening of ABCG2, CD44 and SPP1 Genes SNP Mutation Loci 3.2. Screening of ABCG2, CD44 and SPP1 Genes SNP Mutation Lo annotation results of three genes in BP, CC and MF. (B) Distribution of The sequence map CD44 g 2263A > G site for AG AA and GG (A) The sequence map at CD44 g 2294G > C site for C Figure 3. Sequencing peak map of four SNPs in CD44 gene sequence. The sequence map a CD44-g.2263A > G site for AG, AA and GG (A), The sequence map at CD44-g.2294G > C site for CG GG and CC (B), The sequence map at CD44-g.86895A > G site for AG, AA and GG (C), The se quence map at CD44-g.86978G > A site for AG, AA and GG (D). The arrow indicates the location o the mutation sites. Figure 3. Sequencing peak map of four SNPs in CD44 gene sequence. The sequence map at CD44-g.2263A > G site for AG, AA and GG (A), The sequence map at CD44-g.2294G > C site for CG, GG and CC (B), The sequence map at CD44-g.86895A > G site for AG, AA and GG (C), The se- quence map at CD44-g.86978G > A site for AG, AA and GG (D). The arrow indicates the location of the mutation sites. Figure 3. Sequencing peak map of four SNPs in CD44 gene sequence. The sequence map at CD44- g.2263A > G site for AG, AA and GG (A), The sequence map at CD44-g.2294G > C site for CG, GG and CC (B), The sequence map at CD44-g.86895A > G site for AG, AA and GG (C), The sequence map at CD44-g.86978G > A site for AG, AA and GG (D). The arrow indicates the location of the mutation sites. CD44-g.2263A > G site for AG, AA and GG (A), The sequence map at CD44-g.2294G > C site for C GG and CC (B), The sequence map at CD44-g.86895A > G site for AG, AA and GG (C), The s quence map at CD44-g.86978G > A site for AG, AA and GG (D). The arrow indicates the location the mutation sites. Figure 3. Sequencing peak map of four SNPs in CD44 gene sequence. The sequence map a CD44-g.2263A > G site for AG, AA and GG (A), The sequence map at CD44-g.2294G > C site for CG GG and CC (B), The sequence map at CD44-g.86895A > G site for AG, AA and GG (C), The se quence map at CD44-g.86978G > A site for AG, AA and GG (D). The arrow indicates the location o the mutation sites. 3.2. Screening of ABCG2, CD44 and SPP1 Genes SNP Mutation Loci 3.2. Screening of ABCG2, CD44 and SPP1 Genes SNP Mutation Lo annotation results of three genes in BP, CC and MF. (B) Distribution of The sequence map at ABCG2- g.57261A > G site for AG, AA and GG (A), The sequence map at ABCG2-g.80952G > T site for GT, GG and TT (B), The sequence map at ABCG2-g.94683A > G site for AG, AA and GG (C), The sequence map at ABCG2-g.120017G > A site for AG, GG and AA (D). The arrow indicates the location of the mutation sites. Figure 2. Sequencing peak map of four SNPs in ABCG2 gene sequence. The sequence map a ABCG2-g.57261A > G site for AG、AA and GG (A), The sequence map at ABCG2-g.80952G > T site for GT, GG and TT (B), The sequence map at ABCG2-g.94683A > G site for AG, AA and GG (C), The sequence map at ABCG2-g.120017G > A site for AG, GG and AA (D). The arrow indicates the loca tion of the mutation sites. Figure 2. Sequencing peak map of four SNPs in ABCG2 gene sequence. The sequence map ABCG2-g.57261A > G site for AG、AA and GG (A), The sequence map at ABCG2-g.80952G > T si for GT, GG and TT (B), The sequence map at ABCG2-g.94683A > G site for AG, AA and GG (C), Th sequence map at ABCG2-g.120017G > A site for AG, GG and AA (D). The arrow indicates the loc tion of the mutation sites. Figure 2. Sequencing peak map of four SNPs in ABCG2 gene sequence. The sequence map at ABCG2- g.57261A > G site for AG, AA and GG (A), The sequence map at ABCG2-g.80952G > T site for GT, GG and TT (B), The sequence map at ABCG2-g.94683A > G site for AG, AA and GG (C), The sequence map at ABCG2-g.120017G > A site for AG, GG and AA (D). The arrow indicates the location of the mutation sites. Figure 2. Sequencing peak map of four SNPs in ABCG2 gene sequence. The sequence map at ABCG2-g.57261A > G site for AG、AA and GG (A), The sequence map at ABCG2-g.80952G > T site for GT, GG and TT (B), The sequence map at ABCG2-g.94683A > G site for AG, AA and GG (C), The sequence map at ABCG2-g.120017G > A site for AG, GG and AA (D). The arrow indicates the loca- tion of the mutation sites. Figure 3. Sequencing peak map of four SNPs in CD44 gene sequence. 3.2. Screening of ABCG2, CD44 and SPP1 Genes SNP Mutation Loci 3.2. Screening of ABCG2, CD44 and SPP1 Genes SNP Mutation Lo annotation results of three genes in BP, CC and MF. (B) Distribution of 3.2. Screening of ABCG2, CD44 and SPP1 Genes SNP Mutation Loci 3.2. Screening of ABCG2, CD44 and SPP1 Genes SNP Mutation Lo annotation results of three genes in BP, CC and MF. (B) Distribution of By comparing the sequencing results with the three gene sequences, four SNP loci were found, respectively, in ABCG2 and CD44: ABCG2-g.57261A > G, ABCG2-g.80952G > T, ABCG2-g.94683A > G, ABCG2-g.120017G > A, CD44-g.2263A > G, CD44-g.2294G > C, CD44-g.86895A > G, CD44-g.86978G > A. SPP1 was found two SNP loci SPP1-g.50265G > A and SPP1-g.50315C > T. Figures 2A–D, 3A–D and 4A,B show the sequencing peaks of each SNP locus of ACG2, CD44, and SPP1, respectively. By comparing the sequencing results with the three gene sequences, four SNP lo were found, respectively, in ABCG2 and CD44: ABCG2-g.57261A > G, ABCG2-g.80952G T, ABCG2-g.94683A > G, ABCG2-g.120017G > A, CD44-g.2263A > G, CD44-g.2294G > C CD44-g.86895A > G, CD44-g.86978G > A. SPP1 was found two SNP loci SPP1-g.50265G A and SPP1-g.50315C > T. Figure 2A–D, Figure 3A–D, and Figure 4A,B show the s quencing peaks of each SNP locus of ACG2, CD44, and SPP1, respectively. 3.2. Screening of ABCG2, CD44 and SPP1 Genes SNP Mutation Loci By comparing the sequencing results with the three gene sequences, four SNP loc were found, respectively, in ABCG2 and CD44: ABCG2-g.57261A > G, ABCG2-g.80952G > T, ABCG2-g.94683A > G, ABCG2-g.120017G > A, CD44-g.2263A > G, CD44-g.2294G > C CD44-g.86895A > G, CD44-g.86978G > A. SPP1 was found two SNP loci SPP1-g.50265G > A and SPP1-g.50315C > T. Figure 2A–D, Figure 3A–D, and Figure 4A,B show the se quencing peaks of each SNP locus of ACG2, CD44, and SPP1, respectively. quencing peaks of each SNP locus of ACG2, CD44, and SPP1, respectively. Figure 2. Sequencing peak map of four SNPs in ABCG2 gene sequence. The sequence map ABCG2-g.57261A > G site for AG、AA and GG (A), The sequence map at ABCG2-g.80952G > T si for GT, GG and TT (B), The sequence map at ABCG2-g.94683A > G site for AG, AA and GG (C), Th sequence map at ABCG2-g.120017G > A site for AG, GG and AA (D). The arrow indicates the loc tion of the mutation sites. Figure 2. Sequencing peak map of four SNPs in ABCG2 gene sequence. 3. Results . Functional Annotation and Signal Pathway Analysis of ABCG2, CD44 and SPP1 Genes The proﬁles of three genes were in the biological process (BP), cell component (CC), and molecular function (MF) categories by GO analysis, respectively. The BP analysis showed that the three genes were classiﬁed into the cellular process, localization, metabolic process, response to stimulus, and single-organism process. The CC annotation revealed that the three genes were involved in the cell, cell part, and organelle. The MF classiﬁcation of genes suggested that three genes were involved in binding. GO annotation analysis elucidated that the three genes participated in the regulation of production performance by multiple indispensable activities (Figure 1A). Animals 2023, 13, x FOR PEER REVIEW 5 of 17 Figure 1. GO annotation and KEGG pathway enrichment of ABCG2, CD44 and SPP1 genes. (A) G annotation results of three genes in BP, CC and MF. (B) Distribution of enriched KEGG pathway. Figure 1. GO annotation and KEGG pathway enrichment of ABCG2, CD44 and SPP1 genes. (A) GO annotation results of three genes in BP, CC and MF. (B) Distribution of enriched KEGG pathway. Animals 2023, 13, 89 5 of 15 5 of 15 Furthermore, we performed KEGG pathway enrichment analysis to explore the most active pathways of these genes; the results revealed that these genes were related to signaling molecules and interactions, as well as the immune system and infectious diseases, which are partially but not completely involved (Figure 1B). Figure 1. GO annotation and KEGG pathway enrichment of ABCG2, CD44 and SPP1 genes. (A) G annotation results of three genes in BP, CC and MF. (B) Distribution of enriched KEGG pathway VIEW 6 of 1 Figure 1. GO annotation and KEGG pathway enrichment of ABCG2, CD44 and SPP1 genes. (A) GO 3.3. LD Analysis of SNPS in ABCG2, CD44 and SPP1 Genes 3.3. LD Analysis of SNPS in ABCG2, CD44 and SPP1 Genes LD analysis was performed on the SNP loci of the ABCG2, CD44, and SPP1 genes with SHEsis software, and the results were obtained by D′ value and r2 value. The D′ > 0.7 and r2 > 1/3 might suggest a sufficiently strong LD to be available for plotting [10]. All the D' and r2 values among the different SNP loci of the ABCG2, CD44, and SPP1 genes listed in Table 2. The ABCG2 gene ABCG2-g.57261A > G and ABCG2-g.94683A > G, ABCG2-g.120017G > A, ABCG2-g.80952G > T and ABCG2-g.120017G > A are in a highly linked state (D′ > 0.7, r2 > 1/3), respectively, and the linkage degree between the remaining loci is relatively low (Figure 5A). The CD44 gene CD44-g.2294G > C and CD44-g.86978G > A are in a highly linked state (D′ > 0.7, r2 > 1/3), and there is no high degree of linkage between other loci chain relationship (Figure 5B). There is no high linkage relationship between the two SNPs of the SPP1 gene (D′ > 0.7, r2 < 1/3) (Figure 5C). LD analysis was performed on the SNP loci of the ABCG2, CD44, and SPP1 genes with SHEsis software, and the results were obtained by D′ value and r2 value. The D′ > 0.7 and r2 > 1/3 might suggest a sufﬁciently strong LD to be available for plotting [10]. All the D’ and r2 values among the different SNP loci of the ABCG2, CD44, and SPP1 genes listed in Table 2. The ABCG2 gene ABCG2-g.57261A > G and ABCG2-g.94683A > G, ABCG2-g.120017G > A, ABCG2-g.80952G > T and ABCG2-g.120017G > A are in a highly linked state (D′ > 0.7, r2 > 1/3), respectively, and the linkage degree between the remaining loci is relatively low (Figure 5A). The CD44 gene CD44-g.2294G > C and CD44-g.86978G > A are in a highly linked state (D′ > 0.7, r2 > 1/3), and there is no high degree of linkage between other loci chain relationship (Figure 5B). There is no high linkage relationship between the two SNPs of the SPP1 gene (D′ > 0.7, r2 < 1/3) (Figure 5C). Table 2. The estimated values of LD analysis between mutation sites within the ABCG2, CD44, SPP1 genes. 3.2. Screening of ABCG2, CD44 and SPP1 Genes SNP Mutation Loci 3.2. Screening of ABCG2, CD44 and SPP1 Genes SNP Mutation Lo annotation results of three genes in BP, CC and MF. (B) Distribution of Figure 3. Sequencing peak map of four SNPs in CD44 gene sequence. The sequence map at CD44- g.2263A > G site for AG, AA and GG (A), The sequence map at CD44-g.2294G > C site for CG, GG and CC (B), The sequence map at CD44-g.86895A > G site for AG, AA and GG (C), The sequence map at CD44-g.86978G > A site for AG, AA and GG (D). The arrow indicates the location of the mutation sites. 6 of 15 7 of 17 Animals 2023, 13, 89 Animals 2023, 13, x FOR Figure 4. Sequencing peak map of two SNPs in SPP1 gene sequence. The sequence map at SPP1-g.50265G > A site for GA, GG and AA (A), The sequence map at SPP1-g.50315C > T site for CT, TT and CC (B). The arrow indicates the location of the mutation sites. Figure 4. Sequencing peak map of two SNPs in SPP1 gene sequence. The sequence map at SPP1- g.50265G > A site for GA, GG and AA (A), The sequence map at SPP1-g.50315C > T site for CT, TT and CC (B). The arrow indicates the location of the mutation sites. igure 4. Sequencing peak map of two SNPs in SPP1 gene sequence. The sequence map at PP1-g.50265G > A site for GA, GG and AA (A), The sequence map at SPP1-g.50315C > T site for CT, TT and CC (B). The arrow indicates the location of the mutation sites. Figure 4. Sequencing peak map of two SNPs in SPP1 gene sequence. The sequence map at SPP1- g.50265G > A site for GA, GG and AA (A), The sequence map at SPP1-g.50315C > T site for CT, TT and CC (B). The arrow indicates the location of the mutation sites. 3.3. LD Analysis of SNPS in ABCG2, CD44 and SPP1 Genes 3.3. LD Analysis of SNPS in ABCG2, CD44 and SPP1 Genes ABCG2 Loci g.57261A > G g.80952G > T g.94683A > G g.120017G > A g.57261A > G D′ = 0.61 D′ = 0.99 D′ = 1.00 g.80952G > T r2 = 0.20 D′ = 0.48 D′ = 1.00 g.94683A > G r2 = 0.38 r2 = 0.16 D′ = 0.53 g.120017G > A r2 = 0.35 r2 = 0.61 r2 = 0.26 CD44 Loci g.2263A > G g.2294G > C g.86895A > G g.86978G > A g.2263A > G D′ = 0.47 D′ = 0.97 D′ = 0.71 g.2294G > C r2 = 0.05 D′ = 0.44 D′ = 1.00 g.86895A > G r2 = 0.07 r2 = 0.07 D′ = 0.44 g.86978G > A r2 = 0.05 r2 = 0.46 r2 = 0.03 SPP1 Loci g.50265G > A g.50315C > T g.50265G > A D′ = 0.78 g.50315C > T r2 = 0.22 Table 2. The estimated values of LD analysis between mutation sites within the ABCG2, CD44, SPP1 genes. ABCG2 Loci g.57261A > G g.80952G > T g.94683A > G g.120017G > A g.57261A > G D′ = 0.61 D′ = 0.99 D′ = 1.00 g.80952G > T r2 = 0.20 D′ = 0.48 D′ = 1.00 g.94683A > G r2 = 0.38 r2 = 0.16 D′ = 0.53 g.120017G > A r2 = 0.35 r2 = 0.61 r2 = 0.26 CD44 Loci g.2263A > G g.2294G > C g.86895A > G g.86978G > A g.2263A > G D′ = 0.47 D′ = 0.97 D′ = 0.71 g.2294G > C r2 = 0.05 D′ = 0.44 D′ = 1.00 g.86895A > G r2 = 0.07 r2 = 0.07 D′ = 0.44 g.86978G > A r2 = 0.05 r2 = 0.46 r2 = 0.03 SPP1 Loci g.50265G > A g.50315C > T g.50265G > A D′ = 0.78 g.50315C > T r2 = 0.22 Table 2. The estimated values of LD analysis between mutation sites within the ABCG2, CD44, SPP1 genes. Table 2. The estimated values of LD analysis between mutation sites within the ABCG2, CD44, SPP1 genes. 7 of 15 Animals 2023, 13, 89 R PEER REVIEW Figure 5. The linkage map of SNPs loci of ABCG2, CD44, SPP1 genes. LD value D′ > 0.7, r2 > 1/3 in- dicates relatively high linkage strength. (A) The LD map of 4 SNPs loci of ABCG2 gene. (B) The LD map of 4 SNPs loci of CD44 gene. 3.3. LD Analysis of SNPS in ABCG2, CD44 and SPP1 Genes 3.3. LD Analysis of SNPS in ABCG2, CD44 and SPP1 Genes (C) The LD map of 2 SNPs loci of SPP1 gene. Figure 5. The linkage map of SNPs loci of ABCG2, CD44, SPP1 genes. LD value D′ > 0.7, r2 > 1/3 indicates relatively high linkage strength. (A) The LD map of 4 SNPs loci of ABCG2 gene. (B) The LD map of 4 SNPs loci of CD44 gene. (C) The LD map of 2 SNPs loci of SPP1 gene. age map of SNPs loci of ABCG2, CD44, SPP1 genes. LD value D′ > 0.7, r2 > 1/3 in- high linkage strength. (A) The LD map of 4 SNPs loci of ABCG2 gene. (B) The LD i of CD44 gene. (C) The LD map of 2 SNPs loci of SPP1 gene. Figure 5. The linkage map of SNPs loci of ABCG2, CD44, SPP1 genes. LD value D′ > 0.7, r2 > 1/3 indicates relatively high linkage strength. (A) The LD map of 4 SNPs loci of ABCG2 gene. (B) The LD map of 4 SNPs loci of CD44 gene. (C) The LD map of 2 SNPs loci of SPP1 gene. equency and Allele Frequency of SNP Loci in ABCG2, CD44 and SPP1 Genes 3.4. Genotype Frequency and Allele Frequency of SNP Loci in ABCG2, CD44 and SPP1 Genes equency and Allele Frequency of SNP Loci in ABCG2, CD44 and SPP1 Genes 3.4. Genotype Frequency and Allele Frequency of SNP Loci in ABCG2, CD44 and SPP1 Genes q y q y f in Table 2, among the four SNP loci of ABCG2 gene, ABCG2-g.57261A > G 0952G > T were located in intron 1, ABCG2-g.94683A > G in intron 5, and 7G > A in intron 13. ABCG2 gene has three genotypes for each SNP locus t genotypes of ABCG2-g.57261A > G, ABCG2-g.80952G > T, A > G and ABCG2-g.120017G > A were GG, GT, AG, and AG, respec- test, the 4 SNPs of ABCG2 gene all reached HWE (p > 0.05). Among the D44 gene, CD44-G.2263a > G and CD44-G.2294G > C were located in intron G.86895A > G and CD44-G.86978G > A were located in exon 17. > C, CD44-g.2263A > G, CD44-g.86895A > G and CD44-g.86978G > A also notypes, respectively. The dominant gene of CD44-g.2263A > G, > G and CD44-g.86978G > A was AA, and the CD44-g.2294G > C was CG. 3.3. LD Analysis of SNPS in ABCG2, CD44 and SPP1 Genes 3.3. LD Analysis of SNPS in ABCG2, CD44 and SPP1 Genes wo SNP loci of SPP1 also had three genotypes respectively and the As shown in Table 2, among the four SNP loci of ABCG2 gene, ABCG2-g.57261A > G and ABCG2-g.80952G > T were located in intron 1, ABCG2-g.94683A > G in intron 5, and ABCG2-g.120017G > A in intron 13. ABCG2 gene has three genotypes for each SNP locus The dominant genotypes of ABCG2-g.57261A > G, ABCG2-g.80952G > T, ABCG2-g.94683A > G and ABCG2-g.120017G > A were GG, GT, AG, and AG, respectively. After χ2 test, the 4 SNPs of ABCG2 gene all reached HWE (p > 0.05). Among the four SNPs of CD44 gene, CD44-G.2263a > G and CD44-G.2294G > C were located in intron 2, and CD44-G.86895A > G and CD44-G.86978G > A were located in exon 17. CD44-g.2294G > C, CD44-g.2263A > G, CD44-g.86895A > G and CD44-g.86978G > A also had three genotypes, respectively. The dominant gene of CD44-g.2263A > G, CD44-g.86895A > G and CD44-g.86978G > A was AA, and the CD44-g.2294G > C was CG. Similarly, the two SNP loci of SPP1 also had three genotypes, respectively, and the dominant genotypes were GG and CT. After χ2 test, all SNP loci of the three genes were in HWE (p > 0.05) (Table 3). q y q y f in Table 2, among the four SNP loci of ABCG2 gene, ABCG2-g.57261A > G 0952G > T were located in intron 1, ABCG2-g.94683A > G in intron 5, and 7G > A in intron 13. ABCG2 gene has three genotypes for each SNP locus t genotypes of ABCG2-g.57261A > G, ABCG2-g.80952G > T, A > G and ABCG2-g.120017G > A were GG, GT, AG, and AG, respec- test, the 4 SNPs of ABCG2 gene all reached HWE (p > 0.05). Among the D44 gene, CD44-G.2263a > G and CD44-G.2294G > C were located in intron G.86895A > G and CD44-G.86978G > A were located in exon 17. > C, CD44-g.2263A > G, CD44-g.86895A > G and CD44-g.86978G > A also notypes, respectively. The dominant gene of CD44-g.2263A > G, > G and CD44-g.86978G > A was AA, and the CD44-g.2294G > C was CG. 3.3. LD Analysis of SNPS in ABCG2, CD44 and SPP1 Genes 3.3. LD Analysis of SNPS in ABCG2, CD44 and SPP1 Genes two SNP loci of SPP1 also had three genotypes respectively and the As shown in Table 2, among the four SNP loci of ABCG2 gene, ABCG2-g.57261A > G and ABCG2-g.80952G > T were located in intron 1, ABCG2-g.94683A > G in intron 5, and ABCG2-g.120017G > A in intron 13. ABCG2 gene has three genotypes for each SNP locus The dominant genotypes of ABCG2-g.57261A > G, ABCG2-g.80952G > T, ABCG2-g.94683A > G and ABCG2-g.120017G > A were GG, GT, AG, and AG, respectively. After χ2 test, the 4 SNPs of ABCG2 gene all reached HWE (p > 0.05). Among the four SNPs of CD44 gene, CD44-G.2263a > G and CD44-G.2294G > C were located in intron 2, and CD44-G.86895A > G and CD44-G.86978G > A were located in exon 17. CD44-g.2294G > C, CD44-g.2263A > G, CD44-g.86895A > G and CD44-g.86978G > A also had three genotypes, respectively. The dominant gene of CD44-g.2263A > G, CD44-g.86895A > G and CD44-g.86978G > A was AA, and the CD44-g.2294G > C was CG. Similarly, the two SNP loci of SPP1 also had three genotypes, respectively, and the dominant genotypes were GG and CT. After χ2 test, all SNP loci of the three genes were in HWE (p > 0.05) (Table 3). Animals 2023, 13, 89 8 of 15 Table 3. Genotype and allele frequency of SNPs of ABCG2, CD44 and SPP1 genes. 3.3. LD Analysis of SNPS in ABCG2, CD44 and SPP1 Genes 3.3. LD Analysis of SNPS in ABCG2, CD44 and SPP1 Genes SNP Locus Location Number Genotype Genotype Frequency Allele Allele Number Allele Frequency χ2 HWE ABCG2-g.57261A > G Intron 1 77 AA 0.077 A 554 0.277 1.000 400 AG 0.401 G 1444 0.723 522 GG 0.522 ABCG2-g.80952G > T Intron 1 162 GG 0.162 G 826 0.413 0.735 502 GT 0.503 T 1172 0.587 335 TT 0.335 ABCG2-g.94683A > G Intron 5 256 AA 0.256 A 996 0.498 0.849 484 AG 0.485 G 1002 0.502 259 GG 0.259 ABCG2-g.120017G > A Intron 13 272 AA 0.272 A 1040 0.521 0.994 496 AG 0.497 G 958 0.479 231 GG 0.231 CD44-g.2263A > G Intron 2 660 AA 0.661 A 1617 0.809 0.722 297 AG 0.297 G 381 0.191 42 GG 0.042 CD44-g.2294G > C Intron 2 250 CC 0.250 C 1007 0.504 0.955 507 CG 0.508 G 991 0.496 242 GG 0.242 CD44-g.86895A > G exon 17 542 AA 0.543 A 1468 0.735 0.961 384 AG 0.384 G 530 0.265 73 GG 0.073 CD44-g.86978G > A exon 17 464 AA 0.465 A 1356 0.679 0.93 428 AG 0.428 G 642 0.321 107 GG 0.107 SPP1-g. 50265G > A Intron 1 44 AA 0.044 A 440 0.220 0.856 352 AG 0.352 G 1558 0.780 603 GG 0.604 SPP1-g. 50315 C > T Intron 1 326 CC 0.326 C 1134 0.568 0.916 482 CT 0.483 T 864 0.432 191 TT 0.191 Note: SNP, single nucleotide polymorphisms; χ2 (HWE), Hardy–Weinberg equilibrium χ2 value. Table 3. Genotype and allele frequency of SNPs of ABCG2, CD44 and SPP1 genes. 3.5. Correlation Analysis between Gene SNP Sites and Production Performances 3.5. Correlation Analysis between Gene SNP Sites and Production Performances y f 3.5.1. Association Analysis of Four SNP Loci in ABCG2 Gene and Production Traits 3.5.1. Association Analysis of Four SNP Loci in ABCG2 Gene and Production Traits The correlation analysis results of the four SNP loci of the BCG2 gene with different genotypes of the Holstein cattle somatic cell score, urea nitrogen in milk, daily milk yield, milk protein rate, and milk fat rate are shown in Table 4. The effect of ABCG2-g.57261A > G locus on milk fat rate reached a signiﬁcant level (p < 0.01). The milk fat rate of GG- type cows was the highest, and the AA genotype was the lowest. 3.3. LD Analysis of SNPS in ABCG2, CD44 and SPP1 Genes 3.3. LD Analysis of SNPS in ABCG2, CD44 and SPP1 Genes The ABCG2-g.94683A > G locus had a highly signiﬁcant effect (p < 0.01) on the urea nitrogen content in milk, which was signiﬁcantly higher in the milk of AA-type cows than in AG and GG types. The ABCG2-g.80952G > T locus had a highly signiﬁcant (p < 0.01) effect on milk yield, milk protein rate, somatic cell score, and urea nitrogen in cows, with TT-type cows having signiﬁcantly lower measured daily milk yield than GG and GT types, while TT-type cows had signiﬁcantly higher milk protein rate, somatic cell score, and urea nitrogen than GG and GT types. ABCG2-g.120017G > A had signiﬁcant effects on milk yield, milk protein percentage, somatic cell score, and urea nitrogen (p < 0.01) and had signiﬁcant effects on milk fat percentage (p < 0.05). The milk protein percentage, milk fat percentage, somatic cell score, and urea nitrogen of AG cows were signiﬁcantly higher than those of AA and GG cows. A square visual diagram of ABCG2 gene difference expression is shown in Figure 6. 3.5.2. Association Analysis of 4 SNP Loci of CD44 Gene and Production Traits SNP Locus Genotype Record Number Tested Day Milk Yield (kg) Milk Fat Rate (%) Milk Protein Rate (%) Somatic Cell Score Urea Nitrogen (mg/dL) CD44-g.2263A > G AA 11,107 34.64 ± 0.10 3.91 ± 0.01 3.28 ± 0.00 2.94 ± 0.02 14.03 ± 0.04 a AG 5181 34.36 ± 0.14 3.90 ± 0.01 3.30 ± 0.01 2.88 ± 0.03 13.64 ± 0.06 b GG 706 33.52 ± 0.40 3.97 ± 0.04 3.32 ± 0.01 2.77 ± 0.07 13.43 ± 0.17 b p 0.081 0.354 0.090 0.052 0.000 ** CD44-g.2294G > C CC 4404 34.24 ± 0.15 3.88 ± 0.01 b 3.29 ± 0.01 2.98 ± 0.03 a 13.59 ± 0.07 c CG 8462 34.61 ± 0.11 3.92 ± 0.01 a 3.28 ± 0.00 2.93 ± 0.02 a 14.05 ± 0.05 a GG 4128 34.59 ± 0.16 3.90 ± 0.02 ab 3.29 ± 0.01 2.80 ± 0.03 b 13.86 ± 0.07 b p 0.114 0.010 * 0.109 0.001 0.000 CD44-g.86895A > G AA 9266 34.37 ± 0.11 b 3.91 ± 0.01 b 3.30 ± 0.00 a 2.88 ± 0.02 13.79 ± 0.05 b AG 6538 34.51 ± 0.13 b 3.89 ± 0.01 b 3.28 ± 0.00 a 2.95 ± 0.02 13.85 ± 0.06 b GG 1190 35.61 ± 0.29 a 3.97 ± 0.03 a 3.24 ± 0.01 b 2.94 ± 0.06 14.82 ± 0.13 a p 0.003 0.007 0.000 0.127 0.000 CD44-g.86978G > A AA 8087 34.45 ± 0.11 3.93 ± 0.01 a 3.29 ± 0.00 2.87 ± 0.02 13.92 ± 0.05 AG 7177 34.56 ± 0.12 3.90 ± 0.01 ab 3.29 ± 0.00 2.93 ± 0.02 13.88 ± 0.05 GG 1730 34.61 ± 0.25 3.86 ± 0.02 b 3.27 ± 0.01 3.04 ± 0.05 13.75 ± 0.11 p 0.252 0.004 ** 0.097 0.316 0.060 Note: Data in the same column with different lowercase letters on the shoulder indicate signiﬁcant differences (p < 0.05); * indicates that the differences reach a signiﬁcant level (p < 0.05); ** indicates that the differences reach a highly signiﬁcant level (p < 0.01). Table 4. Effects of different genotypes of SNPs of ABCG2 gene on production performance of Holstein. 3.5.2. Association Analysis of 4 SNP Loci of CD44 Gene and Production Traits Effects of different genotypes of SNPs of ABCG2 gene on production performance of Holstein Note: Data in the same column with different lowercase letters on the shoulder indicate signiﬁcant differences (p < 0.05); * indicates that the differences reach a signiﬁcant level (p < 0.05); indicates that the differences reach a highly signiﬁcant level (p < 0.01). Table 5. Effects of different genotypes of SNPs of CD44 gene on production performance of Holstein. 3.5.2. Association Analysis of 4 SNP Loci of CD44 Gene and Production Traits SNP Locus Genotype Record Number Tested Day Milk Yield Milk Fat Percentage Milk Protein Percentage Somatic Cell Score Urea Nitrogen (mg/dL) ABCG2-g.57261A > G AA 1275 34.00 ± 0.20 3.87 ± 0.02 b 3.28 ± 0.01 2.98 ± 0.04 13.83 ± 0.09 AG 6729 34.68 ± 0.11 3.90 ± 0.01 ab 3.28 ± 0.00 2.94 ± 0.02 13.90 ± 0.05 GG 8990 34.45 ± 0.13 3.92 ± 0.01 a 3.29 ± 0.01 2.88 ± 0.03 13.88 ± 0.06 p 0.154 0.004 0.388 0.063 0.856 ABCG2-g.80952G > T GG 2713 34.68 ± 0.20 a 3.92 ± 0.02 3.28 ± 0.01 b 2.82 ± 0.04 b 13.84 ± 0.09 b GT 8438 34.71 ± 0.11 a 3.90 ± 0.01 3.28 ± 0.00 b 2.91 ± 0.02 a 13.86 ± 0.05 b TT 5843 34.15 ± 0.13 b 3.90 ± 0.01 3.29 ± 0.00 a 2.95 ± 0.03 a 13.94 ± 0.06 a p 0.001 0.230 0.001 0.000 0.002 ABCG2-g.94683A > G AA 4329 34.61 ± 0.15 3.91 ± 0.01 3.28 ± 0.01 2.92 ± 0.03 14.01 ± 0.07 a AG 8059 34.47 ± 0.11 3.90 ± 0.01 3.29 ± 0.00 2.93 ± 0.02 13.81 ± 0.05 b GG 4606 34.49 ± 0.15 3.91 ± 0.01 3.29 ± 0.01 2.87 ± 0.03 13.88 ± 0.07 ab p 0.163 0.207 0.456 0.529 0.004 ABCG2-g.120017G > A AA 4551 34.37 ± 0.15 b 3.91 ± 0.01 a 3.29 ± 0.01 a 2.87 ± 0.03 b 13.71 ± 0.07 b AG 8419 34.77 ± 0.11 a 3.91 ± 0.01 a 3.28 ± 0.00 b 2.95 ± 0.02 a 14.01 ± 0.05 a GG 4024 34.14 ± 0.16 b 3.89 ± 0.01 b 3.29 ± 0.01 a 2.87 ± 0.03 b 13.82 ± 0.07 b p 0.007 0.041 * 0.001 0.005 0.000 ** Note: Data in the same column with different lowercase letters on the shoulder indicate signiﬁcant differences (p < 0.05); * indicates that the differences reach a signiﬁcant level (p < 0.05); indicates that the differences reach a highly signiﬁcant level (p < 0.01). Table 4. 3.5.2. Association Analysis of 4 SNP Loci of CD44 Gene and Production Traits SNP Locus Genotype Record Number Tested Day Milk Yield (kg) Milk Fat Rate (%) Milk Protein Rate (%) Somatic Cell Score Urea Nitrogen (mg/dL) CD44-g.2263A > G AA 11,107 34.64 ± 0.10 3.91 ± 0.01 3.28 ± 0.00 2.94 ± 0.02 14.03 ± 0.04 a AG 5181 34.36 ± 0.14 3.90 ± 0.01 3.30 ± 0.01 2.88 ± 0.03 13.64 ± 0.06 b GG 706 33.52 ± 0.40 3.97 ± 0.04 3.32 ± 0.01 2.77 ± 0.07 13.43 ± 0.17 b p 0.081 0.354 0.090 0.052 0.000 CD44-g.2294G > C CC 4404 34.24 ± 0.15 3.88 ± 0.01 b 3.29 ± 0.01 2.98 ± 0.03 a 13.59 ± 0.07 c CG 8462 34.61 ± 0.11 3.92 ± 0.01 a 3.28 ± 0.00 2.93 ± 0.02 a 14.05 ± 0.05 a GG 4128 34.59 ± 0.16 3.90 ± 0.02 ab 3.29 ± 0.01 2.80 ± 0.03 b 13.86 ± 0.07 b p 0.114 0.010 * 0.109 0.001 0.000 CD44-g.86895A > G AA 9266 34.37 ± 0.11 b 3.91 ± 0.01 b 3.30 ± 0.00 a 2.88 ± 0.02 13.79 ± 0.05 b AG 6538 34.51 ± 0.13 b 3.89 ± 0.01 b 3.28 ± 0.00 a 2.95 ± 0.02 13.85 ± 0.06 b GG 1190 35.61 ± 0.29 a 3.97 ± 0.03 a 3.24 ± 0.01 b 2.94 ± 0.06 14.82 ± 0.13 a p 0.003 0.007 0.000 0.127 0.000 CD44-g.86978G > A AA 8087 34.45 ± 0.11 3.93 ± 0.01 a 3.29 ± 0.00 2.87 ± 0.02 13.92 ± 0.05 AG 7177 34.56 ± 0.12 3.90 ± 0.01 ab 3.29 ± 0.00 2.93 ± 0.02 13.88 ± 0.05 GG 1730 34.61 ± 0.25 3.86 ± 0.02 b 3.27 ± 0.01 3.04 ± 0.05 13.75 ± 0.11 p 0.252 0.004 ** 0.097 0.316 0.060 Note: Data in the same column with different lowercase letters on the shoulder indicate signiﬁcant differences (p < 0.05); * indicates that the differences reach a signiﬁcant level (p < 0.05); indicates that the differences reach a highly signiﬁcant level (p < 0.01). Table 5. Effects of different genotypes of SNPs of CD44 gene on production performance of Holstein. 3.5.2. Association Analysis of 4 SNP Loci of CD44 Gene and Production Traits The association analysis results of different genotypes of the 4 SNP loci of CD44 gene with somatic cell score, milk urea nitrogen, measured daily milk yield, milk protein percentage, and milk fat percentage of Holstein cattle are shown in Table 5. The effect of CD44-g.2263A > G locus on milk urea nitrogen reached a very signiﬁcant level (p < 0.01), and the urea nitrogen content of AA cows was signiﬁcantly higher than that of AG and GG genotypes. The CD44-g.86978G > A locus had a very signiﬁcant effect on the milk fat rate of dairy cows (p < 0.01), and the AA type was the highest. CD44-g.2294G > C locus had extremely signiﬁcant effects on SCS and milk urea nitrogen contents (p < 0.01) and had signiﬁcant effects on milk fat percentage (p < 0.05). CG-type cows had a higher milk fat Animals 2023, 13, 89 9 of 15 9 of 15 percentage and milk urea nitrogen content, which were signiﬁcantly higher than CC-type cows and CG-type cows; CC-type cows and CG-type cows had higher SCS. The CD44- G.86895A > G locus had signiﬁcant effects on milk yield, milk fat rate, protein rate, and milk urea nitrogen content of dairy cows (p < 0.01). The milk yield, milk fat percentage, and milk urea nitrogen content of GG dairy cows were the highest, as they were signiﬁcantly higher than those of other genotypes, and the protein percentage of AA and AG cows was the highest. A square visual diagram of CD44 gene difference expression is shown in Figure 7. higher than those of other genotypes, and the protein percentage of AA and AG cows was the highest. A square visual diagram of CD44 gene difference expression is shown in Figure 7. Table 4. Effects of different genotypes of SNPs of ABCG2 gene on production performance of Holstein. 3.5.2. Association Analysis of 4 SNP Loci of CD44 Gene and Production Traits SNP Locus Genotype Record Number Tested Day Milk Yield Milk Fat Percentage Milk Protein Percentage Somatic Cell Score Urea Nitrogen (mg/dL) ABCG2-g.57261A > G AA 1275 34.00 ± 0.20 3.87 ± 0.02 b 3.28 ± 0.01 2.98 ± 0.04 13.83 ± 0.09 AG 6729 34.68 ± 0.11 3.90 ± 0.01 ab 3.28 ± 0.00 2.94 ± 0.02 13.90 ± 0.05 GG 8990 34.45 ± 0.13 3.92 ± 0.01 a 3.29 ± 0.01 2.88 ± 0.03 13.88 ± 0.06 p 0.154 0.004 0.388 0.063 0.856 ABCG2-g.80952G > T GG 2713 34.68 ± 0.20 a 3.92 ± 0.02 3.28 ± 0.01 b 2.82 ± 0.04 b 13.84 ± 0.09 b GT 8438 34.71 ± 0.11 a 3.90 ± 0.01 3.28 ± 0.00 b 2.91 ± 0.02 a 13.86 ± 0.05 b TT 5843 34.15 ± 0.13 b 3.90 ± 0.01 3.29 ± 0.00 a 2.95 ± 0.03 a 13.94 ± 0.06 a p 0.001 0.230 0.001 0.000 0.002 ABCG2-g.94683A > G AA 4329 34.61 ± 0.15 3.91 ± 0.01 3.28 ± 0.01 2.92 ± 0.03 14.01 ± 0.07 a AG 8059 34.47 ± 0.11 3.90 ± 0.01 3.29 ± 0.00 2.93 ± 0.02 13.81 ± 0.05 b GG 4606 34.49 ± 0.15 3.91 ± 0.01 3.29 ± 0.01 2.87 ± 0.03 13.88 ± 0.07 ab p 0.163 0.207 0.456 0.529 0.004 ABCG2-g.120017G > A AA 4551 34.37 ± 0.15 b 3.91 ± 0.01 a 3.29 ± 0.01 a 2.87 ± 0.03 b 13.71 ± 0.07 b AG 8419 34.77 ± 0.11 a 3.91 ± 0.01 a 3.28 ± 0.00 b 2.95 ± 0.02 a 14.01 ± 0.05 a GG 4024 34.14 ± 0.16 b 3.89 ± 0.01 b 3.29 ± 0.01 a 2.87 ± 0.03 b 13.82 ± 0.07 b p 0.007 0.041 * 0.001 0.005 0.000 ** Note: Data in the same column with different lowercase letters on the shoulder indicate signiﬁcant differences (p < 0.05); * indicates that the differences reach a signiﬁcant level (p < 0.05); ** indicates that the differences reach a highly signiﬁcant level (p < 0.01). Table 5. Effects of different genotypes of SNPs of CD44 gene on production performance of Holstein. 3.5.2. Association Analysis of 4 SNP Loci of CD44 Gene and Production Traits Note: Data in the same column with different lowercase letters on the shoulder indicate signiﬁcant differences (p < 0.05); * indicates that the differences reach a signiﬁcant level (p < 0.05); ** indicates that the differences reach a highly signiﬁcant level (p < 0.01). 3.5.3. Association Analysis of Two SNP Loci in SPP1 Gene and Production Traits 3.5.3. Association Analysis of Two SNP Loci in SPP1 Gene and Production Traits The correlation analysis results of the two SNP loci of the SPP1 gene between different genotypes and Holstein cattle somatic cell score, urea nitrogen in milk, daily milk yield, milk protein rate, and milk fat rate are shown in Table 6. The SPP1-g.50265G > A locus had a signiﬁcant level of effect on milk yield p < 0.05) and had a highly signiﬁcant effect on milk urea nitrogen content (p < 0.01). The milk yield and milk urea nitrogen content of the AG and GG cows were signiﬁcantly higher than those of the AA cows. The SPP1-g.50315 C > T locus had a signiﬁcant effect on the milk fat rate of dairy cows (p < 0.05), and the CT-type and TT-type were signiﬁcantly higher than the CC-type. A square visual diagram of SPP1 gene difference expression is shown in Figure 8. 10 of 15 10 of 15 Animals 2023, 13, 89 Figure 6. Effect of different genotypes in ABCG2 four SNP loci on production traits in Chinese Holstein cattle. (A) The milk fat percentage of GG-type cows in the ABCG2-g.57261A > G locus was signiﬁcantly higher than that of AA-type and AG-type cows. (B) ABCG2-g.94683A > G locus had signiﬁcantly higher urea nitrogen content in the milk of AA type cows than AG and GG types. In ABCG2-g.80952G > T locus, the tested day milk yield (C) of TT type cows was signiﬁcantly lower than that of GG and GT type cows; somatic cell score (D), milk protein percentage (E), and urea nitrogen (F) of TT type cows were signiﬁcantly higher than GG and GT types. In ABCG2-g.120017G > A locus, AA type cows tested day milk yield (G) was signiﬁcantly higher than that of AG and GG type cows, and the milk protein percentage (H), milk fat percentage (I), somatic cell score (J) and urea nitrogen (K) of AG type cows were signiﬁcantly higher than those of AA and GG type cows. 3.5.2. Association Analysis of 4 SNP Loci of CD44 Gene and Production Traits Data in the same column with different lowercase letters on the shoulder indicate signiﬁcant differences (p < 0.05). Figure 6. Effect of different genotypes in ABCG2 four SNP loci on production traits in Chinese Figure 6. Effect of different genotypes in ABCG2 four SNP loci on production traits in Chinese Holstein cattle. (A) The milk fat percentage of GG-type cows in the ABCG2-g.57261A > G locus was signiﬁcantly higher than that of AA-type and AG-type cows. (B) ABCG2-g.94683A > G locus had signiﬁcantly higher urea nitrogen content in the milk of AA type cows than AG and GG types. In ABCG2-g.80952G > T locus, the tested day milk yield (C) of TT type cows was signiﬁcantly lower than that of GG and GT type cows; somatic cell score (D), milk protein percentage (E), and urea nitrogen (F) of TT type cows were signiﬁcantly higher than GG and GT types. In ABCG2-g.120017G > A locus, AA type cows tested day milk yield (G) was signiﬁcantly higher than that of AG and GG type cows, and the milk protein percentage (H), milk fat percentage (I), somatic cell score (J) and urea nitrogen (K) of AG type cows were signiﬁcantly higher than those of AA and GG type cows. Data in the same column with different lowercase letters on the shoulder indicate signiﬁcant differences (p < 0.05). Figure 6. Effect of different genotypes in ABCG2 four SNP loci on production traits in Chinese Holstein cattle. (A) The milk fat percentage of GG-type cows in the ABCG2-g.57261A > G locus was signiﬁcantly higher than that of AA-type and AG-type cows. (B) ABCG2-g.94683A > G locus had signiﬁcantly higher urea nitrogen content in the milk of AA type cows than AG and GG types. In ABCG2-g.80952G > T locus, the tested day milk yield (C) of TT type cows was signiﬁcantly lower than that of GG and GT type cows; somatic cell score (D), milk protein percentage (E), and urea nitrogen (F) of TT type cows were signiﬁcantly higher than GG and GT types. In ABCG2-g.120017G > A locus, AA type cows tested day milk yield (G) was signiﬁcantly higher than that of AG and GG type cows, and the milk protein percentage (H), milk fat percentage (I), somatic cell score (J) and urea nitrogen (K) of AG type cows were signiﬁcantly higher than those of AA and GG type cows. 3.5.2. Association Analysis of 4 SNP Loci of CD44 Gene and Production Traits Data in the same column with different lowercase letters on the shoulder indicate signiﬁcant differences (p < 0.05). 11 of 15 Animals 2023, 13, 89 Figure 7. Effect of different genotypes in CD44 four SNP loci on production traits in Chinese Holstein cattle. In CD44-g.2263A > G locus, AA type urea nitrogen content of was signiﬁcantly higher than that of AG and GG genotypes (A). In CD44-g.2294G > C locus, milk fat percentage (B) and milk urea nitrogen content of CG-type (D) cows were signiﬁcantly higher than CC and CG types. Higher SCS for CC- and CG-type cows (C). In CD44-g.86978G > A locus, Type AA cows have the highest milk fat percentage (E). In CD44-g.86895A > G locus, tested day milk yield (F), milk fat rate (G), and milk urea nitrogen content (I) of GG-type cows were signiﬁcantly higher than those of AA, AG types; AA and AG cows had the highest protein rates (H). Data in the same column with different lowercase letters on the shoulder indicate signiﬁcant differences (p < 0.05). Figure 7. Effect of different genotypes in CD44 four SNP loci on production traits in Chinese Holstein cattle. In CD44-g.2263A > G locus, AA type urea nitrogen content of was signiﬁcantly higher than that of AG and GG genotypes (A). In CD44-g.2294G > C locus, milk fat percentage (B) and milk urea nitrogen content of CG-type (D) cows were signiﬁcantly higher than CC and CG types. Higher SCS for CC- and CG-type cows (C). In CD44-g.86978G > A locus, Type AA cows have the highest milk fat percentage (E). In CD44-g.86895A > G locus, tested day milk yield (F), milk fat rate (G), and milk urea nitrogen content (I) of GG-type cows were signiﬁcantly higher than those of AA, AG types; AA and AG cows had the highest protein rates (H). Data in the same column with different lowercase letters on the shoulder indicate signiﬁcant differences (p < 0.05). Animals 2023, 13, 89 12 of 15 12 of 15 Table 6. Effects of different genotypes of SNPs of SPP1 gene on production performance of Holstein. SNP Locus Genotype Record Number Tested Day Milk Yield Milk Fat Rate Milk Protein Rate Somatic Cell Score Urea Nitrogen SPP1-g. 3.5.2. Association Analysis of 4 SNP Loci of CD44 Gene and Production Traits SPP1-g.50265G > A and SPP1-g.50315C > T loci on the productive performance of different genotypes. tested day milk yield (A) and milk urea nitrogen content (C) were signiﬁcantly higher in AG and GG types of cows than in AA-type; CT and TT types of cows were signiﬁcantly higher than the CC-type, milk fat rate (B) of CT and TT types were signiﬁcantly higher than that of CC-type. Data in the same column with different lowercase letters on the shoulder indicate signiﬁcant differences (p < 0.05). Figure 8. SPP1-g.50265G > A and SPP1-g.50315C > T loci on the productive performance of different genotypes. tested day milk yield (A) and milk urea nitrogen content (C) were signiﬁcantly higher in AG and GG types of cows than in AA-type; CT and TT types of cows were signiﬁcantly higher than the CC-type, milk fat rate (B) of CT and TT types were signiﬁcantly higher than that of CC-type. Data in the same column with different lowercase letters on the shoulder indicate signiﬁcant differences (p < 0.05). 3.5.2. Association Analysis of 4 SNP Loci of CD44 Gene and Production Traits 50265G > A AA 745 33.61 ± 0.37 b 3.89 ± 0.03 3.29 ± 0.01 2.90 ± 0.07 13.23 ± 0.16 b AG 6121 34.47 ± 0.13 a 3.90 ± 0.01 3.29 ± 0.00 2.93 ± 0.02 13.78 ± 0.06 a GG 10,128 34.60 ± 0.10 a 3.92 ± 0.01 3.28 ± 0.00 2.90 ± 0.02 13.99 ± 0.04 a p 0.016 * 0.215 0.619 0.843 0.000 ** SPP1-g. 50315 C > T CC 5427 34.50 ± 0.13 3.88 ± 0.01 b 3.28 ± 0.00 2.89 ± 0.03 14.02 ± 0.06 CT 8272 34.50 ± 0.11 3.92 ± 0.01 a 3.29 ± 0.00 2.93 ± 0.02 13.81 ± 0.05 TT 3295 34.54 ± 0.18 3.91 ± 0.01 a 3.28 ± 0.01 2.90 ± 0.03 13.83 ± 0.08 p 0.420 0.016 * 0.360 0.396 0.134 Note: Data in the same column with different lowercase letters on the shoulder indicate signiﬁcant differences (p < 0.05); * indicates that the differences reach a signiﬁcant level (p < 0.05); ** indicates that the differences reach a highly signiﬁcant level (p < 0.01). Table 6. Effects of different genotypes of SNPs of SPP1 gene on production performance of Holstein. Note: Data in the same column with different lowercase letters on the shoulder indicate signiﬁcant differences (p < 0.05); * indicates that the differences reach a signiﬁcant level (p < 0.05); ** indicates that the differences reach a highly signiﬁcant level (p < 0.01). Figure 8. SPP1-g.50265G > A and SPP1-g.50315C > T loci on the productive performance of different genotypes. tested day milk yield (A) and milk urea nitrogen content (C) were signiﬁcantly higher in AG and GG types of cows than in AA-type; CT and TT types of cows were signiﬁcantly higher than the CC-type, milk fat rate (B) of CT and TT types were signiﬁcantly higher than that of CC-type. Data in the same column with different lowercase letters on the shoulder indicate signiﬁcant differences (p < 0.05). Figure 8. SPP1-g.50265G > A and SPP1-g.50315C > T loci on the productive performance of different genotypes. tested day milk yield (A) and milk urea nitrogen content (C) were signiﬁcantly higher in AG and GG types of cows than in AA-type; CT and TT types of cows were signiﬁcantly higher than the CC-type, milk fat rate (B) of CT and TT types were signiﬁcantly higher than that of CC-type. Data Figure 8. 4. Discussion However, the SCS of cows with these two genotypes are also relatively high, which may increase the burden of udder function with the increase in milk production, aggravate the shedding of mammary epithelial cells, and increase the SCS of cows. Olsen [14] found an SNP mutation in the ABCG2 gene in exon 9, resulting in a mutation of the amino acid from tyrosine to cysteine, but the differences in the ﬁve lactation performances among the three genotypes were not signiﬁcant. In the selection of dairy cow genotypes, it is necessary to comprehensively select the SNPs genotypes corresponding to the target traits, so we are required to use different methods to ﬁnd more SNP loci that affect dairy cow traits. y Most of the existing studies on CD44 have focused on disease treatment. CD44 gene rs13347 polymorphism is signiﬁcantly associated with elevated cancer risk in Asians [15]. Deletion of the CD44 gene has been reported to reduce proinﬂammatory cytokines [16], and treatment of affected mice with CD44 antibodies reduced inﬂammation [17]. Previous studies on CD44 in cattle have shown that CD44 is related to dairy production traits. Jiang [18] believed that the CD44 gene could affect the synthesis of triglyceride in bovine mammary epithelial cells, thereby affecting the lipid content in milk. A total of 4 SNP loci were found in this study, CD44-g.2263A > G and CD44-g.2294G > C were located in the intron region, CD44-G.86895A > G and CD44-G.86978G > A were located in the exon region, the CD44-g.86895A > G site is a nonsense mutation, and the CD44-g.86978G > A site is a missense mutation. Association analysis found that the milk urea nitrogen content of cows with CD44-g.2263A > G locus AA genotype was signiﬁcantly higher than that of AG and GG genotypes. The CD44-g.86978G > A locus had a signiﬁcant effect on milk fat percentage in cows, with the AA type having the highest. Cows with the CD44-G. 2294G > C CG genotype had a higher milk fat percentage and milk urea nitrogen content, which were signiﬁcantly higher than those with CC and CG genotypes, while cows with CC and CG genotypes had higher SCS. Cows carrying the CD44-g.86895A > G locus GG genotype had the highest milk yield, milk fat percentage, and milk urea nitrogen content, which were signiﬁcantly higher than other genotypes, while cows with AA and AG genotypes had the highest protein rate. 4. Discussion Previous studies analyzed the association between ABCG2 gene polymorphism and milk yield and milk composition of Kalanfris cattle and found that the three genotypes of SNP on exon 14 had signiﬁcant effects on the milk fat percentage of cows [11]. Tantia identiﬁed the quantitative trait loci of Indian cattle (Bos indicus) and river cattle (Bubalus bubalis) and found that the allele of ABCG2 had a signiﬁcant impact on high milk fat yield, high fat, and protein percentage [12]. In this study, four polymorphisms were found in the ABCG2 gene, and all these polymorphisms were located in the intronic region. Existing studies have shown that introns can generate splicing signals through their own sequence properties to affect transcript splicing, thereby affecting protein sequences and individual phenotypes [13]. The results of this study showed that ABCG2-G.57261A > G locus had a signiﬁcant effect on the milk fat percentage of dairy cows, and GG type was the highest. Milk urea nitrogen content in the AA genotype of ABCG2-G.94683A > G locus was signiﬁcantly higher than that in the AG and GG genotypes. Milk protein percentage, somatic cell score and urea nitrogen of cows carrying ABCG2-G.80952G > T TT genotype were signiﬁcantly higher than those of the GG and GT genotypes. Milk protein percentage, milk fat percentage, somatic cell score, and urea nitrogen were signiﬁcantly higher in cows carrying the AG genotype at the ABCG2-g.120017G > A locus than in the AA and GG types. The effects of ABCG2-G.80952G > T and ABCG2-G.120017G > A on production traits of dairy cows showed similar trends, probably because of the high linkage Animals 2023, 13, 89 13 of 15 13 of 15 relationship between them. The results of this study show that ABCG2 polymorphisms have signiﬁcant effects on dairy cows’ milk fat rate, milk yield, milk protein rate, and other production traits, which are basically consistent with previous studies. In the selection of dairy cows, the selection of ABCG2-g.57261A > G site GG type and ABCG2-g.120017G > A site AG type dairy cows should be increased in view of improving milk fat rate. In order to improve milk production and milk protein rate of dairy cows, the selection of TT-type cows at ABCG2-g.80952G > T site and AG-type cows at ABCG2-g.120017G > A site can be strengthened. 4. Discussion From the results of this study, the genotypes of the three SNP loci have a signiﬁcant impact on the milk fat rate of dairy cows, which proves that CD44 is related to the lipid metabolism of dairy cows. This result is consistent with previous studies. In order to improve milk fat percentage, the selection of dairy cows carrying CD44-G.86978G > A AA genotype, CD44-G.2294G > C CG genotype and CD44-G.86895A > G GG genotype should be strengthened. There are more and more studies on SPP1 gene polymorphism, which provides a powerful reference for disease treatment and improving livestock and poultry production. y Maria [19] conducted an association analysis of the SPP1 gene polymorphism in Salda sheep and found that rs161844011 locus in exon 7 was associated with SCS. Liang [20] found that people carrying the CC genotype and C allele of the rs11730582 locus of the SPP1 gene had a reduced risk of developing breast cancer. Furthermore, SPP1 gene polymorphism was signiﬁcantly correlated with lactation persistence of dairy cows, and G was its dominant allele [21]. Mello analyzed the association between SPP1 gene polymorphism and milk yield of Girorando cows and found that 305-day cows carrying the T allele at position g.8514C > T had higher milk yield, but it did not reach a signiﬁcant level [22]. The SNPs of the two SPP1 genes obtained in this study were located in the intronic region. The milk yield and milk urea nitrogen content of the AG and GG genotypes of SPP1-G. 50265G > A were signiﬁcantly higher than those of the AA genotypes. SPP1-g.50315 C > T loci CT Animals 2023, 13, 89 14 of 15 14 of 15 and TT genotypes had signiﬁcantly higher milk fat percentages than the CC types. The selection of cows carrying the G allele at the SPP1-G.50265G > A locus can be strengthened in future dairy breeding to improve milk yield. From the perspective of improving the milk fat rate of dairy cows, the selection of dairy cows carrying the SPP1-g.50315 C > T locus T allele can be strengthened. 5. Conclusions This work performed GO annotation and KEGG pathway enrichment analysis on the ABCG2, CD44 and SPP1 genes and analyzed the association between the ABCG2, CD44, and SPP1 genes and Chinese Holstein production performance. The milk yield, milk fat rate, milk protein rate, somatic cell score, and blood urea nitrogen content of Holstein cattle with different genotypes of these 10 SNP loci were different. Among them, ABCG2-G.80952G > T locus, ABCG2-G.120017G > A locus and CD44-G.2294G > C locus had signiﬁcant effects on the somatic cell score (p < 0.01). The present study elucidated that ABCG2, CD44 and SPP1 could be selected for marker-assisted selection and will beneﬁt future precise molecular breeding. Author Contributions: Conceptualization, Y.S.; data curation, X.W.; validation, D.L.; formal analysis, X.L.; investigation, Y.M. and T.Z.; writing—original draft preparation, X.W. and D.L.; writing—review and editing, Y.S.; supervision, Z.Y.; project administration, X.L.; funding acquisition, Y.S. and Z.Y. All authors have read and agreed to the published version of the manuscript. Funding: This work was supported by the National Natural Science Foundation of China (No. 31902147, 31872324), the Open Project Program of Joint International Research Laboratory of Agri- culture and Agri-Product Safety, the Ministry of Education of China, Yangzhou University (JILAR- KF202013). Funding: This work was supported by the National Natural Science Foundation of China (No. 31902147, 31872324), the Open Project Program of Joint International Research Laboratory of Agri- culture and Agri-Product Safety, the Ministry of Education of China, Yangzhou University (JILAR- KF202013). Institutional Review Board Statement: All animal experiments were approved by the Yangzhou University Institutional Animal Care and Use Committee (license number: SYXK [Su] 2017-0044). Institutional Review Board Statement: All animal experiments were approved by the Yangzhou University Institutional Animal Care and Use Committee (license number: SYXK [Su] 2017-0044). Informed Consent Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: The data presented in this study are available in this article. Conﬂicts of Interest: The authors declare no conﬂict of interest. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. El-Bayomi, K.M.; El Araby, I.E.; Awad, A.; Zaglool, A.W. Molecular study of some candidate genes affecting milk production traits in Holstein cattle. Alex. J. Vet. Sci. 2015, 45, 146–150. 2. Thakur, S.; Singh, P.K.; Rathour, R.; Variar, M.; Prashanthi, S.K.; Gopalakrishnan, S.; Singh, A.; Singh, U.; Chand, D.; Singh, N.K. Genotyping and development of single-nucleotide polymorphism (SNP) markers associated with blast resistance genes in rice using GoldenGate assay. Mol. Breed. 2014, 34, 1449–1463. [CrossRef] 3. Gao, L.; Jia, J.; Kong, X. A SNP-Based Molecular Barcode for Characterization of Common Wheat. PLoS ONE 2016, 11, e0150947. [CrossRef] [PubMed] 3. Gao, L.; Jia, J.; Kong, X. A SNP-Based Molecular Barcode for Characterization of Common Wheat. PLoS ONE 2016, 11, e0150947. [CrossRef] [PubMed] 4. Sun, Y.; Lan, X.; Lei, C.; Zhang, C.; Chen, H. Haplotype combination of the bovine CFL2 gene sequence variants and association [CrossRef] [PubMed] 4. Sun, Y.; Lan, X.; Lei, C.; Zhang, C.; Chen, H. Haplotype combination of the bovine CFL2 gene sequence variants and association 4. Sun, Y.; Lan, X.; Lei, C.; Zhang, C.; Chen, H. Haplotype combination of the bovine CFL2 gene sequence variants and association with growth traits in Qinchuan cattle. Gene 2015, 563, 136–141. [CrossRef] 4. Sun, Y.; Lan, X.; Lei, C.; Zhang, C.; Chen, H. Haplotype combination of the bovine with growth traits in Qinchuan cattle. Gene 2015, 563, 136–141. [CrossRef] , ; , ; , ; g, ; , p yp with growth traits in Qinchuan cattle. Gene 2015, 563, 136–141. [CrossRef] g 5. Sun, Y.; Liu, K.; Huang, Y.; Lan, X.; Chen, H. Differential expression of FOXO1 during development and myoblast differentiation of Qinchuan cattle and its association analysis with growth traits. Sci. China Life Sci. 2018, 61, 826–835. [CrossRef] 6. Sun, Y.; Xu, C.; Yang, Z.; Li, M.; Chen, Z.; Xu, T.; Zhang, H.; Mao, Y. The polymorphism of bovine Coﬁlin-1 gene sequence variants and association analysis with growth traits in Qinchuan cattle. Anim. Biotechnol. 2022, 33, 63–69. [CrossRef] 6. Sun, Y.; Xu, C.; Yang, Z.; Li, M.; Chen, Z.; Xu, T.; Zhang, H.; Mao, Y. The polymorphism of bovine Coﬁlin 1 gene sequence variants and association analysis with growth traits in Qinchuan cattle. Anim. Biotechnol. 2022, 33, 63–69. [CrossRef] 7. Lu, X.; Arbab, A.A.I.; Abdalla, I.M.; Liu, D.; Zhang, Z.; Xu, T.; Su, G.; Yang, Z. , A.P.; Chakravarty, A.K.; Mir, M.A.; Arya, A.; Singh, M. Identiﬁcation of Genetic Variants in ABCG2 Gene In uction Traits in Dairy Cattle. Indian J. Anim. Res. 2020, 54, 275–281. [CrossRef] References Genetic Parameter Estimation and Genome-Wide Association Study-Based Loci Identiﬁcation of Milk-Related Traits in Chinese Holstein. Front. Genet. 2021, 12, 799664. [CrossRef] 8. Lu, X.; Abdalla, I.M.; Nazar, M.; Fan, Y.; Zhang, Z.; Wu, X.; Xu, T.; Yang, Z. Genome-Wide Association Study on Reproduction- Related Body-Shape Traits of Chinese Holstein Cows. Animals 2021, 11, 1927. [CrossRef] 7. Lu, X.; Arbab, A.A.I.; Abdalla, I.M.; Liu, D.; Zhang, Z.; Xu, T.; Su, G.; Yang, Z. Genetic Parameter Estimation and Genome-Wide Association Study-Based Loci Identiﬁcation of Milk-Related Traits in Chinese Holstein. Front. Genet. 2021, 12, 799664. [CrossRef] 8 Lu X ; Abdalla I M ; Nazar M ; Fan Y; Zhang Z ; Wu X ; Xu T ; Yang Z Genome-Wide Association Study on Reproduction- 7. Lu, X.; Arbab, A.A.I.; Abdalla, I.M.; Liu, D.; Zhang, Z.; Xu, T.; Su, G.; Yang, Z. Genetic Parameter Estimation and Genome-Wide Association Study-Based Loci Identiﬁcation of Milk-Related Traits in Chinese Holstein. Front. Genet. 2021, 12, 799664. [CrossRef] 8. Lu, X.; Abdalla, I.M.; Nazar, M.; Fan, Y.; Zhang, Z.; Wu, X.; Xu, T.; Yang, Z. Genome-Wide Association Study on Reproduction- l d d h f h l l [ f] y p 9. Shi, Y.Y.; Lin, H.E. SHEsis, a powerful software platform for analyses of linkage disequilibrium, haplotype construction, and genetic association at polymorphism loci. Cell Res. 2005, 15, 97–108. [CrossRef] 9. Shi, Y.Y.; Lin, H.E. SHEsis, a powerful software platform for analyses of linkage disequilibrium, haplotype construction, and genetic association at polymorphism loci. Cell Res. 2005, 15, 97–108. [CrossRef] 10. Ardlie, K.G.; Kruglyak, L.; Seielstad, M. Patterns of linkage disequilibrium in the human genome. Nat. Rev. Genet. 2002, 3, 299–309. [CrossRef] 10. Ardlie, K.G.; Kruglyak, L.; Seielstad, M. Patterns of linkage disequilibrium in the human genome. Nat. Rev. Genet. 2002, 3, 299–309. [CrossRef] 11. Singh, A.P.; Chakravarty, A.K.; Mir, M.A.; Arya, A.; Singh, M. Identiﬁcation of Genetic Variants in ABCG2 Gene Inﬂuencing Milk Production Traits in Dairy Cattle. Indian J. Anim. Res. 2020, 54, 275–281. [CrossRef] 11. Singh, A.P.; Chakravarty, A.K.; Mir, M.A.; Arya, A.; Singh, M. Identiﬁcation of Genetic Variants in ABCG2 Gene Inﬂuencing Milk Production Traits in Dairy Cattle. Indian J. Anim. Res. 2020, 54, 275–281. [CrossRef] Animals 2023, 13, 89 15 of 15 15 of 15 12. Tantia, M.S.; Vijh, R.K.; Mishra, B.P.; Mishra, B.; Kumar, S.; Sodhi, M. DGAT1 and ABCG2 polymorphism in Indian cattle (Bos indicus) and buffalo (Bubalus bubalis) breeds. References BMC Vet. Res. 2006, 2, 32. [CrossRef] [PubMed] ) ( ) 13. Shaul, O. How introns enhance gene expression. Int. J. Biochem. Cell Biol. 2017, 91, 145–155. [CrossRef] g p 14. Olsen, H.G.; Nilsen, H.; Hayes, B.; Berg, P.R.; Svendsen, M.; Lien, S.; Meuwissen, T. Genetic suppo nucleotide in the ABCG2 gene affecting milk composition of dairy cattle. BMC Genet. 2007, 8, 32. [Cros g g p y , , [ ] [ ] 15. Shi, J.; Duan, Y.; Pan, L.; Zhou, X. Positive association between CD44 gene rs13347 C>T polymorphism and risk of cancer in d l h [ f] [ b d] g g p y 15. Shi, J.; Duan, Y.; Pan, L.; Zhou, X. Positive association between CD44 gene rs13347 C>T polymorphism and risk of cancer in Asians: A systemic review and meta-analysis OncoTargets Ther 2016 9 3493–3500 [CrossRef] [PubMed] 15. Shi, J.; Duan, Y.; Pan, L.; Zhou, X. Positive association between CD44 gene rs13347 C>T polymorph Asians: A systemic review and meta-analysis. OncoTargets Ther. 2016, 9, 3493–3500. [CrossRef] [PubMe y y g 16. Fu, Q.; Hou, L.; Xiao, P.; Guo, C.; Chen, Y.; Liu, X. CD44 deﬁciency leads to decreased proinﬂammatory cytokine production in lung induced by PCV2 in mice. Res. Vet. Sci. 2014, 97, 498–504. [CrossRef] [PubMed] y y g 16. Fu, Q.; Hou, L.; Xiao, P.; Guo, C.; Chen, Y.; Liu, X. CD44 deﬁciency leads to decreased proinﬂammatory cytokine production in lung induced by PCV2 in mice Res Vet Sci 2014 97 498 504 [CrossRef] [PubMed] 16. Fu, Q.; Hou, L.; Xiao, P.; Guo, C.; Chen, Y.; Liu, X. CD44 deﬁciency leads to decreased proinﬂammator lung induced by PCV2 in mice. Res. Vet. Sci. 2014, 97, 498–504. [CrossRef] [PubMed] 17. Raﬁ-Janajreh, A.Q.; Chen, D.; Schmits, R.; Mak, T.W.; Grayson, R.L.; Sponenberg, D.P.; Nagarkatti, M.; Nagarkatti, P.S. Evidence for the involvement of CD44 in endothelial cell injury and induction of vascular leak syndrome by IL-2. J. Immunol. 1999, 163, 1619–1627. 18. Jiang, P.; Fang, X.; Xia, L.; Li, X.; Wang, M.; Yang, R.; Zhao, Z. The effect of CD44 gene on triglyceride synthesis of bovine mammary epithelial cells. J. Anim. Sci. 2018, 96, 136–137. [CrossRef] 19. Dettori, M.L.; Pazzola, M.; Petretto, E.; Vacca, G.M. Association Analysis between SPP1, POFUT1 and PRLR Gene Variation and Milk Yield, Composition and Coagulation Traits in Sarda Sheep. Animals 2020, 10, 1216. [CrossRef] 20. Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. References Liang, L.; Lu, G.; Pan, G.; Deng, Y.; Liang, J.; Liang, L.; Liu, J.; Tang, Y.; Wei, G. A Case-Control Study of the Association Between the SPP1 Gene SNPs and the Susceptibility to Breast Cancer in Guangxi, China. Front. Oncol. 2019, 9, 1415. [CrossRef] p y g 21. Bissonnette, N. Short communication: Genetic association of variations in the osteopontin gene (SPP1) with lactation persistency in dairy cattle. J. Dairy Sci. 2018, 101, 456–461. [PubMed] y J y , , [ ] 22. De Mello, F.; Cobuci, J.A.; Martins, M.F.; Silva, M.V.; Neto, J.B. Association of the polymorphism g.8514C>T in the osteopontin gene (SPP1) with milk yield in the dairy cattle breed Girolando. Anim. Genet. 2012, 43, 647–648. [CrossRef] [PubMed] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
https://openalex.org/W4205294761	https://link.springer.com/content/pdf/10.1007/s00068-021-01858-y.pdf	English	null	Lessons learned from terror attacks: thematic priorities and development since 2001—results from a systematic review	European journal of trauma and emergency surgery	2,022	cc-by	19,631	Abstract Purpose The threat of national and international terrorism remains high. Preparation is the key requirement for the resilience of hospitals and out-of-hospital rescue forces. The scientific evidence for defining medical and tactical strategies often feeds on the analysis of real incidents and the lessons learned derived from them. This systematic review of the literature aims to identify and systematically report lessons learned from terrorist attacks since 2001.i Methods PubMed was used as a database using predefined search strategies and eligibility criteria. All countries that are part of the Organization for Economic Cooperation and Development (OECD) were included. The time frame was set between 2001 and 2018. Results Finally 68 articles were included in the review. From these, 616 lessons learned were extracted and summarized into 15 categories. The data shows that despite the difference in attacks, countries, and casualties involved, many of the lessons learned are similar. We also found that the pattern of lessons learned is repeated continuously over the time period studied. Conclusions The lessons from terrorist attacks since 2001 follow a certain pattern and remained constant over time. There- fore, it seems to be more accurate to talk about lessons identified rather than lessons learned. To save as many victims as possible, protect rescue forces from harm, and to prepare hospitals at the best possible level it is important to implement the lessons identified in training and preparation. Keywords Terror attacks · Evaluation · Lessons learned · Emergency preparedness · Public health preparedness · Mass casualties Keywords Terror attacks · Evaluation · Lessons learned · Emergency preparedness · Public health preparedness · Mass casualties Lessons learned from terror attacks: thematic priorities and development since 2001—results from a systematic review Nora Schorscher1 · Maximilian Kippnich1 · Patrick Meybohm1 · Thomas Wurmb2 Received: 22 September 2021 / Accepted: 29 November 2021 / Published online: 13 January 2022 © The Author(s) 2022 2 Subsection Emergency and Disaster Relief Medicine, Department of Anaesthesiology, Intensive Care, Emergency and Pain Medicine, University Hospital Wuerzburg, Oberduerrbacherstrasse 6, 97080 Wuerzburg, Germany European Journal of Trauma and Emergency Surgery (2022) 48:2613–2638 https://doi.org/10.1007/s00068-021-01858-y European Journal of Trauma and Emergency Surgery (2022) 48:2613–2638 https://doi.org/10.1007/s00068-021-01858-y European Journal of Trauma and Emergency Surgery (2022) 48:2613–2638 https://doi.org/10.1007/s00068-021-01858-y REVIEW ARTICLE 1 Department of Anaesthesiology, Intensive Care, Emergency and Pain Medicine, University Hospital Wuerzburg, Wuerzburg, 97080, Germany Materials and methods 3. Articles dealing exclusively with chemical, biological, radiological and nuclear (CBRN) terrorism Study design and search strategy 4. Articles dealing with a narrow point of view and only dealing with specific types of injuries such as burns or psychiatry This is a systematic review of the literature with the focus on lessons learned from terror attacks. A comprehensive literature search was performed to identify articles report- ing medical and surgical management of terrorist attacks and lessons learned derived from them. PubMed was used as database. The first search term concentrated on terror- ism, the second on medical/surgical management and the third on evaluation and lessons learned. Adapted PRISMA guidelines were used and all articles were checked and reported against its checklist [12]. 5. Articles not written in English. Articles dealing exclusively with chemical, biological, radiological and nuclear terrorism (CBRN-attacks) were excluded from the literature-search. The reason for this is the large number of special problems and issues associated with this type of incident. To address this adequately, a separate literature search would be necessary. The search terms were formulated as an advanced search in PubMed in the following way: Search: ((Ter- ror* OR Terror* Attack* OR Terrorism* OR Mass Casult* Incident* OR Mass Shooting* OR Suicide Attack* OR Suicide Bomb* OR Rampage* OR Amok) AND (Prehos- pital Care* OR Emergenc* Medical* Service* OR Emer- genc* Service* OR Emergenc* Care* OR Rescue Mis- sion* OR Triage* OR Disaster* Management* OR First* Respon)) AND (Lesson Learn* OR Quality Indicator* OR Evaluation* OR Analysis* OR Review* OR Report* OR Deficit* OR Problem*). Background The emergency management of terrorist attacks has been one of the prominent topics in disaster and emergency medi- cine before the SARS-CoV-2 pandemic. The most recent attacks have shown that this particular threat is still present and highly relevant today [1–4]. The idea of “stopping the dying as well as the killing”, which has been coined by Park et al. after the London Bridge and Borough Market attacks in 2017, emphasizes the urgent need to focus on emergency management and early medical and surgical intervention [5]. Rescue systems and hospitals must prepare themselves to manage terrorist attacks in order to save as many lives as possible and to return rescue forces from the missions unscathed. As it is impossible to conduct prospective, high-quality scientific studies, the definition of these medical and tactical strategies relies on the analysis of 1 Department of Anaesthesiology, Intensive Care, Emergency and Pain Medicine, University Hospital Wuerzburg, Wuerzburg, 97080, Germany 2 Subsection Emergency and Disaster Relief Medicine, Department of Anaesthesiology, Intensive Care, Emergency and Pain Medicine, University Hospital Wuerzburg, Oberduerrbacherstrasse 6, 97080 Wuerzburg, Germany (0123 1 3456789) 3 2614 N. Schorscher et al. Eligibility criteria and study selection real incidents and the lessons learned derived from them. After the Paris terror attacks in 2015 for example, impor- tant publications, describing the events of the night of the 13th of November 2015, were published [6, 7]. Two publications, one by the French Health Ministry and one by Carli et al., about the “Parisian night of terror” have gone a step further and have clearly described the les- sons learned from these attacks [8, 9]. Importantly, experts agree on the importance of the scientific and systematic evaluation of the most recent terror attacks [10]. Challen et al. proved the existence of a large body of literature on the topic in 2012 already, but questioned its validity and generalisability. The authors based their conclusion on a review, which focused on emergency planning for any kind of disaster [11]. Time frame: The attack on the World Trade Centre in New York, the Pentagon in Arlington, and the crash of a hijacked airliner in 2001 is considered the event that brought international terrorism onto the world stage with the beginning of the new millennium. The attacks have been documented and analysed in great detail. For this rea- son, this analysis starts in 2001 and ends with the terrorist attacks in London and Manchester in 2017. The search history was extended to the year 2018. Included countries: Terrorism is a worldwide phenome- non. Attempting to evaluate the data of all terrorist attacks that have occurred since 2001 seems impossible due to the extremely high number. The work therefore focuses mainly on Western-oriented democracies, for which a terrorist attack is still a relatively rare event and whose infrastruc- ture and emergency services recently had to adapt to this challenge. The Organization for Economic Cooperation and Development (OECD)—countries therefore represent a reasonable selection of countries for this study. More than ever, the principle applies, that the prepa- ration for extraordinary disastrous incidents is the deci- sive prerequisite for successful management. The lack of preparedness for the SARS-CoV-2 pandemic has taught modern society this lesson. With the aim to identify and systematically report the lessons learned from terrorist attacks as an important basis for preparation, we conducted the presented systematic review of the literature. Exclusion criteria: Exclusion criteria: 1. Articles reporting mass casualty incidents without a ter- roristic background 2. Personal reports without any clear defined lessons learned Data abstraction The lessons learned from each included article were extracted according to the inclusion and exclusion cri- teria. Duplicated data was excluded. As expected, there was a vast number of individual lessons learned. To sum- marize the results, it was imperative to divide them into categories. As a basis for developing the categories exist- ing systems were used. The reporting system of Fattah et al. defines 6 categories, but these were not sufficient to 1 3 2615 Lessons learned from terror attacks: thematic priorities and development since 2001—results… Fig. 1 Process to identify the articles included in the systematic review represent all types of lessons learned [13]. Wurmb et al. had recently developed 13 clusters of quality indicators [14], some of which we were able to adopt. However, both systems focused on categories that serve to describe the overall setting of a rescue mission and were therefore not fully suitable for clustering lessons learned. Finally these 15 categories were used for clustering the lessons learned: represent all types of lessons learned [13]. Wurmb et al. had recently developed 13 clusters of quality indicators [14], some of which we were able to adopt. However, both systems focused on categories that serve to describe the overall setting of a rescue mission and were therefore not fully suitable for clustering lessons learned. Finally these 15 categories were used for clustering the lessons learned: • Preparedness/planning/training • Tactics/organisation/logistics • Medical treatment and Injuries • Equipment and supplies • Staffing • Command • Communication • Zoning and safety scene • Triage • Patient flow and distribution • Team spirit • Role Understanding • Cooperation and multidisciplinary approach • Psychiatric support • Record keeping • Preparedness/planning/training • Tactics/organisation/logistics • Medical treatment and Injuries • Equipment and supplies • Staffing • Command • Communication • Zoning and safety scene • Triage • Patient flow and distribution • Team spirit • Role Understanding • Cooperation and multidisciplinary approach • Psychiatric support • Record keeping • Cooperation and multidisciplinary approach Fig. 1 Process to identify the articles included in the systematic review • Psychiatric support • Record keeping After defining the categories, the lessons learned were assigned to them. Where applicable, the lessons learned were divided into “pre-incident”, “during incident” and “post-incident” within the different categories. most frequently, while the category “team spirit” was ranked last in this list. Data abstraction To obtain a graphical overview over the entire study period, the frequency with which the categories were men- tioned per year were colour-coded and presented in a matrix (Fig. 3). Results A summary of all lessons learned assigned to the 15 cat- egories can be found in Table 3. The extended PubMed Search yielded 1635 articles out of which 1434 articles were excluded on title selection only. The abstracts of the remaining 201 articles were evaluated and finally 68 articles were included in the analysis (Fig. 1). To evaluate the quality of the included studies, the PRISMA evaluation was used and all articles were checked and reported against its checklist and then rated as either high quality (HQ), acceptable quality (AQ) or low quality (LQ) paper (Table 1) [12]. The extended PubMed Search yielded 1635 articles out of which 1434 articles were excluded on title selection only. The abstracts of the remaining 201 articles were evaluated and finally 68 articles were included in the analysis (Fig. 1). Discussion i To evaluate the quality of the included studies, the PRISMA evaluation was used and all articles were checked and reported against its checklist and then rated as either high quality (HQ), acceptable quality (AQ) or low quality (LQ) paper (Table 1) [12]. This systematic review is the first of its kind to review the vast amount of literature dealing with lessons learned from terror attacks. It thus contributes to a better understanding of the consequences of terror attacks since 2001. It also brings order to the multitude of defined lessons learned and allows for an overview of all the important findings. A total of 616 lessons learned were assigned to the 15 categories. If a lesson matched more than one category, it was assigned to all matching categories. Therefore, multiple entries occur in some cases. Table 2 shows the distribution of categories across all included articles, while Fig. 2 shows the number of articles in which each category appears. In this figure, the publications are assigned to the respective categories. This provides an overview of the number of arti- cles dealing with each category. An overview of the distribu- tion over time is later given in Fig. 3. Lessons learned within the category “tactics/organisation/logistics” were mentioned i Our data has shown that, despite the difference in attacks, countries, social and political systems and casualties involved, many of the lessons learned and issues identified are similar. Important to note was the fact that time of article release did not relate to content. Many articles written after the London attacks in 2005 formulated similar if not the same lessons learned as articles written in 2017 about Utoya [36, 52]. This is a major point of concern as it indicates, that despite the knowledge about the issues and the existence of already developed, excellent concepts [56, 79, 80], their 1 3 1 3 2616 N. Schorscher et al. Authors Year Incident site Study type PRISMA Roccaforte et al. [15] 2001 USA 9/11 Retrospective AQ Martinez et al.[16] 2001 USA 9/11 Eye Witness AQ Cook et al. [17] 2001 USA 9/11 Eye Witness AQ Tamber et al. [18] 2001 USA 9/11 Expert Opinion AQ Simon et al. [19] 2001 USA 9/11 Review/Report AQ Mattox et al. [20] 2001 USA 9/11 Review/Report AQ Shapira et al. [21] 2002 Israel General Review HQ Frykberg et al. [22] 2002 Multiple Review/Report HQ Garcia-Castrillo et al. Discussion [64] 2016 Utoya, Norway Review/Report AQ Table 1 Overview of all included articles with PRISMA evaluation 1 3 Lessons learned from terror attacks: thematic priorities and development since 2001—results… Lessons learned from terror attacks: thematic priorities and development since 2001—results… 2617 p p Table 1 (continued) Authors Year Incident site Study type PRISMA Raid et al. [65] 2016 Paris, France Personal Review AQ Philippe et al. [8] 2016 Paris, France Government Review HQ Traumabase et al. [66] 2016 Paris, France Personal Review HQ Gregory et al. [67] 2016 Paris, France Review/Report AQ Ghanchi et al. [68] 2016 Paris, France Review/Report AQ Khorram-Manesh et al. [69] 2016 Multiple Review/Report HQ Goralnick et al. [10] 2017 Paris/Boston Expert Opinion AQ Lesaffre et al. [70] 2017 Paris, France Review/Report AQ Wurmb et al. [71] 2018 Würzburg, Germany Lessons Learned HQ Brandrud et al. [72] 2017 Utoya, Norway Review/Report HQ Carli et al. [9] 2017 Paris/Nice, France Review/Report HQ Borel et al. [73] 2017 Paris, France Review/Report AQ Bobko et al. [74] 2018 San Bernadino, USA Review/Report AQ Chauhan et al. [75] 2018 Multiple Review/Report HQ Hunt et al. [76] 2018 London/Manchester, UK Review/Report HQ Hunt et al. [77] 2018 London/Manchester, UK Review/Report HQ Hunt et al. [78] 2018 London/Manchester, UK Review/Report HQ HQ high quality, AQ acceptable quality, LQ low quality, USA United States of America, UK United King- dom HQ high quality, AQ acceptable quality, LQ low quality, USA United States of America, UK United King- dom RE: Rapid extraction to safety A: Assessment by medical providers T: Transport to definitive care. successful implementation and continuous improvements seem to be lacking. RE: Rapid extraction to safety A: Assessment by medical providersi One of these well-developed concepts, the Tactical Com- bat Casualty Care (TCCC), began as a special operations medical research programme in 1996 and is now an integral part of the US Army's trauma care [79]. The Committee on TCCC, which was established in 2001, ensures that the TCCC guidelines are regularly updated [79]. Many of the lessons learned listed in our review are an integral part of these guidelines and are addressed with concrete options for action. For Example, the principles of Tactical Evacuation Care provide detailed instructions on the management of casualties under the special conditions of evacuation from a danger zone [81]. Discussion [23] 2003 Madrid, Spain Review/Report AQ Shamir et al. [24] 2004 Israel Review/Report HQ Einav et al. [25] 2004 Israel Guidelines HQ Almogy et al. [26] 2004 Israel Review/Report AQ Rodoplu et al. [27] 2004 Istanbul, Turkey Retrospective Study AQ Kluger et al. [28] 2004 Israel Review/Report AQ Gutierrez de Ceballos et al. [29] 2005 Madrid, Spain Retrospective Study AQ Kirschbaum et al. [30] 2005 USA 9/11 Lessons Learned HQ Aschkenazy-Steuer et al. [31] 2005 Israel Retrospective Study HQ Lockey et al. [32] 2005 London, UK Retrospective Study HQ Hughes et al. [33] 2006 London, UK Review/Report AQ Shapira et al. [34] 2006 Israel Review/Report AQ Aylwin et al. [35] 2006 London, UK Review/Report HQ Mohammed et al. [36] 2006 London, UK Review/Report AQ Bland et al. [37] 2006 London, UK Personal Review AQ Leiba et al. [38] 2006 Israel Review/Report HQ Singer et al. [39] 2007 Israel Review/Report HQ Schwartz et al. [40] 2007 Israel Review/Report AQ Gomez et al. [41] 2007 Madrid, Spain Review/Report AQ Bloch et al. [42] 2007 Israel Review/Report AQ Bloch et al. [43] 2007 Israel Review/Report AQ Barnes et al. [44] 2007 London, UK Government Evaluation HQ Carresi et al. [45] 2008 Madrid, Spain Review/Report HQ Raiter et al. [46] 2008 Israel Review/Report HQ Shirley et al. [47] 2008 London, UK Review/Report HQ Almgody et al. [48] 2008 Multiple Review/Report AQ Turegano-Fuentes et al. [49] 2008 Madrid, Spain Review/Report AQ Pinkert et al. [50] 2008 Israel Review/Report HQ Pryor et al. [51] 2009 USA 9/11 Review/Report HQ Lockey et al. [52] 2012 Utoya, Norway Review/Report AQ Sollid et al. [53] 2012 Utoya, Norway Review/Report AQ Gaarder et al. [54] 2012 Utoya, Norway Review/Report AQ No authors listed [55] 2013 Boston USA Review/Report AQ Jacobs et al. [56] 2013 USA General Review AQ Gates et al. [57] 2014 Boston, USA Review/Report AQ Wang et al. [58] 2014 Multiple General Review HQ Ashkenazi et al. [59] 2014 Israel Overall Review AQ Thompson et al. [60] 2014 Multiple Retrospective AQ Rimstad et al. [61] 2015 Oslo, Norway Retrospective AQ Goralnick et al. [62] 2015 Boston, USA Retrospective AQ Hirsch et al. [6] 2015 Paris, France Personal Review HQ Lee et al. [63] 2016 San Bernadino, USA Personal Review HQ Pedersen et al. Discussion Moreover, the lack of knowledge on how to deal with injuries caused by firearms or explosive devices, which was mentioned in many articles, could be remedied by a consistent integration of the TCCC guidelines into the training and drills of emergency service staff. T: Transport to definitive care. Consistent implementation of these points in training and practice would be an important step towards improving preparation for terror attacks. A good example of the successful implementation of an interprofessional concept is the 3 Echo concept (Enter, Evaluate, Evacuate) [80]. It was developed and introduced with the goal to optimize the management of mass shooting incidents. At the beginning of concept development stood the identification of deficits. Those deficits correspond to those that we found in the presented systematic review. The introduction of the concept in training and practice has led to successful management of a mass shooting event in Min- neapolis, Minnesota, USA in 2012 [80]. This outlines once again the importance of translating lessons learned into concrete concepts, to integrate them into the training and to practice them regularly in interprofessional drills. Just as the 3 Echo concept is based on interprofessional coop- eration, the Joint Emergency Services Interoperability Prin- ciples (JESIP) project is also based on this principle [82]. It is the standard in Great Britain for the interprofessional cooperation of emergency services in major emergencies or disasters. Through simple instructions and a clear concept, both the aspect of planning and preparation as well as the concrete management of operations are taken care of [82]. f Another concept that deals with the management of ter- rorist attacks and mass shootings is the Medical Disaster Preparedness Concept “THREAT”, which was published after the Hartford Consensus Conference in 2013 [56]. The authors defined seven deficits as lessons learned and recom- mended concrete measures to address them. These lessons were included in our review and were mentioned in one form or the other in many of the articles. The defined THREAT concept components were: T: Threat suppression In interpreting the lessons learned in this systematic review, the question arises whether they are specific to H: Haemorrhage control 1 3 3 2618 N. Schorscher et al. 1 3 Table 2 Distribution of the 15 clusters across all included articles Study Year 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Roccaforte et al. Discussion [15] 2001 x x x x x x Martinez et al.[16] 2001 x x x x x x x x Cook et al. [17] 2001 x x x x x x x Tamber et al.[18] 2001 x x x x x x Simon et al.[19] 2001 x x x x x x x Mattox et al. [20] 2001 x x x x x x Shapira et al. [21] 2002 x x x x x x x x x x x x x Frykberg et al. [22] 2002 x x x x x x x x x x x Garcia-Castrillo et al. [23] 2003 x x x x x Shamir et al.[24] 2004 x x x x x x x x x x Einav et al. [25] 2004 x x x x x x x Almogy et al. [26] 2004 x x x x x Rodoplu et al. [27] 2004 x x x x x x Kluger et al. [28] 2004 x x x x x Gutierrez de Ceballos et al. [29] 2005 x x x x x Kirschbaum et al. [30] 2005 x x x x x x x x x x x x Aschkenazy-Steuer et al. [31] 2005 x x x x x x x x x Lockey et al. [32] 2005 x x x x x x x x Hughes et al. [33] 2006 x x x x x x Shapira et al. [34] 2006 x x x x x Aylwin et al. [35] 2006 x x x x x x x x Mohammed et al. [36] 2006 x x x x x x x x Bland et al. [37] 2006 x x x x x x x Leiba et al. [38] 2006 x x x x x x x x Singer et al. [39] 2007 x x x x x x x x x x x x x Schwartz et al. [40] 2007 x x x x x Gomez et al. [41] 2007 x x x x x x x Bloch et al. [42] 2007 x x x Bloch et al. Discussion [43] 2007 x x x x x x Barnes et al.[44] 2007 x x x x x x x x x Carresi et al.[45] 2008 x x x x x x x x x Raiter et al.[46] 2008 x x x x x Shirley et al.[47] 2008 x x x x x x x Almgody et al. [48] 2008 x x x x x x x Turegano-Fuentes et al. [49] 2008 x x x x x x x Pinkert et al. [50] 2008 x x x x x x Lockey et al. [52] 2012 x x x x x x Sollid et al. [53] 2012 x x x x x x Gaarder et al. [54] 2012 x x x x x x x NN et al. [55] 2013 x x x x x x x Jacobs et al. [56] 2013 x x x x x x x x x Gates et al. [57] 2014 x x x x x x Wang et al. [58] 2014 x x x x x Ashkenazi et al. [59] 2014 x x x Thompson et al. [60] 2014 x x x x x x Rimstad et al. [61] 2015 x x x x Goralnick et al. [62] 2015 x x x x x x x Hirsch et al. [6] 2015 x x x x x x x x x Lee et al. [63] 2016 x x x x x x x x x x Pedersen et al. [64] 2016 x x x x x x Table 2 Distribution of the 15 clusters across all included articles 1 Lessons learned from terror attacks: thematic priorities and development since 2001—results… Lessons learned from terror attacks: thematic priorities and development since 2001—results… 2619 Table 2 (continued) Study Year 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Raid et al. [65] 2016 x x x x x x x x x Philippe et al. [8] 2016 x x x x x x x x Traumabase et al. [66] 2016 x x x x Gregory et al. [67] 2016 x x x x x Ghanchi et al. [68] 2016 x x x x x x x Khorram-Manesh et al. [69] 2016 x x x x x x x x x x x Goralnick et al. [10] 2017 x x x x x x x x Lesaffre et al. Discussion [70] 2017 x x x x x x x Brandrud et al. [72] 2017 x x x x x x x x x x x Carli et al. [9] 2017 x x x x x x x x x x x x Borel et al. [73] 2017 x x x x x x x x x x x Wurmb et al. [71] 2018 x x x x x x x x Bobko et al. [74] 2018 x x x x x x x x x Chauhan al. [75] 2018 x x x x x x x x x Hunt et al. [76] 2018 x x x x x x x x x x x x x Hunt et al. [77] 2018 x x x x x x x x x x x Hunt et al. [78] 2018 x x x x x x 1—Tactics/organization/logistics, 2—Communication, 3—Preparedness/planning/training 4—Triage, 5—Patient flow and distribution, 6— Cooperation/multi-disciplinary approach, 7—Command, 8—Staffing, 9—Medical treatment and type of injuries, 10—Equipment/supplies, 11— Zoning/scene safety, 12—Psych support, 13—Record keeping, 14—Role understanding, 15—Team spirit 1—Tactics/organization/logistics, 2—Communication, 3—Preparedness/planning/training 4—Triage, 5—Patient flow and distribution, 6— Cooperation/multi-disciplinary approach, 7—Command, 8—Staffing, 9—Medical treatment and type of injuries, 10—Equipment/supplies, 11— Zoning/scene safety, 12—Psych support, 13—Record keeping, 14—Role understanding, 15—Team spirit 1—Tactics/organization/logistics, 2—Communication, 3—Preparedness/planning/training 4—Triage, 5—Patient flow and distribution, 6— Cooperation/multi-disciplinary approach, 7—Command, 8—Staffing, 9—Medical treatment and type of injuries, 10—Equipment/supplies, 11— Zoning/scene safety, 12—Psych support, 13—Record keeping, 14—Role understanding, 15—Team spirit terrorist attacks. Our review deals exclusively with lessons learned from terrorist attacks. Other publications, however, non-terrorist mass shootings and disasters. Turner et reported the results of a systematic review of the lite Fig. 2 Number of articles mentioning each of the 15 categories Fig. 2 Number of articles mentioning each of the 15 categories non-terrorist mass shootings and disasters. Turner et al. reported the results of a systematic review of the litera- ture on prehospital management of mass casualty civilian terrorist attacks. Our review deals exclusively with lessons learned from terrorist attacks. Other publications, however, systematically addressed the management of terrorist and 1 3 2620 N. Schorscher et al. Fig. 3 Categories of lessons learned from terror attacks—development since 2001 Fig. 3 Categories of lessons learned from terror attacks—development since 2001 shootings [83]. 3 Discussion The authors identified the need for integra- tion of tactical emergency medical services, improved cross- service education on effective haemorrhage control, the need for early and effective triage by senior clinicians and the need for regular mass casualty incident simulations [83] as key topics. Those correspond congruently with the lessons learned from terrorist attacks that were found and presented in this systematic review.i the included articles dealt with terrorist attacks, the lessons learned are again very comparable to the results of this sys- tematic review. Challen et al. published the results from a scoping review in 2012 [11]. The authors stated that “although a large body of literature exists, its validity and generalisability is unclear” [11]. They concluded that the type and struc- ture of evidence that would be of most value for emergency planners and policymakers has yet to be identified. If try- ing to summarise the development since that statement it can be assumed that on one hand sound concepts have been developed and implemented. On the other hand however, the lessons learned in recent terror attacks still emphasize similar issues as compared to those from the beginning of the analysis in 2001, showing that there is still work to be done. It should be mentioned at this point, that there was a federal conducted evaluation process in Germany after the European terror attacks in 2015/2016. The lessons learned Hugelius et al. performed a review study and identified five challenges when managing mass casualty incidents or disaster situations [84]. These were “to identify the situa- tion and deal with uncertainty”, “to balance the mismatch between contingency plan and reality”, “to establish func- tional crisis organisation”, “to adapt the medical response to actual and overall situation” and “to ensure a resilient response” [84]. The authors included 20 articles, of which 5 articles dealt with terror and mass shooting (including the terror attacks in Paris and Utoya). Discussion Although only 25% of 3 Lessons learned from terror attacks: thematic priorities and development since 2001—results… 2621 p p Table 3 lessons learned assigned to the 15 overwhelming categories Lessons learned Tactics/organization/logistics Pre-incident 1 Offer a detailed manual for potential terror attacks 2 Need for having a solid disaster plan for each hospital 3 Have a national standard for major incidents and a preparedness con- cept/disaster response plan 4 Adequate trauma centre concepts on national level 5 Use trauma guidelines 6 Conduct updated disaster plans/drills 7 Active pre-planned protocols—pre hospital protocol + hospital protocol 8 All hospitals should be included in contingency planning 9 Do not base disaster plan on average surge rates 10 Standardisation in hospital incident planning 11 Have an emergency plan for preparedness 12 Use standard Protocols but keep flexibility 13 Establishment of various anti-terror contingency plans (hijack/bombing/ shooting) 14 Mini disasters as basis for escalation (flu season) 15 Crisis management based on knowledge and data collection During the incident 16 Activate contingency/emergency plans soon 17 Organisation of trauma teams that stay with a patient 18 Cancellation of all elective surgery/discharge of all non-urgent patients 19 Establish a public information centre close to hospital 20 Alert all hospitals 21 Prehospital and hospital coordination + communication is necessary 22 Crowd control is important 23 Maximise surge capacity 24 Distance to hospital site is major distribution factor 25 Evacuation of the less critically ill to further away hospitals 26 Importance of controlled access to hospitals 27 Avoid main gate syndrome—overwhelmed resources at the closest hospital 28 Avoid overcrowding in the ER 29 Activation of white plan—all hospitals/all staff/empty beds → no short- age 30 Recruit help from outside early on 31 Do not forget flexibility 32 Combination of civil defence and emergency medical services 33 Designated treatment area 34 Rapid scene clearance—highly organised und efficient 35 Flexibility across incident sites/hospitals 36 Vehicle coordination and rapid accumulation 37 Set principles rather than fixed protocols to allow for flexibility 38 Importance of quick evacuationfi Lessons learned Pre-incident 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 During the incident 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 Tactics/organization/logistics Offer a detailed manual for potential terror attacks Need for having a solid disaster plan for each hospital Have a national standard for major incidents and a preparedness con- cept/disaster response plan Adequate trauma centre concepts on national level Use trauma guidelines Conduct updated disaster plans/drills Active pre-planned protocols—pre hospital protocol + hospital protocol All hospitals should be included in contingency planning Do not base disaster plan on average surge rates Standardisation in hospital incident planning Have an emergency plan for preparedness Use standard Protocols but keep flexibility Establishment of various anti-terror contingency plans (hijack/bombing/ shooting) Mini disasters as basis for escalation (flu season) Crisis management based on knowledge and data collection Activate contingency/emergency plans soon Organisation of trauma teams that stay with a patient Cancellation of all elective surgery/discharge of all non-urgent patients Establish a public information centre close to hospital Alert all hospitals Prehospital and hospital coordination + communication is necessary Crowd control is important Maximise surge capacity Distance to hospital site is major distribution factor Evacuation of the less critically ill to further away hospitals Importance of controlled access to hospitals Avoid main gate syndrome—overwhelmed resources at the closest hospital Avoid overcrowding in the ER Activation of white plan—all hospitals/all staff/empty beds → no short- age Recruit help from outside early on Do not forget flexibility Combination of civil defence and emergency medical services Designated treatment area Rapid scene clearance—highly organised und efficient Flexibility across incident sites/hospitals Vehicle coordination and rapid accumulation Set principles rather than fixed protocols to allow for flexibility Importance of quick evacuation Ambulance stacking area to allow access and reduce traffic jam Important to declare major incident as soon as possible Manage uncertainties and scene Coordination of rescue—especially HEMS Rapid logistical response Divide emergency response into stages break into smaller parts 1 3 Offer a detailed manual for potential terror attacks Need for having a solid disaster plan for each hospital Have a national standard for major incidents and a preparedness con- cept/disaster response plan Adequate trauma centre concepts on national level Use trauma guidelines Conduct updated disaster plans/drills Active pre-planned protocols—pre hospital protocol + hospital protocol All hospitals should be included in contingency planning Do not base disaster plan on average surge rates Standardisation in hospital incident planning Have an emergency plan for preparedness Use standard Protocols but keep flexibility Establishment of various anti-terror contingency plans (hijack/bombing/ shooting) Mini disasters as basis for escalation (flu season) Crisis management based on knowledge and data collection Activate contingency/emergency plans soon Organisation of trauma teams that stay with a patient Cancellation of all elective surgery/discharge of all non-urgent patients Establish a public information centre close to hospital Alert all hospitals Prehospital and hospital coordination + communication is necessary Crowd control is important Maximise surge capacity Distance to hospital site is major distribution factor Evacuation of the less critically ill to further away hospitals Importance of controlled access to hospitals Avoid main gate syndrome—overwhelmed resources at the closest hospital Avoid overcrowding in the ER Activation of white plan—all hospitals/all staff/empty beds → no short- age Recruit help from outside early on Do not forget flexibility Combination of civil defence and emergency medical services Designated treatment area Rapid scene clearance—highly organised und efficient Flexibility across incident sites/hospitals Vehicle coordination and rapid accumulation Set principles rather than fixed protocols to allow for flexibility Importance of quick evacuation Ambulance stacking area to allow access and reduce traffic jam Important to declare major incident as soon as possible Manage uncertainties and scene Coordination of rescue—especially HEMS Rapid logistical response Divide emergency response into stages break into smaller parts 1 3 N. Discussion Schorscher et al. Discussion 2622 Table 3 (continued) Lessons learned Tactics/organization/logistics 45 Adaptation of decisions taken 46 Early decision by incidence command 47 No headquarter at frontline 48 Peri-incident intensive care manageme 49 Critical mortality is reduced by rapid ment 50 Use ICU staff for resuscitation and tria 51 Four step approach to terror attacks: a capabilities, analysis of gaps, develo 52 Experienced personnel should treat pa 53 Empty hospital immediately 54 Focus on increasing bed capacity espe 55 Constant update on resources and surg 56 Trauma leaders must be aware of bed 57 Combined activation of major inciden 58 Early activation of surge capacity 59 Crucial interaction/communication be ties 60 Fullback structures but flexibility and 61 Tactical management—get an overvie 62 Prehospital damage control—military 63 Regional resource mobilisation vital 64 Have a plan but use continuous reasse response strategy 65 Use METHANE to assess incident 66 Clear escalation plan 67 Coordination and collaboration should inter-regional, multiagency and mult 68 Improved forensic management 69 Logistic is important for operational s 70 Maintaining access to other emergenc 71 Gradual De-escalation – part of contin 72 Issue: recognition of situational aspec evolving risk 73 Cockpit view due to HEMS—helpful 74 Limited mobilisation at remote hospit 75 Incident commander appoints: liaison personnel officer; logistics officer; da officer; patient/family information of 76 “ABCD response”: assess incident siz sonnel, perform initial casualty care 77 Authority and command structure—tw tional vs medical management 78 Med Students used as runners 79 Tape fixed with name/specialty Table 3 (continued) Lessons learned 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 Tactics/organization/logistics g g Adaptation of decisions taken Early decision by incidence commander needed No headquarter at frontline Peri-incident intensive care management—forward deployment Critical mortality is reduced by rapid advanced major incident manage- ment Use ICU staff for resuscitation and triage Four step approach to terror attacks: analysis of scenario; description of capabilities, analysis of gaps, development of operational framework Experienced personnel should treat patient and not take on organisation Empty hospital immediately Focus on increasing bed capacity especially ICU beds Constant update on resources and surge limitation of all hospitals Trauma leaders must be aware of bed capacities Combined activation of major incident plans (all EMS services) Early activation of surge capacity Crucial interaction/communication between hospital/police/municipali- ties Fullback structures but flexibility and improvisation important Tactical management—get an overview and do not get stuck in details Prehospital damage control—military concepts in civilian setting Regional resource mobilisation vital Have a plan but use continuous reassessment and modification of response strategy Use METHANE to assess incident Clear escalation plan Coordination and collaboration should be planned and practised at intra/ inter-regional, multiagency and multiprofessional levels Improved forensic management Logistic is important for operational strategic roles Maintaining access to other emergencies MI/stroke, etc. Discussion Gradual De-escalation – part of contingency plan Issue: recognition of situational aspect and severity + complexity— evolving risk Cockpit view due to HEMS—helpful in big sweep of casualties Limited mobilisation at remote hospitals Incident commander appoints: liaison officer; public information officer; personnel officer; logistics officer; data officer; medical command officer; patient/family information officer “ABCD response”: assess incident size and severity, alert backup per- sonnel, perform initial casualty care, and provide definitive treatment Authority and command structure—two command posts—administra- tional vs medical management Med Students used as runners Tape fixed with name/specialty Delays should be expected Disruption in transport—lengthens rescue effort Guidelines on biochemical warfare Structural organisation important Clear and well-structured coordination Management of uninjured survivors and relatives—good communica- tion 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 Use METHANE to assess incident Maintaining access to other emergencies MI/stroke, etc. Discussion Limited mobilisation at remote hospitalsfi Incident commander appoints: liaison officer; public information officer; personnel officer; logistics officer; data officer; medical command officer; patient/family information officer 1 2623 Lessons learned from terror attacks: thematic priorities and development since 2001—results… Table 3 (continued) Lessons learned Tactics/organization/logistics 86 Development of operational framework 87 Assessment and re-evaluation of disaster plans 88 ED as epicentre 89 Most senior emergency physician directs traffic/surgeons overseas area—triage not by most senior personnel 90 Volunteer surges difficult to manage but can be helpful 91 Need to increase morgue facilities 92 Improved alert system 93 Clear communication, organization and decision making skills 94 Robust and simple organisation and command Post-incident 95 Clinical representation at strategic level to facilitate cooperation between networks/regions 96 Support from neighbouring regions during terror 97 Develop a network of capacities and capabilities which is constantly updated 98 Gaps in provision of rehab services—acute phase vs long term phase 99 Access to legal and financial support for victims 100 Importance of evaluation and improvement of emergency plans 101 Analysis based on past incidences 102 Early debriefing 103 Quickest possible return to normality 104 Quick return to normality—ongoing care for normal patients Lessons learned Communication Pre-incident 1 Terror awareness—train the public—communicate 2 Establish Improved alert system 3 Public engagement and empowerment—communication and teaching 4 Clear communication, organization and decision making skills During the incident 6 Delays in communication should be expected 7 Radio Equipment vital as often all other communication lines lost 8 Importance of reliable information 9 Effective intra-hospital communication 10 Constant update on resources and limitation of all hospitals 11 Better communication between disaster agencies 12 Importance of communication between different rescue teams 13 Identification vests help communication and command structures—clear roles 14 Intra and interhospital communication is important 15 Importance of public communication centre 16 Communication between disaster scene/EMS and hospital is often big problem 17 Use of protected phone lines and walkie-talkies 18 Early information/communication from site to assess severity 19 Early on radio/bleep system—later use of mobile phones possible 20 Clear, well-structured communication and coordination 21 Increase supplies through early communication with vendors 22 Bleeps and cable phones as cell service is often unreliable 23 Multiple scenes create difficult command and communication problems 2623 Lessons learned from terror attacks: thematic priorities and development since 2001—results… Table 3 (continued) Lessons learned Tactics/organization/logistics 86 Development of operational framework 87 Assessment and re-evaluation of disaster plans 88 ED as epicentre 89 Most senior emergency physician directs traffic/surgeons overseas area—triage not by most senior personnel 90 Volunteer surges difficult to manage but can be helpful 91 Need to increase morgue facilities 92 Improved alert system 93 Clear communication, organization and decision making skills 94 Robust and simple organisation and command Post-incident 95 Clinical representation at strategic level to facilitate cooperation between networks/regions 96 Support from neighbouring regions during terror 97 Develop a network of capacities and capabilities which is constantly updated 98 Gaps in provision of rehab services—acute phase vs long term phase 99 Access to legal and financial support for victims 100 Importance of evaluation and improvement of emergency plans 101 Analysis based on past incidences 102 Early debriefing 103 Quickest possible return to normality 104 Quick return to normality—ongoing care for normal patients Lessons learned Communication Pre-incident 1 Terror awareness—train the public—communicate 2 Establish Improved alert system 3 Public engagement and empowerment—communication and teaching 4 Clear communication, organization and decision making skills During the incident 6 Delays in communication should be expected 7 Radio Equipment vital as often all other communication lines lost 8 Importance of reliable information 9 Effective intra-hospital communication 10 Constant update on resources and limitation of all hospitals 11 Better communication between disaster agencies 12 Importance of communication between different rescue teams 13 Identification vests help communication and command structures—clear roles 14 Intra and interhospital communication is important 15 Importance of public communication centre 16 Communication between disaster scene/EMS and hospital is often big problem 17 Use of protected phone lines and walkie-talkies 18 Early information/communication from site to assess severity 19 Early on radio/bleep system—later use of mobile phones possible 20 Clear, well-structured communication and coordination 21 Increase supplies through early communication with vendors 22 Bleeps and cable phones as cell service is often unreliable 23 Multiple scenes create difficult command and communication problems 2623 Lessons learned from terror attacks: thematic priorities and development since 2001—results… Table 3 (continued) Lessons learned Tactics/organization/logistics 86 Development of operational framework 87 Assessment and re-evaluation of disaster plans 88 ED as epicentre 89 Most senior emergency physician directs traffic/surgeons overseas area—triage not by most senior personnel 90 Volunteer surges difficult to manage but can be helpful 91 Need to increase morgue facilities 92 Improved alert system 93 Clear communication, organization and decision making skills 94 Robust and simple organisation and command Post-incident 95 Clinical representation at strategic level to facilitate cooperation between networks/regions 96 Support from neighbouring regions during terror 97 Develop a network of capacities and capabilities which is constantly updated 98 Gaps in provision of rehab services—acute phase vs long term phase 99 Access to legal and financial support for victims 100 Importance of evaluation and improvement of emergency plans 101 Analysis based on past incidences 102 Early debriefing 103 Quickest possible return to normality 104 Quick return to normality—ongoing care for normal patients Lessons learned Communication Pre-incident 1 Terror awareness—train the public—communicate 2 Establish Improved alert system 3 Public engagement and empowerment—communication and teaching 4 Clear communication, organization and decision making skills Lessons learned from terror attacks: thematic priorities and development since 2001—results… 2623 1 N. Discussion Schorscher et al. Discussion 2624 Table 3 (continued) Lessons learned Communication 24 Communication between rescue services is vitally important 25 Do not solely rely on mobile phones—danger of collapse 26 Establish a public information centre close to hospital 27 Use robust communication methods 28 Communication lines often fail—be prepared 29 Management of uninjured survivors and relatives—good com tion 30 Concentrate initially on relaying as much information as poss 31 Important information: (1) the nature of the event (2) the estim number and severity of casualties; (3) the exact location of t (4) the primary routes of approach and evacuation; (5) estim of arrival at the nearest hospital 32 Use megaphones if adequate 33 Turn off all non-critical mobile cell phones during terror even ment implementation) 34 Communication centre for relatives 35 No media inside hospital—media centre set up 36 Importance of communication mechanisms during terror 37 Communication with public—use of media 38 Good telecommunication system—with backup options 39 Create database of victims/casualties 40 Importance of communication/coordination between incident hospitals 41 Importance of even distribution between hospitals—commun 42 Early press briefings to stop hysteria 43 Communication failure will always happen 44 Good care despite communication failure—hence senior well personnel 45 Communication-use of standardised operational terms 46 Good in-hospital communication between specialties 47 Decision making without all information—lack of communic unavoidable 48 Public Reassurance through good communication 49 Restricted internet access to avoid breakage 50 Communication with relatives 51 Better communication of patient information between prehosp hospital setting 52 Communication channel between police, EMS and hospitals 53 Public relations and communication 54 Readiness of hospitals—good communication and preparation 55 Mutual communication systems 56 Better Integration of operators of different rescue chains + com tion 57 Provide patient lists to police to ease communication/informa ering for relatives 58 Importance of patient hand over communication Table 3 (continued) Lessons learned 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 Communication Communication between rescue services is vitally important Do not solely rely on mobile phones—danger of collapse Establish a public information centre close to hospital Use robust communication methods Communication lines often fail—be prepared Management of uninjured survivors and relatives—good communica- tion Concentrate initially on relaying as much information as possible Important information: (1) the nature of the event (2) the estimated number and severity of casualties; (3) the exact location of the event; (4) the primary routes of approach and evacuation; (5) estimated time of arrival at the nearest hospital Use megaphones if adequate Turn off all non-critical mobile cell phones during terror event (govern ment implementation) Communication centre for relatives No media inside hospital—media centre set up Importance of communication mechanisms during terror Communication with public—use of media Good telecommunication system—with backup options Create database of victims/casualties Importance of communication/coordination between incident site and hospitals Importance of even distribution between hospitals—communication Early press briefings to stop hysteria Communication failure will always happen Good care despite communication failure—hence senior well-trained personnel Communication-use of standardised operational terms Good in-hospital communication between specialties Decision making without all information—lack of communication unavoidable Public Reassurance through good communication Restricted internet access to avoid breakage Communication with relatives Better communication of patient information between prehospital and hospital setting Communication channel between police, EMS and hospitals Public relations and communication Readiness of hospitals—good communication and preparation Mutual communication systems Better Integration of operators of different rescue chains + communica- tion Provide patient lists to police to ease communication/information gath- ering for relatives Importance of patient hand over communication Effective communication—improve information sharing Sharing of corporate knowledge—communication of information Good communication and situational awareness—use liaison officers Media policy and communication—robust and well informed Consider radio control mechanisms Confidentiality when it comes to communication with media Importance of communication mechanisms during terror Communication with public—use of media 1 2625 Lessons learned from terror attacks: thematic priorities and development since 2001—results… Table 3 (continued) Lessons learned Communication 65 Security and privacy issues when it comes to media communication 66 Quick and clear communication with relatives—to avoid information gathering via social media Lessons learned Preparedness/planning/training Pre-incident 1 Practise/drill—important! Discussion 2 Terror awareness—train the public 3 Trained prehospital personnel is a crucial factor 4 Update disaster plans—train them 5 Different sort of drills to prepare (manager drills/full scale drills) 6 Training is most important 7 Have and follow a pre-existing plan—based on experience 8 Thorough good quality preparation 9 Good prehospital care systems improve survival 10 Training of triage to reduce over and under triage 11 Debrief early and in a structured way 12 Preparation for incidents and injury types 13 Be prepared: have 1–3 months supply of surgical disposables 14 All hospitals should be included in contingency planning 15 All hospitals should be prepared to act as evacuation hospital—drills and training 16 Importance of damage control concepts—training 17 Cancellation of all elective surgical procedure 18 Emptying of ICU and wards 19 Importance of planning, coordination, training, financial support and well equipped medical services 20 Clear out hospital during latent phase 21 Have a major incident plan—have it rehearsed 22 Analysis based on past incidences 23 Analysis of gaps between scenario and response needed 24 Pre-event preparedness crucial—extensive planning improve outcome 25 Train core of nurses in emergency medicine skills 26 Have an emergency plan even if not a level one trauma centre 27 Rehearsal of emergency plan 28 Every hospital should be prepared for a major incident with terrorist background -solid emergency plans in situ 29 Importance of thorough analysis and short fallings 30 Good mix between planning and improvisation 31 A major incident plan is necessary—on a local as well as regional level 32 Meticulous planning 33 Extensive education 34 Regular review of the contingency plans 35 Emergency and disaster preparation and planning is crucial 36 All hospitals should be ATLS trained and have major incident drills Lessons learned from terror attacks: thematic priorities and development since 2001—results… 2625 Practise/drill—important! Discussion Table 3 (continued) Lessons learned Preparedness/planning/training 43 Teaching/training/education—best preparation 44 Disaster training best preparation for reality—systematic multidiscipli- nary training/drills 45 Train for new pattern of injuries 46 Readiness of hospitals—good communication and preparation 47 Public engagement and empowerment—communication and teaching 48 Staff training in combat medicine—cooperation with the military 49 Greater investment, integration, standardisation of disaster medicine 51 Multidisciplinary training—including police/fire service 52 Monthly multidisciplinary trauma training 53 Train the public/police in first aid/bleeding control 54 Importance of evaluation and improvement of emergency plans 55 Emergency preparedness based on planning/training/learning 56 Competence through continuous planning/training/drills 57 Cooperation: teaching of medical staff by military 58 Teaching of trauma management to med students 59 Therapy of paediatric cases—training is essential 60 Anticipation and planning—Plan Blanc obligatory 61 Awareness of tactical threat—idea of hazardous area response team 62 Training in trauma management 63 Planning and training—the value of organised learning 64 National process for debriefing and lessons learned 65 Regional standards for training Lessons learned Command During the incident 1 Strict command and control structures with designated hierarchy 2 Establish incident command system/centre—this is important 3 Early command and control structure—be prepared to rebuild 4 Avoid improvisation in command structure 5 Identification vests help communication and command structures—clear roles 6 Most senior medical officer = commander 7 Prompt and vigorous leadership 8 Civil defence coordinates and has overall command—clear structure 9 Importance of chain of command 10 Command structures—medical director vs administrative director 11 Incident commander appoints: liaison officer; public information officer; personnel officer; logistics officer; data officer; medical command officer; patient/family information officer 12 Chain of command: most senior official from all important specialties plus hospital admin 13 Multiple scenes create difficult command and communication problems 14 Have experienced decision maker 15 Command and control—regular trauma meetings 16 Importance of EMS command centre 17 Accept chaos phase—command structures will follow 18 Importance of local command structures—most senior official = com- mander in chief 19 Communication/cooperation between managers of different EMS 20 Work within established command and control structures N. Schorscher et al. Discussion Terror awareness—train the public Trained prehospital personnel is a crucial factor Update disaster plans—train them Different sort of drills to prepare (manager drills/full scale drills) Training is most important Have and follow a pre-existing plan—based on experience Thorough good quality preparation Good prehospital care systems improve survival Training of triage to reduce over and under triage Debrief early and in a structured way Preparation for incidents and injury types Be prepared: have 1–3 months supply of surgical disposables All hospitals should be included in contingency planning All hospitals should be prepared to act as evacuation hospital—drills and training Importance of damage control concepts—training Cancellation of all elective surgical procedure Emptying of ICU and wards Importance of planning, coordination, training, financial support and well equipped medical services Clear out hospital during latent phase Have a major incident plan—have it rehearsed Analysis based on past incidences Analysis of gaps between scenario and response needed Pre-event preparedness crucial—extensive planning improve outcome Train core of nurses in emergency medicine skills Have an emergency plan even if not a level one trauma centre Rehearsal of emergency plan Every hospital should be prepared for a major incident with terrorist background -solid emergency plans in situ Importance of thorough analysis and short fallings Good mix between planning and improvisation A major incident plan is necessary—on a local as well as regional level Meticulous planning Extensive education Regular review of the contingency plans Emergency and disaster preparation and planning is crucial All hospitals should be ATLS trained and have major incident drills Regional major incident plan to help allocate resources Have and activate contingency plans soon Be prepared for uncertainty and unsafe environment Having experience best preparation for next incident Training saves lives Drills based on past experiences 1 3 2626 N. Schorscher et al. Discussion 2626 Table 3 (continued) Lessons learned Preparedness/planning/training 43 Teaching/training/education—best preparation 44 Disaster training best preparation for reality—systematic multidiscipli- nary training/drills 45 Train for new pattern of injuries 46 Readiness of hospitals—good communication and preparation 47 Public engagement and empowerment—communication and teaching 48 Staff training in combat medicine—cooperation with the military 49 Greater investment, integration, standardisation of disaster medicine 51 Multidisciplinary training—including police/fire service 52 Monthly multidisciplinary trauma training 53 Train the public/police in first aid/bleeding control 54 Importance of evaluation and improvement of emergency plans 55 Emergency preparedness based on planning/training/learning 56 Competence through continuous planning/training/drills 57 Cooperation: teaching of medical staff by military 58 Teaching of trauma management to med students 59 Therapy of paediatric cases—training is essential 60 Anticipation and planning—Plan Blanc obligatory 61 Awareness of tactical threat—idea of hazardous area response team 62 Training in trauma management 63 Planning and training—the value of organised learning 64 National process for debriefing and lessons learned 65 Regional standards for training Lessons learned Command During the incident 1 Strict command and control structures with designated hierarchy 2 Establish incident command system/centre—this is important 3 Early command and control structure—be prepared to rebuild 4 Avoid improvisation in command structure 5 Identification vests help communication and command structures—clear roles 6 Most senior medical officer = commander 7 Prompt and vigorous leadership 8 Civil defence coordinates and has overall command—clear structure 9 Importance of chain of command 10 Command structures—medical director vs administrative director 11 Incident commander appoints: liaison officer; public information officer; personnel officer; logistics officer; data officer; medical command officer; patient/family information officer 12 Chain of command: most senior official from all important specialties plus hospital admin 13 Multiple scenes create difficult command and communication problems 14 Have experienced decision maker 15 Command and control—regular trauma meetings 16 Importance of EMS command centre 17 Accept chaos phase—command structures will follow 18 Importance of local command structures—most senior official = com- mander in chief 19 Communication/cooperation between managers of different EMS 20 Work within established command and control structures Table 3 (continued) Lessons learned Preparedness/planning/training 43 Teaching/training/education—best preparation 44 Disaster training best preparation for reality—systematic multidiscipli- nary training/drills 45 Train for new pattern of injuries 46 Readiness of hospitals—good communication and preparation 47 Public engagement and empowerment—communication and teaching 48 Staff training in combat medicine—cooperation with the military 49 Greater investment, integration, standardisation of disaster medicine 51 Multidisciplinary training—including police/fire service 52 Monthly multidisciplinary trauma training 53 Train the public/police in first aid/bleeding control 54 Importance of evaluation and improvement of emergency plans 55 Emergency preparedness based on planning/training/learning 56 Competence through continuous planning/training/drills 57 Cooperation: teaching of medical staff by military 58 Teaching of trauma management to med students 59 Therapy of paediatric cases—training is essential 60 Anticipation and planning—Plan Blanc obligatory 61 Awareness of tactical threat—idea of hazardous area response team 62 Training in trauma management 63 Planning and training—the value of organised learning 64 National process for debriefing and lessons learned 65 Regional standards for training Lessons learned Command During the incident 1 S i d d l i h d i d hi h Preparedness/planning/training Strict command and control structures with designated hierarchy Establish incident command system/centre—this is important Early command and control structure—be prepared to rebuild Avoid improvisation in command structure Identification vests help communication and command structures—clear roles Most senior medical officer = commander Prompt and vigorous leadership Civil defence coordinates and has overall command—clear structure Importance of chain of command Command structures—medical director vs administrative director Incident commander appoints: liaison officer; public information officer; personnel officer; logistics officer; data officer; medical command officer; patient/family information officer Chain of command: most senior official from all important specialties plus hospital admin Multiple scenes create difficult command and communication problems Have experienced decision maker Command and control—regular trauma meetings Importance of EMS command centre Accept chaos phase—command structures will follow Importance of local command structures—most senior official = com- mander in chief Communication/cooperation between managers of different EMS Work within established command and control structures Accept chaos phase—command structures will follow Communication/cooperation between managers of different EMS f Work within established command and control structures 1 2627 Lessons learned from terror attacks: thematic priorities and development since 2001—results… Table 3 (continued) Lessons learned Command 21 Clear distinction between command/control and casualty treatment 22 Lead by senior clinicians 23 Effective decision making—command is crucial 24 Command structures need to be robust 25 EMS command structures are vital 26 Dual command—ambulance/tactical commander vs medical com- mander 27 Command and control vs collaboration—both important 28 Flexible leadership 29 Leadership through ER physicians 30 Central Command—Health emergencies crisis management centre 31 Central command in hospital—director of medical operations 32 Good crisis management/command important 33 Multidisciplinary management 34 Clear communication, organization and decision making skills vital 35 Robust and simple organisation and command 36 Crisis management based on knowledge and data collection 37 Solid command structures and leadership based on experience and knowledge 38 Tactical management—get an overview and do not get stuck in details 39 Leadership/coordination through experienced healthcare professionals 40 Tactical command post in safe zone Lessons learned Triage Pre-incident 1 Establish national triage guidelines 2 Improve triage skills 3 Reproducible triage standards 4 Triage according to three ECHO—coloured cards 5 Casualty disposition framework with an effective enhanced triage process During the incident 6 Priority is quick triage, evacuation and transport to hospital 7 Establish casualty collection points/triage simple and early 8 Multiple triage areas—staff with freelancers 9 Coloured tags for triage 10 Use START system—simple triage rapid treatment 11 Doctors not deployed in red zone -triage in safe zone 12 Triage by most senior personnel 13 In-hospital triage according to ATLS 14 Systematic planning for triage, stabilisation and evacuation to hospital through chain of treatment stations 15 Triage at a distant site to disaster 16 Importance of triage—good triager—absolute authority 17 Deploy small medical teams for 2nd triage 18 Senior general surgeon triages at hospital entrance 19 Triage on arrival at hospital entrance as prehospital triage not necessar- ily reliable 20 Rapid primary triage—evacuation of the critical ill to nearest hospital (evacuation hospital) for stabilisation 21 Beware of undertriage essons learned from terror attacks: thematic priorities and development since 2001—results… Lessons learned from terror attacks: thematic priorities and development since 2001—results… 2627 21 1 3 N. Discussion Schorscher et al. Discussion 25 Primary evacuation of mildly injured patients to distant hospitals 26 Treat patient in level 2 trauma centres and only transfer if necessary to level 1 trauma centres 27 Divert non urgent patients to hospitals further away from incident site 28 Survivor reception centres to alleviate hospitals 29 Primary and balanced distribution between hospitals 30 Timely evacuation out of unsafe zone 31 Overload of patients at close by hospitals is huge problem 32 Fast track route for minor injuries 33 Patient flow—evacuation to cold zones 34 Directed quick patient flow to relieve triage area 35 Secondary patient flow according to capacity and specialty 36 Relocation of current patients 37 Cooperation between hospitals and trauma centres—recognise your limits and transfer 38 Tourniquet use und quick transfer to definite care 39 Track patients through hospital is a difficult task 40 Casualty clearing station—part of patient flow 41 Casualty disposition framework with an effective enhanced triage process 42 Safe transfer and handover of existing patients Lessons learned Cooperation and multidisciplinary approach Pre-incident 1 Common goal is an important benefit 2 Cross organisational planning important 3 Communication channel between police, EMS and hospitals 4 Staff training in combat medicine—cooperation with the military 5 Awareness of tactical threat—idea of hazardous area response team 6 Sharing of corporate knowledge—communication of information 7 Clinical representation at strategic level to facilitate cooperation between networks/ regions 8 Simultaneous search/rescue/treatment During the incident 9 Better communication between disaster agencies is important 10 Importance of communication between different rescue teams 11 Especially trauma patients need teamwork and good cooperation (surgery/anaesthetic) 12 Cooperation of the entire medical system—prehospital as well as hospital 13 Increase supplies through early communication with vendors 14 Collaboration with police to deliver supplies 15 Police command centre within hospital 16 Chain of command: most senior official from all important specialties plus hospital admin 17 Communication between rescue services vitally important 18 Good teamwork is crucial 19 Triple: anaesthetist, trauma surgeon abdominal surgeon lead assessment and allocation to definite care 20 Multidisciplinary meetings 21 Most senior emergency physician directs traffic/surgeons overseas area—triage not by most senior personnel 22 Flexibility of services important—interaction/cooperation important 23 Possibility for emergency services to cooperate and communicate 24 Combined activation of major incident plans (all EMS services) 25 Joint field command post 26 Cooperation and communication between hospitals and all emergency services 27 Dual surgical command-triage N. Schorscher et al. Discussion 2628 Table 3 (continued) Lessons learned Triage 22 Importance of triage at incident site 23 Importance of retriage at hospital 24 Importance of triage concepts in general—avoid undertriage 25 Primary in-hospital survey through surgeons and anaesthetists 26 Diligence in triage 27 Large amount of over triage—no negative consequences/overtriage does not kill 28 Establishment of triage areas in hospital 29 Tertiary survey day after 30 Repeated effective triage maintains hospital surge capacity 31 Idea to establish triage hospital 32 Rapid primary survey and triage—delay of secondary survey 33 Most senior emergency physician directs traffic/surgeons overseas area—triage not by most senior personnel 34 Prehospital as well as hospital triage is vitally important 35 Importance of good primary triage 36 Frequent reassessment and triage 37 Quick triage—scoop and run—repeated triage at hospital 38 Quick effective good basic triage—reduction of overtriage 39 Improved triage through physician/paramedic teams 40 Enough equipment but mainly quick triage and transport 41 Deliberate overtriage 42 Directed quick patient flow to relieve triage area 43 Inadequate triage results in critically injured patients—retriage is vital 44 Outside triage area—not in hospital 45 Triage: absolute vs relative emergencies 46 Crisis teams to organise triage 47 Continuous retriage—similar triage system preclinical and in hospital 48 Triage outside hot zone—no therapy in hot zone if not trained 49 Most important triage point: able to walk vs not able to walk Lessons learned Staffing Pre-incident Triage Importance of triage at incident site Importance of retriage at hospital Importance of triage concepts in general—avoid undertriage Primary in-hospital survey through surgeons and anaesthetists Diligence in triage Large amount of over triage—no negative consequences/overtriage does not kill Establishment of triage areas in hospital Tertiary survey day after Repeated effective triage maintains hospital surge capacity Idea to establish triage hospital Rapid primary survey and triage—delay of secondary survey Most senior emergency physician directs traffic/surgeons overseas area—triage not by most senior personnel Prehospital as well as hospital triage is vitally important Importance of good primary triage Frequent reassessment and triage Quick triage—scoop and run—repeated triage at hospital Quick effective good basic triage—reduction of overtriage Improved triage through physician/paramedic teams Enough equipment but mainly quick triage and transport Deliberate overtriage Directed quick patient flow to relieve triage area Inadequate triage results in critically injured patients—retriage is vital Outside triage area—not in hospital Triage: absolute vs relative emergencies Crisis teams to organise triage Continuous retriage—similar triage system preclinical and in hospital Triage outside hot zone—no therapy in hot zone if not trained Most important triage point: able to walk vs not able to walk Staffing 22 Importance of triage at incident site 23 Importance of retriage at hospital 24 Importance of triage concepts in general—avoid undertriage 25 Primary in-hospital survey through surgeons and anaesthetists 26 Diligence in triage 27 Large amount of over triage—no negative consequences/overtriage does not kill 28 Establishment of triage areas in hospital 29 Tertiary survey day after 30 Repeated effective triage maintains hospital surge capacity 31 Idea to establish triage hospital 32 Rapid primary survey and triage—delay of secondary survey 33 Most senior emergency physician directs traffic/surgeons overseas area—triage not by most senior personnel 34 Prehospital as well as hospital triage is vitally important 35 Importance of good primary triage 36 Frequent reassessment and triage 37 Quick triage—scoop and run—repeated triage at hospital 38 Quick effective good basic triage—reduction of overtriage 39 Improved triage through physician/paramedic teams 40 Enough equipment but mainly quick triage and transport 41 Deliberate overtriage 42 Directed quick patient flow to relieve triage area 43 Inadequate triage results in critically injured patients—retriage is vital 44 Outside triage area—not in hospital 45 Triage: absolute vs relative emergencies 46 Crisis teams to organise triage 47 Continuous retriage—similar triage system preclinical and in hospital 48 Triage outside hot zone—no therapy in hot zone if not trained 49 Most important triage point: able to walk vs not able to walk Lessons learned Staffing Pre-incident 1 Deploy trained prehospital personnel 2 Staff imprints lessons from mini-disasters and use this experience 3 Establishment of human resource pools—especially with volunteers 4 Too few nurses—improve incentives 5 Description of relevant capabilities of the medical system 6 Staff training in combat medicine—cooperation with the military 7 Up-to-date list of available staffing important During the incident 8 Descale as soon as possible → rest time for staff 9 Staff Safety is a major concern 10 Freelancers are important but difficult to manage 11 Multiple triage areas—possible staffing with freelancers 12 Quick response—increase staffing as soon as possible 13 Maximal increase of staffing needed—most important factor 14 Forward deployment of anaesthetist—allows for continuity of care 15 Relieve staff after 8–12 h for breaks 16 Optimise utilisation of manpower and supplies Deploy trained prehospital personnel Staff imprints lessons from mini-disasters and use this experience Establishment of human resource pools—especially with volunteers Too few nurses—improve incentives Description of relevant capabilities of the medical system Staff training in combat medicine—cooperation with the military Up-to-date list of available staffing important Descale as soon as possible → rest time for staff Staff Safety is a major concern Freelancers are important but difficult to manage Multiple triage areas—possible staffing with freelancers Quick response—increase staffing as soon as possible Maximal increase of staffing needed—most important factor Forward deployment of anaesthetist—allows for continuity of care Relieve staff after 8–12 h for breaks Optimise utilisation of manpower and supplies 1 Lessons learned from terror attacks: thematic priorities and development since 2001—results… essons learned from terror attacks: thematic priorities and development since 2001—results… 2629 Lessons learned from terror attacks: thematic priorities and development since 2001—results… Table 3 (continued) Lessons learned Staffing 17 Primary survey through surgeons and anaesthetists 18 ED staffed with nurse/doctor combo at each bed 19 Gather information and personnel during latent phase 20 Helicopters to transport staff and equipment 21 Triple: anaesthetist trauma surgeon abdominal surgery lead assessment and allocation to definite care 22 Efficient staff allocation 23 Pre hospital physicians useful 24 Using tags for triage—no resuscitation efforts until enough staffing 25 Train core of nurses in emergency medicine skills 26 Different specialties (ENT/psych) needed 27 Spread out teams to attend more patients 28 Too much staff available in ER—overcrowding 29 Good care despite communication failure—hence senior well trained personnel 30 Triage by senior medical officers 31 Keep track of staff showing up 32 Keep personnel in reserve/on standby 33 Experienced staff is vitally important 34 Surge in equipment and staff vital 35 Safety of personnel—idea of SWAT paramedics—therapy under fire 36 Increase blood bank staff 37 Photography staff/service to document injury Post-incident 38 Follow up on personnel—psychological and physiological Lessons learned Patient flow and distribution Pre-incident 1 Large number of mildly injured patients need to be expected and swiftly dealt with 2629 2629 Lessons learned from terror attacks: thematic priorities and development since 2001—results… Table 3 (continued) Lessons learned Staffing 17 Primary survey through surgeons and anaesthetists 18 ED staffed with nurse/doctor combo at each bed 19 Gather information and personnel during latent phase 20 Helicopters to transport staff and equipment 21 Triple: anaesthetist trauma surgeon abdominal surgery lead assessment and allocation to definite care 22 Efficient staff allocation 23 Pre hospital physicians useful 24 Using tags for triage—no resuscitation efforts until enough staffing 25 Train core of nurses in emergency medicine skills 26 Different specialties (ENT/psych) needed 27 Spread out teams to attend more patients 28 Too much staff available in ER—overcrowding 29 Good care despite communication failure—hence senior well trained personnel 30 Triage by senior medical officers 31 Keep track of staff showing up 32 Keep personnel in reserve/on standby 33 Experienced staff is vitally important 34 Surge in equipment and staff vital 35 Safety of personnel—idea of SWAT paramedics—therapy under fire 36 Increase blood bank staff 37 Photography staff/service to document injury Post-incident 38 Follow up on personnel—psychological and physiological Lessons learned Patient flow and distribution Pre-incident 1 Large number of mildly injured patients need to be expected and swiftly dealt with 2 Provide enough equipment but tailor to quick triage and transport During the incident 3 Majority of survivors are self-rescuer 4 Establish safe way for self-rescuer/non invalid patients 5 Increase ICU capacity move patients and unlock new areas 6 Patient flow—division between different hospital to avoid overload/right patient to the right hospital 7 Fast forward casualty flow 8 Coordinated distribution of casualties to hospitals 9 Log of most severely injured patients and their whereabouts 10 Quick redistribution of patients to clear ER for new ones 11 Use recovery room for monitoring unstable patient 12 Second wave of patient transfer between hospitals to avoid resource overstretching 13 Misdistribution between hospitals is a huge problem 14 Unidirectional patient flow—quick emptying of ED—one way pathway of care 15 Walking wounded redirected to satellite areas 16 Early evaluation of patients by senior doctors—early estimation of ICU capacity/oper- ating capacity needed 17 Transport off ICU patients to different hospitals needs to be thought of 18 Rapid removal from critically ill patients out of an unsafe environment 19 Transferring patients rapidly to definite care—rapid scene clearance 20 Consider the need for secondary transport (interhospital) 21 Distinction between circle 1 and circle 2 hospitals—direction of casualties accordingly 22 Quick evacuation of casualties—if stable enough severely injured patients to trauma hospitals p p Table 3 (continued) Lessons learned Staffing 17 Primary survey through surgeons and anaesthetists 18 ED staffed with nurse/doctor combo at each bed 19 Gather information and personnel during latent phase 20 Helicopters to transport staff and equipment 21 Triple: anaesthetist trauma surgeon abdominal surgery lead assessment and allocation to definite care 22 Efficient staff allocation 23 Pre hospital physicians useful 24 Using tags for triage—no resuscitation efforts until enough staffing 25 Train core of nurses in emergency medicine skills 26 Different specialties (ENT/psych) needed 27 Spread out teams to attend more patients 28 Too much staff available in ER—overcrowding 29 Good care despite communication failure—hence senior well trained personnel 30 Triage by senior medical officers 31 Keep track of staff showing up 32 Keep personnel in reserve/on standby 33 Experienced staff is vitally important 34 Surge in equipment and staff vital 35 Safety of personnel—idea of SWAT paramedics—therapy under fire 36 Increase blood bank staff 37 Photography staff/service to document injury Post-incident 38 Follow up on personnel—psychological and physiological Lessons learned Patient flow and distribution Pre-incident 1 Large number of mildly injured patients need to be expected and swiftly dealt with 2 Provide enough equipment but tailor to quick triage and transport During the incident 3 Majority of survivors are self-rescuer 4 Establish safe way for self-rescuer/non invalid patients 5 Increase ICU capacity move patients and unlock new areas 6 Patient flow—division between different hospital to avoid overload/right patient to the right hospital 7 Fast forward casualty flow 8 Coordinated distribution of casualties to hospitals 9 Log of most severely injured patients and their whereabouts 10 Quick redistribution of patients to clear ER for new ones 11 Use recovery room for monitoring unstable patient 12 Second wave of patient transfer between hospitals to avoid resource overstretching 13 Misdistribution between hospitals is a huge problem 14 Unidirectional patient flow—quick emptying of ED—one way pathway of care 15 Walking wounded redirected to satellite areas 16 Early evaluation of patients by senior doctors—early estimation of ICU capacity/oper- ating capacity needed 17 Transport off ICU patients to different hospitals needs to be thought of 18 Rapid removal from critically ill patients out of an unsafe environment 19 Transferring patients rapidly to definite care—rapid scene clearance 20 Consider the need for secondary transport (interhospital) 21 Distinction between circle 1 and circle 2 hospitals—direction of casualties accordingly 22 Quick evacuation of casualties—if stable enough severely injured patients to trauma hospitals Large number of mildly injured patients need to be expected and swiftly dealt with Provide enough equipment but tailor to quick triage and transport g y j p p y Provide enough equipment but tailor to quick triage and transport Majority of survivors are self-rescuer Establish safe way for self-rescuer/non invalid patients Increase ICU capacity move patients and unlock new areas Patient flow—division between different hospital to avoid overload/right patient to the right hospital Fast forward casualty flow Coordinated distribution of casualties to hospitals Log of most severely injured patients and their whereabouts Quick redistribution of patients to clear ER for new ones Use recovery room for monitoring unstable patient Second wave of patient transfer between hospitals to avoid resource overstretching Misdistribution between hospitals is a huge problem Unidirectional patient flow—quick emptying of ED—one way pathway of care Walking wounded redirected to satellite areas Early evaluation of patients by senior doctors—early estimation of ICU capacity/oper- ating capacity needed Transport off ICU patients to different hospitals needs to be thought of Rapid removal from critically ill patients out of an unsafe environment Transferring patients rapidly to definite care—rapid scene clearance Consider the need for secondary transport (interhospital) Distinction between circle 1 and circle 2 hospitals—direction of casualties accordingly Quick evacuation of casualties—if stable enough severely injured patients to trauma hospitals Increase ICU capacity move patients and unlock new areas Coordinated distribution of casualties to hospitals 1 3 2630 N. Discussion Schorscher et al. 1 Table 3 (continued) Lessons learned Patient flow and distribution 23 ED as epicentre—clear ED quick 24 Establish different treatment areas: fast track, psychiatric, major trauma, etc. Discussion 2630 Table 3 (continued) Lessons learned Patient flow and distribution 23 ED as epicentre—clear ED quick 24 Establish different treatment areas: fast track, psychiatric, major trauma, etc. Discussion 25 Primary evacuation of mildly injured patients to distant hospitals 26 Treat patient in level 2 trauma centres and only transfer if necessary to level 1 trauma centres 27 Divert non urgent patients to hospitals further away from incident site 28 Survivor reception centres to alleviate hospitals 29 Primary and balanced distribution between hospitals 30 Timely evacuation out of unsafe zone 31 Overload of patients at close by hospitals is huge problem 32 Fast track route for minor injuries 33 Patient flow—evacuation to cold zones 34 Directed quick patient flow to relieve triage area 35 Secondary patient flow according to capacity and specialty 36 Relocation of current patients 37 Cooperation between hospitals and trauma centres—recognise your limits and transfer 38 Tourniquet use und quick transfer to definite care 39 Track patients through hospital is a difficult task 40 Casualty clearing station—part of patient flow 41 Casualty disposition framework with an effective enhanced triage process 42 Safe transfer and handover of existing patients Lessons learned Cooperation and multidisciplinary approach P i id t Common goal is an important benefit Cross organisational planning important Communication channel between police, EMS and hospitals Staff training in combat medicine—cooperation with the military Awareness of tactical threat—idea of hazardous area response team Sharing of corporate knowledge—communication of information Clinical representation at strategic level to facilitate cooperation between networks/ regions Simultaneous search/rescue/treatment Common goal is an important benefit Cross organisational planning important Communication channel between police, EMS and hospitals Staff training in combat medicine—cooperation with the military Awareness of tactical threat—idea of hazardous area response team Sharing of corporate knowledge—communication of information Clinical representation at strategic level to facilitate cooperation between networks/ regions Simultaneous search/rescue/treatment Better communication between disaster agencies is important Importance of communication between different rescue teams Especially trauma patients need teamwork and good cooperation (surgery/anaesthetic) Cooperation of the entire medical system—prehospital as well as hospital Increase supplies through early communication with vendors Collaboration with police to deliver supplies Police command centre within hospital Chain of command: most senior official from all important specialties plus hospital admin Communication between rescue services vitally important Good teamwork is crucial Triple: anaesthetist, trauma surgeon abdominal surgeon lead assessment and allocation to definite care Multidisciplinary meetings Most senior emergency physician directs traffic/surgeons overseas area—triage not by most senior personnel Flexibility of services important—interaction/cooperation important Possibility for emergency services to cooperate and communicate Combined activation of major incident plans (all EMS services) Joint field command post Cooperation and communication between hospitals and all emergency services Dual surgical command-triage 1 essons learned from terror attacks: thematic priorities and development since 2001—results… Lessons learned from terror attacks: thematic priorities and development since 2001—results… 2631 1 3 Table 3 (continued) Lessons learned Cooperation and multidisciplinary approach 28 Cooperation between police and EMS 29 Methodical multidisciplinary care delivery 31 Good cooperation/collaboration between services is vital 32 Good interdisciplinary cooperation is vital 30 Command and control vs collaboration—both important 33 Multidisciplinary care saves lives 34 Cooperation between EMS and police/fire services 35 Multidisciplinary training—including police/fire service 36 Multi-professional networks/interaction including mental health 37 Cooperation between hospitals and trauma centres—recognise your limits and transfer 38 Crucial interaction/communication between hospital/police/municipalities 39 Provide patient lists to police to ease communication/information gathering for relatives 40 Good communication between incident site and hospital 41 Law enforcement medical commander—cross over between specialties/cooperation 42 Cooperation between civilian rescue teams and military 43 Good communication and situational awareness—use liaison officers 44 Coordination and collaboration should be planned and practised at intra/inter-regional, multiagency and multiprofessional levels 45 Support from neighbouring regions during terror Lessons learned Equipment and supplies Pre-incident 1 Functioning equipment is vitally important (broadband internet) 2 Constant resource evaluation 3 Combat medical care—reduced level of treatment per patient due to resource insuf- ficiencies 4 Need for appropriate equipment + supplies 5 Increase supply of available blood products 6 Mobile multiple casualty carts and disaster supply carts with equipment are helpful 7 Increase supplies through early communication with vendors 8 Assess Need for chemical and radiological monitors 9 Description of relevant capabilities of medical system 10 Provide megaphones 11 Provide protective personal equipment 12 Install mobile mass casualty vehicles with additional supplies 13 Increase and storage of supplies 14 Supply chains need to be reliable/organised well 15 Regional major incident plan to help allocate resources During the incident 16 Restrict laboratory and radiology testing 17 Protection of medical assets 18 Increase equipment—prep minor OR for major casualties 19 Rapid primary triage—only evacuation of the critical ill to nearest hospital (evacuation hospital) for stabilisation—to avoid resource overstretching 20 Second wave of patient transfer to avoid resource overstretching 21 Optimise utilisation of manpower and supplies 22 Collaboration with police to deliver supplies 23 Helicopters to transport staff and equipment 24 Basic equipment important and needed 25 Use of radio systems 26 Basic first aid kits on buses/trains 27 Allocation of resources difficult especially with multiple incidents 28 Enough equipment but mainly quick triage and transport 29 More advanced equipment including CBRN 30 Allocate resources to correct diagnosis 31 Extensive use of tourniquet Table 3 (continued) Lessons learned Cooperation and multidisciplinary approach 28 Cooperation between police and EMS 29 Methodical multidisciplinary care delivery 31 Good cooperation/collaboration between services is vital 32 Good interdisciplinary cooperation is vital 30 Command and control vs collaboration—both important 33 Multidisciplinary care saves lives 34 Cooperation between EMS and police/fire services 35 Multidisciplinary training—including police/fire service 36 Multi-professional networks/interaction including mental health 37 Cooperation between hospitals and trauma centres—recognise your limits and transfer 38 Crucial interaction/communication between hospital/police/municipalities 39 Provide patient lists to police to ease communication/information gathering for relatives 40 Good communication between incident site and hospital 41 Law enforcement medical commander—cross over between specialties/cooperation 42 Cooperation between civilian rescue teams and military 43 Good communication and situational awareness—use liaison officers 44 Coordination and collaboration should be planned and practised at intra/inter-regional, multiagency and multiprofessional levels 45 Support from neighbouring regions during terror Lessons learned Equipment and supplies Pre-incident 1 Functioning equipment is vitally important (broadband internet) 2 Constant resource evaluation 3 Combat medical care—reduced level of treatment per patient due to resource insuf- ficiencies 4 Need for appropriate equipment + supplies 5 Increase supply of available blood products 6 Mobile multiple casualty carts and disaster supply carts with equipment are helpful 7 Increase supplies through early communication with vendors 8 Assess Need for chemical and radiological monitors 9 Description of relevant capabilities of medical system 10 Provide megaphones 11 Provide protective personal equipment 12 Install mobile mass casualty vehicles with additional supplies 13 Increase and storage of supplies 14 Supply chains need to be reliable/organised well 15 Regional major incident plan to help allocate resources During the incident 16 Restrict laboratory and radiology testing 17 Protection of medical assets 18 Increase equipment—prep minor OR for major casualties 19 Rapid primary triage—only evacuation of the critical ill to nearest hospital (evacuation hospital) for stabilisation—to avoid resource overstretching 20 Second wave of patient transfer to avoid resource overstretching 21 Optimise utilisation of manpower and supplies 22 Collaboration with police to deliver supplies 23 Helicopters to transport staff and equipment 24 Basic equipment important and needed 25 Use of radio systems 26 Basic first aid kits on buses/trains 27 Allocation of resources difficult especially with multiple incidents 28 Enough equipment but mainly quick triage and transport 29 More advanced equipment including CBRN 30 Allocate resources to correct diagnosis 31 Extensive use of tourniquet Cooperation between police and EMS Methodical multidisciplinary care delivery Good cooperation/collaboration between services is vital Good interdisciplinary cooperation is vital Command and control vs collaboration—both important Multidisciplinary care saves lives Cooperation between EMS and police/fire services Multidisciplinary training—including police/fire service Multi-professional networks/interaction including mental health Cooperation between hospitals and trauma centres—recognise your limits and transfer Crucial interaction/communication between hospital/police/municipalities Provide patient lists to police to ease communication/information gathering for relatives Good communication between incident site and hospital Law enforcement medical commander—cross over between specialties/cooperation Cooperation between civilian rescue teams and military Good communication and situational awareness—use liaison officers Coordination and collaboration should be planned and practised at intra/inter-regional, multiagency and multiprofessional levels Support from neighbouring regions during terror Support from neighbouring regions during terror Functioning equipment is vitally important (broadband internet) Constant resource evaluation Combat medical care—reduced level of treatment per patient due to resource insuf- ficiencies Need for appropriate equipment + supplies Increase supply of available blood products Mobile multiple casualty carts and disaster supply carts with equipment are helpful Increase supplies through early communication with vendors Assess Need for chemical and radiological monitors Description of relevant capabilities of medical system Provide megaphones Provide protective personal equipment Install mobile mass casualty vehicles with additional supplies Increase and storage of supplies Supply chains need to be reliable/organised well Regional major incident plan to help allocate resources Restrict laboratory and radiology testing Protection of medical assets Increase equipment—prep minor OR for major casualties Rapid primary triage—only evacuation of the critical ill to nearest hospital (evacuation hospital) for stabilisation—to avoid resource overstretching Second wave of patient transfer to avoid resource overstretching Optimise utilisation of manpower and supplies Collaboration with police to deliver supplies Helicopters to transport staff and equipment Basic equipment important and needed Use of radio systems Basic first aid kits on buses/trains Allocation of resources difficult especially with multiple incidents Enough equipment but mainly quick triage and transport More advanced equipment including CBRN Allocate resources to correct diagnosis Extensive use of tourniquet 1 3 1 3 N. Discussion Schorscher et al. Discussion Schorscher et al. 2634 2634 N. Schorscher et al. Table 3 (continued) Lessons learned Psychiatric support 22 1/3 of victims develop post traumatic stress disorder (PTSB) 23 Psychological support—informal and formal Treatment 24 Improve bereavement support 25 Psychological first aid approach including self help 26 Bereavement nurses—24/7 access in the first 48 h 27 Monitor high risk groups of PTSD Lessons learned Record keeping Pre-incident 1 Create database of victims/casualties 2 Identification difficulties of victims—improve documentation to allow quicker identifica- tion 3 Improvement in identification: INTERPOL Disaster Victim Identification Standard 4 Standardised documentation at regional level/need for national casualty identification system 5 Patient identification difficult task—standardized identification and documentation systems During the incident 6 Written documentation strapped to patient 7 Early start of data collection 8 Good record keeping is essential 9 Lead agency to solely deal with record keeping 10 Importance of data collection of casualties at the scene 11 Importance of documentation—which patient has already been triaged 12 Better communication of patient information between prehospital and hospital setting 13 Detailed documentation of the disaster operation 14 Crisis management based on knowledge and data collection 15 Track patients through hospital—this is a difficult task 16 Photography staff/service to document injury 17 Importance of patient identification to allow for family reunification/bereavement Lessons learned Role understanding 1 Clear identification methods of roles—tags/vests—helps communication and command structures 2 Dedicated roles with clear defined duties during event—command and control physician; discharge/ patient flow organiser; ED supervisor 3 Assigned roles in disaster plan 4 Flexibility but clear roles 5 Know your capabilities/professional role 6 Low post traumatic stress disorder with good preparation, debriefing and high role clarity 7 Clear defined roles help to give security and confidence and improve outcome Lessons learned Team spirit 1 Keep team spirit up 2 Form coalition to keep up spirit and improve 3 Staff solidarity and professionalism vital 4 Public engagement and empowerment—communication and teaching 5 Professionalism and team spirit important for success 6 Mutual support important casualty incidents related to terror attacks [86]. Finally it is important to mention, that hospitals and rescue systems must prepare not only for terrorist attacks, but also for a wide spectrum of disasters. Ultimately, this is the key to increased resilience and successful mission management. were published in 2020 by Wurmb et al. and were very com- parable to those of this systematic review [85]. Discussion 2632 63 Table 3 (continued) Lessons learned 32 33 34 35 36 37 Lessons learned Pre-incident 1 2 3 4 5 6 7 8 9 10 During the incident 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Table 3 (continued) Lessons learned Equipment and supplies 32 Challenge of technology-equipment may fail 33 Back up resources—mobilise equipment and staff 34 Use of clotting devices/tourniquet 35 Surge capacity in equipment and staff is vital 36 Avoid main gate syndrome—overwhelmed resources at the closest hospital 37 Regional resource mobilisation is vital Lessons learned Medical treatment + type of injury Pre-incident 1 Use critical mortality rate as indicator for assessing medical care 2 Terror attack cause different/specific injury patterns 3 Except many blast injuries 4 Average ISS Score of ICU admission 5 Professional abilities are important 6 Train for new pattern of injuries 7 Medical management and knowledge vitally important 8 Stop the bleeding—tourniquet use—train as basic first aid 9 Integration of TCCC to ATLS 10 Improve therapy of paediatric cases—training During the incident 11 Evacuate patients as soon as possible 12 Rapid treatment is important 13 Use START system—simple triage rapid treatment 14 Combat medical care—reduced level of treatment per patient due to resource insuf- ficiencies 15 Early aggressive resuscitation predicts survival 16 Available surgical capacity needs to be increased 17 Restrict laboratory and radiology testing—minimal investigations 18 Only damage control surgery—the rest must wait 19 Medical treatment dependent on type of attack 20 Rapid provision of definite care 21 Therapy according to ATLS guidelines 22 Predominance of minor injuries during terrorist bombings (secondary/tertiary blast effect) and worried well patients 23 Critical injury appears roughly in 1/3rd of the cases 24 Blast injury results often in immediate death—if not there is often a combination with ear injury 25 Only 5% ISS > 15; 2% ISS > 25 26 Main injuries: blunt trauma, blast injury, penetrating wounds, burns 27 Rapid removal from critically ill patients out of an unsafe environment—scoop and run Therapy 28 Damage control treatment and mind set to increase surge capacity 29 Using tags for on scene triage—no resuscitation efforts until enough staffing 30 Treat patient in level 2 trauma centres and only transfer if necessary to level 1 trauma centres 31 Damage control treatment—no provision of individual definite care 32 Use ATLS/PHLTS standards 33 Use tactical combat casualty care + haemorrhage control 34 Roughly 10% suffer major injury 35 Schedule operations according to urgency 36 Extensive use of tourniquet 37 Offer immediate access to ORli Use critical mortality rate as indicator for assessing medical care Terror attack cause different/specific injury patterns Except many blast injuries Average ISS Score of ICU admission Professional abilities are important Train for new pattern of injuries Medical management and knowledge vitally important Stop the bleeding—tourniquet use—train as basic first aid Integration of TCCC to ATLS Improve therapy of paediatric cases—training Evacuate patients as soon as possible Rapid treatment is important Use START system—simple triage rapid treatment Combat medical care—reduced level of treatment per patient due to resource insuf- ficiencies Early aggressive resuscitation predicts survival Available surgical capacity needs to be increased Restrict laboratory and radiology testing—minimal investigations Only damage control surgery—the rest must wait Medical treatment dependent on type of attack Rapid provision of definite care Therapy according to ATLS guidelines Predominance of minor injuries during terrorist bombings (secondary/tertiary blast effect) and worried well patients Critical injury appears roughly in 1/3rd of the cases Blast injury results often in immediate death—if not there is often a combination with ear injury Only 5% ISS > 15; 2% ISS > 25 Main injuries: blunt trauma, blast injury, penetrating wounds, burns Rapid removal from critically ill patients out of an unsafe environment—scoop and run Therapy Damage control treatment and mind set to increase surge capacity Using tags for on scene triage—no resuscitation efforts until enough staffing Treat patient in level 2 trauma centres and only transfer if necessary to level 1 trauma centres Damage control treatment—no provision of individual definite care Use ATLS/PHLTS standards Use tactical combat casualty care + haemorrhage control Roughly 10% suffer major injury Schedule operations according to urgency Extensive use of tourniquet Offer immediate access to OR Patient therapy/flow: tourniquet use und quick transfer to definite care Safety vitally important—extent of therapy based on situational safety Early aggressive resuscitation predicts survival Available surgical capacity needs to be increased Restrict laboratory and radiology testing—minimal investigations Only damage control surgery—the rest must wait Medical treatment dependent on type of attack Medical treatment dependent on type of attack Blast injury results often in immediate death—if not there is often a combination with ear injury Main injuries: blunt trauma, blast injury, penetrating wounds, burns Rapid removal from critically ill patients out of an unsafe environment—scoop and run Therapy 1 3 essons learned from terror attacks: thematic priorities and development since 2001—results… 2633 Lessons learned from terror attacks: thematic priorities and development since 2001—results… Table 3 (continued) Lessons learned Zoning and scene safety Pre-incident 1 Full personal protective equipment and knowledge of the prehospital environment helpful 2 Beware of hospitals being soft targets 3 Safety of personnel—idea of SWAT paramedics—therapy under fire 4 Awareness of tactical threat—idea of hazardous area response team During the incident 5 Security at all hospital entrances—consider immediate lockdown 6 Simultaneous search/rescue/treatment—beware of security risks of this concept 7 Scene safety and scene control—beware of loss of rescue personnel—safety first 8 Beware second hit principle—protect trained personnel 9 Establish a safe way for self-rescuer 11 Safety of staff paramount 12 Rapid removal from critically ill patients out of an unsafe environment 13 Scene safety—important but huge problem hence rapid evacuation 14 Awareness for explosive devices being carried into hospital 10 Doctors not in red zone—triage in safe zone 15 Continuous assessment of scene safety 16 Safety first—triage/command outside danger zone 17 Manage uncertainties and scene 18 Evacuation problematics due to scene and geographical environment 19 Importance of scene safety and terror control 20 Scene safety—secondary attack/collapsing buildings/explosive Device 21 Conventional rescue teams out of danger zone 22 Operating capacity within on scene dressing station-tactical physicians as concept 23 Scene safety—zoning (exclusion zone) 24 Scene safety: develop best compromise btw safety of responders, immediate care and fast extraction 25 Triage outside hot zone—no therapy in hot zone if not trained 26 Tactical command post in safe zone 27 Scene safety cannot be guaranteed 28 Safety vitally important—extent of therapy based on situational safety 29 Challenges of being in the hot zone—multifaceted and continuously evolving 30 Recognition of situational aspect and severity + complexity—evolving risk Lessons learned Psychiatric support Post-incident Lessons learned from terror attacks: thematic priorities and development since 2001—results… 2633 Zoning and scene safety 1 1 Full personal protective equipment and knowledge of the prehospital environment helpful 2 Beware of hospitals being soft targets 3 Safety of personnel—idea of SWAT paramedics—therapy under fire 4 Awareness of tactical threat—idea of hazardous area response team During the incident 5 Security at all hospital entrances—consider immediate lockdown 6 Simultaneous search/rescue/treatment—beware of security risks of this concept 7 Scene safety and scene control—beware of loss of rescue personnel—safety first 8 Beware second hit principle—protect trained personnel 9 Establish a safe way for self-rescuer 11 Safety of staff paramount 12 Rapid removal from critically ill patients out of an unsafe environment 13 Scene safety—important but huge problem hence rapid evacuation 14 Awareness for explosive devices being carried into hospital 10 Doctors not in red zone—triage in safe zone 15 Continuous assessment of scene safety 16 Safety first—triage/command outside danger zone 17 Manage uncertainties and scene 18 Evacuation problematics due to scene and geographical environment 19 Importance of scene safety and terror control 20 Scene safety—secondary attack/collapsing buildings/explosive Device 21 Conventional rescue teams out of danger zone 22 Operating capacity within on scene dressing station-tactical physicians as concept 23 Scene safety—zoning (exclusion zone) 24 Scene safety: develop best compromise btw safety of responders, immediate care and fast extraction 25 Triage outside hot zone—no therapy in hot zone if not trained 26 Tactical command post in safe zone 27 Scene safety cannot be guaranteed 28 Safety vitally important—extent of therapy based on situational safety 29 Challenges of being in the hot zone—multifaceted and continuously evolving 30 Recognition of situational aspect and severity + complexity—evolving risk Lessons learned Psychiatric support Post-incident 1 Early psychiatric help is important 2 Site for acute stress disorder therapy needed 3 Good psychological support is necessary and important 4 Importance of post-traumatic stress disorder treatment groups 6 Do not underestimate the psychological and physical effects on health care workers 7 Psychological support for emergency services/healthcare worker/staff 8 Debriefing as stress relief 9 Psychiatric support before discharge for all patients 10 Psychological support for mildly injured patients 11 Set up survivor groups/psychological support 13 Psychological support to reduce long term impact of terrorism 14 Establishment of mental health counselling for staff 15 Psychiatric illness as hazard for emergency personnel 16 Establish psychological support centre 17 Low PTSD with good preparation, debriefing and high role clarity 18 Psychological follow up for staff and patients 19 Multiprofessional networks/interaction inclusive Mental Health 20 Everyone should be seen by psychiatric experts 21 Psychological care—Increase psychological support short and long term Early psychiatric help is important Site for acute stress disorder therapy needed Good psychological support is necessary and important Importance of post-traumatic stress disorder treatment groups Do not underestimate the psychological and physical effects on health care workers Psychological support for emergency services/healthcare worker/staff Debriefing as stress relief Psychiatric support before discharge for all patients Psychological support for mildly injured patients Set up survivor groups/psychological support Psychological support to reduce long term impact of terrorism Establishment of mental health counselling for staff Psychiatric illness as hazard for emergency personnel Establish psychological support centre Low PTSD with good preparation, debriefing and high role clarity Psychological follow up for staff and patients Multiprofessional networks/interaction inclusive Mental Health Everyone should be seen by psychiatric experts Psychological care—Increase psychological support short and long term 1 3 N. Limitations This systematic review has several limitations. Due to the vast amount of information only PubMed was used as a source. From the authors' point of view, this is a formal disadvantage, but it does not change the significance of the study as in contrast to the question of therapy effectiveness or the comparison of two forms of therapy, the aim here is to systematically present lessons learned. To get even more information, the data search could have been extended to other databases (e.g. Cochrane Library, Web of science) and the grey literature. Given the number of included articles, it is questionable whether this would have sig- nificantly changed the central message of the study. It is even possible that this would have made a systematic presentation and discussion even more difficult. CBRN attacks have been excluded from the research. The reason for that was that many special aspects have to be taken into account in these attacks. Nevertheless CBRN attacks are an important topic, which would need further exploration in the future. The restriction to OECD countries certainly causes a special view on the lessons learned and is thus also a source of bias. However, the aim was to look spe- cifically at countries where terror attacks are a rather rare event and rescue forces and hospitals are often unfamiliar with managing these challenges. Special injury patterns associated with terror attacks were not considered. This reduces the overall spectrum of included articles, but from the authors' point of view, a consideration of these would have exceeded the scope of this review. Funding Open Access funding enabled and organized by Projekt DEAL. The authors did not receive support from any organization for the submitted work. No funding was received to assist with the prepara- tion of this manuscript. No funding was received for conducting this study. No funds, grants, or other support was received. Discussion Furthermore the terror and disaster surgical care (TDSC®) course was developed in 2017 by the Deployment, Disaster, Tactical Surgery Working Group of the German Trauma Society to enhance the preparation of hospitals to manage mass 1 3 Lessons learned from terror attacks: thematic priorities and development since 2001—results… 2635 following versions of the manuscript. All authors read and approved the final manuscript. Consent for publication Not applicable. Open Access This article is licensed under a Creative Commons Attri- bution 4.0 International License, which permits use, sharing, adapta- tion, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Declarations Conflict of interest The authors have no conflicts of interest to declare that are relevant to the content of this article. Availability of data and materials Not applicable. Code availability Not applicable. Code availability Not applicable. Ethics approval Not applicable. Ethics approval Not applicable. Consent to participate Not applicable. References London bombings July 2005: the immediate pre-hospital medical response. Resuscitation. 2005;66:ix–xii. https://doi.org/ 10.1016/j.resuscitation.2005.07.005. 15. Roccaforte JD. The World Trade Center attack. Observations from New York’s Bellevue Hospital. Crit Care. 2001;5:307–9. https:// doi.org/10.1186/cc1056. j 33. Hughes G. The London bombings of 7 July 2005: what is the main lesson? Emerg Med J. 2006;23:666. https://doi.org/10.1136/emj. 2006.039586. 16. Martinez C, Gonzalez D. The World Trade Center attack: doctors in the fire and police services. Crit Care. 2001;5:304–6. https:// doi.org/10.1186/cc1055. 34. Shapira SC. Lessons learnt from the front line in Israel. BMJ. 2006;333:862. https://doi.org/10.1136/bmj.333.7573.862-a. g 17. Cook L. The World Trade Center attack: the paramedic response: an insider’s view. Crit Care. 2001;5:301–3. https://doi.org/10. 1186/cc1054. 35. Aylwin CJ, König TC, Brennan NW, Shirley PJ, Davies G, Walsh MS, Brohi K. Reduction in critical mortality in urban mass casu- alty incidents: analysis of triage, surge, and resource use after the London bombings on July 7, 2005. Lancet. 2006;368:2219–25. https://doi.org/10.1016/S0140-6736(06)69896-6. 18. Tamber PS, Vincent J-L. The World Trade Center attack: lessons for all aspects of health care. Crit Care. 2001;5:299–300. https:// doi.org/10.1186/cc1053. 36. Mohammed AB, Mann HA, Nawabi DH, Goodier DW, Ang SC. Impact of London’s terrorist attacks on a major trauma center in London. Prehosp Disaster Med. 2006;21:340–4. https://doi.org/ 10.1017/s1049023x00003988. 19. Simon R, Teperman S. The World Trade Center attack: lessons for disaster management. Crit Care. 2001;5:318–20. https://doi. org/10.1186/cc1060. 20. Mattox K. The World Trade Center attack disaster prepared- ness: health care is ready, but is the bureaucracy? Crit Care. 2001;5:323–5. https://doi.org/10.1186/cc1062. 37. Bland SA. HEMS training and the 7th July 2005: a personal per- spective. J R Nav Med Serv. 2006;92:130–5. 21. Shapira SC, Shemer J. Medical management of terrorist attacks. Isr Med Assoc J. 2002;4:489–92. 38. Leiba A, Blumenfeld A, Hourvitz A, Weiss G, Peres M, Schwartz D, et al. A four-step approach for establishment of a national medical response to mega-terrorism. Prehosp Disaster Med. 2006;21:436–40. https://doi.org/10.1017/s1049023x00004167. 22. Frykberg ER. Medical management of disasters and mass cas- ualties from terrorist bombings: how can we cope? J Trauma. 2002;53:201–12. https://doi.org/10.1097/00005373-20020 8000-00001. 39. Singer AJ, Singer AH, Halperin P, Kaspi G, Assaf J. Medical les- sons from terror attacks in Israel. J Emerg Med. 2007;32:87–92. https://doi.org/10.1016/j.jemermed.2006.06.007. 23. García-Castrillo Riesgo L, García MA. Terrorism in Spain: emer- gency medical aspects. Prehosp Disaster Med. 2003;18:148–51. https://doi.org/10.1017/s1049023x00000911. 40. Schwartz D, Pinkert M, Leiba A, Oren M, Haspel J, Levi Y, et al. References terrorist attacks: what have we learnt? Lancet. 2017;390:2735– 8. https://doi.org/10.1016/S0140-6736(17)31590-8. 27. Rodoplu U, Arnold JL, Tokyay R, Ersoy G, Cetiner S, Yücel T. Mass-casualty terrorist bombings in Istanbul, Turkey, Novem- ber 2003: report of the events and the prehospital emergency response. Prehosp Disaster Med. 2004;19:133–45. https://doi. org/10.1017/s1049023x00001643.f p g 10. Goralnick E, van Trimpont F, Carli P. Preparing for the next ter- rorism attack: lessons From Paris, Brussels, and Boston. JAMA Surg. 2017;152:419–20. https://doi.org/10.1001/jamasurg.2016. 4990. 11. Challen K, Lee ACK, Booth A, Gardois P, Woods HB, Goodacre SW. Where is the evidence for emergency planning: a scoping review. BMC Public Health. 2012;12:542. https://doi.org/10. 1186/1471-2458-12-542. g 28. Kluger Y, Mayo A, Soffer D, Aladgem D, Halperin P. Functions and principles in the management of bombing mass casualty inci- dents: lessons learned at the Tel-Aviv Souraski Medical Center. Eur J Emerg Med. 2004;11:329–34. https://doi.org/10.1097/00063 110-200412000-00006. 12. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JPA, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ. 2009;339: b2700. https://doi.org/10.1136/bmj.b2700. 29. Gutierrez de Ceballos JP, Turégano Fuentes F, Perez Diaz D, Sanz Sanchez M, Martin Llorente C, Guerrero Sanz JE. Casualties treated at the closest hospital in the Madrid, March 11, terrorist bombings. Crit Care Med. 2005;33:S107–12. https://doi.org/10. 1097/01.ccm.0000151072.17826.72. 13. Fattah S, Rehn M, Lockey D, Thompson J, Lossius HM, Wis- borg T. A consensus based template for reporting of pre-hos- pital major incident medical management. Scand J Trauma Resusc Emerg Med. 2014;22:5. https://doi.org/10.1186/ 1757-7241-22-5. 30. Kirschenbaum L, Keene A, O’Neill P, Westfal R, Astiz ME. The experience at St. Vincent’s Hospital, Manhattan, on September 11: preparedness, response, and lessons learned. Crit Care Med. 2005;33:S48-52. https://doi.org/10.1097/01.ccm.0000151067. 76074.21. 14. Wurmb T, Justice P, Dietz S, Schua R, Jarausch T, Kinstle U, et al. Qualitätsindikatoren für rettungsdienstliche Einsätze bei Terroranschlägen oder anderen Bedrohungslagen : Eine Pilots- tudie nach dem Würzburger Terroranschlag vom Juli 2016 [Qual- ity indicators for rescue operations in terrorist attacks or other threats : a pilot study after the Würzburg terrorist attack of July 2016]. Anaesthesist. 2016;2017(66):404–11. https://doi.org/10. 1007/s00101-017-0298-0. 31. Aschkenasy-Steuer G, Shamir M, Rivkind A, Mosheiff R, Shushan Y, Rosenthal G, et al. Clinical review: the Israeli experience: con- ventional terrorism and critical care. Crit Care. 2005;9:490–9. https://doi.org/10.1186/cc3762. 32. Lockey DJ, Mackenzie R, Redhead J, Wise D, Harris T, Weaver A, et al. References The first thing that stands out is that most lessons learned followed a certain pattern which repeated itself over the entire time frame considered in the systematic review. It can be assumed that in many cases it is therefore less a matter of lessons learned than of lessons identified. Although sound concepts exist, they do not seem to be sufficiently imple- mented in training and practice. This clearly shows that the improvement process has not yet been completed and a great deal of work still needs to be done. The important practical consequence is to implement the lessons identified in train- ing and preparation. This is mandatory to save as many vic- tims of terrorist attacks as possible, to protect rescue forces from harm and to prepare hospitals and public health at the best possible level. 1. Vienna shooting: Gunman hunted after deadly 'terror' attack. BBC News. 03.11.2020. 2. Keskinkılıç OZ. Hanau istüberall. Panorama. 19.02.2021. 3. Kar-Gupta S. Driver who rammed Paris police pledged allegiance to Islamic State: prosecutor. Reuters Media. 28.04.2020. 4. Brewis H. ISIS claims responsibility for Streatham terror attack carried out by Sudesh Amman. Evening Standard. 03.02.2020. 5. Park CL, Langlois M, Smith ER, Pepper M, Christian MD, Davies GE, Grier GR. How to stop the dying, as well as the killing, in a terrorist attack. BMJ. 2020;368: m298. https://doi. org/10.1136/bmj.m298. 6. Hirsch M, Carli P, Nizard R, Riou B, Baroudjian B, Baubet T, et al. The medical response to multisite terrorist attacks in Paris. Lancet. 2015;386:2535–8. https://doi.org/10.1016/S0140- 6736(15)01063-6. 7. Haug CJ. Report from Paris. N Engl J Med. 2015;373:2589–93. https://doi.org/10.1056/NEJMp1515229. 8. Philippe J-M, Brahic O, Carli P, Tourtier J-P, Riou B, Vallet B. French Ministry of Health’s response to Paris attacks of 13 November 2015. Crit Care. 2016;20:85. https://doi.org/10.1186/ s13054-016-1259-8. Author contributions All authors contributed substantially to the study conception and design. Material preparation, data collection and analy- sis were performed by NS and TW. The first draft of the manuscript was written by NS and TW and all authors commented on previous and 9. Carli P, Pons F, Levraut J, Millet B, Tourtier J-P, Ludes B, et al. The French emergency medical services after the Paris and Nice 9. Carli P, Pons F, Levraut J, Millet B, Tourtier J-P, Ludes B, et al. The French emergency medical services after the Paris and Nice 1 2636 N. Schorscher et al. protocol. Ann Surg. 2004;239:295–303. https://doi.org/10.1097/ 01.sla.0000114014.63423.55. Med. 2007;22:181–5. https://doi.org/10.1017/s1049023x000046 23 Crit Care. 2016;20:37. https://doi.org/10.1186/s13054-016-1202-z. p g 49. Turégano-Fuentes F, Pérez-Díaz D, Sanz-Sánchez M, Ortiz AJ. Overall assessment of the response to terrorist bombings in trains, Madrid, 11 March 2004. Eur J Trauma Emerg Surg. 2008;34:433. https://doi.org/10.1007/s00068-008-8805-2. 66. Paris terrorist attack. Early lessons from the intensivists. Crit Care. 2016;20:88. https://doi.org/10.1186/s13054-016-1246-0. p g 50. Pinkert M, Lehavi O, Goren OB, Raiter Y, Shamis A, Priel Z, et al. Primary triage, evacuation priorities, and rapid primary distribu- tion between adjacent hospitals–lessons learned from a suicide bomber attack in downtown Tel-Aviv. Prehosp Disaster Med. 2008;23:337–41. https://doi.org/10.1017/s1049023x00005975. 67. Gregory TM, Bihel T, Guigui P, Pierrart J, Bouyer B, Magrino B, et al. Terrorist attacks in Paris: surgical trauma experience in a referral center. Injury. 2016;47:2122–6. https://doi.org/10.1016/j. injury.2016.08.014. 68. Ghanchi A. Insights into French emergency planning, response, and resilience procedures from a hospital managerial perspective following the Paris Terrorist Attacks of Friday, November 13, 2015. Disaster Med Public Health Prep. 2016;10:789–94. https:// doi.org/10.1017/dmp.2016.21.i 51. Pryor JP. The 2001 world trade center disaster: summary and eval- uation of experiences. Eur J Trauma Emerg Surg. 2009;35:212– 24. https://doi.org/10.1007/s00068-009-9902-6. 52. Lockey DJ. The shootings in Oslo and Utøya island July 22, 2011: lessons for the International EMS community. Scand J Trauma Resusc Emerg Med. 2012;20:4. https://doi.org/10.1186/ 1757-7241-20-4. 69. Khorram-Manesh A. Europe on fire; medical management of ter- ror attacks—new era and new considerations. Bull Emerg Trauma. 2016;4:183–5.f 53. Sollid SJ, Rimstad R, Rehn M, Nakstad AR, Tomlinson A-E, Strand T, et al. Oslo government district bombing and Utøya island shooting July 22, 2011: the immediate prehospital emer- gency medical service response. Scand J Trauma Resusc Emerg Med. 2012;20:3. https://doi.org/10.1186/1757-7241-20-3. 70. Lesaffre X, Tourtier J-P, Violin Y, Frattini B, Rivet C, Stibbe O, et al. Remote damage control during the attacks on Paris: lessons learned by the Paris Fire Brigade and evolutions in the rescue system. J Trauma Acute Care Surg. 2017;82:S107–13. https://doi. org/10.1097/TA.0000000000001438. p g 54. Gaarder C, Jorgensen J, Kolstadbraaten KM, Isaksen KS, Skattum J, Rimstad R, et al. The twin terrorist attacks in Norway on July 22, 2011: the trauma center response. J Trauma Acute Care Surg. 2012;73:269–75. https://doi.org/10.1097/TA.0b013e31825a787f. 71. Wurmb T, Schorscher N, Justice P, Dietz S, Schua R, Jarausch T, et al. Structured analysis, evaluation and report of the emer- gency response to a terrorist attack in Wuerzburg, Germany using a new template of standardised quality indicators. Scand J Trauma Resusc Emerg Med. 2018;26:87. Med. 2007;22:181–5. https://doi.org/10.1017/s1049023x000046 23 59. Ashkenazi I, Turégano-Fuentes F, Einav S, Kessel B, Alfici R, Olsha O. Pitfalls to avoid in the medical management of mass casualty incidents following terrorist bombings: the hospital per- spective. Eur J Trauma Emerg Surg. 2014;40:445–50. https://doi. org/10.1007/s00068-014-0403-x. 43. Bloch YH, Schwartz D, Pinkert M, Blumenfeld A, Avinoam S, Hevion G, et al. Distribution of casualties in a mass-casualty incident with three local hospitals in the periphery of a densely populated area: lessons learned from the medical management of a terrorist attack. Prehosp Disaster Med. 2007;22:186–92. https:// doi.org/10.1017/s1049023x00004635. 60. Thompson J, Rehn M, Lossius HM, Lockey D. Risks to emer- gency medical responders at terrorist incidents: a narrative review of the medical literature. Crit Care. 2014;18:521. https://doi.org/ 10.1186/s13054-014-0521-1. g 44. Barnes R et al. —7 July Review Committee follow-up report MB2. doc. https://www.london.gov.uk/sites/default/files/gla_migrate_ files_destination/archives/assembly-reports-7july-report.pdf 61. Rimstad R, Sollid SJ. A retrospective observational study of med- ical incident command and decision-making in the 2011 Oslo bombing. Int J Emerg Med. 2015;8:4. https://doi.org/10.1186/ s12245-015-0052-9. i y p j y p p 45. Carresi AL. The 2004 Madrid train bombings: an analysis of pre- hospital management. Disasters. 2008;32:41–65. https://doi.org/ 10.1111/j.1467-7717.2007.01026.x. 62. Goralnick E, Halpern P, Loo S, Gates J, Biddinger P, Fisher J, et al. Leadership during the Boston marathon bombings: a quali- tative after-action review. Disaster Med Public Health Prep. 2015;9:489–95. https://doi.org/10.1017/dmp.2015.42. j 46. Raiter Y, Farfel A, Lehavi O, Goren OB, Shamiss A, Priel Z, et al. Mass casualty incident management, triage, injury distribution of casualties and rate of arrival of casualties at the hospitals: lessons from a suicide bomber attack in downtown Tel Aviv. Emerg Med J. 2008;25:225–9. https://doi.org/10.1136/emj.2007.052399. 63. Lee C, Walters E, Borger R, Clem K, Fenati G, Kiemeney M, et al. The San Bernardino, California, terror attack: two emer- gency departments’ response. West J Emerg Med. 2016;17:1–7. https://doi.org/10.5811/westjem.2016.1.29720. 47. Shirley PJ, Mandersloot G. Clinical review: the role of the inten- sive care physician in mass casualty incidents: planning, organisa- tion, and leadership. Crit Care. 2008;12:214. https://doi.org/10. 1186/cc6876. 64. Pedersen MJB, Gjerland A, Rund BR, Ekeberg Ø, Skogstad L. Emergency preparedness and role clarity among rescue workers during the Terror Attacks in Norway July 22, 2011. PLoS ONE. 2016. https://doi.org/10.1371/journal.pone.0156536. 48. Almgody G, Bala M, Rivkind AI. The approach to suicide bombing attacks: changing concepts. Eur J Trauma Emerg Surg. 2008;34:319. https://doi.org/10.1007/s00068-008-8901-3. 65. Service Médical du RAID. Tactical emergency medicine: les- sons from Paris marauding terrorist attack. References Significance of a Level-2, “selective, secondary evacuation” hos- pital during a peripheral town terrorist attack. Prehosp Disaster Med. 2007;22:59–66. https://doi.org/10.1017/s1049023x000043 50. g 24. Shamir MY, Weiss YG, Willner D, Mintz Y, Bloom AI, Weiss Y, et al. Multiple casualty terror events: the anesthesiologist’s perspective. Anesth Analg. 2004;98:1746–52. https://doi.org/10. 1213/01.ane.0000121307.33776.84. 41. Gómez AM, Domínguez CJ, Pedrueza CI, Calvente RR, Lillo VM, Canas JM. Management and analysis of out-of-hospital health- related responses to simultaneous railway explosions in Madrid. Spain Eur J Emerg Med. 2007;14:247–55. https://doi.org/10.1097/ MEJ.0b013e3280bef7c2. 25. Einav S, Feigenberg Z, Weissman C, Zaichik D, Caspi G, Kot- ler D, Freund HR. Evacuation priorities in mass casualty terror- related events: implications for contingency planning. Ann Surg. 2004;239:304–10. https://doi.org/10.1097/01.sla.0000114013. 19114.57. 42. Bloch YH, Leiba A, Veaacnin N, Paizer Y, Schwartz D, Kraskas A, et al. Managing mild casualties in mass-casualty incidents: lessons learned from an aborted terrorist attack. Prehosp Disaster 26. Almogy G, Belzberg H, Mintz Y, Pikarsky AK, Zamir G, Riv- kind AI. Suicide bombing attacks: update and modifications to the 1 3 2637 Lessons learned from terror attacks: thematic priorities and development since 2001—results… Med. 2007;22:181–5. https://doi.org/10.1017/s1049023x000046 23. Med. 2007;22:181–5. https://doi.org/10.1017/s1049023x000046 23 https://doi.org/10.1186/ s13049-018-0555-5. 55. The Boston Marathon bombings: a post-event review of the robust emergency responseED Manag. 2013:25:73–78. 72. Brandrud AS, Bretthauer M, Brattebø G, Pedersen MJ, Håpnes K, Møller K, et al. Local emergency medical response after a terrorist attack in Norway: a qualitative study. BMJ Qual Saf. 2017;26:806–16. https://doi.org/10.1136/bmjqs-2017-006517. 56. Jacobs LM, Wade DS, McSwain NE, Butler FK, Fabbri WP, East- man AL, et al. The Hartford Consensus: THREAT, a medical disaster preparedness concept. J Am Coll Surg. 2013;217:947–53. https://doi.org/10.1016/j.jamcollsurg.2013.07.002.i 57. Gates JD, Arabian S, Biddinger P, Blansfield J, Burke P, Chung S, et al. The initial response to the Boston marathon bomb- ing: lessons learned to prepare for the next disaster. Ann Surg. 2014;260:960–6. https://doi.org/10.1097/SLA.0000000000 000914. 73. Borel M, Rousseau R, Le Saché F, Pariente D, Castro S, Delay M, et al. Organization in response to massive afflux of war victims in civilian practice—experimental feedback from the November 2015 Paris terrorist attacks. J Visc Surg. 2017;154(Suppl 1):S3–7. https://doi.org/10.1016/j.jviscsurg.2017.07.007. 74. Bobko JP, Sinha M, Chen D, Patterson S, Baldridge T, Eby M, et al. A tactical medicine after-action report of the San Bernardino terrorist incident. West J Emerg Med. 2018;19:287–93. https://doi. org/10.5811/westjem.2017.10.31374. 58. Wang D-W, Liu Y, Jiang M-M. Review on emergency medical response against terrorist attack. Mil Med Res. 2014;1:9. https:// doi.org/10.1186/2054-9369-1-9. 1 3 2638 N. Schorscher et al. 82. JESIP—working together, Saving Lives. 24.06.2021. https://www. jesip.org.uk/jesip-the-programme. Accessed 24 Jun 2021. 75. Chauhan R, Conti BM, Keene D. Marauding terrorist attack (MTA): prehospital considerations. Emerg Med J. 2018;35:389– 95. https://doi.org/10.1136/emermed-2017-206959.i 83. Turner CDA, Lockey DJ, Rehn M. Pre-hospital manage- ment of mass casualty civilian shootings: a systematic litera- ture review. Crit Care. 2016;20:362. https://doi.org/10.1186/ s13054-016-1543-7. p g 76. Hunt P. Lessons identified from the 2017 Manchester and Lon- don terrorism incidents. Part 1: introduction and the prehospital phase. BMJ Mil Health. 2020;166:111–4. https://doi.org/10.1136/ jramc-2018-000934. 84. Hugelius K, Becker J, Adolfsson A. Five challenges when man- aging mass casualty or disaster situations: a review study. Int J Environ Res Public Health. 2020. https://doi.org/10.3390/ijerp h17093068. j 77. Hunt P. Lessons identified from the 2017 Manchester and Lon- don terrorism incidents. Part 3: the postincident and recovery phase. BMJ Mil Health. 2020;166:120–4. https://doi.org/10.1136/ jramc-2018-000936.i 85. Wurmb T, Franke A, Schorscher N, Kowalzik B, Helm M, Boh- nen R, et al. Emergency response to terrorist attacks: results of the federal-conducted evaluation process in Germany. Eur J Trauma Emerg Surg. 2020;46:725–30. https://doi.org/10.1007/ s00068-020-01347-8. Med. 2007;22:181–5. https://doi.org/10.1017/s1049023x000046 23 j 78. Hunt P. Lessons identified from the 2017 Manchester and London terrorism incidents. Part two: the reception and definitive care (hospital) phases. BMJ Mil Health. 2020;166:115–9. https://doi. org/10.1136/jramc-2018-000935. g j 79. Butler FK. Tactical combat casualty care: update 2009. J Trauma. 2010;69(Suppl 1):S10–3. https://doi.org/10.1097/TA.0b013e3181 e4220c. 86. Achatz G, Friemert B, Trentzsch H, Hofmann M, Blätzinger M, Hinz-Bauer S, et al. Terror and disaster surgical care: train- ing experienced trauma surgeons in decision making for a MASCAL situation with a tabletop simulation game. Eur J Trauma Emerg Surg. 2020;46:717–24. https://doi.org/10.1007/ s00068-020-01441-x. 80. Autrey AW, Hick JL, Bramer K, Berndt J, Bundt J. 3 Echo: con- cept of operations for early care and evacuation of victims of mass violence. Prehosp Disaster Med. 2014;29:421–8. https://doi.org/ 10.1017/S1049023X14000557. 81. Drew B, Montgomery HR, Butler FK. Tactical combat casualty care (TCCC) guidelines for medical personnel: 05 November 2020. J Spec Oper Med. 2020;20:144–51. 1 3 3
https://openalex.org/W4287683380	https://zenodo.org/records/4267991/files/GSCBPS-2020-0206.pdf	English	null	Bergenia ciliata as antibacterial agent	Zenodo (CERN European Organization for Nuclear Research)	2,020	cc-by	4,075	Bergenia ciliata as antibacterial agent Sayed Suliman Shah 1, †, Dawood Shah 1, Ibrahim Khan 2, *, †, Muhammad Ilyas 3, Sohail Ahmad Jan 4 and Imran Khan 1 1 Department of Chemistry, Government Degree College No.2 Mardan, Abdul Wali Khan Universit 3 Department of Botany, Bacha Khan University Charsadda, KP Pakistan. Publication history: Received on 28 June 2020; ; revised on 03 August 2020; accepted on 05 August 2020 Publication history: Received on 28 June 2020; ; revised on 03 August 2020; accepted on 05 Augu Article DOI: https://doi.org/10.30574/gscbps.2020.12.2.0206 Abstract The search for alternatives to combat antibiotics has been increased due to the upsurge in cases of antibiotic-resistant pathogens that has become a global phenomenon which threatens the human health. Medicinal plants provide multiple health benefits because these plants contain bioactive phytochemicals that are used in medicines as alternatives to various ailments of human beings. Bergenia ciliata is a perennial herb belongs to the family Saxifragaceae that consists of about 30 genera and 580 species worldwide. It has large, evergreen and leathery leaves with the pink to dark purple flowers and has various traditional and modern medical uses. This plant is considered is one of the most effective antibacterial agents. The present review emphasizes the antibacterial potential of Bergenia ciliata by compiling various researches and fragmented information. The most important phytochemicals derived from Bergenia ciliata launched on the pharmaceutical market are also discussed. This attempt will help the researchers to develop and synthesize new antibacterial drugs from Bergenia ciliata against drug-resistive pathogens. Keywords: Bergenia ciliata; Antibacterial agent; Drug-resistive pathogens; Pharmacology; Bioactive compounds Copyright © 2020 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons  Corresponding author: Khan Ibrahim or(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0. GSC Biological and Pharmaceutical Sciences, 2020, 12(02), 037-045 GSC Biological and Pharmaceutical Sciences, 2020, 12(02), 037-045 Corresponding author: Khan Ibrahim Copyright © 2020 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0. Copyright © 2020 Author(s) retain the copyright of this article. This article is published un 1. Introduction The interrelation of human beings with their surrounding herbs and medicinal plants draws its roots to the ancient civilizations [1]. Approximately 270,000 plant species are discovered till present time, while the existence of 400,000 plant species is possible in mother earth [2,3]. Plants are nature’s gift to the humans as they are the main source of many bioactive substances and modern as well as traditional drugs [4]. Approximately 35,000 to 70,000 plant species are used for medicines in the world and about 80 percent of world’s population used medicinal plants for the treatment of various diseases i.e. arthritis, diabetes, eye infection, loss of hair, loose motion, skin disease, abortion, kidney stone, cough, hyper tension, stomach diseases and loss of alertness etc. [5]. Plants are the main source of medicinal compounds. These medicinal compounds, which provide a great help in the discovery of new medicines, isolated from plants are being studied by many researchers worldwide for their various medicinal activities [6]. Bergenia ciliata (winter begonia or hairy bergenia) is a perennial herb belongs to the family Saxifragaceae which consists of about 30 genera and 580 species worldwide [7]. This plant is 30-60 cm in height and has large, evergreen and leathery leaves with pink to dark purple flowers [8]. Family Saxifragaceae has various biological properties including antibacterial, antiviral, antioxidant, anti-ulcer, anti-diabetic, analgesic, anti-plasmodial, antitumor, anti- tussive activity, antineoplastic, diuretic and anti-inflammatory properties [9]. Bergenia ciliata is considered as a miracle Shah et al. / GSC Biological and Pharmaceutical Sciences, 2020, 12(02), 037-045 herb due to its use in treatment of many diseases viz; gastrointestinal problems, malaria and kidney stone etc. [10]. This plant is found between altitudes of 800–3000 m in Pakistan (mainly in Northern areas), Afghanistan, South Tibet, India (mainly in Himalayas and West Bengal), Bhutan, Meghalaya, Bhutan and Nepal [11,12]. Bergenia ciliata contains various phytochemicals viz; tannins, metarbin, terpenoids, mucilage, bergenin, flavonoids, sitoindoside, steroids, saponins, afzelechin, coumarins, gallic acid, tannic acid, paashanolactone, glucosides, quercitin, albumen and mineral salts [13]. Generally there are about 58 different phytochemicals present in Bergenia ciliata species of which some are categorized into 11 categories viz; alcohols, fatty acids, phenols, carboxylic acids, terpenoids, flavonoids, glycosides, nitro compounds, sterol, cinnamic acid and volatile organic compounds [10]. herb due to its use in treatment of many diseases viz; gastrointestinal problems, malaria and kidney stone etc. [10]. 1. Introduction This plant is found between altitudes of 800–3000 m in Pakistan (mainly in Northern areas), Afghanistan, South Tibet, India (mainly in Himalayas and West Bengal), Bhutan, Meghalaya, Bhutan and Nepal [11,12]. Bergenia ciliata contains various phytochemicals viz; tannins, metarbin, terpenoids, mucilage, bergenin, flavonoids, sitoindoside, steroids, saponins, afzelechin, coumarins, gallic acid, tannic acid, paashanolactone, glucosides, quercitin, albumen and mineral salts [13]. Generally there are about 58 different phytochemicals present in Bergenia ciliata species of which some are categorized into 11 categories viz; alcohols, fatty acids, phenols, carboxylic acids, terpenoids, flavonoids, glycosides, nitro compounds, sterol, cinnamic acid and volatile organic compounds [10]. The innovation of antibiotics is the most important discovery in the 20th century which has helped the human being in his everlasting battle against pathogens. But some species of pathogens had developed a significant resistance against the discovered antibiotics [14]. An utmost focus of contemporary medicine is the search for new strategies and substances to combat infectious diseases. This problem is very staid because many bacteria are very resistant to antibacterial drugs and leads to difficulty in treatment of infectious diseases. For example, Staphylococcus aureus, a Gram-positive, round-shaped bacterium, is highly resistant to antibiotics [15]. Such a resistance of microorganisms towards antibiotics causes approximately 25, 000 deaths per year in the European Union and about 23, 000 deaths in the USA. This problem is approaching towards more dangerous levels due to the scarcity of development of new antibiotics as over the past twenty five years only two classes of antibiotics viz; lipopeptides and oxazolidinones have been developed [16,17]. The only solution for this problem is the synthesis and discoveries of new antibacterial and other compounds to combat the resistance of pathogens against the antibiotics. Therefore, it is essential to develop and synthesis new antibiotics to control the resistance of pathogens. Due to the great importance of antibacterial drugs developed and synthesized from plants we gather the utmost fragmented literature about antibacterial activities of Bergenia ciliata and its important phytochemicals. Moreover, this review will help the researchers across the world to develop new antibacterial drugs from Bergenia ciliata against the resistive pathogens. 2. Bergenia ciliata in traditional medicines Traditional medicinal system has been utilized since prehistoric times and draws its roots to ancient civilization. About 2500 plant species have medicinal value and about 6000 plants have been recognized to be used in traditional, folk and herbal medicine [18]. Bergenia ciliata is an attractive medicinal plant which attracts many researchers due to various medicinal activities. It is also used in traditional medicines for the treatment of many diseases i.e. skin diseases, muscular/skeletal disorders, pulmonary infections, gastrointestinal infections, fever, eye diseases ,worm infection , diarrhea, respiratory diseases, renal disorders, fever , oral infections, cancer, and gynecological disorders (Figure 1) [7]. Traditional healers used it mainly in the form of powder. Juice of dried rhizome of Bergenia ciliata in combination with honey has been used by post-partum women against the gastrointestinal disorders. Its rhizome has also been used by adults as an anti-helimintic. Boiled roots of Bergenia ciliata in combination with salt have been reported to be used for the cure of asthma. Its use for the treatment and prevention of piles and cancer is also reported [7,19,20]. Figure 1 Some important traditional uses of Bergenia ciliata Figure 1 Some important traditional uses of Bergenia ciliata 38 38 Shah et al. / GSC Biological and Pharmaceutical Sciences, 2020, 12(02), 037-045 Figure 2 Chemical structures of compounds 1-5 isolated from Bergenia ciliata. 3. Phytochemistry of Bergenia ciliata Bergenia ciliata is an important medicinal herbal plant that contains many important bioactive compounds. Many researchers have evaluated various phytochemicals from Bergenia ciliata (Figure 2; Figure 3). Generally there are about 58 phytochemicals present in Bergenia ciliata species of which 48 volatile organic compounds are categorized into 11 categories viz; alcohols, fatty acids, phenols, carboxylic acids, terpenoids, flavonoids, glycosides, nitro compounds, sterol, cinnamic acid and volatile organic compounds [10]. Some of the important classes of compounds along with their IUPAC names and general formula isolated from Bergenia ciliata are given in Table 1. Figure 2 Chemical structures of compounds 1-5 isolated from Bergenia ciliata. Figure 2 Chemical structures of compounds 1-5 isolated from Bergenia ciliata. Figure 2 Chemical structures of compounds 1-5 isolated from Bergenia ciliata. 39 Shah et al. / GSC Biological and Pharmaceutical Sciences, 2020, 12(02), 037-045 Table 1 Some important phytochemicals isolated from Bergenia ciliata. S. No. Phytochemicals IUPAC names General formula Ref. Phenolic contents 1 Bergenin (2R,3S,4S,4aR,10bS)-3,4,8,10-tetrahydroxy-2-(hydroxymethyl)-9- methoxy-3,4,4a,10b-tetrahydro-2H-pyrano[3,2-c]isochromen-6-one C14H16O9 [21] 2 Gallic acid 3,4,5-Trihydroxybenzoic acid C6H2(OH)3COOH or C7H6O5 [20] 3 Tannic acid [2,3-dihydroxy-5-[[(2R,3R,4S,5R,6S)-3,4,5,6-tetrakis[[3,4-dihydroxy-5- (3,4,5-trihydroxybenzoyl) oxybenzoyl]oxy] oxan-2- yl]methoxycarbonyl]phenyl] 3,4,5-trihydroxybenzoate C76H52O46 [22] Alcohols 4 3-Pentanol 3-Pentanol C5H12O [23] 5 Terpinen-4-ol Terpinen-4-ol C10H18O [24] 6 Pentanol Pentan-1-ol C8H18O [7] Flavonoids 7 (+)-Afzelechin (2R,3S)-2-(4-hydroxyphenyl)-3,4-dihydro-2H-chromene-3,5,7-triol C15H14O5 [25] 8 Quercetin 3-o-α-L- arbinofuranoxide 3-[(2S,3R,4R,5S)-3,4-dihydroxy-5-(hydroxymethyl)oxolan-2-yl]oxy-2- (3,4-dihydroxyphenyl)-5,7-dihydroxychromen-4-one C20H18O11 [26] Fatty acids 9 2-Methyl butanoic acid 2-Methyl butanoic acid C5H10O2 [27] 10 Decanoic acid Decanoic acid C10H20O2 [28] 11 Nonanoic acid Nonanoic acid C9H18O2 [27] Terpenoids 12 Camphor 1,7,7-Trimethylbicyclo[2.2.1]heptan-2-one C10H16O [29] 13 Glucoside 4,11,11-trimethyl-8-methylene-bicyclo[7.2.0]undec-4-ene C15H24 [7] Terpenes 14 Linalool 3,7-dimethylocta-1,6-dien-3-ol C10H18O [30] 15 Beta-phellandrene 3-methylidene-6-propan-2-ylcyclohexene C10H16 [31] Table 1 Some important phytochemicals isolated from Bergenia ciliata. General formula 40 40 Shah et al. / GSC Biological and Pharmaceutical Sciences, 2020, 12(02), 037-045 Shah et al. / GSC Biological and Pharmaceutical Sciences, 2020, 12(02), 037-045 Figure 3 Chemical structures of compounds 6-15 isolated from Bergenia ciliata. nia ciliata as antibacterial agent Figure 3 Chemical structures of compounds 6-15 isolated from Bergenia ciliata. Shah et al. / GSC Biological and Pharmaceutical Sciences, 2020, 12(02), 037-045 Shah et al. / GSC Biological and Pharmaceutical Sciences, 2020, 12(02), 037-045 inhibitory effect against three pathogens, namely Escherichia coli, Bacillus subtilis and Pseudomonas aeruginosa [30]. It was published that rhizome extracts (in methanol, ethanol, water and n-hexane) of Bergenia ciliata showed significant antibacterial potential against four bacterial strains; Klebsiella pneumonia, Staphylococcus auereus, Bacillus subtilis and Escherichia coli. Among all the extracts the methanolic extract showed greater inhibitory potential (12.9±0.5 to 11.8±0.7mm) while the lowest inhibition was showed by aqueous fraction (4.5±0.3 to 3.5±0.1mm) [32]. Another study reported that ethanolic rhizome extracts of Bergenia ciliata showed remarkable antibacterial activities against selected pathogens Salmonella typhimurium (inhibition zone 12±1.0mm) and Escherichia coli (inhibition zone 11.33±0.57mm) [33]. The ethanolic, chloroform, butanolic, hexane, ethyl acetate and aqueous roots and leaves extracts of Bergenia ciliata were evaluated for their antibacterial inhibitory activity against Gram positive and Gram negative strains viz; Staphylococcus auereus, Bacillus subtilis, micrococcus and Bacillus megalerium. It was observed from the results that root extracts of Bergenia ciliata have higher antibacterial activities as compared to the leaves extracts [34]. The antibacterial activities of callus extracts of Bergenia ciliata were determined against selected pathogens viz; Pseudomonas aeruginosa, Staphylococcus aureus, Pseudomonas aeruginosa, Pseudomonas aeruginosa, Staphylococcus aureus and Pseudomonas aeruginosa [35]. The methanol, ethyl acetate and hexane extracts of Bergenia ciliata were evaluated against the selected bacterial strains including Nocardia tenerifensis, Bacillus megaterium and Bacillus subtilis. Among all the extracts the methanolic extract was found to have better antibacterial potential [20]. The antibacterial potential of the crude extract of Bergenia ciliata was determined against six bacterial strains including Bordetella bronchiseptica, Salmonella Setubal, Escherichia coli, Micrococcus luteus, Staphylococcus aureus and Salmonella typhimurium. The results showed that the crude extract significantly inhibited all the selected pathogens [36]. In vitro ethanolic , hexane, distilled water and butanolic extracts of Bergenia ciliata also showed significantly antibacterial activities against Escherichia coli, Staphylococcus aureus, Pseudomonas aeruginosa and Streptococcus faecalis bacteria [37]. Leaves extracts of Bergenia ciliata showed significant antibacterial potential against Staphylococcus aureus and Bacillus megaterium with an inhibition zone of 8–12 mm and 10–20 mm respectively [38]. Another study also evaluated the antibacterial potential of ethanolic extracts of Bergenia ciliata against seven selected bacteria including Bacillus subtilis, Klebsiella pneumonia, Staphylococcus aureus, Escherichia coli, Salmonella typhi, Shigella dysentriae and Sacchromyces cerevisiae with the inhibition zone 22.8±0.15 mm, 18.6±0.15 mm, 24.0±0.10 mm, 23.7±0.25 mm, 22.8±0.15 mm, 29.4±0.47 mm and 23.4±0.26 mm respectively [39]. 4. Bergenia ciliata as antibacterial agent The renewed interest of researchers in medicinal plants allowed them to investigate medicinal plants, such as Bergenia ciliata, and search new strategies and substances to combat infectious diseases. Several pharmacological studies reported that Bergenia ciliata is a potent antibacterial agent (Table 2). In present study we have reviewed different extracts of Bergenia ciliata that can be used against pathogenic bacteria. The antibacterial activities of three medicinal plants including Bergenia ciliata have been investigated. The acetone extracts of the plants showed remarkable anti- 41 Shah et al. / GSC Biological and Pharmaceutical Sciences, 2020, 12(02), 037-045 5. Conclusion The present study gathers the fragmented literature about the antibacterial potential of Bergenia ciliata. The phytochemicals present in of Bergenia ciliata exhibit various biological activities including antibacterial, antioxidants, antifungal, antihaemolytic and cytotoxic. This mini-review would be beneficial for researchers in future to study the medicinally important constituents of Bergenia ciliata to develop new drugs that can be used to combat the antibiotic- resistant pathogens. Based on current review, Bergenia ciliata is considered is one of the most important antibacterial agent which attributed to its phytochemical constituents but some of the constituents of Bergenia ciliata has adverse effects on human health. Therefore, in-depth and intensive studies are required in order to isolate the antibacterial molecules and evaluate the safety, dosage and possible side effects. Shah et al. / GSC Biological and Pharmaceutical Sciences, 2020, 12(02), 037-045 Table 2 Antibacterial activities of Bergenia ciliata. Extract Susceptible bacteria Ref. Acetone extract Escherichia coli, Bacillus subtilis and Pseudomonas aeruginosa [30] Methanol, ethanol, water and n- hexane extracts Klebsiella pneumonia, Staphylococcus auereus, Bacillus subtilis and Escherichia coli [32] Ethanol extract Salmonella typhimurium and Escherichia coli [33] Ethanol, chloroform, butanol, hexane, ethyl acetate and aqueous extracts Staphylococcus auereus, Bacillus subtilis, micrococcus and Bacillus megalerium [34] Callus extract Pseudomonas aeruginosa, Staphylococcus aureus, Pseudomonas aeruginosa, Pseudomonas aeruginosa, Staphylococcus aureus and Pseudomonas aeruginosa [35] Methanol, ethyl acetate and hexane extracts Nocardia tenerifensis, Bacillus megaterium and Bacillus subtilis [20] Crude extract Bordetella bronchiseptica, Salmonella Setubal, Escherichia coli, Micrococcus luteus, Staphylococcus aureus and Salmonella typhimurium [36] Ethanol, hexane, distilled water and butane extracts Escherichia coli, Staphylococcus aureus, Pseudomonas aeruginosa and Streptococcus faecalis [37] Leaves extracts Staphylococcus aureus and Bacillus megaterium [38] Ethanol extract Bacillus subtilis, Klebsiella pneumonia, Staphylococcus aureus, Escherichia coli, Salmonella typhi, Shigella dysentriae and Sacchromyces cerevisiae [39] Table 2 Antibacterial activities of Bergenia ciliata. Extract Susceptible bacteria Ref. Acetone extract Escherichia coli, Bacillus subtilis and Pseudomonas aeruginosa [30] Methanol, ethanol, water and n- hexane extracts Klebsiella pneumonia, Staphylococcus auereus, Bacillus subtilis and Escherichia coli [32] Ethanol extract Salmonella typhimurium and Escherichia coli [33] Ethanol, chloroform, butanol, hexane, ethyl acetate and aqueous extracts Staphylococcus auereus, Bacillus subtilis, micrococcus and Bacillus megalerium [34] Callus extract Pseudomonas aeruginosa, Staphylococcus aureus, Pseudomonas aeruginosa, Pseudomonas aeruginosa, Staphylococcus aureus and Pseudomonas aeruginosa [35] Methanol, ethyl acetate and hexane extracts Nocardia tenerifensis, Bacillus megaterium and Bacillus subtilis [20] Crude extract Bordetella bronchiseptica, Salmonella Setubal, Escherichia coli, Micrococcus luteus, Staphylococcus aureus and Salmonella typhimurium [36] Ethanol, hexane, distilled water and butane extracts Escherichia coli, Staphylococcus aureus, Pseudomonas aeruginosa and Streptococcus faecalis [37] Leaves extracts Staphylococcus aureus and Bacillus megaterium [38] Ethanol extract Bacillus subtilis, Klebsiella pneumonia, Staphylococcus aureus, Escherichia coli, Salmonella typhi, Shigella dysentriae and Sacchromyces cerevisiae [39] Table 2 Antibacterial activities of Bergenia ciliata. able 2 Antibacterial activities of Bergenia ciliata. 42 Shah et al. / GSC Biological and Pharmaceutical Sciences, 2020, 12(02), 037-045 Disclosure of conflict of interest Disclosure of conflict of interest All the authors declare that they have no conflict of interest. Acknowledgments Acknowledgments We would like to thank our elder brothers Said Iftikhar Ali Shah, Mohammad Halim and Muhammad Saleem for their encouragement, moral support, personal attention and care. References Molecular analyses of genetic variability in the populations of Bergenia ciliata in Indian Himalayan Region (IHR). Physiology and Molecular Biology of Plants, 26(5), 975–984. [13] Zafar R, Ullah H, Zahoor M and Sadiq A. (2019). Isolation of bioactive compounds from Bergenia ciliata (haw.) Sternb rhizome and their antioxidant and anticholinesterase activities. BMC Complementary and Alternative Medicine, 19(1) 1–13. [14] González-Bello C. (2017). Antibiotic adjuvants – A strategy to unlock bacterial resistance to antibiotics. Bioorganic and Medicinal Chemistry Letters, 27(18), 4221–4228. [15] Catteau L, Zhu L, Bambeke FV and Quetin-Leclercq J. (2018). Natural and hemi-synthetic pentacyclic triterpenes as antimicrobials and resistance modifying agents against Staphylococcus aureus: a review. Phytochemistry Reviews, 17(5), 1129–1163. [16] Barker WT, Chandler CE, Melander RJ, Ernst RK and Melander C. (2019). Tryptamine derivatives disarm colistin resistance in polymyxin-resistant gram-negative bacteria. Bioorganic and Medicinal Chemistry, 27(9), 1776– 1788. [17] Martens E and Demain AL. (2017). The antibiotic resistance crisis, with a focus on the United States. Journal of Antibiotics, 70(5), 520–526. [18] Ganesan P and Pradeepa P. (2017). Development and characterization of nanofibrous mat from PVA/Tridax Procumbens (TP) leaves extracts. Wound Medicine, 19(3), 15–22. [19] Saha S, Shrivastav PS and Verma RJ. (2014). Antioxidative mechanism involved in the preventive efficacy of Bergenia ciliata rhizomes against experimental nephrolithiasis in rats. Pharmaceutical Biology, 52(6), 712–722. [20] Singh M, Pandey N, Agnihotri V, Singh KK and Pandey A. (2017). Antioxidant, antimicrobial activity and bioactive compounds of Bergenia ciliata Sternb: A valuable medicinal herb of Sikkim Himalaya. Journal of Traditional and Complementary Medicine, 7(2), 152–157. [21] Chauhan R, Ruby KM and Dwivedi J. (2013). Secondary metabolites found in Bergenia species: A compendious review. International Journal of Pharmacy and Pharmaceutical Sciences, 5(1), 9–16. [22] Walter NS, Bagai U and Kalia S. (2013). Antimalarial activity of Bergenia ciliata (Haw.) Sternb. against Plasmodium berghei. Parasitology Research, 112(9), 3123–3128. [23] Gyawali R and Kim K-S. (1970). Bioactive volatile compounds of three medicinal plants from Nepal. Kathmandu University Journal of Science, Engineering and Technology, 8(1), 51–62. [24] Engl T, Kumar P, Javed MS, Tewari AK and Bisht KS. (2017). Phytochemical investigation and in vitro antifungal activity of essential oil from the rhizome of Bergenia stracheyi (Hook.f. & Thomson) Engl. Journal of Drug Delivery and Therapeutics, 7(1), 37–43. [25] Dhalwal K, Shinde VM, Biradar YS and Mahadik KR. (2008). Simultaneous quantification of bergenin, catechin, and gallic acid from Bergenia ciliata and Bergenia ligulata by using thin-layer chromatography. References [1] Khan I, Najeebullah, Ali SM and Shinwari ZK. (2016). Phytopharmacological and ethnomedicinal uses of the Genus Berberis (Berberidaceae): A review. Tropical Journal of Pharmaceutical Research, 15(9), 2047–2057. [2] Verma SK, Gond SK, Mishra A, SharmaVK, Kumar J, Singh DK, Kumar A and Kharwar RN. (2017). Fungal endophytes representing diverse habitats and their role in plant protection. Developments in Fungal Biology and Applied Mycology, 135–157. [3] Deák B, Valkó O, Török P, Kelemen A, Bede A, Csathó AI and Tóthmérész B. (2018). Landscape and habitat filters jointly drive richness and abundance of specialist plants in terrestrial habitat islands. Landscape Ecology, 33(7), 117–1132. [4] Tungmunnithum D, Thongboonyou A, Pholboon A and Yangsabai A. (2018). Flavonoids and other phenolic compounds from medicinal plants for pharmaceutical and medical aspects: An overview. Medicines, 5(3), 93. [5] Ambasta SK, Kumari S, yadav AK, Trivedi I, Prasad B and Sinha UK. (2016). Medicinal plants of Bihar and its neighboring region which needs attention for their conservation. European Journal of Biomedical and Pharmaceutical Sciences, 3(4), 544–55. [6] Majolo F, Delwing LKdOB, Marmitt DJ, Bustamante-Filho IC and Goettert MI. (2019). Medicinal plants and bioactive natural compounds for cancer treatment: Important advances for drug discovery. Phytochemistry Letters, 31(4), 196–207. [7] Ahmad M, Butt MA, Zhang G, Sultana S, Tariq A and Zafar M. (2018). Bergenia ciliata: A comprehensive review of its traditional uses, phytochemistry, pharmacology and safety. Biomedicine and Pharmacotherapy, 97(5), 708– 721. [8] Żbikowska B, Franiczek R, Sowa A, Połukord G, Krzyzanowska B and Sroka Z. (2017). Antimicrobial and Antiradical Activity of Extracts Obtained from Leaves of Five Species of the Genus Bergenia: Identification of Antimicrobial Compounds. Microbial Drug Resistance, 23(6), 771–780. [9] Timalsena S and Lamichhane PP. (2019). Astible Rivularis: Bioactive Compounds and Pharmacological Functions. Chinese Journal of Integrative Medicine, 25(10), 795–799. [10] Hussain A, Kanth M, Shrivastva PK, Sharma M, Tripath J and Khan MA. (2019). Phytochemical analysis of the rhizomes of Bergenia ciliata (How) Sternb. Journal of Drug Delivery and Therapeutics, 9(3), 412–416. [11] Phull A-R, Abbas Q, Ali A, Raza H, kim S-J, Zia M and Ul-haq I. (2016). Antioxidant, cytotoxic and antimicrobial activities of green synthesized silver nanoparticles from crude extract of Bergenia ciliata. Future Journal of Pharmaceutical Sciences, 2(1), 31–36. 43 Shah et al. / GSC Biological and Pharmaceutical Sciences, 2020, 12(02), 037-045 [12] Tiwari V, Meena B, Nair NK and Rana TS. (2020). [31] Byahatti VV, Pai KV and D’Souza MG. (2010). Effect of Phenolic Compounds from Bergenia ciliata (Haw.) Sternb. leaves on Experimental kidney stones. Ancient science of life, 30(1), 14–7. Shah SS, Shah D, Khan I, Ilyas M, Jan SA and Khan I. (2020). Bergenia ciliata as antibacterial agent. GSC Biological and Pharmaceutical Sciences, 12(2), 37-45. References Journal of Food Composition and Analysis, 21(6), 496–500. [26] Kumar V and Tyagi D. (2013). Review on phytochemical, ethnomedical and biological studies of medically useful genus Bergenia. International Journal of Current Microbiology and Appied Sciences, 2(5), 328–334. [27] Kour H, Raina R, Verma PK, Pankaj NK and Singh SP. (2019). Phytochemical ingredients and pharmacological properties of Bergenia ciliata. Journal of Veterinary Pharmacology and Toxicology, 18(1), 1-10. [28] Shirsat VA, Dhainje VM, Krishnapriya M and Ghone S. (2018). Identification of potential antioxidants by in-vitro activity guided fractionation of Bergenia ligulata. Pharmacognosy Magazine, 4(15), 78-84. [29] Nasreen S, Safeer S, Dar KK, Andleeb S, Ejaz M, Khan MA and Ali S. (2018). Etiology of hepatocellular carcinoma and treatment through medicinal plants: a comprehensive review. Oriental Pharmacy and Experimental Medicine, 18(3), 187–197. [30] Shankar KG, Fleming AT, Vidhya R and Pradhan N. (2016). Synergistic efficacy of three plant extracts, Bergenia ciliata, Acorus calamus and Dioscorea bulbifera for antimicrobial activity. International Journal of Pharma and Bio Sciences, 7(4), 619–628. [31] Byahatti VV, Pai KV and D’Souza MG. (2010). Effect of Phenolic Compounds from Bergenia ciliata (Haw.) Sternb. leaves on Experimental kidney stones. Ancient science of life, 30(1), 14–7. 44 44 Shah et al. / GSC Biological and Pharmaceutical Sciences, 2020, 12(02), 037-045 Shah et al. / GSC Biological and Pharmaceutical Sciences, 2020, 12(02), 037-045 [32] Yousaf S, Kaukab G, Gul H, Khalid N, Kausar R, Ahmed H, Ajab H and Gulfraz M. (2018). Pharmacological and phytochemical analysis of Bergenia ciliata leaf and rhizome extracts. Pakistan Journal of Pharmaceutical Sciences, 31(5), 1911–1916. [33] Verma R, Tapwal A, Kumar D and Puri S. (2019). Assessment of antimicrobial potential and phytochemical profiling of ethnomedicinal plant Bergenia ciliata (Haw.) Sternb. In Western Himalaya. Journal of Microbiology, Biotechnology and Food Sciences, 9(1), 15–20. [34] Pokhrel P, Parajuli RR, Tiwari AK and Banerjee J. (2014). A short glimpse on promising pharmacological effects of Begenia ciliata. Journal of Applied Pharmaceutical Research, 2(1), 1–6. [35] Khan UA, Rahman H2, Niaz Z, Qasim M, Khan J, Tayyaba and Rehman B. (2013). Antibacterial activity of some medicinal plants against selected human pathogenic bacteria. European Journal of Microbiology and Immunology, 3(4), 272–274. [36] Ahmed M, Phul AR, Ihsan-ul-Haq, Bibi G, Mazhar K, Ur-Rehman T, Zia M and Mirza B. (2016). Antioxidant, anticancer and antibacterial potential of Zakhm-e-hayat rhizomes crude extract and fractions. Pakistan Journal of Pharmaceutical Sciences, 29(3), 895–902. [37] Pokhrel P, Banerjee J, Dahal P, Khanal H, Gupta AK and Dey BK. (2014). Phytochemical Screening and Biological Evaluation of Different Parts of Plant Bergenia ciliata. Journal of Pharmacognosy and Phytochemistry, 3(4), 220- 224. [38] Rajbhandari M, Mentel R, Jha PK, Chaudhary RP, Bhattarai S, Gewali MB, Karmacharya N, Hipper M and Lindequist U. (2009). Antiviral Activity of Some Plants Used in Nepalese Traditional Medicine. Evidence-Based Complementary and Alternative Medicine, 6(4), 517–522. [39] Khan N, Abbasi AM, Dastagir G, Nazir A, Shah GM, Shah MM and Shah MH. (2014). Ethnobotanical and antimicrobial study of some selected medicinal plants used in Khyber Pakhtunkhwa (KPK) as a potential source to cure infectious diseases. BMC Complementary and Alternative Medicine, 14, 1–10. How to cite this article How to cite this article 45 45
https://openalex.org/W2152805652	https://mobilednajournal.biomedcentral.com/counter/pdf/10.1186/1759-8753-3-5	English	null	How does selfing affect the dynamics of selfish transposable elements?	Mobile DNA	2,012	cc-by	7,901	Abstract Background: Many theoretical models predicting the dynamics of transposable elements (TEs) in genomes, populations, and species have already been proposed. However, most of them only focus on populations of sexual diploid individuals, and TE dynamics in populations partly composed by autogamous individuals remains poorly investigated. To estimate the impact of selfing on TE dynamics, the short- and long-term evolution of TEs was simulated in outcrossing populations with various proportions of selfing individuals. Results: Selfing has a deep impact on TE dynamics: the higher the selfing rate, the lower the probability of invasion. Already known non-equilibrium dynamics (complete loss, domestication, cyclical invasion of TEs) can all be described whatever the mating system. However, their pattern and their respective frequencies greatly depend on the selfing rate. For instance, in cyclical dynamics resulting from interactions between autonomous and non- autonomous copies, cycles are faster when the selfing rate increases. Interestingly, an abrupt change in the mating system from sexuality to complete asexuality leads to the loss of all the elements over a few hundred generations. In general, for intermediate selfing rates, the transposition activity remains maintained. Conclusions: Our theoretical results evidence that a clear and systematic contrast in TE content according to the mating system is expected, with a smooth transition for intermediate selfing rates. Several parameters impact the TE copy number, and all dynamics described in allogamous populations can be also observed in partly autogamous species. This study thus provides new insights to understand the complex signal from empirical comparison of closely related species with different mating systems. The complex co-evolutionary dynamics between TEs and genomes was also enlightened by several population genetic models (see [12] for a review). Usually, the mod- els assume that the number of copies grows to an equi- librium after a burst of amplification, this equilibrium being achieved when replicative transposition balances natural selection against insertions [13,14]. However, independent biological data on TE activity from several elements including Alu elements in humans [15], copia elements, and LTR retrotranposons in plants [16,17] or in Drosophila [18], as well as Class II elements [19], clearly show that the transposition rate is not constant. Indeed, transposition bursts are frequently observed. Changes in activity can be due to epigenetic factors, such as modified methylation patterns [20], and explain why some natural populations have variable TE activity. © 2012 Boutin et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. How does selfing affect the dynamics of selfish transposable elements? Thibaud S Boutin1,2, Arnaud Le Rouzic1 and Pierre Capy1,2* Abstract In any case, if an equilibrium could be reached on a short scale, maintaining such a status over long evolu- tionary stages seems unrealistic. Investigations based on * Correspondence: pierre.capy@legs.cnrs-gif.fr 1LEGS, CNRS UPR9034, IDEEV FR3284, Avenue de la terrasse, Bat 13, 91198 Gif-sur-Yvette, France Full list of author information is available at the end of the article Boutin et al. Mobile DNA 2012, 3:5 http://www.mobilednajournal.com/content/3/1/5 Background Transposable elements (TEs) are now considered as major components of genomes. These DNA sequences are able to invade, to multiply, and to spread across spe- cies despite their deleterious impact (on average) on their host. The capacity to generate structural and func- tional variability illustrates their impact on genome evo- lution [1-3]. If their ubiquity and their features can be only explained by selfishness [4,5], they are also able to enhance genome evolvability and to promote molecular exaptation. Their influence on genome evolution is sup- ported by an increasing number of observations [6,7], including gene regulation via epigenetics landmarks [8-10], or the emergence of the V(D)J immune system in mammals [11]. * Correspondence: pierre.capy@legs.cnrs-gif.fr 1LEGS, CNRS UPR9034, IDEEV FR3284, Avenue de la terrasse, Bat 13, 91198 Gif-sur-Yvette, France Full list of author information is available at the end of the article Boutin et al. Mobile DNA 2012, 3:5 http://www.mobilednajournal.com/content/3/1/5 Page 2 of 9 Boutin et al. Mobile DNA 2012, 3:5 http://www.mobilednajournal.com/content/3/1/5 Page 2 of 9 probability Z, and reproduces proportionally to its fit- ness w (see ‘Methods’ section, parameters of the model are summarized in Table 1). The selective impact si of each TE insertion can vary from highly deleterious to advantageous according to the distribution probability of selective impacts (see [22]). A copy duplicates through replicative transposition (rate u), but can also be elimi- nated by deletion (e.g., excision with no reinsertion, or after ectopic recombination between two copies in direct orientation) with a rate v. Mutations occurring at rate m decrease the copy activity. Copies with no activ- ity are non-autonomous, they are still trans-mobilizable; thanks to the transposition machinery produced by autonomous copies. In the simulations, only the ele- ments able to transpose (i.e., autonomous and non- autonomous elements) were explicitly modeled, immo- bile relic copies are considered as deleted. simulations assuming random mating and accounting for parameters like copy activity, copy impact on host fitness, and mutation on TE activity [21,22], show that non-equilibrium dynamics are likely and predict cycles of re-amplification similar to those observed in ‘Lotka - Volterra’ prey-predator dynamics [23]. ‘Molecular domestication’ events (TE insertions reaching fixation in the host population because of their positive impact on their fitness) can also be reproduced in silico with mod- els allowing a low rate of beneficial insertions [22]. Population invasion A l p A population was considered as invaded by the TE family when one copy per individual on average is pre- sent 100 generations after the initial introduction. Copies spread from their single ancestor during an initial burst of transposition, when most of the copies are highly active. This invasion is expected to be facili- tated by outcrossing [33]: the more selfing, the lower the probability of invasion. This is illustrated in Figure 1 for two population sizes. In addition, the distribution of copy number greatly varies according to the selfing rate. After 100 generations, when Z = 0.9, most individuals from successfully invaded populations have a single copy, while when Z = 0 (i.e., random mating), the aver- age copy number per individual is much higher. Exam- ples of the copy number distribution per individual after 100 generations are provided in Additional file 1. It is generally assumed that selfing has a strong func- tional impact on genomes and on their components, including selfish elements such as TEs [30]. During the last decade, a few models focusing on the impact of self- ing on TE dynamics were developed [31,32]. As expected, they predict a decline of the probability of TE invasion due to reduced exchanges between selfers: the selfish invasion strategy of TEs relies on being trans- mitted to the offspring at higher frequency than expected under Mendelian segregation, and becomes ineffective in absence of outcrossing. So far, little is known about TE dynamics in partial selfers beyond the unrealistic assumption of the transposition-selection equilibrium [32]. Here, we propose to extend the existing theoretical corpus in two directions: (i) assessing the impact of intermediate selfing rates on non-equilibrium TE dynamics, and (ii) evaluating the influence of the mating system on the evolutionary properties of TEs and of their host species. To this end, the fate of TEs in popu- lations with various selfing rates was followed in silico from the very first steps of the genome invasion to the long-term evolution (10,000 generations). In addition, we simulated the consequences of switching from an outcrossing to a self-fertilizing population. Population invasion A l Table 1 Parameters of the model Parameter Definition N Number of individuals in the population Z Probability to reproduce by self-fertilization w Fitness of an individual n Number of TE copies in the genome si Effect of insertion i on fitness μs Average effect of TE insertions on fitness ss Standard deviation of the effect on fitness Ps > 0 Proportion of TE insertions that improve the fitness u Replicative transposition rate v Deletion rate ai Activity of a copy i (ability to produce the transposition machinery) m Mutation rate (toward a less active copy) sm Standard deviation of the effect of a mutation on activity Table 1 Parameters of the model Parameter Definition Background By analogy, it was recently proposed that genomes could be seen as ecosystems in which TEs families are co-evol- ving species [24-26]. TEs being sexually transmitted parasites, almost all models assume random mating. Surprisingly, little is known on how TE copies behave in completely selfing species. Yet, hermaphroditism is widespread in the living world, particularly in plants, but also in about one-third of animal species [27]. Self-crossing enables individuals to reproduce on their own by self-fertilization; it gener- ally induces inbreeding depression or decrease in genetic variation, but can also be advantageous when mates are scarce due to the rarefaction of the partners [28,29]. Results Initially, each population has only one full active copy. The probability of adaptive insertion was fixed to Ps > 0 = 0.1%, this parameter does not affect the pattern (data not shown). The Y -axis is the percentage of invasion achieved (the frequency of the simulations with at least 1 copy per individual on average at generation 100), and the X-axis is the selfing rate (Z). Two population sizes were tested (N = 100 and N = 1000). The error bars represent the maximum-likelihood 95% confidence intervals, calculated assuming a binomial distribution of the invasion probability. Figure 1 Effect of selfing rate on the probability of invasion. Initially, each population has only one full active copy. The probability of adaptive insertion was fixed to Ps > 0 = 0.1%, this parameter does not affect the pattern (data not shown). The Y -axis is the percentage of invasion achieved (the frequency of the simulations with at least 1 copy per individual on average at generation 100), and the X-axis is the selfing rate (Z). Two population sizes were tested (N = 100 and N = 1000). The error bars represent the maximum-likelihood 95% confidence intervals, calculated assuming a binomial distribution of the invasion probability. On one hand, when the selfing rate is high (Z ≥0.9), the invasion frequency is very small after 100 genera- tions. However, in successfully invaded populations, active copies can be maintained for a while (up to 1,000 generations, data not shown) with apparent bursts in the 100 first generations (Additional file 2). These bursts do not correspond to an increase in copy number in the whole population, they are rather due to the accumula- tion of homozygous copies in some lineages until selec- tion removes them. The invasion of a TE family is a stochastic process and is partly driven by genetic drift, but contrary to the well-known influence of population size on the fixation probability of deleterious mutations, the invasion frequency of TEs is virtually insensitive to the number of individuals. This effect, already documen- ted in [34], is due to the deterministic aspect of TE invasion above a copy number threshold in the popula- tion, which does not depend directly on the population size. Results Moreover, there is no or little effect of the prob- ability of adaptive insertion during the first steps of invasion whatever the selfing rate: such rare insertions are unlikely to happen soon enough to influence the early dynamics (data not shown). Figure 2A and 2B show that the frequency of the cycles increases with the selfing rate in large populations (N = 10,000). This phenomenon is clearer when cycles are maintained over a long period, e.g., when the prob- ability of adaptive insertion is relatively low (Figure 2A), but it is still detectable when this probability is high (Ps > 0 = 0.1%) before the fixation of adaptive insertions (Figure 2B). When the selfing rate increases, the fixation of adaptive insertion is delayed, while the copy number is about the same at the top of the first peak (compari- son between Z = 0 or Z = 0.5). As the probability of adaptive insertions increases, advantageous elements can be fixed more often (Figure 2B), provided that their positive effect is strong enough to overcome genetic drift. These beneficial insertions do not need to maintain their transposition activity to sur- vive, and they eventually become inactive. The presence of a substantial amount on non-autonomous copies that are not purged by natural selection or deletions prevents the reinvasion of autonomous copies as observed in the cyclic dynamics, and transposition activity tends to be lost. At the end of this ‘molecular domestication’ pro- cess, other neutral and deleterious copies are progres- sively eliminated from the genome by selection, deletion, and drift, leading to the maintenance of few adaptive insertions that are totally inactive (a = 0). Results TE dynamics from the first generation after introduction to their long-term evolution were explored by indivi- dual-based simulations. In a population of N diploid hermaphrodites, each parent can be a selfer with Boutin et al. Mobile DNA 2012, 3:5 http://www.mobilednajournal.com/content/3/1/5 Page 3 of 9 Boutin et al. Mobile DNA 2012, 3:5 http://www.mobilednajournal.com/content/3/1/5 Figure 1 Effect of selfing rate on the probability of invasion. Initially, each population has only one full active copy. The probability of adaptive insertion was fixed to Ps > 0 = 0.1%, this parameter does not affect the pattern (data not shown). The Y -axis is the percentage of invasion achieved (the frequency of the simulations with at least 1 copy per individual on average at generation 100), and the X-axis is the selfing rate (Z). Two population sizes were tested (N = 100 and N = 1000). The error bars represent the maximum-likelihood 95% confidence intervals, calculated assuming a binomial distribution of the invasion probability. growing number of deleterious insertions [34]. In our model, the decrease in copy number is due to their elimination by natural selection and to a reduction of the general activity of transposition (regulation). After the initial burst, TE dynamics may go through cycles of re-invasion, molecular domestication, or loss (Figure 2), in the same way as for random mating populations [22]. Stable transposition-selection equilibria were not observed with the tested set of parameters. The accumulation of mutations leads to the emer- gence of low-activity and non-autonomous copies. The reduction of the global transposition activity is due to the decline of the production of efficient transposition machinery. As long as the average activity in a genome is non-null, both autonomous and non-autonomous copies can transpose. In such a situation, the non-auton- omous copies parasitize the autonomous elements, and cycles of decline/re-invasion can be maintained up to the end of the simulation (10,000 generations). Previous analysis has shown that these cycles can be explained by interactions between autonomous and non-autonomous copies, similarly to prey-predator (or host-parasite) interactions [22,25]. Autonomous copies are equivalent to hosts, as they are able to spread and maintain by themselves, but non-autonomous copies act as parasites: they can transpose only when autonomous copies are present in the same genome, forcing them to share the transposition machinery. Figure 1 Effect of selfing rate on the probability of invasion. Figure 1 Effect of selfing rate on the probability of invasion. Beyond the initial burst In the absence of a copy number control mechanism, the initial invasion may drive the host population to extinction, owing to the genetic load generated by a Boutin et al. Mobile DNA 2012, 3:5 http://www.mobilednajournal.com/content/3/1/5 Boutin et al. Mobile DNA 2012, 3:5 Page 4 of 9 Page 4 of 9 http://www.mobilednajournal.com/content/3/1/5 Figure 2 Effect of selfing on the long-term dynamics after invasion of the initial copy. Representative dynamics of the long term invasion (10,000 generations from 1 initial copy within large and small populations (A, B : N = 10,000 and C : N = 100) and different selfing rates Z = 0, Z = 0.5, Z = 0.75). A single simulation is presented for each case. The level of gray is proportional to the frequency at which each copy number is present in the population. The initial burst of transposition is followed by a decrease in the number of TEs due to the emergence of non- autonomous copies. Active transposition can persist when the probability of adaptive insertion is low Ps > 0 = 0.01% (A) and ‘domesticated’ copies can be fixed (B) when this probability is high (Ps > 0 = 0.1%). The frequency of cycles tends to increase with the selfing rate for any value of the probability of adaptive insertion. In opposition to large population size, the loss of the TE family is systematic in small populations (C). Figure 2 Effect of selfing on the long-term dynamics after invasion of the initial copy. Representative dynamics of the long term invasion (10,000 generations from 1 initial copy within large and small populations (A, B : N = 10,000 and C : N = 100) and different selfing rates Z = 0, Z = 0.5, Z = 0.75). A single simulation is presented for each case. The level of gray is proportional to the frequency at which each copy number is present in the population. The initial burst of transposition is followed by a decrease in the number of TEs due to the emergence of non- autonomous copies. Active transposition can persist when the probability of adaptive insertion is low Ps > 0 = 0.01% (A) and ‘domesticated’ copies can be fixed (B) when this probability is high (Ps > 0 = 0.1%). The frequency of cycles tends to increase with the selfing rate for any value of the probability of adaptive insertion. Beyond the initial burst In opposition to large population size, the loss of the TE family is systematic in small populations (C). improved purging efficiency in large population-size spe- cies [35,36]. Molecular domestication is consistently observed when the probability of adaptive insertion is high (Ps > 0 = 0.5%, data not shown). For intermediate values, the proportion of domestication increases with the probabil- ity of adaptive insertions, and decreases with the selfing rate. Selfing populations also tend to fix less adaptive insertions on average. Transition to selfing Figure 3 Effect of switching from outcrossing to self-fertilization. The two figures represent typical dynamics of a TE family simulated within a selfing population which has outcrosser ancestors. The initial population was originally a strict outcrossing population (N = 2000), already invaded by TEs, in which an abrupt transition to self-fertilization is performed. The new population thus has a historical background with numerous copies, both autonomous and non-autonomous, and few domesticated insertions (Ps > 0 = 0.01%). Very rapidly after the transition to selfing (generation 0), almost all insertions become homozygous (even number of copies in all lineages). TEs are then systematically lost after few hundred generations, including adaptive insertions. hundred generations, but active TEs always disappear eventually. The model on which our simulations are based is ori- ginal, and provides significant novelties compared to the existing literature: (i) the reproduction mode is flexible and allows partial selfing; (ii) TEs are affected by muta- tion, which generates inactive copies, similar to those found in sequenced genomes; (iii) transposition regula- tion occurs spontaneously as a consequence of copy polymorphism (the transposition rate depends on the ratio between autonomous and non-autonomous copies); and (iv) simulations are initiated from a single copy, instead of TEs artificially distributed among indivi- duals. In previous study, dealing with diploid random mating populations (e.g., [22]), short- and long-term evolution was generally disconnected. Indeed, long-term simulations were initiated with many TE copies (often one copy per individual on average), so that the element cannot be immediately lost by drift. In this study, these two stages of the TEs dynamics are not disconnected and simulations were followed from the initial introduc- tion of a single copy up to 10,000 generations. Our results being similar to those previously published for the same sets of parameters, disconnection of short- and long-term evolution seems to have little effect on the dynamics observed, which strengthens previously pub- lished work. Transition to selfing A family of TEs could invade a population with a very high selfing rate but has low chances to reach a high copy number if they are deleterious. However, an interesting situation can be found when a TE-rich outcrossing population switches to selfing. This was simulated with an abrupt transition in the mating sys- tem after the initial transposition burst. A few genera- tions after the shift, the selfing population is de facto subdivided into competing lineages with slightly dif- ferent TE composition (Figure 3). The lineages with the lowest copy number can persist longer. Finally, all TEs are lost after a few thousand generations. The loss remains a stochastic process, and in small popu- lations (N = 100), the mean number of TEs can increase stochastically within lineages for several The analysis of the joint effect of selfing and popula- tion size is also interesting. Indeed, while selfing has a clear effect on the long-term evolutionary dynamics of TEs in large populations (N = 10, 000), its impact is lowered by genetic drift in small populations (N = 100), in which TEs are generally lost after few thousand gen- erations whatever the mating system (Figure 2C). Indeed, genetic drift also affects TE sequences, and increases the chances to lose all active copies in the population when the average copy number is low. This observation confirms previous findings [22], and high- lights a mechanism acting in a direction opposite to the Page 5 of 9 Boutin et al. Mobile DNA 2012, 3:5 Boutin et al. Mobile DNA 2012, 3:5 http://www.mobilednajournal.com/content/3/1/5 http://www.mobilednajournal.com/content/3/1/5 Figure 3 Effect of switching from outcrossing to self-fertilization. The two figures represent typical dynamics of a TE family simulated within a selfing population which has outcrosser ancestors. The initial population was originally a strict outcrossing population (N = 2000), already invaded by TEs, in which an abrupt transition to self-fertilization is performed. The new population thus has a historical background with numerous copies, both autonomous and non-autonomous, and few domesticated insertions (Ps > 0 = 0.01%). Very rapidly after the transition to selfing (generation 0), almost all insertions become homozygous (even number of copies in all lineages). TEs are then systematically lost after few hundred generations, including adaptive insertions. Model interpretation These figures are quite high com- pared to empirically measured rates (e.g., [37]), but dynamical parameters are known to scale the dynamics of the process without affecting the pattern [22,38] (i.e., decreasing the transposition rate slows down the dynamics by the same order of magnitude). Interest- ingly, the reproduction mode itself might influence transposition parameters. For instance, it is likely that ectopic recombinations (and thus opportunities for TE deletion) are less frequent in selfers. Our model did not consider such complex mechanisms, but this evidences how simplistic our understanding of TE dynamics remains. Several non-exclusive mechanisms have been proposed to explain the discrepancies between theory and obser- vations. Recurrent horizontal transfers of active elements in self-fertilizing or asexual species could partly explain how genomic parasites are maintained in spite of unfa- vorable mating systems. Ancient asexual species are rare, as they are expected to be outcompeted by more evolvable sexual relatives, but an interesting exception can be found in Bdelloid Rotifers, in which asexual lineages are maintained for tens of millions years par- tially because of massive horizontal genes transfers com- pensating the absence of sexual genetic exchanges [47]. TEs were part of this alien DNA material, and some of them have slightly been amplified since these animals display a low content in potentially active repeated TE sequences [48]. In fact, TEs are more likely to be later- ally transferred than any other genetics elements because they are fully equipped for genomic mobility. Such a phenomenon seems to have an important role in the spread of TEs between Eucaryotes [49], and horizon- tally transmitted TEs have been found in both animals [50-53] and plants such as the genus Oryza that includes selfing species [54,55]. An interesting observation is that amplification-loss cycles, suggested by genomic data and described theore- tically in [23], can be also observed in autogamy, but with a higher frequency (Figure 2). This accelerated evo- lution can be attributed to the non-random distribution of TE copies among individuals in partial selfers: as illu- strated in Additional file 1, the variance in copy number (and thus the variance in fitness) increases with selfing (the distribution is wider than the expected Poisson under random mating), thus increasing the efficiency of natural selection and accelerating the elimination of TEs. Model interpretation Our results confirm that the mating system has a deep impact on TE dynamics. Most of the main dynamics already described in random-mating populations were found. Depending on the properties of the TE family and of the host population (including population size and selfing rate), the initial burst can be followed by the loss of all copies, by the fixation of a few beneficial insertions (molecular domestication), or by cycles of reinvasions. As in random-mating populations, the population size has little influence on the first steps of TE invasion [34]. However, the TE dynamics in small populations are less stable, and lead more often to the loss of TE activity. Partial selfing affects the outcome of the model, and as a general rule, TEs are less likely to maintain (either as active parasites or as ‘domesti- cated’ insertions) in selfing populations. The probabil- ity of invasion decreases with the selfing rate, but when it occurs, the initial burst of transposition remains comparable to outcrossers (except for a slight delay), followed by one of the typical dynamics (cycles, loss, or domestication), depending on the mutation rate and the frequency of beneficial insertions. Thereby, the efficiency of TEs as genomic parasites decreases regularly with the selfing rate, but rare suc- cessful invasions can still occur even in populations with 90% selfers. For modeling purposes, the system had to be simpli- fied. For instance, only mobile copies are tracked, and the loss of mobility is assimilated to deletion. As a result, some populations may quickly lose all of their TEs in the simulations, while their genome may still contain relic copies. Our purpose being to investigate Boutin et al. Mobile DNA 2012, 3:5 http://www.mobilednajournal.com/content/3/1/5 Page 6 of 9 Boutin et al. Mobile DNA 2012, 3:5 http://www.mobilednajournal.com/content/3/1/5 whose genome is made of about 80% of TEs [42-44]. In addition, the comparison of the distribution of some TE families in Arabidopsis genus remains incon- clusive. As expected, the self-fertilizing Arabidopsis thaliana has less polymorphic TE insertion sites than its outcrossing sister-species Arabidopsis lyrata, and also displays less TE copies in total [45,46]. However, the contrast remains much weaker than predicted by theory. the influence of sexual reproduction, both transposition and deletion rates were maintained constant (u = 0.1, v = 0.001, respectively). Model interpretation The number of ‘domesticated’ copies is also lower in selfers: in absence of recombination (or in exclusively homozygous individuals after several generations of self- ing), adaptive insertions always remain associated to deleterious elements, which prevents their spread, according to the Hill-Robertson effect [39,40]. Another explanation for the weak contrast between selfing and outcrossing genomes lies in the nature of the deleterious effect of repeated sequences [56]. While heterozygous insertions are likely to be involved in potentially catastrophic ectopic recombinations, homo- zygous copies may be closer to neutrality. TEs could thus be protected from natural selection in mostly homozygote selfers [31,32,57]. If biologically important, this phenomenon should be detected by comparing the TE-content of autosomes versus X chromosomes (hemi- zygous in males), expected to display less copies. Unfor- tunately, X chromosomes also experience lower recombination rates, which has the opposite effect [58]. Here again, the evidence is thus equivocal: homozygosity actually reduces ectopic recombinations in Drosophila melanogaster [59], but the X chromosome of mice and humans is particularly rich in L 1 TE sequences [60,61]. In addition, other mechanisms such as X inactivation may interfere with the expected pattern. The recent comparison between the genomes of A. lyrata and A. thaliana, in which the proportion of TE is lower, does not suggest selection against ectopic recombinations, Our results also highlight the consequences of a shift from outcrossing to selfing, a frequent evolutionary transition among hermaphrodite flowering plants [41]. Such an event increases homozygosity, and subsequently leads to the formation of divergent lineages with variable TE copy number. As a consequence, the variability in TE content (and accordingly, the variance in fitness) is much larger between lineages than within lineages, explaining the rapid loss of TEs in the simulations. Nevertheless, this model does not explain why some selfing species still maintain active TEs in their genomes. Transposable elements The model used for the simulations is individual-based and stochastic, inspired from the framework described in [22]. Within a genome, each copy can be duplicated with a maximum rate of transposition u = 0.1, and may be excised with a rate v = 0.001. In addition, each copy is defined by its own activity ai, i.e., the capacity of the Host population Th l i The population is based on a constant number of N diploid hermaphrodites with a genome made up of 5 chromosomes of 100 cM each. Throughout the non- overlapping generations, individuals are able to repro- duce proportionally to their fitness w. The fitness is defined by w = 1 + Σ si, where si ∼N(μs = −0.01, σs) is the selective impact of the insertion i. The variance σ 2 s of this distribution conditions the probability of adaptive insertions, Psi > 0 (see [22] for more details). Z being the selfing rate, each offspring has a probability Z of having a unique parent (which provide two gametes) and a probability 1 - Z of having two parents randomly drawn in the population, each of them providing one gamete. Gametes are generated by drawing the number of recombinations in a Poisson distribution, spreading the recombinations out uniformly in the genome, and copying randomly one of the chromosomes until a recombination is met, at which point the copy continues from the homologous chromosome. The recombination rate is the same in selfers than in outcrossers. Conclusion In this article, our intention was to determine, using a realistic individual-based model, how TEs evolve in spe- cies with different selfing rates. We confirmed the well- known result that selfing tends to reduce TE content, by preventing the invasion, and by decreasing the number of copies. The dynamics described in outcrossing species are also found in partly self-fertilizing ones, but the details of the invasion process may be significantly affected. Interestingly, TE invasion remains possible even at high selfing rates. Our models also suggest that auto- magous populations fail at fixing rare beneficial inser- tions. Because selfing lessens by several ways the potentially beneficial mutagenic impact of TE copies, our results strengthen the catastrophic evolutionary conse- quences of the loss of sexual reproduction. If the mating system changes from outcrossing to selfing, the genome content in TE is strongly affected. However, sequenced genomes show that species classified as self-fertilizing often maintain a significant amount of active TEs, which contradicts the theory. Among other explanations, this observation suggests that species classified as selfers might experience rare outcrossing events, hindering the eradication of their genomic parasites. Moreover, hori- zontal transfers can play an important role in the mainte- nance of TE-mediated evolvability in selfing lineages. Simulations f In most of the simulations, a single fully active copy was introduced in one of the individuals of the population. The dynamics of the element was followed from genera- tion 1 to 10,000. To analyze the transition from cross fertilization to strict self-fertilization, the set of para- meters used was one of those which led to a cyclic dynamics (see legend of Figure 2). Then, the mating sys- tem was switched (from Z = 0 to Z = 0.9 or Z = 1) at the top of the first wave. Boutin et al. Mobile DNA 2012, 3:5 http://www.mobilednajournal.com/content/3/1/5 Boutin et al. Mobile DNA 2012, 3:5 http://www.mobilednajournal.com/content/3/1/5 Boutin et al. Mobile DNA 2012, 3:5 http://www.mobilednajournal.com/content/3/1/5 element i to produce a functional transposition machin- ery. Thus, the effective rate of transposition is u · ¯a, where ¯a is the mean activity of all (n) TE copies of a genome defined as ¯a = 1 n ai. Mutation of copy activity occurs with a rate m = 0.001. The activity of a mutated element is a′ i = ai −\|δ\|, where δ ∼N(μ = 0, σm = 1). On average, it thus takes one or two mutations to turn a fully active element (ai = 1) to an inactive copy (ai = 0), which is still trans-mobilizable. Here, we did not con- sider the possibility for an element to lose its mobility in presence of transposition machinery: mutational events leading to immobilization are associated with deletions (the element disappears from the pool of observable copies). but rather supports a selection against negative impact of insertions on surrounding genes [62]. Whatever is the focus of the natural selection, its effi- ciency depends directly on the effective host population size (Ne). In a full selfing population, Ne is reduced by half [63]. It was recently argued that population process could be a key factor in the patterning of TE distribu- tion in plants [64]. The rare empirical population studies suggest a demographic impact (more copies in bottle- necked populations) among plants like Arabidopsis [45,65], but also in animals like nematodes [66]. Finally, observed differences might be due to non-deterministic factors: non-equilibrium dynamics are common in our simulations, and two species considered at two different stages of their cyclical dynamics may differ considerably in their TE content. In general, genetic drift may also play an important role in the persistence of active copies, as shown here and previously [22]. Confrontation with data Models of TE dynamics predict that selfish genetic ele- ments should be unable to invade the genome of auto- gamous or asexual host species. We have also shown that TE copies should rapidly be lost after a shift from outcrossing to selfing. However, this prediction is not strongly supported by empirical data. Indeed, TEs are a major component of plant genomes, including selfing species such as the common wheat Triticum aestivum, Page 7 of 9 Page 7 of 9 Author details 1 23. Le Rouzic A, Capy P: Population genetics models of competition between transposable element subfamilies. Genetics 2006, 174(2):785-793. 1LEGS, CNRS UPR9034, IDEEV FR3284, Avenue de la terrasse, Bat 13, 91198 Gif-sur-Yvette, France. 2Université Paris-Sud, Orsay, France. 1LEGS, CNRS UPR9034, IDEEV FR3284, Avenue de la terrasse, Bat 13, 91198 Gif-sur-Yvette, France. 2Université Paris-Sud, Orsay, France. 24. Brookfield J: The ecology of the genome - mobile DNA elements and their hosts. Nat Rev Genet 2005, 6(2):128-136. their hosts. Nat Rev Genet 2005, 6(2):128-136. Abbreviations 18. Vieira C, Biémont C: Transposition rate of the 412 retrotransposable element is independent of copy number in natural populations of Drosophila simulans. Mol Biol Evol 1997, 14(2):185-188. LTR: long terminal repeats; TE: transposable element. References 1. McDonald J: Evolution and consequences of transposable elements. Curr Opin Genet Dev 1993, 3:855-864. 1. McDonald J: Evolution and consequences of transposable elements. Curr Opin Genet Dev 1993, 3:855-864. 31. Wright S, Schoen D: Transposon dynamics and the breeding system. Genetica 1999, 107:139-148. 2. Capy P, Gasperi G, Biémont C, Bazin C: Stress and transposable elements: co-evolution or useful parasites? Heredity 2000, 85:101-106. 2. Capy P, Gasperi G, Biémont C, Bazin C: Stress and transposable elements: co-evolution or useful parasites? Heredity 2000, 85:101-106. 32. Morgan MT: Transposable element number in mixed mating populations. Genet Res 2001, 77(3):261-275. 3. Hua-Van A, Le Rouzic A, Boutin TS, Filee J, Capy P: The struggle for life of the genome’s selfish architects. Biol Direct 2011, 6:19. 3. Hua-Van A, Le Rouzic A, Boutin TS, Filee J, Capy P: The struggle for life of the genome’s selfish architects. Biol Direct 2011, 6:19. 33. Hickey DA: Selfish DNA: a sexually-transmitted nuclear parasite. Genetics 1982, 101(3-4):519-531. 4. Orgel LE, Crick FHC: Selfish DNA: the ultimate parasite. Nature 1980, 284:604-607. 4. Orgel LE, Crick FHC: Selfish DNA: the ultimate parasite. Nature 1980, 284:604-607. 34. Le Rouzic A, Capy P: The first steps of transposable elements invasion: parasitic strategy vs. genetic drift. Genetics 2005, 169:1033-1043. 5. Doolittle WF, Sapienza C: Selfish genes, the phenotype paradigm and genome evolution. Nature 1980, 284(5757):601-603. 5. Doolittle WF, Sapienza C: Selfish genes, the phenotype paradigm and genome evolution. Nature 1980, 284(5757):601-603. 35. Lynch M, Conery JS: The origins of genome complexity. Science 2003, 302(5649):1401-1404. 6. Volff JN: Turning junk into gold: domestication of transposable elements and the creation of new genes in eukaryotes. Bioessays 2006, 28(9):913-922. 36. Lynch M: The Origins of Genome Architecture. 1 edition. Sunderland, MA: Sinauer Associates Inc; 2007. 37. Nuzhdin SV, Mackay TFC: The genomic rate of transposable element movement in Drosophila melanogaster. Mol Biol Evol 1995, 12:180-181. 7. Sinzelle L, Izsvak Z, Ivics Z: Molecular domestication of transposable elements: from detrimental parasites to useful host genes. Cell Mol Life Sci 2009, 66(6):1073-1093. 7. Sinzelle L, Izsvak Z, Ivics Z: Molecular domestication of transposable elements: from detrimental parasites to useful host genes. Cell Mol Life Sci 2009, 66(6):1073-1093. 38. Dolgin ES, Charlesworth B: The fate of transposable elem populations. Genetics 2006, 174(2):817-827. 8. Slotkin RK, Martienssen R: Transposable elements and the epigenetic regulation of the genome. Nat Rev Genet 2007, 8(4):272-285. 39. Authors’ contributions 25. Le Rouzic A, Dupas S, Capy P: Genome ecosystem and transposable elements species. Gene 2007, 390(1-2, Sp. Iss. SI):214-220. TSB and PC designed the research. TSB and ALR carried the research out and analyzed the data. ALR conceived the core software. TSB, ALR, and PC interpreted the data and wrote the article. All the authors read and approved the final manuscript. 26. Venner S, Feschotte C, Biemont C: Dynamics of transposable elements: towards a community ecology of the genome. Trends Genet 2009, 25(7):317-323. 27. Jarne P, Auld JR: Animals mix it up too: the distribution of self- fertilization among hermaphroditic animals. Evolution 2006, 60(9):1816-1824. Acknowledgements 19. Quesneville H, Nouaud D, Anxolabehere D: P elements and MITE relatives in the whole genome sequence of Anopheles gambiae. BMC Genom 2006, 7:214. The authors thank Tatiana Giraud for helpful discussions, and the two anonymous reviewers for detailed comments and suggestions. This study was financially supported by the University Paris-Sud 11 and the CNRS through the following institutions: UPR9034, IFR 115: Génomes, Transcriptomes, Protéomes; GDR 2157: Les éléments transposables: du génome aux populations; PPF Bioinforamtique et Biomathématiques de l’Université Paris-Sud 11. ALR was supported by the European grant EVOLGA ERG-256507. 20. Tsukahara S, Kobayashi A, Kawabe A, Mathieu O, Miura A, Kakutani T: Bursts of retrotransposition reproduced in Arabidopsis. Nature 2009, 461(7262):423-426. 21. Kaplan N, Darden T, Langley CH: Evolution and extinction of transposable elements in mendelian populations. Genetics 1985, 109:459-480. 22. Le Rouzic A, Boutin TS, Capy P: Long-term evolution of transposable elements. Proc Natl Acad Sci USA 2007, 104(49):19375-19380. Additional material Additional file 1: Distribution of the copy number at generation 100. This figure shows the distribution of copy number per individual a Additional file 1: Distribution of the copy number at generation 100. This figure shows the distribution of copy number per individual at Page 8 of 9 Boutin et al. Mobile DNA 2012, 3:5 http://www.mobilednajournal.com/content/3/1/5 14. Capy P, Anxolabéhère D, Langin T: The strange phylogenies of transposable elements: are horizontal transfers the only explantation? Trends Genet 1994, 10:7-12. the 100th generation for different selfing rates in a small population (N = 100). Additional file 2: First step of invasion. This figure shows the dynamics of TEs invasion for different selfing rates in a small population (N = 100). A single simulation is presented for each case. The level of gray is proportional to the frequency at which each copy number is present in the population at each generation. 15. Xing J, Hedges D, Han K, Wang H, Cordaux R, Batzer M: Alu element mutation spectra: molecular clocks and the effect of DNA methylation. J Mol Biol 2004, 344(3):675-682. 16. Hawkins JS, Proulx SR, Rapp RA, Wendel JF: Rapid DNA loss as a counterbalance to genome expansion through retrotransposon proliferation in plants. Proc Natl Acad Sci USA 2009, 106(42):17811-17816. 17. Vitte C, Panaud O: Formation of solo-LTRs through unequal homologous recombination counterbalances amplifications of LTR retrotransposons in rice Oryza sativa L. Mol Biol Evol 2003, 20(4):528-540. Received: 7 November 2011 Accepted: 7 March 2012 Published: 7 March 2012 29. Ghiselin MT: The evolution of hermaphroditism among animals. Q Rev Biol 1969, 44(2):189-208. 30. Wright SI, Ness RW, Foxe JP, Barrett SCH: Genomic consequences of outcrossing and selfing in plants. Int J Plant Sci 2008, 169:105-118. Competing interests Th h d l h The authors declare that they have no competing interests. The authors declare that they have no competing interests. 28. Darwin C: Cross and Self-Fertilization in Plants London: Murray; 18 Received: 7 November 2011 Accepted: 7 March 2012 Published: 7 March 2012 Received: 7 November 2011 Accepted: 7 March 2012 Published: 7 March 2012 Boutin et al. Mobile DNA 2012, 3:5 http://www.mobilednajournal.com/content/3/1/5 Boutin et al. Mobile DNA 2012, 3:5 http://www.mobilednajournal.com/content/3/1/5 Page 9 of 9 44. Charles M, Belcram H, Just J, Huneau C, Viollet A, Couloux A, Segurens B, Carter M, Huteau V, Coriton O, Appels R, Samain S, Chalhoub B: Dynamics and differential proliferation of transposable elements during the evolution of the B and A genomes of wheat. Genetics 2008, 180(2):1071-1086. doi:10.1186/1759-8753-3-5 Cite this article as: Boutin et al.: How does selfing affect the dynamics of selfish transposable elements? Mobile DNA 2012 3:5. 45. Lockton S, Gaut BS: The evolution of transposable elements in natural populations of self-fertilizing Arabidopsis thaliana and its outcrossing relative Arabidopsis lyrata. BMC Evol Biol 2010, 10:10. 46. Hollister JD, Smith LM, Guo YL, Ott F, Weigel D, Gaut BS: Transposable elements and small RNAs contribute to gene expression divergence between Arabidopsis thaliana and Arabidopsis lyrata. Proc Natl Acad Sci USA 2011, 108(6):2322-2327. 47. Gladyshev EA, Meselson M, Arkhipova IR: Massive horizontal gene transfer in bdelloid rotifers. Science 2008, 320(5880):1210-1213. 48. Gladyshev EA, Arkhipova IR: Genome structure of bdelloid rotifers: shaped by asexuality or desiccation? J Hered 2010, 101(Suppl 1):S85-S93. 49. Schaack S, Gilbert C, Feschotte C: Promiscuous DNA: horizontal transfer of transposable elements and why it matters for eukaryotic evolution. Trends Ecol Evol 2010, 25(9):537-546. 50. Loreto ELS, Carareto CMA, Capy P: Revisiting horizontal transfer of transposable elements in Drosophila. Heredity 2008, 100(6):545-554. 51. Gilbert C, Pace JK, Feschotte C: Horizontal SPINning of transposons. Commun Integr Biol 2009, 2(2):117-119. 52. Gilbert C, Schaack S, Pace JK, Brindley PJ, Feschotte C: A role for host- parasite interactions in the horizontal transfer of transposons across phyla. Nature 2010, 464(7293):1347-1350. 53. Gilbert C, Hernandez SS, Flores-Benabib J, Smith EN, Feschotte C: Rampant horizontal transfer of SPIN transposons in squamate reptiles. Mol Biol Evol 2011, 29(2):503-515. 54. Diao X, Freeling M, Lisch D: Horizontal transfer of a plant transposon. PLoS Biol 2006, 4:e5. 55. Roulin A, Piegu B, Wing RA, Panaud O: Evidence of multiple horizontal transfers of the long terminal repeat retrotransposon RIRE1 within the genus Oryza. Plant J 2008, 53(6):950-959. y 56. Nuzhdin SV: Sure facts, speculations, and open questions about evolution of transposable elements. Genetica 1999, 107:129-137. 57. Charlesworth D, Charlesworth B: Transposable elements in inbreeding and outbreeding populations. Genetics 1995, 140:415-417. 58. Charlesworth B, Lapid A, Canada D: The distribution of transposable elements within and between chromosomes in a population of Drosophila melanogaster. I. References Hill WG, Robertson A: The effect of linkage on limits to artificial selection Genet Res 1966, 8(3):269-294. 9. Bossdorf O, Richards CL, Pigliucci M: Epigenetics for ecologists. Ecol Lett 2008, 11(2):106-115. 9. Bossdorf O, Richards CL, Pigliucci M: Epigenetics for ecologists. Ecol Lett 2008, 11(2):106-115. 40. Felsenstein J: The evolutionary advantage of recombination. Genetics 1974, 78(2):737-756. 10. Richards EJ: Population epigenetics. Curr Opin Genet Dev 2008, 18(2):221-226. 10. Richards EJ: Population epigenetics. Curr Opin Genet Dev 2008, 18(2):221-226. 41. Goldberg EE, Kohn JR, Lande R, Robertson KA, Smith SA, Igić B: Species selection maintains self-incompatibility. Science 2010, 330(6003):493-495. 11. Agrawal A, Eastman QM, Schatz DG: Transposition mediated by RAG1 and RAG2 and its implications for the evolution of the immune system. Nature 1998, 394:744-751. 42. Smith DB, Flavell RB: Characterisation of the wheat genome by renaturation kinetics. Chromosoma 1975, 50:223-242. 43. Sabot F, Guyot R, Wicker T, Chantret N, Laubin B, Chalhoub B, Leroy P, Sourdille P, Bernard M: Updating of transposable element annotations from large wheat genomic sequences reveals diverse activities and gene associations. Mol Genet Genom 2005, 274(2):119-130. 12. Le Rouzic A, Deceliere G: The models of population genetics of transposable elements. Genet Res 2005, 85:171-181. transposable elements. Genet Res 2005, 85:171-181. 13. Charlesworth B, Charlesworth D: The population dynamics of transposable elements. Genet Res 1983, 42(1):1-27. Boutin et al. Mobile DNA 2012, 3:5 http://www.mobilednajournal.com/content/3/1/5 Element frequencies and distribution. Genet Res 1992, 60(2):103-114. 59. Montgomery EA, Huang SM, Langley CH, Judd BH: Chromosome rearrangement by ectopic recombination in Drosophila melanogaster: genome structure and evolution. Genetics 1991, 129(4):1085-1098. 60. Boyle AL, Ballard SG, Ward DC: Differential distribution of long and short interspersed element sequences in the mouse genome: chromosome karyotyping by fluorescence in situ hybridization. Proc Natl Acad Sci USA 1990, 87(19):7757-7761. 61. Korenberg JR, Rykowski MC: Human genome organization: Alu, lines, and the molecular structure of metaphase chromosome bands. Cell 1988, 53(3):391-400. 62. Hu TT, Pattyn P, Bakker EG, Cao J, Cheng JF, Clark RM, Fahlgren N, Fawcett JA, Grimwood J, Gundlach H, Haberer G, Hollister JD, Ossowski S, Ottilar RP, Salamov AA, Schneeberger K, Spannagl M, Wang X, Yang L, Nasrallah ME, Bergelson J, Carrington JC, Gaut BS, Schmutz J, Mayer KFX, de Peer YV, Grigoriev IV, Nordborg M, Weigel D, Guo YL: The Arabidopsis lyrata genome sequence and the basis of rapid genome size change. Nat Genet 2011, 43(5):476-481. Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: Submit your next manuscript to BioMed Central and take full advantage of: 63. Pollak E: On the theory of partially inbreeding finite populations. I. Partial selfing. Genetics 1987, 117(2):353-360. 63. Pollak E: On the theory of partially inbreeding finite populations. I. Partial selfing. Genetics 1987, 117(2):353-360. 64. Tenaillon MI, Hollister JD, Gaut BS: A triptych of the evolution of plant transposable elements. Trends Plant Sci 2010, 15(8):471-478. g 64. Tenaillon MI, Hollister JD, Gaut BS: A triptych of the evolution of plant transposable elements. Trends Plant Sci 2010, 15(8):471-478. 65. Lockton S, Ross-Ibarra J, Gaut BS: Demography and weak selection drive patterns of transposable element diversity in natural populations of Arabidopsis lyrata. Proc Natl Acad Sci USA 2008, 105(37):13965-13970. 66. Dolgin ES, Charlesworth B, Cutter AD: Population frequencies of transposable elements in selfing and outcrossing Caenorhabditis nematodes. Genet Res 2008, 90(4):317-329. 66. Dolgin ES, Charlesworth B, Cutter AD: Population frequencies of transposable elements in selfing and outcrossing Caenorhabditis nematodes. Genet Res 2008, 90(4):317-329.
https://openalex.org/W2137872210	https://europepmc.org/articles/pmc3472176?pdf=render	English	null	Partitioning additive genetic variance into genomic and remaining polygenic components for complex traits in dairy cattle	BMC genomic data	2,012	cc-by	8,427	RESEARCH ARTICLE Open Access Abstract Background: Low cost genotyping of individuals using high density genomic markers were recently introduced as genomic selection in genetic improvement programs in dairy cattle. Most implementations of genomic selection only use marker information, in the models used for prediction of genetic merit. However, in other species it has been shown that only a fraction of the total genetic variance can be explained by markers. Using 5217 bulls in the Nordic Holstein population that were genotyped and had genetic evaluations based on progeny, we partitioned the total additive genetic variance into a genomic component explained by markers and a remaining component explained by familial relationships. The traits analyzed were production and fitness related traits in dairy cattle. Furthermore, we estimated the genomic variance that can be attributed to individual chromosomes and we illustrate methods that can predict the amount of additive genetic variance that can be explained by sets of markers with different density. Results: The amount of additive genetic variance that can be explained by markers was estimated by an analysis of the matrix of genomic relationships. For the traits in the analysis, most of the additive genetic variance can be explained by 44 K informative SNP markers. The same amount of variance can be attributed to individual chromosomes but surprisingly the relation between chromosomal variance and chromosome length was weak. In models including both genomic (marker) and familial (pedigree) effects most (on average 77.2%) of total additive genetic variance was explained by genomic effects while the remaining was explained by familial relationships. Results: The amount of additive genetic variance that can be explained by markers was estimated by an analysis of the matrix of genomic relationships. For the traits in the analysis, most of the additive genetic variance can be explained by 44 K informative SNP markers. The same amount of variance can be attributed to individual chromosomes but surprisingly the relation between chromosomal variance and chromosome length was weak. In explained by 44 K informative SNP markers. The same amount of variance can be attributed to individual chromosomes but surprisingly the relation between chromosomal variance and chromosome length was weak. In models including both genomic (marker) and familial (pedigree) effects most (on average 77.2%) of total additive genetic variance was explained by genomic effects while the remaining was explained by familial relationships. Abstract Conclusions: Most of the additive genetic variance for the traits in the Nordic Holstein population can be explained using 44 K informative SNP markers. By analyzing the genomic relationship matrix it is possible to predict the amount of additive genetic variance that can be explained by a reduced (or increased) set of markers. For the population analyzed the improvement of genomic prediction by increasing marker density beyond 44 K is limited. Keywords: Genomic variance, Polygenic variance, Chromosomes, Complex traits, Dairy cattle genetic markers selection (GS), can reduce the cost of running an intensive breeding program due to potential reductions in the num- ber of individuals tested for own or progeny performance, a shorter generation interval [2], and simultaneously greatly enhance the genetic gain from the program without a concomitant extra increase in the accumulation of inbreeding [3]. Currently, this has led to implementation of GS methods in many large scale commercial dairy cattle breeding programs [4-6]. Implementation in other animal and plant species are being initiated and is expected to in- crease considerably in the near future [7]. © 2012 Jensen et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 * Correspondence: Just.Jensen@agrsci.au.dk Department of Molecular Biology and Genetics, Centre for Quantitative Genetics and Genomics, Aarhus University, Research Centre Foulum, DK-8830, Tjele, Denmark Just Jensen, Guosheng Su and Per Madsen Just Jensen, Guosheng Su and Per Madsen Just Jensen, Guosheng Su and Per Madsen Background A typical model trait in humans is height that has been investigated in many studies. This is a trait mea- sured with an accuracy that is comparable to the accuracy of daughter yield deviations of dairy sires that are progeny-tested using large daughter groups. Height in humans is known to have a heritability around 0.8 [11]. In several well designed large scale studies with the use of very high density SNPs, numerous loci have been identi- fied that are significantly associated with human height, yet each of them typically only account for a very small fraction of total phenotypic variance. Collectively these loci are only responsible for up to 5% of the total pheno- typic variance in human height [11,12]. This has lead to discussions among human geneticists about the missing heritability of complex traits [13]. Suggested explanations have included dominance and epistatic interactions, geno- type by environment interactions and common epigenetic factors causing resemblance between relatives [14]. Recently, [19,20], proposed a one-step method combining marker based genomic relationships and pedigree-based relationships into a single relationship matrix. An import- ant factor in this procedure is that marker-based and pedigree-based relationship coefficients must be expressed on the same scale, i.e. the variance of the genomic and the classical additive genetic effects must refer to the same base population, and the proportion of total genetic variance explained by markers and the remaining genetic variance must be known. Currently implemented multistep proce- dures may also need to be on the same scale in order to en- sure derivation of optimum combined predictions. The purpose of this study was to evaluate the amount of additive genetic variation in production and fitness related traits in dairy cattle, to quantify the amount of additive genetic variation that can be explained using genomic markers with different density, and to quantify the amount of genomic variance that can be ascribed to individual chromosomes. The value of increasing the density of mar- ker information for predicting genetic merit was also assessed using subsets of available markers. g As mentioned above the total variance explained by pre- viously identified causal loci is usually only a small fraction of total genetic variance in the populations investigated. In GWAS very stringent significance thresholds are necessary due to the very large number of statistical tests that are con- ducted when searching the whole genome using high dens- ity SNP marker panels. Background Low cost genotyping of individuals or families using gen- omic markers with constantly increasing density is cur- rently being introduced in genetic improvement programs for agricultural animal and crop species. Use of dense gen- omic markers can increase the accuracy of predicting addi- tive genetic merit especially for selection of candidates that do not yet have own or progeny records [1]. Such applica- tion of dense genomic markers, usually called genomic Correspondence: Just.Jensen@agrsci.au.dk Department of Molecular Biology and Genetics, Centre for Quantitative Genetics and Genomics, Aarhus University, Research Centre Foulum, DK-8830, Tjele, Denmark The current industry standard in dairy cattle breeding is use of 50 K chips such as the Illumina Bovine SNP50 Page 2 of 9 Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 BeadChip [8] but other options, ranging from low density 3 K chips to high density 800 K chips, are also commer- cially available. Increasing the density of genetic markers is expected to increase the amount of genetic variance that can be explained by markers due to increased linkage dis- equilibrium between markers and causative loci [9]. A fur- ther step is the use of complete sequencing of individual genomes either based on direct sequence data or based on sequence data for selected individuals and imputing geno- types for animals that has been genotyped with lower mar- ker density If not all genetic variance can be explained by markers then, in order to ensure optimal predictions, the remaining genetic variance should be accounted for in other ways. A simple approach is to combine predicted breeding values based on genomic information with traditional breeding values based on pedigree using selection index theory [16]. An alternative method is to include both genomic and pedigree relationships in the analysis simultaneously. Previ- ous studies have report that a model including a residual polygenic effect slightly increases reliability and reduce bias in prediction of future records [17,18]. Such a method requires the partitioning of genetic variance accounted for by genomic information and remaining genetic variance accounted for by pedigree relationships. In human genetics very high density chips have been used in large scale studies [10]. However, genome wide as- sociation studies (GWAS) have generally not been able to identify individual genes that can explain a large propor- tion of total genetic variance of complex traits recorded in humans. Background This will only allow loci with large effects to become statistically significant. However, [10] showed that a considerable proportion of additive genetic variance can be explained by a very large number of SNPs and their effects can be predicted simultaneously using ap- propriate statistical models. Such models also includes the SNPs with small effects as long as they are associated with the trait of interest This association may be either due to SNPs being located within causative loci, being in linkage disequilibrium (LD) with causative genes, or due to markers tracing parts of familial additive genetic relationships among genotyped animals [15]. These results are well in line with results from dairy cattle where a large proportion of additive genetic can be explained using dense markers. Methods Data Data on deregressed proofs (DRP) were used as response variable in the present study, which were derived from the Nordic Holstein genetic evaluations official spring 2011 run. Production traits included milk production (milk), fat production (fat), protein production (protein). Fitness related traits included female fertility (fertility), other dis- eases than mastitis (health), and mastitis (mastitis). The specific trait definitions and recording procedures are detailed in http://www.nordicebv.info/Forside.htm and [21]. The Nordic Genetic Evaluation (NAV) system for dairy cattle includes all recorded dairy cattle in Denmark, Sweden and Finland and published proofs is standardized so that the variance of predicted breeding values for bulls belonging to a recent two birth year cohort was 100. Since the reliabilities of published proofs vary among traits the variance of the deregressed proofs also vary among the Page 3 of 9 Page 3 of 9 Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 accounted for by markers on a specific autosomal chromo- some and go is the vector of additive genetic effects accounted for by markers on all remaining chromosomes. 1 is a vector of ones and Za and Zg are incidence matrices relating observations in y to additive genetic effects in a and in gx, gc or go, respectively. Subscript x on vectors in different models indicates that definitions vary with model. Due to computational constraints specific effects of mar- kers on single chromosomes was estimated for only one chromosome at a time, and each analysis included effects of markers on the specific chromosome and the combined effects of markers on all other chromosomes. Therefore model (4) was run one time for each trait and for each chromosome. traits analyzed. The deregression removes the shrinkage of the published proofs and the DRP, therefore, is similar to progeny group means corrected for all non-genetic effects influencing the traits analyzed. The ratios of variance com- ponents are dependent of the accuracy of the individual DRP. All estimates of variance ratios etc. are shown at the average reliability of the bulls included in the study. The deregressed proofs were merged with marker records of individual bulls that were typed using Illumina Bovine SNP50 BeadChip (Illumina, San Diego, California, US). Only bulls with both genotype records and dereg- ressed proof for at least one trait in the database were included in the analysis. A summary of the data used are shown in Table 1. Methods Data In total there were 5217 bulls which were born in the period from 1989 to 2006. The parameter μ was considered as a fixed effect in all models and all other effects were assumed random nor- mally distributed effects with variances var ax ð Þ ¼ Aσ2 a , where A is the additive genetic relationship matrix com- puted from the full pedigree, and var gx ð Þ ¼ Gσ2 g, where A total of 47152 SNP markers were available in the raw marker data, and after removing markers with minor allele frequency (MAF) < 0.01 and non-informative markers that were a simple linear function of another marker, a total of 44012 markers remained for analysis. A sire-dam pedigree of all bulls included in the analysis was constructed from the official pedigree file from NAV (http://www.nordicebv. info). The pedigree file included a total of 42144 animals and most bulls were traced to the middle of the previous century or earlier. G ¼ M P ð Þ M P ð Þ 0 2 P m j¼1 pj 1 pj ð5Þ ð5Þ In (5) M is an allele sharing matrix containing the num- ber of copies of the second allele, and P is a matrix con- taining twice the population frequency of the second allele, i.e.2pj, and m is the number of marker loci included in computation of G. The division with 2Pm j¼1pj 1 pj in (5) is included to ensure that the scale of A and G are comparable but will otherwise not influence the predic- tions from the model but only the scale of model para- meters. Finally, var e ð Þ ¼ Dσ2 e where D is a diagonal matrix containing weights proportional to the effective number of records in each DRP. The linear models (1)-(4) are based on different ways of describing the relationship among animals. For genomic relationships the methods used are detailed in [16]. In short, the relationship matrix A, based on pedigree information uses probability of iden- tity by descent, whereas the genomic relationship matrix G based on marker information use probability of identity by state. Analysis of genomic relationships Calculate Gm using then SNPs in group one and calculate Gt using the SNPs in group two, assuming that the SNPs in group two are the causal variants. 3. Use linear regression ð6Þ gtjk ¼ α þ βgmjk þ e ð6Þ for j ≤k. The diagonal dominance of Gt and Gm is removed by subtracting 1.0 from the diagonal elements before estimating α and β. In order to separate effects of polygenic/familial gen- etic relationships from genomic relationships model (3) were run. In this model the covariance among animals due to additive genetic relationships and due to genomic relationships (markers) both were included. Averaged across all traits total genetic variance estimated in model (3) was 101.7% of total genetic variance estimated in the animal model (1) with a range from 98.4% for protein to 106.3% for mastitis (computed from results in Table 2). Model (3) is thus able to explain all genetic variance in the population and estimates of total genetic variance are essentially identical to estimates from the classical individual animal model, which is expected to yield un- biased estimates of additive genetic variance. Averaged This procedure is repeated for different N and the re- lation between β and N can be determined empirically. To obtain an unbiased estimate of Gt we want E Gt G m ¼ G m . This can be accomplished by comput- ing: G m ¼ β Gm I ð Þ þ I ð7Þ ð7Þ The genomic covariance matrices are diagonally dom- inant with diagonal elements close to unity. If all diag- onal elements in Gmis unity the adjustment in (7) corresponds to adjusting estimated variance components by β. In other words the estimate of genetic variance obtained from model (2) is biased downwards with an amount proportional to β. Analysis of genomic relationships used as reference in the standardization predicted breed- ing values. The ratio of additive genetic variance over the total variance (VR) corresponds to a heritability of the DRP. DRP are basically progeny group means so it would be expected that the proportion of additive gen- etic variance in DRP increases with increasing progeny group size. In general dairy bulls in the Nordic countries are tested on large groups of progeny. For the produc- tion traits, VR were in the range 0.92 for milk to 0.97 for protein, which clearly indicate that the DRP have high accuracy. For the fitness traits VR were between 0.75 for health and 0.82 for mastitis. The lower level of VR for the fitness traits is due to the lower underlying heritability of such traits since they are based on the same progeny groups as the production traits. Additive genetic differences between individuals are due to, generally unknown, causative genes. If the genotypes at all causative loci were known, the true genomic rela- tionship matrix (Gt) with regard to the trait of interest could be computed based on all the causative loci. In practice this is not possible and instead we compute G based on the marker data only and here we name this Gm. The accuracy of Gm to describe the genetic covari- ance among individuals sharing the same causative genes (Gt) depends on the linkage disequilibrium between the markers and the causative genes. The accuracy of the genomic relationship matrix can be assessed using the procedure of [10] who regard Gm as an estimate of Gt. The procedure includes the following steps: For all traits analyzed, the genetic (total genomic) vari- ance estimated in model (2) was lower than the additive genetic variance estimated using the classical individual animal model (1). The reason for this is that the genomic relationship matrix (G) do not trace all relationships due to sharing of causative alleles. However, the difference is small and generally the genomic information accounts for between 92% and 98% of the total additive genetic vari- ance depending on the trait in question. These results are well in line with results of [17] and [16] who used regres- sions on future data to validate their models. 1. Randomly sample 2 N SNPs across the genome and divide them in two groups of equal size. 2. Models The data were analyzed using the following models: ð1Þ y ¼ 1μ þ Zaa1 þ e1 ð1Þ y ¼ 1μ þ Zg g2 þ e2 ð2Þ y ¼ 1μ þ Zg g3 þ Za a3 þ e3 ð3Þ y ¼ 1μ þ Zg gc þ Zg go þ Za a4 þ e4 ð4Þ y ¼ 1μ þ Zaa1 þ e1 ð1Þ y ¼ 1μ þ Zg g2 þ e2 ð2Þ y ¼ 1μ þ Zg g3 þ Za a3 þ e3 ð3Þ y ¼ 1μ þ Zg gc þ Zg go þ Za a4 þ e4 ð4Þ y ¼ 1μ þ Zg g3 þ Za a3 þ e3 ð4Þ y ¼ 1μ þ Zg gc þ Zg go þ Za a4 þ e4 where μ is the general mean, ax is the vectors of additive genetic effects not accounted for by genetic markers (for model 1 this reduces to the classical individual animal model since no markers are included in the model), gx is the vector of the additive genetic effects accounted for by markers, gc is the vector of additive genetic effects Table 1 Mean and standard deviation of deregressed bull proofs Abbreviation Trait n mean Sd Milk Milk yield 4398 97.41 13.21 Fat Fat yield 4398 96.99 12.23 Protein Protein yield 4398 95.44 14,55 Fertility Female fertility 4415 99.44 16.90 Health Health index 4240 96.69 19.22 Mastitis Mastitis resistance 4398 95.98 11.70 Table 1 Mean and standard deviation of deregressed bull proofs The number of markers included in computation of G in models (2) and (3) were 44012. The same markers were used in model (4) but markers were split into mar- kers for one chromosome at a time and the markers on all other chromosomes pooled such that two genomic relationship of same size were computed. Model (4) was used in this way for all 29 autosomal chromosomes. All analysis including estimation of variance compo- nents using Restricted Maximum Likelihood were con- ducted using the DMU software [22,23]. Page 4 of 9 Page 4 of 9 Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 Analysis of genomic relationships Table 2 Additive genetic and genomic variances for production and fitness traits estimated in models 1, 2, 3 and 4 Model 1 Model 2 Model 3 Model 4 Trait ^σa2 VR1 aÞ ^σg2 VR1 gÞ ^σg2 ^σa;2 VR1 aþgÞ P^σa2 Milk 138.24 0.92 134.18 0.88 119.49 20.87 0.93 115.3 Fat 113.10 0.91 109.33 0.87 93.61 22.36 0.94 90.5 Protein 143.16 0.97 132.99 0.88 106.67 34.26 0.96 103.0 Fertility 151.74 0.78 142.42 0.74 110.38 40.10 0.78 106.5 Health 141.57 0.65 136.70 0.63 101.84 42.60 0.66 98.4 Mastitis 99.19 0.82 97.30 0.79 81.77 23.67 0.85 79.0 1) Ratio of genetic variance in model over total variance. Variance components A summary of the estimated variance components from all models (1) to (4) are shown in Table 2. Estimates of additive genetic variance using a classical individual ani- mal model (model (1)) ranged from 99.19 to 151.74 on the deregressed scale from the Nordic Genetic Evalu- ation of Dairy Cattle. The scale of the deregressed proofs are somewhat arbitrary since it depends on the accuracy of breeding values for the bulls included in the cohort Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 Page 5 of 9 Page 5 of 9 Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 The estimates of variance due to individual chromosomes were then summed over the models. For all traits the sum of variances due to individual chromosomes is slightly smaller than the total genomic variance for each trait in model (3). The sum of chromosomal variance as a per- centage of the total genomic variance estimated in model (3) varied between 96% and 97% for all traits (Results not shown). This indicates that covariance between genomic effects on different chromosomes is positive but weak. The estimates of variance due to individual chromosomes were then summed over the models. For all traits the sum of variances due to individual chromosomes is slightly smaller than the total genomic variance for each trait in model (3). The sum of chromosomal variance as a per- centage of the total genomic variance estimated in model (3) varied between 96% and 97% for all traits (Results not shown). This indicates that covariance between genomic effects on different chromosomes is positive but weak. over the six traits 77.2% of total genetic variance esti- mated in model (3) is attributed to genomic relation- ships among bulls and the remaining 22.8% is attributed to familial genetic relationships among bulls traced via the additive genetic relationship matrix. For the produc- tion traits this percentage varied between 75% and 85% and for the fitness related traits it varied between 71% and 78% (percentages computed from Table 2). This indicates that the genomic relationship matrix G is able to catch a large proportion of the genetic variation in the population when based on the SNP density used in the current study. Estimates of variance components due to individual chromosomes are shown in Figure 1. Discussion and conclusions Figure 2 Expected proportion of total additive genetic variance traced by increasing number of markers. The records analyzed in this paper were DRPs which were derived from the routine genetic evaluations of dairy cattle in Denmark, Sweden and Finland. Such DRPs are similar to progeny group means adjusted for non-genetic effects. Therefore, a very large proportion of phenotypic variance in analyzed DRP was due to additive genetic effects. For all traits analyzed more than 92% of all additive genetic variance could be explained using 44 K SNP markers. For models including both polygenic additive genetic (pedi- gree) effects and genomic (marker) effects, the latter accounted for between 71% and 85% of all additive genetic variance. Estimation of genomic variance of each individ- ual chromosome showed that 96%–97% of all genomic (marker) variance could be attributed to individual chro- mosomes. Inclusion of polygenic familial effects in the models ensured that potential linkage disequilibrium across chromosomes was already taken into account. Most of the additive genetic variance in the population analyzed could be explained by genetic markers. The effect of redu- cing (increasing) the number of genetic markers on gen- omic prediction could be predicted by estimating the accuracy of the genomic relationship matrix. it would be expected that chromosomal variance was closely related to chromosome length. The R2 values for a linear regression model of chromosomal variance on chromosome length varied between 0.11 and 0.21 (Figure 1). This indicates that QTL are not evenly distributed over the genome. This is clearly the case for milk and fat that had very low R2 due to the major gene segregating as men- tioned above. In general the relation between variance due to individual chromosomes might also be dependent on the model and method used as well as the underlying popula- tion structure in the data analyzed. Variance components The estimates are plotted against the length of individual chromosomes measured as the number of nucleotides (in Mb) between first and last marker on each chromosome. Note that the scale of the estimated variances varies by trait in Fig- ure 1. Very large variances for milk and fat were found on chromosome 14 where it is known that a gene of very large effect segregates in the Holstein population [24]. As can clearly be seen from the figure the relation between chromosomal variance and chromosome length is weak. If QTL were evenly distributed over the genome Additive genetic variance due to individual chromosomes Results from analysis using model (4), where genomic variance due to individual chromosomes were estimated, is also summarized in Table 2. Only the genomic variance summed over all chromosomes is shown. The chromo- somal variance was estimated in 29 individual models where the effect of an individual chromosome and the combined effect of all other chromosomes were estimated. Figure 1 Estimates of genomic variance (y axis) due to individual chromosomes in relation to chromosome length (x axis) in Mb. Figure 1 Estimates of genomic variance (y axis) due to individual chromosomes in relation to chromosome length (x axis) in Mb. Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 Page 6 of 9 expected proportion of additive genetic variance explained by markers reaches 0.96 when all 44 K markers were used. Figure 2 Expected proportion of total additive genetic variance traced by increasing number of markers. The additive genetic variance explained by different number of markers were investigated using model (2) by varying the number of markers used to compute the gen- omic relationship matrix. Results averaged over all traits are shown in Table 3. Results using all 44 K markers are the same as in the analysis presented in Table 2 and are repeated here in relative form for reference. As the num- ber of markers included in the analysis decreases the pro- portion of total additive genetic variance estimated as genomic variance also decreases. As can be seen from Figure 2. The estimates of total genomic variance followed the expectation (proportional to β) quite closely when the numbers of markers included in the genomic relationship matrix were varied. Amount of variance explained depending on number of SNPs The procedure of [10] were used to estimate the proportion of additive genetic variance that can be explained by mar- kers. This expected proportion (β) is shown in (Figure 2) as a function of number of markers included in the genomic relationship matrix. The expected proportion of additive genetic variance explained by markers is less than 0.85 when the number of markers is below 5 K but increases rapidly until between 15 K and 20 K markers is used in the estima- tion of genomic variance. Further increases in the number of markers only increase the expected proportion of additive genetic variance explained by markers marginally. The Variance components Surprisingly the estimates of variance due to individual chromosomes only showed a weak relationship with chromosome length (Figure 1). The results on individual chromosomes are in contrast to results of [26] who found strong relations between chromosomal variance and chromosome length. These authors used a matrix of kin- ships between individuals as genomic covariance matrix and estimated the variance due to individual chromosomes as the difference between models using all markers and models using all markers but the ones on the specific chromosome in question. Such an indirect procedure is necessary when using kinship matrices, because such matrices normally are singular if they are based on a lim- ited number of markers. We repeated our analysis using methods as in [26]. Generally variance estimates due to genetic similarity matrices are larger than variances due to the genomic relationship matrix used in this paper because of different scales. However, the correlations between esti- mates of variances due to individual chromosomes and chromosome length were even lower than those presented in this paper and several estimates were negative! This method, therefore, was not further pursued [27]. Also esti- mate genomic variance due to individual chromosomes. They used a Bayesian approach that allocated equal prior variance to each SNP. Chromosomes with many SNPs, therefore, received more prior variance than chromosomes with few(er) SNPs. They were able to identify genomic regions with larger contribution to genomic variance due to known major genes. However, most SNPs had small effects and therefore there were strong associations be- tween the amount of variance per chromosome and chromosome length or equally number of SNPs per chromosome. Effects of individual SNPs are composed of effects due to co-segregation with closely linked QTL and effects due to LD with QTL elsewhere in the genome, the latter generated by familial relationships in the population. This effect tends to smooth all genomic variance over all SNPs and may therefore give an unclear picture of how much genomic variance can be ascribed to each chromo- some. Clearly more research on partitioning of genomic variance into effects of individual chromosomes, chromo- some segments or grouping of markers that are expected to be located near causative genes etc. is needed. The amount of additive genetic variance that can be explained by markers obviously cannot exceed the total additive genetic variance in the population. Variance components Most of the total phenotypic variance in the traits analyzed was additive genetic due to the use of deregressed proofs, which average out any dominance deviations across mul- tiple daughters. Such proofs, of course, are functions of the Table 3 Expected (β) and estimated relative amount of additive genetic variance explained by different number of markers No of Markers β Estimated proportion of genetic variance explained by markers 44012 0.960 0.936 22006 0.930 0.918 11003 0.909 0.880 Table 3 Expected (β) and estimated relative amount of additive genetic variance explained by different number of markers No of Markers β Estimated proportion of genetic variance explained by markers 44012 0.960 0.936 22006 0.930 0.918 11003 0.909 0.880 Table 3 Expected (β) and estimated relative amount of additive genetic variance explained by dif ed (β) and estimated relative amount of additive genetic variance explained by different number o Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 Page 7 of 9 Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 Page 7 of 9 Additive genetic variance due to individual chromosomes When summing over chromosomes the estimates of gen- omic variance due to individual chromosomes yielded total genomic variances that were similar to the total genomic variance in model (2) where this quantity was estimated directly. The method therefore seems able to yield esti- mates of genomic variance due to individual chromosomes. Surprisingly the estimates of variance due to individual chromosomes only showed a weak relationship with chromosome length (Figure 1). The results on individual chromosomes are in contrast to results of [26] who found strong relations between chromosomal variance and chromosome length. These authors used a matrix of kin- ships between individuals as genomic covariance matrix and estimated the variance due to individual chromosomes as the difference between models using all markers and models using all markers but the ones on the specific chromosome in question. Such an indirect procedure is necessary when using kinship matrices, because such matrices normally are singular if they are based on a lim- ited number of markers. We repeated our analysis using methods as in [26]. Generally variance estimates due to genetic similarity matrices are larger than variances due to the genomic relationship matrix used in this paper because of different scales. Variance components However, the correlations between esti- mates of variances due to individual chromosomes and chromosome length were even lower than those presented in this paper and several estimates were negative! This method, therefore, was not further pursued [27]. Also esti- mate genomic variance due to individual chromosomes. They used a Bayesian approach that allocated equal prior variance to each SNP. Chromosomes with many SNPs, therefore, received more prior variance than chromosomes with few(er) SNPs. They were able to identify genomic regions with larger contribution to genomic variance due to known major genes. However, most SNPs had small effects and therefore there were strong associations be- tween the amount of variance per chromosome and chromosome length or equally number of SNPs per chromosome. Effects of individual SNPs are composed of effects due to co-segregation with closely linked QTL and effects due to LD with QTL elsewhere in the genome, the latter generated by familial relationships in the population. This effect tends to smooth all genomic variance over all SNPs and may therefore give an unclear picture of how much genomic variance can be ascribed to each chromo- some. Clearly more research on partitioning of genomic variance into effects of individual chromosomes, chromo- some segments or grouping of markers that are expected to be located near causative genes etc. is needed. Amount of variance explained by genomic markers procedures and definitions used in the recording system and methods used in the genetic evaluation. For produc- tion traits (milk, fat, and protein) the additive genetic vari- ance was between 91% and 97% of total phenotypic variance. The high level is expected because of the large sizes of progeny group used in the testing procedures for dairy bulls in all the Nordic countries and because of the high heritability of the underlying traits. For the fitness related traits (fertility, health, and mastitis) the relative amount of additive genetic variance is lower reflecting the lower accuracy of fitness related DRP. Additive genetic variance due to individual chromosomes When summing over chromosomes the estimates of gen- omic variance due to individual chromosomes yielded total genomic variances that were similar to the total genomic variance in model (2) where this quantity was estimated directly. The method therefore seems able to yield esti- mates of genomic variance due to individual chromosomes. Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 Page 8 of 9 population. The analysis of the genomic relationship matrices revealed that a large proportion of the total addi- tive genetic variance in the Nordic Holstein population was expected to be explained by a set of 44 K markers. Analysis of both production and fitness related traits showed that the amount of variance accounted for by markers in the Nordic Holstein population was close to the expectations from regression based analysis of the gen- omic relationship matrix. Estimates of genomic variance closely followed expectation when the number of markers included in computation of genomic relationship matrix was varied. The amount of additive genetic variance that can be explained by genomic markers depends on several factors: Number of markers on causative sites, markers in linkage disequilibrium with causative genes due to close “historical” linkage at population level, and finally linkage disequilibrium among markers and genes at family level, due to the family structure in the population. With 44 K markers spread over the genome the number of markers within causative sites probably is limited. The linkage dis- equilibrium between markers and causative genes is very dependent on effective population size [28]. The Holstein cattle population has a low effective population size and, therefore, there will be relatively few recombination events in the recent history of the breed. In practice this means that there will be considerable linkage disequilibrium be- tween markers and causative genes. This is also supported by the fact that most of the total additive genetic variation was explained by genomic relationships and not by addi- tive genetic relationships based on pedigree when the model includes both relationship matrices. The family structure in dairy cattle populations creates linkage dis- equilibrium between markers and causative genes even if they are on different chromosomes and this also helps the markers in being able to explain most of the additive gen- etic variance in the population. Analysis of the structure of the genomic relationship matrices might be of considerable value in deciding on avenue for future development of typing strategies when using genomic markers. Such analysis also could give extra insight in the effects of population structure and population history on effectiveness of future selection programs using genomic selection in other breeds or in other species. Authors’ contributions JUJ conceived the study and conducted all analysis, GS developed and implemented algorithms for computing genomic relationship matrices and PM maintained the DMU software package used in the statistical analysis. JUJ edited the manuscript based on extensive input from all authors who have read and approved the final manuscript. The analysis of genomic relationship matrices showed that a high proportion of additive genetic variance can be expected to be explained using 44 k genomic markers in this population of dairy cattle. This leaves limited room for further improvements of predictive ability of genomic models by including more markers. One of the current trends in use of genomic markers is to move from 50 K marker chips to 800 K marker chips or even complete se- quencing of whole genomes for individual animals. Our results indicate that the advantages of this route may be limited. In fact including several orders of more markers than used in this study may turn out to be counterpro- ductive. Extremely dense markers will include more mar- kers on most causative sites and given knowledge of variation in the causative genes there is no extra informa- tion in the remaining markers. Alternative models that better can distinguish between causative genes and non in- formative markers might be of great value in future Competing interests The authors declare no competing interests. Abbreviations GS G i l Abbreviations GS: Genomic selection; DRP: Deregressed proof; Milk: Milk production; Fat: Fat production; Protein: Protein production; Fertility: Female fertility; Health: Veterinary treatments for diseases other than mastitis; Mastitis: Mastitis. Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 In summary we estimated the amount of additive genetic variance that can be explained using dense SNP marker panels. In the Holstein population analyzed, almost all the additive genetic variance could be explained using 44 K SNP markers. The amount of additive genetic variance that is expected to be explained by markers could be predicted from analysis of the genomic relationship matrix. Further increases in marker density will have limited effects on pre- dictive accuracy unless better methods distinguishing be- tween markers with real effects and markers with no effect are used. Results presented in this study can be used to de- termine the weight given to marker relationships and to fa- milial relationships in one step prediction methods where these sources of relationships are combined and in two step methods where information based on genomic rela- tionships must be combined with information form poly- genic relationships. 1. Meuwissen TH, Hayes BJ, Goddard ME: Prediction of total genetic value using genome-wide dense marker maps. Genetics 2001, 157:1819–1829. 2. Schaeffer LR: Strategy for applying genome wide selection in dairy cattle. Journal of Animal Breeding and Genetics 2006, 123:218–223. 3. Buch LH, Sørensen MK, Berg P, Pedersen LD, Sørensen AC: Genomic selection strategies in dairy cattle: Strong positive interaction between Variance components The classical individual animal model (1) yields unbiased estimates of population additive genetic variance. Comparing results from model (2) (genomic model) with results from model (1) (animal model) clearly illustrates this. For the traits milk and fat 97% of the additive genetic variance can be explained by markers. The estimates of additive genetic variance due to individual chromosomes clearly show that these traits are influenced by a major gene that contributes to the large proportion of additive genetic variance. Statis- tical models fitting effects of individual marker genes might be a better alternative in these cases. For protein and fertility the amount of additive genetic variance that can be explained by markers is somewhat lower than for milk and fat. This indicate that models used for these traits should include an effect to account for addi- tive genetic effects not accounted for by markers or alterna- tively more markers are needed to ensure that most of the additive genetic variance can be explained by markers. This is clearly seen from the results of model (3) where both mar- ker effects and classical polygenic familial additive genetic effects were included in the model. However, in general, a large proportion of the total additive genetic variance was explained by markers in this study. This is in considerable contrast to [10] who found that only 45% of additive genetic variance in human height was explained using a marker panel of 294,831 SNPs. One major difference between that study and the current study is that the individuals in the human study were nearly unrelated whereas the bulls in the current study have very many relationships due to the highly structured breeding programs used in commercial dairy cat- tle. Consequently, the genomic variance (σ2 g ) include three different sources of variation: segregation of observed SNPs that cause functional differences in the genes affecting the trait analyzed, co-segregation of markers and causative genes due to physical linkage at population level and linkage dis- equilibrium at family level. A second difference is that the effective population size in dairy cattle is small with esti- mates in Danish Holstein of Ne=49 [25], whereas the effect- ive population size in humans is expected to be several orders larger than in dairy cattle. Amount of variance explained by genomic markers Obviously genetic markers cannot explain more than all the total (additive) genetic variation present in the Acknowledgments The Danish Cattle Federation, FaBa Co-Op, Swedish Dairy Association and Nordic Genetic evaluation is thanked for providing the data. External funding for this study, including the extensive genotyping, were provided by the Danish Ministry for Food, Agriculture and Fisheries project “Genomic Selection – from function to efficient utilization in cattle breeding” Grant no 3405-10-0137 and the Milk Levy Fund, Viking Genetics and Nordic Genetic Evaluation. The first author was funded through a grant from Aarhus University. Received: 1 March 2012 Accepted: 31 May 2012 Published: 13 June 2012 Received: 1 March 2012 Accepted: 31 May 2012 Published: 13 June 2012 Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 26. Pimentel ECG, Erbe M, Koenig S, Simianer H: Frontiers: Genome Partitioning of Genetic Variation for Milk Production and Composition Traits in Holstein Cattle. Genomics: Frontiers in Livestock; 2011:2. 27. Cole JB, VanRaden PM, O’Connell JR, Van Tassell CP, Sonstegard TS, Schnabel RD, et al: Distribution and location of genetic effects for dairy traits. Journal of dairy science 2009, 92:2931–2946. 28. Goddard M: Genomic selection: prediction of accuracy and maximisation of long term response. Genetica 2009, 136:245–257. doi:10.1186/1471-2156-13-44 Cite this article as: Jensen et al.: Partitioning additive genetic variance into genomic and remaining polygenic components for complex traits in dairy cattle. BMC Genetics 2012 13:44. use of genotypic information and intensive use of young bulls on genetic gain. Journal of Animal Breeding and Genetics 2012, 129:138-151. use of genotypic information and intensive use of young bulls on genetic gain. Journal of Animal Breeding and Genetics 2012, 129:138-151. 4. Mc Hugh N, Meuwissen THE, Cromie AR, Sonesson AK: Use of female information in dairy cattle genomic breeding programs. Journal of dairy science 2011, 94:4109–4118. 5. VanRaden P, Cooper T: The genomic evaluation system in the United States: Past, present, future. Journal of dairy science 2011, 94:3202–3211. 6. Lund M, de Ross S, de Vries A, Druet T, Ducrocq V, Fritz S, et al: A common reference population from four European Holstein populations increases reliability of genomic predictions. Genetics Selection Evolution 2011, 43:43. doi:10.1186/1471-2156-13-44 Cite this article as: Jensen et al.: Partitioning additive genetic variance into genomic and remaining polygenic components for complex traits in dairy cattle. BMC Genetics 2012 13:44. doi:10.1186/1471-2156-13-44 Cite this article as: Jensen et al.: Partitioning additive genetic variance into genomic and remaining polygenic components for complex traits in dairy cattle. BMC Genetics 2012 13:44. 7. Jannink JL, Lorenz AJ, Iwata H: Genomic selection in plant breeding: from theory to practice. Briefings in Functional Genomics 2010, 9:166–177. 8. Matukumalli LK, Lawley CT, Schnabel RD, Taylor JF, Allan MF, Heaton MP, et al: Development and characterization of a high density SNP genotyping assay for cattle. PLoS One 2009, 4:e5350. 9. Harris BL, Johnson DL, Spelman RJ, Sattler JD: Genomic selection in New Zealand and the implications for national genetic evaluation. ICAR Technical Series 2009, 13:325–330. 10. Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 Yang J, Benyamin B, McEvoy BP, Gordon S, Henders AK, Nyholt DR, et al: Common SNPs explain a large proportion of the heritability for human height. Nature genetics 2010, 42:565–569. 11. Visscher PM, Hill WG, Wray NR: Heritability in the genomics era\|[mdash]\| concepts and misconceptions. Nature Reviews Genetics 2008, 9:255–266. 12. Gudbjartsson DF, Walters GB, Thorleifsson G, Stefansson H, Halldorsson BV, Zusmanovich P, et al: Many sequence variants affecting diversity of adult human height. Nature genetics 2008, 40:609–615. 13. Maher B: Personal genomes: The case of the missing heritability. Nature 2008, 456:18. 14. Johannes F, Colot V, Jansen RC: PersPeCTives. Epigenome dynamics: a quantitative genetics perspective. NATuRe ReVIeWS\| genetics 2008, 9:883. 15. Habier D, Fernando RL, Dekkers JCM: The Impact of Genetic Relationship Information on Genome-Assisted Breeding Values. Genetics 2007, 177:2389–2397. 16. VanRaden PM: Efficient methods to compute genomic predictions. Journal of dairy science 2008, 91:4414–4423. 17. Liu Z, Seefried FR, Reinhardt F, Rensing S, Thaller G, Reents R: Impacts of both reference population size and inclusion of a residual polygenic effect on the accuracy of genomic prediction. Genetics, selection, evolution: GSE 2011, 43:19. 18. Gao H, Christensen O, Madsen P, Nielsen U, Zhang Y, Lund M, et al: Comparison on genomic predictions using three GBLUP methods and two single-step blending methods in the Nordic Holstein population. Genetics Selection Evolution 2012, 44:8. 19. Christensen OF, Lund MS: Genomic prediction when some animals are not genotyped. Genet Sel Evol 2010, 42(2):1-8. 20. Legarra A, Aguilar I, Misztal I: A relationship matrix including full pedigree and genomic information. Journal of dairy science 2009, 92:4656–4663. 21. Danish Cattle Federation: Principles of Danish Cattle Breeding: Recording of Production Data, Recording of Breeding Data, Calculation Methods, Breeding Values. Eight Edition. Report The Danish Agricultural Advisory Center 2006;126. 22. Jensen J, Mantysaari EA, Madsen P, Thompson R: Residual maximum likelihood estimation of (co) variance components in multivariate mixed linear models using average information. Jour Ind Soc Ag Statistics 1997, 49:215–236. 23. Madsen P, Jensen J. A users guide to DMU. A package for analysing multivariate mixed models, Version 6, Release 5.0. 2010. (http://dmu.agrsci. dk/dmuv6_guide.5.0.pdf). Report. References 1. Meuwissen TH, Hayes BJ, Goddard ME: Prediction of total genetic value using genome-wide dense marker maps. Genetics 2001, 157:1819–1829. 2. Schaeffer LR: Strategy for applying genome wide selection in dairy cattle. Journal of Animal Breeding and Genetics 2006, 123:218–223. 1. Meuwissen TH, Hayes BJ, Goddard ME: Prediction of total genetic value using genome-wide dense marker maps. Genetics 2001, 157:1819–1829. 2. Schaeffer LR: Strategy for applying genome wide selection in dairy cattle. Journal of Animal Breeding and Genetics 2006, 123:218–223. 3. Buch LH, Sørensen MK, Berg P, Pedersen LD, Sørensen AC: Genomic selection strategies in dairy cattle: Strong positive interaction between 1. Meuwissen TH, Hayes BJ, Goddard ME: Prediction of total genetic value using genome-wide dense marker maps. Genetics 2001, 157:1819–1829. 2. Schaeffer LR: Strategy for applying genome wide selection in dairy cattle. 1. Meuwissen TH, Hayes BJ, Goddard ME: Prediction of total genetic value using genome-wide dense marker maps. Genetics 2001, 157:1819–1829. 3. Buch LH, Sørensen MK, Berg P, Pedersen LD, Sørensen AC: Genomic selection strategies in dairy cattle: Strong positive interaction between Page 9 of 9 Jensen et al. BMC Genetics 2012, 13:44 http://www.biomedcentral.com/1471-2156/13/44 Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: Submit your next manuscript to BioMed Central and take full advantage of: 24. Grisart B, Coppieters W, Farnir Fdr, Karim L, Ford C, Berzi P, et al: Positional Candidate Cloning of a QTL in Dairy Cattle: Identification of a Missense Mutation in the Bovine DGAT1 Gene with Major Effect on Milk Yield and Composition. Genome Research 2002, 12:222–231. • Convenient online submission 25. Sørensen AC, Sørensen MK, Berg P: Inbreeding in Danish Dairy Cattle Breeds. Journal of dairy science 2005, 88:1865–1872.
https://openalex.org/W1486011393	http://sedici.unlp.edu.ar/bitstream/handle/10915/140393/Documento.pdf?sequence=1	English	null	Immiscible fluid flow through nanotubes	Theoretical Chemistry accounts	2,003	cc-by	2,919	Regular article Immiscible ﬂuid ﬂow through nanotubes Interfaces between coexisting phases are under intense scientiﬁc investiga- tion, because of the signiﬁcant role they play in technological applications. They are critical to the performance of detergents and soaps, to chemical engineering separation processes such as absorption and distillation, to petroleum recovery and the function of biological membranes, to mention just a few examples [2]. An important role for molecular dynamics (MD) simulation is to explore the assumptions used in experimental results. At the molecular level, information about the liquid/liquid interface is contained in the density and orientational proﬁles, although there is no direct experimental measurement of the orientational proﬁle. In contrast, MD simulation can provide infor- mation about the form of these orientational distribu- tion functions [3]. Therefore, the macroscopic ﬂow description must be augmented with the results of the microscopic physics. In related problems, MD simulations of slow channel ﬂow of a viscous inhomogeneous ﬂuid [1, 4] and water conduction through the hydrophobic channel of a From the Proceedings of the 28th Congreso de Quı´micos Teo´ ricos de Expresio´ n Latina (QUITEL 2002) Correspondence to: J. Maran˜ o´ n e-mail: maranon@venus.ﬁsica.unlp.edu.ar The low-Reynolds-numbers ﬂow of continuum Newton ﬂuids has been described by the Stokes equations; however, there remain a number of unresolved questions regarding the ﬂuid interface [1]. Interfaces between coexisting phases are under intense scientiﬁc investiga- tion, because of the signiﬁcant role they play in technological applications. They are critical to the performance of detergents and soaps, to chemical engineering separation processes such as absorption and distillation, to petroleum recovery and the function of biological membranes, to mention just a few examples [2]. An important role for molecular dynamics (MD) simulation is to explore the assumptions used in experimental results. At the molecular level, information about the liquid/liquid interface is contained in the density and orientational proﬁles, although there is no direct experimental measurement of the orientational proﬁle. In contrast, MD simulation can provide infor- mation about the form of these orientational distribu- tion functions [3]. Therefore, the macroscopic ﬂow description must be augmented with the results of the microscopic physics. The criterion for the stability of the interface between two liquids was the equality of the chemical potential in both phases at the interface. For this condition to occur, the pressure components normal to the interface in the two phases must be equal. Theor Chem Acc (2003) 110: 410–413 DOI 10.1007/s00214-003-0495-6 Theor Chem Acc (2003) 110: 410–413 DOI 10.1007/s00214-003-0495-6 Regular article Immiscible ﬂuid ﬂow through nanotubes J. Maran˜ o´ n Di Leo1, J. Maran˜ o´ n2 1 Departamento de Aerona´ utica, Facultad de Ingenierı´a, Universidad Nacional de La Plata, 1, 47, 1900 La Plata, Argentina 2 Departamento de Fı´sica, Facultad de Ciencias Exactas, Universidad Nacional de La Plata, C.C. 67, 1900 La Plata, Argentina R i d 8 S t b 2002 / A t d 22 M 2003 / P bli h d li 25 N b 2003 Regular article Immiscible ﬂuid ﬂow through nanotubes 1 Departamento de Aerona´ utica, Facultad de Ingenierı´a, Universidad Nacional de La Plata, 1, 47, 1900 La Plata, Argentina 2 Departamento de Fı´sica, Facultad de Ciencias Exactas, Universidad Nacional de La Plata, C.C. 67, 1900 La Plata, Argentina Received: 8 September 2002 / Accepted: 22 May 2003 / Published online: 25 November 2003 Springer-Verlag 2003 Received: 8 September 2002 / Accepted: 22 May 2003 / Published online: 25 November 2003 Springer-Verlag 2003 Received: 8 September 2002 / Accepted: 22 May 2003 / Published online: 25 November 2003 Springer-Verlag 2003 Abstract. We have analyzed the properties that charac- terize the water/oil interface when ﬂow occurs through a prismatic ﬂexible surface of van der Waals particles at 300 K. Thermodynamics and dynamics properties of the ﬂow as density proﬁles, interface width and diﬀusion coeﬃcients were calculated. We have observed a com- plex mechanism, the back-and-forth movement of the interface when advancing through a nanotube. carbon nanotube [5] have been carried out. To this end, we have studied with MD simulation the slow ﬂow of the water/hydrocarbon interface through a rectangular prismatic network (a parallelogram Lx:Ly:Lz = 6.44:3.0:3.0 nm, net constant 0.15 nm), of 3520 single- wall van der Waals particles. We carried out a MD simulation with the aim of characterizing the water/oil interface from its thermodynamic and diﬀusion proper- ties. Keywords: Interface – Conﬁnement – Flow – Nanotube Several MD simulation runs were performed on the water/oil interface using (periodic) boxes. We started with two separate simulations. One of pure SPC/E water [6] (846 molecules) and the other of a mixture of equal numbers of ethane, propane and butane molecules (60 molecules). Later, by joining the two boxes along the x direction a new box was constructed. Then we went to the previously mentioned prismatic surface conﬁning the 860 water molecules (SPC/E, with density 0.928 g/cm3) and 60 hydrocarbon molecules (butane, propane and ethane, with total density 0.2904 g/cm3). The x direction is perpendicular to the interfaces. These conditions generated two interfaces, the mutual independence of which requires that the two liquid slabs attain bulk properties. Keywords: Interface – Conﬁnement – Flow – Nanotube The low-Reynolds-numbers ﬂow of continuum Newton ﬂuids has been described by the Stokes equations; however, there remain a number of unresolved questions regarding the ﬂuid interface [1]. Regular article Immiscible ﬂuid ﬂow through nanotubes We therefore performed NPT runs with pressure scaling in the x direction, perpendic- ular to the interface, by coupling the 1055 Pa (1 atm) hydrostatic bath, and allowing changes in the box length only in that direction. In this way the system was allowed to adjust its bulk pressure by changing the volume of the box. To avoid undesirable eﬀects of long-range Coulomb interactions, we applied a charge group criterion for the interacting atoms within the cutoﬀ(0.9 nm). Conse- quently, only eﬀective dipole–dipole interactions were present. Then, the box went to the prismatic surface. In related problems, MD simulations of slow channel ﬂow of a viscous inhomogeneous ﬂuid [1, 4] and water conduction through the hydrophobic channel of a From the Proceedings of the 28th Congreso de Quı´micos Teo´ ricos de Expresio´ n Latina (QUITEL 2002) Because we considered the NPT simulation suﬃcient to obtain a stable water/oil interface, we ran a simula- tion keeping the box volume constant for a period of Correspondence to: J. Maran˜ o´ n e-mail: maranon@venus.ﬁsica.unlp.edu.ar Correspondence to: J. Maran˜ o´ n e-mail: maranon@venus.ﬁsica.unlp.edu.ar 411 0.5 ns. The criterion for the equilibrium was the stability of the potential energy of the system. 0.5 ns. The criterion for the equilibrium was the stability of the potential energy of the system. In the static situation both interfaces are symmetric, but in our case the symmetry is broken. However, with the object to improve the statistics we symmetrized the data from the two interfaces. The proﬁle density at Va ¼ 0:09 nm/ps is displayed in Fig. 1. Similar proﬁles were observed for diﬀerent mean velocities. p gy y The forces and conﬁguration energies were computed using the modiﬁed GROMOS87 package [7]. Only oxygen–oxygen, oxygen–carbon, carbon–carbon, oxy- gen–van der Waals particle and carbon–van der Waals particle interactions were considered, i.e. Lennard-Jones and electrostatic interactions. The hydrogen atoms interact only through Coulomb forces and the van der Waals particles do not interact with each other through Lennard-Jones forces. Periodic boundary conditions in the ﬂow direction were applied. The boundaries in the other direction were formed with the prismatic surface. The van der Waals particles were attached to a lattice site by a spring force (spring constant k ¼ 80) for the thermal roughness control of the prismatic surface [8]; besides with this k value the walls retain their integrity over the duration of the calculation. Regular article Immiscible ﬂuid ﬂow through nanotubes Deformations of the elastic surface are important factors that produce instability in the complex ﬂux, increasing the probability of wall disintegration at larger Reynolds numbers. The van der Waal particles were particles with mass and the van der Waals parameter of the oxygen atom. Besides, to attach the particles to each lattice site, each van der Waals particle had covalent bonds with the ﬁrst neigh- bor. For more details see Ref. [9]. The simulation shows that the water density proﬁle exhibits a pronounced layering eﬀect as in the free sim- ulation, and from this data, we can observe a well- deﬁned interface. The density proﬁle is rather diﬀerent from that reported by others authors [11, 12, 13] in nonconﬁned systems. The interface is placed at 4.18 nm; it corresponds to the middle point of the distance over which the water density proﬁle drops from 90 to 10%. The water density dropped more slowly to zero than the densities of the hydrocarbons did. The peak of the hydrocarbon mixture density proﬁle in the interfacial region is markedly much lower and wider than the similar one in a nonconﬁned system. The interface width obtained in our simulations was about 0:39 0:02 nm (Fig. 2) and was much narrower than the interface of the conﬁned equilibrium system Fig. 1. Density proﬁles of water/oil interface in the x direction. The dotted line is the water density; the solid line is the hydrocarbon mixture [ ] We started several NVT MD simulations. The tem- perature was controlled by coupling the system to a thermal bath at 300 K with a time constant of 0.3 ps. A twin-range cutoﬀfor nonbonded forces of 0.9/0.9 nm was used. The covalent bond lengths for water and hydrocarbon molecules, as the van der Waals particles of the prismatic network, were constrained by the SHAKE procedure (tolerance 0.0001 nm). The system was run using a time step of 2 fs. Trajectories were collected every ten time steps for further analysis. Both dynamics simulations were run on a Pentium computer. p We simulated the ﬂow, of low Reynolds numbers of order of less than 0.1, through a nanotube by intro- ducing an equivalent procedure to the gravity force ap- plied to each atom [1, 4, 10]. This ﬂow is generated by applying diﬀerent uniform acceleration to each ﬂuid in the direction of the nanotube. Regular article Immiscible ﬂuid ﬂow through nanotubes Because, we are dealing with small systems and short times, many trial-and-error runs were carried out in order to extract an average velocity, Va, from important thermal noise. Fig. 1. Density proﬁles of water/oil interface in the x direction. The dotted line is the water density; the solid line is the hydrocarbon mixture Fig. 2. Interface width versus water average velocity, Va In order to obtain a large Va spectrum, a series of simulations – about ten simulations – were done. From the atom trajectory analysis, relations between Va and diﬀerent properties that characterize the interface ﬂow were obtained. One important thermodynamics interface property is the density proﬁle of ﬂuids along the ﬂow. To this end, the density proﬁle was calculated using 64 slabs parallel to the yz plane along the x direction by averaging the time frames over the last 60 ps of the NVT simulation. Using the center-of-mass of water as a reference frame, the densities were calculated. We comment that the two proﬁles have suitable normalization: the water proﬁle with oxygen bulk den- sity and the hydrocarbon proﬁle with the corresponding bulk density of the molecular center of mass. Fig. 2. Interface width versus water average velocity, Va 412 Fig. 4. Velocity proﬁle of the water phase at Va ¼ 0:09 nm/ps (0.87 nm). Besides, the molecular water/hydrocarbon ratio (W/H), that characterizes the hydrophobicity of the interface, W/H = 8 1, is larger than in conﬁned equilibrium systems, W/H = 5. As a consequence of this behavior, fewer water and hydrocarbon molecules are involved in the interface owing to driven ﬂux forces increasing the hydropho- bicity of the interface. Further complementary infor- mation is given in Fig. 3a and b concerning to the cross- sectional densities of the water and hydrocarbon mixture for the ﬂux with Va ¼ 0:09 nm/ps. These proﬁles are not diﬀerent with respect to each ﬂux and also for zero-ﬂux proﬁles. The water proﬁle exhibits a pronounced layer- ing eﬀect as in the free simulation. The ﬁrst-layer mini- mum is 0.5 nm closes to the wall than the zero ﬂux. The hydrocarbon cross-sectional density shows that the molecules remain close to the wall over all the simula- tion. Fig. 4. Velocity proﬁle of the water phase at Va ¼ 0:09 nm/ps We report some dynamics properties of this complex ﬂow at Va ¼ 0:09 nm/ps. Regular article Immiscible ﬂuid ﬂow through nanotubes A proﬁle velocity of the water phase is displayed in Fig. 4. We can see that it is not Bernoulli ﬂow. The proﬁle represents a typical turbulent ﬂow of the ‘‘plug’’ type. calculated the dynamics diﬀusion coeﬃcients, Df, of water versus Va for diﬀerent ﬂows. Figure 5 shows that this relation is rather linear in Va. We can study quan- titatively the eﬀect of Va on diﬀusion in the ﬂow. Con- sider the diﬀusivity of water in the ﬂow direction: Continuing our analysis of the dynamics properties of the ﬂow, starting from the atom trajectories, we Df ¼ De þ CVa; ð1Þ ð1Þ Df ¼ De þ CVa; where De ¼ 2 105 is the diﬀusion coeﬃcient of water at equilibrium (Va ¼ 0:0) and C ¼ 149:5 ps/nm. This relation diﬀers from the quadratic dependence of Va of Tayor–Aris theory which was obtained by simulation with van der Waals particles of Poiseuille ﬂow [1]. Fig. 3. a Cross-sectional density of the water phase. b Cross- sectional density of the hydrocarbon phase e at equilibrium (Va ¼ 0:0) and C ¼ 149:5 ps/nm. Thi relation diﬀers from the quadratic dependence of Va o Tayor–Aris theory which was obtained by simulation with van der Waals particles of Poiseuille ﬂow [1]. An enlightening question emerges when a complex mechanism during the advance of the interface through a nanotube is observed. This mechanism can be ex plained by Fig. 6, where the center of mass of the water phase evolves in the ﬂow direction with the simulation time. This quasi-periodic movement of the center of mass the back-and-forth motion, has a positive Va for each cycle. Therefore, the mechanism produces a real advance of the ﬂuids. We also observe that from the constan value of the interface width for all Va (Fig, 2) and, in addition, the proﬁle velocity of the plug ﬂow (Fig. 4), the Fig. 3. a Cross-sectional density of the water phase. b Cross- sectional density of the hydrocarbon phase Fig. 5. Df versus Va of the water phase An enlightening question emerges when a complex mechanism during the advance of the interface through a nanotube is observed. This mechanism can be ex- plained by Fig. 6, where the center of mass of the water phase evolves in the ﬂow direction with the simulation time. Regular article Immiscible ﬂuid ﬂow through nanotubes This quasi-periodic movement of the center of mass, the back-and-forth motion, has a positive Va for each cycle. Therefore, the mechanism produces a real advance of the ﬂuids. We also observe that from the constant value of the interface width for all Va (Fig, 2) and, in addition, the proﬁle velocity of the plug ﬂow (Fig. 4), the Fig. 3. a Cross-sectional density of the water phase. b Cross- sectional density of the hydrocarbon phase Fig. 5. Df versus Va of the water phase Fig. 3. a Cross-sectional density of the water phase. b Cross- sectional density of the hydrocarbon phase Fig. 3. a Cross-sectional density of the water phase. b Cross- sectional density of the hydrocarbon phase Fig. 5. Df versus Va of the water phase Fig. 3. a Cross-sectional density of the water phase. b Cross- sectional density of the hydrocarbon phase Fig. 5. Df versus Va of th Fig. 5. Df versus Va of the water phase 413 Fig. 6. Center-of-mass (CMx) evolution of the water phase with simulation time Acknowledgements. This work was partly supported by the CON- ICET, Argentina. J.M. is a Scientiﬁc Researcher of CONICET. References 1. Koplik J, Banavar JR, Willemsem JF (1988) Phys Rev Lett 60:1282 2. Provata A, Prassas VD, Theorodou DN (1997) J Chem Phys 107:5125, and references there in 3. Sokhan VP, Tildesley DJ (1997) Mol Phys 92:625, and references therein 4. Koplik J, Banavar JR, Willemsem JF (1989) Phys Fluids A 1:781 5. Hummer G, Rasaiah JC, Noworyta JP (2001) Nature 414:188 d C G i S ( ) h Ch 6. Berendsen HJC, Grigera JR, Straatsma T (6269) J Phys Chem 91:1987 7. GR Oningen Molecular Simulation package 8. Murad S, Ravi P, Powles JG (1993) J Chem Phys 98:977 Fig. 6. Center-of-mass (CMx) evolution of the water phase with simulation time 9. Maran˜ o´ n Di Leo J, Maran˜ o´ n J (2003) J Mol Struct (THEO- CHEM) 623:159 ) 10. Hannon L, Lie GC, Clementi R (1986) Phys Lett A 119:174 10. Hannon L, Lie GC, Clementi R (1986) Phys Lett A 119:174 11. Lynse P (1987) J Chem Phys 86:4177 11. Lynse P (1987) J Chem Phys 86:4177 oil phase takes care of all displacements of the water phase. All this behavior could be seen by means of the molecular animation program GOPENMOL. We think that the atomic roughness and the elastical nature of the prismatic surface are possibly responsible for this movement. 12. Locker Carpenter I, Hehre WJ (1990) J Phys Chem 94:531 13. van Buuren AR, Marrink S-J, Berendsen HJC (1993) J Phys Chem 97:9206, and reference therein
https://openalex.org/W3104471925	https://link.springer.com/content/pdf/10.1007/JHEP10(2019)043.pdf	English	null	Testing dark matter with Cherenkov light — prospects of H.E.S.S. and CTA for exploring minimal supersymmetry	The Journal of high energy physics/The journal of high energy physics	2,019	cc-by	17,121	Open Access, c⃝The Authors. Article funded by SCOAP3. Published for SISSA by Springer Received: July 17, 2019 Accepted: September 9, 2019 Published: October 4, 2019 Testing dark matter with Cherenkov light — prospects of H.E.S.S. and CTA for exploring minimal supersymmetry https://doi.org/10.1007/JHEP10(2019)043 Contents 1 Introduction 1 2 Indirect detection with VHE gamma rays 3 2.1 Observation of the Galactic Center region with H.E.S.S. 3 2.2 Dark matter prospects with CTA in the inner Galactic halo 3 2.3 Statistical method for sensitivity computation 4 2.4 Expected signal and background events 5 3 The MSSM and details of the scan 8 3.1 The p9MSSM 8 3.2 p9MSSM scanning setup and constraints 8 4 Results 14 4.1 CTA sensitivity to the p9MSSM parameter space 14 4.2 Discussion of results 15 4.3 Underabundant neutralinos 19 4.4 Study of benchmark points 20 5 Conclusions 22 Contents 1 Introduction 1 2 Indirect detection with VHE gamma rays 3 2.1 Observation of the Galactic Center region with H.E.S.S. 3 2.2 Dark matter prospects with CTA in the inner Galactic halo 3 2.3 Statistical method for sensitivity computation 4 2.4 Expected signal and background events 5 3 The MSSM and details of the scan 8 3.1 The p9MSSM 8 3.2 p9MSSM scanning setup and constraints 8 4 Results 14 4.1 CTA sensitivity to the p9MSSM parameter space 14 4.2 Discussion of results 15 4.3 Underabundant neutralinos 19 4.4 Study of benchmark points 20 5 Conclusions 22 1 3 3 3 4 5 8 8 8 14 14 15 19 20 22 JHEP10(2019)043 4.3 Underabundant neutralinos 4.4 Study of benchmark points 5 Conclusions 5 Conclusions Testing dark matter with Cherenkov light — prospects of H.E.S.S. and CTA for exploring minimal supersymmetry JHEP10(2019)043 Andrzej Hryczuk,a Krzysztof Jod lowski,a Emmanuel Moulin,b Lucia Rinchiuso,b Leszek Roszkowski,c,a Enrico Maria Sessoloa and Sebastian Trojanowskid,a aNational Centre for Nuclear Research, Pasteura 7, 02-093 Warsaw, Poland bIRFU, CEA,D´epartement de Physique des Particules, Universit´e Paris-Saclay, F-91191 Gif-sur-Yvette, France cAstrocent, Nicolaus Copernicus Astronomical Center Polish Academy of Sciences, ul. Bartycka 18, 00-716 Warsaw, Poland dConsortium for Fundamental Physics, School of Mathematics and Statistics, University of Sheﬃeld, Hounsﬁeld Road, Sheﬃeld, S3 7RH, U.K. E-mail: andrzej.hryczuk@ncbj.gov.pl, krzysztof.jodlowski@ncbj.gov.pl, emmanuel.moulin@cea.fr, lucia.rinchiuso@cea.fr, leszek.roszkowski@ncbj.gov.pl, enrico.sessolo@ncbj.gov.pl, s.trojanowski@sheffield.ac.uk Andrzej Hryczuk,a Krzysztof Jod lowski,a Emmanuel Moulin,b Lucia Rinchiuso,b Leszek Roszkowski,c,a Enrico Maria Sessoloa and Sebastian Trojanowskid,a aNational Centre for Nuclear Research, Pasteura 7, 02-093 Warsaw, Poland bIRFU, CEA,D´epartement de Physique des Particules, Universit´e Paris-Saclay, F-91191 Gif-sur-Yvette, France cAstrocent, Nicolaus Copernicus Astronomical Center Polish Academy of Sciences, ul. Bartycka 18, 00-716 Warsaw, Poland dConsortium for Fundamental Physics, School of Mathematics and Statistics, University of Sheﬃeld, Hounsﬁeld Road, Sheﬃeld, S3 7RH, U.K. E-mail: andrzej.hryczuk@ncbj.gov.pl, krzysztof.jodlowski@ncbj.gov.pl, emmanuel.moulin@cea.fr, lucia.rinchiuso@cea.fr, leszek.roszkowski@ncbj.gov.pl, enrico.sessolo@ncbj.gov.pl, s.trojanowski@sheffield.ac.uk leszek.roszkowski@ncbj.gov.pl, enrico.sessolo@ncbj.gov.pl, s.trojanowski@sheffield.ac.uk Abstract: We provide an updated and improved study of the prospects of the H.E.S.S. and Cherenkov Telescope Array (CTA) experiments in testing neutralino dark matter in the Minimal Supersymmetric Standard Model with nine free parameters (p9MSSM). We include all relevant experimental constraints and theoretical developments, in particular a calculation of the Sommerfeld enhancement for both present-day annihilations and the relic abundance. We perform a state-of-the-art analysis of the CTA sensitivity with a log- likelihood test ratio statistics and apply it to a numerical scan of the p9MSSM parameter space focusing on a TeV scale dark matter. We ﬁnd that, assuming Einasto proﬁle of dark matter halo in the Milky Way, H.E.S.S. has already been able to nearly reach the so-called thermal WIMP value, while CTA will go below it by providing a further improvement of at least an order of magnitude. Both H.E.S.S. and CTA are sensitive to several cases for which direct detection cross section will be below the so-called neutrino ﬂoor, with H.E.S.S. being sensitive to most of the wino region, while CTA also covering a large fraction of the ∼1 TeV higgsino region. We show that CTA sensitivity will be further improved in the monochromatic photon search mode for both single-component and underabundant dark matter. Keywords: Supersymmetry Phenomenology ArXiv ePrint: 1905.00315 Open Access, c⃝The Authors. Article funded by SCOAP3. 1 Introduction Dark matter (DM) is the dominant component of matter in the universe but its nature re- mains unknown. Dark matter in the form of weakly interacting massive particles (WIMPs) has attracted a great deal of interest in the last decades, and a worldwide experimental eﬀort is underway to unveil its fundamental properties (for recent reviews see, e.g., [1, 2]). WIMP candidates appear in many extensions of the Standard Model (SM), among which a notable example is supersymmetry (SUSY). A multi-faceted approach has been developed to search for WIMP DM that exploits the complementarity of direct detection strategies, in which one attempts to detect WIMPs scattering oﬀthe target nuclei, indirect detec- tion, which seeks detecting products of WIMP self annihilations, and the production at high-energy colliders. The so-far null experimental searches carried out at colliders and in underground lab- oratories for the direct detection of DM have pushed the WIMP mass scale into the TeV range. In the SUSY framework, this is in agreement with expectations for the scale of soft SUSY-breaking consistent with the discovery at the LHC of a 125 GeV Higgs boson. To be able to study TeV WIMPs at colliders requires center-of-mass energy beyond that of the Large Hadron Collider (LHC), and direct detection faces the lowered number density of DM particles due to the larger DM mass. – 1 – It is in this mass regime where indirect detection with gamma rays may also play a major role. While a continuous part of the gamma-ray ﬂux is expected to drop for energies close to the DM mass, pronounced line-like features that appear there provide a distinctive signature of TeV DM over astrophysical backgrounds. The quest for such spectral features further motivates the searches carried out with instruments with large eﬀective area at TeV energies, such as ground-based arrays of Imaging Atmospheric Cherenkov Telescopes (IACTs). The currently operating IACTs H.E.S.S. [3], MAGIC [4] and VERITAS [5], as well as the Fermi-LAT [6] experiment on satellite and the ground-based water tank array HAWC [7], have done deep observation campaigns in the Galactic Center (GC) of the Milky Way and nearby dwarf galaxy satellites of the Milky Way. The next-generation Cherenkov Telescope Array (CTA) is expected to start data taking within a decade. 2.1 Observation of the Galactic Center region with H.E.S.S. The GC region is the brightest source of DM annihilation in gamma rays (for a review see e.g. [21]). However, it harbors numerous astrophysical sources that shine in very high energy (E ≳100 GeV) gamma rays. Among them are H.E.S.S. J1745-290 [22], a strong emission spatially coincident with the supermassive black hole Sagittarius A*, the super- nova/pulsar wind nebula G09+01 [23], the supernova remnant H.E.S.S. J1745-303 [24], and a diﬀuse emission extending along the Galactic plane [25, 26]. The rich observational dataset obtained from deep observations of the GC region by the H.E.S.S. phase-I in- strument has been used to look for continuum and line signals from DM annihilations. Standard analyses of H.E.S.S.-I observations of the GC region provided about 250 hours of live time in the inner 1◦of the GC. JHEP10(2019)043 H.E.S.S. searches have been performed with 10 years of data of the 4-telescope ar- ray towards the GC for the continuum [27] and mono-energetic gamma line [3] channels, respectively, using a 2-dimensional likelihood ratio test statistics to look for any possible excess over the measured background. In order to avoid modeling the complex standard astrophysical background in the GC region, a region of ±0.3◦in Galactic latitude along the Galactic plane has been excluded from the dataset together with a disk of 0.4◦ra- dius centered at the position of J1745-303. No excess in the signal region with respect to background was found and some of the strongest constraints on TeV DM were derived in various annihilation channels [3, 27]. 1 Introduction JHEP10(2019)043 The GC region is arguably the most promising astrophysical environment to detect DM annihilation signals in very high energy (VHE, E ≳100 GeV) gamma rays due to its relative proximity and the expected large accumulation of DM. However, the GC region is known to be also populated with numerous standard astrophysical emitters in VHE gamma rays. The detection of sharp spectral features expected from TeV DM annihilations would then be key to provide convincing signatures against the smoother energy spectra exhibited by astrophysical backgrounds. The purpose of this work is to improve and update on previous papers [8–16] that have explored the observational status and prospects of detecting neutralino DM within the Minimal Supersymmetric Standard Model (MSSM). In our analysis we focus on the upcoming CTA [17] and therefore on the heavy neutralino as DM in the nine-parameter version of the MSSM (p9MSSM) that will be deﬁned below. Our analysis improves previous works by: (i) deriving the projected CTA sensitivity via a state-of-the-art binned likelihood analysis to be used by the CTA Collaboration, (ii) using up-to-date experimental constraints and numerical tools that include, e.g., 13 TeV LHC data and (iii) taking into account the Sommerfeld enhancement (SE) for all points in the scan, whereas previous works included it only as an estimate or only in some selected sectors of parameter space. The paper is organized as follows. In section 2 we provide an overview of the recent input in VHE gamma-ray results from the H.E.S.S. experiment in the context of searches for heavy DM. An update of the CTA sensitivity to DM searches in the GC region using the latest Monte Carlo simulations of the CTA instrument response functions is provided. In section 3 we brieﬂy describe the p9MSSM, scanning methodology and experimental constraints applied in the analysis. In section 4 we compare the results of our scans with the reach of current and planned indirect and direct detection experiments. We stress the importance of CTA to provide coverage of one of the most interesting cases, the ∼1 TeV higgsino region, as emphasized in [18] (see also [19, 20] for a recent work and review) that otherwise would remain unexplored. Finally, we present our conclusions in section 5. – 2 – 2.2 Dark matter prospects with CTA in the inner Galactic halo The central region of the Milky Way is also a prime target for DM search with the planned CTA [17]. CTA is envisaged as a two-site observatory to be built at the Paranal site (Chile) in the Southern hemisphere and at La Palma (Spain) in the Northern hemisphere. As the GC region can be observed under favorable and eﬃcient conditions from the Southern hemisphere, the Chilean site of CTA is best suited to explore the GC region. The CTA observation strategy plans a survey of the GC region as a key-science observation program for DM searches [28]. A deep multi-year observation program is planned in the form of an extended and homogeneous survey of the inner several degrees of the GC. The CTA ﬂux sensitivity is expected to improve by up to about one order of magnitude compared to H.E.S.S. and the energy resolution reaches 15% at about 100 GeV down to better than 5% in the TeV energy range. The performance of CTA used in this work is based on instrumental Monte Carlo simulations performed for the Southern array which comprises 4 Large-Size Telescopes (LSTs), 25 Mid-Size Telescopes (MSTs), and 70 Small-Size Telescopes (SSTs). See ref. [29] for further details. Following the methodology presented in ref. [30], in this work the sensitivity to DM annihilation signals for CTA observations of the GC region is computed using the latest publicly-available instrument response functions (IRFs) of the CTA Southern site at average zenith angle 20◦[31]. The main background for IACTs measurements consists of hadronic (proton and nuclei) cosmic rays (CRs) as well as electron and positron CRs, with a dominant contribution – 3 – from the protons. In order to eﬃciently separate gamma rays from an overwhelming CR background, eﬃcient discrimination techniques based of the shower image topology have been developed [32]. However, due to ﬁnite CR rejection of IACTs, a residual background consisting of misreconstructed CRs identiﬁed as gamma rays is unavoidable. The expected residual background determination for CTA has been performed through extensive Monte Carlo simulations [29]. Here, we use the so-called prod3b version of the instrument response functions that include the residual background determinations. The region of interest for the DM signal extraction with CTA extends up to ±5◦from the GC both in Galactic longitude and latitude. The overall region is split into squared pixels of side 0.5◦. 2.2 Dark matter prospects with CTA in the inner Galactic halo A homogeneous exposure of 500 hours is assumed over the entire ﬁeld of view. The energy-diﬀerential residual background rate and acceptance are extracted from ref. [31], and the energy threshold is taken at 30 GeV. All observations are assumed to be taken at 20◦zenith angle. The IRFs depends on the chosen analysis cuts. All simulations are based on the CTA-South site performance according to an event selection optimized for 50 hours of observation. JHEP10(2019)043 The above-mentioned IRFs are provided for on-axis measurement, i.e. for emission located near the center of the ﬁeld of view (FoV). In case of emission distant from the center, the IRFs have been computed as a function of the oﬀ-axis angle and the CTA ﬂux sensitivity has been computed accordingly. The radius of the FoV region in which the ﬂux sensitivity is within a factor 2 of the one at the center is more than 3 degrees above several hundred GeV [31]. A possible CTA GC survey can make use of a regular grid of pointing positions. Provided that the distance between two nearby pointings is close enough, an overall spatially homogeneous sensitivity can be obtained. At a few degree distance, the sensitivity reached from a single pointing position degrades signiﬁcantly but is expected to be compensated by nearby pointings. An optimized and quantitative pointing position strategy for the GC survey with CTA to achieve the best possible sensitivity in the inner several degrees of the GC is much beyond the scope of this work. In what follows we will assume that an homogeneous ﬂux sensitivity in the overall region of interest with a 500 hour ﬂat exposure can be achieved provided that the overall adequate amount of observation time is granted to the GC survey to fulﬁll this goal. 2.3 Statistical method for sensitivity computation A dedicated 3-dimensional likelihood ratio test statistics technique has been developed to exploit the spectral and spatial features of the expected DM signal with respect to the background. The spatial pixels are deﬁned as squared pixels of 0.5◦between ±5◦in both Galactic longitude and latitudes. 20 energy bins are taken logarithmically-spaced between energies from 10 GeV to 100 TeV, following ref. [31]. The likelihood function for DM searches is deﬁned as a product of the Poisson proba- bilities of event counting in the signal and background regions in the i-th energy bin, j-th Galactic longitude bin, and k-th Galactic latitude bin. It reads (2.1) Lijk (sijk, bijk) = Pois (sijk + bijk, mijk) Pois (αjkbijk, nijk) , (2.1) – 4 – where the likelihood function follows a Poisson distribution given by Pois(λ, n) = λne−λ/n! while αjk corresponds to the ratio of the solid angle size of the background over the signal regions. The measured count numbers in the signal and background regions are mijk and nijk, respectively. Following H.E.S.S.’ strategy for DM searches in the GC region, the OFF region (see below) measurements are taken in the same observational and instrumental conditions as for the signal measurement, which does not require any further acceptance correction for the background determination. In this case, αjk is taken to be 1. In the context of a counting experiment using ON-OFF measurements [3], the signal is searched in an ON region where the measured number of events is mijk. The expected number of background events in the signal region, bijk, is determined from the measurement of number of events in control (OFF) regions, nijk, with no or little expected searched signal. sijk is the expected signal in the signal region. In order to compute an expected sensitivity, no excess between the ON and the OFF regions is assumed, i.e. mijk ≡nijk.where the likelihood function follows a Poisson distribution given by Pois(λ, n) = λne−λ/n! while αjk corresponds to the ratio of the solid angle size of the background over the signal regions. The measured count numbers in the signal and background regions are mijk and nijk, respectively. Following H.E.S.S.’ strategy for DM searches in the GC region, the OFF region (see below) measurements are taken in the same observational and instrumental conditions as for the signal measurement, which does not require any further acceptance correction for the background determination. 2.3 Statistical method for sensitivity computation In this case, αjk is taken to be 1. In the context of a counting experiment using ON-OFF measurements [3], the signal is searched in an ON region where the measured number of events is mijk. The expected number of background events in the signal region, bijk, is determined from the measurement of number of events in control (OFF) regions, nijk, with no or little expected searched signal. sijk is the expected signal in the signal region. In order to compute an expected sensitivity, no excess between the ON and the OFF regions is assumed, i.e. mijk ≡nijk. JHEP10(2019)043 The total likelihood function L is the product of the individual likelihood functions over each ijk bin deﬁned as L = Y ijk Lijk. The log-likelihood ratio test statistics (LLRTS) is deﬁned as: ˆ is deﬁned as: LLRTS = −2 ln L sijk,ˆˆbijk) L ˆsijk,ˆbijk) , (2.2) (2.2) where the single and double carets indicate unconditional and conditional maximization, respectively [33]. For each mass, a LLRTS is computed and a LLRTS value equal to 2.71 for one degree of freedom corresponds to an one-sided upper limit at 95% C.L. on σv0. The expected sensitivity is computed for a 100% branching ratio in each of the channels W +W −, ZZ, hh, Zh, c¯c, b¯b, t¯t, e+e−, µ+µ−, τ +τ −and γγ. 2.4 Expected signal and background events The expected photon ﬂux from pair-annihilation of DM particles of mass mDM in a region of solid angle ∆Ωin the sky can be expressed as dΦDM γ dE ∆Ω, E = σv0 8 π m2 DM dNγ(E) dE × J ∆Ω , (2.3) (2.3) – 5 – Proﬁles Einasto (E) NFW Cored Einasto (CE) ρs (GeVcm−3) 0.079 0.307 0.079 rs (kpc) 20.0 21.0 20.0 αs 0.17 − 0.17 rc (kpc) − − 3.0 Table 1. Parameters of the Einasto, NFW and Cored Einasto DM proﬁles at the GC considered in this work. Table 1. Parameters of the Einasto, NFW and Cored Einasto DM proﬁles at the GC considered in this work. where σv0 is the total annihilation cross section to all primary channels providing photons in the ﬁnal sates, and dNγ(E)/dE is the photon spectrum per annihilation. J(∆Ω) is the so-called J-factor deﬁned as the integral of the square of the DM density ρ along the line of sight s and over ∆Ωby JHEP10(2019)043 J ∆Ω ≡ Z ∆Ω dΩ Z ∞ 0 ds ρDM r(s, θ) 2 . (2.4) (2.4) s is the distance along the line of sight from the observer and is related to the radial distance r in the coordinates centered at GC by r = s2 + r2 ⊙−2 r⊙s cos θ 1/2, where θ is the angle between the direction of observation and the GC, and r⊙= 8.5 kpc is the distance from the Sun to the GC. We consider a cuspy DM distribution at the GC for which suitable parametrizations are the Einasto [34] and Navarro-Frenk-White (NFW) [35] proﬁles deﬁned as ρE(r) = ρs exp −2 αs r rs αs −1 and ρNFW(r) = ρs " r rs 1 + r rs 2#−1 , (2.5) (2.5) where normalization ρs, scale radius rs, and power index αs are given in table 1, following ref. [36]. The local DM density is taken to be ρ⊙= 0.39 GeV cm−3 [37]. Since the DM density in the GC is rather uncertain we also consider a Cored Einasto proﬁle with a core radius rc such that ρCE(r < rc) = ρE(rc) and ρCE(r ≥rc) = ρE(r). We note that the presence of possible DM substructures is known to play a subdominant role in DM searches towards the GC and is not therefore considered here (for a discussion see, e.g., ref. 2.4 Expected signal and background events [38] and references therein). The expected DM signal count number in the ijk-th bin writes as sijk = Tobs Z ∆Ei dE dΦDM γ dE (∆Ωjk, E) Aγ eﬀ(E) G (mDM −E) (2.6) (2.6) where dΦDM γ /dE is deﬁned in eq. (2.3), Aγ eﬀ(E) is the gamma-ray energy-dependent eﬀective area, Tobs is the observation time, and G(mDM −E) is a Gaussian function centered at the DM mass mDM of width taken as the CTA energy resolution in order to reproduce the eﬀect of the energy resolution on the theoretical signal spectrum. The DM spectrum dNγ(E)/dE is taken from ref. [39] for continuum channels. The monoenergetic gamma-ray line is a Dirac delta function centered at mDM. The signal count rate in the ijk-th bin is integrated over the spatial pixel of solid angle size ∆Ωjk and energy bin width ∆Ei. – 6 – 10−27 10−26 10−25 10−24 100 200 400 600 1000 2000 5000 σv0 [cm3s−1] mχ [GeV] Cored Einasto b t ⌧ Z W + h NFW b t ⌧ Z W + h Einasto b t ⌧ Z W + h Projected CTA sensitivity - 95% C.L. (a) 10−29 10−28 10−27 10−26 100 200 400 600 1000 2000 5000 σvγγ [cm3s−1] mχ [GeV] Projected CTA sensitivity to γγ - 95% C.L. H.E.S.S. 2018 (Einasto) Einasto Cored Einasto NFW (b) Figure 1. 95% C.L. CTA projected sensitivities to the velocity-weighted annihilation cross section versus DM mass mχ, derived from observations of the inner Galactic halo assuming 500 hour homogeneous exposure for three separate halo proﬁles: Cored Einasto (dashed double-dotted), NFW (dotted) and Einasto (dashed lines). (a) The CTA sensitivity is given for the speciﬁc ﬁnal states introduced in the text. (b) The CTA sensitivity is derived for a monochromatic γ line. The solid line represents the current 95% C.L. observed upper limit from H.E.S.S. obtained for the Einasto DM proﬁle [3]. Line texture is the same as in (a). 10−27 10−26 10−25 10−24 100 200 400 600 1000 2000 5000 σv0 [cm3s−1] mχ [GeV] Cored Einasto b t ⌧ Z W + h NFW b t ⌧ Z W + h Einasto b t ⌧ Z W + h Projected CTA sensitivity - 95% C.L. (a) 10−29 10−28 10−27 10−26 100 200 400 600 1000 2000 5000 σvγγ [cm3s−1] mχ [GeV] Projected CTA sensitivity to γγ - 95% C.L. H.E.S.S. 1All sensitivity limits, including more annihilation channels, for all three halo proﬁles considered here, can be found in the supplementary material on the arXiv. Limits provided there also extend to 100 TeV. 2.4 Expected signal and background events 2018 (Einasto) Einasto Cored Einasto NFW (b) (b) (a) JHEP10(2019)043 Figure 1. 95% C.L. CTA projected sensitivities to the velocity-weighted annihilation cross section versus DM mass mχ, derived from observations of the inner Galactic halo assuming 500 hour homogeneous exposure for three separate halo proﬁles: Cored Einasto (dashed double-dotted), NFW (dotted) and Einasto (dashed lines). (a) The CTA sensitivity is given for the speciﬁc ﬁnal states introduced in the text. (b) The CTA sensitivity is derived for a monochromatic γ line. The solid line represents the current 95% C.L. observed upper limit from H.E.S.S. obtained for the Einasto DM proﬁle [3]. Line texture is the same as in (a). The CR background count number is given by nijk = Tobs Z ∆Ei Z ∆Ωjk dE dΩdΓCR dEdΩ(Ω, E) , (2.7) (2.7) where dΓCR/dEdΩis the energy-diﬀerential residual background rate per steradian. The background modeling follows the Monte Carlo procedure outlined in the papers, refs. [29, 40]. 1 1 A detailed modeling of the spectral and spatial extrapolation of the Galactic Diﬀuse Emission measured by Fermi-LAT in the TeV energy range is beyond the scope of this paper and we neglect it in our computation. A band of ±0.3◦in Galactic latitudes is excluded from the ROIs as being populated by numerous standard astrophysical sources of VHE gamma rays. A 0.4◦radius disk is removed at the position of HESS J1745-303, one of the brightest TeV gamma-ray sources in the overall ROI. We present in ﬁgure 1a1 the projected CTA 95% C.L. sensitivity to DM annihilation as a function of DM mass mχ. For this ﬁgure the DM particle is assumed to annihilate into the speciﬁc SM ﬁnal states described in the legend with 100% branching fraction. The exclusion lines are computed for each of the three diﬀerent choices of the DM Galactic halo proﬁle. The CTA 95% C.L. sensitivity to monochromatic γ-ray lines for the same three choices of halo proﬁle is featured in ﬁgure 1b. Note that the current monochromatic γ-ray H.E.S.S. bound, ref. [3], is more constraining than the corresponding Fermi-LAT monochromatic bound, ref. [6], in the mass range m ∼ > 300 GeV. – 7 – 3.1 The p9MSSM The simplest realization of SUSY that is also phenomenologically viable is the R-parity conserving MSSM, where the lightest supersymmetric particle (LSP) is stable and may be identiﬁed as a thermally-produced DM candidate. We will assume that the LSP is the light- est neutralino. However, with over 100 free soft-breaking parameters, it is almost impossible (nor actually even necessary) to study the MSSM in complete generality. Therefore, one has to study more constraining models with a speciﬁc high-scale mechanism for SUSY breaking (e.g. CMSSM/mSUGRA) or to consider a p(henomenological)MSSM [9, 41]. The latter is based on the following assumptions: (i) CP conservation, (ii) Minimal Flavor Vio- lation at the electroweak scale, (iii) degenerate ﬁrst two generations of sfermion soft-mass parameters and (iv) negligible Yukawa couplings and trilinear couplings (A-terms) for the ﬁrst two generations. In our numerical scan we consider the p9MSSM where in addition to the above, we set the gluino mass, the third-generation down-type right soft squark mass, and the ﬁrst two generations of soft slepton masses at 20 TeV, which decouples them (see table 2). The p9MSSM provides a suﬃciently generic parametrization and coverage of the DM properties of the MSSM with CP and R-parity conservation. It captures a rich electroweak scale phenomenology with multiple possibilities regarding its UV-completion, while being suﬃcient for our purpose of exploring heavy neutralinos as DM. Indeed, adding more MSSM parameters to the scan would not alter our results in any signiﬁcant way. 3 The MSSM and details of the scan Low energy-scale supersymmetry (SUSY) is the most thoroughly studied scenario for new physics that provides solutions to the problems of the SM — e.g. hierarchy problem, lack of DM candidate, uniﬁcation of gauge interactions. Despite null results for any new physics signal at the LHC or direct and indirect detection experiments searching for DM, SUSY remains an attractive candidate for new physics, especially in light of the discovery at the LHC of a Higgs boson with mass not far above the Z boson mass. Indeed, the experimental data have so far only excluded models based on optimistic expectations founded on purely theoretical, or aesthetic, arguments, like naturalness. JHEP10(2019)043 3.2 p9MSSM scanning setup and constraints We apply the projected sensitivity reach of CTA as calculated in section 2 to the case of the MSSM parametrized by 9 free input parameters. The parameters that we scan over and their ranges are shown in table 2. We employ the Multinest v.2.7 [42, 43] package for the scan, using ﬂat priors. In order to ensure the best coverage of the parameter space of the model, several independent scans with 20, 000 live points each have been performed and the resulting points have been combined when presenting the results. The supersymmetric spectrum is calculated with SPheno v4.0.3 [44, 45]. We allow the bino mass M1 and the µ parameter to assume negative values in order to accommodate blind spots in DM direct detection [46, 47], which stem from the vanishing hχχ coupling for certain combinations of parameters (see also [48] for a recent discussion). The remaining gaugino mass parameter M2 is kept positive, starting from the a minimal value of 100 GeV, allowed by the LEP – 8 – Parameter Range bino mass −10 < M1 < 10 wino mass 0.1 < M2 < 10 gluino mass M3 = 20 trilinear couplings −30 < At = Ab = Aτ < 30 pseudoscalar mass 0.1 < mA < 10 µ parameter −10 < µ < 10 3rd gen. left soft squark mass 0.1 < m e Q3 < 30 3rd gen. right up soft squark mass 0.1 < m˜tR < 30 3rd gen. right down soft squark mass m˜bR = 20 1st/2nd gen. soft squark masses m e Q1,2 = m ˜dR,˜sR = 20 soft slepton masses 0.1 < m˜τR = meL3 < 10 soft slepton masses m˜eR,˜µR = meL1,2 = 20 ratio of Higgs doublet VEVs 1 < tan β < 62 Nuisance parameter Central value, error Top pole mass mt (GeV) (173.34, 0.76) [49] Table 2. Ranges of the p9MSSM parameters used in our scans. All masses and trilinear couplings are given in TeV. Parameter Range bino mass −10 < M1 < 10 wino mass 0.1 < M2 < 10 gluino mass M3 = 20 trilinear couplings −30 < At = Ab = Aτ < 30 pseudoscalar mass 0.1 < mA < 10 µ parameter −10 < µ < 10 3rd gen. left soft squark mass 0.1 < m e Q3 < 30 3rd gen. right up soft squark mass 0.1 < m˜tR < 30 3rd gen. 3.2 p9MSSM scanning setup and constraints We additionally impose a 10% theoretical uncertainty on the calculation to partially take into account the eﬀects of, e.g., loop corrections [54, 55], variations in the renormalization scheme and scale [56], and modiﬁcations to the QCD equations of state [57–59]. At the typical mass scale tested by CTA and H.E.S.S. (∼1 to a few TeV) the SE plays an important role and strongly aﬀects both the calculation of the present-day neutralino annihilation cross section, σv0, and, to a lesser degree also the determination of the thermal neutralino relic density [60–62]. An accurate treatment of the freeze-out process thus requires the incorporation of the SE coming from multiple exchanges of all the gauge bosons and of the SM Higgs and applied to all co-annihilation channels. At present, the only public code that gives the relic density with the SE included for a generic neutralino and all possible co-annihilation partners in the general MSSM is DarkSE — a package written for DarkSUSY v5 [63].2 JHEP10(2019)043 A complete numerical treatment of the SE is very CPU-expensive and thus cannot be handled automatically in a scan. Therefore, we have adopted a two-step approach: 1) in the scan we use micrOMEGAs and include the SE by rescaling the result using a grid of the enhancements in the M2-µ plane following the procedure of [10]; 2) the ﬁnal points are then post-processed with the accurate SE treatment using full DarkSE code. Sommerfeld enhancement is also included in the computation of the present-day σv0, as well as for σvγγ and σvZγ.3 Ideally, one could use an approximate simpliﬁed treatments of the SE for the ﬁrst step, as in, e.g., [69, 70], but unfortunately these are known only for simple setups — there is no known method for estimating the SE for the relic density with co- annihilations, and there is also no simple functional dependence on either the input nor physical parameters. Note that in our analysis we do not take into account possible bound-state formation of strongly interacting co-annihilating particles. This eﬀect was noticed and ﬁrst discussed for a simple toy model in a recent work [71] and potentially can apply to the regions of the MSSM parameter space featuring one or more squarks almost degenerate in mass with the neutralino, particularly if the latter lies around the TeV scale. Implementing bound-state formation in our code would go far beyond the scope of this analysis. 3.2 p9MSSM scanning setup and constraints right down soft squark mass m˜bR = 20 1st/2nd gen. soft squark masses m e Q1,2 = m ˜dR,˜sR = 20 soft slepton masses 0.1 < m˜τR = meL3 < 10 soft slepton masses m˜eR,˜µR = meL1,2 = 20 ratio of Higgs doublet VEVs 1 < tan β < 62 Nuisance parameter Central value, error Top pole mass mt (GeV) (173.34, 0.76) [49] Table 2. Ranges of the p9MSSM parameters used in our scans. All masses and trilinear couplings are given in TeV. JHEP10(2019)043 able 2. Ranges of the p9MSSM parameters used in our scans. All masses and trilinear couplings e given in TeV. Table 2. Ranges of the p9MSSM parameters used in our scans. All masses and trilinear couplings are given in TeV. bounds on charginos. Most of the third generation sfermion masses are allowed to assume a broad range of values in between being almost mass degenerate with the lightest neutralino up to tens of TeV. The former regime allows for eﬃcient co-annihilations to occur in the early Universe when the DM relic density is determined, while the latter, in case of squarks, can more easily lead to a correct value of the Higgs boson mass, mh, thanks to an increase of the characteristic SUSY scale. As discussed above, the remaining sfermion mass parameters and the gluino mass M3 are ﬁxed at 20 TeV. They do not play any real role in a further discussion. The SUSY mass parameters are deﬁned at the scale of the geometrical average of the physical stop masses, MSUSY = (m˜t1m˜t2)1/2. The ratio of the Higgs doublets’ vevs, tan β, and the top quark pole mass, mt, which is treated here as a nuisance parameter, are deﬁned at the electroweak symmetry breaking (EWSB) scale. We assume a Gaussian distribution for mt, whose central value and experimental error are given in [49], mt = (173.34±0.76) GeV. Our numerical scans are driven by a global likelihood function, which incorporates a standard set of constraints described below. Dark matter relic density. The constraint with the strongest impact on our numerical result is given by the measurement of the relic abundance of DM, as given by Planck [50], Ωχh2 = 0.120 ± 0.001. (3.1) (3.1) To calculate the relic density we employ micrOMEGAs v.5.0.6 [51, 52] supplemented – 9 – by DarkSE [53]. 2DarkSE does not take into account some recent theoretical developments relative to the most proper way of implementing the SE computation. In particular, the code includes oﬀ-diagonal terms in the annihilation matrix [64–66] exclusively in the pure wino limit. In this respect it provides a less accurate determination than that of a new program that is currently being developed [67], which has been already used in several phenomenological studies [13, 68]. However, DarkSE also presents an additional functionality of having the SE implemented for sfermion co-annihilation, which is a necessary ingredient for the scan performed in this work. 3It has been checked that the zero-velocity limit of these cross sections gives essentially the same result as when averaged over the Maxwellian velocity distribution of DM in the GC, with only minimal percent level diﬀerences in the close proximity of the SE resonance in the wino region. 3.2 p9MSSM scanning setup and constraints While this eﬀect might modify the value of the predicted relic density for some points, these could only be sporadic cases with very strong co-annihilations with squarks. Another eﬀect that in principle could have some impact on the discussed limits is the modiﬁcation of the end point of the energy spectrum of photons produced in the present-day 3It has been checked that the zero-velocity limit of these cross sections gives essentially the same result as when averaged over the Maxwellian velocity distribution of DM in the GC, with only minimal percent level diﬀerences in the close proximity of the SE resonance in the wino region. – 10 – DM annihilation due to soft and collinear gauge boson emission. Such processes, though formally of higher order, are enhanced by large Sudakov logarithms especially at energy scales much larger than the electroweak one. This has been noticed in the context of DM annihilation in [72] and explicitly seen in the wino annihilation computation at one-loop [73] while ﬁnally approached with resummation techniques in [74–78]. In [79, 80] it was argued that for the neutralino annihilation the modiﬁcation due to fully resummed exclusive cross section is most relevant in multi-TeV regime. However, more recently [77, 78] showed that for the nearly whole energy regime of interest for CTA, high precision calculations in fact would require resummation of Sudakov logarithms. Nevertheless, due to the fact that it is currently not possible to directly apply the framework of [79] or [77] to generic p9MSSM neutralinos, and since the expected corrections in the DM mass range of our interest are typically much lower than astrophysical uncertainties, we do not include this eﬀect in our scan. JHEP10(2019)043 When performing the numerical scans, we study two commonly discussed cases: 1. the thermal relic density saturates eq. (3.1), in which case we use a Gaussian distri- bution for the relic density, 1. the thermal relic density saturates eq. (3.1), in which case we use a Gaussian distri- bution for the relic density, 2. the thermal relic density does not exceed the value given in eq. (3.1), in which case we use a half-Gaussian distribution — with relic density imposed only as an upper bound. In the former case, we assume that no deviations from the standard cosmological history of the Universe took place, as well as that the lightest neutralino is the only DM particle. 3.2 p9MSSM scanning setup and constraints In the future, the neutralino scattering cross section on neutrons and protons can also be constrained by their interactions inside neutron stars and white dwarfs [19, 97]. The corresponding limits, however, depend on additional astrophysical assumptions, as well as progress in observations and, therefore, they are not discussed further below. JHEP10(2019)043 Collider constraints. The TeV mass-range particle spectrum of the MSSM is very poorly constrained by direct SUSY searches at colliders (see, e.g., [14, 15, 98, 99] and references therein), including the most recent data from the LHC. In our case, since we focus on the parameter space characterized by colored sparticles lying in the multi-TeV range, only very few points are aﬀected by LHC bounds, with negligible impact on the results shown in section 4. For completeness, we also take into account LEP and Tevatron limits on SUSY particles [100]. Higgs physics. The Higgs mass determination and Higgs-sector LHC measurements in general can show their eﬀect on the MSSM parameter space under probe in DM searches. Indirect constraints on the stop mass and mixing from the Higss mass measurement aﬀect the extension of the regions potentially subject to stop co-annihilation; bounds on the mass and couplings of heavy Higgs bosons can end up inﬂuencing somewhat the shape of the funnel regions. In here, the Higgs sector is constrained with HiggsBounds−5.2.0beta [101, 102] and HiggsSignals −2.2.1beta [103], while additional constraints from searches for heavy Higgs decays to τ +τ −are implemented following [104, 105]. Flavor physics. We calculate a few ﬂavor observables with Superiso Relic v4.0 [86]. The parameter space of the MSSM is potentially sensitive, in particular, to the bounds from rare decays in b →sll processes and radiative decays like b →sγ, which can constrain scan points characterized by large tan β values, and/or relatively light non-SM Higgs bosons, squarks, and charginos/neutralinos. We use the following experimental determinations: BR (B →Xsγ) = (3.27 ± 0.14) × 10−4, (3.2) BR B0 s →µ+µ− = 3.0 ± 0.6+0.3 −0.2 × 10−9 (3.3) (3.2) (3.3) (3.2) ( ) (3.3) where, following, e.g., ref. [106], in eq. (3.2) we give the calculated average [107] of the determinations in refs. [108–112], and in eq. (3.3) we report the most recent LHCb mea- surement, based on 8 TeV collision data [113]. We thus implicitly assume that eq. 3.2 p9MSSM scanning setup and constraints In the latter case, the neutralino cannot be a single particle comprising the DM. We then present results of CTA sensitivity to underabundant neutralinos with local density rescaled by the square of the ratio of the neutralino density to the Planck [50] value. On the other hand, the neutralino relic density can also be signiﬁcantly aﬀected by deviations from the standard cosmological history of the Universe, e.g., if neutralino freeze- out occurs during an extended reheating period [81–83] (see also [84, 85] for recent studies) or in presence of additional non-thermal production. In this case, the neutralino can be a single DM particle even though its standard freeze-out relic density does not saturate the Planck value. In order to accommodate such scenarios, we additionally present results for the aforementioned case 2 but without rescaling σv0. Dark matter direct detection. The steady progress observed in recent years in direct detection (DD) searches for DM in underground liquid noble gas detectors has led to strong upper limits on the spin-independent cross section of the neutralino scattering oﬀnucleons. We include the most recent DD bounds here. For this we employ SuperIso Relic v4.0 [86] and the DDCalc v.2.0.0 package [87], assuming the Standard Halo Model (SHM) and the following values for the relevant astrophysical parameters: ρ0 = 0.39 GeV/cm3, vrot = 220 km/s, vesc = 544 km/s. We note that slight modiﬁcations to the SHM that might be suggested by e.g. recent data release by the GAIA Collabora- tion [88], see e.g. [89, 90] for further discussion, would have minor impact on our results. – 11 – The experimental limits that we take into account are the following: PandaX-2 [91], PICO- 60 [92], and the most recent results from the XENON1T collaboration [93]. Almost universally in the parameter space of the MSSM the bounds on the spin- dependent cross section of the neutralino scattering oﬀthe proton or neutron cannot com- pete with the corresponding DD bounds on the spin-independent cross section. The current bounds on σSD p for mχ of our interest come from the searches by the IceCube Collabora- tion for neutrinos coming from the center of the Milky Way [94], the Earth [95], and the Sun [96]. Since they are less constraining than the aforementioned DD bounds, and also the indirect detection bounds described below, we do not consider them here. 3.2 p9MSSM scanning setup and constraints (3.3) has superseded an older statistical combination of CMS and LHCb measurements with 7 and 8 TeV data [114]. Note that very recently the ATLAS Collaboration has presented a – 12 – measurement of BR B0 s →µ+µ− , from a combination of data taken during their 8 TeV and 13 TeV runs, which agrees with eq. (3.3): BR B0 s →µ+µ− = (2.8+0.8 −0.7) × 10−9 [115]. We impose the bounds of eqs. (3.2) and (3.3) at the 95% C.L., a posteriori on the points belonging to the 2 σ region of the proﬁle likelihood. This reduces the number of viable points in the scan by approximately 2%. Other potentially relevant ﬂavor observables like BR(B± →τντ) or the Bs mass mixing measurement ∆MBs are not constraining at the mass scale relevant for this paper. Note that we do not include constraints from observables that are currently showing a 2–3 σ discrepancy with the SM, like the diﬀerential branching ratios and angular ob- servables in B0 →K∗0µ+µ−[116, 117], or the branching ratio measurements that have recently provided tantalizing hints of lepton ﬂavor nonuniversality [118–122]. It is known that these anomalies cannot be explained consistently in the MSSM (see, e.g., ref. [123]) and that, if conﬁrmed to higher statistical signiﬁcance with further release of data, will require new physics beyond the particle content of the MSSM. For analogous reasons, we do not apply to the parameter space the constraint from the measurement [124] of the muon anomalous magnetic moment, which shows a 3.5 σ discrepancy with the SM expectation, δ (g −2)µ = (27.4 ± 7.6) × 10−10 [125]. It is well known that this value cannot be accom- modated in the regions of the MSSM parameter space that feature a TeV-scale LSP, see, e.g., ref. [126]. In this case too, if the anomaly were to be conﬁrmed by upcoming Fermilab data [127], it will require a BSM explanation lying outside of the MSSM parameter space relevant for the current analysis. One should keep in mind, however, that is possible to extend the MSSM minimally by a U(1) gauge group, so that the all of the above-mentioned ﬂavor anomalies become consistent with TeV-scale neutralinos with the exact same DM properties as in the vanilla MSSM [128]. JHEP10(2019)043 Dark matter indirect detection. 3.2 p9MSSM scanning setup and constraints The indirect detection constraints on neutralino DM, that are the main subject of this study, are not included in the likelihood function when performing initial numerical scans of the parameter space of the MSSM. Instead, we carefully study them by postprocessing the results obtained in these scans. This leads to a better understanding of their impact on the allowed parameter space. The most constraining data for the TeV-scale mass range are currently provided by H.E.S.S. A more detailed description of DM ID limits from H.E.S.S. and future projections has been described in details in section 2. When presenting the results below, we also take into account Fermi-LAT limits on DM-induced γ-rays that correspond to 6 years of data and observation of 28 dSphs [129]. These data are in principle most constraining in the MSSM for neutralinos of mixed gauge composition with a mass of a few hundred GeV, which are, however, already strongly bounded by the null DD results. They might also provide a complementary probe on the low-energy tail of spectra from the annihilation of winos including SE. We illustrate this below for a ﬁxed annihilation ﬁnal state into a b¯b pair. We have also veriﬁed numerically, following Superiso Relic v4.0 [86], that taking into account a complete list of annihilation ﬁnal states leads to similar results. – 13 – We note that σv can also be constrained by requiring that the CMB spectrum is not aﬀected too much by the pre- and post-recombination energy injection from DM annihila- tions [130–132]. However, for the heavy DM of our interest, this eﬀect typically leads to less stringent bounds than null searches for DM annihilation signal in the GC by H.E.S.S. and in dSphs by Fermi-LAT (see [13] for recent discussion). A recent determination [133] of the bounds on the neutralino annihilation cross section from AMS-02 antiproton cosmic-ray (CR) data [134] has proven to be competitive with H.E.S.S. diﬀuse γ-ray searches in the ∼TeV mass range. We have veriﬁed that this is in general true also in the context of our scans using SuperIso Relic v4.0 [86] that employs a semi-analytic approach to solving the propagation equations following [135]. However, the limits that one derives from the AMS-02 data depend on the assumed CR propagation model and suﬀers from large astrophysical uncertainties (see, e.g., refs. [136, 137]). 3.2 p9MSSM scanning setup and constraints For this reason, we do not discuss them in details in the following section, which focuses on DM-induced γ-ray signal. JHEP10(2019)043 4.1 CTA sensitivity to the p9MSSM parameter space 4.1 CTA sensitivity to the p9MSSM parameter space For each point in the scan, we compute the H.E.S.S. limit for the present-day annihila- tion cross section, σv0, and the corresponding projected sensitivity of CTA. We use the 95% C.L. bounds and projections for annihilation to pure channels (see ﬁgure 1a). In case of annihilation ﬁnal states for which H.E.S.S. limits have not been reported by the collab- oration, we employ the most relevant existing bounds. In particular, for hh ﬁnal state we use ZZ limit, for ﬁnal states with c and s quarks — b¯b limit, for the lightest quarks — τ +τ −limit and for e+e−we employ µ+µ−limit. Instead, for CTA we derive bounds for a more complete set of annihilation ﬁnal states, as discussed in section 2.4. In order to verify whether a particular point in the p9MSSM parameter space is within current bounds and future sensitivities, we combine limits obtained for pure annihilation ﬁnal states by taking their average weighted by the branching-ratios corresponding to those channels. In section 4.4 we show that this procedure is suﬃcient for our purpose, by comparing our results for several benchmark scenarios to a more detailed treatment in which photon spectra are carefully combined prior to obtaining the CTA limit. For channels with non-SM particles in the ﬁnal state, e.g., the neutral MSSM Higgs particles, A0 and H, we employ the bounds computed for the SM Higgs; for the charged MSSM Higgs particle, H±, we use the bounds derived for W ±. While the non-SM annihi- lation ﬁnal states typically do not play a dominant role in our analysis, they might become important for selected points in the parameter space. For these points, we have veriﬁed our results against a more detailed procedure in which decays of the non-SM particles were taken into account employing HDECAY [138, 139] before generating the combined photon spectrum using ref. [39]. In the plots below we only show points that belong to the 95% C.L. region of the global proﬁle-likelihood, i.e. we select ∆χ2 ≤5.99 from the best-ﬁt point, where ∆χ2 = −2 ln (L/Lmax). – 14 – Figure 2. Distribution of p9MSSM points with ∆χ2 ≤5.99 in the (mχ, σv0) plane. The color coding reﬂects the composition of the lightest neutralino, as discussed in the text and according to the legend. The current upper 95% C.L. limits from H.E.S.S. 4.1 CTA sensitivity to the p9MSSM parameter space [27] applied to the p9MSSM are indicated with a black solid line. The projected CTA sensitivity applied to the p9MSSM is shown as a thick (Einasto), or thin (Cored Einasto) dashed double-dotted line. All points above the line will be probed at the ∼95% C.L. The Fermi-LAT [129] b¯b mode from dwarf spheroidal galaxies is shown as a dashed line. To highlight the complementarity between the continuous and monochromatic photon search, we denote the points whose σvγγ is within reach (assuming Einasto halo proﬁle) at CTA by dark gray triangles. JHEP10(2019)043 Figure 2. Distribution of p9MSSM points with ∆χ2 ≤5.99 in the (mχ, σv0) plane. The color coding reﬂects the composition of the lightest neutralino, as discussed in the text and according to the legend. The current upper 95% C.L. limits from H.E.S.S. [27] applied to the p9MSSM are indicated with a black solid line. The projected CTA sensitivity applied to the p9MSSM is shown as a thick (Einasto), or thin (Cored Einasto) dashed double-dotted line. All points above the line will be probed at the ∼95% C.L. The Fermi-LAT [129] b¯b mode from dwarf spheroidal galaxies is shown as a dashed line. To highlight the complementarity between the continuous and monochromatic photon search, we denote the points whose σvγγ is within reach (assuming Einasto halo proﬁle) at CTA by dark gray triangles. 4.2 Discussion of results We present in ﬁgure 2 the scan points in the plane (mχ, σv0) of the present-day annihilation cross section of the neutralino versus its mass. The color code used in ﬁgure 2 refers to the gauge composition of the neutralino LSP, which, by construction, in the MSSM is never a 100% pure eigenstate. How “pure” a certain mass eigentate is depends on the elements of the unitary matrix, Z, diagonalizing the neutralino mass matrix after EWSB. In green we show the points with the LSP containing at least 90% of the pure bino gauge eigenstate (i.e., that is, in the basis of gauge eigenstes {bino, wino, down-type higgsino, up-type higgsino}, we require \|Z11\|2 > 0.9 for these points). In blue, the points for which it is for at least 90% a wino (\|Z12\|2 > 0.9). In cyan we show a mixture of these two gaugino states, with the additional constraint that the higgsino composition remain below 10%, \|Z13\|2 + \|Z14\|2 < 0.1. In – 15 – red we show neutralinos that are dominated for at least 90% by their higgsino fraction (\|Z13\|2 + \|Z14\|2 > 0.9). We ﬁnally point out that only very few points characterized by a mixture of a gaugino and a large higgsino component appear, marked in gold, in the plot, as they are in strong tension with the latest bounds from direct detection searches. We also note that due to our general focus on TeV-scale neutralino DM, which is of most relevance for H.E.S.S and CTA, our scanning procedure is not tuned to thoroughly explore the parameter space of the p9MSSM corresponding to light neutralinos with masses around the EWSB scale. For this reason, we do not show in our plots points corresponding to the region where mχ ≈mh/2 where the correct neutralino DM relic density can be obtained thanks to eﬃcient resonance annihilations via the Higgs boson exchange. We note, however, that the expected annihilation cross section for such light neutralino DM lies well below the reach of CTA. The same is also true for another instance of supersymmetry at the electroweak scale that has recently been discussed in the context of a collection of mild excesses present in the LHC data [99]. JHEP10(2019)043 We show in ﬁgure 2 with a solid black line the current 95% C.L. 4.2 Discussion of results upper bound on σv0 from 254 hours of observation of the GC at H.E.S.S., under the Einasto proﬁle assumption, applied to the points of the p9MSSM. Importantly, when deriving these results, as well as CTA sensitivity lines discussed below, we take into account all the points obtained in the scan of the parameter space including the ones that violate some of other constraints and, therefore, are not shown in the plot. In particular, the presence of such otherwise excluded points allows us to determine the position of the H.E.S.S. limit in the region with a light neutralino and large σv0 which is virtually excluded by current bounds. The latest observations exclude points whose neutralino is strongly dominated by the wino component (in blue, and some in cyan), for which the annihilation cross section in the present day has a large SE [140, 141]. The plot updates ﬁgure 5(a) of ref. [10] and is in agreement with e.g. ref. [11]. Compositions of the neutralino very close to a pure wino state are in very strong tension with H.E.S.S. with continuum observations as well as monochromatic line searches. The H.E.S.S. bound, on the other hand, does not bite into the ∼1 TeV (nearly pure) higgsino region of the parameter space, corresponding to the red points in ﬁgure 2, for which the SE is less pronounced. Upcoming increased statistics can tighten the bound but, realistically, batches of new data are at this point not expected to bring qualitative improvements to the current picture. It is CTA, with an eﬀective area that is by about a factor 10 larger than that of H.E.S.S.’ at 1 TeV, that will be probing more deeply into the higgsino region of the parameter space. We show with a dash-dotted black line our projection of the sensitivity of CTA in the p9MSSM in searches for DM-induced diﬀuse photon ﬂux, with 500 hours of observation of the GC and under the Einasto proﬁle assumption. In addition, we overlap gray triangles to the points that are within the sensitivity of the CTA γ-ray line search. As was described in section 2, we factor in a detailed treatment of the statistical uncertainties, and the likelihood function is calculated with an improved design of the ROIs of the Galactic Plane with respect to previous analyses [10, 142]. 4.2 Discussion of results The higgsino region of the parameter space is likely to be tested in its near entirety by CTA, and the same is true for points with – 16 – bino-dominated neutralinos with annihilation cross section around the thermal freeze-out value (light dotted line). We note that the actual CTA limit on the p9MSSM parameter space cannot be per- fectly represented by a single line due to number of possible neutralino DM annihilation ﬁnal states that need to be taken into account. However, we have veriﬁed that, for mχ ≳1 TeV the approximate limit that we present reproduces very well the true CTA sensitivity. For lower masses, the line shown in ﬁgure 2 corresponds to a conservative approach, i.e., all the points lying above the line are within the CTA reach. We follow a similar strategy to obtain the approximate H.E.S.S. limit shown as a solid line in ﬁgure 2. The points that will remain untested, almost all characterized by a nearly pure bino- like composition of the LSP, are those for which the neutralino annihilation cross section is too small to yield the correct relic density, and thus either feature spectra with sparticles that co-annihilate in the early Universe with the LSP (near mass degeneracy between the bino-like neutralino and one or more sfermions), or spectra that include one or more Higgs bosons of mass within a few hundred GeV of 2mχ, which provide a means for funnels, or resonant s-channel annihilation of the LSP in the early Universe due to the thermal broadening of the energy distribution. As is well known, the speciﬁcs of these spectra are very model-dependent. Moreover, their realization in explicit high-scale completions can encounter model building challenges and/or require some ﬁne tuning of the initial parameters. JHEP10(2019)043 This is unlike in the case of (nearly pure) higgsinos and winos, which do fall inside the sensitivity of large IACTs, and for which the correct value of the relic density emerges naturally once the mass of the LSP is around either 1 TeV, or ∼2.5 −3 TeV, respectively, quite independently of the model details of the rest of the sparticle spectrum. Note, however, that for higgsino points with masses larger than about 1.3 TeV, shown outside of the CTA sensitivity in ﬁgure 2, one also relies on additional mechanisms like coannihilations with squarks in the early Universe to preserve the correct relic density. 4.2 Discussion of results These points tend to feature lower present-day annihilation cross section than lighter higgsinos, and they are consequently more diﬃcult to probe. The projected sensitivity of CTA shown in ﬁgure 2 is obtained in the two limiting cases of Einasto and Cored Einasto DM halo proﬁles. A sensitivity line corresponding to the NFW proﬁle can be easily obtained by multiplying the projected line for Einasto case by the factor of about 2.5 obtained from ﬁgure 1a. In ﬁgure 2, we also show with the dot-dashed line the projected CTA sensitivity reach obtained for the Cored Einasto proﬁle deﬁned in section 2.4. As can be seen, in this case CTA can still play an important role by probing the entire wino-like neutralino DM scenario which would otherwise remained not fully tested by the H.E.S.S. observations. In ﬁgure 3 we show the p9MSSM points in the (mχ, σSI p ) plane. The most recent XENON1T 90% C.L. upper limit [93] is shown by a solid violet line. The XENON1T results are included in the global likelihood function, and that explains the absence above the line of any point belonging to the ∼2σ region of the proﬁle likelihood. The onset of the irreducible neutrino background is denoted by a solid black line. The color code is the same as in ﬁgure 2 and we additionally overlap violet triangles to points excluded by the – 17 – Figure 3. Distribution of p9MSSM points with ∆χ2 ≤5.99 in the (mχ, σSI p ) plane with color coding as in ﬁgure 2. Points excluded by H.E.S.S. (Einasto, both continuous and monochromatic photon search) are denoted by violet triangles, while those within the sensitivity of CTA (Einasto, both continuous and monochromatic photon search) are denoted by black triangles. The most recent limit from the XENON1T Collaboration [93], which is included in the likelihood function, is denoted by a purple solid line, while the onset of the irreducible neutrino background is denoted by a black solid line. JHEP10(2019)043 Figure 3. Distribution of p9MSSM points with ∆χ2 ≤5.99 in the (mχ, σSI p ) plane with color coding as in ﬁgure 2. Points excluded by H.E.S.S. (Einasto, both continuous and monochromatic photon search) are denoted by violet triangles, while those within the sensitivity of CTA (Einasto, both continuous and monochromatic photon search) are denoted by black triangles. 4.2 Discussion of results The most recent limit from the XENON1T Collaboration [93], which is included in the likelihood function, is denoted by a purple solid line, while the onset of the irreducible neutrino background is denoted by a black solid line. H.E.S.S. bound on σv0. Black triangles are overlapped to points within our projection of the sensitivity of CTA in the Einasto proﬁle. The necessity of using both direct and indirect detection strategies to cover the most substantial portions of the parameter space of the MSSM with high-mass DM has been pointed out in the literature since early after the discovery of the Higgs boson at the LHC [10]. We show the power of complementarity of direct and indirect detection in ﬁg- ure 4, where we project the points of the p9MSSM to the (σSI p , σv0) plane. The color code is the same as in ﬁgure 2 and ﬁgure 3. The future reach of direct underground searches with noble liquids is bound to bite into the parameter space from right to left, until it reaches the irreducible neutrino background, shown here as a shaded region (recall that the value of σSI p characteristic of the neutrino “ﬂoor” for direct DM searches depends on the DM mass, hence the boundary of the shaded area is jagged in ﬁgure 4). To guide the eye, we add a vertical dashed gray line denoting the neutrino background limit σSI p ≈4 · 10−12 pb taken at mχ ≈2 TeV. Conversely, the sensitivity of IACTs gradually improves from the top down, providing a complementary means of testing the parameter space. The H.E.S.S. bound is denoted in the ﬁgure by a dashed black horizontal line while the projected sensitivity of CTA is denoted by a dashed double-dotted horizontal line. – 18 – Figure 4. The points of the p9MSSM in the (σSI p , σv0) plane. Color coding is the same as in ﬁgure 2. The upper limit on σSI p is determined by the current sensitivity of XENON1T, included in the global likelihood function. The shaded region covers the points lying below the irreducible neutrino ﬂoor. To guide the eye we add a vertical dashed line with reference value of the neutrino background limit taken at mχ = 2 TeV. The dashed horizontal line denotes the H.E.S.S. 95% C.L. 4.3 Underabundant neutralinos As discussed in section 3.2, the neutralino can be a good DM candidate even when its thermally produced relic abundance is diﬀerent from the total DM relic density in the Universe. It can then either be one of several DM components, or might even remain the only DM particle but in non-standard cosmological scenarios. In this subsection, we present the results of two scans corresponding to the cases in which the relic density constraint is imposed as an upper bound only, by means of a half Gaussian distribution. The corresponding results can be seen in ﬁgures 5a and 5b where only the points that belong to the 95% C.L. region of the global proﬁle-likelihood are shown in the (mχ, σv0) plane. In ﬁgure 5a we rescale σv0 by (Ωχh2/0.12)2 which corresponds to the case when neu- tralino DM can provide only a partial contribution to the total ΩDMh2. Similarly, we rescale the DM DD cross section σSI p by Ωχh2/0.12 when imposing the corresponding con- straints. As can be seen in the plot, underabundant higgsino-like and wino-like neutralinos with masses of order few hundred GeV are typically beyond the reach of CTA. There are, however, some higgsino-like points that can be probed by the CTA monochromatic photon (σvγγ + 1 2σvγZ) search even though these points lie below the projected CTA sensitivity in searches for DM-induced diﬀuse photon spectrum (dash-dotted line in the plot). These points are denoted by gray triangles. The crucial impact of the monochromatic line search As discussed in section 3.2, the neutralino can be a good DM candidate even when its thermally produced relic abundance is diﬀerent from the total DM relic density in the Universe. It can then either be one of several DM components, or might even remain the only DM particle but in non-standard cosmological scenarios. In this subsection, we present the results of two scans corresponding to the cases in which the relic density constraint is imposed as an upper bound only, by means of a half Gaussian distribution. The corresponding results can be seen in ﬁgures 5a and 5b where only the points that belong to the 95% C.L. region of the global proﬁle-likelihood are shown in the (mχ, σv0) plane. In ﬁgure 5a we rescale σv0 by (Ωχh2/0.12)2 which corresponds to the case when neu- tralino DM can provide only a partial contribution to the total ΩDMh2. 4.2 Discussion of results upper limit, taken at mχ ≈2.5 TeV, while the dashed double-dotted horizontal line denotes the approximate CTA reach, taken at mχ ≈1 TeV. JHEP10(2019)043 Figure 4. The points of the p9MSSM in the (σSI p , σv0) plane. Color coding is the same as in ﬁgure 2. The upper limit on σSI p is determined by the current sensitivity of XENON1T, included in the global likelihood function. The shaded region covers the points lying below the irreducible neutrino ﬂoor. To guide the eye we add a vertical dashed line with reference value of the neutrino background limit taken at mχ = 2 TeV. The dashed horizontal line denotes the H.E.S.S. 95% C.L. upper limit, taken at mχ ≈2.5 TeV, while the dashed double-dotted horizontal line denotes the approximate CTA reach, taken at mχ ≈1 TeV. 4.3 Underabundant neutralinos Similarly, we rescale the DM DD cross section σSI p by Ωχh2/0.12 when imposing the corresponding con- straints. As can be seen in the plot, underabundant higgsino-like and wino-like neutralinos with masses of order few hundred GeV are typically beyond the reach of CTA. There are, however, some higgsino-like points that can be probed by the CTA monochromatic photon (σvγγ + 1 2σvγZ) search even though these points lie below the projected CTA sensitivity in searches for DM-induced diﬀuse photon spectrum (dash-dotted line in the plot). These points are denoted by gray triangles. The crucial impact of the monochromatic line search – 19 – (a) (b) Figure 5. (a) Distribution of points with ∆χ2 ≤5.99 in (mχ, (Ωχh2/0.12)2 · σv0) plane for underabundant neutralinos. Color and line texture coding is the same as in ﬁgure 2. (b) Same as in (a) but without rescaling. (a) (b) JHEP10(2019)043 (b) (a) Figure 5. (a) Distribution of points with ∆χ2 ≤5.99 in (mχ, (Ωχh2/0.12)2 · σv0) plane for underabundant neutralinos. Color and line texture coding is the same as in ﬁgure 2. (b) Same as in (a) but without rescaling. is even more pronounced for heavier neutralinos with masses mχ ≈1 TeV. In particular, it is worth stressing that, e.g., an underabundant wino-like neutralino DM with mχ ≈1 TeV can be discovered by CTA in monochromatic-line searches with no corresponding signal in the diﬀuse spectrum searches. For even heavier, but still underabundant, wino-like neutralinos, CTA can provide a good way of indirectly detecting them in both types of searches. In ﬁgure 5b we show the results that correspond to a scenario with the neutralino being the only DM particle and having its production in the early Universe supplemented by, e.g., some non-thermal contribution. Notably, this allows one to consider neutralino DM with signiﬁcantly larger values of the annihilation cross section and, therefore, much better prospects for discovery in future indirect searches. In particular, in this scenario CTA could easily discover higgsino-like neutralino DM with the mass of order a few hundred GeV in both diﬀuse photon and monochromatic-line searches. As can be seen in the plot, the Fermi-LAT limits [129] bite into the low mass region of the parameter space, where IACTs lose sensitivity. 4.3 Underabundant neutralinos This is illustrated in ﬁgure 5b by a dashed line for ﬁxed annihilation ﬁnal state into a b¯b pair, which well represents the position of the exclusion bound we would obtain when imposing Fermi-LAT as a constraint in the likelihood. This scenario is also independently constrained by DD searches of DM, which are taken into account in our scanning procedure. 4.4 Study of benchmark points In the previous section we have computed the H.E.S.S. limits and CTA sensitivity in the p9MSSM by combining the bounds shown in ﬁgure 1a weighted by the branching fractions to the appropriate ﬁnal states. We have already noted, however, that, in principle, a more robust procedure should be applied. It would involve summing over all weighted spectra of annihilation ﬁnal states and then using up-to-date instrument response functions and background estimates to obtain the limit as described in detail in section 2. The full – 20 – Benchmark BM1 BM2 BM3 BM4 BM5 BM6 BM7 points mχ 1099 1765 1840 531 1516 2288 997 [GeV] Branching W +W −0.23 bb 0.35 W +W −0.64 bb 0.85 tt 0.17 τ +τ −0.26 τ +τ −0.39 fractions bb 0.23 W +W −0.29 hA 0.14 τ +τ −0.14 bb 0.16 γτ +τ −0.22 tt 0.37 tt 0.21 ZZ 0.24 ZH 0.14 tt 0.01 hA 0.16 bb 0.21 bb 0.22 ZZ 0.20 τ +τ −0.05 γW +W −0.08 ZH 0.16 γµ+µ−0.14 γτ +τ −0.01 Zh 0.06 γW +W −0.04 W +H−0.16 γe+e−0.13 τ +τ −0.04 Zh 0.03 W −H+ 0.157 tt 0.03 Ωχh2 0.10 0.16 0.13 0.13 - - - Main χ± 1 χχ →SM χ± 1 , χ2 χχ →SM χ± 1 , χ2 χχ →SM slepton mechanism coann. A-funnel coann. t-channel coann. t-channel coann. σv0 1.98 · 10−26 8.18 · 10−27 1.08 · 10−26 1.12 · 10−26 7.63 · 10−28 4.54 · 10−31 2.14 · 10−32 [cm3s−1] σv95% C.L. (simpliﬁed) 0 7.64 · 10−27 7.82 · 10−27 6.59 · 10−27 7.29 · 10−27 8.84 · 10−27 7.82 · 10−27 6.65 · 10−27 [cm3s−1] σv95% C.L. (full calc.) 0 7.92 · 10−27 8.05 · 10−27 6.06 · 10−27 7.47 · 10−27 9.34 · 10−27 6.50 · 10−27 4.87 · 10−27 [cm3s−1] JHEP10(2019)043 Table 3. Selected benchmark points characterized by diﬀerent properties. The main annihilation mechanism at freeze-out corresponds to the ﬁnal state with the largest branching ratio. The theo- retical value of the cross section σv0 is given as well as the relic density Ωχh2. CTA sensitivity is reported for the simpliﬁed and full scheme computation. The sensitivity is expressed as 95% C.L. upper limits. The LLRTS value is derived according to eq. (2.2) for the given σv0 in the two computation schemes. Table 3. Selected benchmark points characterized by diﬀerent properties. 4.4 Study of benchmark points The color code tracks the branching fractions for dominant present-day annihilation channels, while the height of the column gives the theoretical value of the cross section σv0. The relative diﬀerence between the full and simpliﬁed procedure is highlighted in percentage. Benchmark points to the right of the vertical dashed line do not yield the correct relic abundance and feature annihilation cross section much below the projected limits. They are characterized by a large number of diﬀerentiated annihilation channels and, in particular, include large fraction to τ +τ −. These are the spectra producing the maximal diﬀerence between the two computational methods. It follows as a consequence that in the physically relevant region analyzed in section 4.2 the simpliﬁed procedure gives a very good approximation of the full calculation. 1 % 2 % 0 % -8 % 8 % -16 % -26 % x10-1 x10-5 x10-4 higgsino bino bino bino- wino full simplified BM1 1.099TeV BM2 1.765TeV BM3 1.84TeV BM4 0.531TeV BM5 1.516TeV BM6 2.288TeV BM7 0.997TeV 10-27 10-26 2x10-26 2x10-27 5x10-27 σv [cm3/s] W-H+ hH ZH Zh γe+e- γμ+μ- γτ+τ- τ+τ- bb tt ZZ W+W- JHEP10(2019)043 Figure 6. Comparison between σv95% C.L. (full calc.) 0 and σv95% C.L. (simpliﬁed) 0 for the 7 Figure 6. Comparison between σv0 and σv0 for the 7 selected representative benchmark points given in table 3. The color code tracks the branching fractions for dominant present-day annihilation channels, while the height of the column gives the theoretical value of the cross section σv0. The relative diﬀerence between the full and simpliﬁed procedure is highlighted in percentage. Benchmark points to the right of the vertical dashed line do not yield the correct relic abundance and feature annihilation cross section much below the projected limits. They are characterized by a large number of diﬀerentiated annihilation channels and, in particular, include large fraction to τ +τ −. These are the spectra producing the maximal diﬀerence between the two computational methods. It follows as a consequence that in the physically relevant region analyzed in section 4.2 the simpliﬁed procedure gives a very good approximation of the full calculation. of detecting neutralino DM within the p9MSSM, even if their relic density was altered in the desired way by assuming a modiﬁed cosmological history. of detecting neutralino DM within the p9MSSM, even if their relic density was altered in the desired way by assuming a modiﬁed cosmological history. 4.4 Study of benchmark points However, it is interesting to note that this discrepancy is not due to the low statistics of the signal coming from neutralinos with a small value of the annihilation cross section. It actually persists if one multiplies σv0 by an appropriate factor that brings σv0 close to the projected CTA sensitivity reach. Therefore, it could potentially also aﬀect some analyses performed for other models of new physics in which a particle DM candidate features mixed leptonic-hadronic ﬁnal annihilation states, and could lead to a sizable discrepancy between the true reach of indirect detection experiments and the one determined by the simpliﬁed approach (or similar). 4.4 Study of benchmark points The main annihilation mechanism at freeze-out corresponds to the ﬁnal state with the largest branching ratio. The theo- retical value of the cross section σv0 is given as well as the relic density Ωχh2. CTA sensitivity is reported for the simpliﬁed and full scheme computation. The sensitivity is expressed as 95% C.L. upper limits. The LLRTS value is derived according to eq. (2.2) for the given σv0 in the two computation schemes. procedure, on the other hand, has the disadvantage of being extremely time and CPU consuming. In this section, we test the simpliﬁed treatment against the more accurate one for some carefully selected representative benchmark scenarios. For that purpose, we choose 7 benchmark points with diﬀerent neutralino properties and diverse annihilation ﬁnal states. The physical properties of these points are summarized in table 3. Among these points, BM5−BM7 fail to provide the thermally produced relic abundance consistent with Ωχh2 ≈0.12 in the standard freeze-out scenario, but could do this, e.g., in modifed cosmological scenarios. The last two rows of the table show the diﬀerence between the 95% C.L. CTA projected sensitivities computed with the simpliﬁed and full procedures. The dependence of this diﬀerence on ﬁnal states and speciﬁc branching ratios is shown in ﬁgure 6. We ﬁnd good agreement, better than 10%, for typical points corresponding to higgsino-like, wino-like, mixed bino-wino and some pure bino-like neutralinos. The biggest discrepancy (up to 25%) occurs for BM6 and BM7 which are bino-like neutralinos that annihilate primarily to leptons (note the diﬀerent shape of the limit for the τ +τ − annihilation ﬁnal state in ﬁgure 1a) but that also exhibit a signiﬁcant branching fraction into hadronic ﬁnal states. Such points are not found in ﬁgure 2–ﬁgure 5b as their thermal relic density would overclose the Universe. Moreover, their σv0 is orders of magnitude below the CTA sensitivity, hence they would be irrelevant for determining CTA prospects – 21 – 1 % 2 % 0 % -8 % 8 % -16 % -26 % x10-1 x10-5 x10-4 higgsino bino bino bino- wino full simplified BM1 1.099TeV BM2 1.765TeV BM3 1.84TeV BM4 0.531TeV BM5 1.516TeV BM6 2.288TeV BM7 0.997TeV 10-27 10-26 2x10-26 2x10-27 5x10-27 σv [cm3/s] W-H+ hH ZH Zh γe+e- γμ+μ- γτ+τ- τ+τ- bb tt ZZ W+W- Figure 6. Comparison between σv95% C.L. (full calc.) 0 and σv95% C.L. (simpliﬁed) 0 for the 7 selected representative benchmark points given in table 3. 5 Conclusions In this work we performed an updated and improved study of the reach of CTA in test- ing neutralino DM in minimal supersymmetric scenarios. The results were compared with the most recent bounds on σv0, as a function of DM mass, obtained by H.E.S.S. We con- ducted the analysis in the framework of the 9-parameter MSSM, or p9MSSM. We included – 22 – the most recent constraints from DM direct detection searches, ﬂavor physics, and Higgs searches, and constructed a state-of-the-art likelihood ratio test statistic approach to an- alyze the CTA sensitivity. The direct constraints on sparticle masses from the LHC are also included, although they are known to be of very limited impact for the parameter space leading to TeV-scale DM. Furthermore, on the theoretical side we reﬁned the cal- culations of DM relic abundance and present-day annihilation cross section by taking into account the Sommerfeld enhancement eﬀect for a completely generic mixed neutralino and its co-annihilation partners. In particular, for the ﬁrst time sfermion co-annihilations were considered with Sommerfeld eﬀect included in a scanning framework. Having all these improvements implemented, we performed numerical scans of the p9MSSM parameter space focusing on a TeV scale neutralino DM. We ﬁnd that, assuming the Einasto proﬁle of DM halo in the Milky Way, H.E.S.S. has been able to nearly reach the so-called thermal WIMP value, while CTA will go below it by providing a further improvement of at least an order of magnitude. The results show that both H.E.S.S. and CTA are sensitive to several cases for which direct detection cross section will be below the so-called neutrino ﬂoor, with H.E.S.S. being sensitive to most of the wino region, while CTA also covering a large fraction of the 1 TeV higgsino region. We additionally show the extent to which the CTA sensitivity will be further improved in the monochromatic photon search mode for both single-component and underabundant DM. JHEP10(2019)043 While we focused on the Einasto proﬁle when presenting the results for the p9MSSM, we also studied two other DM proﬁles, namely the standard NFW proﬁle and the version of the Einasto proﬁle with a core with conservative radius rc = 3 kpc, for which we presented the most up-to-date CTA sensitivities in searches relevant for a number of ﬁxed annihilation ﬁnal states. These can be easily combined to derive actual results for any model of new physics predicting heavy WIMP DM. 5 Conclusions In particular, when applied to the p9MSSM, the aforementioned Cored Einasto proﬁle leads to substantially weaker current bounds and future sensitivity reaches. In this case, the H.E.S.S. limits do not completely exclude the region of the parameter space with wino-like neutralino DM. Instead, CTA will be able to fully probe this important scenario. Acknowledgments We would like to thank Alexander Pukhov for providing a reﬁned micrOMEGAs module that calculates annihilation cross section to monochromatic photons. We would like to thank Martin Vollmann for useful discussions. This research has made use of the CTA instrument response functions provided by the CTA Consortium and Observatory, see http://www.cta- observatory.org/science/cta-performance/ (version prod3b-v1) for more details. K.J., L.Rosz. and S.T. are supported by the National Science Centre (NCN) research grant No. 2015/18/A/ST2/00748. L.Rosz. is also supported by the National Science Centre research grant No. 2016/22/M/ST9/00583 and by “AstroCeNT: Particle Astrophysics Science and Technology Centre” project that is carried out within the International Research Agendas programme of the Foundation for Polish Science co-ﬁnanced by the European Union under the European Regional Development Fund. E.M.S. is supported in part by the National – 23 – Science Centre (Poland) under the research Grant No. 2017/26/D/ST2/00490. S.T. is sup- ported in part by the Lancaster-Manchester-Sheﬃeld Consortium for Fundamental Physics under STFC grant: ST/P000800/1. The use of the CIS computer cluster at the National Centre for Nuclear Research in Warsaw is gratefully acknowledged. Open Access. This article is distributed under the terms of the Creative Commons Attribution License (CC-BY 4.0), which permits any use, distribution and reproduction in any medium, provided the original author(s) and source are credited. References JHEP10(2019)043 [1] G. Arcadi et al., The waning of the WIMP? A review of models, searches and constraints, Eur. Phys. J. C 78 (2018) 203 [arXiv:1703.07364] [INSPIRE]. [2] L. Roszkowski, E.M. Sessolo and S. Trojanowski, WIMP dark matter candidates and searches — current status and future prospects, Rept. Prog. Phys. 81 (2018) 066201 [arXiv:1707.06277] [INSPIRE]. [3] HESS collaboration, Search for γ-Ray Line Signals from Dark Matter Annihilations in the Inner Galactic Halo from 10 Years of Observations with H.E.S.S., Phys. Rev. Lett. 120 (2018) 201101 [arXiv:1805.05741] [INSPIRE]. [4] J. Aleksi´c et al., Optimized dark matter searches in deep observations of Segue 1 with MAGIC, JCAP 02 (2014) 008 [arXiv:1312.1535] [INSPIRE]. [4] J. Aleksi´c et al., Optimized dark matter searches in deep observations of Segue 1 with MAGIC, JCAP 02 (2014) 008 [arXiv:1312.1535] [INSPIRE]. [5] VERITAS collaboration, Dark Matter Constraints from a Joint Analysis of Dwarf Spheroidal Galaxy Observations with VERITAS, Phys. Rev. D 95 (2017) 082001 [arXiv:1703.04937] [INSPIRE]. [5] VERITAS collaboration, Dark Matter Constraints from a Joint Analysis of Dwarf Spheroidal Galaxy Observations with VERITAS, Phys. Rev. D 95 (2017) 082001 [arXiv:1703.04937] [INSPIRE]. [6] Fermi-LAT collaboration, Updated search for spectral lines from Galactic dark matter interactions with pass 8 data from the Fermi Large Area Telescope, Phys. Rev. D 91 (2015) 122002 [arXiv:1506.00013] [INSPIRE]. [7] HAWC collaboration, A Search for Dark Matter in the Galactic Halo with HAWC, JCAP 02 (2018) 049 [arXiv:1710.10288] [INSPIRE]. [8] A. Fowlie, K. Kowalska, L. Roszkowski, E.M. Sessolo and Y.-L.S. Tsai, Dark matter and collider signatures of the MSSM, Phys. Rev. D 88 (2013) 055012 [arXiv:1306.1567] [INSPIRE]. [9] M. Cahill-Rowley, J.L. Hewett, A. Ismail and T.G. Rizzo, Lessons and prospects from the pMSSM after LHC Run I, Phys. Rev. D 91 (2015) 055002 [arXiv:1407.4130] [INSPIRE]. [10] L. Roszkowski, E.M. Sessolo and A.J. Williams, Prospects for dark matter searches in the pMSSM, JHEP 02 (2015) 014 [arXiv:1411.5214] [INSPIRE]. [11] M.E. Cabrera-Catalan, S. Ando, C. Weniger and F. Zandanel, Indirect and direct detection prospect for TeV dark matter in the nine parameter MSSM, Phys. Rev. D 92 (2015) 035018 [arXiv:1503.00599] [INSPIRE]. [12] ATLAS collaboration, Summary of the ATLAS experiment’s sensitivity to supersymmetry after LHC Run 1 — interpreted in the phenomenological MSSM, JHEP 10 (2015) 134 [arXiv:1508.06608] [INSPIRE]. [13] M. Beneke, A. Bharucha, A. Hryczuk, S. Recksiegel and P. Ruiz-Femenia, The last refuge of mixed wino-Higgsino dark matter, JHEP 01 (2017) 002 [arXiv:1611.00804] [INSPIRE]. References – 24 – [14] A. Arbey, M. Boudaud, F. Mahmoudi and G. Robbins, Robustness of dark matter constraints and interplay with collider searches for New Physics, JHEP 11 (2017) 132 [arXiv:1707.00426] [INSPIRE]. [15] GAMBIT collaboration, A global ﬁt of the MSSM with GAMBIT, Eur. Phys. J. C 77 (2017) 879 [arXiv:1705.07917] [INSPIRE]. [16] S. Abel, D.G. Cerde˜no and S. Robles, The Power of Genetic Algorithms: what remains of the pMSSM?, arXiv:1805.03615 [INSPIRE]. [17] CTA Consortium collaboration, Science with the Cherenkov Telescope Array, World Scientiﬁc, Singapore (2019) [arXiv:1709.07997] [INSPIRE]. JHEP10(2019)043 JHEP10(2019)043 [18] K. Kowalska, L. Roszkowski, E.M. Sessolo, S. Trojanowski and A.J. Williams, Looking for supersymmetry: 1 TeV WIMP and the power of complementarity in LHC and dark matter searches, in Proceedings of 50th Rencontres de Moriond, QCD and high energy interactions, La Thuile Italy (2015), pg. 195 [arXiv:1507.07446] [INSPIRE]. [19] R. Krall and M. Reece, Last Electroweak WIMP Standing: Pseudo-Dirac Higgsino Status and Compact Stars as Future Probes, Chin. Phys. C 42 (2018) 043105 [arXiv:1705.04843] [INSPIRE]. [20] K. Kowalska and E.M. Sessolo, The discreet charm of higgsino dark matter — a pocket review, Adv. High Energy Phys. 2018 (2018) 6828560 [arXiv:1802.04097] [INSPIRE]. [21] C. van Eldik, Gamma rays from the Galactic Centre region: a review, Astropart. Phys. 71 (2015) 45 [arXiv:1505.06055] [INSPIRE]. [22] H.E.S.S. collaboration, Spectrum and variability of the Galactic Center VHE gamma-ray source HESS J1745-290, Astron. Astrophys. 503 (2009) 817 [arXiv:0906.1247] [INSPIRE]. [23] H.E.S.S. collaboration, Very high energy gamma rays from the composite SNR G0.9+0.1, Astron. Astrophys. 432 (2005) L25 [astro-ph/0501265] [INSPIRE]. [24] H.E.S.S. collaboration, Exploring a SNR/Molecular Cloud Association Within HESS J1745-303, Astron. Astrophys. 483 (2008) 509 [arXiv:0803.2844] [INSPIRE]. [25] H.E.S.S. collaboration, Discovery of very-high-energy gamma-rays from the galactic centr ridge, Nature 439 (2006) 695 [astro-ph/0603021] [INSPIRE]. [26] H.E.S.S. collaboration, Acceleration of petaelectronvolt protons in the Galactic Centre, Nature 531 (2016) 476 [arXiv:1603.07730] [INSPIRE]. [27] H.E.S.S. collaboration, Search for dark matter annihilations towards the inner Galactic halo from 10 years of observations with H.E.S.S, Phys. Rev. Lett. 117 (2016) 111301 [arXiv:1607.08142] [INSPIRE]. [28] E. Moulin et al., Dark Matter Programme, in Science with the Cherenkov Telescope Array, World Scientiﬁc, Singapore (2019), pg. 45. [29] T. Hassan et al., Monte Carlo Performance Studies for the Site Selection of the Cherenkov Telescope Array, Astropart. Phys. 93 (2017) 76 [arXiv:1705.01790] [INSPIRE]. [30] L. Rinchiuso et al., Hunting for Heavy Winos in the Galactic Center, Phys. Rev. References D 98 (2018) 123014 [arXiv:1808.04388] [INSPIRE]. [31] CTA’s expected baseline performance, https://www.cta-observatory.org/science/cta-performance/. [31] CTA’s expected baseline performance, https://www.cta-observatory.org/science/cta-performance/. – 25 – [32] F. Aharonian, J. Buckley, T. Kifune and G. Sinnis, High energy astrophysics with ground-based gamma ray detectors, Rept. Prog. Phys. 71 (2008) 096901 [INSPIRE]. [33] G. Cowan, K. Cranmer, E. Gross and O. Vitells, Asymptotic formulae for likelihood-based tests of new physics, Eur. Phys. J. C 71 (2011) 1554 [Erratum ibid. C 73 (2013) 2501] [arXiv:1007.1727] [INSPIRE]. [34] J. Einasto, On the Construction of a Composite Model for the Galaxy and on the Determination of the System of Galactic Parameters, Trudy Alma-Ata Astroﬁz. Inst. 5 (1965) 87. [35] J.F. Navarro, C.S. Frenk and S.D.M. White, A Universal density proﬁle from hierarchical clustering, Astrophys. J. 490 (1997) 493 [astro-ph/9611107] [INSPIRE]. JHEP10(2019)043 [36] L. Pieri, J. Lavalle, G. Bertone and E. Branchini, Implications of High-Resolution Simulations on Indirect Dark Matter Searches, Phys. Rev. D 83 (2011) 023518 [arXiv:0908.0195] [INSPIRE]. [37] R. Catena and P. Ullio, A novel determination of the local dark matter density, JCAP 08 (2010) 004 [arXiv:0907.0018] [INSPIRE]. [38] J.M. Gaskins, A review of indirect searches for particle dark matter, Contemp. Phys. 57 (2016) 496 [arXiv:1604.00014] [INSPIRE]. [39] M. Cirelli et al., PPPC 4 DM ID: A Poor Particle Physicist Cookbook for Dark Matter Indirect Detection, JCAP 03 (2011) 051 [Erratum ibid. 1210 (2012) E01] [arXiv:1012.4515] [INSPIRE]. [40] CTA Consortium collaboration, Monte Carlo design studies for the Cherenkov Telescope Array, Astropart. Phys. 43 (2013) 171 [arXiv:1210.3503] [INSPIRE]. [41] MSSM Working Group collaboration, The Minimal supersymmetric standard model: Group summary report, in GDR (Groupement De Recherche) — Supersymetrie, Montpellier France (1998) [hep-ph/9901246] [INSPIRE]. [42] F. Feroz and M.P. Hobson, Multimodal nested sampling: an eﬃcient and robust alternative to MCMC methods for astronomical data analysis, Mon. Not. Roy. Astron. Soc. 384 (2008) 449 [arXiv:0704.3704] [INSPIRE]. [43] F. Feroz, M.P. Hobson and M. Bridges, MultiNest: an eﬃcient and robust Bayesian inference tool for cosmology and particle physics, Mon. Not. Roy. Astron. Soc. 398 (2009) 1601 [arXiv:0809.3437] [INSPIRE]. [44] W. Porod, SPheno, a program for calculating supersymmetric spectra, SUSY particle decays and SUSY particle production at e+e−colliders, Comput. Phys. Commun. 153 (2003) 275 [hep-ph/0301101] [INSPIRE]. [45] W. Porod and F. Staub, SPheno 3.1: Extensions including ﬂavour, CP-phases and models beyond the MSSM, Comput. Phys. Commun. 183 (2012) 2458 [arXiv:1104.1573] [INSPIRE]. [46] J.R. Ellis, A. Ferstl and K.A. References Olive, Reevaluation of the elastic scattering of supersymmetric dark matter, Phys. Lett. B 481 (2000) 304 [hep-ph/0001005] [INSPIRE]. [47] C. Cheung, L.J. Hall, D. Pinner and J.T. Ruderman, Prospects and Blind Spots for Neutralino Dark Matter, JHEP 05 (2013) 100 [arXiv:1211.4873] [INSPIRE]. – 26 – [48] T. Han, F. Kling, S. Su and Y. Wu, Unblinding the dark matter blind spots, JHEP 02 (2017) 057 [arXiv:1612.02387] [INSPIRE]. [49] ATLAS, CDF, CMS, D0 collaboration, First combination of Tevatron and LHC measurements of the top-quark mass, arXiv:1403.4427 [INSPIRE]. [50] Planck collaboration, Planck 2018 results. VI. Cosmological parameters, arXiv:1807.06209 [INSPIRE]. [51] G. B´elanger, F. Boudjema, A. Pukhov and A. Semenov, MicrOMEGAs: A Program for calculating the relic density in the MSSM, Comput. Phys. Commun. 149 (2002) 103 [hep-ph/0112278] [INSPIRE]. JHEP10(2019)043 JHEP10(2019)043 [52] G. B´elanger, F. Boudjema, A. Pukhov and A. Semenov, MicrOMEGAs: Version 1.3, Comput. Phys. Commun. 174 (2006) 577 [hep-ph/0405253] [INSPIRE]. [53] A. Hryczuk, The Sommerfeld enhancement for scalar particles and application to sfermion co-annihilation regions, Phys. Lett. B 699 (2011) 271 [arXiv:1102.4295] [INSPIRE]. [54] N. Baro, F. Boudjema and A. Semenov, Full one-loop corrections to the relic density in the MSSM: A Few examples, Phys. Lett. B 660 (2008) 550 [arXiv:0710.1821] [INSPIRE]. [55] N. Baro, F. Boudjema, G. Chalons and S. Hao, Relic density at one-loop with gauge boson pair production, Phys. Rev. D 81 (2010) 015005 [arXiv:0910.3293] [INSPIRE]. [56] J. Harz, B. Herrmann, M. Klasen, K. Kovarik and P. Steppeler, Theoretical uncertainty of the supersymmetric dark matter relic density from scheme and scale variations, Phys. Rev. D 93 (2016) 114023 [arXiv:1602.08103] [INSPIRE]. [57] M. Hindmarsh and O. Philipsen, WIMP dark matter and the QCD equation of state, Phys. Rev. D 71 (2005) 087302 [hep-ph/0501232] [INSPIRE]. [58] M. Laine and Y. Schr¨oder, Quark mass thresholds in QCD thermodynamics, Phys. Rev. D 73 (2006) 085009 [hep-ph/0603048] [INSPIRE]. [59] M. Drees, F. Hajkarim and E.R. Schmitz, The Eﬀects of QCD Equation of State on the Relic Density of WIMP Dark Matter, JCAP 06 (2015) 025 [arXiv:1503.03513] [INSPIRE]. [60] J. Hisano, S. Matsumoto, M. Nagai, O. Saito and M. Senami, Non-perturbative eﬀect on thermal relic abundance of dark matter, Phys. Lett. B 646 (2007) 34 [hep-ph/0610249] [INSPIRE]. [61] M. Cirelli, A. Strumia and M. Tamburini, Cosmology and Astrophysics of Minimal Dark Matter, Nucl. Phys. B 787 (2007) 152 [arXiv:0706.4071] [INSPIRE]. [62] A. Hryczuk, R. Iengo and P. References Ullio, Relic densities including Sommerfeld enhancements in the MSSM, JHEP 03 (2011) 069 [arXiv:1010.2172] [INSPIRE]. [63] T. Bringmann, J. Edsj¨o, P. Gondolo, P. Ullio and L. Bergstr¨om, DarkSUSY 6: An Advanced Tool to Compute Dark Matter Properties Numerically, JCAP 07 (2018) 033 [arXiv:1802.03399] [INSPIRE]. [64] M. Beneke, C. Hellmann and P. Ruiz-Femenia, Non-relativistic pair annihilation of nearly mass degenerate neutralinos and charginos I. General framework and S-wave annihilation, JHEP 03 (2013) 148 [Erratum ibid. 1310 (2013) 224] [arXiv:1210.7928] [INSPIRE]. [65] M. Beneke, C. Hellmann and P. Ruiz-Femenia, Non-relativistic pair annihilation of nearly mass degenerate neutralinos and charginos III. Computation of the Sommerfeld enhancements, JHEP 05 (2015) 115 [arXiv:1411.6924] [INSPIRE]. – 27 – [66] M. Beneke, C. Hellmann and P. Ruiz-Femenia, Heavy neutralino relic abundance with Sommerfeld enhancements - a study of pMSSM scenarios, JHEP 03 (2015) 162 [arXiv:1411.6930] [INSPIRE]. [67] M. Beneke et al., in preparation. [67] M. Beneke et al., in preparation. [68] M. Beneke et al., Relic density of wino-like dark matter in the MSSM, JHEP 03 (2016) 119 [arXiv:1601.04718] [INSPIRE]. [69] T.R. Slatyer, The Sommerfeld enhancement for dark matter with an excited state, JCAP 02 (2010) 028 [arXiv:0910.5713] [INSPIRE]. [70] S. El Hedri, A. Kaminska and M. de Vries, A Sommerfeld Toolbox for Colored Dark Sectors, Eur. Phys. J. C 77 (2017) 622 [arXiv:1612.02825] [INSPIRE]. JHEP10(2019)043 [71] J. Harz and K. Petraki, Radiative bound-state formation in unbroken perturbative non-Abelian theories and implications for dark matter, JHEP 07 (2018) 096 [arXiv:1805.01200] [INSPIRE]. [72] P. Ciafaloni, D. Comelli, A. Riotto, F. Sala, A. Strumia and A. Urbano, Weak Corrections are Relevant for Dark Matter Indirect Detection, JCAP 03 (2011) 019 [arXiv:1009.0224] [INSPIRE]. [73] A. Hryczuk and R. Iengo, The one-loop and Sommerfeld electroweak corrections to the Wino dark matter annihilation, JHEP 01 (2012) 163 [Erratum ibid. 1206 (2012) 137] [arXiv:1111.2916] [INSPIRE]. [74] M. Baumgart, I.Z. Rothstein and V. Vaidya, Calculating the Annihilation Rate of Weakly Interacting Massive Particles, Phys. Rev. Lett. 114 (2015) 211301 [arXiv:1409.4415] [INSPIRE]. [75] M. Bauer, T. Cohen, R.J. Hill and M.P. Solon, Soft Collinear Eﬀective Theory for Heavy WIMP Annihilation, JHEP 01 (2015) 099 [arXiv:1409.7392] [INSPIRE]. [76] G. Ovanesyan, T.R. Slatyer and I.W. Stewart, Heavy Dark Matter Annihilation from Eﬀective Field Theory, Phys. Rev. Lett. 114 (2015) 211302 [arXiv:1409.8294] [INSPIRE]. [77] M. Beneke, A. Broggio, C. Hasner and M. Vollmann, Energetic γ-rays from TeV scale dark matter annihilation resummed, Phys. Lett. References B 786 (2018) 347 [arXiv:1805.07367] [INSPIRE]. [78] M. Beneke, A. Broggio, C. Hasner, K. Urban and M. Vollmann, Resummed photon spectrum from dark matter annihilation for intermediate and narrow energy resolution, JHEP 08 (2019) 103 [arXiv:1903.08702] [INSPIRE]. [79] M. Baumgart et al., Resummed Photon Spectra for WIMP Annihilation, JHEP 03 (2018) 117 [arXiv:1712.07656] [INSPIRE]. [80] M. Baumgart et al., Precision Photon Spectra for Wino Annihilation, JHEP 01 (2019) 036 [arXiv:1808.08956] [INSPIRE]. [81] G.F. Giudice, E.W. Kolb and A. Riotto, Largest temperature of the radiation era and its cosmological implications, Phys. Rev. D 64 (2001) 023508 [hep-ph/0005123] [INSPIRE]. [82] N. Fornengo, A. Riotto and S. Scopel, Supersymmetric dark matter and the reheating temperature of the universe, Phys. Rev. D 67 (2003) 023514 [hep-ph/0208072] [INSPIRE]. – 28 – [83] G. Gelmini, P. Gondolo, A. Soldatenko and C.E. Yaguna, The Eﬀect of a late decaying scalar on the neutralino relic density, Phys. Rev. D 74 (2006) 083514 [hep-ph/0605016] [INSPIRE]. [84] L. Roszkowski, S. Trojanowski and K. Turzy´nski, Neutralino and gravitino dark matter with low reheating temperature, JHEP 11 (2014) 146 [arXiv:1406.0012] [INSPIRE]. [85] M. Drees and F. Hajkarim, Neutralino Dark Matter in Scenarios with Early Matter Domination, JHEP 12 (2018) 042 [arXiv:1808.05706] [INSPIRE]. [86] A. Arbey, F. Mahmoudi and G. Robbins, SuperIso Relic v4: A program for calculating dark matter and ﬂavour physics observables in Supersymmetry, Comput. Phys. Commun. 239 (2019) 238 [arXiv:1806.11489] [INSPIRE]. JHEP10(2019)043 [87] The GAMBIT Dark Matter Workgroup collaboration, DarkBit: A GAMBIT module for computing dark matter observables and likelihoods, Eur. Phys. J. C 77 (2017) 831 [arXiv:1705.07920] [INSPIRE]. [88] Gaia collaboration, Gaia Data Release 2, Astron. Astrophys. 616 (2018) A1 [arXiv:1804.09365] [INSPIRE]. [89] L. Necib, M. Lisanti and V. Belokurov, Inferred Evidence For Dark Matter Kinematic Substructure with SDSS-Gaia, arXiv:1807.02519 [INSPIRE]. [90] N.W. Evans, C.A.J. O’Hare and C. McCabe, Reﬁnement of the standard halo model for dark matter searches in light of the Gaia Sausage, Phys. Rev. D 99 (2019) 023012 [arXiv:1810.11468] [INSPIRE]. [91] PandaX-II collaboration, Dark Matter Results From 54-Ton-Day Exposure of PandaX-II Experiment, Phys. Rev. Lett. 119 (2017) 181302 [arXiv:1708.06917] [INSPIRE]. [92] PICO collaboration, Dark Matter Search Results from the PICO-60 C3F8 Bubble Chamber, Phys. Rev. Lett. 118 (2017) 251301 [arXiv:1702.07666] [INSPIRE]. [93] XENON collaboration, Dark Matter Search Results from a One Ton-Year Exposure of XENON1T, Phys. Rev. Lett. 121 (2018) 111302 [arXiv:1805.12562] [INSPIRE]. References [94] IceCube collaboration, Search for Neutrinos from Dark Matter Self-Annihilations in the center of the Milky Way with 3 years of IceCube/DeepCore, Eur. Phys. J. C 77 (2017) 627 [arXiv:1705.08103] [INSPIRE]. [95] IceCube collaboration, First search for dark matter annihilations in the Earth with the IceCube Detector, Eur. Phys. J. C 77 (2017) 82 [arXiv:1609.01492] [INSPIRE]. [96] IceCube collaboration, Search for annihilating dark matter in the Sun with 3 years of IceCube data, Eur. Phys. J. C 77 (2017) 146 [Erratum ibid. C 79 (2019) 214] [arXiv:1612.05949] [INSPIRE]. [97] M. Baryakhtar, J. Bramante, S.W. Li, T. Linden and N. Raj, Dark Kinetic Heating of Neutron Stars and An Infrared Window On WIMPs, SIMPs and Pure Higgsinos, Phys. Rev. Lett. 119 (2017) 131801 [arXiv:1704.01577] [INSPIRE]. [98] K. Kowalska, Phenomenological MSSM in light of new 13 TeV LHC data, Eur. Phys. J. C 76 (2016) 684 [arXiv:1608.02489] [INSPIRE]. [99] GAMBIT collaboration, Combined collider constraints on neutralinos and charginos, Eur. Phys. J. C 79 (2019) 395 [arXiv:1809.02097] [INSPIRE]. – 29 – [100] Particle Data Group collaboration, Review of Particle Physics, Phys. Rev. D 98 (2018) 030001. [101] P. Bechtle, O. Brein, S. Heinemeyer, G. Weiglein and K.E. Williams, HiggsBounds: Confronting Arbitrary Higgs Sectors with Exclusion Bounds from LEP and the Tevatron, Comput. Phys. Commun. 181 (2010) 138 [arXiv:0811.4169] [INSPIRE]. [102] P. Bechtle, S. Heinemeyer, O. St˚al, T. Stefaniak and G. Weiglein, Applying Exclusion Likelihoods from LHC Searches to Extended Higgs Sectors, Eur. Phys. J. C 75 (2015) 421 [arXiv:1507.06706] [INSPIRE]. [103] P. Bechtle, S. Heinemeyer, O. St˚al, T. Stefaniak and G. Weiglein, HiggsSignals: Confronting arbitrary Higgs sectors with measurements at the Tevatron and the LHC, Eur. Phys. J. C 74 (2014) 2711 [arXiv:1305.1933] [INSPIRE]. JHEP10(2019)043 [104] ATLAS collaboration, Search for Minimal Supersymmetric Standard Model Higgs Bosons H/A in the ττ ﬁnal state in up to 13.3 fb−1 of pp collisions at √s = 13 TeV with the ATLAS Detector, ATLAS-CONF-2016-085 (2016). [105] CMS collaboration, Search for a neutral MSSM Higgs boson decaying into ττ with 12.9 fb−1 of data at √s = 13 TeV, CMS-PAS-HIG-16-037 (2016). [106] The GAMBIT Flavour Workgroup collaboration, FlavBit: A GAMBIT module for computing ﬂavour observables and likelihoods, Eur. Phys. J. C 77 (2017) 786 [arXiv:1705.07933] [INSPIRE]. [107] M. Misiak and M. Steinhauser, Weak radiative decays of the B meson and bounds on MH± in the Two-Higgs-Doublet Model, Eur. Phys. J. C 77 (2017) 201 [arXiv:1702.04571] [INSPIRE]. References [108] BaBar collaboration, Measurement of the B →Xsγ branching fraction and photon energy spectrum using the recoil method, Phys. Rev. D 77 (2008) 051103 [arXiv:0711.4889] [INSPIRE]. [109] BaBar collaboration, Exclusive Measurements of b →sγ Transition Rate and Photon Energy Spectrum, Phys. Rev. D 86 (2012) 052012 [arXiv:1207.2520] [INSPIRE]. [110] BaBar collaboration, Precision Measurement of the B →Xsγ Photon Energy Spectrum, Branching Fraction and Direct CP Asymmetry ACP (B →Xs+dγ), Phys. Rev. Lett. 109 (2012) 191801 [arXiv:1207.2690] [INSPIRE]. [111] Belle collaboration, Measurement of the ¯B →Xsγ Branching Fraction with a Sum of Exclusive Decays, Phys. Rev. D 91 (2015) 052004 [arXiv:1411.7198] [INSPIRE]. [112] Belle collaboration, Measurement of the inclusive B →Xs+dγ branching fraction, photon energy spectrum and HQE parameters, in Proceedings of 38th International Conference on High Energy Physics (ICHEP 2016), Chicago U.S.A. (2016) [arXiv:1608.02344] [INSPIRE]. [113] LHCb collaboration, Measurement of the B0 s →µ+µ−branching fraction and eﬀective lifetime and search for B0 →µ+µ−decays, Phys. Rev. Lett. 118 (2017) 191801 [arXiv:1703.05747] [INSPIRE]. [114] CMS and LHCb collaborations, Observation of the rare B0 s →µ+µ−decay from the combined analysis of CMS and LHCb data, Nature 522 (2015) 68 [arXiv:1411.4413] [INSPIRE]. – 30 – [115] ATLAS collaboration, Study of the rare decays of B0 s and B0 mesons into muon pairs using data collected during 2015 and 2016 with the ATLAS detector, JHEP 04 (2019) 098 [arXiv:1812.03017] [INSPIRE]. [116] LHCb collaboration, Diﬀerential branching fractions and isospin asymmetries of B →K(∗)µ+µ−decays, JHEP 06 (2014) 133 [arXiv:1403.8044] [INSPIRE]. [117] LHCb collaboration, Angular analysis of the B0 →K∗0µ+µ−decay using 3 fb−1 of integrated luminosity, JHEP 02 (2016) 104 [arXiv:1512.04442] [INSPIRE]. [118] BaBar collaboration, Measurement of an Excess of ¯B →D(∗)τ −¯ντ Decays and Implications for Charged Higgs Bosons, Phys. Rev. D 88 (2013) 072012 [arXiv:1303.0571] [INSPIRE]. JHEP10(2019)043 [119] LHCb collaboration, Measurement of the ratio of branching fractions B( ¯B0 →D∗+τ −¯ντ)/B( ¯B0 →D∗+µ−¯νµ), Phys. Rev. Lett. 115 (2015) 111803 [Erratum ibid. 115 (2015) 159901] [arXiv:1506.08614] [INSPIRE]. [120] Belle collaboration, Measurement of the τ lepton polarization and R(D∗) in the decay ¯B →D∗τ −¯ντ, Phys. Rev. Lett. 118 (2017) 211801 [arXiv:1612.00529] [INSPIRE]. [121] LHCb collaboration, Test of lepton universality with B0 →K∗0ℓ+ℓ−decays, JHEP 08 (2017) 055 [arXiv:1705.05802] [INSPIRE]. [122] LHCb collaboration, Search for lepton-universality violation in B+ →K+ℓ+ℓ−decays, Phys. Rev. Lett. 122 (2019) 191801 [arXiv:1903.09252] [INSPIRE]. [123] W. Altmannshofer and D.M. Straub, New physics in b →s transitions after LHC run 1, Eur. Phys. J. References C 75 (2015) 382 [arXiv:1411.3161] [INSPIRE]. [124] Muon g-2 collaboration, Final Report of the Muon E821 Anomalous Magnetic Moment Measurement at BNL, Phys. Rev. D 73 (2006) 072003 [hep-ex/0602035] [INSPIRE]. [125] M. Davier, Update of the Hadronic Vacuum Polarisation Contribution to the muon g-2, Nucl. Part. Phys. Proc. 287-288 (2017) 70 [arXiv:1612.02743] [INSPIRE]. [126] K. Kowalska, L. Roszkowski, E.M. Sessolo and A.J. Williams, GUT-inspired SUSY and the muon g −2 anomaly: prospects for LHC 14 TeV, JHEP 06 (2015) 020 [arXiv:1503.08219] [INSPIRE]. [127] Muon g-2 collaboration, Muon (g-2) Technical Design Report, arXiv:1501.06858 [INSPIRE]. [128] L. Darm´e, K. Kowalska, L. Roszkowski and E.M. Sessolo, Flavor anomalies and dark matter in SUSY with an extra U(1), JHEP 10 (2018) 052 [arXiv:1806.06036] [INSPIRE]. [129] Fermi-LAT, DES collaboration, Searching for Dark Matter Annihilation in Recently Discovered Milky Way Satellites with Fermi-LAT, Astrophys. J. 834 (2017) 110 [arXiv:1611.03184] [INSPIRE]. [130] N. Padmanabhan and D.P. Finkbeiner, Detecting dark matter annihilation with CMB polarization: Signatures and experimental prospects, Phys. Rev. D 72 (2005) 023508 [astro-ph/0503486] [INSPIRE]. [131] S. Galli, F. Iocco, G. Bertone and A. Melchiorri, CMB constraints on Dark Matter models with large annihilation cross-section, Phys. Rev. D 80 (2009) 023505 [arXiv:0905.0003] [INSPIRE]. – 31 – [132] T.R. Slatyer, N. Padmanabhan and D.P. Finkbeiner, CMB Constraints on WIMP Annihilation: Energy Absorption During the Recombination Epoch, Phys. Rev. D 80 (2009) 043526 [arXiv:0906.1197] [INSPIRE]. [133] A. Cuoco, J. Heisig, M. Korsmeier and M. Kr¨amer, Constraining heavy dark matter with cosmic-ray antiprotons, JCAP 04 (2018) 004 [arXiv:1711.05274] [INSPIRE]. [134] AMS collaboration, Antiproton Flux, Antiproton-to-Proton Flux Ratio, and Properties of Elementary Particle Fluxes in Primary Cosmic Rays Measured with the Alpha Magnetic Spectrometer on the International Space Station, Phys. Rev. Lett. 117 (2016) 091103. [135] M. Boudaud, M. Cirelli, G. Giesen and P. Salati, A fussy revisitation of antiprotons as a tool for Dark Matter searches, JCAP 05 (2015) 013 [arXiv:1412.5696] [INSPIRE]. JHEP10(2019)043 JHEP10(2019)043 [136] C. Evoli, I. Cholis, D. Grasso, L. Maccione and P. Ullio, Antiprotons from dark matter annihilation in the Galaxy: astrophysical uncertainties, Phys. Rev. D 85 (2012) 123511 [arXiv:1108.0664] [INSPIRE]. [137] M. Cirelli and G. Giesen, Antiprotons from Dark Matter: Current constraints and future sensitivities, JCAP 04 (2013) 015 [arXiv:1301.7079] [INSPIRE]. [138] A. Djouadi, J. Kalinowski and M. Spira, HDECAY: A Program for Higgs boson decays in the standard model and its supersymmetric extension, Comput. Phys. Commun. 108 (1998) 56 [hep-ph/9704448] [INSPIRE]. [139] A. References Djouadi, J. Kalinowski, M. Muehlleitner and M. Spira, HDECAY: Twenty++ years after, Comput. Phys. Commun. 238 (2019) 214 [arXiv:1801.09506] [INSPIRE]. [140] T. Cohen, M. Lisanti, A. Pierce and T.R. Slatyer, Wino Dark Matter Under Siege, JCAP 10 (2013) 061 [arXiv:1307.4082] [INSPIRE]. [141] A. Hryczuk, I. Cholis, R. Iengo, M. Tavakoli and P. Ullio, Indirect Detection Analysis: Wino Dark Matter Case Study, JCAP 07 (2014) 031 [arXiv:1401.6212] [INSPIRE]. [142] V. Lefranc, E. Moulin, P. Panci and J. Silk, Prospects for Annihilating Dark Matter in the inner Galactic halo by the Cherenkov Telescope Array, Phys. Rev. D 91 (2015) 122003 [arXiv:1502.05064] [INSPIRE]. – 32 –
https://openalex.org/W2930777234	http://jrpb.unram.ac.id/index.php/jrpb/article/download/107/79	Indonesian	null	MODIFIKASI ASPEK RASIO SUDU KINCIR ANGIN DAN GEAR RASIO PADA AERATOR BEBASIS TEKNOLOGI POMPA SPIRAL BERTENAGA AIR	Jurnal Ilmiah Rekayasa Pertanian dan Biosistem	2,019	cc-by-sa	4,286	Diterima: Januari 2019 Disetujui: Maret 2019 Diterima: Januari 2019 Disetujui: Maret 2019 ABSTRACT Productivity of shrimp farm in Indonesia can be enhanced by maintaining quality of water using aerator. This research aims to increase rotation velocity through aerator modification utilizing spiral pump that moved by a wind turbine. The research method was conducted by several steps, i.e. (1) construction of modification design of aspect ratio and gear ratio, (2) preparation of wind turbine material, (3) assembly of tool, and (4) test performance. Modification was applied in the velocity requirement of aerator with aspect of wind turbine ratio and gear ratio. The aspect ratio used on the wind turbine propeller were 0.833 and 2.167, whereas the gear ratio used were 1:10 and 1:5. Measured parameter were wind velocity, aerator performance (wind turbine velocity, spiral pump velocity, power, discharge, and head), and water quality (dissolved oxygen). The result of the performance test showed the wind velocity ranged from 0.62 – 4.65 m/s. The measured theoretical rotational velocity was 45.4632. While the value of rotational reduction was approximately 0.41. The lowest power was obtained by the lowest wind velocity of 0.62 m/s. Generated power was 0.1967 Watt. Furthermore, the value of power, water discharge, and head were approximately 24.02834 watt, 15.52008 cm3/s, 30.16 cm respectively. While the dissolved oxygen was 4.4 – 11.9 mg/L. Aerator using 0.833 aspect ratio and 1:10 gear ratio similar with aerator using 2.167 aspect ratio and 1:5 gear ratio can produce better rotational velocity, power, and water discharge. However, the unmodified aerator head produced higher value. Whereas, both aerator can produce dissolved oxygen of 4 – 11 mg/L. Keywords: aerator, aspect ratio, gear ratio, spiral pump, savonius 1Jurusan Keteknikan Pertanian, Fakultas Teknologi Pertanian, Universitas Brawijaya, Jl. Veteran, Malang 65145 1Jurusan Keteknikan Pertanian, Fakultas Teknologi Pertanian, Universitas Brawijaya, Jl. Veteran, Malang 65145 Gunomo Djoyowasito1,), Rifaldi Aprilianto1, Ary Mustofa Ahmad1, Musthofa Lutfi1 1Jurusan Keteknikan Pertanian, Fakultas Teknologi Pertanian, Universitas Brawijaya, Jl. Veteran, Malang 65145 Jurnal Ilmiah Rekayasa Pertanian dan Biosistem, Vol. 7, No. 1, Maret 2019 Jurnal Ilmiah Rekayasa Pertanian dan Biosistem, Vol. 7, No. 1, Maret 2019 JRPB, Vol. 7, No. 1, Maret 2019, Hal. 75-84 DOI: 10.29303/jrpb.v7i1.107 ISSN 2301-8119, e-ISSN 2443-1354 Tersedia online di http://jrpb.unram.ac.id/ MODIFIKASI ASPEK RASIO SUDU KINCIR ANGIN DAN GEAR RASIO PADA AERATOR BEBASIS TEKNOLOGI POMPA SPIRAL BERTENAGA AIR Gunomo Djoyowasito1,), Rifaldi Aprilianto1, Ary Mustofa Ahmad1, Musthofa Lutfi1 ABSTRAK Peningkatan produktivitas tambak udang di Indonesia dilakukan dengan tetap menjaga kualitas air tambak menggunakan aerator. Penelitian ini bertujuan untuk meningkatkan kecepatan putar dengan modifikasi aerator yang memanfaatkan teknologi pompa spiral yang kemudian digerakkan oleh kincir angin. Metode penelitian dilakukan dengan beberapa tahapan, yaitu: (1) 75 Jurnal Ilmiah Rekayasa Pertanian dan Biosistem, Vol. 7, No. 1, Maret 2019 Pembuatan rancangan modifikasi aspek rasio dan gear rasio (2) Persiapan Komponen kincir angin dan gear rasio, (3) Perakitan alat, (4) Uji kinerja alat. Modifikasi dilakukan pada kebutuhan kecepatan putar aerator dengan aspek rasio kincir angin dan gear rasio. Aspek rasio yang digunakan pada sudu kincir angin sebesar 0,833 dan 2,167, sementara gear rasio yang digunakan adalah 1:10 dan 1:5. Parameter yang diukur adalah kecepatan angin, kinerja aerator (kecepatan putar kincir angin, kecepatan putar pompa spiral, daya, debit, tinggi tekan), dan kualitas air (oksigen terlarut). Hasil uji kinerja alat dengan kecepatan angin berkisar antara 0,62 - 4,65 m/s. Kecepatan putar teoritis yang terukur adalah sebesar 45,4632. Sementara untuk reduksi putar memiliki nilai rata-rata sebesar 0,41. Daya teoritis terkecil yang dihasilkan oleh kecepatan angin terkecil, yaitu sebesar 0,62 m/s. Daya yang dihasilkan sebesar 0,1967 Watt. Kemudian daya, debit air, dan tinggi tekan rata-rata 24,02834 watt, 15.52008 cm3/s, 30,16 cm. Sementara oksigen terlarut yang dihasilkan berkisar antara 4,4 – 11,9 mg/L. Aerator dengan aspek rasio 0,833 dan gear rasio 1:10 serta aerator dengan aspek rasio 2,167 dan gear rasio 1:5 mampu menghasilkan kecepatan putar, daya, dan debit air yang lebih baik. Namun, untuk tinggi tekan aerator yang belum dimodifikasi menghasilkan nilai yang lebih tinggi. Sementara kedua aerator mampu menghasilkan oksigen terlarut berkisar 4 – 11 mg/L. Kata kunci: aerator, aspek rasio, gear rasio, pompa spiral, savonius 3. Metode Beberapa alat yang digunakan dalam perancangan, yaitu las listrik dan karbit, roller, cutting, gerinda, kunci L dan kunci pas. Sementara alat yang digunakan untuk pengujian, yaitu windmeter digital, tachometer, DO meter, rollmeter, gelas ukur, wadah penampung. Penelitian ini dilakukan pada bulan Juni - Juli 2018, dengan beberapa tahapan yaitu: (1) Pembuatan rancangan modifikasi aspek rasio, dan gear rasio (2) Persiapan Komponen kincir angin dan gear rasio, (3) Perakitan alat, (5) Uji kinerja alat. Desain dari keseluruhan aerator pompa spiral dapat dilihat pada gambar 1. METODE PENELITIAN penelitian air tambak, Aquades sebagai bahan untuk kalibrasi DO meter. PENDAHULUAN aerator sebagai sistem aerasi. Dimana pada umumnya masyarakat menggunakan tenaga diesel atau tenaga listrik untuk menggerakkan aerator seperti pada aerator tipe kincir air (Utama, dkk., 2013). Tambak merupakan suatu wadah yang dapat digunakan untuk budidaya ikan, rumput laut, rajungan, kepiting, udang, dan hewan lainnya yang dapat hidup di air payau (Ismayani, 2017). Salah satu tambak yang memiliki potensi hasil yang baik adalah tambak udang, dimana berdasarkan data FAO (2015) dalam Hakim (2017) di Indonesia sendiri telah banyak melakukan ekspor komoditi udang yang membuat Indonesia berada posisi kedua penghasil udang terbesar selama kurun waktu 2004 – 2013 dengan jumlah 13.147.297 ton. Permasalahan utama yang sering ditemukan dalam kegagalan produksi tambak udang adalah buruknya kualitas air selama masa pemeliharaan (Arsad, dkk., 2017). Kualitas air ini sangat dipengaruhi oleh oksigen terlarut yang merupakan kebutuhan dasar kehidupan organisme didalam air (Salmin, 2005). Kekurangan oksigen dapat menyebakan organisme mudah tertular penyakit, stress, menghambat pertumbuhan, dan dapat menyebabkan kematian sehingga menurunkan produktivitasnya (Kordi dan Tacung, 2007). Upaya yang dilakukan untuk mengatasi permasalahan ini para petani tambak udang menggunakan p Berdasarkan permasalahan tersebut maka dibuat aerator yang memanfaatkan teknologi pompa spiral yang merupakan pompa yang sederhana yang dapat digerakkan oleh arus air, sungai, dan angin (Kassab et al., 2005). Untuk menggerakkan aerator ini dimanfaatkan energi angin yang melimpah dengan memanfaatkan teknologi kincir angin Savonius Tipe-U yang mampu berputar pada kecepatan angin rendah (Latif, 2013). Dimana untuk menghubungkan putaran kincir angin dan putaran poma spiral maka digunakan differential gear dengan rasio gear tertentu. Namun penggunaan pompa spiral membutuhkan kecepatan putar yang tinggi (Marwanto dan Asral, 2017). Sehingga dilakukan modifikasi aspek rasio sudu kincir angin dan gear rasio untuk meningkatkan kecepatan putar yang dihasilkan kincir angin sebagai penggerak pompa spiral. Dimana dengan kecepatan putar pompa spiral yang meningkat diharapkan pompa spiral mampu melakukan proses aerasi yang lebih maksimal. 76 Jurnal Ilmiah Rekayasa Pertanian dan Biosistem, Vol. 7, No. 1, Maret 2019 2. Bahan Bahan yang digunakan dalam penelitian ini antara lain plat galvalum sebagai bahan dalam pembuatan kincir angin, plat strip sebagai bahan kerangka kincir angin, besi as 2,54 cm sebagai poros kincir angin, mur, baut dan rivet sebagai pengunci alat, air tambak sebagai media perlakuan, kayu, terpal, untuk membuat sekatan pada tambak sebagai kolam Sementara desain modifikasi aspek rasio sudu kincir angin dan gear rasio dapat dilihat pada gambar 2. Dimana modifikasi dilakukan dengan menambah tinggi sudu kincir angin sebesar 75 cm dan mengganti gear driver dengan roda gigi bediameter 7,5 cm. Gambar 1. Rancangan Aerator Pompa Spiral Gambar 1. Rancangan Aerator Pompa Spiral Gambar 2. Desain Rancangan Modifikasi Aspek rasio sudu kincir angin dan roda gigi Gambar 1. Rancangan Aerator Pompa Spiral Gambar 1. Rancangan Aerator Pompa Spiral Gambar 2. Desain Rancangan Modifikasi Aspek rasio sudu kincir angin dan roda gigi Gambar 2. Desain Rancangan Modifikasi Aspek rasio sudu kincir angin dan roda gig 77 77 Jurnal Ilmiah Rekayasa Pertanian dan Biosistem, Vol. 7, No. 1, Maret 2019 Rancang Bangun Alat Rancang Bangun Alat g g Proses pembuatan alat dilakukan perencanaan desain alat yang menggunakan software Autocad 2010. Perancangan model aerator tambak pompa spiral bertenaga air ini dilakukan dengan pendekatan struktural dan fungsional. Rancangan struktural digunakan untuk mengetahui komponen dan dimensi secara detail dari setiap komponen tersebut pada model yang dibuat. Rancangan fungsional digunakan untuk mengetahui fungsi dari setiap komponen model alat. Hasil penerapan dari rancangan model yang dibuat dapat dilihat pada gambar 3 di bawah ini. p y p – 20.00 dengan selang waktu pengukuran yaitu 3 jam. 1. Kecepatan angin 1. Kecepatan angin Kecepatan angin diukur dengan windmeter. 2. Kecepatan putar teoritis (rpm) 2. Kecepatan putar teoritis (rpm) 2. Kecepatan putar teoritis (rpm) Kecepatan putar teoritis dihitung dengan rumus p p ( p ) Kecepatan putar teoritis dihitung dengan rumus NT =(( 𝑣 𝐷 𝑥 𝜋 ) 𝑥 60)…….………..…(1) NT =(( 𝑣 𝐷 𝑥 𝜋 ) 𝑥 60)…….………..…(1) 3. Reduksi putar turbin 3. Reduksi putar turbin Reduksi putar turbin dihitung dengan rumus NA = NT x Red atau Red = NA / NT…(2) 3. Reduksi putar turbin Reduksi putar turbin dihitung dengan rumus NA = NT x Red atau Red = NA / NT…(2) Gambar 3. Hasil Rancangan Aerator Pompa Spiral Reduksi putar turbin dihitung dengan rumus NA = NT x Red atau Red = NA / NT…(2) Dimana NA adalah kecepatan putar aktual yang diukur dengan tachometer. 4. Daya angin teoritis (watt) Daya angin teoritis dihitung dengan rumus Daya angin teoritis dihitung dengan rumus PT = 0.5 x ρ x (D x t1) x v3…......…(3) 5. Daya angin aktual (watt) Daya angin aktual dihitung dengan rumus PA = PT x Eff atau PA = 2 𝜋 𝑛 …...(4) Asumsi: Menurut Park (1981), efisiensi maksimal yang dapat dihasilkan oleh kincir angin savonius adalah sekitar 59,3 %. 6. Debit air Debit air dapat dihitung dengan rumus Q = 𝑉 𝑡2 …………………………...…(5) 7. Oksigen terlarut (DO) Kandungan oksigen terlarut diukur dengan menggunakan DO meter. Gambar 3. Hasil Rancangan Aerator Pompa Spiral Bagian dan fungsi alat sebagai berikut: (A) Kincir angin sebagai penggerak, (B) Kerangka penyangga sebagai penyangga alat, (C) Differential gear sebagai transmisi, (D) Pompa spiral untuk pengaerasi air tambak, (E) Pelampung untuk mengapungkan alat. PA = PT x Eff atau PA = 2 𝜋 𝑛 …...(4) HASIL DAN PEMBAHASAN Pengujian pada penelitian ini dilakukan untuk menguji kelayakan alat hasil modifikasi sehingga bisa berfungsi dengan baik. Pengujian ini dilakukan selama 5 hari berturut-turut. Dimana setiap harinya dilakukan dari pukul 08.00 Hasil Modifikasi Alat Modifikasi dilakukan pada bagian kincir angin yang ditambahkan aspek rasio sudunya yaitu perbandingan tinggi dan diameter sudu dan differential gear dimana dilakukan penggantian rasio gearnya. Hasil rancangan dapat dilihat pada gambar 4 dan gambar 5. 7. Oksigen terlarut (DO) Kandungan oksigen terlarut diukur dengan menggunakan DO meter. 78 Jurnal Ilmiah Rekayasa Pertanian dan Biosistem, Vol. 7, No. 1, Maret 2019 Gambar 4. Hasil Modifikasi Aspek Rasio Sudu Kincir Angin Gambar 5. Hasil Modifikasi Gear Rasio h/d = 2,167 h/d = 0,833 Gear rasio 1: 10 Gear rasio 1: 5 Gambar 4. Hasil Modifikasi Aspek Rasio Sudu Kincir Angin h/d = 2,167 h/d = 0,833 pengukuran yang dilakukan setiap 3 jam dari pukul 08.00 – 20.00 WIB dapat dilihat bahwa nilai maksimum rata – rata kecepatan angin selalu terjadi pada pukul 14.00 yaitu sebesar 4.266 m/s dan nilai minimum rata – rata kecepatan angin selalu terjadi pada pikul 08.00 WIB yaitu sebesar 1,588 m/s. Kondisi ini dapat dilihat pada gambar 7. Gambar 4. Hasil Modifikasi Aspek Rasio Sudu Kincir Angin Gambar 4. Hasil Modifikasi Aspek Rasio Sudu Kincir Angin Gear rasio 1: 5 Gear rasio 1: 10 Gambar 5. Hasil Modifikasi Gear Rasio Gear rasio 1: 10 Gear rasio 1: 5 Gambar 6. Grafik rata – rata kecepatan angin dan hari pengamatan Gambar 7. Grafik rata – rata kecepatan angin dan waktu pengamatan 2.39 2.822 2.888 2.598 2.388 0 1 2 3 4 1 2 3 4 5 Kecepatan Angin (m/det) Hari Ke- 1.588 2.068 4.266 3.468 1.696 0 1 2 3 4 5 8 11 14 17 20 Kecepatan Angi (m/det) Pengamatan Pada Jam Gambar 6. Grafik rata – rata kecepatan angin dan hari pengamatan 2.39 2.822 2.888 2.598 2.388 0 1 2 3 4 1 2 3 4 5 Kecepatan Angin (m/det) Hari Ke- Gambar 6. Grafik rata – rata kecepatan angin dan hari pengamatan 2.39 2.822 2.888 2.598 2.388 0 1 2 3 4 1 2 3 4 5 Kecepatan Angin (m/det) Hari Ke- Gear rasio 1: 5 Gear rasio 1: 10 Gear rasio 1: 10 Gambar 5. Hasil Modifikasi Gear Rasio Hasil modifikasi aspek rasio sudu yang dilakukan berfungsi untuk meningkatkan kecepatan putar kincir angin, hal ini dikarenakan pada kincir angin savonius semakin besar perbedaan aspek rasio sudu (h/d) maka semakin besar kecepatan putar yang dihasilkan (Brusca et. al., 2014). Hasil Pengukuran Potensi Angin Hasil pengukuran potensi angin dilakukan selama 5 hari berturut-turut. Dari hasil pengukuran tersebut, dapat diketahui bahwa besarnya potensi angin yang tersedia di wilayah tambak Desa Wedani - Cerme berkisar antara 0,62 – 4,65 m/s dengan kecepatan angin rata-rata yang diperoleh sebesar 2,6172 m/s. Dari hasil pengukuran kecepatan angin setiap harinya, dapat diketahui bahwa kondisi angin di wilayah tambak tersebut tidak selalu sama kondisi tersebut dapat dilihat pada gambar 6. Sementara dari Perbedaan kecepatan angin ini terjadi disebabkan oleh adanya perbedaan tekanan. Menurut Kartasapoetra (2004) di dalam Pramono, dkk., (2016) daerah yang memiliki tekanan rendah memiliki suhu yang lebih tinggi dibandingkan daerah yang memiliki tekanan tinggi Hasil Modifikasi Alat Hasil Modifikasi pada differential gear dengan gear rasio 1:5 ini berfungsi untuk meningkatkan kecepatan putar pompa spiral, Hal ini dikarenakan semakin kecil rasio gear yang digunakan akan semakin besar torsi yang dibutuhkan dan semakin bertambah kecepatan yang dihasilkan (Yunianto, 2016). Gambar 6. Grafik rata – rata kecepatan angin dan hari pengamatan Gambar 7. Grafik rata – rata kecepatan angin dan waktu pengamatan 1.588 2.068 4.266 3.468 1.696 0 1 2 3 4 5 8 11 14 17 20 Kecepatan Angi (m/det) Pengamatan Pada Jam Gambar 7. Grafik rata – rata kecepatan angin dan waktu pengamatan 1.588 2.068 4.266 3.468 1.696 0 1 2 3 4 5 8 11 14 17 20 Kecepatan Angi (m/det) Pengamatan Pada Jam Gambar 7. Grafik rata – rata kecepatan angin dan waktu pengamatan Jumlah Putaran (rpm) Berdasarkan pengukuran kecepatan putar kincir angin dan pompa spiral dapat diketahui bahwa kincir angin pada aerator 79 Jurnal Ilmiah Rekayasa Pertanian dan Biosistem, Vol. 7, No. 1, Maret 2019 pompa spiral yang telah dimodifikasi mampu berputar pada kecepatan angin minimal 1,51 m/s. Sementara kecepatan putar kincir angin yang dihasilkan berkisar antara 2 – 62 rpm, sementara kecepatan putar pompa spiral berkisar antara 0,4 – 12,4 rpm. Dimana dari data yang diperoleh dapat dibuat grafik seperti pada gambar 8. teoritis terjadi pada hari pertama pukul 08.00 yaitu sebesar 10,77. Sementara untuk nilai reduksi putaran kincir angin terbesar terjadi pada hari ke 4 pukul 14.00 yaitu sebesar 0,77. Gambar 9. Grafik Hubungan Kecepatan Angin Dengan Kecepatan Putar Teoritis dan Aktual Gambar 8. Grafik Hubungan Kecepatan Angin dengan Kecepatan Putar Turbin Masing – Masing Aerator Gambar 9. Grafik Hubungan Kecepatan Angin Dengan Kecepatan Putar Teoritis dan Aktual Berdasarkan gambar 9 dapat dilihat bahwa nilai kecepatan putar teoritis terus bertambah seiring dengan bertambahnya nilai kecepatan angin. Sementara nilai kecepatan putar aktual juga memiliki nilai yang bertambah seiring bertambahnya nilai kecepatan angin, namun untuk beberapa data dapat dilihat bahwa nilai kecepatan putar aktual kincir angin ada yang mengalami penurunan dengan bertambahnya nilai kecepatan angin. Kecepatan putar aktual dipengaruhi oleh kondisi pompa spiral yang berisi air yang membuat berat putaran kincir angin. Beban differential gear dan arah angin yang tidak searah dengan putaran kincir angin. Gambar 8. Grafik Hubungan Kecepatan Angin dengan Kecepatan Putar Turbin Masing – Masing Aerator Berdasarkan gambar 8 dapat dilihat bahwa semakin besar nilai kecepatan angina, semakin bertambah juga nilai kecepatan putar kincir angin dan pompa spiral. Sehingga dapat diketahui bahwa kecepatan angin dan besar rotasi dihasilkan memiliki hubungan yaitu berbanding lurus. Berdasarkan hal tersebut, hasil ini telah sesuai dengan penelitian yang dilakukan oleh Aryanto, dkk., (2013), yaitu kecepatan angin berbanding lurus dengan putaran yang dihasilkan. Debit Air dan Tinggi Tekan g p Putar Pompa Spiral Dan Debit Air g p Putar Pompa Spiral Dan Debit Air Berdasarkan data yang didapatkan dapat dilihat bahwa nilai rata – rata kecepatan angin 2,6172 m/s. dari kecepatan angin tersebut dapat menghasilkan rata – rata debit aliran sebesar 15,52008 cm3/s dan tinggi tekan rata - rata sebesar 30,16 cm. Kemudian pada kecepatan angin minimal yang mampu memutar kincir angin dan pompa spiral sebesar 1,51 m/s, dimana dengan kecepatan angin tersebut dapat dihasilkan kecepatan putar turbin dan pompa spiral sebesar 11 dan 2,2 rpm yang kemudian menghasilkan debit air sebesar 2,667 cm3/s, dan tinggi tekan sebesar 9 cm. Sedangkan untuk kecepatan angin maksimal yakni sebesar 4,65 m/s mampu Gambar 12 Grafik Hubungan Kecepatan Putar Pompa Spiral Dan Tinggi Tekan Gambar 12 Grafik Hubungan Kecepatan Putar Pompa Spiral Dan Tinggi Tekan Kecepatan Putar Teoritis dan Reduksi Putar Turbin Berdasarkan perhitungan yang dilakukan dengan persamaan 4 dan 5 didapatkan nilai daya teoritis terkecil yang dihasilkan oleh kecepatan angin terkecil, yaitu sebesar 0,62 m/s daya yang dihasilkan sebesar 0,1967 Watt. Sementara untuk nilai kecepatan angin terbesar, yaitu sebesar 4,65 m/s, daya yang dihasilkan sebesar 82,949 Watt. Berdasarkan penelitian yang telah dilakukan menunjukkan bahwa rata – rata kecepatan putar teoritis yang terukur adalah sebesar 45,4632. Sementara untuk reduksi putar memiliki nilai rata – rata sebesar 0,41. Kemudian nilai kecepatan putar teoritis terbesar terjadi pada hari ke 4 pukul 14.00 yaitu sebesar 80,776. Sementara nilai terkecil kecepatan putar 80 Jurnal Ilmiah Rekayasa Pertanian dan Biosistem, Vol. 7, No. 1, Maret 2019 Sementara daya aktual terkecil adalah 0,117 watt, dan nilai daya aktual terbesar adalah 49,189 Watt. untuk menghasilkan kecepatan putar turbin dan pompa spiral sebesar 62 dan 12,4 rpm yang kemudian menghasilkan debit aliran sebesar 38 cm3/s dengan tinggi tekan sebesar 83 cm. Gambar 10. Grafik Hubungan Kecepatan Angin Dan Daya Berdasarkan gambar 11 dan gambar 12 dapat dilihat bahwa seiring bertambahnya nilai kecepatan putar pompa spiral nilai debit air dan tinggi tekan juga mengalami kenaikan. Sehingga dapat disimpulkan bahwa hubungan kecepatan putar pompa spiral dengan debit air dan tinggi tekan adalah berbanding lurus. Kondisi ini sesuai dengan penelitian yang dilakukan oleh Marwanto dan Asral (2017) dimana semakin cepat pompa spiral berputar maka akan semakin besar debit air yang dihasilkan. Gambar 10. Grafik Hubungan Kecepatan Angin Dan Daya Berdasarkan gambar 10 dapat dilihat bahwa nilai daya yang dihasilkan oleh kincir angin semakin bertambah seiring bertambahnya nilai kecepatan angin. Sehingga dapat disimpulkan bahwa nilai daya dan kecepatan angin adalah berbanding lurus. Kesimpulan tersebut telah sesuai dengan penelitian yang telah dilakukan oleh Nanang, dkk., (2016) bahwa semakin besar kecepatan angin maka akan semakin besar daya yang dihasilkan. Gambar 11. Grafik Hubungan Kecepatan Putar Pompa Spiral Dan Debit Air Gambar 11. Grafik Hubungan Kecepatan Oksigen Terlarut (Disolved Oxygen) Berdasarkan data yang didapatkan nilai oksigen terlarut yang ada di tambak 81 Jurnal Ilmiah Rekayasa Pertanian dan Biosistem, Vol. 7, No. 1, Maret 2019 Desa Wedani berkisar antara 4,4 – 11,9 mg/L dengan rata – rata sebesar 8,016 mg/L. Nilai oksigen terlarut maksimum terjadi pada hari ke 3 pukul 17.00. sementara nilai oksigen terlarut terjadi pada hari pertama pukul 08.00. aerator. Dimana dari pukul 08.00 – 17.00 oksigen mengalami peningkatan sementara pada pukul 17.00 – 20.00 oksigen terlarut mengalami penurunan. Penurunan yang terjadi pada malam hari disebabkan oleh aerator yang sudah tidak dapat bekerja optimal karena menurunnya kecepatan angin, kemudian karena sudah tidak adanya fotosintesis, dan oksigen yang yang kemudian digunakan oleh semua organisme yang hidup didalam tambak maupun disekitar tambak sepanjang malam. hal ini sesuai dengan penelitian yang dilakukan oleh (Sinaga, dkk., 2016). Tabel 1. Nilai rata – rata oksigen terlarut pada waktu pengamatan yang berbeda pada aerator yang telah dimodifikasi Jam ke Kec. Angin (m/det) Oksigen Terlarut (mg/L) Sebelu m Sesuda h 8.00 1.588 5.44 5.44 11.00 2.068 5.44 7.1 14.00 4.266 7.1 9.8 17.00 3.468 9.8 10.82 20.00 1.696 10.82 6.92 Rata - Rata 2.6172 7.72 8.016 KESIMPULAN Brusca, S., Lanzafame, R., Messina, M., (2014). Design of A Vertical-Axis Wind Turbine: How The Aspect Ratio Affects The Turbine’s Performance. International Journal of Energy and Environmental Engineering. Vol. 5. Pp. 333 - 340 Berdasarkan hasil pengujian dan penelitian aerator pompa spiral bertenaga bayu yang telah dimodifikasi dapat di simpulkan bahwa: 1. Kecepatan angin yang terukur adalah antara 0,62 – 4,65 m/s. Ferreira, N.C, C. Bonnetti, and W.Q. Seiffert. (2011). Hydrological and Water Quality Indices as Management Tools in Marine Shrimp Culture. Aquaculture 318: 425–433 2. Kecepatan angin minimum yang mampu menggerakkan kincir angin adalah 1,51 m/s. 3. Kecepatan putar teoritis yang terukur adalah sebesar 45,4632. Sementara untuk reduksi putar memiliki nilai rata – rata sebesar 0,41. Daya teoritis terkecil yang dihasilkan oleh kecepatan angin terkecil yaitu sebesar 0,62 m/s daya yang dihasilkan sebesar 0,1967 Watt. Hakim, Luqman. 2017. Performa Budidaya udang vename (Litopenaeus Vannamei) Semi Intensif di Desa Purworejo Kecamatan Pasir Sakti Kabupaten Lampung Timur. Skripsi. Lampung: Universitas Lampung 4. Pada kecepatan angin yang sama nilai kecepatan putar kincir angin, pompa spiral, daya dan debit air memiliki nilai yang lebih tinggi. Namun untuk tinggi tekan yang dihasilkan aerator yang telah dimodifikasi memiliki nilai yang lebih rendah. Oksigen terlarut yang dihasilkan berkisar antara 4,4 – 11,9 mg/L 4. Pada kecepatan angin yang sama nilai kecepatan putar kincir angin, pompa spiral, daya dan debit air memiliki nilai yang lebih tinggi. Namun untuk tinggi tekan yang dihasilkan aerator yang telah dimodifikasi memiliki nilai yang lebih rendah. Oksigen terlarut yang dihasilkan berkisar antara 4,4 – 11,9 mg/L Ismayani, Erika. 2017. Manajemen Usaha Budidaya Udang Vanname (Litopeneaus Vannamei) (Studi Kasus Tambak Pt. Beroro Jaya Vanname Di Kabupaten Konawe Selatan). Tesis. Kendari: Universitas Halu Oleo 2017 Kassab, Sadek Z, A. A.A Naby, and El Sayed I.A.B. (2005). Coil Pump Performance Under Variable Operating Conditions. Ninth International Water Technology Conference, IWTC9. Perbandingan Alat yang Telah Dimodifikasi dan Belum Dimodifikasi Berdasarkan tabel 2 dapat dilihat bahwa pada aerator pompa spiral yang telah dimodifikasi mengalami peningkatan kinerja. Dimana pada kecepatan angin yang sama nilai kecepatan putar kincir angin, pompa spiral, daya dan debit air memiliki nilai yang lebih tinggi. Namun untuk tinggi tekan yang dihasilkan aerator yang telah dimodifikasi memiliki nilai yang lebih rendah yang disebabkan adanya penyumbatan pada nozel oleh lumpur. Berdasarkan tabel 1, nilai oksigen terlarut terbesar rata – rata terjadi pada pukul 17.00 dan oksigen terlarut terjadi pada pukul 08.00 pagi. Selain itu dapat dilihat pula bahwa rata – rata oksigen terlarut mengalami peningkatan seiring waktu pengukuran atau pengoperasian Tabel 2 Tabel perbandingan aerator belum modifikasi dan telah di modifikasi Tabel 2 Tabel perbandingan aerator belum modifikasi dan telah di modifikasi Tabel 2 Tabel perbandingan aerator belum modifikasi dan telah di modifikasi No. Pembanding Belum Modifikasi Telah Dimodifikasi 1 Kecapatan angin 0,45 – 6,71 m/s 0,62 – 4,65 m/s 2 Kecepatan angin terendah untuk memutar kincir angin 2,5 m/s 1,51 m/s 3 Kecepatan putar kincir angin 22 – 64 rpm 2 – 62 rpm 4 Kec. Putar pompa spiral 2,2 – 6,4 rpm 0,4 – 12,4 rpm 5 Daya 0,03 – 81,73 watt 0,197 – 82,94 watt 6 Debit air 7 - 18,33 cm3/s 0,667 – 38 cm3/s 7 Tinggi tekan 42 – 118 cm 3 – 83 cm 8 Oksigen terlarut 4,5 – 11,1 mg/L 4,4 – 11,9 mg/L Pembanding Belum Modifikasi Telah Dimodifikasi t i 0 45 6 71 / Sementara oksigen terlarut yang dihasilkan memiliki range yang hampir sama dan merupakan kondisi oksigen terlarut yang sesuai dengan yang dibutuhkan spesies udang yaitu 4 – 11 Kemudian kecepatan angin minimum yang mampu menggerakkan kincir angin pada aerator yang dimodifikasi berada pada kecepatan angin yang lebih rendah yaitu 1,51. m/s 82 Jurnal Ilmiah Rekayasa Pertanian dan Biosistem, Vol. 7, No. 1, Maret 2019 Aryanto, Firman, Mara, I Made, dan Nuarsa, Made. (2013). Pengaruh Kecepatan Angin Dan Variasi Jumlah Sudu Terhadap Unjuk Kerja Turbin Angin Poros Horizontal. Jurnal Dinamika Teknik Mesin, 3(1). ISSN: 2088-088X. mg/L. Menurut Ferreira, et al. (2011) konsentrasi minimal DO (oksigen terlarut) yang dibutuhkan spesies udang agar dapat bertahan hidup adalah sebesar 4–6 mg/L, sedangkan kandungan DO yang optimal bagi udang berkisar antara 7-11 mg/L. mg/L. Menurut Ferreira, et al. Perbandingan Alat yang Telah Dimodifikasi dan Belum Dimodifikasi (2011) konsentrasi minimal DO (oksigen terlarut) yang dibutuhkan spesies udang agar dapat bertahan hidup adalah sebesar 4–6 mg/L, sedangkan kandungan DO yang optimal bagi udang berkisar antara 7-11 mg/L. DAFTAR PUSTAKA Arsad, Sulastri, Afandy, Ahmad, Atika P. Purwadhi, Maya, Betrina V., Saputra, Dhira K., dan Buwono, Nanik Retno. (2017). Studi Kegiatan Budidaya Pembesaran Udang Vaname (Litopenaeus Vannamei) dengan Penerapan Sistem Pemeliharaan Berbeda. Jurnal Ilmiah Perikanan dan Kelautan, 9(1). ISSN: 2085-5842 Kordi, M.G.H.K., dan Tacung, A.B. (2007). Pengelolaan Kualitas Air Dalam Budidaya Perairan. Jakarta: PT Rineka Cipta. Latif, Melda. (2013). Efisiensi Prototipe Turbin Savonius pada Kecepatan Angin Rendah. Jurnal Rekayasa Elektrika, 10(3). 83 Jurnal Ilmiah Rekayasa Pertanian dan Biosistem, Vol. 7, No. 1, Maret 2019 Suh Oksigen Terlarut dan pH Secara Vertikal Selama 24 Jam di Danau Kelapa Gading Kabupaten Asahan Sumatera Utara. Omni-Akuatika, 12(2), November 2016: 114 -12. Suh Oksigen Terlarut dan pH Secara Vertikal Selama 24 Jam di Danau Kelapa Gading Kabupaten Asahan Sumatera Utara. Omni-Akuatika, 12(2), November 2016: 114 -12. Marwanto, dan Asral. (2017). Pengujian Pompa Spiral Dengan Kincir Air Pada Aliran Irigasi. Jurnal Teknik, 4(2). Nanang, Rahmat, Gunarto, dan Sarwon, Eko. (2016). Study Eksperimental Berbagai Macam Jenis Sudu Turbin Angin Sumbu Horisontal Skala Laboratorium. Respository UM Pontianak. Diakses pada 30 juli 2018. http://repository.unmuhpnk.ac.id/id/ eprint/149. Utama, Aditya Rifa, Mubarrok, Muhammad Fasih, Ardiansyah, Antomi, Hendra, Fathoni, Muhammad Januar, dan Dikairono, Rudy. (2013). Eco Aerator: Inovasi Penyuplai Oksigen Dengan Teknologi Vertical Axis Wind Dan Archimedes' Screw Guna Menurunkan Biaya Operasional Petani Tambak. e-Proceeding PIMNAS. Surabaya: Institut Teknologi Sepuluh Nopember Surabaya. Park, J. 1981. The Wind Power Book. California: Chesire Book Pramono, Yuri, Warsito, dan Syafriadi. (2016). Monitoring Data Kecepatan Dan Arah Angin Secara Real Time Melalui Web. Jurnal Teori Dan Aplikasi Fisika, 4(02). Yunianto, Rizky Firmansyah. (2016). Analisa Pengaruh Variasi Perbandingan Roda Gigi Transmisi Untuk Memaksimalkan Daya Listrik Pada Turbin Angin Savonius Bertingkat. Skripsi. Malang: Universitas Islam Malang. Salmin. (2005). Oksigen Terlarut (DO) dan Kebutuhan Oksigen Biologi (BOD) sebagai Salah Satu Indikator untuk Menentukan Kualitas Perairan. Oseana XXX(3): 21-26. Sinaga, Eva Lia Risky, Muhtadi, Ahmad, dan Bakti, Darma. (2016). Profil 84
https://openalex.org/W2806826615	https://europepmc.org/articles/pmc6028868?pdf=render	English	null	Metachronous Peritoneal Metastases After Adjuvant Chemotherapy are Associated with Poor Outcome After Cytoreduction and HIPEC	Annals of surgical oncology	2,018	cc-by	6,291	Metachronous Peritoneal Metastases After Adjuvant Chemotherapy are Associated with Poor Outcome After Cytoreduction and HIPEC Nina R. Sluiter, MD1, Koen P. Rovers, MD2, Youssra Salhi, BSc1, Stijn L. Vlek, MD1, Veerle M. H. Coupe´, MD, PhD3, Henk M. W. Verheul, MD, PhD4, Geert Kazemier, MD, PhD1, Ignace H. J. T. de Hingh, MD, PhD2, and Jurriaan B. Tuynman, MD, PhD1 1Department of Surgery, VU University Medical Center, Amsterdam, The Netherlands; 2Department of Surgery, Catharina Hospital Eindhoven, Eindhoven, The Netherlands; 3Department of Epidemiology and Biostatistics, VU University Medical Center, Amsterdam, The Netherlands; 4Department of Medical Oncology, VU University Medical Center, Amsterdam, The Netherlands Catherina Hospital Eindhoven containing 345 CRC patients treated with the intent of HIPEC were utilized. Results. High Peritoneal Cancer Index (PCI) scores were associated with worse DFS [hazard ratio (HR) 1.04, 95% conﬁdence interval (CI) 1.00–1.08, p = 0.040] and OS (HR 1.11, 95% CI 1.07–1.15, p \ 0.001) in multivariate anal- ysis. Furthermore, patients with peritoneal recurrence within 1 year following adjuvant chemotherapy had worse DFS (HR 2.13, 95% CI 1.26–3.61, p = 0.005) and OS (HR 2.76, 95% CI 1.45–5.27, p = 0.002) than patients who did not receive adjuvant chemotherapy or patients with peri- toneal recurrence after 1 year. ORIGINAL ARTICLE – COLORECTAL CANCER ORIGINAL ARTICLE – COLORECTAL CANCER The Author(s) 2018 First Received: 27 February 2018; Published Online: 31 May 2018 N. R. Sluiter, MD e-mail: n.sluiter@vumc.nl ABSTRACT Catherina Hospital Eindhoven containing 345 CRC patients treated with the intent of HIPEC were utilized. Introduction. Cytoreduction and hyperthermic intraperi- toneal chemotherapy (HIPEC) improve the survival of colorectal cancer (CRC) patients with peritoneal metas- tases. Patient selection is key since this treatment is associated with high morbidity. Patients with peritoneal recurrence within 1 year after previous adjuvant chemotherapy are thought to beneﬁt less from HIPEC treatment; however, no published data are available to assist in clinical decision making. This study assessed whether peritoneal recurrence within 1 year after adjuvant chemotherapy was associated with survival after HIPEC treatment. p Results. High Peritoneal Cancer Index (PCI) scores were associated with worse DFS [hazard ratio (HR) 1.04, 95% conﬁdence interval (CI) 1.00–1.08, p = 0.040] and OS (HR 1.11, 95% CI 1.07–1.15, p \ 0.001) in multivariate anal- ysis. Furthermore, patients with peritoneal recurrence within 1 year following adjuvant chemotherapy had worse DFS (HR 2.13, 95% CI 1.26–3.61, p = 0.005) and OS (HR 2.76, 95% CI 1.45–5.27, p = 0.002) than patients who did not receive adjuvant chemotherapy or patients with peri- toneal recurrence after 1 year. Conclusion. Peritoneal recurrence within 1 year after previous adjuvant chemotherapy, as well as high PCI scores, are associated with poor survival after cytoreduc- tion and HIPEC. These factors should be considered in order to avoid high-morbidity treatment in patients who might not beneﬁt from such treatment. Methods. Peritoneal recurrence within 1 year after adju- vant chemotherapy, as well as other potentially prognostic clinical and pathological variables, were tested in univari- ate and multivariate analysis for correlation with primary outcomes, i.e. overall survival (OS) and disease-free sur- vival (DFS). Two prospectively collected databases from the VU University Medical Center Amsterdam and Cytoreductive surgery (CRS) combined with hyper- thermic intraperitoneal chemotherapy (HIPEC) is currently the only potentially curative treatment for colorectal cancer (CRC) patients with limited peritoneal metastases (PM).1,2 This approach increases median survival rates from 12 to 16 months after treatment with systemic chemotherapy alone 3–5 to 33–45 months, translating to a 5-year survival rate of 35%.4,6,7 However, the combination of extensive surgery and HIPEC is associated with relatively high morbidity and mortality rates of 16–64% and 1–5%,8–10 respectively. Hence, it is of utmost importance to carefully select patients who will beneﬁt most from this treatment. Electronic supplementary material The online version of this article (https://doi.org/10.1245/s10434-018-6539-x) contains supplementary material, which is available to authorized users. Ann Surg Oncol (2018) 25:2347–2356 https://doi.org/10.1245/s10434-018-6539-x Ann Surg Oncol (2018) 25:2347–2356 https://doi.org/10.1245/s10434-018-6539-x ABSTRACT The Author(s) 2018 First Received: 27 February 2018; Published Online: 31 May 2018 N. R. Sluiter, MD e-mail: n.sluiter@vumc.nl 2348 N. R. Sluiter et al. N. R. Sluiter et al. Patient Selection and Data Collection Patient Selection and Data Collection In recent years, there has been wide interest in the identiﬁcation of prognostic factors in patients with PM, both clinically 11 and biologically.12,13 Nevertheless, the lack of randomized controlled trials and the large hetero- geneity of published studies limit the use of these variables in clinical practice. The current body of prognostic char- acteristics include the Peritoneal Cancer Index (PCI; a score for peritoneal tumor burden) and biological tumor characteristics such as primary tumor differentiation.11 Some have combined variables to predict outcome after CRS and HIPEC using nomograms such as the Peritoneal Surface Disease Severity Score (PSDSS)11 or the Colorectal Peritoneal Metastases Prognostic Surgical Score (COMPASS).14,15 These characteristics rely heavily on intra- or postoperative ﬁndings, whereas selection of HIPEC patients should ideally take place before the start of treatment. Therefore, identiﬁcation and validation of new prognostic variables that can be assessed preoperatively is warranted. The present study was approved by the Medical Ethics Review Committee of the VU University Medical Center (2018.124). Patients with PM of colorectal adenocarcinoma who underwent surgery with the intent of CRS and HIPEC were considered for inclusion in this study, whereas patients with synchronous metastases and low-grade appendiceal mucinous neoplasms (LAMN), as well as patients without histologically proven PM, were excluded. The following clinicopathological and follow-up data were collected from the records of both institutions: age, sex, American Society of Anesthesiologists (ASA) score, information on comorbidity and primary tumor character- istics, prior treatment, and prior surgical scores (PSS). A PSS of 0 was recorded when there was no history of abdominal surgery or only a biopsy; a PSS of 1 was recorded when abdominal surgery was performed in one of the abdominal regions; and a PSS of 2 was recorded for surgery in two to ﬁve regions, and a PSS of 3 for surgery in more than ﬁve regions. In clinical practice, patients developing PM despite treatment with adjuvant chemotherapy after primary tumor surgery seem to beneﬁt less from CRS and HIPEC, espe- cially if PM are diagnosed within 1 year after primary tumor resection, or even during chemotherapeutic treat- ment. ABSTRACT Accordingly, several studies exclude patients with PM development or progression despite systemic chemotherapy from CRS and HIPEC.6,16,17 Moreover, certain guidelines encourage consideration of this factor in the CRS and HIPEC selection process; 18,19 however, no data are currently available to support this decision in clinical practice. This study aimed to assess whether peri- toneal recurrence within 1 year after adjuvant chemotherapy is associated with poor survival in CRC patients after CRS and HIPEC. Intraoperatively, the extent of peritoneal disease was quantiﬁed using the PCI, a numeric score that combines the lesion size (0–3) with the amount of affected abdomino- pelvic regions (to a maximum of 13) to a score from 0 to 39.21,22 After completion of CRS, resection outcome was determined according to the maximal size of residual tumor tissue: an R1 (complete) resection was scored when no macroscopically visible tumor was left behind; an R2a resection was scored when the tumor was smaller than 2.5 mm; and an R2b resection was scored when the residual tumor was larger than 2.5 mm.23 Postoperatively, hospital complications were docu- mented and scored according to the Common Terminology Criteria for Adverse Events (CTCAE) v4.0 grading sys- tem.24 Follow-up data, including recurrences and death, were obtained from both hospitals. Operative Treatment To examine whether patients who developed PM after adjuvant chemotherapy had worse outcomes, patients were divided into four groups based on administration of adju- vant chemotherapy (yes vs. no) and time to diagnosis of PM after primary tumor resection (within 1 year vs. after more than 1 year). These categories will be referred to as follows: (1) PM within 1 year without adjuvant chemotherapy; (2) PM after more than 1 year without adjuvant chemotherapy; (3) PM within 1 year after adju- vant chemotherapy; and (4) PM more than 1 year after adjuvant chemotherapy. A diagnosis of PM was based on regular follow-up after resection of the primary tumor, consisting of carcinoembryonic antigen (CEA) measure- ments, and ultrasound and computed tomography (CT) scans, according to the Dutch guidelines.25 Patients from the VU University Medical Center and the Catharina Hospital Eindhoven, two tertiary referral centers, were included in this study. Both hospitals performed CRS and HIPEC according to the same standardized protocol.7 Cytoreductive surgery consisted of complete debulking, stripping of the affected peritoneum, and removal of the omentum.20 When deemed necessary, multiorgan resec- tions were carried out. Subsequently, if a macroscopic complete resection was achieved, either oxaliplatin (460 mg/m2 body surface) or mitomycin C (35 mg/m2 body surface) was installed in the peritoneal cavity, with a target temperature of 39–41 C for 30 or 90 min, respectively. Clinical Prognosticators for HIPEC Surgery 2349 Statistical Analysis Associations between clinicopathological variables were tested using the Fisher’s exact test or Chi square test for two categorical/dichotomous variables, or the independent t test or one-way analysis of variance (ANOVA) for a continuous, normally distributed variable with a dichoto- mous or categorical variable, respectively. A signiﬁcant difference was assumed for a p value \ 0.05 (two-sided test). If necessary, variables were dichotomized to provide a minimum of ten events per category in the survival analysis. Dichotomization was performed on the basis of mean values for continuous variables. Overall survival (OS) and disease-free survival (DFS) were deﬁned as the time (in months) from the date of CRS and HIPEC to the date of death from any cause or date of recurrence, respectively. Operative Treatment Univariate associations between OS or DFS and clinicopathological variables that could be determined preoperatively were tested using the Kaplan–Meier method TABLE 1 Baseline characteristics of all patients Characteristic n (%) General characteristics All 175 Female sex 93 (53.1) Age, years Mean (SD) 61.7 (10.3) ASA classiﬁcation I–II 150 (85.7) III 25 (14.3) Primary tumor characteristics Location Colon 159 (90.9) Rectum 16 (9.1) Tumor differentiation Good/moderate 127 (84.1) Poor 20 (13.3) Signet cell 4 (2.6) Tumor histology Adenocarcinoma 138 (81.7) Mucinous adenocarcinoma 31 (18.3) T stage T1–3 107 (61.8) T4 66 (38.2) N stage N0 66 (37.9) N1–2 108 (62.1) Distant metastases 13 (7.4) Stage 1–2 62 (35.8) 3–4 111 (64.2) Perioperative treatment Primary tumor: adjuvant chemotherapy 111 (64.2) Primary tumor: adjuvant chemotherapy type Oxaliplatin 2 (1.8) Capecitabine 4 (3.6) CAPOX 71 (64.0) FOLFOX 14 (12.6) 5-FU 1 (0.9) Unknown 19 (17.1) Development of PM B 1, no chemotherapy 30 (17.1) [ 1 year, no chemotherapy 34 (19.4) B 1 year after chemotherapy 36 (20.6) [ 1 year after chemotherapy 75 (42.9) HIPEC: neoadjuvant chemotherapy 21 (12.1) HIPEC: adjuvant chemotherapy 72 (41.4) Prior surgical score 0 13 (7.6) 1 8 (4.7) TABLE 1 continued Characteristic n (%) 2 140 (81.9) 3 10 (5.8) HIPEC characteristics Operative procedure Open CRS and HIPEC 138 (78.9) Laparoscopic CRS and HIPEC 3 (1.7) Open–close 34 (19.4) PCI Mean (SD) 12 (8) HIPEC chemotherapy Mitomycin C 128 (90.8) Oxaliplatin 13 (9.2) Resection score R1 135 (77.1) R2a 6 (3.5) R2b 34 (19.4) SAE Total 94 (53.7) Grade I: mild 12 (6.9) Grade II: moderate 35 (20.0) Grade III: severe 32 (18.3) Grade IV: life-threatening 13 (7.4) Grade V: death 2 (1.1) Reoperation 32 (18.3) ASA American Society of Anesthesiologists, CAPOX capecitabine ? oxaliplatin, CRS cytoreductive surgery, FOLFOX leucovorin ? 5-FU ? oxaliplatin, HIPEC hyperthermic intraperi- toneal therapy, PCI Peritoneal Cancer Index, PM peritoneal metastases, SAE serious adverse event, SD standard deviation, 5-FU 5-ﬂuorouracil Associations between clinicopathological variables were tested using the Fisher’s exact test or Chi square test for two categorical/dichotomous variables, or the independent t test or one-way analysis of variance (ANOVA) for a continuous, normally distributed variable with a dichoto- mous or categorical variable, respectively. A signiﬁcant difference was assumed for a p value \ 0.05 (two-sided test). If necessary, variables were dichotomized to provide a minimum of ten events per category in the survival analysis. Operative Treatment Dichotomization was performed on the basis of mean values for continuous variables. Overall survival (OS) and disease-free survival (DFS) were deﬁned as the time (in months) from the date of CRS and HIPEC to the date of death from any cause or date of recurrence, respectively. Univariate associations between OS or DFS and clinicopathological variables that could be determined preoperatively were tested using the Kaplan–Meier method 2350 N. R. Sluiter et al. Oncologic Outcomes (log-rank test). Variables with a p value B 0.1 in univariate survival analysis were included in a multivariate Cox regression analysis. Variable selection in the Cox model was performed using backward selection, with a threshold p value of 0.1 for exclusion from the model. Statistical analyses were performed using the Statistical Package for Social Sciences (SPSS) version 23 for Windows (IBM Corporation, Armonk, NY, USA). (log-rank test). Variables with a p value B 0.1 in univariate survival analysis were included in a multivariate Cox regression analysis. Variable selection in the Cox model was performed using backward selection, with a threshold p value of 0.1 for exclusion from the model. Statistical analyses were performed using the Statistical Package for Social Sciences (SPSS) version 23 for Windows (IBM Corporation, Armonk, NY, USA). A complete overview of univariate survival analysis is presented in Table 2. Figure 1 depicts the survival curves of the statistically signiﬁcant variables in univariate analysis. RESULTS The median DFS of the whole cohort was 12.0 months (95% CI 8.8–13.2). Patients with a diagnosis of PM within 1 year after adjuvant chemotherapy had a median DFS of 6.0 months (95% CI 4.1–7.9), compared with 20.0 months (95% CI 7.1–32.9) in patients with PM within 1 year without adjuvant chemotherapy, 9.0 months (95% CI 4.5–13.5) in patients with PM after more than 1 year without adjuvant chemotherapy, and 13.0 months (95% CI 10.2–15.8) in patients with PM more than 1 year after adjuvant chemotherapy (p \ 0.001) (Fig. 1a). Further- more, patients with a PCI [ 11 had a DFS of 8.0 months (95% CI 4.9–11.1), compared with 13.0 months (95% CI 9.6–16.4) in patients with lower PCI scores (p = 0.002) (Fig. 1b). Other factors that were considered in univariate survival analysis included gender, sex, age, ASA classiﬁ- cation, poor or signet cell tumor differentiation, mucinous tumor type, primary tumor location, primary tumor stage, perioperative treatment, and type of chemotherapy, and were therefore not associated with DFS (Table 2). Operative Treatment TABLE 2 Overview of univariate survival analysis Characteristic DFS p valuea OS p valuea n Median DFS (95% CI) n Median OS (95% CI) General characteristics All patients 138 12.0 (8.8–13.2) 175 27.0 (20.6–33.4) Sex Male 58 11.0 (9.4–12.6) 0.367 82 26.0 (9.7–42.3) 0.768 Female 80 12.0 (8.1–15.9) 93 28.0 (21.6–34.3) Age, years B 60 56 12.0 (8.2–15.8) 0.657 69 28.0 (18.5–37.5) 0.487 [ 60 82 11.0 (8.7–13.3) 106 27.0 (17.9–36.1) ASA classiﬁcation I–II 121 12.0 (9.7–14.3) 0.309 150 27.0 (20.1–33.8) 0.171 III 17 9.0 (3.8–14.2) 25 16.0 (8.8–23.2) Primary tumor characteristics Location Colon 126 12.0 (10.0–14.0) 0.072 159 28.0 (21.7–34.3) 0.237 Rectum 12 6.0 (4.7–7.3) 16 19.0 (2.9–35.1) Differentiation Good/moderate 110 12.0 (9.8–14.2) 0.918 127 35.0 (21.6–48.4) 0.003 Poor/signet cell 16 9.0 (7.2–10.8) 24 9.0 (2.0–16.0) Histology Adenocarcinoma 117 12.0 (9.8–14.2) 0.301 138 29.0 (19.3–38.7) 0.392 Mucinous 19 9.0 (6.2–11.8) 31 23.0 (16.0–30.1) T stage T1–3 85 11.0 (8.8–13.2) 0.387 107 24.0 (15.0–33.0) 0.918 T4 51 14.0 (9.7–18.3) 66 29.0 (16.7–41.3) N stage N0 53 14.0 (6.7–21.3) 0.181 66 35.0 (20.6–49.4) 0.790 N1–2 84 11.0 (8.9–13.1) 108 24.0 (19.5–28.5) Distant metastases No 128 11.0 (8.7–13.3) 0.896 162 28.0 (21.0–35.0) 0.610 Yes 10 11.0 (2.0–20.0) 13 24.0 (18.4–29.6) Stage Stage 1–2 49 14.0 (6.1–21.9) 0.169 62 35.0 (18.4–51.6) 0.982 Stage 3–4 88 11.0 (9.0–13.0) 112 24.0 (19.8–28.2) Perioperative treatment Primary tumor: adjuvant chemotherapy No 52 12.0 (8.0–16.0) 0.194 64 35.0 (22.8–47.2) 0.658 Yes 86 11.0 (9.3–12.7) 111 24.0 (18.4–29.6) Development of PM after primary tumor resection B 1 year, no chemotherapy 25 20.0 (7.1–32.9) \ 0.001 30 42.0 (17.7–66.4) \ 0.001 [ 1 year, no chemotherapy 27 9.0 (4.5–13.5) 34 24.0 (15.9–32.1) B 1 year after chemotherapy 27 6.0 (4.1–7.9) 36 18.0 (11.7–24.3) [ 1 year after chemotherapy 59 13.0 (10.2–15.8) 75 56.0 (28.9–83.2) HIPEC: neoadjuvant chemotherapy No 124 12.0 (10.0–14.0) 0.781 154 27.0 (19.9–34.1) 0.565 Yes 14 9.0 (5.6–12.4) 21 24.0 (8.6–39.4) Clinical Prognosticators for HIPEC Surgery 2351 TABLE 2 continued Characteristic DFS p valuea OS p valuea n Median DFS (95% CI) n Median OS (95% CI) HIPEC: adjuvant chemotherapy No 77 11.0 (9.0–13.0) 0.496 102 24.0 (12.7–35.3) 0.167 Yes 61 12.0 (9.6–14.4) 72 28.0 (12.4–43.6) Prior surgical score 0–2 129 11.0 (8.8–13.2) 0.577 161 24.0 (16.8–31.2) 0.075 3 9 21.0 (8.2–33.8) 10 Not reached HIPEC/PM characteristics PCI B 11 91 13.0 (9.6–16.4) 0.002 94 56.0 (–) \ 0.001 [ 11 46 8.0 (4.9–11.1) 77 15.0 (9.4–20.6) HIPEC chemotherapy type Mitomycin C 126 11.0 (9.0–13.0) 0.649 128 37.0 (26.3–47.7) 0.922 Oxaliplatin 12 22.0 (0–51.5) 13 29.0 (–) ASA American Society of Anesthesiologists, CI conﬁdence interval, DFS disease-free survival, HIPEC hyperthermic intraperitoneal therapy, OS overall survival, PCI Peritoneal Cancer Index, PM peritoneal metastases aLog-rank test ASA American Society of Anesthesiologists, CI conﬁdence interval, DFS disease-free survival, HIPEC hyperthermic intraperitoneal therapy, OS overall survival, PCI Peritoneal Cancer Index, PM peritoneal metastases aLog-rank test aLog-rank test Baseline Characteristics In the VU University Medical Center and Catharina Hospital Eindhoven, 345 patients with peritoneally metastasized CRC were treated with the intent of CRS and HIPEC from January 2008 until May 2016. After exclusion of patients with synchronous metastases, other pathology subtypes, and patients without histologically proven PM, 175 patients were selected for further analysis. Table 1 represents the baseline characteristics of all patients. The majority of baseline characteristics did not differ signiﬁcantly between the group of patients with PM within 1 year after adjuvant chemotherapy and the groups of patients who did not receive adjuvant chemotherapy or were diagnosed with PM after more than 1 year (electronic supplementary Table 1). As expected, lymph node involvement at the time of primary tumor resection was more frequently diagnosed in patients who received adju- vant chemotherapy (p \ 0.001). 2352 N. R. Sluiter et al. Baseline Characteristics 1d). Third, a poor or signet cell tumor differentiation was associated with a worse median OS of 9.0 months (95% CI 2.0–16.0), compared with 35.0 months (95% CI 21.6–48.4) in patients with well to moderately differentiated primary tumors (p = 0.003) (Fig. 1e). b FIG. 1 Kaplan–Meier curves of all patients. Graphs (a) and (b) depict the disease-free survival curves: (a) patients with PM within 1 year without chemotherapy, versus PM after more than 1 year without chemotherapy, versus PM within 1 year after chemotherapy, versus PM more than 1 year after chemotherapy; (b) patients with a PCI B 11 versus patients with a PCI [ 11. Graphs c–e depict the overall survival curves: (c) patients with PM within 1 year without chemotherapy, versus PM after more than 1 year without chemotherapy, versus PM within 1 year after chemotherapy, versus PM more than 1 year after chemotherapy; (d) patients with a PCI B 11 versus patients with a PCI [ 11; (e) patients with poor/ signet cell differentiation versus patients with good/moderate differentiation. PM peritoneal metastases, PCI Peritoneal Cancer Index Multivariate Analysis Both high PCI (HR 1.04 for each PCI point, 95% CI 1.00–1.08, p = 0.040) and peritoneal recurrence within 1 year after adjuvant chemotherapy (HR 2.13, 95% CI 1.26–3.61, p = 0.005) remained signiﬁcant in the multi- variate DFS model (reference category: PM more than 1 year after chemotherapy) (Table 3). In the multivariate OS model, only peritoneal recurrence within 1 year after adjuvant chemotherapy (HR 2.76, 95% CI 1.45–5.27, p \ 0.002) and high PCI scores (HR 1.11 for each PCI point, 95% CI 1.07–1.15, p \ 0.001) were associated with worse OS (reference category: PM more than 1 year after chemotherapy) (Table 3). Variables that had a p value B 0.1 in univariate survival analysis, but were excluded from the ﬁnal model after multivariate Cox regression analysis, were primary tumor location (p = 0.130) for DFS and primary tumor differentiation (p = 0.118) and PSS (p = 0.376) for OS. Overall Survival The median OS of the entire cohort was 27.0 months (95% CI 20.6–33.4). Patients with a diagnosis of PM within 1 year after adjuvant chemotherapy had worse median OS (18.0 months, 95% CI 11.7–24.3) than patients with PM within 1 year without adjuvant chemotherapy (42.0 months, 95% CI 17.7–66.4), patients with PM after more than 1 year without adjuvant chemotherapy (24.0 months, 95% CI 15.9–32.1), and patients with PM more than 1 year after adjuvant chemotherapy (56.0 months, 95% CI 28.9–83.2) (p \ 0.001). PCI scores [ 11 were associated with worse median OS (15 months, Baseline Characteristics p<0.001 15.0 months 56.0 months d Overall survival - PCI PCI ≤11 (n=94) PCI >11 (n=77) All patients (n=175) c Overall survival - Development of PM e Overall survival - Differentiation p=0.003 9.0 months 35.0 months Good/moderate diﬀerenaon (n=127) Poor/signet cell diﬀerenaon (n=24) All paents (n=175) a Disease-free survival - Development of PM PM ≤ 1 year, no chemotherapy (n=25) PM > 1 year, no chemotherapy (n=27) PM ≤ 1 year after chemotherapy (n=27) PM > 1 year after chemotherapy (n=59) All patients (n=138) p<0.001 p=0.002 b Disease-free survival - PCI 8.0 months 13.0 months PCI ≤11 (n=91) PCI >11 (n=46) All patients (n=138) 27.0 months 27.0 months 27.0 months 13.0 months 12.0 months PM ≤ 1 year, no chemotherapy (n=30) PM > 1 year, no chemotherapy (n=34) PM ≤ 1 year after chemotherapy (n=36) PM > 1 year after chemotherapy (n=75) All patients (n=175) 42.0 months 56.0 months 18.0 months 24.0 months p<0.001 20.0 months 9.0 months 6.0 months 12.0 months 1,0 0,8 0,6 0,4 0,2 0,0 0 20 40 Disease-free survival (months) Disease-free survival (months) Overall survival (months) Overall survival (months) Overall survival (months) 60 80 1,0 0,8 0,6 0,4 0,2 0,0 0 20 40 60 80 1,0 0,8 0,6 0,4 0,2 0,0 0 20 40 60 80 1,0 0,8 0,6 0,4 0,2 0,0 0 20 40 60 80 1,0 0,8 0,6 0,4 0,2 0,0 0 20 40 60 80 a Disease-free survival - Development of PM p<0.001 13.0 months 20.0 months 9.0 months 6.0 months 12.0 months 1,0 0,8 0,6 0,4 0,2 0,0 0 20 40 Disease-free survival (months) 60 80 a Disease-free survival - Development of PM c p 27.0 months 42.0 months 56.0 months 18.0 months 24.0 months p<0.001 Overall survival (months) 1,0 0,8 0,6 0,4 0,2 0,0 0 20 40 60 80 p<0.001 15.0 months 56.0 months d Overall survival - PCI 27.0 months Overall survival (months) 1,0 0,8 0,6 0,4 0,2 0,0 0 20 40 60 80 p=0.002 b Disease-free survival - PCI 8.0 months 13.0 months PCI ≤11 (n=91) PCI >11 (n=46) All patients (n=138) 12.0 months Disease-free survival (months) 1,0 0,8 0,6 0,4 0,2 0,0 0 20 40 60 80 p<0.001 15.0 months PCI ≤11 (n=94) PCI >11 (n=77) All patients (n=175) e Overall survival - Differentiation p=0.003 9.0 months 35.0 months Good/moderate diﬀerenaon (n=127) Poor/signet cell diﬀerenaon (n=24) All paents (n=175) 27.0 months 27.0 months Overall survival (months) Overall survival (months) 0,2 0,0 0 20 40 60 80 1,0 0,8 0,6 0,4 0,2 0,0 0 20 40 60 80 2353 Clinical Prognosticators for HIPEC Surgery 95% CI 9.4–20.6) compared with patients with a PCI B 11 (56.0 months, 95% CI –) (p \ 0.001; Fig. CI conﬁdence interval, DFS disease-free survival, OS overall sur- vival, PCI Peritoneal Cancer Index, PM peritoneal metastases, PSS prior surgical score DISCUSSION A single-center study including 21 cases found a median OS of 28 months in patients who did not respond to adjuvant chemotherapy, which led the authors to conclude that the survival of this group was comparable with survival rates in other HIPEC patients6; however, this study did not include a control group. Another retrospec- tive study compared 19 patients with tumor progression despite neoadjuvant chemotherapy with the same number of patients with stable disease under neoadjuvant chemotherapy, resulting in an insigniﬁcant survival dif- ference between these groups.27 The relatively small groups in both studies might hamper these outcomes and conclusions. To provide the answers necessary to take our results into the clinical practice, a prospective study should com- pare HIPEC with systemic chemotherapy in patients with PM within 1 year after chemotherapy. Until then, CRS and HIPEC may be considered a valid option in this group of patients if the peritoneal tumor burden is limited (PCI B 11). Therefore, whether the limited DFS observed in this cohort of patients with early PM development justiﬁes the high morbidity and mortality rates of the CRS and HIPEC procedure should be carefully considered. Next to identi- ﬁcation of clinical prognosticators, future research should focus on identifying molecular tissue and blood-borne characteristics to select patients with biologically favorable tumor characteristics responding best to further aggressive therapies such as CRS and HIPEC. Additionally, identiﬁ- cation of biomarkers and clinical prognostic factors could potentially help us optimize the HIPEC procedure, aiming for a personalized treatment rather than the current one- size-ﬁts-all approach. This can be illustrated by selection of the most effective chemotherapeutic drug based on the prediction of individual chemotherapeutic responses,32 which may eventually result in a tailored and more robust HIPEC treatment, leading to improved oncologic outcomes in subgroups with relatively poor prognosis. High PCI emerged as a second characteristic that was associated with poor outcome in the present cohort. This variable is an established prognostic factor and is widely used for the identiﬁcation of patients with curable peri- toneal disease, deﬁned as a PCI B 20.11 The present results, supported by the available literature, emphasize the importance of performing a diagnostic laparoscopy to determine PCI indices prior to HIPEC surgery, especially since current imaging techniques lack sensitivity for detection of PM.28,29 No other variables were associated with DFS and OS in the multivariate models. DISCUSSION The diagnosis of PM relatively shortly after prior chemotherapy can be assessed preoperatively in contrast to well-established prognostic factors that are obtained after or during the operation, such as PCI, resection scores and combined prognostic scores, including the PSDSS 11 and the relatively novel COMPASS.14,15 Hence, the selection of patients with early peritoneal recurrence after prior chemotherapy might help identify patients who beneﬁt less from HIPEC, in this way preventing unnecessary exposure to an invasive procedure that may even harm the patient. The association between early peritoneal recurrences after adjuvant chemotherapy and poor outcome could be explained by a more aggressive biological tumor behavior, leading to a poor response to chemotherapy and, subse- quently, a poor response to HIPEC treatment. taken into account. First, our data are possibly subjected to selection bias that is associated with cohort studies. Sec- ond, based on our results, we can conclude that patients with early peritoneal recurrence after adjuvant chemotherapy have decreased survival; however, there is no solid evidence for an alternative treatment. The effect of second-line chemotherapy in this subgroup of patients, with a poor response to prior chemotherapy, has yet to be proven. Palliative treatment regimens are heterogeneous in this patient group, which makes it hard to gather valid retrospective data from a control cohort. Thereby, new options arise for patients not suitable for HIPEC treatment, including Pressurized IntraPeritoneal Aerosol Chemother- apy (PIPAC). This innovative therapy consists of repetitive intraperitoneal administration of aerosolized chemotherapy and has already been shown to be feasible and safe in end- stage PM originating from several primaries.30 For CRC patients, PIPAC with oxaliplatin, whether in combination with systemic chemotherapy or not, showed encouraging results.31 Future studies in well-deﬁned populations should demonstrate the potential role of PIPAC in patients with irresectable PM of CRC since the current evidence for PIPAC in this population is scarce. Several other study groups have already excluded patients with rapid PM development or progression under systemic chemotherapy from CRS and HIPEC.6,16,17 In addition, various guidelines recommend the use of this selection criterion in the workup of potential HIPEC can- didates.18,19 In contrast, two relatively small retrospective studies argued that there is no reason for excluding patients with quick peritoneal relapses after prior chemotherapy from HIPEC. DISCUSSION TABLE 3 Final model resulting from multivariate survival analysis Variable Hazard rate (95% CI) p value Disease-free survival PCI 1.04 (1.00–1.08) 0.040 Development of PM [ 1 year after chemotherapy Reference 0.001 B 1 year, no chemotherapy 0.57 (0.31–1.06) 0.075 [ 1 year, no chemotherapy 1.20 (0.67–2.19) 0.535 B 1 year after chemotherapy 2.13 (1.26–3.61) 0.005 Overall survival PCI 1.11 (1.07–1.15) \ 0.001 Development of PM [ 1 year after chemotherapy Reference 0.019 B 1 year, no chemotherapy 1.34 (0.62–2.92) 0.454 [ 1 year, no chemotherapy 1.89 (0.95–3.76) 0.071 B 1 year after chemotherapy 2.76 (1.45–5.27) 0.002 Variables with a p value B 0.1 in univariate survival analysis were included in a multivariate Cox regression analysis: (1) primary tumor location, development of PM, and PCI for DFS; (2) primary tumor differentiation, development of PM, PSS, and PCI for OS CI conﬁdence interval, DFS disease-free survival, OS overall sur- vival, PCI Peritoneal Cancer Index, PM peritoneal metastases, PSS prior surgical score TABLE 3 Final model resulting from multivariate survival analysis Variable Hazard rate (95% CI) p value This prospective cohort study shows that peritoneal recurrence within 1 year after previous adjuvant chemotherapy and a high initial PCI index are the most important risk factors for poor DFS and OS in patients with peritoneally metastasized CRC. This effect was indepen- dent of primary tumor stage or tumor differentiation. In 175 patients treated with the intent of CRS and HIPEC in two tertiary referral centers, patients with PM diagnosed within 1 year after adjuvant chemotherapy had worse median DFS (HR 2.13, p = 0.005) and OS (HR 2.76, p \ 0.001) than patients not treated with adjuvant chemotherapy or patients who were diagnosed with PM after more than 1 year in both univariate and multivariate analysis. A high PCI was the only other variable associated with DFS and OS in the multivariate models (HR 1.04, p = 0.040; and HR 1.11, p \ 0.001, respectively). Cytoreduction and HIPEC is currently the preferred treatment option for patients with PM of CRC,1, 2 but due to the relatively high morbidity rates associated with this procedure, the need for careful patient selection should be emphasized.8–10 Preoperative decision making is crucial, warranting practical clinical and pathological prognostic CI conﬁdence interval, DFS disease-free survival, OS overall sur- vival, PCI Peritoneal Cancer Index, PM peritoneal metastases, PSS prior surgical score 2354 N. R. Sluiter et al. factors. REFERENCES 1. Mirnezami R, Mehta AM, Chandrakumaran K, et al. Cytore- ductive surgery in combination with hyperthermic intraperitoneal chemotherapy improves survival in patients with colorectal peritoneal metastases compared with systemic chemotherapy alone. Br J Cancer. 2014;111(8):1500–1508. 16. Elias D, Delperro JR, Sideris L, et al. Treatment of peritoneal carcinomatosis from colorectal cancer: impact of complete cytoreductive surgery and difﬁculties in conducting randomized trials. Ann Surg Oncol. 2004;11(5):518–521. 17. Quenet F, Goere D, Mehta SS, et al. Results of two bi-institu- tional prospective studies using intraperitoneal oxaliplatin with or without irinotecan during HIPEC after cytoreductive surgery for colorectal carcinomatosis. Ann Surg. 2011;254(2):294–301. 2. Verwaal VJ, Zoetmulder FA. Follow-up of patients treated by cytoreduction and chemotherapy for peritoneal carcinomatosis of colorectal origin. Eur J Surg Oncol. 2004;30(3):280–285. 3. Franko J, Ibrahim Z, Gusani NJ, Holtzman MP, Bartlett DL, Zeh HJ 3rd. Cytoreductive surgery and hyperthermic intraperitoneal chemoperfusion versus systemic chemotherapy alone for col- orectal peritoneal carcinomatosis. Cancer. 2010;116(16):3756–3762. 18. Esquivel J. Cytoreductive surgery and hyperthermic intraperi- toneal chemotherapy for colorectal cancer: survival outcomes and patient selection. J Gastrointest Oncol. 2016;7(1):72–78. 19. Dube P, Sideris L, Law C, et al. Guidelines on the use of cytoreductive surgery and hyperthermic intraperitoneal chemotherapy in patients with peritoneal surface malignancy arising from colorectal or appendiceal neoplasms. Curr Oncol. 2015;22(2):e100–112. 4. Verwaal VJ, van Ruth S, de Bree E, et al. Randomized trial of cytoreduction and hyperthermic intraperitoneal chemotherapy versus systemic chemotherapy and palliative surgery in patients with peritoneal carcinomatosis of colorectal cancer. J Clin Oncol. 2003;21(20):3737–3743. 20. Sugarbaker PH. Patient selection and treatment of peritoneal carcinomatosis from colorectal and appendiceal cancer. World J Surg. 1995;19(2):235–240. 5. Bloemendaal AL, Verwaal VJ, van Ruth S, Boot H, Zoetmulder FA. Conventional surgery and systemic chemotherapy for peri- toneal carcinomatosis of colorectal origin: a prospective study. Eur J Surg Oncol. 2005;31(10):1145–1151. 21. Elias D, Blot F, El Otmany A, et al. Curative treatment of peri- toneal carcinomatosis arising from colorectal cancer by complete resection and intraperitoneal chemotherapy. Cancer. 2001;92(1):71–76. 6. Elias D, Gilly F, Boutitie F, et al. Peritoneal colorectal carcino- matosis treated with surgery and perioperative intraperitoneal chemotherapy: retrospective analysis of 523 patients from a multicentric French study. J Clin Oncol. 2010;28(1):63–68. 22. Sugarbaker PH. Intraperitoneal chemotherapy and cytoreductive surgery for the prevention and treatment of peritoneal carcino- matosis and sarcomatosis. Semin Surg Oncol. 1998;14(3):254–261. 7. Kuijpers AM, Mirck B, Aalbers AG, et al. CONCLUSIONS perioperative intraperitoneal chemotherapy for colorectal peri- toneal carcinomatosis. World J Surg. 2010;34(1):70–78. 10. Ihemelandu CU, McQuellon R, Shen P, Stewart JH, Votano- poulos K, Levine EA. Predicting postoperative morbidity following cytoreductive surgery with hyperthermic intraperi- toneal chemotherapy (CS ? HIPEC) with preoperative FACT-C (Functional Assessment of Cancer Therapy) and patient-rated performance status. Ann Surg Oncol. 2013;20(11):3519–3526. The present study found support for considering both PCI and early peritoneal recurrence after adjuvant therapy in patient selection prior to CRS and HIPEC. Prospective trials might help us move forward by conﬁrming or rejecting factors associated with outcome in retrospective studies, ultimately providing clear guidelines to identify the right patients for the right treatment. 11. Kwakman R, Schrama AM, van Olmen JP, et al. Clinicopatho- logical parameters in patient selection for cytoreductive surgery and hyperthermic intraperitoneal chemotherapy for colorectal cancer metastases: a meta-analysis. Ann Surg. 2016;263(6):1102–1111. DISCLOSURES Nina R. Sluiter, Koen P. Rovers, Youssra Salhi, Stijn L. Vlek, Veerle M. H. Coupe´, Henk M. W. Verheul, Geert Kazemier, Ignace H. J. T. de Hingh, and Jurriaan B. Tuynman declare no conﬂicts of interest. DISCLOSURES Nina R. Sluiter, Koen P. Rovers, Youssra Salhi, Stijn L. Vlek, Veerle M. H. Coupe´, Henk M. W. Verheul, Geert Kazemier, Ignace H. J. T. de Hingh, and Jurriaan B. Tuynman declare no conﬂicts of interest. 12. Sluiter NR, de Cuba EM, Kwakman R, et al. Versican and vas- cular endothelial growth factor expression levels in peritoneal metastases from colorectal cancer are associated with survival after cytoreductive surgery and hyperthermic intraperitoneal chemotherapy. Clin Exp Metastasis. 2016;33(4):297–307. 13. de Cuba EM, de Hingh IH, Sluiter NR, et al. Angiogenesis-re- lated markers and prognosis after cytoreductive surgery and hyperthermic intraperitoneal chemotherapy for metastatic col- orectal cancer. Ann Surg Oncol. 2016;23(5):1601–1608. OPEN ACCESS This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://crea tivecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. 14. Simkens GA, van Oudheusden TR, Nieboer D, et al. Develop- ment of a prognostic nomogram for patients with peritoneally metastasized colorectal cancer treated with cytoreductive surgery and HIPEC. Ann Surg Oncol. 2016;23(13):4214–4221. 15. Demey K, Wolthuis A, de Buck van Overstraeten A, et al. DISCUSSION Importantly, the primary tumor stage was not a signiﬁcant risk factor for poor DFS and OS. To our knowledge, this is currently the largest study showing that the development of PM following adjuvant therapy is a poor prognostic factor after treatment with CRS and HIPEC. However, some limitations should be Clinical Prognosticators for HIPEC Surgery 2355 CONCLUSIONS External validation of the prognostic nomogram (COMPASS) for patients with peritoneal carcinomatosis of colorectal cancer. Ann Surg Oncol. 2017;24(12):3604–3608. REFERENCES Cytoreduction and HIPEC in the Netherlands: nationwide long-term outcome fol- lowing the Dutch protocol. Ann Surg Oncol. 2013;20(13):4224–4230. 23. Verwaal VJ, van Ruth S, Witkamp A, Boot H, van Slooten G, Zoetmulder FA. Long-term survival of peritoneal carcinomatosis of colorectal origin. Ann Surg Oncol. 2005;12(1):65–71. 8. Randle RW, Doud AN, Levine EA, et al. Peritoneal surface disease with synchronous hepatic involvement treated with Cytoreductive Surgery (CRS) and hyperthermic intraperitoneal chemotherapy (HIPEC). Ann Surg Oncol. 2015;22(5):1634–1638. 24. Trotti A, Colevas AD, Setser A, et al. CTCAE v3.0: development of a comprehensive grading system for the adverse effects of cancer treatment. Semin Radiat Oncol. 2003;13(3):176–181. 25. Werkgroep Colorectaal Carcinoom. Richtlijn Colorectaal Carci- noom. 2014. http://www.oncoline.nl/colorectaalcarcinoom. Accessed 1 Feb 2018. 9. Saxena A, Yan TD, Morris DL. A critical evaluation of risk factors for complications after cytoreductive surgery and 2356 N. R. Sluiter et al. 26. Klaver YL, de Hingh IH, Boot H, Verwaal VJ. Results of cytoreductive surgery and hyperthermic intraperitoneal chemotherapy after early failure of adjuvant systemic chemotherapy. J Surg Oncol. 2011;103(5):431–434. 30. Grass F, Vuagniaux A, Teixeira-Farinha H, Lehmann K, Demartines N, Hubner M. Systematic review of pressurized intraperitoneal aerosol chemotherapy for the treatment of advanced peritoneal carcinomatosis. Br J Surg. 2017;104(6):669–678. 27. Passot G, Vaudoyer D, Cotte E, et al. Progression following neoadjuvant systemic chemotherapy may not be a contraindica- tion to a curative approach for colorectal carcinomatosis. Ann Surg. 2012;256(1):125–129. 31. Demtroder C, Solass W, Zieren J, Strumberg D, Giger-Pabst U, Reymond MA. Pressurized intraperitoneal aerosol chemotherapy with oxaliplatin in colorectal peritoneal metastasis. Colorectal Dis. 2016;18(4):364–371. 28. Koh JL, Yan TD, Glenn D, Morris DL. Evaluation of preopera- tive computed tomography in estimating peritoneal cancer index in colorectal peritoneal carcinomatosis. Ann Surg Oncol. 2009;16(2):327–333. 32. Kwakman R, de Cuba EM, de Winter JP, et al. Tailoring heated intraperitoneal mitomycin C for peritoneal metastases originating from colorectal carcinoma: a translational approach to improve survival. Br J Cancer. 2015;112(5):851–856. 29. Chang MC, Chen JH, Liang JA, Huang WS, Cheng KY, Kao CH. PET or PET/CT for detection of peritoneal carcinomatosis: a meta-analysis. Clin Nucl Med. 2013;38(8):623–629.
https://openalex.org/W2815351724	https://ojs.unpkediri.ac.id/index.php/matematika/article/download/11992/907	Indonesian	null	Meningkatkan Hasil Belajar Materi Garis Dan Sudut Dengan Pendekatan Aktivitas Pada Siswa Kelas VII Semester 2 SMP Negeri 1 Ngantru Tulungagung	Jurnal math educator Nusantara : wahana publikasi karya tulis ilmiah di bidang pendidikan matematika/Jurnal Math Educator Nusantara : wahana publikasi karya tulis ilmiah di bidang pendidikan matematika	2,018	cc-by	3,522	Jurnal Math Educator Nusantara (JMEN) Wahana publikasi karya tulis ilmiah di bidang pendidikan matematika p-issn: 2459-9735 e-issn: 2580-9210 http://ojs.unpkediri.ac.id/index.php/matematika Article received : 07 April 2018, article revised : 24 Mei 2018, article published: 30 Mei 2018 Abstrak: Terwujudnya kondisi pembelajaran siswa aktif merupakan harapan dari semua komponen pendidikan termasuk masyarakat dan praktisi pendidikan. Oleh sebab itu dalam kegiatan pembelajaran dituntut satu stategi pembelajaran yang direncanakan oleh guru dengan mengedepankan keaktifan siswa dalam kegiatan belajar mengajar. Melalui kegiatan belajar yang menekankan pada aktivitas siswa diharapkan mampu meningkatkan motivasi dan hasil belajar yang sesuai dengan tujuan pendidikan di sekolah. Ada beberapa jenis aktivitas yang disampaikan oleh para ahli, diantaranya: (1) kegiatan-kegiatan visual, (2) kegiatan-kegiatan lisan (oral), (3) mendengarkan, (4) menulis, (5) menggambar, (6) metric, (7) mental, dan (8) emosional. Paparan data data dalam pembahasan penelitian tindakan (action research) ini pada dasarnya menjabarkan tentang upaya peningkatan aktivitas belajar siswa dalam kegiatab belajar melalui kegiatan pembelajaran berbasis aktivitas. Ada beberapa hal cara meningkatkan keterlibatan siswa dalam upaya meningkatkan aktivitas belajar, diantaranya: (1) guru diharapkan dapat mengenal dan membantu anak-anak yang kurang terlibat. Selidiki apa yang menyebabkannya dan usaha apa yang bisa dilakukan untuk meningkatkan partisipasi anak tersebut, (2)guru harus menyiapkan mental dan kemandirian belajar siswa secara tepat, dan (3) sesuaikan pengajaran dengan kebutuhan-kebutuhan individual siswa. Hasil penelitian dapat disimpulkan bahwa guru dalam upayanya meningkatkan hasil belajar siswa dengan menggunakan pendekatan pembelajaran berbasis aktifitas berhasil. Keberhasilan bisa diamati dari jumlah ketuntasan belajar individu siswa (nilai 75) dengan ketuntasan kelas rata-rata 86,2 % ,rata-rata nilai siswa 83,3,dan terjadi peningkatan 31 % dari tahun pelajaran 2015-2016 dengan kategori tindakan cukup baik setelah menuntaskan siklus kedua. Kata Kunci: Aktivitas, hasil belajar garis dan sudut, kelas VII SMP Meningkatkan Hasil Belajar Materi Garis dan Sudut Dengan Pendekatan Aktivitas Pada Siswa Kelas VII Semester 2 SMP Negeri 1 Ngantru Tulungagung Bagus Nirwono Guru matematika SMPN 1 Ngantru Kecamatan Ngantru Kabupaten Tulungagung E-mail: nirwonob@gmail.com PENDAHULUAN Matematika sudah lama menjadi momok bagi kebanyakan siswa. Hal ini dapat diidentifikasi dari pendapat negatif kebanyakan siswa. Keadaan ini membuat kebanyakan siswa mudah menyerah belajar matematika apalagi ada pendapat yang mengatakan bahwa tidak bisa matematika itu wajar karena memang sulit. Temannya banyak dan sebagainya. Halnya dengan siswa kelas VII F SMPN 1 Ngantru pada tahun pelajaran 2015-2016 mempunyai nilai rata-rata 62,5 dengan ketuntasan klasikal mencapai 55,2 % pada materi garis dan sudut jauh dari KKM yang telah ditentukan yaitu 75 dengan ketuntasan klasikal 85% Untuk mengatasi masalah diatas, guru harus mengambil langkah bijaksana. Berdasar diskusi dengan teman sejawat maka diperoleh beberapa solusi.Salah satu langkah yang ditempuh adalah pembelajaran aktif. Menurut Suparno, dkk (2002) siswa yang aktif dalam proses pembelajaran dicirikan oleh dua aktivitas, yaitu aktivitas dalam berpikir {minds-on), dan aktivitas dalam berbuat {hands-on). Perbuatan nyata siswa dalam pembelajaran merupakan hasil keterlibatan berpikir siswa terhadap kegiatan belajarnya. Dengan demikian DOI: doi.org/10.29407/jmen.v4i01.11992 71 \| Jurnal Math Educator Nusantara, Vol. 4 No. 1, Mei 2018, pp. 70 - 78 proses siswa aktif dalam kegiatan belajar mengajar merupakan suatu kegiatan pembelajaran yang harus dilaksanakan secara terus menerus dan tiada henti. Hal ini dapat dilakukan apabila interaksi antara guru dan siswa terjalin dengan baik. Sebab menurut Usman (2002) interaksi dan hubungan timbal balik antara guru dan siswa itu merupakan syarat utama bagi berlangsungnya proses belajar mengajar. Dalam aktivitas pembelajaran di sekolah, guru harus mengusahakan agar siswa dapat melakukan pengamatan yang efektif agar memperoleh hasil pembelajaran yang sebaik- baiknya. Dalam mengajar, hendaknya memberikan kesempatan kepada siswa untuk melakukan pengamatan yang sebaik-baiknya. Beberapa hal yang dapat dilakukan guru untuk membantu siswa melakukan pengamatan yang baik dalam pembelajaran adalah sebagai berikut: (a) Pengamatan akan lebih efektif kepada rangsangan-rangsangan yang mempunyai struktur dan bentuk yang jelas. Oleh karena itu, hal-hal yang akan dipelajari hendaknya mempunyai struktur dan organisasi yang jelas endaknya mempunyai struktur dan organisasi yang jelas. a. Pengamatan kepada sesuatu yang dekat akan lebih berkesan. Oleh karena itu, siswa diberi banyak kesempatan untuk lebih dekat dengan hal-hal yang akan dipelajari. a. Pengamatan kepada sesuatu yang dekat akan lebih berkesan. Oleh karena itu, siswa diberi banyak kesempatan untuk lebih dekat dengan hal-hal yang akan dipelajari. . waktu guru mengajar, sebaiknya dimulai dengan pengalaman-pengalaman siswa. b. waktu guru mengajar, sebaiknya dimulai dengan pengalaman-pengalaman siswa. c. Pengamatan dimulai dengan keseluruhan, baru kemudian kepada bagian-bagian. Copyright © 2018, JMEN (Jurnal Math Educator Nusantara), Wahana Publikasi Karya Tulis Ilmiah di Bidang Pendidikan Matematika ISSN 2459-9735 (print), ISSN 2580-9210 (online) PENDAHULUAN Oleh karena itu dalam memberikan bahan yang akan diajarkan, sebaiknya dimulai dengan keseluruhan, baru kemudian kepada bagian-bagian yang lebih khusus. c. Pengamatan dimulai dengan keseluruhan, baru kemudian kepada bagian bagian. Oleh karena itu dalam memberikan bahan yang akan diajarkan, sebaiknya dimulai dengan keseluruhan, baru kemudian kepada bagian-bagian yang lebih khusus. d. Pengamatan dipengaruhi oleh peringkat perkembangan individu. Oleh karena itu, pengajaran hendaknya disesuaikan dengan peringkat perkembangan individu, terutama peringkat perkembangan kognitif. d. Pengamatan dipengaruhi oleh peringkat perkembangan individu. Oleh karena itu, pengajaran hendaknya disesuaikan dengan peringkat perkembangan individu, terutama peringkat perkembangan kognitif. e. Terdapat perbedaan individual dalam pengamatan. Tiap individu mempunyai macam gaya pengamatan (ada gaya visual, auditif, taktis, dan kinestetik). Oleh karena itu pembelajaran hendaknya disesuaikan dengan gaya pengamatan masing-masing siswa. Beberapa faktor dapat menimbulkan terjadinya kesalahan atau kelainan pengamatan, seperti rangsangan yang kurang jelas, kurangnya perhatian siswa, pengalaman di masa lampau, kurang baiknya alat indera, lingkungan yang mengganggu, dan sebagainya. Menurut Hamalik (2001) mengatakan bahwa pengajaran yang efektif adalah pengajaran yang menyediakan kesempatan belajar sendiri atau melakukan aktivitas sendiri. Dalam kemajuan metodologi dewasa ini asas aktivitas lebih ditonjolkan melalui suatu program unit activity, sehingga kegiatan belajar siswa menjadi dasar untuk mencapai tujuan dan hasil belajar yang lebih memadai. Ada beberapa jenis aktivitas yang disampaikan oleh para ahli, diantaranya: (1) kegiatan-kegiatan visual, (2) kegiatan-kegiatan lisan (oral), (3) mendengarkan, (4) menulis, (5) menggambar, (6) metrik, (7) mental, dan (8) emosional.(Hamalik : 2001) Dari beberapa macam aktivitas tersebut menunjukkan bahwa dalam kegiatan pembelajaran, aktivitas siswa sangat diperlukan dalam memenuhi tujuan pembelajaran. Sehingga dalam suatu kegiatan pembelajaran, aktivitas siswa harus disesuaikan dengan materi pembelajaran yang disampaikan oleh guru atau masalah yang sedang dibahas. Nirwono, B., Meningkatkan Hasil Belajar Materi Garis dan... 72 Kurikulum Matematika harus memberikan pengalaman belajar yang melibatkan siswa pada proses dan produk dalam ilmu pengetahuan dan teknologi. Pendekatan yang dipakai dalam kurikulum diharapkan akan mendorong siswa menjadi pelajar yang aktif dan fleksibel. Secara khusus pendekatan ini akan: (1) memperhatikan pcrbedaan individu siswa, (2) memberikan kesempatan yang sama kepada semua siswa imtuk mempelajari konsep-konsep esensial, (3) membekali siswa dengan ketrampilan untuk memahami dunia melalui penyelidikan, dan (4) membekali siswa dengan ketrampilan baik untuk memilih alat-alat yang sesuai maupun bahan-bahan yang diperlukan. Kurikulum Matematika perlu memberikan pengalaman belajar yang membantu siswa memenuhi kebutuhan pribadi, sosial, lingkungan dan ekonomi. Copyright © 2018, JMEN (Jurnal Math Educator Nusantara), Wahana Publikasi Karya Tulis Ilmiah di Bidang Pendidikan Matematika ISSN 2459-9735 (print), ISSN 2580-9210 (online) PENDAHULUAN Standar kompetensi (SK) dalam pembelajaran materi dan sudut kelas VII SMP adalah Geometri (5) memahami hubungan garis dengan garis, garis dengan sudut, sudut dengan sudut serta menentukan ukurannya. Yang menjadi Kompetensi dasar (KD) adalah : (1) Menentukan hubungan antara dua garis ,serta besar dan jenis sudut.(2) Memahami sifat-sifat sudut yang terbentuk jika dua garis berpotongan atau dua garis sejajar berpotongan dengan garis lain.(3) Melukis sudut (4) Membagi sudut. Kegiatan pembelajaran sesuai KD 5.1 adalah (a) kedudukan dua garis seperti sejajar,berimpit,berpotongan dan bersilangan (b) satuan sudut yang sering digunakan seperti derajat dan radian. (c) pengukuran sudut dengan busur derajat (d) jenis-jenis sudut seperti sudut lancip,siku-siku dan tumpul. Kegiatan pembelajaran sesuai KD 5.2 adalah (a) sifat sudut akibat perpotongan dua garis (b) sifat sudut akibat perpotongan dua garis sejajar dengan sebuah garis. (c) Menyelesaikan soal yang berhubungan dengan sifat sudut. Kegiatan pembelajaran sesuai KD 5.3 adalah (a) Melukis sudut dengan penggaris dan busur derajat (b) Memindahkan sudut dengan penggaris dan jangka (c) Menggunakan jangka dan penggaris melukis sudut 60 dan90 (d) melukis sudut siku-siku dengan sepasang penggaris berbentuk segitiga. Kegiatan pembelajaran sesuai KD 5.4 adalah (a) Membagi sudut menjadi dua sama besar deengan penggaris dan jangka (b) Menggunakan penggaris dan jangka untuk melukis sudut 30, 45, 120 dan 150. Berdasarkan pada pendapat diatas menunjukkan bahwa aktifitas siswa dalam kegiatan belajar mengajar mutlak diperlukan. Namun yang lebih penting lagi dalam meningkatkan aktivitas siswa tersebut adalah kemampuan guru dalam merencanakan suatu kegiatan belajar mengajar tersebut dalam mencapai tujuan pembelajaran. : "Jika pendekatan pembelajaran yang selama ini digunakan guru mata pelajaran Matematika dalam materi garis dan sudut, diganti dengan pendekatan pembelajaran berbasis aktivitas, maka dimungkinkan akan terjadi peningkatan prestasi belajar siswa pada materi itu". Penelitian tindakan kelas ini bertujuan untuk mengetahui dan mendeskripsikan (1) Pengaruh penggunaan pendekatan aktivitas dalam pembelajaran garis dan sudut terhadap motivasi belajar siswa (2) pengaruh penggunaan pendekatan berbasis aktivitas dalam pembelajaran garis dan sudut terhadap prestasi belajar siswa. Sedangkan penelitian memberi manfaat antara lain (1) Mengembangkan kemampuan profesi dan meningkatkan ketrampilan dalam proses pembelajaran oleh guru (2) mengembangkan proses pembelajaran agar lebih 73 \| Jurnal Math Educator Nusantara, Vol. 4 No. 1, Mei 2018, pp. 70 - 78 73 aktif, kreatif, dan menyenangkan (bermakna) bagi siswa (3) Bagi sekolah untuk peningkatan proses dan hasil pembelajaran. aktif, kreatif, dan menyenangkan (bermakna) bagi siswa (3) Bagi sekolah untuk peningkatan proses dan hasil pembelajaran. METODE Dalam penelitian ini digunakan jenis studi kasus. Alasan pemilihan karena berusaha meneliti pengaruh (refleksi) suatu tindakan yang diberikan pada unit sosial secara mendalam. Masalah yang dijadikan penekanan adalah bagaimana perilaku siswa bisa terpengaruh oleh proses pembelajaran yang menggunakan pendekatan aktifitas sehingga prestasi belajarnya meningkat atau belajarnya tuntas. Supaya diperoleh gambaran yang jelas dan mengenai sasaran maka data diambil dari satu obyek saja kemudian di teliti lebih mendalam. Yang digunakan sebagai subyek penelitian adalah siswa kelas VII-A SMPN 1 Ngantru Tulungagung Jawa Timur pada semester genap tahun pelajaran 2016-2017 yang memiliki kemampuan heterogen. Penelitian yang dilakukan meliputi tiga tahapan yaitu (1) tahap persiapan, (2) tahap pelaksanaan, (3) tahap pelaporan. Pada tahap pelaksanaan terbagi atas 3 siklus. Pengumpulan data yang digunakan dalam penelitian ini adalah (1) observasi, (2) wawancara, (3) angket, (4) dokumentasi. Untuk memperoleh keabsakan temuan, peneliti menggunakan tehnik trianggulasi yaitu menguji kebenaran data tertentu dengan membandingkannya dengan data yang diperoleh dari sumber lain, pada berbagai fase penelitian lapangan, pada waktu yang berlainan dan sering menggunakan metode yang berlainan. Pada penelitian ini analisa data dilakukan terus menerus, mulai siklus tindakan awal sampai siklus tindakan terakhir. Tehnik analisa data yang digunakan adalah (1) reduksi data, (2) display, (3) kesimpulan dan verifikasi. Skor hasil tes siswa dalam mengerjakan soal–soal yang meliputi skor hasil tes pengetahuan prasyarat yang diberikan sebelum tindakan, hasil tes pada setiap akhir tindakan, dan hasil pekerjaan siswa dalam kegiatan pembelajaran. Hasil pekerjaan tersebut akan digunakan unuk melihat peningkatan pemahaman dan pen capaian hasil belajar siswa. satu Presentase ketuntasan individual ( Agustina dalam tim STKIP : 2007) adalah perbandingan jumlah skor yang dicapai siswa dengan skor maksimum/ideal dikalikan seratus persen.Siswa dianggap tuntas secara individu, jika ketuntasan yang dicapai minimal 75% (Kriteria Ketuntasan Minimal atau KKM ) .Persentase ketuntasan kelas (Agustina dalam Tim STKIP: 2007) adalah perbandingan jumlah siswa yang tuntas dengan jumlah siswa dikalikan seratus persen.Kelas/Klasikal dianggap tuntas jika mencapai 85 %. Copyright © 2018, JMEN (Jurnal Math Educator Nusantara), Wahana Publikasi Karya Tulis Ilmiah di Bidang Pendidikan Matematika ISSN 2459-9735 (print), ISSN 2580-9210 (online) METODE Kriteria taraf keberhasilan tindakan (Suharsimi dalam tim STKIP:2007 ) dapat ditentukan sebagai berikut : 75 % < NR < 100 % : Sangat Baik 25 % < NR < 50 % : Cukup Baik 50 % < NR < 75 % : Baik 0 % < NR < 25 % : Kurang Baik 75 % < NR < 100 % : Sangat Baik 25 % < NR < 50 % : Cukup Baik 50 % < NR < 75 % : Baik 0 % < NR < 25 % : Kurang Baik Copyright © 2018, JMEN (Jurnal Math Educator Nusantara), Wahana Publikasi Karya Tulis Ilmiah di Bidang Pendidikan Matematika ISSN 2459-9735 (print), ISSN 2580-9210 (online) Copyright © 2018, JMEN (Jurnal Math Educator Nusantara), Wahana Publikasi Karya Tulis Ilmiah di Bidang Pendidikan Matematika ISSN 2459-9735 (print), ISSN 2580-9210 (online) Nirwono, B., Meningkatkan Hasil Belajar Materi Garis dan... \| 7 74 Copyright © 2018, JMEN (Jurnal Math Educator Nusantara), Wahana Publikasi Karya Tulis Ilmiah di Bidang Pendidikan Matematika ISSN 2459-9735 (print), ISSN 2580-9210 (online) Siklus 1 Paparan data dalam pembahasan penelitian tindakan {action research) ini pada dasarnya menjabarkan tentang upaya peningkatan prestasi belajar siswa melalui kegiatan pembelajaran berbasis aktivitas. Kedatangan guru ke kekelas mendapati (1) Kesiapan pembelajaran pendidik baik dan peserta didik cukup (2) Seting pembelajaran baik (3) Penampilan pendidik baik, penampilan peserta dan lingkungan sedang serta usaha pendidik memperbaiki peserta dan lingkungan baik (4) Kegiatan membuka pembelajaran baik (5) Kegiatan inti pembelajaran sedang (6) Kegiatan menutup pembelajaran sedang (7) Pengalaman pembelajaran bagi pendidik baik bagi peserta sedang Kegiatan pembelajaran sesuai KD 5.1 adalah (a) kedudukan dua garis seperti sejajar,berimpit,berpotongan dan bersilangan (b) satuan sudut yang sering digunakan seperti derajat dan radian. (c) pengukuran sudut dengan busur derajat (d) jenis-jenis sudut seperti sudut lancip,siku-siku dan tumpul.Kegiatan pembelajaran sesuai KD 5.2 adalah (a) sifat sudut akibat perpotongan dua garis (b) sifat sudut akibat perpotongan dua garis sejajar dengan sebuah garis. (c) Menyelesaikan soal yang berhubungan dengan sifat sudut. Selanjutnya untuk membuktikan keefektifan penggunaan pembelajaran berbasis aktivitas dalam kegiatan belajar mengajar, selanjutnya strategi pembelajaran ini ditindaklanjuti pada kegiatan siklus berikutnya Kegiatan pada siklus 2, pada dasarnya sama dengan apa yang dilakukan oleh peneliti dalam kegiatan belajar mengajar yang dilakukan pada siklus 1 ini, yaitu dilaksanakan selama dua kali pertemuan masing-masing pertemuan 2 x 40 menit. Ada pun hasil dan kegiatan belajar mengajar pada kegiatan siklus 2 ini, secara rinci akan dipaparkan sebagai berikut di bawah ini.(1).Kesiapan pembelajaran pendidik baik dan peserta didik baik (2 ).Seting pembelajaran baik(3).Penampilan pendidik baik, penampilan peserta dan lingkungan sedang serta usaha pendidik memperbaiki peserta dan lingkungan baik (4).Kegiatan membuka pembelajaran baik (5).Kegiatan inti pembelajaran baik (6).Kegiatan menutup pembelajaran baik (7).Pengalaman pembelajaran bagi pendidik baik bagi peserta baik (8). Kegiatan pembelajaran sesuai KD 5.3 adalah (a) Melukis sudut dengan penggaris dan busur derajat (b) Memindahkan sudut dengan penggaris dan jangka (c) Menggunakan jangka dan penggaris melukis sudut 60 dan90 (d) melukis sudut siku-siku dengan sepasang penggaris berbentuk segitiga.Kegiatan pembelajaran sesuai KD 5.4 adalah (a) Membagi sudut menjadi dua sama besar deengan penggaris dan jangka (b) Menggunakan penggaris dan jangka untuk melukis sudut 30, 45, 120 dan 150. Copyright © 2018, JMEN (Jurnal Math Educator Nusantara), Wahana Publikasi Karya Tulis Ilmiah di Bidang Pendidikan Matematika ISSN 2459-9735 (print), ISSN 2580-9210 (online) 75 \| Jurnal Math Educator Nusantara, Vol. 4 No. 1, Mei 2018, pp. Copyright © 2018, JMEN (Jurnal Math Educator Nusantara), Wahana Publikasi Karya Tulis Ilmiah di Bidang Pendidikan Matematika ISSN 2459-9735 (print), ISSN 2580-9210 (online) Siklus 1 70 - 78 Tabel 1 : Hasil penilaian tes siklus 1 dan 2 NO Rentang Nilai Frekwensi Siklus 1 Frekwensi siklus 2 Kategori Prestasi Belajar Keterangan 1. 95 N 100 1 2 Sangat Baik Naik 1 2 85 N 94 11 13 Baik Naik 2 3 75 N 84 17 18 sedang Naik 1 4 65 N 74 4 2 Cukup baik Naik 2 5 55 N 64 2 1 kurang Naik 1 6 N ≤ 54 1 0 Kurang sekali Naik 1 Total: 36 36 Naik 8 frekwensi siklus 1 frekwensi siklus 2 Column1 0 5 10 15 20 < 55 55 - 64 65 - 74 75 - 84 85 - 94 < 55 frekwensi siklus 1 frekwensi siklus 2 Column1 frekwensi siklus 1 frekwensi siklus 2 Column1 0 5 10 15 20 < 55 55 - 64 65 - 74 75 - 84 85 - 94 < 55 frekwensi siklus 1 frekwensi siklus 2 Column1 Presentase ketuntasan individual diperoleh dari jumlah nilai yang diperoleh siswa dibagi nilai ideal.Siswa dianggap tuntas secara individu, jika skor yang dicapai minimal 75 ( Kriteria Ketuntasan Minimal atau KKM ) .Ketuntasan individu setelah siklus 1 adalah 29 dari 36 siswa atau 80,5 % dengan nilai rata-rata 80,4 .Sedangkan ketuntasan individu setelah siklus 2 adalah 35 dari 36 siswa atau 92 % dengan nilai rata-rata 83,3.Nilai rata-rata individu siklus 1 dan 2 adalah 81,9.Terjadi peningkatan rata-rata nilai yang diperoleh siswa pada pembelajaran materi garis dan sudut dari tahun pelajaran 2015-2016 yang sebesar 62,5 menjadi 81,9 atau 19,4 hasil pendekatan pembelajaran aktivitas pada tahun pelajaran 2016-2017.Peningkatan 19,4 atau 31 % dari 62,5 menurut Suharsimi (dalam Tim STKIP:2007) termasuk dalam taraf keberhasilan tindakan yang cukup baik. Presentase ketuntasan kelas/klasikal diperoleh dari jumlah siswa yang tuntas dibagi jumlah siswa. Ketuntasan klasikal dapat tercapai jika jumlah siswa yang tuntas mencapai Copyright © 2018, JMEN (Jurnal Math Educator Nusantara), Wahana Publikasi Karya Tulis Ilmiah di Bidang Pendidikan Matematika ISSN 2459-9735 (print), ISSN 2580-9210 (online) Nirwono, B., Meningkatkan Hasil Belajar Materi Garis dan... 76 minimal 85 % dari jumlah siswa keseluruhan di suatu kelas.Ketuntasan kelas rata-rata hasil penilaian proses siklus 1 dan 2 adalah 86,3 %. Terjadi peningkatan ketuntasan kelas pembelajaran materi garis dan sudut dari tahun pelajaran 2015-2016 yang sebesar 55,2% menjadi 86,3% atau 30,9% hasil pendekatan pembelajaran aktivitas pada tahun pelajaran 2016-2017.Menurut Suharsimi dalam Tim STKIP:2007) termasuk dalam taraf keberhasilan tindakan yang cukup baik. Siklus 1 Hasil penilaian proses menunjukan bahwa 8 dari 36 siswa mengalami kenaikan dari siklus 1 ke siklus 2 atau 22,2 %.Menurut Suharsimi (dalam Tim STKIP:2007)termasuk dalam taraf keberhasilan tindakan yangkurang baik. Tabel 2 : Rekapitulasi angket motivasi pengaruh dari Pendekatan Pembelajaran Berbasis Aktivitas Pada materi Garis Dan Sudut No Ang ket Kriteria Yang Diukur pendapat Sebelum siklus 1 Sesudah siklus 2 Keteran gan skor Ya Tdk Skor Ya Tdk Skor 1 Senang belajar 14 22 1,39 26 10 1,72 Naik 33% 2 Mudah memahami materi pelajaran 15 21 1,50 22 14 1,61 Naik 21% 3 Termotivasi untuk belajar 16 20 1,44 26 10 1,72 Naik 28% 4 Meningkatkan kemampuan berpikir tingkat tinggi 10 26 1,28 19 17 1,53 Naik 25% 5 Berani menyampaikan pendapat 12 24 1,33 25 11 1,81 Naik 48% Rata-rata 13,4 22,6 1,37 23,6 13,4 1,68 Naik 31% l 2 : Rekapitulasi angket motivasi pengaruh dari Pendekatan Pembelajaran Berbasis Aktivitas Pada materi Garis Dan Sudut Copyright © 2018, JMEN (Jurnal Math Educator Nusantara), Wahana Publikasi Karya Tulis Ilmiah di Bidang Pendidikan Matematika ISSN 2459-9735 (print), ISSN 2580-9210 (online) Copyright © 2018, JMEN (Jurnal Math Educator Nusantara), Wahana Publikasi Karya Tulis Ilmiah di Bidang Pendidikan Matematika ISSN 2459-9735 (print), ISSN 2580-9210 (online) 77 \| Jurnal Math Educator Nusantara, Vol. 4 No. 1, Mei 2018, pp. 70 - 78 sebelum siklus 1 sesudah siklus 2 Column1 0 20 40 60 80 100 120 140 160 180 200 sebelum siklus 1 sesudah siklus 2 Column1 sebelum siklus 1 sesudah siklus 2 Column1 Berdasarkan pada akhir kegiatan siklus 2 tersebut; peneliti meiakukan refleksi dari hasil kegiatan penelitian sebagai berikut: (1) terlihat ada peningkatan motivasi siswa dalam mengikuti kegiatan belajar mengajar dibandingkan dengan kegiatan belajar mengajar sebelumnya (menggunakan strategi tradisional), (2) beberapa siswa cepat dalam mempelajari materi yang disampaikan oleh guru, (3) beberapa siswa sudah ada keberanian dalam menyampaikan pendapat, dan (4) kegiatan diskusi sudah terkesan hidup dan berjalan, tidak lagi didominasi oleh siswa yang pandai. Berdasarkan data pengamatan dan obeservasi peneliti selama kegiatan penelitian tindakan, dapat diperoleh data aktivitas siswa dalam kegiatan belajar mengajar siswa bahwa aktivitas siswa dalam kegiatan belajar dengan pendekatan berbasis aktivitas menunjukkan ada peningkatan dibandingkan dengan strategi yang digunakan oleh guru sebelumnya. Hal ini dapat dilihat data yang didapatkan oleh peneliti dan hasil belajar siswa dalam kegiatan belajar mengajar, menunjukkan motivasi belajar siswa mulai meningkat. SIMPULAN Berdasarkan temuan dan pembahasan dapat ditarik kesimpulan sebagai berikut (1) Pendekatan aktivitas membuat siswa aktif dalam belajar. (2) Pendekatan Pembelajaran Berbasis Aktivitas dapat meningkatkan prestasi belajar siswa pada materi garis dan sudut Untuk memaksimalkan hasil belajar, pembelajaran tentang teorema geometri bagi siswa SMP sebaiknya mempertimbangkan cara pandang dan hasil penelitian ini Berdasarkan temuan dan pembahasan dapat ditarik kesimpulan sebagai berikut (1) Pendekatan aktivitas membuat siswa aktif dalam belajar. (2) Pendekatan Pembelajaran Berbasis Aktivitas dapat meningkatkan prestasi belajar siswa pada materi garis dan sudut Untuk memaksimalkan hasil belajar, pembelajaran tentang teorema geometri bagi siswa SMP sebaiknya mempertimbangkan cara pandang dan hasil penelitian ini. Copyright © 2018, JMEN (Jurnal Math Educator Nusantara), Wahana Publikasi Karya Tulis Ilmiah di Bidang Pendidikan Matematika ISSN 2459-9735 (print), ISSN 2580-9210 (online) Nirwono, B., Meningkatkan Hasil Belajar Materi Garis dan... 78 DAFTAR PUSTAKA Hamalik, O: 2001. Proses Belajar Mengajar. Jakarta : PT. Bumi Aksara Hamalik, O: 2002. Perencanaan Pengajaran Berdasarkan Pendekatan Sistem. Jakarta: PT. Bumi Aksara Lampiran Peraturan Menteri Pendidikan no 22 Tahun 2006. Tentang Standar Isi Lampiran Peraturan Menteri Pendidikan no 23 Tahun 2006. Tentang Standar Kompetensi Lulusan Lampiran Peraturan Menteri Pendidikan no 41 Tahun 2007. Tentang Standar Proses Materi Pelatihan Terintegrasi,Buku 1 :Jakarta.Depdiknas.Dirjen Dikdasmen,Direktorat pembinaan SMP. Materi Pelatihan Terintegrasi,Buku 2 :Jakarta.Depdiknas.Dirjen Dikdasmen,Direktorat pembinaan SMP. Materi Pelatihan Terintegrasi,Buku 3 :Jakarta.Depdiknas.Dirjen Dikdasmen,Direktorat pembinaan SMP. Suparno, P., Rohandi, R., Sukadi, G., Kartono, S: 2001. Reformasi Pendidikan Sebuah Rekomendasi Yogyakarta: Penerbit Kanisius. Tim STKIP PGRI Tulungagung : 2007.Makalah Penelitian Tindakan Kelas (Action Reseac).Seminar Menjadi Guru Profesional. Usman, Uzer, M : 2002. Menjadi Guru Profesional. Edisi Kedua. Cetakan ke empat belas. Bandung : PT Remaja Rosdakarya. Wilis,D,Ratna : 1988.Teori-teori Belajar.Jakarta.P2LPTK Zuriah, N.: 2003. Penelitian Tindakan dalam bidang Pendidikan dan Sosial. Edisi Pertama. Malang: Bayu Media Publishing. Copyright © 2018, JMEN (Jurnal Math Educator Nusantara), Wahana Publikasi Karya Tulis Ilmiah di Bidang Pendidikan Matematika ISSN 2459-9735 (print), ISSN 2580-9210 (online)
https://openalex.org/W2589870514	http://old.scielo.br/pdf/rbeaa/v21n2/1415-4366-rbeaa-21-02-0083.pdf	English	null	Urban solid waste in the production of Lafoensia pacari seedlings	Revista Brasileira de Engenharia Agrícola e Ambiental	2,017	cc-by	4,029	A B S T R A C Th Key words: sewage sludge urban waste biosolid This study aimed to verify the potential of urban solid wastes as substrate for production of seedlings of Lafoensia pacari. Five treatments were tested, four with solid wastes and one standard substrate, namely: sewage sludge from Alegria Wastewater Treatment Plant (WTP); sewage sludge from Ilha do Governador WTP; sewage sludge from Sarapuí WTP; domestic garbage compost (Fertlurb); and a commercial substrate made of biostabilized pine bark (standard substrate). The wastes received 20% (in volume) of shredded coconut fiber. At 105 days after sowing, the seedlings were evaluated for different quality parameters. Seedlings produced with Sarapuí WTP sewage sludge showed the best results in all the parameters, followed by seedlings produced with sewage sludge from Alegria and Ilha do Governador WTPs, which did not differ. Seedlings produced with domestic garbage compost showed satisfactory results, higher than the ones observed for seedlings produced with commercial substrate. The urban solid wastes with 20% of coconut fiber showed high potential and can be recommended for the composition of substrate in the production of Lafoensia pacari seedlings. Resíduos sólidos urbanos na produção de mudas de Lafoensia pacari Palavras-chave: lodo de esgoto lixo domiciliar biossólido Palavras-chave: lodo de esgoto lixo domiciliar biossólido Ref. 013-2016 – Received 3 Mar, 2016 • Accepted 18 Nov, 2016 • Published 5 Jan, 2017 Urban solid waste in the production of Lafoensia pacari seedlings Alan H. M. de Abreu1, Leticia B. Marzola2, Lucas A. de Melo3, Paulo S. dos S. Leles4, Elton L. S. Abel2 & Jorge M. Alonso1 1 Universidade Federal Rural do Rio de Janeiro/Programa de Pós Graduação em Ciências Ambientais e Florestai alanhenriquem@gmail.com (Corresponding author); j_makh@hotmail.com 1 Universidade Federal Rural do Rio de Janeiro/Programa de Pós Graduação em Ciências Ambientais e Florestais. Seropédica, RJ. E-mail: alanhenriquem@gmail.com (Corresponding author); j_makh@hotmail.com 2 Universidade Federal Rural do Rio de Janeiro. Seropédica, RJ. E-mail: engflorestalleticia@gmail.com; eltonabel@florestal.eng.br 3 Universidade Federal de Lavras/Departamento de Silvicultura. Lavras, MG. E-mail: lucas.amaral@dfc.ufla.br 4 Universidade Federal Rural do Rio de Janeiro/Departamento de Silvicultura. Seropédica, RJ. E-mail: pleles@ufrrj.br q @g ( p g ); j_ @ 2 Universidade Federal Rural do Rio de Janeiro. Seropédica, RJ. E-mail: engflorestalleticia@gmail.com; eltonabel@florestal.eng.br 3 Universidade Federal de Lavras/Departamento de Silvicultura. Lavras, MG. E-mail: lucas.amaral@dfc.ufla.br 4 Universidade Federal Rural do Rio de Janeiro/Departamento de Silvicultura. Seropédica, RJ. E-mail: pleles@ufrrj.br 2 Universidade Federal Rural do Rio de Janeiro. Seropédica, RJ. E-mail: engflorestalleticia@gmail.com; eltonabel@flore 3 Universidade Federal de Lavras/Departamento de Silvicultura. Lavras, MG. E-mail: lucas.amaral@dfc.ufla.br 4 Universidade Federal Rural do Rio de Janeiro/Departamento de Silvicultura. Seropédica, RJ. E-mail: pleles@ufrrj.br Introduction Currently, urban solid wastes constitute a large environ- mental liability, occupying large volumes in the landfills and decreasing the life span of these spaces. The recycling of these materials aiming at the formulation of substrates to produce forest seedlings could be a more sustainable way of disposal, besides guaranteeing the long-term supply of organic matter and low cost for nursery owners (Caldeira et al., 2012a). In this context, the domestic urban garbage and sewage sludge emerge as two of the main wastes to be recycled. The use of sewage sludge in the production of forest seedlings has been successfully tested and the positive result on the growth of the seedlings has been attributed to the high concentrations of nutrients and organic matter contained in this material (Faria et al., 2013; Rocha et al., 2013; Trazzi et al., 2014). The Fertlurb urban waste compost was provided by the Municipal Urban Cleaning Company of Rio de Janeiro (COMLURB). The Fertlurb is produced from the composting of organic resides present in the domestic garbage, from the city of Rio de Janeiro. These organic wastes are collected, sorted and composted until they are properly stabilized for use. It should be pointed out that, because of the large amount of plastic and glass fragments, the compost needed to be standardized before use by sieving through a 10 mm mesh. The domestic urban garbage, properly separated and composted, can also be used as substrate for various purposes (Nóbrega et al., 2008; Silva et al., 2014). The reuse of urban garbage in the formulation of substrates for seedlings production has high environmental value, because it returns to the productive cycle a material that, if treated as waste, has high cost of final destination (Zapata et al., 2005).f The tested treatments were: T1 – Sewage sludge from Alegria WTP; T2 – Sewage sludge from Ilha do Governador WTP; T3 – Sewage sludge from Sarapuí WTP; T4 - Fertlurb (urban garbage compost) and T5 – Commercial substrate based on stabilized pine bark residues (standard substrate used in forest nurseries). The five treatments were mixed with 20% shredded coconut fiber, to improve the porosity of the substrates. After formulation of the substrates, representative samples were collected for chemical analysis (Table 1). R E S U M O Objetivou-se, neste trabalho, verificar o potencial de resíduos sólidos urbanos como substratos na produção de mudas de Lafoensia pacari. Cinco tratamentos foram testados dentre os quais quatro resíduos urbanos e um substrato padrão, sendo: lodo de esgoto da Estação de Tratamento de Esgotos (ETE) Alegria; lodo de esgoto da ETE Ilha do Governador; lodo de esgoto da ETE Sarapuí; composto de lixo urbano (Fertlurb) e substrato comercial à base de casca de pinheiro bioestabilizada (substrato padrão). Foram acrescidos, aos resíduos, 20% em volume de fibra de coco triturada; aos105 dias após a semeadura avaliaram-se as variáveis referentes à qualidade de mudas; as mudas produzidas com substrato à base lodo de esgoto da ETE Sarapuí apresentaram os melhores resultados em todas as variáveis, seguidos pelas produzidas com os lodos das ETE Alegria e Ilha do Governador, que não diferiram entre si. As mudas produzidas com composto de lixo urbano apresentaram resultados satisfatórios e superiores ao observado nas mudas produzidas com substrato comercial. Os resíduos sólidos urbanos acrescidos de 20% de fibra de coco apresentaram potencial elevado podendo ser recomendados para a composição de substratos na produção de mudas de Lafoensia pacari. Ref. 013-2016 – Received 3 Mar, 2016 • Accepted 18 Nov, 2016 • Published 5 Jan, 2017 Ref. 013-2016 – Received 3 Mar, 2016 • Accepted 18 Nov, 2016 • Published 5 Jan, 2017 Alan H. M. de Abreu et al. Alan H. M. de Abreu et al. 84 used in the experiment differed regarding the dehydration method. The sludges of Alegria and Ilha do Governador WTPs remained approximately 90 days on drying beds under full sun, until reaching moisture close to 30%, while the sludge from the Sarapuí WTP was dehydrated through thermal drying, exposed to approximately 350 ºC for 30 min, coming out of the dryer with moisture close to 10%. After these processing steps and stabilization, representative samples were collected and subjected to the chemical and microbiological analyses required by the CONAMA Resolution No. 375/2006 (Brasil, 2006). Introduction The experimental design was completely randomized, with five treatments and four replicates, and each replicate consisted of 18 seedlings.hi In order to test the effectiveness in the use of these wastes, it is important to prioritize species that are used for various purposes, such as Lafoensia pacari St. Hill., also known as ‘dedaleiro’. This species is widely used in plantations of forest restoration and urban landscaping, due to its plasticity, small and medium size, and large white-yellowish flowers (Lorenzi, 2002; Carvalho, 2006). In this context, this study aimed to evaluate the potential of using urban solid waste in the composition of substrates for the production of Lafoensia pacari seedlings. The containers consisted of 280 cm3 tubes filled with the substrates corresponding to each treatment. After filling, the tubes were arranged on beds under full sun and only then sowing was performed. For protection against adverse weather, the trays were covered with a 50% shade screen, from sowing until the plants reached a height close to 5 cm, approximately at 30 days after sowing. Irrigation was performed using micro- sprinklers, three times a day. Since the aim was to evaluate the potential of the substrates, no complementary chemical fertilizations were performed. At 60 days after sowing, when the seedlings were approximately 15 cm high, they were spaced by changing the density to occupy 50% of the tray, thus decreasing the competition for light.t R. Bras. Eng. Agríc. Ambiental, v.21, n.2, p.83-87, 2017. Alan H. M. de Abreu et al. Material and Methods The experiment was carried out from October 2013 to January 2014, in the forest nursery of the Forest Institute located at the Federal Rural University of Rio de Janeiro (UFRRJ), in Seropédica, RJ, Brazil (22° 45’ 27” S; 43° 41’ 46” W; 30 m). The climate of the region, according to Köppen’s classification, is Aw (Brasil, 1980), tropical with summer rains. According to the data of the last 20 years of the weather station of the Agricultural Research Company of Rio de Janeiro (PESAGRO-RJ), the mean annual rainfall is 1245 mm, with driest period in June, July and August, and water excess in December, January and February. The mean time of insolation is 2527 h, the mean annual evaporation is 1576 mm and the relative air humidity is 69%. At 105 days after sowing, shoot height (H) and collar diameter (D) were measured. After data compilation, the six central seedlings of each plot were selected, totaling 24 Table 1. Total contents (%) of macronutrients and organic matter present in the substrates formulated with urban solid wastes and in commercial substrate The sewage sludges used in the experiment were provided by the Water and Sewerage State Company of Rio de Janeiro (CEDAE) and came from three wastewater treatment plants (WTP): Alegria WTP, Ilha do Governador WTP and Sarapuí WTP, which received only sewage of domestic origin. The three stations have activated sludge secondary treatment systems and compaction of the secondary sludge, through centrifuges. After the dewatering in centrifuges, the sludges 1Sulfuric digestion - Kjeldhal distillation; 2Mehlich extractor; 31 mol L-1 KCl extractor; 4Walkley- Black method; OM - Organic matter Substrates 1N 2P 2K 3Ca 3Mg 4MO Alegria WTP 3.88 0.62 0.52 0.24 0.02 54.4 Ilha WTP 3.34 1.12 0.21 0.69 0.03 57.9 Sarapuí WTP 4.23 1.73 0.18 0.88 0.02 51.6 Fertlurb 0.81 0.53 0.12 0.69 0.38 33.0 Commercial substrate 0.49 0.17 0.16 1.49 0.58 31.0 R. Bras. Eng. Agríc. Ambiental, v.21, n.2, p.83-87, 2017. Urban solid waste in the production of Lafoensia pacari seedlings 85 seedlings per treatment, for the evaluation of shoot dry matter (SDM), root dry matter (RDM) and total dry matter (TDM). The obtained data were used to calculate the variables height/diameter ratio (H/D) and shoot dry matter/root dry matter ratio (SDM/RDM) (Carneiro, 1995). Material and Methods In addition, the Dickson quality index was also calculated, which considers the calculations, the robustness (H/D ratio) and the equilibrium of biomass distribution in the seedling (TDM and SDM/RDM ratio), considering the results of various important variables used for quality evaluation (Dickson et al., 1960). The higher the value of DQI, the better the quality of the seedlings in the lot, theoretically.h concentration of nutrients in this waste, especially N and P (Table 1). These nutrients are highly required in the initial growth stages of the seedlings, with important role in the initial impulse of shoot growth, which justifies the higher means in height and diameter in the treatments with higher contents of these nutrients (Faria et al., 2013; Trazzi et al., 2014). The higher growth of Lafoensia pacari seedlings observed in the sludge from the Sarapuí WTP is justified, because it is the only one, among the evaluated sludges, subjected to thermal drying. Carvalho et al. (2015) concluded that, for activated sludges, as those used in the present study, thermal drying preserves better the nutrients and maintains biodegradable fractions of the carbon, and this process is more indicated for the production of sludge to be used as organic fertilizer. These authors also mention that, in longer drying processes, such as those applied to the sludges of Alegria and Ilha do Governador WTPs, which use drying beds, besides a probable leaching of nutrients, the presence of moisture and oxygen allows microorganisms to consume part of the C present in the material, especially the soluble C, which tends to be the most readily available. The sludge generated this way would be the most indicated to be used as soil conditioner, since its organic matter is more stabilized. The data of all evaluated characteristics and ratios were subjected to analysis of variance and, when significant effects were observed, to Tukey test (p ≤ 0.05). The statistical analyses were performed using the software ‘Sistema de Análise Estatística e Genética’ (SAEG, 2007). Results and Discussion The seedlings produced in the treatment formulated with sludge from the Sarapuí WTP showed significantly higher results in almost all analyzed variables, followed by the seedlings of the treatments with sludge from Alegria and Ilha do Governador WTPs, which did not differ statistically (Table 2). In the sequence, the treatment with Fertlurb urban garbage compost showed means higher than those of the commercial substrate, but lower than those of treatments with sewage sludge. As an intrinsic feature of Lafoensia pacari, its height growth is favored detriment of diameter growth, which explains the relatively high values of the H/D ratio, in all treatments. According to José et al. (2005), seedlings with higher H/D ratio may have difficulty to stand upright after planting; thus, values below 10 are recommended. However, Delarmelina et al. (2013) call the attention for the need of more studies to determine the classes of optimal values of H/D ratio for species of the Brazilian flora, because this index is highly variable depending on the species.h Based on the growth variables of the seedlings, it can be inferred that the sewage sludge positively influenced the growth of Lafoensia pacari seedlings. Various authors have attributed the higher growth of seedlings produced in substrates containing sewage sludge to the high contents of nutrients and organic matter found in this waste (Caldeira et al., 2012b; Delarmelina et al., 2013; Gomes et al., 2013; Rocha et al., 2013). The sludge from the Sarapuí WTP also showed the best values of dry matter variables, which may be related to the greater concentration of nutrients in this sludge, in which the greater supply of N and P also led to greater increment in SDM, RDM and consequently TDM, parameters highly dependent on the nutrition provided by the substrate. Santos et al. (2008), in study with seven native arboreal species, observed that, as the P supply increased, there were increments in shoot and root dry matter production for the studied species. The same can be observed in the present experiment, in which the treatments with sewage sludge, which contain higher concentration of P, also showed higher biomass production. Trazzi et al. (2014), analyzing the potential of sewage sludge as component of the substrate for the production of Tectona grandis seedlings, tested different proportions of sludge, coconut fiber and carbonized rice husk. R. Bras. Eng. Agríc. Ambiental, v.21, n.2, p.83-87, 2017. Literature Cited and may be considered as a strategy of the plant, which invests in root growth instead of shoot growth to absorb as much as nutrients as possible under that condition. The same behavior was observed in this experiment, in which the lowest values of SDM/RDM were found in the treatment with commercial substrate. Brasil. Ministério de Minas e Energia: Departamento Nacional da Produção Mineral. Projeto RADAM Brasil. Folhas sc. 21. Juremo: Geomorfologia, pedologias, vegetação e uso potencial da terra. Rio de Janeiro: MME/DNPM, 1980. 420p. Rio de Janeiro: MME/DNPM, 1980. 420p. Confirming the results of the other quality parameters, previously presented, the highest mean of DQI was also found in seedlings produced with substrate formulated using sludge from the Sarapuí WTP, followed by the treatments with sludge from Alegria and Ilha do Governador WTPs, which did not differ statistically and showed means higher than those of the treatment with urban garbage compost. High DQI values are achieved when the seedlings have low H/D ratio (seedlings without etiolation), balanced values of SDM/RDM (seedlings with root system capable of supplying the shoots) and high value of total organic matter (biomass). On the other hand, the lowest means of DQI were found in the treatment with commercial substrate. Brasil. Ministério do Meio Ambiente. Conselho Nacional do Meio Ambiente. Resolução - CONAMA no 375/2006. Define critérios e procedimentos para o uso agrícola de lodos de esgoto gerados em estações de tratamento de esgoto sanitário e seus produtos derivados. Diário Oficial da República Federativa do Brasil. Brasília: CONAMA, n. 167, 2006. p.141-146. Caldeira, M. V. W.; Delarmelina, W. M.; Lube, S. G.; Gomes, D. R.; Gonçalves, E. O; Alves, A. F. Biossólido na composição de substrato para a produção de mudas de Tectona grandis. Floresta, v.42, p.77-84, 2012a. http://dx.doi.org/10.5380/rf.v42i1.26302 v.42, p.77-84, 2012a. http://dx.doi.org/10.5380/rf.v42i1.263 Caldeira, M. V. W.; Delarmelina, W. M.; Peroni, L.; Gonçalves, E. O.; Silva, A. G. Lodo de esgoto e vermiculita na produção de mudas de eucalipto. Pesquisa Agropecuária Tropical, v.43, p.155-163, 2013. http://dx.doi.org/10.1590/S1983-40632013000200002 These results corroborate the efficiency of the substrates produced with urban solid waste in the quality of Lafoensia pacari seedlings. López et al. (2008), analyzing the use of composted urban waste as substrate, also found results that confirm the efficiency of both sewage sludge and urban garbage compost in the growth of forest seedlings. Caldeira, M. V. W.; Gomes, D. R.; Gonçalves, E. O.; Delarmelina, W. M.; Sperandio, H. Literature Cited V.; Trazzi, P. A. Biossólido como substrato para produção de mudas de Toona ciliata var. australis. Revista Árvore, v.36, p.1009-1017, 2012b. http://dx.doi.org/10.1590/ S0100-67622012000600002 The worst results were obtained in the treatment with commercial substrate, probably due to the lack of complementary fertilization, which is essential when this type of substrate is used. On the other hand, the treatments with sewage sludge and urban garbage compost formed quality seedlings at 105 days after sowing without the need for complementary fertilization. Caldeira et al. (2012a) analyzed the potential of sewage sludge to produce seedlings of Tectona grandis and concluded that this is a viable alternative, because of the good results obtained, saving of fertilizers and environmental benefit.h Carneiro, J. G. de A. Produção e controle de qualidade de mudas florestais. Curitiba: UFPR/FUPEF, 1995. 451p. Carvalho, C. S.; Ribeirinho, V. S.; Andrade, C. A.; Grutzmacher, P.; Pires, A. M. M. Composição química da matéria orgânica de lodos de esgoto. Revista Brasileira de Ciências Agrárias, v.10, p.413-419, 2015. http://dx.doi.org/10.5039/agraria.v10i3a5174 Carvalho, P. E. R. Espécies arbóreas brasileiras. Brasília: Embrapa Informação Tecnológica; Colombo: Embrapa Florestas, 2006. 1039p. Delarmelina, W. M.; Caldeira, M. V. W.; Faria, J. C. T.; Gonçalves, E. O. Uso de lodo de esgoto e resíduos orgânicos no crescimento de mudas de Sesbania virgata (Cav.) Pers. Revista Agroambiente, v.7, p.184-192, 2013. http://dx.doi.org/10.18227/1982-8470ragro. v7i2.888 Therefore, it can be inferred that the use of these urban solid wastes as substrate emerges as an alternative to simplify the production of seedlings in tubes, decreasing the dependence of one of the most crucial and emblematic phases of this production system, which is the need to replace the nutrients in the substrate through basal and/or top-dressing fertilizations. Dickson, A.; Leaf, A. L.; Hosner, J. F. Quality appraisal of white spruce and white pineseedling stock in nurseries. Forest Chronicle, v.36, p.10-13, 1960. http://dx.doi.org/10.5558/tfc36010-1 Faria, J. C. T.; Caldeira, M. V. W.; Delarmelina, W. M.; Lacerda, L. C.; Gonçalves, E. O. Substratos a base de lodo de esgoto na produção de mudas de Senna alata. Comunicata Scientiae, v.4, p.342-351, 2013. Results and Discussion These authors found results that corroborate the positive contribution of the sludge to the growth in height and diameter of forest seedlings and indicated proportions of 80% of sewage sludge and 20% of coconut fiber or 60% of sewage sludge and 40% of carbonized rice husk for the production of teak seedlings. It is possible to observe that the lowest means of SDM/ RDM ratio were found in the treatment with commercial substrate (Table 2). According to Caldeira et al. (2013), the SDM/RDM ratio is lower in environments of lower fertility In general, the sewage sludge showed the best results of height and diameter, in comparison to the other treatments, and the Sarapuí WTP stood out. The best results found in the treatments with sewage sludge are probably due to the higher H – Shoot height; D – Collar diameter; H/D – Height/diameter ratio; SDM – Shoot dry matter; RDM – Root dry matter; TDM –Total dry matter; SDM/RDM – Shoot dry matter/root dry matter ratio; DQI – Dickson quality index; Means followed by the same letter in the column do not differ statistically by Tukey test (p ≥ 0.95) Treatment H (cm) D (mm) H/D SDM RDM TDM SDM/RDM DQI g plant-1 Alegria WTP 62.1 b 6.76 b 09.30 ab 07.65 b 3.12 b 10.78 b 2.51 ab 0.93 b Ilha WTP 66.5 b 6.46 b 10.35 aa 08.59 b 3.10 b 11.68 b 2.78 aa 0.89 b Sarapuí WTP 72.1 a 8.99 a 08.02 ba 10.67 a 4.45 a 15.13 a 2.40 ab 1.45 a Fertlurb 44.0 c 4.53 c 09.77 ab 03.56 c 1.70 c 05.26 c 2.10 ba 0.45 c Commercial substrate 11.2 d 1.17 d 09.53 ab 00.66 d 0.65 d 01.31 d 1.01 ca 0.12 d Table 2. Quality variables for Lafoensia pacari seedlings produced with different urban wastes, at 105 days after sowing R. Bras. Eng. Agríc. Ambiental, v.21, n.2, p.83-87, 2017. Alan H. M. de Abreu et al. 86 Conclusions 1. The urban solid wastes mixed with 20% of coconut fiber showed high potential and can be recommended for the composition of substrates in the production of Lafoensia pacari seedlings. Gomes, D. R.; Caldeira, M. V. W.; Gonçalves, E. O.; Delarmelina, W. M.; Trazzi, P. A. Lodo de esgoto como substrato para produção de mudas de Tectona grandis L. Cerne, v.19, p.123-131, 2013. http:// dx.doi.org/10.1590/S0104-77602013000100015 2. The sewage sludge from the Sarapuí WTP showed the best results, followed by the sludges from Alegria and Ilha do Governador WTPs. The Fertlurb showed satisfactory results, higher than those of the commercial substrate. José, A. C.; Davide, A. C.; Oliveira, S. L. Produção de mudas de aroeira (Schinus terebynthifolius Raddi) para recuperação de áreas degradadas pela mineração de bauxita. Cerne, v.11, p.187- 196, 2005. R. Bras. Eng. Agríc. Ambiental, v.21, n.2, p.83-87, 2017. R. Bras. Eng. Agríc. Ambiental, v.21, n.2, p.83-87, 2017. Acknowledgments López, R.; Cabrera, F.; Madejon, E.; Sancho, F.; Alvarez, J. M. Urban composts as an alternative for peat in forestry nursery growing media. Dynamic Soil, v.2, p.60-66, 2008. The authors thank the Water and Sewerage State Company of Rio de Janeiro (CEDAE) for the support to the research. R. Bras. Eng. Agríc. Ambiental, v.21, n.2, p.83-87, 2017. Urban solid waste in the production of Lafoensia pacari seedlings 87 Lorenzi, H. Árvores brasileiras: Manual de identificação e cultivo de plantas arbóreas nativas do Brasil. v.2. 2.ed. Nova Odessa: Platarum, 2002. 368p. Santos, J. Z. L.; Resende, A. V.; Corte, E. F. Crescimento, acúmulo de fósforo e frações fosfatadas em mudas de sete espécies arbóreas nativas. Revista Árvore, v.32, p.799-807, 2008. http://dx.doi. org/10.1590/S0100-67622008000500003 Nóbrega, R. S. A.; Paula, A. M.; Vilas Boas, R. C.; Nóbrega, J. C. A.; Moreira, F. M. S. Parâmetros morfológicos de mudas de Sesbania virgata (Caz.) Pers e de Anadenanthera peregrina (L.) cultivadas em substrato fertilizado com composto de lixo urbano. Revista Árvore, v.32, p.597-607, 2008. http://dx.doi.org/10.1590/S0100- 67622008000300020 Silva, R. F.; Eitelwein, M. T.; Cherubin, M. R.; Fabbris, C.; Weirich, S.; Pinheiro, R. R. Produção de mudas de Eucalyptus grandis em substratos orgânicos alternativos. Ciência Florestal, v.24, p.609- 619 2014 htt //d d i /10 5902/1980509815745 619, 2014. http://dx.doi.org/10.5902/1980509815745 Trazzi, P. A.; Caldeira, M. V. W.; Reis, E. F.; Silva, A. G. Produção de mudas de Tectona grandis em substratos formulados com biossólido. Cerne, v.20, p.293-302, 2014. http://dx.doi.org/10.15 90/01047760.201420021134 Rocha, J. H. T.; Backes, C.; Diogo, F. A.; Pascotto, C. B.; Borelli, K. Composto de lodo de esgoto como substrato para mudas de eucalipto. Pesquisa Florestal Brasileira, v.33, p.27-36, 2013. http:// dx.doi.org/10.4336/2013.pfb.33.73.331 Zapata, N.; Guerrero, F.; Polo, A. Evaluación de corteza de pino y residuos urbanos como componentes de sustratos de cultivo. Agricultura Técnica, v.65, p.387-400, 2005. http://dx.doi. org/10.4067/S0365-28072005000400004 b SAEG - Sistema para Análises Estatísticas, versão 9.1. Viçosa: UFV, 2007.
https://openalex.org/W2969976636	https://www.nature.com/articles/s41467-019-11695-8.pdf	English	null	High-throughput screen reveals sRNAs regulating crRNA biogenesis by targeting CRISPR leader to repress Rho termination	Nature communications	2,019	cc-by	14,942	1 State Key Laboratory of Trauma, Burns and Combined Injury, Institute of Surgery Research, Daping Hospital, Army Medical University, Chongqing, China. 2 Department of Biomedical Sciences, School of Medicine and Health Sciences, University of North Dakota, Grand Forks, ND, USA. 3 Department of Pediatrics, Ruijin Hospital afﬁliated to Shanghai Jiao Tong University School of Medicine Shanghai, Shanghai, China. 4 Key Laboratory of Biotechnology for Medicinal Plants of Jiangsu Province, Jiangsu Normal University, Xuzhou, Jiangsu, China. 5 Southwestern Medical University, Pulmonary and Allergy Institute, Afﬁliated Hospital, Luzhou, China. 6 Department of Microbiology and Immunology, Dalhousie University, Halifax, NS, Canada. 7 Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai, China. Correspondence and requests for materials should be addressed to L.L. (email: llan@simm.ac.cn) or to J.J. (email: hellojjx@126.com) or to M.W. (email: min.wu@und.edu) ARTICLE NATURE COMMUNICATIONS \| (2019) 10:3728 \| https://doi.org/10.1038/s41467-019-11695-8 \| www.nature.com/naturecommunications ARTICLE ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 C RISPR-Cas systems endow prokaryotes with adaptive and heritable immunity1–3, which employ RNA-guided nucleases for recognizing and destroying invading DNA or RNA4–6. CRISPR-Cas systems function through three phases: spacer acquisition, expression of CRISPR RNA (crRNA), and interference2,7. At the spacer acquisition stage, foreign nucleic acids from an invader would be incorporated into CRISPR loci as new spacers, forming expanded archives of past infections. Following the spacer acquisition is the crRNA biogenesis phase, in which CRISPR loci are transcribed to yield precursor crRNA (pre-crRNA) and are then cleaved by Cas proteins into mature crRNAs. In the inter- ference and ﬁnal phase, crRNA-guided Cas proteins cleave foreign DNA or RNA and mediate their degradation. Although extensive biological and ecological studies have built a framework about the structures and functions of the CRISPR-Cas system1,8,9, we have just begun to understand the fascinating prokaryote immunity10. Importantly, the CRISPR leader sequence exhibits the speciﬁcity to constitute an ideal target substrate for spacer integration during the adaptation phase11, but little is known regarding whether the leader is critical for producing pre-crRNA transcripts during the crRNA biogenesis phase. paired RNA molecules by catalyzing ATP-dependent formation of a 3ʹ→5ʹ phosphodiester bond on single-stranded RNAs, which offers a means for investigating the interaction of bacterial sRNAs and RNA molecules in vivo22. We used T4 RNA ligase 1 to generate sRNA-RNA chimaera to selectively probe the inter- actome for interactions between bacterial sRNAs and CRISPR- Cas system (Fig. 1a). We investigated the effect of T4 RNA ligase 1 expression on P. aeruginosa PA14 strain throughout the growth period (Fig. 1b), which showed a decline in viability for 1 h after IPTG treatment. Therefore, the inducible expression of T4 RNA ligase 1 was maintained up to 1 h for each experiment. g p p The P. aeruginosa PA14 I-F CRISPR-Cas comprises Cas1, Cas3, Csy1–4 complex ﬂanked by two CRISPR loci (Supplemen- tary Fig. 1a). To identify potential sRNAs that target leaders in CRISPR loci, we used the pKH6 vector22 to create an arabinose- inducible vector (pKH6-CRISPR1 leader and pKH6-CRISPR2 leader) and introduced the vector into PA14 containing pKH- t4rnl1, respectively. After IPTG and arabinose treatment, we used a library of 274 P. aeruginosa endogenous sRNAs to detect the ligated chimeric sRNA-CRISPR leader using sRNA-speciﬁc primers and CRISPR leader-speciﬁc primers as described in Fig. 1a. ARTICLE We observed 9 and 25 sRNA-CRISPR leader chimeras for CRISPR1 and CRISPR2 leaders, respectively (Fig. 1c, d, Supplementary Fig. 1b, and Supplementary data 1). Computa- tional analysis using the online IntaRNA tool also predicts interaction between CRISPR loci and sRNAs (Fig. 1e). The difference between Fig. 1d, e is possibly due to the linking between CRISPR leader and sRNAs through 5ʹ monophosphates to 3ʹ hydroxyl groups by T4 RNA ligase 1, but the majority of P. aeruginosa sRNA molecules are transcript products containing 5ʹ triphosphoryl termini. In order to investigate and characterize whether any of these 34 sRNAs interact with and/or regulate CRISPR loci, we constructed each of the sRNA over-expressing plasmids in combination with CRISPR1-lacZ or CRISPR2-lacZ fusion plasmid, and transformed them into PA14 to monitor lacZ activity. Of the 35 sRNAs tested, one sRNA pant391 repressed CRISPR2-lacZ expression by more than twofold, while sRNAs pant463 and PhrS increased CRISPR2-lacZ expression (Fig. 1f). Of note, PhrS had the strongest positive effect on the level of CRISPR locus, which was further investigated (Fig. 1f). Regulatory RNAs are an essential group of molecules that facilitate various aspects in gene expression, such as transcription, RNA processing or stabilization, and translation12,13. Small reg- ulatory RNAs (sRNAs), the primary group of regulatory RNA in bacteria (50- to 400 bp), are a major regulator of numerous metabolic and stress responses in bacteria14–16. In particular, prompt responses to various stimuli are shown to be controlled by sRNAs17,18. However, it is unknown whether sRNAs can modulate CRISPR-Cas adaptive immunity by inﬂuencing the transcriptional activity. Prokaryotes use Rho-dependent termination mechanisms for RNA polymerase (RNAP) recycling in most species of bacteria kingdom19. Rho along with its cofactor NusG bind to the tran- scription terminator pause sites that function as an attenuator20,21. Rho moves along the nascent RNA molecules that enable it to function at RNA polymerase, resulting in the dissociation of RNA polymerase complex and termination of transcription. y Here, we used a combination of approaches to search for candidate sRNAs that may regulate CRISPR-Cas function. Based on the data from GRIL-Seq (global small non-coding RNA target identiﬁcation by ligation/sequencing)22, the studies derived from genome-wide identiﬁcation of sRNA in Pseudomonas aeruginosa23,24 and our initial analyses, we constructed high- throughput library encoding 274 sRNAs to ligate to the CRISPR leader via T4 RNA ligase-catalyzed linking assay. ARTICLE The screened candidate sRNAs that target CRISPR leaders were characterized by functional assays and potential binding domains in the targets were predicted using the IntaRNA computing tool and assessed by biochemical assays. PhrS is shown with the most signiﬁcant interference with Rho-mediated termination by interacting with Type I-F CRISPR leaders, resulting in transcriptional activation of CRISPR loci and then stimulation of CRISPR-Cas adaptive immunity against bacteriophage invasion. As PhrS also shows the similar functions in type I-C/-E CRISPR-Cas systems, demon- strating pervasiveness of sRNA-mediated control of CRISPR-Cas activities. Our data reveal the function of CRISPR leaders, which not only contain a conserved integration host factor to create the ideal target substrate for Cas1-Cas2 to spacer acquisition11,25,26, but also facilitates crRNA biogenesis by controlling CRISPR loci transcription. g ( g ) To detect speciﬁc ligation of candidate targets of PhrS with CRISPR loci, we performed reverse transcription-polymerase chain reaction (RT-PCR) to analyze the ligated products as described in Fig. 1a using phrS-speciﬁc primer and CRISPR locus- speciﬁc primer, followed by induction expression of RNA for up to 20 min in the presence of T4 RNA ligase or an inactive T4 RNA ligase (t4K99N). We noticed that the amplicons of PhrS- CRISPR2 leader chimeras were induced to facilitate the expres- sion of PhrS for up to 20 min, but abrogated by an inactive T4 RNA ligase (Fig. 1g). Sequencing analysis of the amplicons conﬁrms that the junction sequences are the PhrS-CRISPR2 leader chimeras (Fig. 1h). These data indicate that PhrS is a candidate sRNA that interacts with type I-F CRISPR leaders of PA14 strain. PhrS promotes CRISPR2 locus transcription and interference. To investigate the inﬂuence of PhrS on CRISPR-Cas functionality, we evaluated the expression pattern of cas operon or CRISPR loci in the PA14 phrS deletion strain (ΔphrS) vs. the wild-type (WT) strain. Only CRISPR2 locus, not cas operon and CRISPR1 locus, exhibited lower expression in PA14 ΔphrS than WT throughout the survey growth period, but restored expression levels close to the WT upon complementing PA14 ΔphrS (Fig. 2a). We then measured the transformation efﬁciency of CRISPR-Cas on elim- inating CRISPR-targeted plasmids that contained protospacers in High-throughput screen reveals sRNAs regulating crRNA biogenesis by targeting CRISPR leader to repress Rho termination Ping Lin 1,2, Qinqin Pu2, Qun Wu 2,3, Chuanmin Zhou2, Biao Wang2, Jacob Schettler 2, Zhihan Wang 2, Shugang Qin2, Pan Gao 2, Rongpeng Li4, Guoping Li5, Zhenyu Cheng 6, Lefu Lan 7, Jianxin Jiang1 & Min Wu 2 Discovery of CRISPR-Cas systems is one of paramount importance in the ﬁeld of micro- biology. Currently, how CRISPR-Cas systems are ﬁnely regulated remains to be deﬁned. Here we use small regulatory RNA (sRNA) library to screen sRNAs targeting type I-F CRISPR-Cas system through proximity ligation by T4 RNA ligase and ﬁnd 34 sRNAs linking to CRISPR loci. Among 34 sRNAs for potential regulators of CRISPR, sRNA pant463 and PhrS enhance CRISPR loci transcription, while pant391 represses their transcription. We identify PhrS as a regulator of CRISPR-Cas by binding CRISPR leaders to suppress Rho-dependent transcription termination. PhrS-mediated anti-termination facilitates CRISPR locus transcription to gen- erate CRISPR RNA (crRNA) and subsequently promotes CRISPR-Cas adaptive immunity against bacteriophage invasion. Furthermore, this also exists in type I-C/-E CRISPR-Cas, suggesting general regulatory mechanisms in bacteria kingdom. Our ﬁndings identify sRNAs as important regulators of CRISPR-Cas, extending roles of sRNAs in controlling bacterial physiology by promoting CRISPR-Cas adaptation priming. 1 TURE COMMUNICATIONS \| (2019) 10:3728 \| https://doi.org/10.1038 NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 Also shown is reverse transcription-polymerase chain reaction (RT-PCR)-based strategy for determining chimeras of CRISPR leader with sRNA. b T4 RNA ligase or its inactive mutation in t4rnl1 gene with lysine (K) to asparagine (N) affects P. aeruginosa cell growth. c Screening of 274 P. aeruginosa sRNAs library (239 intergenic sRNAs candidate and 35 annotated sRNAs) linking to CRISPR leader by T4 RNA ligase. Pink represents sRNA-containing chimeras; green represents non-target sRNA chimeras. d Detection of chimeras of P. aeruginosa 35 annotated sRNAs linking to CRISPR leader sequences by T4 RNA ligase in vivo, relative to Supplementary Fig. 1b. Pink represents sRNA-containing chimeras; green represents non-target sRNA chimeras. e IntaRNA prediction of P. aeruginosa annotated sRNAs interactions with CRISPR leader. f Overexpression sRNA to screen candidate sRNAs on regulation of CRISPR1-lacZ and CRISPR2-lacZ fusion. g Amplicons were detected for PhrS-CRISPR2 leader chimeras. Primer for targets PhrS with CRISPR leader (as shown in a) was carried out for PCR step. PCR production for PhrS and housekeeping gene (pheS) was carried out to ensure the genes expression in all samples. h Sequencing reads corresponding to PhrS chimeras with CRISPR2 leader by TA-clone sequencing. Results are shown with mean ± SEM from three independent experiments. *P < 0.01, P < 0.05, one-way ANOVA plus Tukey test CRISPR1 (denoted CR1-sp1) or CRISPR2 (denoted CR2-sp1) in PA14 ΔphrS (Supplementary Fig. 1a). Strikingly, mutation of phrS had no effect on CRISPR1-dependent CRISPR interference (Fig. 2b, left), but resulted in equal transformation frequencies of PA14 ΔTCR lacking cas genes when CRISPR2-targeted DNA was used (Fig. 2b, right), reﬂecting a lack of CRISPR2 interference and immunity functionality that is regulated by PhrS. We also observed that CRISPR-sensitive phage JBD25, which targets a spacer in CRISPR1 locus, failed to replicate in PA14 WT, ΔphrS and ΔphrS/p-phrS (Fig. 2c and Supplementary Fig. 1a). Con- versely, CRISPR-sensitive JBD18, which targets a spacer in CRISPR2 locus, was able to replicate in PA14 ΔphrS, but failed to replicate in WT and ΔphrS/p-phrS (Fig. 2c). Taken together, our data demonstrate that PhrS modulates efﬁciency of CRISPR2 interference, hence controlling its functionality. Based on these ﬁndings, we reasoned that transcriptional changes of CRISPR2 locus may be associated with PhrS. Northern blot analysis supported this premise—that PhrS is required for the synthesis of crRNA in CRISPR2 locus (Fig. 2d). NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 ARTICLE a b IPTG treatment Bacterial count (log cfu/ml) 0 5 10 15 1 2 3 4 5 6 7 8 Time (h) PA14 WT PA14 t4rhl1 PA14 t4K99N Binding No binding c CRISPR1 leader CRISPR2 leader 35 annotated sRNAs 239 intergenic sRNAs candidates AmiL CrcZ EsrA ffs nalA NrsZ P5 P7 P8 P9 P13 P14 P15 P18 P26 P27 P30 P32 P34 P35 PesA PhrS Pail PA4406.1 PA5316.1 PhrX PhrY PrrF1 PrrF2 RgsA RnpB RsmW RsmY RsmZ ssrS AmiL CrcZ EsrA ffs nalA NrsZ P5 P7 P8 P9 P13 P14 P15 P18 P26 P27 P30 P32 P34 P35 PesA PhrS Pail PA4406.1 PA5316.1 PhrX PhrY PrrF1 PrrF2 RgsA RnpB RsmW RsmY RsmZ ssrS d e CRISPR1 leader CRISPR2 leader CRISPR1 leader CRISPR2 leader 0 5 10 20 0 5 10 20 0 5 10 20 0 5 10 20 g min, L-ara t4rhl t4K99N t4rhl t4K99N CRISPR1 leader CRISPR2 leader PhrS target PhrS CRISPR2 leader PhrS CRISPR2 leader h f Fold change (log2) ** * CRISPR2-lacZ pant391 pant463 PhrS pant312 pant333 pant342 pant349 pant431 pant35 pant81 pant119 pant187 pant167 pant155 pant184 pant245 pant250 pant439 pant445 pant452 P35 pant509 pant457 P27 Rsmw Empty vector pant391 pant463 PhrS pheS CRISPR2 leader –2 –1 0 1 2 –3 250 50 bp 250 50 bp 200 500 300 CRISPR2 leader CRISPR1 leader RT-PCR cDNA synthesis Total RNA isolation Primer: sRNA (specific) Random primer CRISPR2 leader Total 274 sRNAs library pKH6 CRISPR2 leader sRNA target pKH13-T4rnl: T4 RNA ligase Fold change (log2) CRISPR1-lacZ pant92 pant96 pant103 pant459 pant390 pant185 pant457 P27 Rsmw Empty vector –1.0 –0.5 0.0 0.5 1.0 Fig. 1 T4 RNA ligase-catalyzed ligation of sRNAs to P. aeruginosa CRISPR loci. a Schematic of the formation of sRNAs chimeras with CRISPR leader by T RNA ligase. Two RNA molecules were linked to form pKH6-CRISPR leader plasmid for expressing CRISPR leader and pKH13-t4rhl1 for expressing T4 RN ligase. Also shown is reverse transcription-polymerase chain reaction (RT-PCR)-based strategy for determining chimeras of CRISPR leader with sRNA b T4 RNA ligase or its inactive mutation in t4rnl1 gene with lysine (K) to asparagine (N) affects P. aeruginosa cell growth. c Screening of 274 P. aerugino sRNAs library (239 intergenic sRNAs candidate and 35 annotated sRNAs) linking to CRISPR leader by T4 RNA ligase. Pink represents sRNA-containin chimeras; green represents non-target sRNA chimeras. d Detection of chimeras of P. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 PCR production for PhrS and housekeeping gene (pheS) was carried out to ensure the genes expression in b IPTG treatment Bacterial count (log cfu/ml) 0 5 10 15 1 2 3 4 5 6 7 8 Time (h) PA14 WT PA14 t4rhl1 PA14 t4K99N AmiL CrcZ EsrA ffs lA AmiL CrcZ EsrA ffs lA d e CRISPR1 leader CRISPR2 leader CRISPR1 leader CRISPR2 leader t391 RISPR2 leader RISPR1 leader RT-PCR cDNA synthesis Total RNA isolation Primer: sRNA (specific) Random primer CRISPR2 leader Total 274 sRNAs library pKH6 CRISPR2 leader sRNA target pKH13-T4rnl: T4 RNA ligase b IPTG treatment Bacterial count (log cfu/ml) 0 5 10 15 1 2 3 4 5 6 7 8 Time (h) PA14 WT PA14 t4rhl1 PA14 t4K99N d e CR CR CR CR a CRISPR2 leader CRISPR1 leader RT-PCR cDNA synthesis Total RNA isolation Primer: sRNA (specific) Random primer CRISPR2 leader To sRN pKH6 CRISPR2 leader sRNA target pKH13-T4rnl: T4 RNA ligase b a e d Binding No binding c CRISPR1 leader CRISPR2 leader 35 annotated sRNAs 239 intergenic sRNAs candidates f Fold change (log2) * CRISPR2-lacZ pant391 pant463 PhrS pant312 pant333 pant342 pant349 pant431 pant35 pant81 pant119 pant187 pant167 pant155 pant184 pant245 pant250 pant439 pant445 pant452 P35 pant509 pant457 P27 Rsmw Empty vector pant391 pant463 –2 –1 0 1 2 –3 Primer: sRNA (specific) Fold change (log2) CRISPR1-lacZ pant92 pant96 pant103 pant459 pant390 pant185 pant457 P27 Rsmw Empty vector –1.0 –0.5 0.0 0.5 1.0 c f 0 5 10 20 0 5 10 20 0 5 10 20 0 5 10 20 g min, L-ara t4rhl t4K99N t4rhl t4K99N CRISPR1 leader CRISPR2 leader PhrS target PhrS CRISPR2 leader PhrS CRISPR2 leader h PhrS pheS CRISPR2 leader 250 50 bp 250 50 bp 200 500 300 h g 0 5 10 20 0 5 10 20 0 5 10 20 0 5 10 20 g min, L-ara t4rhl t4K99N t4rhl t4K99N CRISPR1 leader CRISPR2 leader PhrS target PhrS CRISPR2 leader PhrS pheS CRISPR2 leader 250 50 bp 250 50 bp 200 500 300 S PhrS CRISPR2 leader h Fig. 1 T4 RNA ligase-catalyzed ligation of sRNAs to P. aeruginosa CRISPR loci. a Schematic of the formation of sRNAs chimeras with CRISPR leader by T4 RNA ligase. Two RNA molecules were linked to form pKH6-CRISPR leader plasmid for expressing CRISPR leader and pKH13-t4rhl1 for expressing T4 RNA ligase. Results l sRNA library screening identiﬁes regulators of CRISPR loci. T4 RNA ligase 1 (single-stranded RNA ligase 1) links two base- NATURE COMMUNICATIONS \| (2019) 10:3728 \| https://doi.org/10.1038/s41467-019-11695-8 \| www.nature.com/naturecommunications 2 NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 aeruginosa 35 annotated sRNAs linking to CRISPR leader sequences b T4 RNA ligase in vivo, relative to Supplementary Fig. 1b. Pink represents sRNA-containing chimeras; green represents non-target sRNA chimeras. e IntaRNA prediction of P. aeruginosa annotated sRNAs interactions with CRISPR leader. f Overexpression sRNA to screen candidate sRNAs on regulatio of CRISPR1-lacZ and CRISPR2-lacZ fusion. g Amplicons were detected for PhrS-CRISPR2 leader chimeras. Primer for targets PhrS with CRISPR leader ( shown in a) was carried out for PCR step. PCR production for PhrS and housekeeping gene (pheS) was carried out to ensure the genes expression in samples. h Sequencing reads corresponding to PhrS chimeras with CRISPR2 leader by TA-clone sequencing. Results are shown with mean ± SEM from three independent experiments. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 P < 0.01, P < 0.05, one-way ANOVA plus Tukey test a b IPTG treatment Bacterial count (log cfu/ml) 0 5 10 15 1 2 3 4 5 6 7 8 Time (h) PA14 WT PA14 t4rhl1 PA14 t4K99N AmiL CrcZ EsrA ffs nalA NrsZ P5 P7 P8 P9 P13 P14 P15 P18 P26 P27 P30 P32 P34 P35 PesA PhrS Pail PA4406.1 PA5316.1 PhrX PhrY PrrF1 PrrF2 RgsA RnpB RsmW RsmY RsmZ ssrS AmiL CrcZ EsrA ffs nalA NrsZ P5 P7 P8 P9 P13 P14 P15 P18 P26 P27 P30 P32 P34 P35 PesA PhrS Pail PA4406.1 PA5316.1 PhrX PhrY PrrF1 PrrF2 RgsA RnpB RsmW RsmY RsmZ ssrS d e CRISPR1 leader CRISPR2 leader CRISPR1 leader CRISPR2 leader CRISPR2 leader CRISPR1 leader RT-PCR cDNA synthesis Total RNA isolation Primer: sRNA (specific) Random primer CRISPR2 leader Total 274 sRNAs library pKH6 CRISPR2 leader sRNA target pKH13-T4rnl: T4 RNA ligase a b IPTG treatment Bacterial count (log cfu/ml) 0 5 10 15 1 2 3 4 5 6 7 8 Time (h) PA14 WT PA14 t4rhl1 PA14 t4K99N Binding No binding c CRISPR1 leader CRISPR2 leader 35 annotated sRNAs 239 intergenic sRNAs candidates AmiL CrcZ EsrA ffs nalA NrsZ P5 P7 P8 P9 P13 P14 P15 P18 P26 P27 P30 P32 P34 P35 PesA PhrS Pail PA4406.1 PA5316.1 PhrX PhrY PrrF1 PrrF2 RgsA RnpB RsmW RsmY RsmZ ssrS AmiL CrcZ EsrA ffs nalA NrsZ P5 P7 P8 P9 P13 P14 P15 P18 P26 P27 P30 P32 P34 P35 PesA PhrS Pail PA4406.1 PA5316.1 PhrX PhrY PrrF1 PrrF2 RgsA RnpB RsmW RsmY RsmZ ssrS d e CRISPR1 leader CRISPR2 leader CRISPR1 leader CRISPR2 leader 0 5 10 20 0 5 10 20 0 5 10 20 0 5 10 20 g min, L-ara t4rhl t4K99N t4rhl t4K99N CRISPR1 leader CRISPR2 leader PhrS target PhrS CRISPR2 leader PhrS CRISPR2 leader h f Fold change (log2) * * CRISPR2-lacZ pant391 pant463 PhrS pant312 pant333 pant342 pant349 pant431 pant35 pant81 pant119 pant187 pant167 pant155 pant184 pant245 pant250 pant439 pant445 pant452 P35 pant509 pant457 P27 Rsmw Empty vector pant391 pant463 PhrS pheS CRISPR2 leader –2 –1 0 1 2 –3 250 50 bp 250 50 bp 200 500 300 CRISPR2 leader CRISPR1 leader RT-PCR cDNA synthesis Total RNA isolation Primer: sRNA (specific) Random primer CRISPR2 leader Total 274 sRNAs library pKH6 CRISPR2 leader sRNA target pKH13-T4rnl: T4 RNA ligase Fold change (log2) CRISPR1-lacZ pant92 pant96 pant103 pant459 pant390 pant185 pant457 P27 Rsmw Empty vector –1.0 –0.5 0.0 0.5 1.0 Fig. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 1 T4 RNA ligase-catalyzed ligation of sRNAs to P. aeruginosa CRISPR loci. a Schematic of the formation of sRNAs chimeras with CRISPR leader by RNA ligase. Two RNA molecules were linked to form pKH6-CRISPR leader plasmid for expressing CRISPR leader and pKH13-t4rhl1 for expressing T4 R ligase. Also shown is reverse transcription-polymerase chain reaction (RT-PCR)-based strategy for determining chimeras of CRISPR leader with sRN b T4 RNA ligase or its inactive mutation in t4rnl1 gene with lysine (K) to asparagine (N) affects P. aeruginosa cell growth. c Screening of 274 P. aerugin sRNAs library (239 intergenic sRNAs candidate and 35 annotated sRNAs) linking to CRISPR leader by T4 RNA ligase. Pink represents sRNA-contain chimeras; green represents non-target sRNA chimeras. d Detection of chimeras of P. aeruginosa 35 annotated sRNAs linking to CRISPR leader sequences T4 RNA ligase in vivo, relative to Supplementary Fig. 1b. Pink represents sRNA-containing chimeras; green represents non-target sRNA chimeras. e IntaRNA prediction of P. aeruginosa annotated sRNAs interactions with CRISPR leader. f Overexpression sRNA to screen candidate sRNAs on regulat of CRISPR1-lacZ and CRISPR2-lacZ fusion. g Amplicons were detected for PhrS-CRISPR2 leader chimeras. Primer for targets PhrS with CRISPR leader shown in a) was carried out for PCR step. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 P < 0.01, P < 0.05, one-way ANOVA plus Tukey test a Time (h) 6 8 12 16 24 0 100 200 300 β-gal (RFU/s/OD600) cas-lacZ WT ΔphrS ΔphrS/p-phrS a Time (h) 6 8 12 16 24 Time (h) 6 8 12 16 24 0 100 200 300 β-gal (RFU/s/OD600) 0 100 200 300 β-gal (RFU/s/OD600) β-gal (RFU/s/OD600) 1 2 2 1 CRISPR1-lacZ cas-lacZ WT ΔphrS ΔphrS/p-phrS WT ΔphrS ΔphrS/p-phrS a Time (h) 6 8 12 16 24 Time (h) 6 8 12 16 24 0 100 200 300 β-gal (RFU/s/OD600) β-gal (RFU/s/OD600) 0 100 200 250 50 150 CRISPR2-lacZ CRISPR1-lacZ WT ΔphrS ΔphrS/p-phrS WT ΔphrS ΔphrS/p-phrS Time (h) b WT ΔphrS ΔphrS/p-phrS ΔTCR WT ΔphrS ΔphrS/p-phrS ΔTCR 100 101 102 100 101 102 Efficiency of transformation CR2-sp1 (%) Efficiency of transformation CR1-sp1 (%) ns * ns ns b Ic WT ΔphrS ΔphrS/p-phrS ΔTCR CRISPR 1 crRNA CRISPR 2 crRNA 5s RNA d PhrS 50 bp 50 150 80 e WT ΔphrS/pgRNA ΔphrS/pg-CRISPR2 100 101 102 % CFU containing targeted plasmid CR2-sp1 f e WT ΔphrS/pgRNA ΔphrS/pg-CRISPR2 WT ΔphrS/pgRNA ΔphrS/pg-CRISPR2 Efficiency of transformation CR2-sp1 (%) 100 101 102 100 101 102 % CFU containing targeted plasmid CR2-sp1 f WT ΔphrS/pgRNA ΔphrS/pg-CRISPR2 Efficiency of transformation CR2-sp1 (%) 100 101 102 d f g Fig. 2 PhrS stimulates CRISPR2 crRNA transcription and subsequent CRISPR-Cas interference. a cas-lacZ or CRISPR-lacZ activity in PA14 WT and ΔphrS mutant backgrounds throughout the growth period. b Transformation efﬁciency with CR1-sp1 and CR2-sp1 plasmids in PA14 WT or ΔphrS mutant. c Phage plaque assay of JBD18 and JBD25 for PA14 WT, ΔphrS, ΔphrS/p-phrS, and PA14 lacking a CRISPR-Cas system (ΔTCR). d Northern blot of crRNA levels, PhrS, and 5S RNA in PA14 WT and its mutant strains. e Retention of the CRISPR-targeted plasmid CR2-sp1 in the PA14 ΔphrS background strain with pgRNA-CRISPR2 that coexpressed the crRNA in the CRISPR2 locus. f Transformation efﬁciency of CR2-sp1 vector in PA14 ΔphrS background strain with pgRNA-CRISPR2. g The same JBD18 phage was tested on PA14 ΔphrS background strain containing pgRNA empty vector or plasmid expression the indicated CRISPR2. Results are shown with mean ± SEM from three independent experiments. P < 0.01, P < 0.05, one-way ANOVA plus Tukey test creg motif of PhrS is required for regulating CRISPR system. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 Moreover, addition of CRISPR2 locus into PA14 ΔphrS showed sufﬁcient efﬁciency to account for the CRISPR interference and immunity (Fig. 2e, f), implying that CRISPR2 locus was indeed activated by sRNA PhrS. Meanwhile, expression of CRISPR2 locus in PA14 ΔphrS background strain resulted in lowered plaques efﬁciency of JBD18 (Fig. 2g), corresponding with less potent production of crRNA of CRISPR2 locus in PA14 ΔphrS. Altogether, our results demonstrate that PhrS stimulates CRISPR-Cas-dependent immunity and enhances host defense against invasive element correction by production of the potent crRNA of a speciﬁc CRISPR2 locus. NATURE COMMUNICATIONS \| (2019) 10:3728 \| https://doi.org/10.1038/s41467-019-11695-8 \| www.nature.com/naturecommunications 3 ARTICLE NATURE COMMUNICATIONS \| (2019) 10:3728 \| https://doi.org/10.1038/s41467-019-11695-8 \| www.nature.com/naturecommunications NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 a Time (h) 6 8 12 16 24 Time (h) 6 8 12 16 24 Time (h) 6 8 12 16 24 0 100 200 300 β-gal (RFU/s/OD600) 0 100 200 300 β-gal (RFU/s/OD600) β-gal (RFU/s/OD600) 0 100 200 250 50 150 CRISPR2-lacZ CRISPR1-lacZ cas-lacZ WT ΔphrS ΔphrS/p-phrS WT ΔphrS ΔphrS/p-phrS WT ΔphrS ΔphrS/p-phrS b WT ΔphrS ΔphrS/p-phrS ΔTCR WT ΔphrS ΔphrS/p-phrS ΔTCR 100 101 102 100 101 102 Efficiency of transformation CR2-sp1 (%) Efficiency of transformation CR1-sp1 (%) ns * ns ns JBD18 JBD25 Ten fold phage dilution Infecting phage: Bacteria: c WT ΔTCR ΔphrS ΔphrS/p-phrS WT ΔTCR ΔphrS ΔphrS/p-phrS WT ΔphrS ΔphrS/p-phrS ΔTCR CRISPR 1 crRNA CRISPR 2 crRNA 5s RNA f d e WT ΔphrS/pgRNA ΔphrS/pg-CRISPR2 WT ΔphrS/pgRNA ΔphrS/pg-CRISPR2 Efficiency of transformation CR2-sp1 (%) 100 101 102 100 101 102 % CFU containing targeted plasmid CR2-sp1 g Ten fold phage dilution Infecting phage: Bacteria: JBD18 ΔphrS ΔphrS/pg-CRISPR2 PhrS 50 bp 50 150 80 Fig. 2 PhrS stimulates CRISPR2 crRNA transcription and subsequent CRISPR-Cas interference. a cas-lacZ or CRISPR-lacZ activity in PA14 WT and ΔphrS mutant backgrounds throughout the growth period. b Transformation efﬁciency with CR1-sp1 and CR2-sp1 plasmids in PA14 WT or ΔphrS mutant. c Phage plaque assay of JBD18 and JBD25 for PA14 WT, ΔphrS, ΔphrS/p-phrS, and PA14 lacking a CRISPR-Cas system (ΔTCR). d Northern blot of crRNA levels, PhrS, and 5S RNA in PA14 WT and its mutant strains. e Retention of the CRISPR-targeted plasmid CR2-sp1 in the PA14 ΔphrS background strain with pgRNA-CRISPR2 that coexpressed the crRNA in the CRISPR2 locus. f Transformation efﬁciency of CR2-sp1 vector in PA14 ΔphrS background strain with pgRNA-CRISPR2. g The same JBD18 phage was tested on PA14 ΔphrS background strain containing pgRNA empty vector or plasmid expression the indicated CRISPR2. Results are shown with mean ± SEM from three independent experiments. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 ARTICLE pJT-phrScreg pJT-phrScmut SD pJT-phrScreg pJT-phrS a SD c ΔphrS/pJT-phrS ΔphrS/pJT-phrScreg ΔphrS/pJT-phrScreg ΔphrS/pJT-phrScmut ΔphrS 0 100 200 300 400 β-gal (RFU/s/OD600) 1 2 3 4 Time after induction (h) CRISPR2-lacZ ΔphrS/pJT-phrS ΔphrS/pJT-phrScreg ΔphrS/pJT-phrScreg ΔphrS/pJT-phrScmut ΔphrS WT CRISPR2 crRNA 5s RNA b PhrS d ΔphrS/pJT-phrS ΔphrS/pJT-phrScreg ΔphrS/pJT-phrScreg phrS/pJT-phrScmut ΔphrS 10 0 10 1 10 2 Efficiency of transformation CR2-sp1 (%) ns e ΔphrS/pJT-phrS ΔphrS/pJT-phrScreg ΔphrS/pJT-phrScreg ΔphrS/pJT-phrScmut ΔphrS WT Bacteria: CRISPR-sensitive JBD18 Infection phage: Ten fold phage dilution SD g 1: WT pleader CRISPR2-GFP 3: WT pleader CRISPR2-mut1-GFP 4: WT pleader CRISPR2-mut2-GFP 2: WT pleader CRISPR2-Δ-GFP 1 2 3 4 5 1 2 3 4 5 GFP mode Light mode 5: ΔphrS pleader CRISPR2-GFP leaderCRISPR2-Δ leaderCRISPR2-mut1 leaderCRISPR2-mut2 f PhrS leaderCRISPR2 170 184 +502 +490 5′ 5′ 5′ 5′ 3′ 3′ 3′ 5′ 3′ 3′ h TSS RBS 3′ 5′ pleader CRISPR2-GFP PhrS sRNA binding site WT pleaderCRISPR2-GFP WT pleaderCRISPR2-mut1-GFP WT pleaderCRISPR2-mut2-GFP WT pleaderCRISPR2-Δ-GFP ΔphrS pleaderCRISPR2-GFP 50 bp 80 150 80 37 25 kDa 170 200 169 170 200 P GFP Leader region Fig. 3 creg motif of PhrS is essential for crRNA regulation of CRISPR-Cas adaptive immunity. a Diagram of the sRNA PhrS and mutations. creg motif is marked with bold. Eleven nucleotides of creg were deleted in the plasmid pJT-phrSΔcreg. pJT-phrScmut plasmid contained site mutations in creg motif. pJT-phrScreg represent regions with expression of creg motif. b LacZ activity as determined in the PA14 ΔphrS background strain harboring the PhrS expression plasmid pJT-PhrS, plasmid pJT-PhrSΔcreg, plasmid pJT-phrScmut, or plasmid pJT-phrScreg with toluate (2 mM ﬁnal concentration). c Northern blot of CRISPR2 crRNA production in PA14 ΔphrS background strain containing the PhrS expression plasmid pJT-phrS, plasmid pJT-phrSΔcreg, plasmid pJT-phrScmut, or plasmid pJT-phrScreg. d Transformation efﬁciency of CR2-sp1 vector in PA14 ΔphrS background strain within the plasmid pJT-phrS, pJT- phrSΔcreg, pJT-phrScmut, or pJT-phrScreg. e Phage plaque of CRISPR-sensitive phage JBD18 on PA14 ΔphrS background harboring the plasmid pJT-phrS, pJT- phrSΔcreg, pJT-phrScmut, or pJT-phrScreg. f IntaRNA (Freiburg RNA tools) prediction of PhrS interactions with CRISPR2 leader target (upper). A diagram of leader binding site with phrS knockout or mutant with red parts mark (lower). g GFP ﬂuorescence in PA14 strains or ΔphrS transformed with pleaderCRISPR2-GFP, pleaderCRISPR2-Δ-GFP, pleaderCRISPR2-mut1-GFP, or pleaderCRISPR2-mut2-GFP plasmid, respectively. GFP mode is on the left and the visible light of light mode is on the right). NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 h Western blot analysis of GFP in the PA14 WT or ΔphrS transformed with pleaderCRISPR2-GFP, pleaderCRISPR2-Δ- GFP, pleaderCRISPR2-mut1-GFP, or pleaderCRISPR2-mut2-GFP plasmid, 5 μg total proteins were used to western blotting analysis. Results are shown with mean ± SEM from three independent experiments. P < 0.01, P < 0.05, one-way ANOVA plus Tukey test pJT-phrScreg ΔphrS/pJT-phrS ΔphrS/pJT-phrScreg ΔphrS/pJT-phrScreg ΔphrS/pJT-phrScmut ΔphrS 0 100 200 300 400 β-gal (RFU/s/OD600) 1 2 3 4 Time after induction (h) CRISPR2-lacZ b g pJT-phrScmut SD e 170 200 pJT-phrS a SD c ΔphrS/pJT-phrS ΔphrS/pJT-phrScreg ΔphrS/pJT-phrScreg ΔphrS/pJT-phrScmut ΔphrS WT CRISPR2 crRNA 5s RNA PhrS S 50 bp 80 150 80 170 200 b pJT-phrScreg d SD 169 a d c e d ΔphrS/pJT-phrS ΔphrS/pJT-phrScreg ΔphrS/pJT-phrScreg phrS/pJT-phrScmut ΔphrS 10 0 10 1 10 2 Efficiency of transformation CR2-sp1 (%) * ns h A d ΔphrS/pJT-phrS ΔphrS/pJT-phrScreg ΔphrS/pJT-phrScreg phrS/pJT-phrScmut ΔphrS 10 0 10 1 10 2 Efficiency of transformation CR2-sp1 (%) ns e ΔphrS/pJT-phrS ΔphrS/pJT-phrScreg ΔphrS/pJT-phrScreg ΔphrS/pJT-phrScmut ΔphrS WT Bacteria: CRISPR-sensitive JBD18 Infection phage: Ten fold phage dilution f le h FP FP P Time after induction (h) Time after induction (h) leaderCRISPR2-Δ leaderCRISPR2-mut1 leaderCRISPR2-mut2 f PhrS leaderCRISPR2 170 184 +502 +490 5′ 5′ 5′ 5′ 3′ 3′ 3′ 5′ 3′ 3′ TSS RBS 3′ 5′ pleader CRISPR2-GFP PhrS sRNA binding site P GFP Leader region f Δ h WT pleaderCRISPR2-GFP WT pleaderCRISPR2-mut1-GFP WT pleaderCRISPR2-mut2-GFP WT pleaderCRISPR2-Δ-GFP ΔphrS pleaderCRISPR2-GFP 37 25 kDa Δ g 1: WT pleader CRISPR2-GFP 3: WT pleader CRISPR2-mut1-GFP 4: WT pleader CRISPR2-mut2-GFP 2: WT pleader CRISPR2-Δ-GFP 1 2 3 4 5 1 2 3 4 5 GFP mode Light mode 5: ΔphrS pleader CRISPR2-GFP h g al for crRNA regulation of CRISPR-Cas adaptive immunity. a Diagram of the sRNA PhrS and mutations. creg motif is Fig. 3 creg motif of PhrS is essential for crRNA regulation of CRISPR-Cas adaptive immunity. a Diagram of the sRNA PhrS and mutations. creg motif is marked with bold. Eleven nucleotides of creg were deleted in the plasmid pJT-phrSΔcreg. pJT-phrScmut plasmid contained site mutations in creg motif. pJT-phrScreg represent regions with expression of creg motif. b LacZ activity as determined in the PA14 ΔphrS background strain harboring the PhrS expression plasmid pJT-PhrS, plasmid pJT-PhrSΔcreg, plasmid pJT-phrScmut, or plasmid pJT-phrScreg with toluate (2 mM ﬁnal concentration). NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 c Northern blot of CRISPR2 crRNA production in PA14 ΔphrS background strain containing the PhrS expression plasmid pJT-phrS, plasmid pJT-phrSΔcreg, plasmid pJT-phrScmut, or plasmid pJT-phrScreg. d Transformation efﬁciency of CR2-sp1 vector in PA14 ΔphrS background strain within the plasmid pJT-phrS, pJT- phrSΔcreg, pJT-phrScmut, or pJT-phrScreg. e Phage plaque of CRISPR-sensitive phage JBD18 on PA14 ΔphrS background harboring the plasmid pJT-phrS, pJT- phrSΔcreg, pJT-phrScmut, or pJT-phrScreg. f IntaRNA (Freiburg RNA tools) prediction of PhrS interactions with CRISPR2 leader target (upper). A diagram of leader binding site with phrS knockout or mutant with red parts mark (lower). g GFP ﬂuorescence in PA14 strains or ΔphrS transformed with pleaderCRISPR2-GFP, pleaderCRISPR2-Δ-GFP, pleaderCRISPR2-mut1-GFP, or pleaderCRISPR2-mut2-GFP plasmid, respectively. GFP mode is on the left and the visible light of light mode is on the right). h Western blot analysis of GFP in the PA14 WT or ΔphrS transformed with pleaderCRISPR2-GFP, pleaderCRISPR2-Δ- GFP, pleaderCRISPR2-mut1-GFP, or pleaderCRISPR2-mut2-GFP plasmid, 5 μg total proteins were used to western blotting analysis. Results are shown with mean ± SEM from three independent experiments. P < 0.01, P < 0.05, one-way ANOVA plus Tukey test especially creg element in PhrS, and +491 to +502 segments of CRISPR2 (Fig. 3f, upper), supporting the hypothesis that base- pairing between PhrS and CRISPR2 leader is responsible for PhrS-mediated CRISPR locus transcription. To test this idea, GFP reporter containing three variants of CRISPR2 leader (Fig. 3f, lower) were transformed into PA14 WT. Similar to ΔphrS pleaderCRISPR2-GFP strain, GFP-containing strain plating analy- sis demonstrated that mutant-binding sites in the CRISPR2 leader displayed much weaker ﬂuorescence (Fig. 3g). Moreover, western blot analysis of GFP showed that three variants strains had weaker GFP quantity than the WT pleaderCRISPR2-GFP strain (Fig. 3h). These ﬁndings attest that the binding sites in the CRISPR2 locus are highly subject to CRISPR2 locus transcrip- tional regulation. Taken together, these ﬁndings demonstrate that the precise binding sites between creg element of PhrS and CRISPR2 leader is required for the transcription of CRISPR locus to control CRISPR-Cas immunity response. interference (Supplementary Fig. 3c, d), indicating that PhrS- mediated PQS biosynthesis operon had no role in CRISPR-Cas expression or function. As some sRNAs modulate gene function through interaction with chromosomal DNAs14, the “reverse transcription-associated trap (RAT)” assay29 was performed to detect RNA/DNA interaction through pull-down and PCR by interacting with DNA-speciﬁc primers (Supplementary Fig. 4a). NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 In addition to its regulatory function, PhrS has an open reading frames (ORF) that encodes a conserved 37 amino acid peptide (Supplementary Fig. 2a)27. We found that there is no difference of CRISPR2 locus transcription between PA14 ΔphrS and ΔphrS/phrS-ORF (restored expression of internal ORF of PhrS) by lacZ reporter and northern blotting (Supplementary Fig. 2b, c) and similarly no difference of CRISPR-Cas interference was noticed (Supplementary Fig. 2d, e). These data demonstrated that the internal ORF of PhrS-encoded protein had no effect on CRISPR-Cas functionality, indicating that PhrS as a sRNA may act on PA14 CRISPR-Cas adaptive immunity. background strain (Fig. 3a). As shown in Fig. 3b, pJT-phrScreg in PA14 ΔphrS background, but not pJT-phrSΔcreg and pJT-phrScmut, stimulated transcription levels of CRISPR2-lacZ fusion gene (as a reporter) similar to that of pJT-phrS. In addition, northern blotting showed that pJT-phrS or pJT-phrScreg does not cause reduced expression of CRISPR2 locus compared to PA14 WT (Fig. 3c). Furthermore, pJT-phrScreg showed more efﬁcient transformation inhibition in CRISPR-Cas interference, whereas pJT-phrSΔcreg and pJT-phrScmut did not (Fig. 3d). We also found that CRISPR-sensitive JBD18 can replicate on pJT-phrSΔcreg and pJT-phrScmut, but not on pJT-phrScreg (Fig. 3e). Our observations suggest that CRISPR2 regulation is dependent on the creg element of PhrS. y The secondary structures of PhrS were characterized to contain a conserved region (creg element, 12 nt in length)28. To evaluate which motif or region of PhrS is essential for CRISPR-Cas system, we overexpressed three PhrS functional domain variants (pJT-phrSΔcreg, pJT-phrScmut: point mutations introduced into the conserved creg element, and pJT-phrScreg) in PA14 ΔphrS As shown in Fig. 1a, we used T4 RNA ligase 1 to link two base- paired RNA molecules. We hypothesized that PhrS acts as a regulatory molecule by interacting directly with a leader sequence to control CRISPR2 locus transcription. Computational analysis by IntaRNA tool showed the potential interaction between PhrS, NATURE COMMUNICATIONS \| (2019) 10:3728 \| https://doi.org/10.1038/s41467-019-11695-8 \| www.nature.com/naturecommunicatio 4 NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 NATURE COMMUNICATIONS \| (2019) 10:3728 \| https://doi.org/10.1038/s41467-019-11695-8 \| www.nature.com/naturecommunications NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 ARTICLE c d a Leader region TSS RBS qPCR amplicon 3′ 5′ GFP mode Light mode pleaderCRISPR2-GFP PhrS sRNA binding site Growth JBD18 JBD25 Infecting phage ΔphrS h BCM + + – + + + – + GFP intensity e f 5s RNA CRISPR2 crRNA No BCM + BCM g Absorbance, 595 nm b +BCM +BCM No BCM No BCM WT ΔphrS WT WT WT ΔphrS WT ΔphrS WT ΔphrS No BCM + BCM WT ΔphrS WT ΔphrS WT ΔphrS GFP levels ratio +/– BCM 0 50 100 150 200 WT ΔphrS 6 12 16 24 8 0 2000 4000 6000 Fluorescence intensity units Time (h) WT ΔphrS 5 10 15 20 0 0.5 0 1 1.5 2.0 Ten fold phage dilution ΔphrS ΔphrS Time (h) j i leader:PhrS Leader only PhrS only 1:0.1 1:0.5 1:1 1:5 5 0 10 15 20 25 Fold enrichment (RIP/IgG) Rho-His tag RIP leader:PhrS Input 1:0.1 1:0.5 1:1 1:5 Rho leader RNA pull-down 37 25 kDa 50 80 150 bp 80 75 37 50 kDa P GFP Fig. 4 PhrS inhibits Rho-dependent termination with leader region within CRISPR2 locus. a Diagram of pleaderCRISPR2-GFP construct that includes leade region of CRISPR2 locus. “P” represents Ptac promoter. TSS represents transcription start site. RBS represents a ribosome-binding site. The fragment o qPCR is marked as green. b Time-course of GFP levels in PA14 strains (WT and ΔphrS) transformed with pleaderCRISPR2-GFP grew in LB medium. Bacteri density (left) and GFP levels (right) were simultaneously monitored. c GFP ﬂuorescence of PA14 WT and ΔphrS harboring pleaderCRISPR2-GFP reporter d GFP ﬂuorescence of PA14 WT and ΔphrS harboring pleaderCRISPR2-GFP cultured on LB with agar containing 8 μg/ml bicyclomycin (BCM). e Ratio of gfp in PA14 WT or ΔphrS harboring pleaderCRISPR2-GFP without or within BCM. f Western blot of GFP in PA14 WT and ΔphrS transformed with pleaderCRISPR2 GFP, 5 μg total proteins were used to western blotting analysis. g Northern blot of crRNA levels in PA14 WT and ΔphrS with or without BCM. 5S probe loading control. h Tenfold dilution of lysates of CRISPR-sensitive phages JBD18 and JBD25 grew on bacterial lawns of PA14 WT and ΔphrS without or withi BCM. i RNA immunoprecipitation (RIP) was performed to test PhrS affecting Rho loading with leader sequence via His-Rho. Box plots: center line represents median, min to max (whiskers) and individual data points (blue). NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 f Western blot of GFP in PA14 WT and ΔphrS transformed with pleaderCRISPR2- GFP, 5 μg total proteins were used to western blotting analysis. g Northern blot of crRNA levels in PA14 WT and ΔphrS with or without BCM. 5S probe, loading control. h Tenfold dilution of lysates of CRISPR-sensitive phages JBD18 and JBD25 grew on bacterial lawns of PA14 WT and ΔphrS without or within BCM. i RNA immunoprecipitation (RIP) was performed to test PhrS affecting Rho loading with leader sequence via His-Rho. Box plots: center line represents median, min to max (whiskers) and individual data points (blue). j RNA pull-down to investigate PhrS-mediated inhibition of Rho loading with leader region by detection of Rho proteins through western blotting. Bands corresponding to His-Rho were observed in leader RNA pull-down with PhrS in a dose-dependent manner. Results are shown with mean ± SEM from three independent experiments. P < 0.01, P < 0.05, one-way ANOVA plus Tukey test was abolished by brief exposure to Rho inhibitor bicyclomycin (BCM)30 (Supplementary Fig. 5b, lane 3). A GFP reporter with or without the CRISPR2 leader was generated for in vivo experiment (Supplementary Fig. 5c). In comparison with PA14 WT contain- ing pGFP, plating PA14 WT with pleaderCRISPR2-GFP demon- strated signiﬁcantly subdued ﬂuorescence (Supplementary Fig. 5d), whereas ﬂuorescence of pleaderCRISPR2-GFP strain recovered to the control level in the presence of BCM (Supplementary Fig. 5e). Further, quantitative PCR (qPCR) quantiﬁcation of gfp shows that pleaderCRISPR2-GFP strain exhibited a stronger response to BCM than the pGFP control (Supplementary Fig. 5f, g). In summary, these ﬁndings reveal CRISPR2 leader encompassing a Rho-dependent termination signal. growth period (Fig. 4b, right). Considering the Rho activity on CRISPR2 leaders, our data indicate PhrS represses Rho- dependent termination, leading to increased GFP levels. Plating PA14 ΔphrS transformed with pleaderCRISPR2-GFP plasmid displayed markedly reduced ﬂuorescence compared to that of WT (Fig. 4c). Supplementing BCM increased the ﬂuorescence of PA14 ΔphrS (Fig. 4d). Moreover, GFP levels in PA14 WT were altered slightly, while PA14 ΔphrS manifested almost a 100-fold augmentation with BCM treatment conﬁrmed by transcript and protein determination (Fig. 4e, f). This was consistent with the expression of crRNA molecules in the CRISPR2 locus by northern blotting (Fig. 4g). In addition, the plaquing efﬁciency of phage JBD18 was reduced compared to PA14 ΔphrS with BCM treatment (Fig. 4h). NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 j RNA pull-down to investigate PhrS-mediated inhibition of Rho loading wit leader region by detection of Rho proteins through western blotting. Bands corresponding to His-Rho were observed in leader RNA pull-down with PhrS in dose-dependent manner. Results are shown with mean ± SEM from three independent experiments. P < 0.01, P < 0.05, one-way ANOVA plus Tukey tes c a Leader region TSS RBS qPCR amplicon 3′ 5′ GFP mode Light mode pleaderCRISPR2-GFP PhrS sRNA binding site No BCM No BCM WT ΔphrS WT ΔphrS P GFP Growth Absorbance, 595 nm b WT ΔphrS 5 10 15 20 0 0.5 0 1 1.5 2.0 Time (h) b a GFP intensity WT ΔphrS 6 12 16 24 8 0 2000 4000 6000 Fluorescence intensity units Time (h) e f No BCM + BCM WT ΔphrS WT ΔphrS WT ΔphrS GFP levels ratio +/– BCM 0 50 100 150 200 * 37 25 kDa f No BCM + BCM WT ΔphrS WT ΔphrS 37 25 kDa 5s RNA CRISPR2 crRNA No BCM + BCM g WT ΔphrS WT ΔphrS j 50 80 150 bp 80 f e g d +BCM +BCM WT WT ΔphrS ΔphrS d j i leader:PhrS Leader only PhrS only 1:0.1 1:0.5 1:1 1:5 5 0 10 15 20 25 Fold enrichment (RIP/IgG) Rho-His tag RIP leader:PhrS Input 1:0.1 1:0.5 1:1 1:5 Rho leader RNA pull-down 80 75 37 50 kDa i leader:PhrS Leader only PhrS only 1:0.1 1:0.5 1:1 1:5 5 0 10 15 20 25 Fold enrichment (RIP/IgG) Rho-His tag RIP i JBD18 JBD25 Infecting phage ΔphrS h BCM + + – + + + – + Ten fold phage dilution i j I h Fig. 4 PhrS inhibits Rho-dependent termination with leader region within CRISPR2 locus. a Diagram of pleaderCRISPR2-GFP construct that includes leader region of CRISPR2 locus. “P” represents Ptac promoter. TSS represents transcription start site. RBS represents a ribosome-binding site. The fragment of qPCR is marked as green. b Time-course of GFP levels in PA14 strains (WT and ΔphrS) transformed with pleaderCRISPR2-GFP grew in LB medium. Bacteria density (left) and GFP levels (right) were simultaneously monitored. c GFP ﬂuorescence of PA14 WT and ΔphrS harboring pleaderCRISPR2-GFP reporter. d GFP ﬂuorescence of PA14 WT and ΔphrS harboring pleaderCRISPR2-GFP cultured on LB with agar containing 8 μg/ml bicyclomycin (BCM). e Ratio of gfp in PA14 WT or ΔphrS harboring pleaderCRISPR2-GFP without or within BCM. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 We observed no interaction between PhrS and CRISPR2 locus chromosomal DNAs as well as cas1 locus and pheS locus (Sup- plementary Fig. 4b). especially creg element in PhrS, and +491 to +502 segments of CRISPR2 (Fig. 3f, upper), supporting the hypothesis that base- pairing between PhrS and CRISPR2 leader is responsible for PhrS-mediated CRISPR locus transcription. To test this idea, GFP reporter containing three variants of CRISPR2 leader (Fig. 3f, lower) were transformed into PA14 WT. Similar to ΔphrS pleaderCRISPR2-GFP strain, GFP-containing strain plating analy- sis demonstrated that mutant-binding sites in the CRISPR2 leader displayed much weaker ﬂuorescence (Fig. 3g). Moreover, western blot analysis of GFP showed that three variants strains had weaker GFP quantity than the WT pleaderCRISPR2-GFP strain (Fig. 3h). These ﬁndings attest that the binding sites in the CRISPR2 locus are highly subject to CRISPR2 locus transcrip- tional regulation. Taken together, these ﬁndings demonstrate that the precise binding sites between creg element of PhrS and CRISPR2 leader is required for the transcription of CRISPR locus to control CRISPR-Cas immunity response. sRNAs are shown to be powerful regulators because they can modulate both transcription and translation14. Owing to only pre-crRNA transcribed from CRISPR loci, we focused on PhrS inﬂuence on the process of transcription (transcriptional elonga- tion and termination) at CRISPR2 locus. We hypothesized that PhrS may regulate CRISPR2 transcription via inhibition of Rho- dependent transcriptional termination, because there is no G–C rich hairpin loop at CRISPR2 leader for intrinsic termination (Rho-independent termination) to control RNA transcription. To investigate this, we ﬁrst tested whether Rho terminates CRISPR2 transcription using one round transcription reaction method. Transcription of CRISPR2 template (Supplementary Fig. 5a) gave rise to an intact transcript without Rho (Supplementary Fig. 5b, lane 1). However, Rho together with NusG prompted strong transcription termination (Supplementary Fig. 5b, lane 2), which PhrS controls Rho-dependent termination at CRISPR2 locus. We next sought to determine the mechanism how PhrS exerts its function in CRISPR-Cas systems. PhrS as an activator of PqsR synthesis stimulates PQS biosynthesis operon (PqsA-E)28. How- ever, mutation of PqsA-E has no effect on CRISPR2 locus tran- scription (Supplementary Fig. 3a, b) and consequent CRISPR-Cas TURE COMMUNICATIONS \| (2019) 10:3728 \| https://doi.org/10.1038/s41467-019-11695-8 \| www.nature.com/naturecommunications 5 NATURE COMMUNICATIONS \| (2019) 10:3728 \| https://doi.org/10.1038/s41467-019-11695-8 \| www.nature.com/naturecommunications NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 Next, to further test whether the binding sites are required for PhrS-mediated anti-termination in a CRISPR locus, we further generated three variants of the CRISPR2 leader (Figs. 3f and 6a) to perform transcription analysis. We found that anti-termination mediated by PhrS was speciﬁc for the binding sites in the CRISPR2 locus, because PhrS has no effect on Rho-dependent termination for other three variants of a CRISPR2 leader region without PhrS-binding sites (Fig. 6a, left). Moreover, Rho together with NusG leaded to obvious transcription termination (Fig. 6a, right, lane 2), which was not suppressed by PhrS with deﬁciency of PhrS-binding sites in the transcriptional template (Fig. 6a, right, lane 3), but was greatly inhibited by BCM (Fig. 6a, right, lane 4). Furthermore, GFP-containing strain plating analysis also illustrated that the mutant-binding sites of the CRISPR2 leader displayed remarkably attenuated ﬂuorescence (Fig. 6b, upper). Supplementing BCM increased the ﬂuorescence of three variants strains (Fig. 6b, lower). Moreover, qPCR analysis of gfp revealed that three variants strains exhibited signiﬁcantly lower GFP quantity than the WT pleaderCRISPR2-GFP strain (Fig. 6c), consistent with the intensity of the GFP signal (Fig. 6d). GFP expression in the three variants was highly induced after BCM treatment (Fig. 6c, d), indicating that the binding sites of PhrS and CRISPR2 locus are highly subject to the activity of PhrS- mediated anti-termination. Reconstitution of PhrS on anti-termination in CRISPR2 locus. In vitro reconstituted system enabling a single round transcrip- tion assay showed that PhrS, similar to BCM treatment, abolished Rho-mediated robust termination (Fig. 5a), meaning that PhrS interacts directly with CRISPR2 leader to inhibit Rho-dependent termination. Next, we estimated the activity for Rho-mediated termination and PhrS-mediated anti-termination at the CRISPR2 locus. qRT- PCR was performed to determine various regions (UTR [untrans- lated region], 5ʹORF and ORF) of CRISPR2-lacZ transcript in PA14 WT and ΔphrS strains containing translational CRISPR2- lacZ reporter (Fig. 5b). Normalized to “ORF” with each strain, we attained the determination of the termination efﬁciency within lacZ between PA14 WT and ΔphrS. Compared to BCM treatment, the [UTR]/[ORF] or [5ʹUTR]/[ORF] value in PA14 WT without BCM was increased due to powerful Rho-mediated termination (Fig. 5c). Furthermore, a higher value of [UTR]/ [ORF] or [5ʹUTR]/[ORF] in PA14 ΔphrS compared to WT inhibited the efﬁcient Rho-dependent termination (Fig. 5c), consistent with a stronger termination phenotype within the CRISPR2 leader in PA14 ΔphrS strain. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 Collectively, these observations support our notion that PhrS inhibits Rho-dependent transcriptional termi- nation to stimulate CRISPR2 crRNA synthesis against phage infection. g To test whether PhrS inﬂuences Rho function on CRISPR2 leader, GFP ﬂuorescence assay in PA14 WT and ΔphrS contain- ing pleaderCRISPR2-GFP plasmid was utilized (Fig. 4a). PA14 WT and ΔphrS containing GFP reporter cultured in luria broth (LB) medium displayed the same growth rate (Fig. 4b, left). None- theless, the expression level of GFP gradually increased in PA14 WT rather than the ΔphrS mutant strain throughout the surveyed To further verify this concept, overexpression of PhrS resulted in the robust increase of lacZ transcripts, consistent with the response to BCM (Supplementary Fig. 6a, b). Remarkably, BCM treatment did not lead to comparable induction of CRISPR2 6 ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 with pleaderCRISPR2(+x+y)-GFP derivatives with various seg- ments of the CRISPR2 leader (Supplementary Fig. 7a). All strains had equal growth rates (Supplementary Fig. 7b, c, lower), but WT pleaderCRISPR2(+400+600)-GFP exhibited much diminished ﬂuorescence, as well as WT pleaderCRISPR2-GFP intensity, compared to other derivative strains (Supplementary Fig. 7b, c, upper), indicating that Rho termination activity may be attri- butable to a ~ + 400 to +600 segment of the CRISPR2 leader, which overlap the binding sites between PhrS and the CRISPR2 leader. leader in response to PhrS increase (Supplementary Fig. 6c). Altogether, PhrS-mediated repression of Rho in CRISPR2 leader is attributed to the stimulatory effect of PhrS on CRISPR2 transcription. p Rho has to bind RNA to achieve termination21. To investigate the mechanism by which PhrS causes the inhibition of Rho, we performed crosslinked RNA immunoprecipitation (RIP) via His- Rho protein to investigate whether PhrS affects Rho loading or translocation with the CRISPR2 leader. We detected the gradually decreased enrichment of the CRISPR2 leader with increasing doses of PhrS (Fig. 4i), indicating that PhrS affects Rho loading. Consequently, to conﬁrm PhrS-mediated inhibition of the biochemical interaction of the CRISPR2 leader with Rho, we performed RNA pull-down assays. The biotin-labeled CRISPR2 leader was transcribed, hybridized to PhrS, and added His-Rho protein for incubation. Samples captured on streptavidin magnetic beads were detected through western blotting analysis to identify and conﬁrm that Rho binding gradually reduced due to the increased PhrS (Fig. 4j). Altogether, PhrS represses Rho loading to stimulate CRISPR2 transcription. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 Indeed, treatment of BCM drastically reduced the values of [UTR]/[ORF] or [5ʹUTR]/[ORF] in PA14 WT and ΔphrS (Fig. 5c). To further probe the underlying mechanism, we also detected the native CRISPR2 transcript (pre- crRNA) in PA14 Δcsy4 and Δcys4/ΔphrS strains given Csy4 being responsible for pre-crRNA processing into short crRNAs in PA1431. To this end, we found that PA14 Δcsy4 displayed greatly increased expression levels of pre-crRNAs in Δcsy4 strains deﬁcient in PhrS (Fig. 5d). [UTR]/[ORF] and [5ʹUTR]/[ORF] ratio of the transcript of the CRISPR2 locus for PA14 Δcys4/ ΔphrS strain signiﬁcantly exceeded PA14 Δcsy4, as well as BCM treatment (Fig. 5e). Taken together, these data support the cumulative effect of PhrS on Rho anti-termination as determined in vitro. To further substantiate this conclusion, GFP assay was used with PA14 strains: WT, ΔphrS, ΔphrS/pJT-phrScmut carrying points mutations introduced into the creg element and ΔphrS/ pJT-phrScmut/p-phrS, which harbored pleaderCRISPR2-GFP repor- ter plasmid. Of note, GFP-containing strain plating analysis also demonstrated that mutant creg sites of PhrS in ΔphrS/pJT- phrScmut strain, similar to ΔphrS strain, displayed greatly weakened ﬂuorescence, but restored to the control level of the WT upon complementing ΔphrS/pJT-phrScmut strain (Fig. 6e, upper). Supplementing BCM increased the ﬂuorescence of ΔphrS/ pJT-phrScmut strain (Fig. 6e, lower). Moreover, qRT-PCR-based quantiﬁcation and western blot analysis of GFP showed that ΔphrS/pJT-phrScmut strain had much weaker GFP quantity than the WT strain (Fig. 6f, g). Altogether, we conclude that the inhibition of Rho termination is dependent on PhrS direct binding to CRISPR2 leader, resulting in elevated expression of the CRISPR locus in the PA14 I-F CRISPR-Cas system. PhrS on anti-termination is a common event for CRISPR system. To further characterize the role of sRNA PhrS in multiple types CRISPR-Cas regulation, we tested two other types I-C/-E CRISPR-Cas systems, each with at least one CRISPR array, which P. aeruginosa ST277 and SM4386 possess PhrS or its homologs with PA14 strains (Supplementary Fig. 8a). We observed that cas operon expression, same to CRISPR2 (I-E) and CRISPR1 (I-F) was not altered by PhrS (Supplementary Fig. 8b). Remarkably, CRISPR1 locus-associated type I-C and I-E systems, similar to CRISPR2 (I-F), exhibited signiﬁcant reduction in PhrS-deﬁcient strains (Supplementary Fig. 8b). In agreement, complementation of PhrS with plasmids restored the expression of CRISPR1 (I-C), CRISPR1 (I-E), and CRISPR2 (I-F). Direct target of PhrS and CRISPR2 inhibits Rho termination. sRNAs regulate diverse processes through a variety of distinct mechanisms14. NATURE COMMUNICATIONS \| (2019) 10:3728 \| https://doi.org/10.1038/s41467-019-11695-8 \| www.nature.com/naturecommunications NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 a PhrS sRNA inhibits Rho termination at leader sequence of CRISPR2 locu RNAP Rho + NusG PhrS + – BCM + + + – – – + – + – + + – – + + + – a Runoff T In vitro transcription template nP 5′ 1 nt…783 nt Leader region 1 2 3 4 5 nt 200 500 3′ TSS pre-crRNA b d c Δcsy4, –BCM Δcsy4, +BCM Δcsy4/ΔphrS, –BCM Δcsy4/ΔphrS, +BCM e WT, –BCM WT, +BCM ΔphrS, +BCM ΔphrS, –BCM [UTR] [ORF] [5′ORF] [ORF] [ORF] [ORF] [UTR] [ORF] [5′ORF] [ORF] [ORF] [ORF] 0 10 20 30 Relative expression (CRISPR locus) 0 5 10 15 20 25 Relative expression (lacZ) 0.0 0.5 1.0 1.5 2.0 Relative expression (CRISPR locus, 5′ORF) ns ** * ns ns ** * ns * Δcsy4 Δcsy4/ΔphrS Δcsy4, –BCM Δcsy4, +BCM Δcsy4/ΔphrS, –BCM Δcsy4/ΔphrS, +BCM e [UTR] [ORF] [5′ORF] [ORF] [ORF] [ORF] 0 10 20 30 Relative expression (CRISPR locus) ns * ns * e d c e Fig. 5 Reconstituted PhrS-mediated anti-termination acts on CRISPR2 locus. a PhrS sRNA inhibits Rho termination at leader sequence of CRISPR2 locus in vitro. Preformed elongation was transcribed without (lane 1 and 5) or with (lane 2–4) Rho and NusG. The Rho-dependent termination was estimated without (lane 2) or with PhrS sRNA (lane 3) or BCM (lane 4). The bracket represents termination regions (T). b Diagram of CRISPR leader with lacZ ORF and CRISPR array. The fragments for qRT-PCR are labeled as green. The dynamic value of [UTR]/[ORF] was used to assess termination efﬁciency of Rho- dependent termination by PhrS in CRISPR leader, normalized by CRISPR array and lacZ RNA. Increased value represents more termination efﬁciency, but a lower value indicates reduction of termination efﬁciency. c Transcript expression of various lacZ zones quantiﬁed for PA14 WT or ΔphrS before and after BCM treatment. d Levels of CRISPR array measured for the PA14 Δcsy4 and Δcsy4/ΔphrS mutant. e Transcript expression of various CRISPR locus zones quantiﬁed for PA14 Δcsy4 or Δcsy4/ΔphrS before and after BCM treatment. Results are shown with mean ± SEM from three independent experiments. *P < 0.01, P < 0.05, one-way ANOVA plus Tukey test Next, we examined whether PhrS-mediated anti-termination occurs at CRISPR loci in multiple CRISPR-Cas systems. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 RNAP Rho + NusG PhrS + – BCM + + + – – – + – + – + + – – + + + – a d c Δcsy4, –BCM Δcsy4, +BCM Δcsy4/ΔphrS, –BCM Δcsy4/ΔphrS, +BCM e WT, –BCM WT, +BCM ΔphrS, +BCM ΔphrS, –BCM Leader region TSS 5′ nP 3′ Δcsy4 Background strain Leader region TSS 3′ 5′ nP 5′ORF WT Background strain CRISPR2-lacZ Runoff T Condition 1 Condition 2 Condition 3 termination termination [UTR1] [ORF1] [UTR2] [ORF2] < [UTR1] [ORF1] [UTR3] [ORF3] > Termination models [UTR] [ORF] [5′ORF] [ORF] [ORF] [ORF] [UTR] [ORF] [5′ORF] [ORF] [ORF] [ORF] 0 10 20 30 Relative expression (CRISPR locus) 0 5 10 15 20 25 Relative expression (lacZ) 0.0 0.5 1.0 1.5 2.0 Relative expression (CRISPR locus, 5′ORF) ns ** * ns ns ** * ns * In vitro transcription template nP 5′ 1 nt…783 nt Leader region 1 2 3 4 5 ORF qPCR amplicon UTR Δcsy4 Δcsy4/ΔphrS tit pre-crRNA i LacZ ATG nt 200 500 PhrS sRNA binding site PhrS sRNA binding site b 3′ TSS CRISPR2 locus pre-crRNA Fig. 5 Reconstituted PhrS-mediated anti-termination acts on CRISPR2 locus. a PhrS sRNA inhibits Rho termination at leader sequence of CRISPR2 locus in vitro. Preformed elongation was transcribed without (lane 1 and 5) or with (lane 2–4) Rho and NusG. The Rho-dependent termination was estimated without (lane 2) or with PhrS sRNA (lane 3) or BCM (lane 4). The bracket represents termination regions (T). b Diagram of CRISPR leader with lacZ ORF and CRISPR array. The fragments for qRT-PCR are labeled as green. The dynamic value of [UTR]/[ORF] was used to assess termination efﬁciency of Rho- dependent termination by PhrS in CRISPR leader, normalized by CRISPR array and lacZ RNA. Increased value represents more termination efﬁciency, but a lower value indicates reduction of termination efﬁciency. c Transcript expression of various lacZ zones quantiﬁed for PA14 WT or ΔphrS before and after BCM treatment. d Levels of CRISPR array measured for the PA14 Δcsy4 and Δcsy4/ΔphrS mutant. e Transcript expression of various CRISPR locus zones quantiﬁed for PA14 Δcsy4 or Δcsy4/ΔphrS before and after BCM treatment. Results are shown with mean ± SEM from three independent experiments. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 The precise base-paired sites between the creg element of PhrS and the CRISPR2 leader are required for the CRISPR2 locus transcription (Fig. 3). We hypothesized that PhrS acts as a regulatory molecule by interacting directly with a leader sequence to control Rho-dependent in CRISPR2 locus. To test this notion, we evaluated whether a speciﬁc region of the CRISPR2 leader is required, plating PA14 WT was transformed 7 TURE COMMUNICATIONS \| (2019) 10:3728 \| https://doi.org/10.1038/s41467-019-11695-8 \| www.nature.com/naturecommunications ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 *P < 0.01, P < 0.05, one-way ANOVA plus Tukey test Leader region TSS 5′ nP 3′ Δcsy4 Background strain Leader region TSS 3′ 5′ nP 5′ORF WT Background strain CRISPR2-lacZ Condition 1 Condition 2 Condition 3 termination termination [UTR1] [ORF1] [UTR2] [ORF2] < [UTR1] [ORF1] [UTR3] [ORF3] > Termination models ORF qPCR amplicon UTR tit pre-crRNA i LacZ ATG PhrS sRNA binding site PhrS sRNA binding site b CRISPR2 locus RNAP Rho + NusG PhrS + – BCM + + + – – – + – + – + + – – + + + – a d c Δcsy4, –BCM Δcsy4, +BCM Δcsy4/ΔphrS, –BCM Δcsy4/ΔphrS, +BCM e WT, –BCM WT, +BCM ΔphrS, +BCM ΔphrS, –BCM Leader region TSS 5′ nP 3′ Δcsy4 Background strain Leader region TSS 3′ 5′ nP 5′ORF WT Background strain CRISPR2-lacZ Runoff T Condition 1 Condition 2 Condition 3 termination termination [UTR1] [ORF1] [UTR2] [ORF2] < [UTR1] [ORF1] [UTR3] [ORF3] > Termination models [UTR] [ORF] [5′ORF] [ORF] [ORF] [ORF] [UTR] [ORF] [5′ORF] [ORF] [ORF] [ORF] 0 10 20 30 Relative expression (CRISPR locus) 0 5 10 15 20 25 Relative expression (lacZ) 0.0 0.5 1.0 1.5 2.0 Relative expression (CRISPR locus, 5′ORF) ns ** * ns ns ** * ns * In vitro transcription template nP 5′ 1 nt…783 nt Leader region 1 2 3 4 5 ORF qPCR amplicon UTR Δcsy4 Δcsy4/ΔphrS tit pre-crRNA i LacZ ATG nt 200 500 PhrS sRNA binding site PhrS sRNA binding site b 3′ TSS CRISPR2 locus pre-crRNA Fig. 5 Reconstituted PhrS-mediated anti-termination acts on CRISPR2 locus. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 IntaRNA platform analysis showed that these CRISPR loci also possess a potential target site for the creg element of PhrS (Fig. 7a). Monitoring the β-galactosidase activities conﬁrmed that all these CRISPR loci were downregulated in PhrS-deﬁcient strains vs. WT strains (Fig. 7b). Pre-treatment of PhrS-deﬁcient strains with BCM led to an apparent increase of three CRISPR loci (Fig. 7b). Importantly, overexpression of PhrS resulted in upregulation of three CRISPR loci (Fig. 7c). The ﬁndings conﬁrm PhrS-mediated anti-termination for type I-C/-E/-F CRISPR loci. Collectively, these observations illustrate that PhrS blocks Rho-dependent termination by targeting CRISPR leaders, which is likely a common phenomenon in multiple CRISPR-Cas systems among the bacteria kingdom (Fig. 7d). mechanisms, such as changes in RNA conformation12,18. Here, we elaborate that PhrS inhibits Rho-dependent termination to ensure CRISPR transcription, establishing a sophisticated prin- ciple of sRNA-mediated transcription control of CRISPR-Cas adaptive immunity. Furthermore, the formation of PhrS-leader complex impedes Rho loading on RNA molecules in the CRISPR loci. Our results illuminate a concept that CRISPR leaders not only contain a conserved integration host factor to create the ideal target substrate for Cas1-Cas2 during spacer acquisition11,25,26, but also facilitate crRNA biogenesis via control of CRISPR loci transcription. p Co-evolutionary dynamics between bacteria and phage presses the emergence of bacterial defense systems32. Understanding CRISPR-Cas adaptive immune systems in bacteria has improved our knowledge of bacterial biology and phage–host interaction. While insight into the CRISPR-Cas spacer acquisition and interference stage has expeditiously amassed, the molecular machineries for promptly and precisely stimulating CRISPR-Cas adaptive immunity is rather limited. Previous studies show that bacteria control CRISPR-Cas systems through quorum sensing autoinducers33,34. We reveal that PhrS activates the transcription of pre-crRNA in CRISPR-Cas systems. In agreement, PhrS deletion reduces CRISPR-Cas immunity against phage and invading DNA, illustrating an anti-phage mechanism involving NATURE COMMUNICATIONS \| (2019) 10:3728 \| https://doi.org/10.1038/s41467-019-11695-8 \| www.nature.com/naturecommunicatio NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 ARTICLE a RNAP Rho+NusG PhrS Runoff T Template + + + – + + – + + + – + + + + + + + PhrS binding site mutant nP nP In vitro transcription template c leaderCRISPR2 leaderCRISPR2-Δ leaderCRISPR2-mut1 leaderCRISPR2-mut2 leaderCRISPR2 leaderCRISPR2 0.0000 0.0002 0.0004 0.0006 0.0008 0.0010 GFP expression levels No BCM +BCM ns ns d ns + + + – + – – – + – + – + + – – + + + – RNAP Rho+NusG PhrS BCM leaderCRISPR2-Δ GFP mode Light mode No BCM No BCM 1: pleaderCRISPR2-GFP 3: pleaderCRISPR2-mut1-GFP 1 4 b 0 2000 4000 6000 GFP intensity units 8000 10,000 ns No BCM +BCM 1 4 +BCM +BCM 1 4 1 2 3 2 3 2 3 2 3 4 GFP mode Light mode No BCM No BCM 1: WT 4: ΔphrS/pJT-phrScmut/p-phrS 2: ΔphrS 1 2 3 4 e 1 2 3 4 1 4 1 2 3 2 3 4 3: ΔphrS/pJT-phrScmut 0.0000 0.0002 0.0004 0.0006 0.0008 0.0010 GFP expression levels No BCM +BCM ns ns WT ΔphrS/pJT-phrScmut/p-phrS ΔphrS ΔphrS/pJT-phrScmut f g No BCM +BCM 1 2 3 4 1 2 3 4 1: WT 4: ΔphrS/pJT-phrScmut/p-phrS 2: ΔphrS 3: ΔphrS/pJT-phrScmut + BCM + BCM 4: pleaderCRISPR2-mut2-GFP 2: pleaderCRISPR2-Δ-GFP nt 200 500 nt 200 500 kDa 37 25 pleaderCRISPR2-Δ-GFP pleaderCRISPR2-GFP pleaderCRISPR2-mut1-GFP pleaderCRISPR2-mut2-GFP pleaderCRISPR2-Δ-GFP pleaderCRISPR2-GFP pleaderCRISPR2-mut1-GFP pleaderCRISPR2-mut2-GFP Fig. 6 The effect of PhrS sRNA on CRISPR transcription via a direct regulatory target. a Transcription reaction shows that PhrS did not hamper Rho termination on CRISPR2 leader mutants’ template with or without BCM. b GFP ﬂuorescence in PA14 strains containing pleaderCRISPR-GFP, pleaderCRISPR-Δ- GFP, pleaderCRISPR-mut1-GFP, or pleaderCRISPR-mut2-GFP plasmid, respectively, without (upper) or with (lower) BCM. c qPCR for GFP levels with or without BCM for PA14 strains containing pleaderCRISPR-GFP, pleaderCRISPR-Δ-GFP, pleaderCRISPR-mut1-GFP, or pleaderCRISPR-mut2-GFP plasmid, respectively. d CRISPR2 leader mutation affect GFP intensity. e GFP ﬂuorescence in PA14 WT, ΔphrS, ΔphrS/pJT-phrScmut, and ΔphrS/pJT-phrScmut/p-phrS strains transformed with pleaderCRISPR-GFP plasmid without (upper) or with (lower) BCM. f qPCR assay for GFP levels without or with BCM for PA14 WT, ΔphrS, ΔphrS/pJT-phrScmut, and ΔphrS/pJT-phrScmut/p-phrS strains transformed with pleaderCRISPR-GFP plasmid. g Western blot analysis of GFP in the PA14 WT, ΔphrS, ΔphrS/pJT-phrScmut, and ΔphrS/pJT-phrScmut/p-phrS strains transformed with pleaderCRISPR-GFP plasmid. Ten micrograms total proteins were used in western blotting analysis. Results are shown with mean ± SEM from three independent experiments. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 Ten micrograms total proteins were used in western blotting analysis. Results are shown with mean ± SEM from three independent experiments. P < 0.01, P < 0.05, one-way ANOVA plus Tukey test PhrS. PhrS stimulates PqsR synthesis to facilitate synthesis of quinolone signal, which links to oxygen availability to impact the formation of P. aeruginosa bioﬁlms28. Our ﬁndings deﬁne a regulatory role of sRNA PhrS in CRISPR-Cas activity to battle against phage infection, demonstrating an added layer of reg- ulation in CRISPR-Cas adaptive immunity. the formation of a PhrS-leader complex can inhibit Rho- dependent termination rather than stimulate it. Furthermore, our research leads us to believe that PhrS disrupts Rho loading along with the nascent CRISPR leader or 5ʹUTR to control transcriptional process in P. aeruginosa. p p g The environmental cues, for instance cell density, may stimu- late sRNA expression to modulate bacterial behavior via the base- pairing mechanism16,38. Increasing numbers of sRNAs have been characterized, which has signiﬁcantly improved understanding of their biological function. The regions for base-pairing between sRNA and RNA molecules are marked as seed region. Interest- ingly, there is only one seed region in some sRNAs, such as RybB39, whereas numerous seed regions for others matching various RNA molecules, such as Spot 4240. The precise mechanism of the creg element of PhrS directly targeted with the CRISPR2 leader is required for CRISPR locus transcription, revealing that the seed regions for speciﬁc base-pairing lead to the unwinding of structural elements and ultimately repression of the premature transcription termination in a CRISPR locus. p y The general termination factor Rho plays an important role in riboswitch-mediated gene regulation that alters the expression of associated protein-coding regions35,36. Recently, it was shown that long 5ʹUTRs of bacterial genes powerfully facilitate Rho- mediated regulatory signals15. Here, we provide strong evidence that Rho-dependent transcription termination acts at CRISPR leaders, demonstrating that Rho is required for the control of bacterial non-coding RNA transcription. This phenomenon is also seen in type I-C/-E CRISPR-Cas systems, indicating that CRISPR leader perhaps functions as a general target site for Rho-mediated regulation. Importantly, our ﬁndings reveal a surprisingly widespread mechanism in which PhrS mediates anti- termination of transcription by inhibiting the activity of Rho- dependent transcription termination via base-pairing with the regulatory motifs of CRISPR loci leader, which affects the tran- scription of CRISPR loci. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 P < 0.01, P < 0.05, one-way ANOVA plus Tukey test c 0.0000 0.0002 0.0004 0.0006 0.0008 0.0010 GFP expression levels No BCM +BCM ns ns 0 0 pleaderCRISPR2-Δ-GFP pleaderCRISPR2-GFP pleaderCRISPR2-mut1-GFP pleaderCRISPR2-mut2-GFP a RNAP Rho+NusG PhrS Runoff T Template + + + – + + – + + + – + + + + + + + PhrS binding site mutant nP nP In vitro transcription template leaderCRISPR2 leaderCRISPR2-Δ leaderCRISPR2-mut1 leaderCRISPR2-mut2 leaderCRISPR2 leaderCRISPR2 + + + – + – – – + – + – + + – – + + + – RNAP Rho+NusG PhrS BCM leaderCRISPR2-Δ nt 200 500 nt 20 50 c GFP mode Light mode No BCM No BCM 1: WT 4: ΔphrS/pJT-phrScmut/p-phrS 2: ΔphrS 1 2 3 4 e 1 2 3 4 1 4 1 2 3 2 3 4 3: ΔphrS/pJT-phrScmut + BCM + BCM e a 0.0000 0.0002 0.0004 0.0006 0.0008 0.0010 GFP expression levels No BCM +BCM ns ns WT ΔphrS/pJT-phrScmut/p-phrS ΔphrS ΔphrS/pJT-phrScmut f g f d ns 0 2000 4000 6000 GFP intensity units 8000 10,000 ns No BCM +BCM pleaderCRISPR2-Δ-GFP pleaderCRISPR2-GFP pleaderCRISPR2-mut1-GFP pleaderCRISPR2-mut2-GFP GFP mode Light mode No BCM No BCM 1: pleaderCRISPR2-GFP 3: pleaderCRISPR2-mut1-GFP 1 4 b 1 4 +BCM +BCM 1 4 1 2 3 2 3 2 3 2 3 4 4: pleaderCRISPR2-mut2-GFP 2: pleaderCRISPR2-Δ-GFP d b d g No BCM +BCM 1 2 3 4 1 2 3 4 1: WT 4: ΔphrS/pJT-phrScmut/p-phrS 2: ΔphrS 3: ΔphrS/pJT-phrScmut kDa 37 25 g Fig. 6 The effect of PhrS sRNA on CRISPR transcription via a direct regulatory target. a Transcription reaction shows that PhrS did not hamper Rho termination on CRISPR2 leader mutants’ template with or without BCM. b GFP ﬂuorescence in PA14 strains containing pleaderCRISPR-GFP, pleaderCRISPR-Δ- GFP, pleaderCRISPR-mut1-GFP, or pleaderCRISPR-mut2-GFP plasmid, respectively, without (upper) or with (lower) BCM. c qPCR for GFP levels with or without BCM for PA14 strains containing pleaderCRISPR-GFP, pleaderCRISPR-Δ-GFP, pleaderCRISPR-mut1-GFP, or pleaderCRISPR-mut2-GFP plasmid, respectively. d CRISPR2 leader mutation affect GFP intensity. e GFP ﬂuorescence in PA14 WT, ΔphrS, ΔphrS/pJT-phrScmut, and ΔphrS/pJT-phrScmut/p-phrS strains transformed with pleaderCRISPR-GFP plasmid without (upper) or with (lower) BCM. f qPCR assay for GFP levels without or with BCM for PA14 WT, ΔphrS, ΔphrS/pJT-phrScmut, and ΔphrS/pJT-phrScmut/p-phrS strains transformed with pleaderCRISPR-GFP plasmid. g Western blot analysis of GFP in the PA14 WT, ΔphrS, ΔphrS/pJT-phrScmut, and ΔphrS/pJT-phrScmut/p-phrS strains transformed with pleaderCRISPR-GFP plasmid. Discussion CRISPR-Cas systems are discovered throughout diverse microbes, empowering these microorganisms with unique mechanisms for adaptive immunity3. Through a high-throughput approach, we identiﬁed sRNA PhrS as a regulator of CRISPR-Cas functionality to stimulate CRISPR loci transcription. sRNAs possess an impressive effect on bacterial behaviors through a variety of TURE COMMUNICATIONS \| (2019) 10:3728 \| https://doi.org/10.1038/s41467-019-11695-8 \| www.nature.com/naturecommunications 8 NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 aeruginosa SM4386 type I-E CRISPR-Cas system 5′ 3′ 3′ 5′ PhrS leaderCRISPR1 Hybridization Energy –17.8 kcal/mol Multiple types CRISPR-Cas adaptive immunity (I-C, I-E, I-F) No PhrS PhrS Induction Premature termination Anti-termination Rho Rho RNAP Leader 5' sRNA PhrS Growing pre-crRNA RNAP Repress CRISPR locus Stimulate CRISPR locus d pre-crRNA biogenesis pre-crRNA biogenesis s 5′- 3′ 3′ 5′ sRNA-RNA interaction b b ns 0 400 800 1000 200 600 β-gal (RFU/s/OD600) β-gal (RFU/s/OD600) β-gal (RFU/s/OD600) CRISPR1-lacZ WT, –BCM ΔphrS, –BCM WT, +BCM ΔphrS, +BCM 0 400 800 1000 200 600 0 400 800 1000 200 600 CRISPR2-lacZ CRISPR1-lacZ ns ns d c Type I-C Type I-E Type I-F d d b c β-gal (RFU/s/OD600) β-gal (RFU/s/OD600) β-gal (RFU/s/OD600) 0 200 400 500 100 300 0 200 600 800 400 0 200 600 800 400 c CRISPR1-lacZ CRISPR1-lacZ CRISPR2-lacZ WT/pJT19 WT/pJT-phrS Type I-C Type I-E Type I-F Fig. 7 PhrS-mediated anti-termination in CRISPR loci of multiple types CRISPR-Cas systems is a general phenomenon. a IntaRNA (Freiburg RNA tools) prediction of PhrS sRNA interaction with 5ʹUTR of CRISPR locus in multiple types CRISPR-Cas system. b CRISPR-lacZ activity for multiple types CRISPR-Cas system in WT or ΔphrS strain with or without BCM. c CRISPR-lacZ activity for multiple types CRISPR-Cas system in WT without PhrS overexpression, after PhrS overexpression. d Proposed model of PhrS-mediated regulation of CRISPR-Cas adaptive immunity. Bacterial small RNA PhrS inhibits the Rho- dependent termination to promote the expression of CRISPR loci, expanding function of sRNA in activating CRISPR-Cas adaptive immunity. Results are shown with mean ± SEM from three independent experiments. P < 0.01, P < 0.05, one-way ANOVA plus Tukey test (Supplementary data 3). The PCR products were recovered and cloned into a pMD-19 vector (Takara) for sequencing. understanding of their biological action in bacteria, but whether or not other classes of sRNAs or global sRNAs play a pivotal role in regulating CRISPR-Cas systems, especially controlling cas operon expression, remains unfolded. In addition, our study reveals sRNA-mediated control of CRISPR-Cas system via RNA–RNA interaction, indicating that other RNA modiﬁcations at CRISPR loci, such as epigenetic modiﬁcations (M5C or M6A), might also play a part in the transcriptional or post transcrip- tional control of CRISPR-Cas systems, warranting further investigation. Plasmid retention and transformation of efﬁciency assay. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 a 5′ 3′ 3′ A CACGAG 5′ PhrS leaderCRISPR1 P. aeruginosa ST277 type I-C CRISPR-Cas system Hybridization Energy –12.1 kcal/mol 5′ 3′ 3′ 5′ PhrS leaderCRISPR2 P. aeruginosa PA14 type I-F CRISPR-Cas system Hybridization Energy –17.9 kcal/mol P. aeruginosa SM4386 type I-E CRISPR-Cas system 5′ 3′ 3′ 5′ PhrS leaderCRISPR1 Hybridization Energy –17.8 kcal/mol b * ns 0 400 800 1000 200 600 β-gal (RFU/s/OD600) β-gal (RFU/s/OD600) β-gal (RFU/s/OD600) β-gal (RFU/s/OD600) β-gal (RFU/s/OD600) β-gal (RFU/s/OD600) CRISPR1-lacZ WT, –BCM ΔphrS, –BCM WT, +BCM ΔphrS, +BCM 0 400 800 1000 200 600 0 400 800 1000 200 600 CRISPR2-lacZ CRISPR1-lacZ ns ns Multiple types CRISPR-Cas adaptive immunity (I-C, I-E, I-F) No PhrS PhrS Induction Premature termination Anti-termination Rho Rho RNAP Leader 5' sRNA PhrS Growing pre-crRNA RNAP Repress CRISPR locus Stimulate CRISPR locus d 0 200 400 500 100 300 0 200 600 800 400 0 200 600 800 400 c CRISPR1-lacZ CRISPR1-lacZ CRISPR2-lacZ WT/pJT19 WT/pJT-phrS Type I-C pre-crRNA biogenesis pre-crRNA biogenesis s 5′- 3′ 3′ 5′ sRNA-RNA interaction Type I-E Type I-F Type I-C Type I-E Type I-F Fig. 7 PhrS-mediated anti-termination in CRISPR loci of multiple types CRISPR-Cas systems is a general phenomenon. a IntaRNA (Freiburg RNA tools) prediction of PhrS sRNA interaction with 5ʹUTR of CRISPR locus in multiple types CRISPR-Cas system. b CRISPR-lacZ activity for multiple types CRISPR-Cas system in WT or ΔphrS strain with or without BCM. c CRISPR-lacZ activity for multiple types CRISPR-Cas system in WT without PhrS overexpression, after PhrS overexpression. d Proposed model of PhrS-mediated regulation of CRISPR-Cas adaptive immunity. Bacterial small RNA PhrS inhibits the Rho- dependent termination to promote the expression of CRISPR loci, expanding function of sRNA in activating CRISPR-Cas adaptive immunity. Results are shown with mean ± SEM from three independent experiments. P < 0.01, *P < 0.05, one-way ANOVA plus Tukey test a a 5′ 3′ 3′ A CACGAG 5′ PhrS leaderCRISPR1 P. aeruginosa ST277 type I-C CRISPR-Cas system Hybridization Energy –12.1 kcal/mol 5′ 3′ 3′ 5′ PhrS leaderCRISPR2 P. aeruginosa PA14 type I-F CRISPR-Cas system Hybridization Energy –17.9 kcal/mol P. Methods Bacteria and plasmids. Bacterial strains (Supplementary data 2) are derivatives of P. aeruginosa PA14, ST277 and SM4386 strains. We used E. coli DH5a or P. aeruginosa strains to construct the plasmids (Supplementary data 2). E. coli DH5a or P. aeruginosa strains were cultured in LB or LB with agar containing ampicillin (100 μg/ml, Fisher Scientiﬁc), chloramphenicol (12.5 μg/ml, Sigma), tetracycline (10 μg/ml, Sigma), kanamycin (100 μg/ml, Fisher Scientiﬁc), carbenicillin (100 μg/ ml, Fisher Scientiﬁc), gentamicin (75 μg/ml, ACROS) or bicyclomycin (BCM) (8 μg/ml, Santa Cruz). Phage isolation and plaque assay. Phages were isolated from lysogen (Supple- mentary data 2) according to Marino42. Plaque assay on bacterial lawns of PA14 WT or derivatives strains was conducted at 37 °C on LB agar (1.5%) plates with a lower percentage of LBTop agar (0.3%). Added 1 х 108 bacteria cells to 4 ml LBTop agar and transformed to LB agar plate as an even layer. The plates were spotted with 3.5 μl of each phage lysate on the lawn and grown overnight. The observed circular zones of clearing indicate the lysis of the tester strains. β-Galactosidase assay. P. aeruginosa containing lacZ reporters (Supplementary data 2) were grown for β-galactosidase assay for type I-C/-E/-F CRISPR-Cas system according to Joshua P. Ramsay43. Brieﬂy, all integrative lacZ reporter strains were electroporated and grown in LB with tetracycline at 30 °C to detect β-galactosidase assay according to Adrian G. Patterson34. The relative ﬂuorescence intensity was monitored using Bio-TeK Synergy HT Multi-Mode Microplate Reader (Bio-Tek, Winooski, VT). The plate-reader software calculated Vmax to normalize the value of RFU/s/OD600. Library screen for sRNA chimeras by qPCR. We used a pKH6 vector from Dr. Stephen Lory of Harvard Medical School22 to construct pKH6-CRISPR2 and pKH6-CRISPR1 plasmid. These plasmids were each transfected into PA14 con- taining pKH-t4rnl1, respectively. Overnight cultures of PA14 WT with pKH13- t4rnl1 (or pKH13-t4K99N) and pKH6-CRISPR2 (or pKH6-CRISPR1) plasmid were grown in LB broth with antibiotic and diluted the overnight culture to OD600 = 0.01. When OD600 = 0.5, IPTG was added for 1 h incubation and then added the L-arabinose to 20% for different time points. The cells were centrifuged at 12,000 rpm/min for RNA isolation with DNase treatment. One microgram of total RNA was converted to complementary DNA (cDNA) was synthesized with SuperScript III First-Strand Synthesis system (Invitrogen). RT-PCR were per- formed using GoTaq Green Master Mix (Promega) with speciﬁc primers Protein reporter assays. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 The default repressive state of Rho activity can be partially relieved with the alterations of nutrient or metabolic conditions. For example, Rho is active within several well-characterized E. coli riboswitch, such as thiM, which responds to intracellular levels of TPP35,37. Our data suggest that p p In summary, our work discovers a mechanism of sRNA- mediated control of CRISPR-Cas systems, showing that PhrS ﬁnely regulates anti-termination elements to activate the tran- scription of CRISPR loci that functions throughout bacterial CRISPR-Cas adaptive immunity in response to phage infection. There are now ﬁve major classes of sRNAs in bacteria18. Con- tinued identiﬁcation of diverse classes of sRNAs will expand our TURE COMMUNICATIONS \| (2019) 10:3728 \| https://doi.org/10.1038/s41467-019-11695-8 \| www.nature.com/naturecommunications 9 9 ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 The plasmid CR1- sp1 and CR2-sp1 were used to perform plasmid retention and transformation of efﬁciency assay in PA14-WT and its derived mutation strains according to Hoyland-Kroghsbo33,41. For plasmid retention assay, PA14-WT and its derived mutants were electroporated with plasmid CR1-sp1 or CR2-sp1 and cultured in LB. Colony forming units (CFU) were calculated on luria broth with agar containing antibiotic to calculate the percentage of plasmid retention. For transformation of efﬁciency assay, strains transfected with 1 μg CR1-sp1 and CR2-sp1 plasmid were shaken for 1 h at 37 °C and plated on lysogeny broth medium with antibiotic overnight. CFUs were quantiﬁed and transformation of efﬁciency was calculated as the percentage colonies transformed by CR1-sp1 and CR2-sp1 compared with untargeted plasmid. References k The samples were separated by 6% TBE-urea polyacrylamide gels. 10. Ledford, H. Five big mysteries about CRISPR’s origins. Nature 541, 280–282 (2017). 11. McGinn, J. & Marrafﬁni, L. A. Molecular mechanisms of CRISPR-Cas spacer acquisition. Nat. Rev. Microbiol. 17, 7–12 (2018). 12. Dutta, T. & Srivastava, S. Small RNA-mediated regulation in bacteria: a growing palette of diverse mechanisms. Gene 656, 60–72 (2018). g g p 13. Serganov, A. & Nudler, E. A decade of riboswitches. Cell 152, 17–24 (20 13. Serganov, A. & Nudler, E. A decade of riboswitches. Cell 1 13. Serganov, A. & Nudler, E. A decade of riboswitches. Cell 152, 17–24 (2013). 14. Waters, L. S. & Storz, G. Regulatory RNAs in bacteria. Cell 136, 615–628 (2009). 14. Waters, L. S. & Storz, G. Regulatory RNAs in bacteria. Cell 136, 615–628 (2009). 15. Sedlyarova, N. et al. sRNA-mediated control of transcription termination in E. coli. Cell 167, 111–121 e113 (2016). 16. Storz, G., Vogel, J. & Wassarman, K. M. Regulation by small RNAs in bacteria: expanding frontiers. Mol. Cell 43, 880–891 (2011). Western blotting. The samples derived from PA14 WT and its derived mutation strains were separated and transferred to nitrocellulose membranes (GE-Health- care, Pittsburgh, PA). Membranes were incubated with mouse monoclonal anti- body against GFP (Biolegend, San Diego, CA) and His-tag (ThermoFisher Scientiﬁc, Waltham, MA) at 1:5000 for overnight at 4 °C with primary antibodies44,45. After washing, adding corresponding secondary antibodies for 1.5 h. After washing ﬁve times with washing buffer, the protein bands were visualized by chemiluminescence. 17. Sonnleitner, E., Romeo, A. & Blasi, U. Small regulatory RNAs in Pseudomonas aeruginosa. RNA Biol. 9, 364–371 (2012). 18. Hor, J., Gorski, S. A. & Vogel, J. Bacterial RNA biology on a genome scale. Mol. Cell 70, 785–799 (2018). 19. D’Heygere, F., Rabhi, M. & Boudvillain, M. Phyletic distribution and conservation of the bacterial transcription termination factor Rho. Microbiology 159, 1423–1436 (2013). 20. Grylak-Mielnicka, A., Bidnenko, V., Bardowski, J. & Bidnenko, E. Transcription termination factor Rho: a hub linking diverse physiologica processes in bacteria. Microbiology 162, 433–447 (2016). RNA immunoprecipitation (RIP). Anti-His antibody was used for RIP assay. One microgram of His-Rho were used to pull down the RNA of CRISPR leader incu- bated with different concentration of PhrS. References k 1. Makarova, K. S. et al. An updated evolutionary classiﬁcation of CRISPR-Cas systems. Nat. Rev. Microbiol. 13, 722–736 (2015). 2. Hille, F. et al. The biology of CRISPR-Cas: backward and forward. Cell 172, 1239–1259 (2018). 3. Jackson, S. A. et al. CRISPR-Cas: adapting to change. Science 356, aal5056 (2017). Northern blot analysis. Five microgram of total RNA was separated by 6% TBE- urea polyacrylamide gels in 1x TBE and then transferred to Hybond-XL mem- branes (GE-Healthcare, Pittsburgh, PA). Using ultraviolet to cross-linking and hybridizing with gamma32P-ATP labeled oligonucleotide probes (Supplementary table 1). After washing three times to remove unwanted probe, the bands were detected. 4. Barrangou, R. et al. CRISPR provides acquired resistance against viruses in prokaryotes. Science 315, 1709–1712 (2007). 5. Marrafﬁni, L. A. & Sontheimer, E. J. CRISPR interference limits horizontal gene transfer in staphylococci by targeting DNA. Science 322, 1843–1845 (2008). 6. Adli, M. The CRISPR tool kit for genome editing and beyond. Nat. Commun. 9, 1911 (2018). sRNA PhrS synthesis. PhrS templates with T7 promoter sequence (Supplemen- tary table 2) were ampliﬁed by RT-PCR. PhrS sRNA was transcribed with MEGAscript T7 kit according to manufacturer’s protocols with TURBO DNase treatment. 7. Jeon, Y. et al. Direct observation of DNA target searching and cleavage by CRISPR-Cas12a. Nat. Commun. 9, 2777 (2018). 8. Lewis, K. M. & Ke, A. Building the class 2 CRISPR-Cas arsenal. Mol. Cell 65, 377–379 (2017). 9. Strecker, J. et al. Engineering of CRISPR-Cas12b for human genome editing. Nat. Commun. 10, 212 (2019). In vitro transcription. For transcription reaction, the procedure was previously described by Sedlyarova et al.15 with modiﬁcation. Brieﬂy, the initial elongation complex was formed with 75 nM of corresponding transcription templates (Sup- plementary Table 2) and 100 nM RNA polymerase (RNAP) holoenzyme in 100 μl of 40 mM Tris-HCl pH = 8.0; 20 mM MgCl2; 50 mM NaCl; 0.003% IGEPAL; 5 mM β-mercaptoethanol with 2 μl RNase inhibitor (Takara Bio USA, Mountain View, CA). Transcription reaction mix (25 μM GTP, UTP, ATP-P-32) was performed at 22 °C for 5 min incubation and then mixed with 1 μM Rho with NusG proteins with or without the PhrS sRNA or bicyclomycin. Heat the mixture containing 1 mM ATP and 100 μM other NTPs to transcribe at 37 °C with 10 min and stop the reaction with 1x TBE, 8 M Urea, 20 mM EDTA, 0.025% xylencyanol, 0.025% bromophenol blue at 95 °C. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 Received: 27 December 2018 Accepted: 30 July 2019 GFP assay in the liquid culture for measure GFP intensity: corresponding strains containing the reporters were cultured to OD600 = 1.0 and diluted 20-fold to measure GFP in SPECTRAmas GEMINI-XS Spectroﬂuorometer (Molecular Devices, San Jose, CA). RNA isolation and qPCR. RNA puriﬁcation with DNAase I digestion was per- formed by the Direct-zolTM RNA MiniPrep kit (ZYMO RESEARCH, Irvine, CA). cDNA was synthesized using the High Capacity cDNA Reverse Transcription Kit (ThermoFisher Scientiﬁc, Waltham, MA). qPCR was analyzed by Maxima SYBR Green qPCR Master Mix (ThermoFisher Scientiﬁc, Waltham, MA) with gene- speciﬁc primers (Supplementary table 1). ARTICLE ARTICLE References k Further information on research design is available in the Nature Research Reporting Summary linked to this article. 29. Sun, J. et al. A novel antisense long noncoding RNA within the IGF1R gene locus is imprinted in hematopoietic malignancies. Nucl. Acids Res. 42, 9588–9601 (2014). References k The reactants were washed three times with RIPA buffer (150 mM NaCl, 50 mM Tris, 0.5% sodium deoxycholate, 0.1% SDS, 1% NP-40 with RNase inhibitor [Takara Bio USA, Mountain View, CA]) and then washed twice with another washing buffer (1 M NaCl, 50 mM Tris, 0.5% sodium deoxycholate, 0.1% SDS, 1% NP-40). RNA isolation was used Direct-zolTM RNA MiniPrep kit (ZYMO RESEARCH, Irvine, CA). 21. Mitra, P., Ghosh, G., Hafeezunnisa, M. & Sen, R. Rho protein: roles and mechanisms. Annu Rev. Microbiol. 71, 687–709 (2017). 22. Han, K., Tjaden, B. & Lory, S. GRIL-seq provides a method for identifying direct targets of bacterial small regulatory RNA by in vivo proximity ligation. Nat. Microbiol. 2, 16239 (2016). 23. Lu, P., Wang, Y., Hu, Y. & Chen, S. RgsA, an RpoS-dependent sRNA, negatively regulates rpoS expression in Pseudomonas aeruginosa. Microbiology 164, 716–724 (2018). Microbiology 164, 716–724 (2018). Biotinylated RNA pull-down. We used the Biotin RNA labeling mix (Sigma) and T7 RNA polymerase to get the Biotin-CRISPR leader transcripts. Biotin-CRISPR leader incubated with different concentration of PhrS were adsorbed onto strep- tavidin magnetic beads and added His-Rho protein to incubation at 4 °C for 6 h. After washing ﬁve times in RIP buffer (150 mM KCl, 25 mM Tris-HCl (pH 7.4), 0.5 mM DTT. 0.5% NP-40, and protease inhibitors), the eluted samples were detected by western blot. 24. Gómez‐Lozano, M., Marvig, R. L., Molin, S. & Long, K. S. Genome‐wide identiﬁcation of novel small RNAs in Pseudomonas aeruginosa. Environ. Microbiol. 14, 2006–2016 (2012). 25. Wright, A. V. et al. Structures of the CRISPR genome integration complex. Science 357, 1113–1118 (2017). 26. Nunez, J. K., Bai, L., Harrington, L. B., Hinder, T. L. & Doudna, J. A. CRISPR immunological memory requires a host factor for speciﬁcity. Mol. Cell 62, 824–833 (2016). 27. Sonnleitner, E. et al. Detection of small RNAs in Pseudomonas aeruginosa by RNomics and structure-based bioinformatic tools. Microbiology 154, 3175–3187 (2008). Statistical analysis. Values were obtained from three independent experiments, shown as mean ± SEM. P-values were derived with GraphPad (GraphPad Software, LaJolla, CA) using one-way analysis of variance (ANOVA) plus Tukey test. 28. Sonnleitner, E. et al. The small RNA PhrS stimulates synthesis of the Pseudomonas aeruginosa quinolone signal. Mol. Microbiol. 80, 868–885 (2011). Reporting summary. Further information on research design is available in the Nature Research Reporting Summary linked to this article. Reporting summary. Methods Measuring GFP ﬂuorescence: P. aeruginosa PA14 con- taining GFP reporter were cultured on LB with agar for 20 h, which was imaged at GFP ﬂuorescence mode by IVIS XRII system (PerkinElmer, Waltham, MA). 10 NATURE COMMUNICATIONS \| (2019) 10:3728 \| https://doi.org/10.1038/s41467-019-11695-8 \| www.nature.com/naturecommunications Author contributions P.L., M.W. and J.X.J designed experiments and wrote the manuscript. P.L., Q.P., Q.W. and J.S. designed and performed most of the experiments. P.L., M.W. and J.X.J. analyzed data. C.Z., B.W., S.Q., R.L., P.G., G.L., Z.W., Z.C. and L.L. advised on experimental design and manuscript preparation. 35. Bastet, L. et al. Translational control and Rho-dependent transcription termination are intimately linked in riboswitch regulation. Nucl. Acids Res. 45, 7474–7486 (2017). 36. Hollands, K. et al. Riboswitch control of Rho-dependent transcription termination. Proc. Natl Acad. Sci. USA 109, 5376–5381 (2012). ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-019-11695-8 Chen of Wuhan institute of Virology (Chinese Academy of Science) for providing information of sRNA library information, and Dr. Sergei Nechaev and Dr. Catherine Brissette of University of North Dakota for critical reading. This work was supported by National Institutes of Health Grants AI101973-01, R01AI109317-01A1, and R01AI0138203-01 to M.W., UND Post-Doc Pilot Grant; this work was also supported by the Key Program of National Nature Science Foundation of China (81530063) to Jianxin Jiang. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. 31. Haurwitz, R. E., Jinek, M., Wiedenheft, B., Zhou, K. & Doudna, J. A. Sequence- and structure-speciﬁc RNA processing by a CRISPR endonuclease. Science 329, 1355–1358 (2010). 32. Samson, J. E., Magadan, A. H., Sabri, M. & Moineau, S. Revenge of the phages: defeating bacterial defences. Nat. Rev. Microbiol. 11, 675–687 (2013). R01AI0138203-01 to M.W., UND Post-Doc Pilot Grant; this work was also supported by the Key Program of National Nature Science Foundation of China (81530063) to Jianxin Jiang. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. 33. Hoyland-Kroghsbo, N. M. et al. Quorum sensing controls the Pseudomonas aeruginosa CRISPR-Cas adaptive immune system. Proc. Natl Acad. Sci. USA 114, 131–135 (2017). 34. Patterson, A. G. et al. Quorum sensing controls adaptive immunity through the regulation of multiple CRISPR-Cas systems. Mol. Cell 64, 1102–1108 (2016). Competing interests: The authors declare no competing interests. 39. Papenfort, K., Bouvier, M., Mika, F., Sharma, C. M. & Vogel, J. Evidence for an autonomous 5’ target recognition domain in an Hfq-associated small RNA. Proc. Natl Acad. Sci. USA 107, 20435–20440 (2010). Reprints and permission information is available online at http://npg.nature.com/ reprintsandpermissions/ 40. Beisel, C. L. & Storz, G. The base-pairing RNA spot 42 participates in a multioutput feedforward loop to help enact catabolite repression in Escherichia coli. Mol. Cell 41, 286–297 (2011). Peer review information: Nature Communications thanks the anonymous reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available. Peer review information: Nature Communications thanks the anonymous reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available. 41. Lin, P. et al. CdpR Inhibits CRISPR-cas adaptive immunity to lower anti-viral defense while avoiding self-reactivity. iScience 13, 55–68 (2019). Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. 42. Marino, N. D. et al. Discovery of widespread type I and type V CRISPR-Cas inhibitors. Science 362, 240–242 (2018). 43. Ramsay, J. High-throughput ß-galactosidase and ß-glucuronidase assays using ﬂuorogenic substrates. Bio-Protoc. 3, e827–e827 (2013). Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/ licenses/by/4.0/. g 44. Li, R. et al. Type I CRISPR-Cas targets endogenous genes and regulates virulence to evade mammalian host immunity. Cell Res. 26, 1273–1287 (2016). 45. Wu, M., Stockley, P. G. & Martin, W. J. 2nd An improved western blotting technique effectively reduces background. Electrophoresis 23, 2373–2376 (2002). Supplementary Information accompanies this paper at https://doi.org/10.1038/s41467- 019-11695-8. p y 38. Hoe, C. H., Raabe, C. A., Rozhdestvensky, T. S. & Tang, T. H. Bacterial sRNAs: regulation in stress. Int J. Med. Microbiol. 303, 217–229 (2013). Competing interests: The authors declare no competing interests. Data availability 30. Zwiefka, A., Kohn, H. & Widger, W. R. Transcription termination factor rho: the site of bicyclomycin inhibition in Escherichia coli. Biochemistry 32, 3564–3570 (1993). Authors conﬁrm that all data and materials in the study can be obtained from the corresponding author (M.W.) upon reasonable request. 11 NATURE COMMUNICATIONS \| (2019) 10:3728 \| https://doi.org/10.1038/s41467-019-11695-8 \| www.nature.com/naturecommunications Additional information 37. Bastet, L., Turcotte, P., Wade, J. T. & Lafontaine, D. A. Maestro of regulation: riboswitches orchestrate gene expression at the levels of translation, transcription and mRNA decay. RNA Biol. 15, 679–682 (2018). Supplementary Information accompanies this paper at https://doi.org/10.1038/s41467- 019-11695-8. Supplementary Information accompanies this paper at https://doi.org/10.1038/s41467- 019-11695-8. Acknowledgements g We thank Dr. Stephen Lory of Harvard Medical School to give pKH6, pKH13, or lacZ vectors, Dr. Evgeny Nudler of New York University School of Medicine for providing Rho and NusG proteins, Dr. Elisabeth Sonnleitner of University of Vienna for providing pJT-phrS, pJT-phrSΔcreg, pJT-phrScreg, and pJT-phrScmut, Dr. Alan R. Davidson of University of Toronto for providing Pseudomonas aeruginosa SMC4386, Dr. Shiyun © The Author(s) 2019 NATURE COMMUNICATIONS \| (2019) 10:3728 \| https://doi.org/10.1038/s41467-019-11695-8 \| www.nature.com/naturecommunications 12 NATURE COMMUNICATIONS \| (2019) 10:3728 \| https://doi.org/10.1038/s41467-019-11695-8 \| www.nature.com/naturecommunicatio
https://openalex.org/W2982649480	https://publications.goettingen-research-online.de/bitstream/2/92026/3/Anker_et_al-2019-ESC_Heart_Failure.pdf	English	null	Open access efforts begin to bloom: <i>ESC Heart Failure</i> gets full attention and first impact factor	ESC heart failure	2,019	cc-by	4,888	Abstract In 2014, the Heart Failure Association (HFA) of the European Society of Cardiology (ESC) founded the ﬁrst open access journal focusing on heart failure, ESC Heart Failure (ESC-HF). In the ﬁrst 5 years, in ESC-HF we published more than 450 articles. Through ESC-HF, the HFA gives room for heart failure research output from around the world. A transfer process from the European Journal of Heart Failure to ESC-HF has also been installed. As a consequence, in 2018 ESC-HF received 289 submis- sions, and published 148 items (acceptance rate 51%). The journal is listed in Scopus since 2014 and on the PubMed website since 2015. In 2019, we received our ﬁrst impact factor from ISI Web of Knowledge / Thomson-Reuters, which is 3.407 for 2018. This report reviews which papers get best cited. Not surprisingly, many of the best cited papers are reviews and facts & num- bers mini reviews, but original research is also well cited. Received: 27 November 2018; Revised: 24 May 2019; Accepted: 30 July 2019 Correspondence to: Stefan D. Anker, Division of Cardiology and Metabolism, Department of Cardiology, Charité, Campus Virchow-Klinikum, Augustenburger Platz 1, D-13353 Berlin, Germany. Tel: +49-30-450 553463; Fax: +49-30-450 553951. Email: s.anker@cachexia.de Besides PubMed, one can ﬁnd ESC-HF articles via Google Scholar since 2014. ESC-HF is also listed in Scopus since 2014 (see http://www.scopus.com—this is the search system of Elsevier). In the summer of 2018, we received conﬁrmation of our acceptance by ISI Web of Knowledge (Thomson- Reuters) to be considered for assessment of citations and impact factor retrospectively since the start of the journal for their 2018 Journal Citation Report to come out mid-2019. This is where things got interesting and several people performed analyses as to what our impact factor will be. At the HFA congress, we mentioned that it should be at least 3.3—we can now say that it is 3.407 for 2018. We are listed in the ﬁeld ‘Cardiac & Cardiovascular Systems’ at position 49 of 136. This is a great success for the HFA and the editorial board together with our reviewers and for all of the heart failure research community. We have established an open access journal in the ﬁeld of heart failure research. A very big thank you to all of you who contribute to the journal’s progress and success and of course also to our many readers for whom we are doing this. MINI REVIEW MINI REVIEW ESC HEART FAILURE Published online in Wiley Online Libr ESC Heart Failure 2019; 6: 903–908 Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/ehf2.12540 ESC Heart Failure 2019; 6: 903–908 Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/ehf2.12540 ; © 2019 The Authors. ESC Heart Failure published by John Wiley & Sons Ltd on behalf of the European Society of Cardiology. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. Open access efforts begin to bloom: ESC Heart Failure gets full attention and ﬁrst impact factor Stefan D. Anker1,2,3, Stephan von Haehling4,5 and Zoltan Papp6 1Division of Cardiology and Metabolism, Department of Cardiology, Charité, Campus Virchow-Klinikum, Augustenburger Platz 1, D-13353, Berlin, Germany 2Berlin-Brandenburg Center for Regenerative Therapies (BCRT), Berlin, Germany; 3German Centre for Cardiovascular Research (DZHK), Berlin, Germany; 4Department of Cardiology and Pneumology, Heart Center Göttingen, University of Göttingen Medical Center, Georg-August-University, Göttingen, Germany; 5German Centre for Cardiovascular Medicine (DZHK), Göttingen, Germany; 6Division of Clinical Physiology, Department of Cardiology, Faculty of Medicine, University of Debrecen, Debrecen, Hungary Received: 27 November 2018; Revised: 24 May 2019; Accepted: 30 July 2019 *Correspondence to: Stefan D. Anker, Division of Cardiology and Metabolism, Department of Cardiology, Charité, Campus Virchow-Klinikum, Augustenburger Platz 1, D-13353 Berlin, Germany. Tel: +49-30-450 553463; Fax: +49-30-450 553951. Email: s.anker@cachexia.de © 2019 The Authors. ESC Heart Failure published by John Wiley & Sons Ltd on behalf of the European Society of Cardiology. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution an original work is properly cited. Failure published by John Wiley & Sons Ltd on behalf of the European Society of Cardiology. der the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the Abstract Does the world need an open access journal focusing on heart failure, some asked in 2014 when ESC Heart Failure (for short: ESC-HF) had just been launched during the annual congress of the Heart Failure Association (HFA) of the European Society of Cardiology (ESC). Today, we are looking back at ﬁve years of activity with more than 450 published reviews, editorials, facts & numbers mini reviews, letters, and, most importantly, original research articles. In the ﬁrst year, it was not always easy to ﬁll the journal, but today, this is no longer an issue. In 2018, we have received 289 submissions and published 148 items—an acceptance rate of just above 50%. In 2015, our publishers Wiley and we managed to get ESC- HF listed on the PubMed website. This helped our visibility and resulted in a strong increase in submissions following this. For the aﬁcionados of these processes, we know that only a Medline acceptance truly guarantees a full PubMed listing— but our Medline application (submitted March 2017) was suc- cessful, and full MEDLINE indexing started with the 2017 volume. p No. Abstract First author Title Type Times cited Reference 2016 Jujo K Randomized pilot trial comparing tolvaptan with furosemide on renal and neurohumoral effects in acute heart failure Original research articles 29 1 2016 Konishi M Heart failure epidemiology and novel treatments in Japan: facts and numbers Editorials 23 2 2017 Springer J Muscle wasting and sarcopenia in heart failure and beyond: update 2017 Reviews 21 3 2016 Nagarajan V Obesity paradox in heart failure: a heavy matter Reviews 16 4 2016 Arrigo M Effect of precipitating factors of acute heart failure on readmission and long-term mortality Original research articles 16 5 2016 Nunez J Left ventricular ejection fraction recovery in patients with heart failure treated with intravenous iron: a pilot study Short communication 16 6 2017 Luedde M Heart failure is associated with depletion of core intestinal microbiota Original research articles 16 7 2017 Riley JP Palliative care in heart failure: facts and numbers Editorial 15 8 2017 Saitoh M Anorexia, functional capacity, and clinical outcome in patients with chronic heart failure: results from the studies investigating co-morbidities aggravating heart failure (SICA-HF) Original research articles 15 9 2016 Sotiropoulos K Red cell distribution width and mortality in acute heart failure patients with preserved and reduced ejection fraction Original research articles 14 10 2017 Delepaul B Who are patients classiﬁed within the new terminology of heart failure from the 2016 ESC guidelines? Abstract Original research articles 14 11 2016 Hayashi T Subclinical hypothyroidism is an independent predictor of adverse cardiovascular outcomes in patients with acute decompensated heart failure Original research articles 13 12 2017 Barkhudaryan A Cardiac muscle wasting in individuals with cancer cachexia Original research articles 11 13 2016 Sato A Associations of dipeptidyl peptidase-4 inhibitors with mortality in hospitalized heart failure patients with diabetes mellitus Original research articles 10 14 2017 Amin A On admission serum sodium and uric acid levels predict 30 day rehospitalization or death in patients with acute decompensated heart failure Original research articles 10 15 2016 Yoshihisa A The CHA(2)S(2)-VASC score as a predictor of high mortality in hospitalized heart failure patients Original research articles 9 16 2016 Hoshida S Age- and sex-related differences in diastolic function and cardiac dimensions in a hypertensive population Original research articles 9 17 2017 Morishita T Association between matrix metalloproteinase-9 and worsening heart failure events in patients with chronic heart failure Original research articles 9 18 2017 Alma LJ Shared biomarkers between female diastolic heart failure and pre-eclampsia: a systematic review and meta-analysis Review 9 19 2017 Trippel TD Ghrelin and hormonal markers under exercise training in patients with heart failure with preserved ejection fraction: results from the EX-DHF pilot study Original research articles 9 20 2017 Khan MS Renin-angiotensin blockade in heart failure with preserved ejection fraction: a systematic review and meta-analysis Reviews 8 21 2017 Moeckel M The role of procalcitonin in acute heart failure patients Review 8 22 2018 Seropian IM Neutrophil-to-lymphocyte ratio and platelet-to-lymphocyte ratio as predictors of survival after heart transplantation Original research articles 8 23 2016 Searle J Acute heart failure facts and numbers: acute heart failure populations Editorial 7 24 2016 Thomsen MM Varying effects of recommended treatments for heart failure with reduced ejection fraction: meta-analysis of randomized controlled trials in the ESC and ACCF/AHA guidelines Reviews 7 25 2016 Savage HO Development and validation of a novel non-contact monitor of nocturnal respiration for identifying sleep-disordered breathing in patients with heart failure Original research articles 7 26 2016 Aleksova N Barriers to goals of care discussions with hospitalized patients with advanced heart failure: feasibility and performance of a novel questionnaire Original research articles 7 27 2017 Toma M Differentiating heart failure phenotypes using sex-speciﬁc transcriptomic and proteomic biomarker panels Original research articles 7 28 S.D. Anker et 4 904 S.D. Anker et al. Abstract DOI: 10.1002/ehf2.12540 ESC Heart Failure 2019; 6: 903–908 Open access efforts begin to bloom: ESC Heart Failure gets full attention and ﬁrst impact factor 905 No. First author Title Type Times cited Reference 2017 Yoshihisa A Associations between acylcarnitine to free carnitine ratio and adverse prognosis in heart failure patients with reduced or preserved ejection fraction Short communications 7 29 2017 Jaarsma T Sexual function of patients with heart failure: facts and numbers Editorial 7 30 2017 Theidel U Budget impact of intravenous iron therapy with ferric carboxymaltose in patients with chronic heart failure and iron deﬁciency in Germany Original research articles 7 31 2018 Oehman J Focused echocardiography and lung ultrasound protocol for guiding treatment in acute heart failure Original research articles 7 32 2018 Pascual-Figal D Rationale and design of transition: a randomized trial of pre-discharge vs. DOI: 10.1002/ehf2.12540 ESC Heart Failure 2019; 6: 903–908 Abstract Anker et al. is ﬁnding reviewers for your submissions. As many have asked in this context, we do not mind if you inform us about previ- ous reviewer comments in your submission cover letters, but we then also request that you add response comments in your submission package for us to consider. But of course, we will trigger also our own reviewer activities. And for this, a longer rather than shorter list of suggested reviewers would also be very helpful if you want to technically support your submission to ESC-HF. For the aﬁcionados of technical details, we also would like to give you the details as to how our impact factor came about. The Impact Factor 2018 is calculated based on the cita- tions in 2018 to items published in ESC-HF in the two years before that, that is, in 2016 and 2017. This number (which is 402) is then divided by the sum of full articles (i.e. reviews and original papers) published in 2016/2017. In 2016, we pub- lished 35 full papers, and in 2017, there were 83 papers. The sum of these two years was 118, and considering the above mentioned 402 citations to these papers in the literature in 2018, our impact factor was calculated to be 3.407. We hope that together we can keep the journal at a good quality standard, even if more papers are published in the future. We strongly suspect that submissions will further increase. From 2018 to 2019, the projected growth in submissions is >30%. And we apologize already, if some- times there are still some delays. We do our best with the staff in Berlin and Göttingen and Debrecen. And if you sometimes feel that we are like the PLoS-ONE of heart failure, then indeed this is what we wish to be, the major publication threshold being that the work under consider- ation is ‘scientiﬁcally sound’. But there are two major differ- ences, which is that we have the support of a strong society and that a good proportion of the papers we accept also comes out in print. As an indication of which papers published in ESC-HF have received most citations, we include a table that is based on ISI Web of Knowledge, as assessed on 3 August 2019, to show all articles published in 2016–2018 in ESC-HF with >5 citations so far. Abstract Not surprisingly, many of the best cited papers are re- views and facts & numbers mini reviews, but original research is also well cited. (Table 1). There is a lot of work where people say that nothing is really wrong with it. It may just be a little small in study size, lacking the extra control group or extra biomarker tests that some would have hoped for or that simply, for these and many other reasons, may not have priority for the lim- ited publication space in the premier cardiology and heart failure specialist journals. We want to give room for the many other good pieces of heart failure research from around the world. This is also why we have installed a transfer process from the European Journal of Heart Failure to ESC-HF, which is now the source of half of all submis- sions to our journal. At last, some also have asked us about our plans and am- bitions for future impact factors of ESC-HF. We are not striving to get the top of heart failure research—we know that they primarily published elsewhere. We are aiming for an impact factor above 4, but we are committed to es- sentially unlimited growth of the journal. We are aiming for an acceptance rate of 40–50%. We believe that the main job of a journal like ours is to give room for discussion and for any interesting research. We are aiming to get your research published, let you keep the copyright, and make the research available for the whole world to read at no cost. This is the open access model at work for heart fail- ure. This is ESC Heart Failure for you. To manage the increasing number of submissions while maintaining a reasonable time from submission to review de- cision and Early View publication, we have substantially in- creased the size of the editorial board. If some of you feel qualiﬁed and are interested to help in this, please let us know. The biggest problem we face for a timely publication process Abstract post-discharge initiation of sacubitril/valsartan Original research articles 7 33 2016 Meems LMG Plasma calcidiol, calcitriol, and parathyroid hormone and risk of new onset heart failure in a population-based cohort study Original research articles 6 34 2016 Ahmed MB Higher risk for incident heart failure and cardiovascular mortality among community- dwelling octogenarians without pneumococcal vaccination Original research articles 6 35 2016 Yamamoto E The clinical signiﬁcance of plasma neopterin in heart failure with preserved left ventricular ejection fraction Short communications 6 36 2017 Pose A Beneﬁt of tolvaptan in the management of hyponatraemia in patients with diuretic- refractory congestive heart failure: the SEMI-SEC project Original research articles 6 37 2017 Silvetti S Rehospitalization after intermittent levosimendan treatment in advanced heart failure patients: a meta-analysis of randomized trials Original research articles 6 38 2017 Ancion A Serum albumin level and hospital mortality in acute non-ischemic heart failure Original research articles 6 39 2017 Cohen-Solal A Beta blocker dose and markers of sympathetic activation in heart failure patients: interrelationships and prognostic signiﬁcance Original research articles 6 40 2017 Salem K Gender-adjusted and age-adjusted economic inpatient burden of congestive heart failure: cost and disability-adjusted life-year analysis Original research articles 6 41 2017 Jain A The renal-cardiac connection in subjects with preserved ejection fraction: a population based study Original research articles 6 42 2017 Lancellotti P Protocol update and preliminary results of EACVI/HFA Cardiac Oncology Toxicity (COT) Registry of the European Society of Cardiology ESC and HFA paper 6 43 2017 Peled Y The impact of gender mismatching on early and late outcomes following heart transplantation Original research articles 6 44 2017 Koyama S Dynamic changes of serum microRNA-122-5p through therapeutic courses indicates amelioration of acute liver injury accompanied by acute cardiac decompensation Original research articles 6 45 2018 Porto C Association between vitamin D deﬁciency and heart failure risk in the elderly Original research articles 6 46 2018 Ferreira JP Rationale of the ﬁbrotargets study designed to identify novel biomarkers of myocardial ﬁbrosis Original research articles 6 47 2018 Mustroph J Empagliﬂozin reduces Ca/calmodulin-dependent kinase II activity in isolated ventricular cardiomyocytes Short communication 6 48 2018 Smedema JP Right ventricular involvement and the extent of left ventricular enhancement with magnetic resonance predict adverse outcome in pulmonary sarcoidosis Original research articles 6 49 2018 Yoshihisa A Liver ﬁbrosis score predicts mortality in heart failure patients with preserved ejection fraction Original research articles 6 50 Table 1 (continued) DOI: 10.1002/ehf2.12540 ESC Heart Failure 2019; 6: 903–908 906 S.D. References Pascual-Figal D, Wachter R, Senni M, Belohlavek J, Noè A, Carr D, Butylin D. Rationale and design of TRANSITION: a randomized trial of pre-discharge vs. post-discharge initiation of sacubitril/ valsartan. ESC Heart Fail 2018; 5: 327–336. 22. Möckel M, Searle J, Maisel A. The role of procalcitonin in acute heart failure pa- tients. ESC Heart Fail. 2017; 4: 203–208. 23. Seropian IM, Romeo FJ, Pizarro R, Vulcano NO, Posatini RA, Marenchino RG, Berrocal DH, Belziti CA. Neutro- phil-to-lymphocyte ratio and platelet- to-lymphocyte ratio as predictors of survival after heart transplantation. ESC Heart Fail. 2018; 5: 149–156. 13. Barkhudaryan A, Scherbakov N, Springer J, Doehner W. Cardiac muscle wasting in individuals with cancer ca- chexia. ESC Heart Fail. 2017; 4: 458–467. 34. Meems LMG, Brouwers FP, Joosten MM, Lambers Heerspink HJ, Zeeuw D, Bakker SJL, Gansevoort RT, Gilst WH, Harst P, Boer RA. Plasma calcidiol, calcitriol, and parathyroid hormone and risk of new onset heart failure in a population- based cohort study. ESC Heart Fail. 2016; 3: 189–197. 14. Sato A, Yoshihisa A, Kanno Y, Takiguchi M, Miura S, Shimizu T, Nakamura Y, Yamauchi H, Owada T, Sato T, Suzuki S, Oikawa M, Yamaki T, Sugimoto K, Kunii H, Nakazato K, Suzuki H, Saitoh S-i, Takeishi Y. Associations of dipeptidyl peptidase-4 inhibitors with mortality in hospitalized heart failure patients with diabetes mellitus. ESC Heart Fail. 2016; 3: 77–85. 24. Searle J, Frick J, Möckel M. Acute heart failure facts and numbers: acute heart failure populations. ESC Heart Fail. 2016; 3: 65–70. 25. Thomsen MM, Lewinter C, Køber L. Varying effects of recommended treat- ments for heart failure with reduced ejection fraction: meta-analysis of ran- domized controlled trials in the ESC and ACCF/AHA guidelines. ESC Heart Fail. 2016; 3: 235–244. 35. Ahmed MB, Patel K, Fonarow GC, Mor- gan CJ, Butler J, Bittner V, Kulczycki A, Kheirbek RE, Aronow WS, Fletcher RD, Brown CJ, Ahmed A. Higher risk for inci- dent heart failure and cardiovascular mortality among community-dwelling octogenarians without pneumococcal vaccination. ESC Heart Fail. 2016; 3: 11–17. 15. Amin A, Chitsazan M, Shiukhi Ahmad Abad F, Taghavi S, Naderi N. On admis- sion serum sodium and uric acid levels predict 30 day rehospitalization or death in patients with acute decompensated heart failure. ESC Heart Fail. 2017; 4: 162–168. 26. References 4. Nagarajan V, Kohan L, Holland E, Keeley EC, Mazimba S. Obesity paradox in heart failure: a heavy matter. ESC Heart Fail. 2016; 3: 227–234. treated with intravenous iron: a pilot study. ESC Heart Fail. 2016; 3: 293–298. 1. Jujo K, Saito K, Ishida I, Furuki Y, Kim A, Suzuki Y, Sekiguchi H, Yamaguchi J, Ogawa H, Hagiwara N. Randomized pi- lot trial comparing tolvaptan with furo- semide on renal and neurohumoral effects in acute heart failure. ESC Heart Fail. 2016; 3: 177–188. 7. Luedde M, Winkler T, Heinsen FA, Rühlemann MC, Spehlmann ME, Bajrovic A, Lieb W, Franke A, Ott SJ, Frey N. Heart failure is associated with depletion of core intestinal microbi- ota. ESC Heart Fail. 2017; 4: 282–290. 5. Arrigo M, Tolppanen H, Sadoune M, Feliot E, Teixeira A, Laribi S, Plaisance P, Nouira S, Yilmaz MB, Gayat E, Mebazaa A. Effect of precipitating fac- tors of acute heart failure on readmis- sion and long-term mortality. ESC Heart Fail. 2016; 3: 115–121. 2. Konishi M, Ishida J, Springer J, Haehling S, Akashi YJ, Shimokawa H, Anker SD. Heart failure epidemiology and novel treatments in Japan: facts and numbers. ESC Heart Fail. 2016; 3: 145–151. 8. Riley JP, Beattie JM. Palliative care in heart failure: facts and numbers. ESC Heart Fail. 2017; 4: 81–87. 6. Núñez J, Monmeneu JV, Mollar A, Núñez E, Bodí V, Miñana G, García-Blas S, Santas E, Agüero J, Chorro FJ, Sanchis J, López-Lereu MP. Left ventricular ejection fraction recovery in patients with heart failure 9. Saitoh M, Dos Santos MR, Emami A, Ishida J, Ebner N, Valentova M, Bekfani T, Sandek A, Lainscak M, Doehner W, Anker SD, von Haehling S. Anorexia, functional capacity, and clinical outcome 3. Springer J, Springer JI, Anker SD. Mus- cle wasting and sarcopenia in heart fail- ure and beyond: update 2017. ESC Heart Fail 2017; 4: 492–498. DOI: 10.1002/ehf2.12540 ESC Heart Failure 2019; 6: 903–908 907 proteomic biomarker panels. ESC Heart Fail. 2017; 4: 301–311. heart failure events in patients with chronic heart failure. ESC Heart Fail. 2017; 4: 321–330. in patients with chronic heart failure: re- sults from the Studies Investigating Co- morbidities Aggravating Heart Failure (SICA-HF). ESC Heart Fail. 2017; 4: 448–457. 29. Yoshihisa A, Watanabe S, Yokokawa T, Misaka T, Sato T, Suzuki S, Oikawa M, Kobayashi A, Takeishi Y. References Associations be- tween acylcarnitine to free carnitine ra- tio and adverse prognosis in heart failure patients with reduced or pre- served ejection fraction. ESC Heart Fail. 2017; 4: 360–364. 19. Alma LJ, Bokslag A, Maas AHEM, Franx A, Paulus WJ, de Groot CJM. Shared bio- markers between female diastolic heart failure and pre-eclampsia: a systematic review and meta-analysis. ESC Heart Fail. 2017; 4: 88–98. 10. Sotiropoulos K, Yerly P, Monney P, Garnier A, Regamey J, Hugli O, Martin D, Metrich M, Antonietti J-P, Hullin R. Red cell distribution width and mortal- ity in acute heart failure patients with preserved and reduced ejection fraction. ESC Heart Fail. 2016; 3: 198–204. ; 20. Trippel TD, Holzendorf V, Halle M, Gelbrich G, Nolte K, Duvinage A, Schwarz S, Rutscher T, Wiora J, Wachter R, Herrmann-Lingen C, Duengen HD, Hasenfuβ G, Pieske B, Edelmann F. Ghrelin and hormonal markers under exercise training in patients with heart failure with preserved ejection fraction: results from the Ex-DHF pilot study. ESC Heart Fail. 2017; 4: 56–65. 30. Jaarsma T. Sexual function of patients with heart failure: facts and numbers. ESC Heart Fail. 2017; 4: 3–7. 11. Delepaul B, Robin G, Delmas C, Moine T, Blanc A, Fournier P, Roger-Rollé A, Domain G, Delon C, Uzan C, Boudjellil R, Carrié D, Roncalli J, Galinier M, Lairez O. Who are patients classiﬁed within the new terminology of heart fail- ure from the 2016 ESC guidelines? ESC Heart Fail. 2017; 4: 99–104. 31. Theidel U, Väätäinen S, Martikainen J, Soini E, Hardt T, Doehner W. Budget im- pact of intravenous iron therapy with fer- ric carboxymaltose in patients with chronic heart failure and iron deﬁciency in Germany. ESC Heart Fail. 2017; 4: 274–281. 21. Khan MS, Fonarow GC, Khan H, Greene SJ, Anker SD, Gheorghiade M, Butler J. Renin–angiotensin blockade in heart failure with preserved ejection fraction: a systematic review and meta-analysis. ESC Heart Fail. 2016; 4: 402–408. 32. Öhman J, Harjola V-P, Karjalainen P, Lassus J. Focused echocardiography and lung ultrasound protocol for guiding treatment in acute heart failure. ESC Heart Fail. 2018; 5: 120–128. 12. Hayashi T, Hasegawa T, Kanzaki H, Funada A, Amaki M, Takahama H, Ohara T, Sugano Y, Yasuda S, Ogawa H, Anzai T. Subclinical hypothyroidism is an independent predictor of adverse cardiovascular outcomes in patients with acute decompensated heart failure. ESC Heart Fail. 2016; 3: 168–176. 33. References Savage HO, Khushaba RN, Zaffaroni A, Colefax M, Farrugia S, Schindhelm K, TESChler H, Weinreich G, Grueger H, Neddermann M, Heneghan C, Simonds A, Cowie MR. Development and valida- tion of a novel non-contact monitor of nocturnal respiration for identifying sleep-disordered breathing in patients with heart failure. ESC Heart Fail. 2016; 3: 212–219. 36. Yamamoto E, Hirata Y, Tokitsu T, Kusaka H, Tabata N, Tsujita K, Yamamuro M, Kaikita K, Watanabe H, Hokimoto S, Maruyama T, Ogawa H. The clinical sig- niﬁcance of plasma neopterin in heart failure with preserved left ventricular ejection fraction. ESC Heart Fail. 2016; 3: 53–59. 16. Yoshihisa A, Watanabe S, Kanno Y, Takiguchi M, Sato A, Yokokawa T, Miura S, Shimizu T, Abe S, Sato T, Suzuki S, Oikawa M, Sakamoto N, Yamaki T, Sugimoto K, Kunii H, Nakazato K, Suzuki H, Saitoh S, Takeishi Y. The CHA2DS2- VASc score as a predictor of high mortal- ity in hospitalized heart failure patients. ESC Heart Fail. 2016; 3: 261–269. 27. Aleksova N, Demers C, Strachan PH, MacIver J, Downar J, Fowler R, Heyland DK, Ross HJ, You JJ. Barriers to goals of care discussions with hospitalized pa- tients with advanced heart failure: feasi- bility and performance of a novel questionnaire. ESC Heart Fail. 2016; 3: 245–252. 37. Pose A, Almenar L, Gavira JJ, López- Granados A, Blasco T, Delgado J, Aramburu O, Rodríguez A, Manzano L, Manito N. Beneﬁt of tolvaptan in the management of hyponatraemia in pa- tients with diuretic-refractory conges- tive heart failure: the SEMI-SEC project. ESC Heart Fail. 2017; 4: 130–137. 17. Hoshida S, Shinoda Y, Ikeoka K, Fukuoka H, Inui H, Watanabe T. Age- and sex- related differences in diastolic function and cardiac dimensions in a hypertensive population. ESC Heart Fail. 2016; 3: 270–277. 28. Toma M, Mak GJ, Chen V, Hollander Z, Shannon CP, Lam KKY, Ng RT, Tebbutt SJ, Wilson-McManus JE, Ignaszewski A, Anderson T, Dyck JRB, Howlett J, Ezekowitz J, McManus BM, Oudit GY. Differentiating heart failure phenotypes using sex-speciﬁc transcriptomic and 38. Silvetti S, Belletti A, Fontana A, Pollesello P. Rehospitalization after in- termittent levosimendan treatment in advanced heart failure patients: a meta- analysis of randomized trials. ESC Heart Fail. 2017; 4: 595–604. 18. Morishita T, Uzui H, Mitsuke Y, Amaya N, Kaseno K, Ishida K, Fukuoka Y, Ikeda H, Tama N, Yamazaki T, Lee JD, Tada H. References Association between matrix metalloproteinase-9 and worsening DOI: 10.1002/ehf2.12540 ESC Heart Failure 2019; 6: 903–908 908 S.D. Anker et al. Cardiac Oncology Toxicity (COT) Registry of the European Society of Cardiology. ESC Heart Fail. 2017; 4: 312–318. Firat H, Rossignol P, Pizard A, Zannad F. Rationale of the FIBROTARGETS study designed to identify novel bio- markers of myocardial ﬁbrosis. ESC Heart Fail. 2018; 5: 139–148. 39. Ancion A, Allepaerts S, Oury C, Gori AS, Piérard LA, Lancellotti P. Serum albumin level and hospital mortality in acute non-ischemic heart failure. ESC Heart Fail. 2017; 4: 138–145. 44. Peled Y, Lavee J, Arad M, Shemesh Y, Katz M, Kassif Y, Asher E, Elian D, Har-Zahav Y, Goldenberg I, Freimark D. The impact of gender mismatching on early and late outcomes following heart transplantation. ESC Heart Fail. 2017; 4: 31–39. 48. Mustroph J, Wagemann O, Lücht CM Trum M, Hammer KP, Sag CM, Lebek S, Tarnowski D, Reinders J, Perbellini F, Terracciano C. Empagliﬂozin reduces Ca/calmodulin-dependent kinase II ac- tivity in isolated ventricular cardiomyocytes. ESC Heart Fail. 2018; 5: 642–648. 40. Cohen-Solal A, Jacobson AF, Piña IL. Beta blocker dose and markers of sympa- thetic activation in heart failure pa- tients: interrelationships and prognostic signiﬁcance. ESC Heart Fail. 2017; 4: 499–506. 41. Salem K, ElKhateeb O. Gender-adjusted and age-adjusted economic inpatient burden of congestive heart failure: cost and disability-adjusted life-year analy- sis. ESC Heart Fail. 2017; 4: 259–265. 45. Koyama S, Kuragaichi T, Sato Y, Kuwabara Y, Usami S, Horie T, Baba O, Hakuno D, Nakashima Y, Nishino T, Nishiga M, Nakao T, Arai H, Kimura T, Ono K. Dynamic changes of serum microRNA-122-5p through therapeutic courses indicates amelioration of acute liver injury accompanied by acute car- diac decompensation. ESC Heart Fail. 2017; 4: 112–121. 49. Smedema JP, van Geuns RJ, Ector J, Heidbuchel H, Ainslie G, Crijns HJGM. Right ventricular involvement and the extent of left ventricular enhancement with magnetic resonance predict ad- verse outcome in pulmonary sarcoidosis. ESC Heart Fail. 2018; 5: 157–171. 42. Jain A, Scott C, Chen HH. The renal- cardiac connection in subjects with pre- served ejection fraction: a population based study. ESC Heart Fail. 2017; 4: 266–273. 50. Yoshihisa A, Sato Y, Yokokawa T, Sato T, Suzuki S, Oikawa M, Kobayashi A, Yamaki T, Kunii H, Nakazato K, Nakazato SI. Liver ﬁbrosis score predicts mortality in heart failure patients with preserved ejection fraction. ESC Heart Fail. DOI: 10.1002/ehf2.12540 ESC Heart Failure 2019; 6: 903–908 References 2018; 5: 262–270. 43. Lancellotti P, Galderisi M, Donal E, Edvardsen T, Popescu BA, Farmakis D, Filippatos G, Habib G, Lestuzzi C, Santoro C, Moonen M, Jerusalem G, Andarala M, Anker SD, ESC Cardiac Oncology Toxicity Long-Term Registry Investigators. Protocol update and preliminary results of EACVI/HFA 46. Porto CM, Silva VL, da Luz JSB, Filho BM, da Silveira VM. Association between vitamin D deﬁciency and heart failure risk in the elderly. ESC Heart Fail. 2018; 5: 63–74. 47. Ferreira JP, Machu JL, Girerd N, Jaisser F, Thum T, Butler J, González A, Diez J, Heymans S, McDonald K, Gyöngyösi M,
https://openalex.org/W4321457877	https://www.frontiersin.org/articles/10.3389/fclim.2023.994193/pdf	English	null	Dimming the midnight sun? Implications of the Sámi Council's intervention against the SCoPEx project	Frontiers in climate	2,023	cc-by	12,320	TYPE Original Research PUBLISHED 21 February 2023 DOI 10.3389/fclim.2023.994193 TYPE Original Research PUBLISHED 21 February 2023 DOI 10.3389/fclim.2023.994193 © 2023 Oksanen. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. humanitarian perspectives, climate change, solar geoengineering, policy perspectives, humanitarian, Sámi Council, SCoPEx, indigenous peoples OPEN ACCESS OPEN ACCESS EDITED BY Roop Singh, Red Cross Red Crescent Climate Centre, Netherlands REVIEWED BY Duncan McLaren, Lancaster University, United Kingdom Atsushi Ishii, Tohoku University, Japan CORRESPONDENCE Aslak-Antti Oksanen antti.oksanen@bristol.ac.uk SPECIALTY SECTION This article was submitted to Climate Risk Management, a section of the journal Frontiers in Climate RECEIVED 14 July 2022 ACCEPTED 17 January 2023 PUBLISHED 21 February 2023 CITATION Oksanen A-A (2023) Dimming the midnight sun? Implications of the Sámi Council’s intervention against the SCoPEx project. Front. Clim. 5:994193. doi: 10.3389/fclim.2023.994193 COPYRIGHT © 2023 Oksanen. This is an open-access artic distributed under the terms of the Creative Commons Attribution License (CC BY). The u distribution or reproduction in other forums i permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, accordance with accepted academic practice No use, distribution or reproduction is permitted which does not comply with these Aslak-Antti Oksanen School of Sociology, Politics and International Studies, University of Bristol, Bristol, United Kingdom Indigenous peoples are amongst those most vulnerable to the efects of climate change and any potential environmental efects of geoengineering projects. It is therefore not surprising that the Sámi Council decided to take an open stance on the SCoPEx solar geoengineering research project upon ﬁnding out of its planned test ﬂight near Giron/Kiruna, Sweden, in the Sámi people’s domicile area. In their open letter to the to the SCoPEx Advisory Committee, the Swedish Space Corporation and the Swedish government, the Sámi Council objected to the lack of any consultations with the Sámi people and the aims of the project, resulting in cancellation of the ﬂight. As the Sámi Council has a strong track record of leadership among indigenous peoples globally, this intervention has implications for the role of indigenous peoples in relation to the question of geoengineering. This paper uses a discourse analytical method to analyse publicly available sources to map out the background for the Sámi Council’s intervention against the SCoPEx project and its future implications. It ﬁnds that the manner in which the SCoPEx project’s test ﬂight was planned on Sámi domicile area, without any consultations, led the Sámi Council to ﬁnd joint cause with environmental civil society groups opposed to geoengineering. Subsequently, the Sámi Council has taken an active role in rallying further indigenous opposition to the SCoPEx project and by extension geoengineering research. Introduction representatives, as required by the UN Declaration on the Rights of Indigenous Peoples (UNDRIP, 2007) and humanitarian principles of safeguarding those most vulnerable (Sphere Project, 2018, pp. 38, 43), the Sámi learned about it from the Indigenous Environmental Network (IEN), which is part of the anti-geoengineering campaign Hands oﬀMother Earth (HOME). Given the Sámi people’s historical and contemporary experience with settler society’s continuous extractive incursions into Sámi ancestral lands and waters, it is unsurprising that learning about a solar geoengineering research project being planned in their domicile area without seeking their consent would draw a hostile response. From a humanitarian perspective, indigenous peoples are amongst those most vulnerable to climate change. The impacts of climate change on indigenous peoples are global but take diﬀerent forms in diﬀerent regions, ranging from paciﬁc islanders risking displacement by rising sea levels (Voyatzis-Bouillard and Kelman, 2021), to pastoralists in the Sahel suﬀering the consequences of droughts and land (Dieng, 2021) and the myriad negative eﬀects unpredictable weather conditions have on people living close to nature (Jones, 2019; Schramm et al., 2020). The Sámi people, who inhabit northern Fennoscandia and the Russian Kola peninsula have also experienced adverse eﬀects of climate change, with mid-winter thaws resulting in icy snows blocking access to reindeer pastures, forcing Sámi reindeer herders to resort to costly emergency fodder, and unpredictable weather conditions complicating seasonal herd migrations (personal correspondence with Sámi reindeer herders). Geoengineering advocates could therefore expect that the Sámi, along with other indigenous peoples, may be in favor of geoengineering research that could generate knowledge with the potential for drastically mitigating or even reversing global warming. However, there has been declared opposition from indigenous organizations through statements like the Anchorage Declaration. The Sámi have recently joined ranks with indigenous peoples opposing geoengineering research through the Sámi Council’s open letters demanding the cancellation of the planned test ﬂight for Harvard University’s SCoPEx project, which had been scheduled for June 2021 on Sámi domicile lands in Giron/Kiruna (northern Sweden) and the halting of the project altogether. The SCoPEx project aims to explore some of the potential side-eﬀects of Stratospheric Aerosol Injection (SAI) as a way of cooling the planet’s temperature by reﬂecting more sunlight (Low et al., 2022, p. 5). Introduction Rather than embracing the concept of geoengineering as a potential solution for averting climate disaster, most indigenous organizations engaged in the question have rejected it as a distraction from measures necessary to curb climate change, like emission cuts (The Anchorage Declaration, 2009). As the Sámi Council’s intervention directly contributed to the Swedish Space Agency (SSA) canceling SCoPEx planned test ﬂight, while the Great Barrier Reef experiment went ahead with local indigenous participation, it suggests that indigenous peoples have considerable normative power in conferring or denying legitimacy to geoengineering research. The manner in which the SCoPEx project planned its test ﬂight, and the Sámi Council’s reaction appears to have contributed to a hardening of indigenous opposition to geoengineering research, as it has deepened ties between them and environmental civil society organizations, as will be evidenced below. Moreover, the Sámi Council has a strong track record of leadership among indigenous peoples, and has since gone on to write an open letter to Harvard University asking them to scrap the SCoPEx project, co-signed by more than 30 indigenous peoples’ organizations from around the world (Doyle, 2021). This is a highly unfavorable outcome to anybody interested in carrying out such research, caused by the SCoPEx project’s failure to consult the Sámi people on high-stakes research planned on their ancestral lands. Moreover, it fails to live up to international standards on indigenous rights as codiﬁed in the UNDRIP (2007) and humanitarian principles as deﬁned by the Red Cross/Red Crescent led Sphere Project (2018). j This paper will assess the implications of the Sámi Council’s intervention against the SCoPEx project for the role of indigenous peoples in relation to geoengineering research. It takes stock of the background for the intervention, of how indigenous peoples came to matter in global environmental governance and their previous engagement with the question of geoengineering. It will also map out the alliances that have formed between indigenous peoples’ organizations and environmental civil society organizations engaged in the question of geoengineering and how this has played into the Sámi Council’s intervention. In doing this, it will employ Laclau and Mouﬀe’s discourse theory to analyse how discursive framing of the SCoPEx project and the issue of geoengineering has catalyzed the formation of the alliance of indigenous peoples’ organizations and environmental civil society organizations that the Sámi Council decided to join. OPEN ACCESS It is argued that this coalition of indigenous peoples’ organizations and environmental civil society organizations is premised on a discursive framing of an opposition between nature-based solutions to climate change and, geoengineering as representative of a technological solution that allows extractive capitalism to persist. The Sámi Council’s intervention has important humanitarian implications. As indigenous peoples are uniquely vulnerable to any environmental changes resulting from geoengineering, the Sámi Council’s intervention and its outcome sets a precedent of indigenous peoples as stakeholders in the geoengineering question, whose views must be respected and interests safeguarded. 01 Frontiers in Climate frontiersin.org 10.3389/fclim.2023.994193 Oksanen Introduction The paper focuses on the Sámi Council’s and HOME aﬃliated civil society organizations’ discursive framing of the SCoPEx project and SAI research, and has opted not to include a similar assessment of SCoPEx and SAI advocacy discourse, as SCoPEx and SAI advocates, unlike the HOME aﬃliated organizations, appear to have made no eﬀort to inﬂuence the Sámi Council. While indigenous opposition to geoengineering research is outspoken and widespread, it is not unanimous. An ocean fertilization geoengineering experiment was carried out with the consent of a Haida nation band council in Haida Gwaii, British Columbia. However, only one of the island’s two Haida band councils approved the project, and many Haida individuals rejected it, seeing it as having been carried out in bad faith, with the approving band council not expressing the community’s majority opinion (Gannon and Hulme, 2018; Whyte, 2018, p. 298). This experience has likely contributed further to indigenous skepticism toward geoengineering research. A more positive example of geoengineering research involving indigenous people, is that of the University of Sydney and Queensland University-led cloud brightening experiment by the Great Barrier Reef, which was carried out with the consent of the local indigenous Manduburra organization, representing the traditional land and water owner, with an observer taking part in the experiment (Readfearn, 2020; Low et al., 2022, p. 90). The manner in which this research team consulted the local indigenous people and sought their informed consent stands in stark contrast to how the SCoPEx project planned its Giron/Kiruna test ﬂight without consulting the local indigenous Sámi people. Instead of being informed of the planned the test ﬂight through consultations with SCoPEx project Frontiers in Climate frontiersin.org How indigenous peoples came to matter in global environmental governance Indigenous peoples were long marginal in every sense, subject to genocides, forced displacement and assimilation. Their interests were given little regard and formal promises given to them were easily broken. The settler-states’ treatment of indigenous peoples was motivated by a variety of discriminatory attitudes and assumptions that indigenous peoples had no future with the advance of modernity. Attitudes from settler-colonial governments ranged from exterminationist hostility (Quijano, 2000, p. 568), to pity, where indigenous peoples’ demise should be softened (Lundmark, 2008, pp. 138–143), to wanting to “assist” indigenous peoples to overcome their “backwardness” through assimilation to settler culture (Tavanti, 2003, p. 107; Coulthard, 2014, p. 11). Laclau and Mouﬀe term privileged positions of partial ﬁxity within a discourse as nodal points. Society is held together by widely accepted discursive nodal points like “the nation,” “democracy,” and “human rights.” It is with reference to such discursive nodal points that relational identities and diﬀerences in standing between diﬀerent groups come to be accepted as “natural” (Laclau and Mouﬀe, 2001, pp. 113, 135). Tensions still exist between diﬀerent social groups, but in contemporary society, there is such a wide plurality of relational identities and antagonisms that a unifying point of popular mobilization is absent (Laclau and Mouﬀe, 2001, pp. 85– 87, 133–134). Under such conditions, political mobilization entails the articulation of discourses that claim existing discursive nodal points and frame inequalities as unjust with reference to them, while supplanting them with new nodal points (Laclau and Mouﬀe, 2001, p. 189). These discourses redeﬁne relational identities by creating chains of equivalence between diﬀerent social groups that bridge their diﬀerences with reference to their shared interests, in antagonistic opposition to groups seen as threatening those interests (Laclau and Mouﬀe, 2001, pp. 127–134). Thus, it is through the articulation of shared narratives centered on discursive nodal points with wide public acceptance that chains of equivalence can be created to bridge the diﬀerence between various groups in their pursuit of shared political goals. Settler-states’ attitudes toward indigenous peoples only began to change due to indigenous peoples’ active eﬀorts to challenge portrayals of them, which eventually resulted in some successes codiﬁed through national and international principles and legal standards for indigenous peoples’ rights. A key catalyst for this was the emergence of the ﬁrst intercontinental indigenous peoples’ organizations, which gave indigenous peoples an unprecedented capacity to assert themselves on the global stage. Materials and methods Laclau and Mouﬀe’s (2001) discourse theory oﬀers a valuable tool for analyzing the context of the Sámi Council’s intervention against the SCoPEx project, as it enables seeing the shared narratives that 02 frontiersin.org frontiersin.org 10.3389/fclim.2023.994193 10.3389/fclim.2023.994193 Oksanen have brought diﬀerent indigenous and environmental civil society groups together in joint opposition to geoengineering research. to self-determination and human rights, and how indigenous peoples came to be engaged in the question of geoengineering alongside environmental civil society groups. This provides the context for the Sámi people’s intervention against the SCoPEx project, which will be analyzed through the Sámi Council’s open letters and an 80 min public webinar co-hosted by them and environmental civil society organizations that includes highly prominent environmental scholars and campaigners. Laclau and Mouﬀe (2001, pp. 4, 85–87) characterize their work as “post-Marxist,” as they see it as an evolution of Marxist class-based analysis for a time where relational identities are much more ﬂuid and transitory than during the early stages of industrial capitalism when Marxism emerged as a social theory. Rather than seeing contemporary society as characterized by opposition between clearly deﬁned social classes, they see relational identities as inherently ﬂuid (Laclau and Mouﬀe, 2001, pp. 85–87). The concept of articulation is central to Laclau and Mouﬀe’s discourse theory. Articulations exist within the context of a discourse, and deﬁne or modify the relational identities of diﬀerent groups of people, or elements as Laclau and Mouﬀe term such groups. Once these elements assume their relational positions through the practices of articulation, they become moments, with the appearance of clearly deﬁned identities. However, the transformation of elements into moments is always incomplete, as the relational identities remain unstable, because every articulation is non-identical to itself, containing excess and non- conformity to the relational identities it tries to impose on diﬀerent elements (Laclau and Mouﬀe, 2001, pp. 105–114). How indigenous peoples came to matter in global environmental governance This process was initiated through parallel eﬀorts by indigenous peoples on both sides of the USA-Canada border. In Canada, indigenous peoples for the ﬁrst time formed a national umbrella organization through the National Indian Brotherhood (NIB) to resist the implementation of the Canadian government’s 1969 White Paper that would have abolished indigenous status and all accompanying indigenous treaty rights in Canada (McFarlane, 1993, pp. 101–107; Cowger, 1999, p. 3). In the USA, the Civil Rights Movement of the 1960s had inspired the emergence of an American Indian Movement. It hosted the inaugural gathering of the Pan-American International Indian Treaty Council (IITC) in 1974 (Sanders, 1989). The following year the NIB hosted the inaugural conference of the World Council of Indigenous Peoples (WCIP), which in addition to representatives of indigenous peoples from the Americas, also included M¯aori, Aboriginals of Australia, Innuits, and the Sámi People, represented by the Sámi Council (Sanders, 1977, pp. 15–18). The Sámi Council assumed a central role in the WCIP, hosting its second 1977 conference in Giron/Kiruna (the town by which the planned SCoPEx test ﬂight was to take place) (Jull, 2003, p. 25). Though there were signiﬁcant diﬀerences among the indigenous peoples [which in the end would contribute to the dissolution of the WCIP in 1996 (Nyyssönen, 2007, p. 329)], they succeed in creating chains of equivalence bridging their diﬀerence through the shared experience of settler-colonial dispossession and subjugation and opposition to continued colonial governance and extractivism, helping them achieve their shared goals of gaining representation at global forums. The below application will demonstrate how claiming existing discursive nodal points has enabled indigenous peoples to assert themselves in global environmental governance forums and allowed the bridging of diﬀerences between indigenous peoples’ organizations and environmental civil society organizations though the creation of chains of equivalence with reference to joint antagonism toward what geoengineering is seen to represent. Frontiers in Climate frontiersin.org Results Hence their approach is to say no when in doubt, and do so in time, to establish a precedent for accepting them as stakeholders before projects on their territories are initiated and expended with reference to existing sunk costs (Lawrence and Kløcker, 2016, pp. 50–51; Lindahl et al., 2016, p. 20). Since indigenous peoples achieved some success in articulating themselves with the nodal points of self-determination and human rights within the UN, they have worked on transposing these discursive nodal points to global environmental governance to make human rights, and by extension indigenous rights, intrinsic to it. This can be seen in how they have approached two key global environmental governance covenants, the United Nations Framework Convention on Climate Change (UNFCCC) and the Convention on Biological Diversity (CBD) and their annual Conferences of Parties (COPs). Indigenous peoples are highly organized in their approach to these forums, as all accredited indigenous organizations attend the COPs under their respective umbrella organizations, the International Indigenous Peoples Forum on Climate Change (IIPFCC) (Suiseeya et al., 2022, p. 611) and the International Indigenous Forum on Biodiversity (IIFB) (Witter et al., 2015, p. 899). At both forums, the indigenous caucuses have followed the same discursive strategy to embed human rights, and by extension, indigenous rights, within the language of the formal commitments enacted under the umbrella of the framework agreements. With the CBD, they succeeded in incorporating a clause pertaining speciﬁcally to respect for indigenous knowledge, innovation and practices in Article 8j (Witter et al., 2015, p. 899). With the UNFCCC’s Paris Agreement, indigenous peoples failed, despite their joint eﬀorts with non-indigenous civil society organizations, to secure the inclusion of any reference to indigenous rights, or human rights (Suiseeya et al., 2022, pp. 614–622). They did, however, succeed in including a clause in the Paris Agreement on respecting indigenous peoples’ knowledge under article 7.5. Also, symbolically, the Agreement refers to indigenous peoples in the plural, which was a point of contention, with some state delegations preferring the singular “indigenous people” to avoid the self-determination implications of using the plural form of “indigenous peoples” (Suiseeya et al., 2022, pp. 614–622). Results Gaining institutional acceptance of indigenous peoples’ rights and representation at the UN level contributed to transforming outside perceptions of indigenous peoples from relics of the past, destined for cultural extinction, to a permanent feature of the world capable of adapting to modernity, from which settler society can learn, particularly on environmental issues. Thus, the discourse has shifted, from indigenous peoples as an impediment to settler society’s progress, to becoming rights-bearing entities and a reservoir of knowledge of potential value for solving the sustainability problems setter society has encountered. While this instrumentalization of indigenous knowledge to settler society’s needs is problematic (Reid and Chandler, 2018, pp. 259–261) and actively challenged by indigenous people (Deranger et al., 2022, pp. 69–71) there are also indigenous groups that embrace it (Zurba and Papadopoulos, 2021), as it provides a discursive nodal point to leverage indigenous inﬂuence in global environmental forums. Indigenous peoples’ engagement with the question of geoengineering ﬁrst took formal expression through the Anchorage Declaration, which was adopted at the IIPFCC’s preparatory conference in the run-up to the 2009 Copenhagen UNFCCC COP 15 (McLean et al., 2009). The Declaration stated that the signatories challenged states to “abandon false solutions that negatively impact Indigenous Peoples” rights, lands, air, oceans, forests, territories and waters [including] nuclear energy, large-scale dams, geo-engineering techniques, “clean coal,” agro-fuels, plantations, and market-based mechanisms such as carbon trading, the Clean Development Mechanism, and forest oﬀsets (The Anchorage Declaration, 2009). Thus, indigenous peoples successfully bridged their diﬀerences with reference to their shared plight under settler-colonialism and articulated of themselves onto the widely recognized discursive nodal point peoples’ right to self-determination, as embedded within the UN’s wider human rights discourse. However, despite gaining institutional acceptance of the principle of indigenous self- determination, translating it into concrete rights and safeguards has been an uneven process with indigenous peoples remaining highly vulnerable. They know, and have been recurrently reminded, that in projects where strong economic interests are at stake, their chances to inﬂuence are minimal, with “consultations” often amounting to being informed of what has already been decided (Dylan et al., 2013, While the Anchorage declaration only mentioned geoengineering in passing alongside what were characterized as other false solutions to the climate crisis, the 2010 World People’s Conference on Climate Change, in Cochabamba, Bolivia, turned out to be more signiﬁcant in shaping indigenous peoples’ stances in relation to geoengineering. Results This section uses Laclau and Mouﬀe’s discourse theory to analyse the context and content of the Sámi Council’s intervention against the SCoPEx project. It will ﬁrst be shown how indigenous peoples came to matter in questions of global environmental governance by articulating themselves to the discursive nodal points of peoples’ right 03 Oksanen Oksanen 10.3389/fclim.2023.994193 Oksanen 10.3389/fclim.2023.994193 By establishing the IITC and WCIP, indigenous peoples created the organizational capacity to assert themselves on the global stage. They found the discursive nodal point of peoples’ right to self- determination, which the anti-colonial movements of the Global South had successfully embedded within the UN’s human rights agenda (Alam and Al Faruque, 2019, p. 62), to be an eﬀective pathway for articulating demands for indigenous rights. The then ongoing process of codifying the UN’s human rights principles into legally binding covenants provided an entry point through which to begin articulating indigenous claims to self-determination on the global stage. For example, article 1 under both the International Covenant on Civil and Political Rights (ICCPR, 1966) and the International Covenant on Economic, Social and Cultural Rights (ICESCR, 1966), aﬃrm peoples’ right to self-determination, and the ICCPR’s Article 27 aﬃrms minorities’ rights to practice their own culture. Legal challenges by indigenous groups and individuals within states that ratiﬁed these covenants and complaints to the UN’s Human Rights Committee (UNHRC) have resulted in precedents that eﬀectively extended such rights to indigenous peoples (Doyle, 2015; Saul, 2016). And the IITC’s and WCIP’s lobbying toward the UNHRC contributed to initiating a process that led to the formation of the UN Working Group on Indigenous Populations (UNWGIP) in 1982. The UNWGIP’s work eventually led to the formation of the UN Permanent Forum on Indigenous Issues (UNPFII) in 2000 and the passing of the UNDRIP by the UN General Assembly in 2007 (Eide, 2006, pp. 163–164; Lile, 2006; Kemner, 2011, p. 14). The Sámi Council has nominated the Sámi people’s representatives to the UNPFII, who have played highly inﬂuential roles with the Forum, exempliﬁed by the ﬁrst UNPFII Chairperson Ole Henrik Magga and the current UNPFII (at the time of writing) Chairperson, Ánne Nuorgam, being Sámi representatives (Sámiráddi, 2015; Scaﬃdi, 2021). pp. 74–78). They have also found out the hard way, that stopping projects early is often their only chance. Frontiers in Climate frontiersin.org Results The event was organized by civil society organizations disillusioned by the UNFCCC process following COP 15 (Zurba and Papadopoulos, 2021). During it, the ETC 04 frontiersin.org 10.3389/fclim.2023.994193 Oksanen Oksanen demand its termination (Erickson, 2020, pp. 120–121). The Sámi people have found themselves directly at odds with environmental civil society organizations over the issues of wind power and predator conservation. Environmental civil society organizations have enthusiastically backed the expansion of wind power into the Nordic tundra, with little regard for its severely detrimental eﬀect on Sámi reindeer herding or for the Sámi’s territorial rights (Normann, 2021, p. 79) Conservationist organizations see the protection of predators like wolf, wolverine and lynx in reindeer herding areas as essential, with little consideration for their predation on the Sámi’s reindeer (Vidal, 2016). The SCoPEx project, and what it is seen to represent has provided a shared reference point for indigenous organizations and environmental civil society organizations, environmental scholars and activists to paper over their potential diﬀerences and articulate themselves in united opposition to geoengineering research. The below analysis of the open letters and a public webinar co-hosted by the Sámi Council and HOME aﬃliated environmental civil society organizations maps out the discursive nodal points that created the chain of equivalence between them. Group initiated the anti-geoengineering campaign Hands Oﬀ Mother Earth (HOME) (Thiele, 2019, p. 463; Sikka, 2020, pp. 93–94), with other environmental civil society organizations like Friends of the Earth and indigenous organizations, including The Indigenous Environmental Network and La Viva Campesina, among co-founding members (Ribeiro, 2020, pp. 226–227; ETC Group, 2012). HOME published its manifesto in 2018, which declares ﬁrm opposition to all forms of geoengineering research, including shooting aerosols into the stratosphere, cloud brightening, ocean fertilization, and also carbon capture and storage. The manifesto has been signed by numerous, predominantly non-indigenous, environmental civil society organizations, including the Heinrich Böll Foundation (HOME Manifesto, 2018). Thus, while an anti- engineering coalition that included environmental civil society organizations and indigenous peoples’ organizations already existed prior to the Sámi Council’s intervention, indigenous peoples’ organizations made up a small proportion of its declared members. The following sub-section analyses how the Sámi Council articulated themselves as part of that alliance, and how it has since taken a leading role in growing its indigenous contingent. Results From the webinar and the other publicly available information, we can see that there was coordination between the Sámi Council and HOME member organizations—speciﬁcally the Indigenous Environmental Network, the ETC Group, and the Heinrich Böll Foundation—in their opposition to the SCoPEx project’s planned Giron/Kiruna test ﬂight. The Sámi Council’s open letter,1 dated 24.02.2021, addressed to the SSC, the Swedish Government and the SCoPEx Advisory Committee was preceded by a letter sent weeks earlier to the SSC and the Swedish government, co- signed by three aforementioned organizations, Greenpeace Sweden, Friends of the Earth Sweden, Friends of the Earth International and others. There are clear similarities between the letters as both highlights the unpredictability of SAI technology with potential for extreme consequences, the danger of a termination shock, of it deﬂating pressure to transition to carbon neutrality, and the lack of both Swedish representation and consultation with aﬀected indigenous populations as required by UNDRIP (2007) (Open Letter to the SSC and Swedish Government by Greenpeace Sweden, Jordens Vänner/Friends of the Earth Sweden, Naturskyddsföreningen/Swedish Society for Nature Conservation (SSNC), Action Group on Erosion, Technology and Concentration (ETC Group), Biofuelwatch, Center for International Environmental Law (CIEL), Climate Justice Alliance (CJA), Friend of the Earth International, Heinrich Böll Foundation, Indigenous Environmental Network (IEN), WhatNext? (2020); Open Letter to the SCoPEx Advisory Committee by the Sámi Council Open Letter to the SCoPEx Advisory Committee (2021). The Sámi Council’s letter also raises further objections toward the homogenous composition of the SCoPEx Advisory Committee, consisting almost entirely of US citizens and/or residents, and its approach to research ethics as expressed in the draft “Engagement Process for SCoPEx,” which states that no research project should have to answer questions pertaining to the potential moral hazard of geoengineering deployment and surrounding decision making and governance 1 Co-signed by Johanna Sandahl, President, Swedish Society for Nature Conservation, Mikael Sundström Chairperson, Friends of the Earth Sweden and Isadora Wronski, Programme Manager, Greenpeace Sweden. Frontiers in Climate The Sámi Council’s intervention against the SCoPEx project And the message in this letter, is that we as indigenous peoples do not approve of the legitimizing development toward solar geoengineering and nor for it to be conducted in or above our lands or territories and skies, nor in any ecosystem anywhere.” The close partnership that formed between the HOME coalition, represented by the ETC Group, Heinrich Böll Foundation, and the Indigenous Environmental Network, and the Sámi Council in articulating their opposition to the aims of the SCoPEx project can be further discerned from the webinar co-hosted by these organizations live streamed through the ETC Group’s Facebook page on June 6th, 2021 (ETC Group’s Facebook Page, 2021). The normative inﬂuence of the Sámi Council’s intervention against the SCoPEx project can be seen from how the webinar drew the participation of the global inﬂuencers in the climate change debate of Naomi Klein, Vandana Shiva, Bill McKibben and Greta Thunberg in solidarity with the Sámi people. The rhetoric of the webinar contributors shows the discursive nodal points that formed the chains of equivalency within this diverse group and how this was posited in antagonistic opposition to what the SCoPEx project was seen to represent. The central discursive nodal point of the opposition between the extractive, mechanistic mastery over nature mentality and its technological solutions to climate change, and respect for the sanctity of Mother Earth and nature-based solutions also came through strongly in Vandana Shiva’s contribution. Shiva is a physicist who founded the Research Foundation for Science, Technology, and Natural Resource Policy (RFSTN), author of numerous books, including Ecofeminism, Gates to a Global Empire and We Feed the World, and a social and environmental activist. Her contribution began with stressing how the idea of SAI was expressive of the mistaken assumption that control and mastery over nature is possible and desirable: “You are adding pollution to the existing pollution, creating the same instability. Pollution for a pollution problem; deliberate climate change as a solution to climate change is not just insane, but it is, as Einstein said, repeating the mindset that got you into the crisis in the ﬁrst place. And what is that mindset? A mechanistic worldview. The idea of mastery and control, of engineering a solution for everything living.” Vice President, Åsa Larsson Blind represented the Sámi Council at the webinar. The Sámi Council’s intervention against the SCoPEx project In her opening statement, she said that the Sámi Council had been unaware of the SCoPEx project and the larger question of geoengineering, and was alerted to SCoPEx’s planned activities in Giron/Kiruna in February 2021 by Tom Goldtooth of the Indigenous Environmental Network and the ETC Group. This triggered what Larsson Blind described as an “instinctive” response by the Sámi Council as expressed in the open letter: The discursive fault line between geoengineering advocates as adherents of a misguided mastery over nature mentality, and the harmony with nature mentality and nature based-solutions to climate change championed by the webinar contributors, was further reinforced by emphasizing how dimming the sun through SAI would impede photosynthesis, which is a precondition for nature-based solutions to climate change. Holding up a leaf for dramatic eﬀect: “We reacted quite instinctively when we learned what the idea of solar geoengineering is, and what they are attempting to do. We reacted instinctively, because this goes against our world view, that we as humans should live and adapt to nature, and it is against the respect with which we are taught to treat nature and mother earth. And it is the whole idea that the line of thought that has put us in the climate crisis in the ﬁrst place should also be the one thing taking us out of it. And there our position is clear, that now is the time for all society to listen to what mother earth is telling us and nature is telling us now. Our focus should be on ﬁnding a way to a sustainable climate-friendly society. And this is why we have reacted... together with other organizations and allies, with the result of the cancellation of the test in Giron.” “When the sun shines on this green leaf, with chlorophyll, we get photosynthesis. We get the amazing capacity of this leaf to absorb carbon dioxide, but give us oxygen, give us carbohydrates. Blocking the sun means blocking photosynthesis. Blocking the sun means putting food production and food security at risk. And the grazing of the reindeer at risk. And you don’t know, it’s totally unpredictable what it’s going to do. But it’s doubly wrong, because this green leaf is the alternative to fossil fuels. The Sámi Council’s intervention against the SCoPEx project While the Sámi Council has been active in global environmental governance, most clearly exempliﬁed by their Permanent Participant status at the Artic Council (2023), they had up to their intervention against the SCoPEx project, remained outside the discussion on geoengineering. This section applies Laclau and Mouﬀe’s discourse theory to publicly available material, to analyse how the Sámi Council decided to articulate itself in opposition to geoengineering research, alongside environmental civil society groups and inﬂuential environmental scholars and activists, as a result of SCoPEx’s planned test ﬂight in the Sámi domicile area. Before proceeding with the analysis, is it worth pointing out, that though indigenous groups and environmental civil society organizations often work together to achieve their shared interests protecting the natural environment (see, for example, their above referenced collaboration at UNFCCC COPs), there are at times signiﬁcant disagreements between them on what that entails, or what speciﬁc aspects of nature should be protected. The most infamous example of an environmental civil society organization and an indigenous people at loggerheads, is that of how Greenpeace’s anti-sealing campaign of the 1970s and 1980s resulted in the EC/EU’s seal pelt ban, which collapsed the seal pelt market and denied the Inuit’s a vital revenue stream (Rodgers and Ingram, 2019, p. 18). A more recent example, is that of the Canadian Boreal Forest Agreement of 2010, which ended a 3 year long boycott campaign of Canadian forestry products, led by environmental civil society organizations including Greenpeace and Canopy. The agreement was negotiated between 21 Canadian forestry companies and 9 environmental civil society organizations without any consultations with the indigenous peoples residing in the boreal forests. This led the First Nations of Nishnawbe Aski Nation (NAN) council, representing 49 First Nations communities in Northern Ontario, to denounce the agreement for its disregard of indigenous stakeholders and Frontiers in Climate 05 frontiersin.org 10.3389/fclim.2023.994193 10.3389/fclim.2023.994193 Oksanen issues. The letter argues “that precisely because of the extraordinary and particular risks associated with SAI, this technology and SCoPEx cannot be treated like other research.” The letter concludes by stating that research along the lines of SCoPEx cannot be carried out before concerns such as those raised in the letter have been addressed and full global consensus on its acceptability has been reached. world in a joint letter to Harvard University, urging them to put a stop to the SCoPEx project. Frontiers in Climate frontiersin.org The Sámi Council’s intervention against the SCoPEx project There are all these conﬂicts of interest that these scientists who are pushing this [technological solutions] most aggressively, have their own private companies, which beneﬁt from the normalization of the idea of a techno-ﬁx.” Goldtooth connected indigenous opposition to geoengineering research to the longstanding discursive nodal point of indigenous peoples’ right to give or withhold consent on projects aﬀecting them under the UNDRIP, which is nestled within the wider nodal point of the UN’s broader human rights agenda as demonstrated in the prior section: Klein’s focus on Keith’s entrepreneurship highlights how the lines between capitalists backing geoengineering and the scientists advocating for it get blurred when the scientists operate on both sides of the money and know-how equation. Discursively, this serves to homogenize the scientists and their ﬁnancial backers, and particularly those vested in the fossil fuel industry into the “geo clique,” providing a clear opponent to mobilize activists’ energies against. Klein further elaborated on the internal dynamics of the “geo clique,” and how scientists like David Keith provide moral cover the continued wealth accumulation for fossil fuel magnates like Murrey Edwards: “So, lack of consultation is a big concern with our indigenous peoples. We found that out with the Sámi, we found that out in Tucson. This project, they are keeping it secret. I don’t know what they have to hide, but under international standards, under the tool of the UN Declaration on the Rights of Indigenous Peoples, our indigenous peoples, in every region we reside, including Sweden, have a right to the provision of free, prior and informed consent, and the right to give their consent to these kinds of projects.” “It is that moral hazard, that so long as you are holding up the promise of a techno-ﬁx, then you can keep digging it up... and I think it is particularly relevant that David Keith... though he certainly identiﬁes as an environmentalist, and is certainly worried about the climate crisis, is so actively in business with one of the major drivers of said climate crisis. The Sámi Council’s intervention against the SCoPEx project So, not just any old billionaire, like Bill Gates, but someone so embedded in the Tar Sands.” In his concluding remarks, Goldtooth aﬃrmed the discursive nodal point of antagonist opposition between indigenous values and knowledge of living in harmony in nature and the mechanistic mastery over nature mentality that geoengineering is seen as representative of: Klein also supplanted the discursive nodal point of geoengineering being representative of a mastery over nature mentality with how geoengineering projects and their funders reproduced a colonial terra nullius logic in their preference for ﬁnding remote sparsely populated areas for conducting outdoor experiments. As a prior example of such terra nullius predatory logic, she raised the example of the ocean fertilization experiment at Haida Gwaii and how it was carried out by a “rogue geoengineering entrepreneur [Russ George], who was selling this false promise of carbon credits for the Haida’s protection of their forest, if they allowed him to dump a whole bunch of iron ﬁlings in the ocean.” “I just wanted to lift that up very quickly, on the important values of indigenous knowledge, that understands through thousands and thousands of years of our accumulation of our knowledge of the sacredness of Mother Earth and Father Sky. That’s a delicate balance, and that society and humanity and Western sciences must not play God. They have to be held accountable for any technology that they are proposing. Enough damage has been done to the climate and the balance of our ecosystems. But we do have solutions, deﬁnitely.” Bill McKibben, author of books like The End of Nature and Falter, and co-founder of the environmental organization 350.org, spoke along similar lines to Klein, reinforcing the discursive nodal point of geoengineering being a preoccupation of billionaires beneﬁtting from the extractivist status quo. He divided the billionaires into two camps: those who sought to escape the planet altogether and those who sought a technological ﬁx to the climate crisis, both of which distract from the urgent task of decarbonising the global economy: Klein concluded by aﬃrming the discursive battle lines between geoengineering representing extractive capitalist coloniality, and those seeking to live in harmony with nature, by connecting David Keith’s involvement with carbon storage in the Alberta Tar Sands, with indigenous resistance toward the pipeline to the USA, including the blockade against the pipeline that Tom Goldtooth was attending at the time of the webinar. The Sámi Council’s intervention against the SCoPEx project We need to move from hydrocarbons to carbohydrates, the molecule of life, which gives us everything we need.” In Blind’s statement, we can see the ﬁrst discursive nodal point, of a world view of living in harmony with mother earth, which is posited in opposition to technological manipulation of nature through solar geoengineering. Naomi Klein, author of numerous inﬂuential books like No Logo, Shock Doctrine and This Changes Everything and a social and environmental activist, identiﬁed a tangible antagonist to mobilize against by focusing on the capitalist interests backing a technological solution to climate change, terming a small number of geoengineering advocating scientists and their billionaire backers the “geo clique.” She focused particularly on the role of SCoPEx’s lead scientist David Keith, and his relation to Bill Gates and the Alberta Tar Sands bitumen extraction ﬁnancer Murrey Edwards, who were the initial ﬁnancial backers for Keith’s Alberta Tar Sands based carbon capture company Carbon Engineering: Blind went on to state a quite categorical opposition to solar geoengineering, and that the Sámi Council was working to broaden indigenous opposition to it: “But now that we have engaged ourselves in the issue, our position is clear, that this technology of solar geoengineering, that is not a part of our chosen future. So, this is also why we have taken the initiative to gather indigenous peoples from around the Frontiers in Climate 06 frontiersin.org Oksanen 10.3389/fclim.2023.994193 “SCoPEx was presented as a proposed initiative, in the United States, in the southwest a number of years ago, in the Tucson Arizona region, and when we found out, we immediately contacted the tribe of the Yaqui... who live south of Tucson, and they consulted with their spiritual leaders, I talked with them as well. And they were just like appalled.” “This is signiﬁcant, because, David Keith is in business, with Tar Sands companies. His private company, which is an air capture company called Carbon Engineering... Frontiers in Climate frontiersin.org The Sámi Council’s intervention against the SCoPEx project Tom Goldtooth, founder and executive director of the Indigenous Environmental Network and ETC Group board member, joined the webinar late, having driven directly from an indigenous-led blockade against the planned routing of an oil pipeline from the Alberta Tar Sands beneath the Mississippi River in Minnesota. This further reinforced the framing of extractive, fossil fuel driven capitalism and technological solutions to the climate crisis against the alternative of respect for Mother Earth and nature-based solutions. Goldtooth had prior experience with organizing indigenous resistance against a prior planned SCoPEx test ﬂight in Arizona: “Bill Gates, who did the original funding for much of this research, is of the stay at home and come up with some technological miracles, to avoid having to do the real work ahead of us, which is to get oﬀthe coal and gas and oil and replace it with sun and wind and conservation and eﬃciency... Our job is to keep insisting that we have to move fast, that JeﬀBezos aside, there is no planet B, and we are not going to someplace else.... This is really an incredibly beautiful planet, the only one that we know of that supports life, and our job is to ﬁgure out how to get it back to the state that we know works.” 07 frontiersin.org Oksanen 10.3389/fclim.2023.994193 10.3389/fclim.2023.994193 Oksanen Youth climate activist Greta Thunberg’s contribution also centered on the discursive nodal point of how the concept of geoengineering was expressive of the same mastery over nature mindset that created the crisis: sure that we are going to get... it will also show that indigenous peoples are not alone in this, this is a global position, this is a joint position, and I hope that will also make a diﬀerence.” The open letter addressed at Harvard University, dated 04.06.2021, was co-signed by more than 30 indigenous peoples’ organizations from around the world (Doyle, 2021; Sámi Council Open Letter to Harvard University, 2021). This suggests that the Sámi Council’s global leadership among indigenous peoples has potential for signiﬁcant impact in shaping their attitudes toward geoengineering research and in expanding the indigenous contingent of the anti-geoengineering coalition of indigenous peoples’ organizations and environmental civil society groups. The Sámi Council’s intervention against the SCoPEx project The tone adopted in the letter was a bit diﬀerent from that of the Sámi Council’s previous letter to the SCC, as it did not focus on any of the speciﬁcs of SCoPEx planned activities in Giron/Kiruna, but rather, centered on the value-driven, core discursive nodal point of antagonism between living in harmony with Mother Nature, and the mastery over nature mentality expressed though the concept of geoengineering: “When you are in a hole, stop digging. We cannot move out of this crisis with the same mindset that got us into this. We got into this mess by thinking we could control nature, by thinking we can manipulate and thinking we are above nature.... We simply do not know how the earth will respond to this mechanism, if we set these things going.” Thunberg’s reﬂections on the problems with who would control geoengineering once it started also aﬃrmed the framing an antagonistic relation between the beneﬁciaries of the status quo and those disadvantaged by it: “It is probably going to be the high-income countries and the richest people. And we will do things that beneﬁt us, and perhaps even at the expense of other people. Unfortunately, that is most often the peoples living in the Global South and the indigenous peoples around the world. They are always going to suﬀer from these kinds of decisions. And that is the exact opposite of climate justice, and environmental justice, and social justice. It’s like we are trying to do everything we possibly can to avoid reducing emissions... Even manipulating nature, manipulating the sky.” “We note with greatest concern that eﬀorts toward the potential development of solar geoengineering technology are now taken and planned to be performed through ﬁeld tests. SAI (stratospheric aerosol injection) technology builds upon artiﬁcially manipulating the environment and thereafter, if deployed, might generate irreversible changes to natural systems. Climate manipulation strongly contradicts our understanding and experience of how to respect and live in harmony with Mother Nature, and therefore, this technology is not something we see as a part of our chosen future.” Åsa Larsson Blind’s statement in the questions and answers part shows the potential long term implications of the SCoPEx project’s planned test ﬂight in Giron/Kiruna on indigenous peoples’ standing in relation to geoengineering research. The Sámi Council’s intervention against the SCoPEx project When Blind was asked about next steps in relation to the SCoPEx project, she stated that the Sámi Council had written an open letter to Harvard University for which they were collecting co-signatures from indigenous peoples’ organizations from around the world before sending it: From the above, we can see how the SCoPEx project provided a reference point for the deepening of a pre-existing alliance of green civil society organizations and indigenous peoples organizations centered on the HOME campaign, with the ﬁrst step consisting of the Sámi Council opting to join it, followed by its eﬀort to mobilize indigenous peoples from around the world, with some apparent success. As shown above, the key discursive nodal points that constitute the alliance of indigenous peoples’ organizations and environmental civil society organizations, and high proﬁle environmental scholars and activists, is their shared articulation of themselves as respecting the sanctity of Mother Earth, advocating for drastic emission cuts and nature-based solutions to climate change. Geoengineering is posited as embodying the opposite of these values, as it is seen as expressive of colonial, extractivist mastery over nature mentalities, and as a legitimation device for the status quo, providing the billionaires who fund it with the promise of a solution that avoids taking any drastic measures on emission cuts. The framing is one of antagonistic opposition between those ﬁghting for the sanctity of Mother Earth and the predatory elites who try to solve the problems they have created by over-exploiting nature, through further, deliberate manipulation of it, where they stand to beneﬁt as the providers of the solutions to the problems they created. Through the letter “we hope that we can show that we have a joint position and that we actually can inﬂuence the discussion. And to focus on what we should discuss. And I would like to also touch upon what Tom [Goldtooth] said, because I believe that it is very important. I’m convinced that indigenous peoples’ knowledge has the potential to play a signiﬁcant role and a way toward a sustainable future for all.... So, this step that we are taking, is also to that we, as indigenous peoples, we want to contribute to the solutions. And we do not see solar geoengineering as part of that. The Sámi Council’s intervention against the SCoPEx project So, we hope that our letter can help us get into the discussion.” We can see from the below statement from Blind that the SCoPEx project team, through its conduct in failing to consult the Sámi people on their planned test ﬂight in the Sámi domicile area provided an eﬀective antagonistic reference point in relation to which a global indigenous coalition opposed to geoengineering could be mobilized: “We have taken this initiative, but it is on behalf of indigenous voices everywhere. We see the response that we get to this letter, that we have major indigenous organizations from all over the world. This is a joint position. So even if this is our initiative, I see this as a joint cause, and supporting the letter for indigenous voices everywhere, because we have broad support for the letter... is not a Sámi position, it is not an Arctic position, it’s a joint indigenous position from all over the world. And we can also show with the position and the broad support that I am Frontiers in Climate frontiersin.org Discussion From the above, we can see that the SCoPEx project team’s decision to plan a test ﬂight in Giron/Kiruna without consulting Frontiers in Climate 08 10.3389/fclim.2023.994193 10.3389/fclim.2023.994193 Oksanen representatives of the Sámi people has resulted in the Sámi Council joining forces with HOME aﬃliated organizations in opposing geoengineering research and rallying further support from indigenous organizations around the world. This is a concerning outcome to those in favor of geoengineering research, given the Sámi Council’s track record of global leadership among indigenous peoples, as exempliﬁed by its role within the WCIP and having held the UNPFII chairpersonship twice. The potential implications of this to indigenous peoples’ standing on the question of engineering research have already been foreshadowed by the Sámi Council’s success in recruiting indigenous peoples’ organizations from around the world as co-signatories of their open letter to Harvard University, demanding halting of the SCoPEx project. The Sámi Council’s and the HOME coalition’s critical voices on geoengineering were further ampliﬁed by their success in recruiting some of the world’s most inﬂuential environmental scholars and activists to their joint public webinar. consultations with representatives of the Sámi people because they hoped they would be unaware of the project and not challenge it. As geoengineering is envisioned as a potential part of the solution to the humanitarian crisis of climate change, and Sámi reindeer herding depends on a delicate ecological balance, the SCoPEx project also fails to live up to Sphere’s humanitarian protection principle of minimizing risks to those potentially negatively aﬀected by humanitarian programmes (Sphere Project, 2018, p. 38). The SCoPEx project also failed to live up to Sphere’s humanitarian protection principle (4) of helping vulnerable people claim their rights (Sphere Project, 2018, p. 43) by refraining from seeking the Sámi people’s free prior consent as required under Article 32 of the UNDRIP (2007). p q The cancellation of the SCoPEx Kiruna test ﬂight is a good outcome for indigenous peoples, as the Sámi Council’s intervention was credited as a signiﬁcant contributing factor by the media (Doyle, 2021, 2022; Goering, 2021), it can potentially serve as a precedent for indigenous peoples’ right to free and informed consent under Article 32 of the (UNDRIP, 2007) applying to geoengineering research. Discussion With such a precedent, and with such a strong and well-organized coalition of indigenous peoples’ organizations and environmental civil society organizations, it is unlikely that any geoengineering project could be carried out without the knowledge or consent of aﬀected indigenous peoples. This is important from a humanitarian perspective, as indigenous peoples are amongst those most vulnerable to localized climate and weather pattern change that could result from the implementation of geoengineering projects. Reﬂecting the perceived normative power of indigenous people to confer and deny legitimacy to geoengineering research projects, SCoPEx lead scientist, David Keith has said, regarding the feasibility of carrying out the Giron/Kiruna test ﬂight following the Sámi Council’s intervention: “There is no question that in the public battle, if it is Harvard against indigenous peoples, we can’t proceed. That’s just a reality” (Doyle, 2022). It is a signiﬁcant and rare achievement for indigenous peoples to gain that kind of leverage. Eﬀectively the kind of veto power they have unsuccessfully sought against extractive projects in their lands and waters for decades. As demonstrated above, indigenous peoples and environmental civil society organizations do not always agree on how to safeguard nature or what aspects of it should be protected, and there have been instances of open conﬂict between them. Application of Laclau and Mouﬀe’s discourse theory enabled perceiving how the manner in which the SCoPEx project’s test ﬂight in Giron/Kiruna was planned, rather less than transparently, provided a reference point for the articulation of a chain of equivalency between the Sámi council and environmental civil society organizations in opposition to geoengineering research. The SCoPEx project, funded in large part by wealthy individuals like Bill Gates (Tollefson, 2018), lends itself very well to the framing we saw in the webinar. Of it being the product of a small, unaccountable clique of scientists and ﬁnancers pushing a techno-ﬁx to the climate crisis, undermining the urgency for cutting emissions and ﬁnding nature-based solutions. With the reference point of antagonism against the “geo clique” and its predatory mastery over nature mentality, it is easy for indigenous peoples’ organizations and environmental civil societies to unite in the name of their shared goals of respecting the sanctity of Mother Earth and nature-based solutions to climate change. For indigenous organizations, a close relationship with environmental activist groups may bring risks as well as beneﬁts. 3 This version of the SCoPEx Advisory Committee’s (2021) mission statement was updated in June 2021, after receiving the Sámi Council’s open letter, while the statement announcing the formation of the SCoPEx Advisory Committee (2020a) and outlined its mandate contained no reference to indigenous leaders, so the language on indigenous leaders was likely added in reaction to the Sámi Council’s open letter. 2 The SCoPEx Advisory Committee (2021) was established in 2019 to, in its own words, ‘provide oversight on the appropriateness of the research and advice on the governance of SCoPEx’, and to ‘ensure that any and all SCoPEx project is undertaken in a transparent, responsible, and legitimate manner by ensuring that it contributes to scientiﬁc understanding and is guided by meaningful public engagement’. The committee consists of former civil servants, scientists and academics from ﬁelds like physics, environmental law and climate science, with 10 current members at the time of writing, and 2 former members (SCoPEx Advisory Committee, 2020b). Frontiers in Climate frontiersin.org Discussion If the discussion begins with acknowledging all the valuable ways with which they contribute to managing the ecosystem and by extension the climate, and how these practices can be protected and further expanded, it is much more likely that indigenous peoples can see value in interventions that manage the climate. A research project that exempliﬁes this approach is the co-Nordic project “How to preserve the tundra in a warming climate”4 which attempted to use the potential of reindeer for maintaining the climate-cooling ecosystem services that the tundra oﬀers. The project was carried out in collaboration with reindeer herding Sámi (Horstkotte et al., 2017). Such knowledge exchange and co-production between indigenous peoples and researchers holds the key to an open, productive discussion on all tools available for responding to the climate crisis, and how to protect indigenous peoples, their assets and values in the event of any potential intervention. Another risk in working with environmental civil society organizations, is that their and indigenous peoples’ perceptions of what respecting the sanctity of Mother Earth means can diverge. The ETC Group’s reaction to the Great Barrier Reef cloud brightening experiment oﬀers a cautionary tale. As stated above, the research group that carried out the Great Barrier Reef cloud brightening experiment followed the principles of UNDRIP Article 32 and humanitarian principles of safeguarding those most vulnerable, and helping them claim their rights, by seeking and obtaining the prior informed consent of the local indigenous people. The experiment shows how solar geoengineering, or at least marine cloud brightening, can be framed diﬀerently discursively, as it was emphasized by the project team that the cloud brightening technique is premised on speeding up a “natural” process of marine cloud formation by spraying atomised sea water into the air, which the observing Manduburra representative went along with: “We welcome scientiﬁc research where Indigenous people and the rest of Australia work together to maintain the reef ecosystem for future generations... This technology might help prevent bleaching and we like that it uses no chemicals and relies on natural processes” (Readfearn, 2020). The ETC Group did not accept this discursive framing, describing the research project as “risky,” “shockin” and in contravention to a claimed CBD moratorium on outdoor geoengineering experiments (ETC Group, 2020). Discussion On one hand, they serve well to amply indigenous organizations’ voices, as evidenced by the impressive line-up of panelists that participated in the webinar co-hosted by the Sámi Council and environmental civil society organizations. On the other hand, when indigenous organizations side with environmental civil society groups deeply committed to opposing geoengineering, it risks creating the perception of them being dogmatically opposed geoengineering. If public pressure starts building for using geoengineering solutions as a last resort for averting climate disaster, this could result in the side-lining of indigenous organizations with outspoken opposition. By stating that SAI research should not be carried out in any ecosystem in the absence of global consensus, the Sámi Council might have overplayed their hand by setting an impossibly high bar It is also remarkable that the SCoPEx project had previously planned a test ﬂight in Arizona (Chen, 2017), that according to Goldtooth was canceled due to local indigenous peoples’ objections to the lack of consultations. Though the SCoPEx Advisory Committee (2019)2 was established well after this, its mission statement includes making “concerted eﬀorts to consult, especially with those who have experienced historical barriers to participation, including Indigenous and local leaders.” Thus, they should, by their own admission, have known of their obligation to obtain the free, prior informed consent of aﬀected indigenous peoples under Article 32 of the UNDRIP (2007).3 From the outside, it appears that they refrained from 09 Frontiers in Climate frontiersin.org 10.3389/fclim.2023.994193 10.3389/fclim.2023.994193 Oksanen that eﬀectively amounts to outright rejection of SAI, which could lead to their exclusion from future discussion on its governance. reﬂective surfaces to shrubs and trees has long been observed in Alaska, where there is an absence of semi-tame caribou (Sturm et al., 2005). While the Sámi people have been making this signiﬁcant contribution to solar radiation management unwittingly, it is precisely practices like Sámi reindeer herding, that are expressive of the kinds of long-standing practices indigenous peoples have institutionalized over countless generations, which they know to be sustainable and contributing in many ways to upholding the current ecosystem. Such practices must be recognized, understood and supported for a meaningful dialogue on geoengineering with indigenous peoples to be possible. Funding For a constructive dialogue between scientists interested in geoengineering research and indigenous peoples, engagement should start with acknowledgment and understanding of the forms of solar radiation management indigenous peoples area already contributing. The Sámi people have already been involved in research that conﬁrms their signiﬁcant contribution to mitigating climate change by protecting the reﬂective surfaces of the snowy tundras by directing reindeer grazing. Millions of kilometers are natural reﬂective surfaces are now endangered by creeping treelines, as invading trees and shrubs are darkening them (Chapin et al., 2005; Aune et al., 2011). In Fennoscandia, the systematically directed reindeer grazing by the Sámi has contributed to halting the creeping treeline, maintaining vast reﬂective surfaces that could otherwise have been lost (Olofsson et al., 2001, 2009; Pajunen et al., 2012). By contrast, loss of such This paper was informed by research carried out with funding by the British Academy’s Virtual Sandpits Follow on Funding Just Transitions funding scheme, grant number: VSFoFJT\100028. Data availability statement The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author. Discussion The ETC’s reporting on the research contains many of the same tropes identiﬁed above, including that this is a project of the geoengineering-clique pushing for a techno- ﬁx to climate change that they can personally proﬁt from, and that it provides an alibi for unabated fossil fuel driven emissions, distracting from the need for drastic emission cuts (ETC Group, 2020). The report uses a negative photographic image of the white research vessel, making it appear as if the vessel is black and is spraying a black mist into the air. As ETC Group’s reporting never acknowledges the Manduburra’s blessing of the project, it would appear that indigenous peoples’ right to free prior informed consent to projects aﬀecting them only matters to the ETC Group insofar as the indigenous peoples in question share the ETC Group’s opinions on the projects. This example shows that indigenous peoples’ organizations need to have a critical awareness of their non- indigenous partner organizations having their own agendas that are not always aligned with their interests, and that they might not treat them seriously when that is the case. Ethics statement Written informed consent was not obtained from the individual(s) for the publication of any potentially identiﬁable images or data included in this article. 4 https://www.researchgate.net/project/NCoE-Tundra-How-to-preserve- the-tundra-in-a-warming-climate Author contributions The author conﬁrms being the sole contributor of this work and has approved it for publication. Frontiers in Climate References Alam, S., and Al Faruque, A. (2019). From sovereignty to self-determination: emergence of collective rights of indigenous peoples in natural resources management. Georget. Environ. Law Rev. 32, 59–84. Goering, L. (2021). Sweden Rejects Pioneering Test of Solar Geoengineering Tech. London: Thomson Reuters Foundation. Available online at: https://news.trust.org/item/ 20220117030451-2608r/ (accessed July 13, 2022). Greenpeace Sweden, Jordens Vänner/Friends of the Earth Sweden, Naturskyddsföreningen/Swedish Society for Nature Conservation (SSNC), Action Group on Erosion, Technology and Concentration (ETC Group), Biofuelwatch, Center for International Environmental Law (CIEL), Climate Justice Alliance (CJA), Friend of the Earth International, Heinrich Böll Foundation, Indigenous Environmental Network (IEN), WhatNext? (2020). Open Letter to Swedish Government. Geoengineering Monitor. Available online at: https://www.geoengineeringmonitor.org/2021/02/letter-to-the- swedish-space-corporation-on-planned-scopex-test-ﬂight/?print=pdf (accessed July, 13, 2022). Artic Council. (2023). Permanent Participants. Finland: Artic Council. Available online at: https://www.arctic-council.org/about/permanent-participants/ (accessed November 11, 2022). Aune, S., Hofgaard, A., and Söderström, L. (2011). Contrasting climate- and land-use- driven tree encroachment patterns of subarctic tundra in northern Norway and the Kola Peninsula. Can. J. For. Res. 41, 437–449. doi: 10.1139/X10-086 Chapin, F. S., Sturm, M., Serreze, M. C., McFadden, J. P., Key, J. R., Lloyd, A. H., et al. (2005). Role of land-surface changes in arctic summer warming. Science. 310, 657–660. doi: 10.1126/science.1117368 HOME Manifesto (2018). Hands OﬀMothe Earth! Manifesto Against Geoengineering. Geoengineering Monitor. Available online at: https://www.geoengineeringmonitor. org/2018/10/hands-oﬀ-mother-earth-manifesto-against-geoengineering/?print=pdf (accessed July 13, 2022). Chen, S. (2017). Helping Hand or Hubris? College Park: APS News, p. 26. Coulthard, G. S. (2014). Red Skin, White Masks: Rejecting the Colonial Politics of Recognition. Minneapolis, MN: University of Minnesota Press. doi: 10.5749/minnesota/9780816679645.001.0001 Horstkotte, T., Utsi, T. Å., Larsson-Blind, A., Bruggess, P., Johansen, P., Käyhk,ö, J., et al. (2017). Human-animal agency in reindeer management: Sami herders’ perspectives on vegetation dynamics under climate change. Ecosphere 8, e01931. doi: 10.1002/ecs2.1931 Cowger, T. W. (1999). The National Congress of American Indians: The Founding Years. Lincoln, NE: University of Nebraska Press. Deranger, E. T., Sinclair, R. B. G., McGregor, D., and Gobby, J. (2022). Decolonizing climate research and policy: making space to tell our own stories, in our own ways. Commun. Dev. J. 57, 52–73. doi: 10.1093/cdj/bsab050 ICCPR (1966). International Covenant on Civil and Political Rights. New York, NY: United Nations. Available online at: http://www.ohchr.org/EN/ProfessionalInterest/ Pages/CCPR.aspx (accessed July 13, 2022). Dieng, A. (2021). The Sahel: challenges and opportunities. Int. Rev. Red Cross. 103, 765–779. doi: 10.1017/S1816383122000339 ICESCR (1966). International Covenant on Economic, Social and Cultural Rights. New York, NY: United Nations. References Available online at: https://www.ohchr.org/en/instruments- mechanisms/instruments/international-covenant-economic-social-and-cultural-rights (accessed July 13, 2022). Doyle, A. (2021). Indigenous Peoples Urge Harvard to Scrap Solar Geoengineering Project. London: Thomson Reuters Foundation. Available online at: https://news.trust. org/item/20210609202041 gptbr/?source packageandid 45631f30 69bb 41a0 8f5f Jones, R. (2019). Climate change and indigenous health promotion. Global Health Promot. 26, 73–81. doi: 10.1177/1757975919829713 org/item/20210609202041 gptbr/?sourc 10996b3479f4 (accessed July 13, 2022). Doyle, A. (2022). After Sun-Dimming Setback, Geoengineers Seek a Diplomatic Fix. London: Thomson Reuters Foundation. Available online at: https://news.trust.org/item/ 20220117030451-2608r/ (accessed July 13, 2022). Jull, P. (2003). “The politics of sustainable development,” in Indigenous Peoples: Resource Management and Global Rights, eds S. Jerntoft, H. Minde, and R. Nilsen (Delf: Eubron Delft), 121–44. Doyle, C. M. (2015). Indigenous Peoples, Title to Territory, Rights and Resources: The Transformation of the Role of Free Prior and Informed Consent. London: Routledge. Kemner, J. (2011). Lobbying for global indigenous rights: the world council of indigenous peoples (1975–1997). Forum Inter. Am. Res. FIAR 4, 2 Dylan, A., Smallboy, B., and Lightman, E. (2013). “Saying no to resource development is not an option”: economic development in moose Cree ﬁrst nation. Alberta Law Rev. 51, 59–90. doi: 10.3138/jcs.47.1.59 Laclau, E., and Mouﬀe, C. (2001). Hegemony and Socialist Strategy: Towards a Radical Democratic Politics, 2nd ed. London: Verso. Lawrence, R., and Kløcker, L. R. (2016). ““Då är det inte renskötsel”—Konsekvenser av en gruvetablering i Laver, Älvsbyn, för Semisjaur Njarg sameby,” in Stockholm Environment Institute, Project Report (Stockholm: Stockholm Environment Institute), 2016–01. Eide, A. (2006). Rights of indigenous peoples: achievements in international law during the last quarter of a century. Netherlands Yearbook Int. Law 37, 155–212. doi: 10.1017/S0167676806001553 Erickson, B. (2020). Anthropocene futures: linking colonialism and environmentalism in an age of crisis. Environ. Plan. D Soc. Space 38, 111–128. doi: 10.1177/0263775818806514 Lile, H. K. (2006). “A new era for indigenous peoples: the United Nations permanent forum on indigenous issues,” in Gáldu Cála Journal of Indigenous Peoples Rights. No 2/2006 (Kautokeino: Resource Centre for the Rights of Indigenous Peoples). ETC Group (2020). Geoengineers Test Risky Planetary Engineering Scheme in Australia. ETC Group. Available online at: https://www.etcgroup.org/content/geoengineers-test- risky-planetary-engineering-scheme-australia (accessed October 24, 2012). Lindahl, K., Zachrisson, B., Viklund, A., Roine, M., Fjellborg, S., Johansson, D. A., et al. (2016). “Konﬂikter om gruvetablering: Lokalsamhällets aktörer och vägar till hållbarhet,” in Länstyrellsen. Report 2/2016 (Luleå: Länsstyrelsen, Norrbotten). ETC Group. (2012). Hands OﬀMother Earth. ETC Group. Available online at: https:// www.etcgroup.org/content/hands-mother-earth-0 (accessed July, 14, 2022). Acknowledgments I thank my colleagues from the Carbon Elites Collective, Keri Facer (PI), Peter Newell, Pablo Suarez, María Estrada Fuentes, 4 https://www.researchgate.net/project/NCoE-Tundra-How-to-preserve- the-tundra-in-a-warming-climate 10 frontiersin.org frontiersin.org 10.3389/fclim.2023.994193 Oksanen 10.3389/fclim.2023.994193 that could be construed as a potential conﬂict of interest. Jeremy Brice, Antonia Layard, and Kendra Allenby, as our research collaboration on the British Academy funded project on Just Transitions has informed this paper. I also thank Lauri Oksanen for sharing his insights on working with Sámi reindeer herders on the How to preserve the tundra in a warming climate project. Publisher’s note All claims expressed in this article are solely those of the authors and do not necessarily represent those of their aﬃliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Conﬂict of interest The author declares that the research was conducted in the absence of any commercial or ﬁnancial relationships References Low, S., Baum, C. M., and Sovacool, B. K. (2022). Taking it outside: exploring social opposition to 21 early-stage experiments in radical climate interventions. Energy Res. Soc. Sci. 90, 1–21. doi: 10.1016/j.erss.2022.102594 ETC Group’s Facebook Page (2021). Solar Geoengineering: Warnings from Scientists, Indigenous Peoples, Youth, and Climate Activists. ETC Group. (accessed July 13, 2022). Lundmark, L. (2008). Stulet Land: Svensk Makt på Samisk Mark. Stockholm: Ordfront. Gannon, K. E., and Hulme, M. (2018). Geoengineering at the “edge of the world”: exploring perceptions of ocean fertilisation through the Haida Salmon Restoration Corporation. Geo Geogr. Environ. 5, e54. doi: 10.1002/geo2.54 McFarlane, P. (1993). Brotherhood to Nationhood: George Manuel and the Making of the Modern Indian Movement. Toronto: Between the Lines. 11 Frontiers in Climate frontiersin.org Oksanen 10.3389/fclim.2023.994193 McLean, K., Castillo, A., Johnston, S., Gross, T., Vierros, M., and Noa, R. (2009). Report of the Indigenous Peoples’ Global Summit on Climate Change. SCoPEx Advisory Committee (2020b). Advisory Committee Members. SCoPEx Advisory Committee. Available online at: https://scopexac.com/advisory-committee- members/ (accessed November 30, 2022). Normann, S. (2021). Green colonialism in the Nordic context: exploring Southern Saami representations of wind energy development. J. Commun. Psychol. 49, 77–94. doi: 10.1002/jcop.22422 SCoPEx Advisory Committee (2021). Workplan and Operating Guidelines from the Advisory Committee to the Stratospheric Controlled Perturbation Experiment (SCoPEx) Project. Available online at: https://scopexac.com/wp-content/uploads/ 2021/05/SCoPEx-Advisory-Committee-External-Document_Website_Final_5_21.pdf (accessed November 30, 2022). Nyyssönen, J. (2007). Everybody recognized that we were not white”: Sami Identity Politics in Finland, 1945–1990. Thesis submitted for the degree of Doctor Artium Department of History Faculty of Social Sciences University of Tromsø. SCoPEx Advisory Committee. (2019). Statement from Louise Bedsworth, PhD On Serving as Chair of the Advisory Committee for the SCoPEx Project. Available online at: https://scopexac.com/july-29-2019/ (accessed February 8, 2023). Olofsson, J., Kitti, H., Rautiainen, P., Stark, S., and Oksanen, L. (2001). Eﬀects of summer grazing by reindeer on vegetation structure, productivity and nutrient cycling in the North Fennoscandian tundra. Ecography 24, 13–24. doi: 10.1034/j.1600-0587.2001.240103.x Sikka, T. (2020). Activism and neoliberalism: two sides of geoengineering discourse. Capit. Nat. Soc. 31, 84–102 doi: 10.1080/10455752.2018. 1554690 Olofsson, J., Oksanen, L., Callaghan, T., Hulme, P. E., Oksanen, T., and Suominen, O. (2009). Herbivores inhibit climate driven shrub expansion on the tundra. Glob. Chang. Biol. 15, 2681–2693. doi: 10.1111/j.1365-2486.2009.01935.x Sphere Project (2018). Humanitarian Charter and Minimum Standards in Humanitarian Response. Geneva: Sphere Association. Available online at: https:// spherestandards.org/wp-content/uploads/Sphere-Handbook-2018-EN.pdf (accessed February 8, 2023). Pajunen, A., Virtanen, R., and Roininen, H. (2012). References Browsing mediated shrub canopy changes drive composition and species richness in forest-tundra ecosystems. Oikos 121, 1544–1552. doi: 10.1111/j.1600-0706.2011.20115.x Quijano, A. (2000). Coloniality of power, Eurocentrism, and Latin America. Nepantla Views South 1, 533–580. doi: 10.1177/0268580900015002005 Sturm, M., Schimel, J., Michaelson, G., Welker, J. M., Oberbauer, S. F., Liston, G. E., et al. (2005). Winter biological processes could help convert arctic tundra to Shrubland. BioScience. 55, 17–26. doi: 10.1641/0006-3568(2005)055[0017:WBPCHC]2. 0.CO;2 Readfearn, G. (2020). Scientists trial cloud brightening equipment to shade and cool Great Barrier Reef. Guardian. Available online at: https://www.theguardian.com/ environment/2020/apr/17/scientists-trial-cloud-brightening-equipment-to-shade-and- cool-great-barrier-reef (accessed February 8, 2023). Suiseeya, K. R. M., Zanott, L., and Haapala, K. (2022). Navigating the spaces between human rights and justice: cultivating Indigenous representation in global environmental governance. J. Peasant Stud. 49, 604–628. doi: 10.1080/03066150.2020.18 35869 Reid, J., and Chandler, D. (2018). Being in being: contesting the ontopolitics of indigeneity. Eur. Legacy 23, 251–268. doi: 10.1080/10848770.2017.1420284 Ribeiro, S. (2020). “People’s Resistance Against Geoengineering,” in The Routledge Handbook on Ecosocialism, eds L. Brownhill, S. Engel-Di Mauro, T. Giacomini, A. Isla, M. Löwy, and T. E. Turner (London: Routledge), 223–234. doi: 10.4324/97804293414 27-25 Tavanti, M. (2003). Las Abejas: Paciﬁst Resistance and Syncretic Identities in a Globalizing Chiapas. London: Routledge. Tavanti, M. (2003). Las Abejas: Paciﬁst Resistance and Syncretic Identities in a Globalizing Chiapas. London: Routledge. The Anchorage Declaration (2009). Indigenous Peoples’ Global Summit on Climate Change. Anchorage: Indigenous Peoples’ Global Summit on Climate Change. Available online at: https://unfccc.int/resource/docs/2009/smsn/ngo/168.pdf (accessed July 13, 2022). Rodgers, K., and Ingram, D. (2019). Decolonizing environmentalism in the Artic? Greenpeace, complicity and negotiating the contradictions of solidarity in the Inuit Nunangat. Interface J. Soc. Movem. 11, 1–34. Thiele, L. P. (2019). Geoengineering and sustainability. Environ. Polit. 28, 460–479. doi: 10.1080/09644016.2018.1449602 Thiele, L. P. (2019). Geoengineering and sustainability. Environ. Polit. 28, 460–479. doi: 10.1080/09644016.2018.1449602 Sámi Council Open Letter to Harvard University (2021). Indigenous Peoples Call on Harvard to Shut Down the SCoPEx Project. Available online at: https://static1. squarespace.com/static/5dfb35a66f00d54ab0729b75/t/60c0a4bac8e3952583139537/ 1623237819160/Indigenous+Peoples+call+on+Harvard+to+shut+down+the+SCoPEx+ project.pdf (accessed July 13, 2022). Tollefson, J. (2018). First sun-dimming experiment will test a way to cool Earth. Nature 563, 613–615. doi: 10.1038/d41586-018-07533-4 UNDRIP (2007). United Nations Declaration on the Rights of Indigenous Peoples. New York, NY: United Nations. Available online at: https://www.un.org/development/ d /i di l /d l ti th i ht f i di l ht l# UNDRIP (2007). United Nations Declaration on the Rights of Indigenous Peoples. New York, NY: United Nations. References Available online at: https://www.un.org/development/ desa/indigenouspeoples/declaration-on-the-rights-of-indigenous-peoples.html#: $\sim$:text=The%20United%20Nations%20Declaration%20on,%2C%20Bangladesh %2C%20Bhutan%2C%20Burundi%2C (accessed July 13, 2022). Sámi Council Open Letter to the SCoPEx Advisory Committee (2021). Regarding SCoPEx Plans for Test Flights at the Swedish Space Corporation in Kiruna. Kárášjohka/Karasjok, Norway: Sámiráddi. Available online at: https://static1.squarespace. com/static/5dfb35a66f00d54ab0729b75/t/603e2167a9c0b96ﬀb027c8d/1614684519754/ Letter+to+Scopex+Advisory+Committee+24+February.pdf (accessed July 13, 2022). Sámi Council Open Letter to the SCoPEx Advisory Committee (2021). Regarding SCoPEx Plans for Test Flights at the Swedish Space Corporation in Kiruna. Kárášjohka/Karasjok, Norway: Sámiráddi. Available online at: https://static1.squarespace. / / dfb f d b b / / b ﬀb d/ / Vidal, J. (2016). Sami Reindeer Herders Battle Conservationists and Miners to Cling on to Arctic Culture. The Observer. Available online at: https://www.theguardian. com/global-development/2016/feb/21/sami-people-reindeer-herders-arctic-culture#:~: text=Sami%20reindeer%20herders%20battle%20conservationists%20and%20miners %20to%20cling%20on%20to%20Arctic%20culture,-This%20article%20is&text=When %20Europe’s%20indigenous%20Arctic%20people,used%20by%20British%20sheep %20farmers (accessed February 8, 2023). Sámiráddi (2015). Saami Representative to the UN: Permanent Forum. Norway: Sámiráddi. Available online at: https://www.saamicouncil.net/news-archive/saami- representative-to-the-un-permanent-forum (accessed July 13, 2022). Sanders, D. E. (1977). The Formation of the World Council of Indigenous Peoples. Copenhagen: International Work Group for Indigenous Aﬀairs Sanders, D. E. (1989). The UN working group on indigenous populations. Hum. Rights Quart. 11, 406–433. Voyatzis-Bouillard, D., and Kelman, I. (2021). Do climate change interventions impact the determinants of health for paciﬁc island peoples? A literature review. Contemp. Pac. 33, 466–496. doi: 10.1353/cp.2021.0039 Saul, B. (2016). Indigenous Peoples and Human Rights: International and Regional Jurisprudence. Oxford: Hart Publishing. doi: 10.5040/9781474201995 Whyte, K. P. (2018). Indigeneity in geoengineering discourses: some considerations. Ethics Policy Environ. 21 289–307. doi: 10.1080/21550085.2018.15 62529 Scaﬃdi, L. (2021). A Saami Childhood Sets the Stage for UN Indigenous Forum. New York, NY: UN News. Available online at: https://news.un.org/en/audio/2021/04/1090092 (accessed July 13, 2022). Witter, R., Suiseeya, M., Gruby, R. L., Hitchner, R., Maclin, E. M., Bourque, M., et al. (2015). Moments of inﬂuence in global environmental governance. Environ. Polit. 24, 894–912 doi: 10.1080/09644016.2015. 1060036 Schramm, P. J., Al Janabi, A. I., Campbell, L. W., Donatuto, J. L., and Gaughen, S. C. (2020). How indigenous communities are adapting to climate change: insights from the climate-ready tribes initiative. Health Aﬀairs 39, 2153–2159. doi: 10.1377/hlthaﬀ.2020.00997 Zurba, M., and Papadopoulos, A. (2021). Indigenous participation and the incorporation of indigenous knowledge and perspectives in global environmental governance forums: a systematic review. Environ. Manag. 69, 2616. doi: 10.1007/s00267-021- 01566-8 SCoPEx Advisory Committee (2020a). Terms of Reference Established by Harvard University. SCoPEx Advisory Committee. Available online at: https://scopexac. com/terms-of-reference-established-by-harvard-university/ (accessed November 11, 2022). Frontiers in Climate 12 frontiersin.org
https://openalex.org/W4242249483	https://pure.ulster.ac.uk/files/91099694/699428_Manuscript.PDF	English	null	Challenges and opportunities for the future of Brain-Computer Interface in neurorehabilitation	null	2,021	cc-by	10,598	General rights Th i h General rights The copyright and moral rights to the output are retained by the output author(s), unless otherwise stated by the document licence. General rights The copyright and moral rights to the output are retained by the output author(s), unless otherwise stated by the docum Unless otherwise stated, users are permitted to download a copy of the output for personal study or non-commercial research and are permitted to freely distribute the URL of the output. They are not permitted to alter, reproduce, distribute or make any commercial use of the output without obtaining the permission of the author(s). If the document is licenced under Creative Commons, the rights of users of the documents can be found at https://creativecommons.org/share-your-work/cclicenses/. If the document is licenced under Creative Commons, the rights of users of the documents can be found at https://creativecommons.org/share-your-work/cclicenses/. Link to publication record in Ulster University Research Portal Published in: Frontiers in Neuroscience Publication Status: Published online: 02/07/2021 DOI: 10.3389/fnins.2021.699428 Document Version Author Accepted version Document Version Author Accepted version General rights The copyright and moral rights to the output are retained by the output author(s), unless otherwise stated by the document licence. Challenges and opportunities for the future of Brain-Computer Interface in neurorehabilitation Simon, C., Bolton, D., Kennedy, N., Soekadar, S., & Ruddy, K. (2021). Challenges and opportunities for the future of Brain-Computer Interface in neurorehabilitation. Frontiers in Neuroscience, 15, 1-8. Article 699428. Advance online publication. https://doi.org/10.3389/fnins.2021.699428 Link to publication record in Ulster University Research Portal Take down policy a e do po cy The Research Portal is Ulster University's institutional repository that provides access to Ulster's research outputs. Every effort has been made to ensure that content in the Research Portal does not infringe any person's rights, or applicable UK laws. If you discover content in the Research Portal that you believe breaches copyright or violates any law, please contact pure-support@ulster.ac.uk Download date: 24/10/2024 Contribution to the field In the transition from laboratory research to clinical application, BCI for neurorehabilitation remains stuck in the lab despite largely promising findings demonstrating a unique promise for restoring movement in brain injured patients. We take a cautionary standpoint in this review of the state of the art, outlining reasons why current implementations of BCI for clinical intents are fraught due to remaining technical, practical and theoretical issues. We critique each of the barriers to clinical implementation of the technology, providing evidence-based solutions that would advance the state of this field and enable BCI to become more widely used in clinical environments. We believe that this article will impact widely upon neuroscientists, engineers, and clinicians alike, encouraging a consensus approach to promote quality research improving existing BCI implementations. We hope this will pave the way for future work tackling remaining uncertainties regarding the neural mechanisms that underlay how BCI can improve motor function. Ultimately, we hope that this article will accelerate efforts to bring BCI technology closer to the bedside, so that brain injured patients can begin to benefit from what it has to offer. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest Word count: 249 Brain-Computer Interfaces (BCI) provide a unique technological solution to circumvent the damaged motor system. For neurorehabilitation, the BCI can be used to translate neural signals associated with movement intentions into tangible feedback for the patient, when they are unable to generate functional movement themselves. Clinical interest in BCI is growing rapidly, as it would facilitate rehabilitation to commence earlier following brain damage and provides options for patients who are unable to partake in traditional physical therapy. However, substantial challenges with existing BCI implementations have prevented its widespread adoption. Recent advances in knowledge and technology provide opportunities to facilitate a change, provided that researchers and clinicians using BCI agree on standardisation of guidelines for protocols and shared efforts to uncover mechanisms. We propose that addressing the speed and effectiveness of learning BCI control are priorities for the field, which may be improved by multimodal or multi-stage approaches harnessing more sensitive neuroimaging technologies in the early learning stages, before transitioning to more practical, mobile implementations. Clarification of the neural mechanisms that give rise to improvement in motor function is an essential next step towards justifying clinical use of BCI. In particular, quantifying the unknown contribution of non-motor mechanisms to motor recovery calls for more stringent control conditions in experimental work. In review y g p Here we provide a contemporary viewpoint on the factors impeding the scalability of BCI. Further, we provide a future outlook for optimal design of the technology to best exploit its unique potential, and best practices for research and reporting of findings. Colin Simon1, David A. Bolton2, Niamh Kennedy3, Surjo R. Soekadar4, Kathy L. Ruddy1* 1Trinity College Institute for Neuroscience, Trinity College Dublin, Ireland, 2Utah State University, United States, 3Ulster University, United Kingdom, 4Charité – Universitätsmedizin Berlin, Germany Submitted to Journal: Frontiers in Neuroscience Specialty Section: Neuroprosthetics Article type: Mini Review Article Manuscript ID: 699428 Received on: 23 Apr 2021 Revised on: 01 Jun 2021 Journal website link: www.frontiersin.org n rev Submitted to Journal: Frontiers in Neuroscience Specialty Section: Neuroprosthetics Article type: Mini Review Article Manuscript ID: 699428 Received on: 23 Apr 2021 Revised on: 01 Jun 2021 Journal website link: www.frontiersin.org n review Keywords Brain Computer Interface, Stroke, Neurorehabilitation, Transcranial Magnetic Stimulation, Neurofeedback w Author contribution statement CS wrote the first draft and edited the final draft. KR and CS conceptualised the idea for the manuscript. KR, SS, NK and DB worked on subsequent drafts of the manuscript editing and providing feedback. All authors reviewed the final version and were involved in discussions throughout. Abstract Word count: 249 Funding statement KLR and CS are supported by a grant from the Health Research Board of Ireland [HRB-EIA-2019-003]. KR would also like to acknowledge grant funding relating to this work from Enterprise Ireland [SI20203045] and [CS20202099]. Formatted: Font colour: Red Deleted: <#>Clinical Neurotechnology Laboratory, Neurowissenschaftliches Forschungszentrum (NWFZ). Department of Psychiatry and Neurosciences (CCM), Charité , Universitätsmedizin Berlin, Germany¶ w Corresponding Author K.L Ruddy ruddykl@tcd.ie Trinity College Institute of Neuroscience and School of Psychology, Lloyd Building, Trinity College Dublin. College Green, Dublin 2. view y @ Trinity College Institute of Neuroscience and School of Psychology, Lloyd Building, Trinity College Dublin. College Green, Dublin 2. view ct In review Challenges and opportunities for the future of Brain-Computer Interface in neurorehabilitation Simon, C1., Bolton, D.A.E2., Kennedy, N.C3, Soekadar, S.R4., Ruddy, K.L.1 1. Trinity College Institute of Neuroscience and School of Psychology, Trinity College Dublin. Ireland 1. Trinity College Institute of Neuroscience and School of Psychology, Trinity College Dublin. Ireland 2. Department of Kinesiology and Health Science, Utah State University, Logan, Utah, USA 2. Department of Kinesiology and Health Science, Utah State University, Logan, Utah, USA 3. School of Psychology, Ulster University, Northern Ireland, UK Formatted: Font colour: Red 4. Clinical Neurotechnology Laboratory, Neurowissenschaftliches Forschungszentrum (NWFZ), Department of Psychiatry and Neurosciences (CCM), Charité - Universitätsmedizin Berlin, Germany 4. Clinical Neurotechnology Laboratory, Neurowissenschaftliches Forschungszentrum (NWFZ), Department of Psychiatry and Neurosciences (CCM), Charité - Universitätsmedizin Berlin, Germany Deleted: <#>Clinical Neurotechnology Laboratory, Neurowissenschaftliches Forschungszentrum (NWFZ). Department of Psychiatry and Neurosciences (CCM), Charité , Universitätsmedizin Berlin, Germany¶ w Corresponding Author K.L Ruddy Corresponding Author K.L Ruddy Abstract Following a stroke, approximately 77% of survivors are left with some degree of upper limb impairment (Lawrence et al., 2001; Nakayama et al., 1994), which is a key factor in preventing their engagement in normal activities of daily living and rendering them dependant on caregivers. Rehabilitation approaches that actively promote intensive and prolonged functional use of the paretic limb result in the largest gains in movement capability (Veerbeek et al., 2014). However, the gold standard approaches such as Constraint-Induced Movement Therapy (CIMT) require the patient to be capable of producing a sufficient level of functional movement in order to partake (Kwakkel et al., 2015). This prevents participation for those who are more severely impaired, or patients in the early weeks after brain injury who have not yet regained any function. With mounting recent evidence indicating that early intervention is crucial to harness the brain’s endogenous recovery processes (Stinear et al., 2020), therapies that can support the patient to generate appropriate functional patterns of brain activity and motor behaviour are greatly needed, during the period when they are unable to generate actual movement unassisted. In review Abstract Brain-Computer Interfaces (BCI) provide a unique technological solution to circumvent the damaged motor system. For neurorehabilitation, the BCI can be used to translate neural signals associated with movement intentions into tangible feedback for the patient, when they are unable to generate functional movement themselves. Clinical interest in BCI is growing rapidly, as it would facilitate rehabilitation to commence earlier following brain damage and provides options for patients who are unable to partake in traditional physical therapy. However, substantial challenges with existing BCI implementations have prevented its widespread adoption. Recent advances in knowledge and technology provide opportunities to facilitate a change, provided that researchers and clinicians using BCI agree on standardisation of guidelines for protocols and shared efforts to uncover mechanisms. In We propose that addressing the speed and effectiveness of learning BCI control are priorities for the field, which may be improved by multimodal or multi-stage approaches harnessing more sensitive neuroimaging technologies in the early learning stages, before transitioning to more practical, mobile implementations. Clarification of the neural mechanisms that give rise to improvement in motor function is an essential next step towards justifying clinical use of BCI. In particular, quantifying the unknown contribution of non-motor mechanisms to motor recovery calls for more stringent control conditions in experimental work. Here we provide a contemporary viewpoint on the factors impeding the scalability of BCI. Further, we provide a future outlook for optimal design of the technology to best exploit its unique potential, and best practices for research and reporting of findings. review Here we provide a contemporary viewpoint on the factors impeding the scalability of BCI. Further, we provide a future outlook for optimal design of the technology to best exploit its unique potential, and best practices for research and reporting of findings. In review Formatted: Font colour: Red Deleted: Formatted: Font colour: Red Deleted: (Veerbeeck et al., 2014) Formatted: Font colour: Red Formatted: Font colour: Red Deleted: stroke w Brain-Computer Interfaces (BCI) are hailed as a promising approach to overcome paralysis by translating brain signals from movement intentions into computerised or motorised feedback. translating brain signals from movement intentions into computerised or motorised feedback. They can be used to restore, replace, enhance, supplement or improve the natural output of the Central Nervous System (CNS), hence providing opportunities for motor rehabilitation from a range of conditions including spinal cord injury, traumatic brain injury and stroke. Brain-Computer Interface for Neurorehabilitation; Basic premise and scope of the review Non-invasive Brain-Computer (and Brain-Machine) Interfaces provide an advanced technological solution, decoding brain signals directly from the scalp and translating them into movement of a virtual (on screen) or robotic effector. The effector can also be the user’s own limb, with electrical stimulation of muscles triggered by brain activity (Biasiucci et al., 2018; Bai et al., 2020). By closing the disrupted sensorimotor loop and providing tangible feedback, the patient learns to control the effector by motor imagery or movement intentions. Restoring relevant sensory feedback in relation to volitional movement attempts is believed to mobilise the fundamental mechanisms of motor learning (Mrachaz-Kersting et al., 2021). As such, they can engage in mental practise of movement and keep their motor neural circuitry active, warding off the detrimental effects of limb non-use (Buxbaum et al., 2020), its associated white matter degeneration (Egorova et al., 2020), and promote use-dependant neuroplastic processes (Xing & Bai, 2020). Despite largely encouraging evidence suggesting that functional gains are produced exceeding those of standard care (Carvalho et al., 2019; Cervera et al., 2018; Coscia et al., 2019; Hatem et al., 2016; Kovyazina et al., 2019; López-Larraz et al., 2018; McConnell et al., 2017; Monge-Pereira et al., 2017; Raffin & Hummel, 2018; Remsik et al., 2016; Xing & Bai, 2020) substantial challenges with existing BCI implementations have prevented widespread adoption of the technology clinically (Baniqued et al., 2021). Here we provide a viewpoint on the practical, technical and mechanistic factors impeding the scalability of BCI into rehabilitative care packages. Further, we provide a future outlook for optimal design of the technology to best exploit its unique potential, and best practices for research and reporting of findings. view Non-invasive BCI typically consists of three key components: A recorder, a decoder, and an effector. The recorder acquires brain signals from the scalp surface. The decoder analyses the recorded data, and the effector acts upon the information. In most cases the recorder is an electroencephalogram (EEG) detecting scalp electrical fluctuations associated with neuronal activity. In practise, any neural signal could be incorporated into a BCI, and implementations using Magnetoencephalography (MEG) (Buch et al., 2008; Foldes et al., 2015), functional Magnetic Resonance Imaging (fMRI) (Thibault et al., 2018) and functional Near Infrared Spectroscopy (fNIRS) (Soekadar et al., 2021) have also shown merit. The decoder is usually a program run on a computer, extracting desired aspects from the signal and conducting analyses in real-time. Brain-Computer Interface for Neurorehabilitation; Basic premise and scope of the review For BCI to serve as a useful therapeutic tool for neurorehabilitation, solutions that allow users to achieve control within a shorter time frame, and that are effective across a wider range of motor imagery capabilities, are needed to secure the future of this technology. As the neural signals used to drive effectors in an EEG-BCI are heavily influenced by ongoing mental state, the individual capacity to generate high quality mental imagery of movement dictates how easily detectable the relevant motor signals will be (Chavarriaga et al., 2017; Marchesotti et al., 2016). Visual motor imagery, imagining observing yourself perform a movement, produces less pronounced scalp-recorded signals than kinaesthetic motor imagery where the feeling of movement is effectively experienced (Neuper et al., 2005). Kinaesthetic motor imagery produces more easily detectible neural signals, at scalp locations overlaying motor cortical brain regions and additionally modulates corticospinal excitability Formatted: Font colour: Red Formatted: Font colour: Red Deleted: (Corbet et al., 2018). (Biasiucci et al., 2018; Bai et al., 2020(Bai et al., 2020; Biasiucci et al., 2018)). Formatted: Font colour: Red Deleted: (Mrachaz-Kersting et al., 2021; Chavarriaga et al., 2017) Deleted: (Othmer, 2009) Formatted: Font colour: Red Formatted: Font colour: Red Formatted: Font colour: Red Deleted: (XXX Formatted: Font colour: Red Deleted: ) (Majid et al., 2015; Ruddy et al., 2018; Khan et al., 2020) Formatted: Font colour: Red Deleted: (Pfurtscheller et al., 2003; Stieger et al., 2021) Formatted: Font colour: Red Deleted: (Thompson, 2019) In review can be used below motor threshold to provide continuous somatosensory feedback as the BCI signal (Corbet et al., 2018). can be used below motor threshold to provide continuous somatosensory feedback as the BCI signal (Corbet et al., 2018). Brain-Computer Interface for Neurorehabilitation; Basic premise and scope of the review The analysis process may be as simple as measuring amplitude or frequency of ongoing brain activity (Wierzgała et al., 2018), or more complex decompositions of inter-regional connectivity or dynamic changes to spatial patterns of activation (Rathee et al., 2017). The effector of a BCI can take multiple forms. For neurorehabilitation, it may be a device that assists the patient to complete movements, such as a robotic limb (Khan et al., 2020; Soekadar et al., 2019; Tariq et al., 2018), a device that gives virtual (eg. on-screen) feedback to the participant to promote appropriate patterns of neural activity (de Castro-Cros et al., 2020; Kerous et al., 2018; Si-Mohammed et al., 2018), or a trigger to induce electrical stimulation of muscles in order to evoke movement (Bai et al., 2020; Biasiucci et al., 2018). Even in the absence of evoked movement, electrical stimulation In review can be used below motor threshold to provide continuous somatosensory feedback as the BCI signal (Corbet et al., 2018). Practical and technical challenges with clinical implementation of BCI For the BCI participant, learning to control the effector requires multiple practise sessions, viewing continuous feedback and learning by reward (Chavarriaga et al., 2017; Mrachacz- Kersting et al., 2021). While passive/implicit learning is known to play a role in BCI control (Othmer, 2009), most human participants report developing and fine-tuning mental strategies throughout the course of training, usually involving imagination of movement (Khan et al., 2020; Majid et al., 2015; Ruddy et al., 2018), or in the case of brain injured patients, attempts to make movement with the paretic limb (Bai et al., 2020; Balasubramanian et al., 2018; Blokland et al., 2012). Even for neurologically healthy participants, gaining effective control of an EEG-BCI takes many distinct sessions (Pfurtscheller et al., 2003; Stieger et al., 2021). Without seeing tangible results within the first training sessions, it is likely that patients loose motivation to continue investing effort into trying to control the BCI. Another factor known to influence learning is the large variation in individual capability for motor imagery, contributing to the fact that 10-30% (Allison & Neuper, 2010; Lotte & Jeunet, 2015; Vidaurre & Blankertz, 2010) of users never achieve control over the BCI; a phenomenon historically referred to as BCI illiteracy but more recently coined BCI inefficiency (Thompson, 2019). Practical and technical challenges with clinical implementation of BCI Practical and technical challenges with clinical implementation of BCI For the BCI participant, learning to control the effector requires multiple practise sessions, viewing continuous feedback and learning by reward (Chavarriaga et al., 2017; Mrachacz- Kersting et al., 2021). While passive/implicit learning is known to play a role in BCI control (Othmer, 2009), most human participants report developing and fine-tuning mental strategies throughout the course of training, usually involving imagination of movement (Khan et al., 2020; Majid et al., 2015; Ruddy et al., 2018), or in the case of brain injured patients, attempts to make movement with the paretic limb (Bai et al., 2020; Balasubramanian et al., 2018; Blokland et al., 2012). Even for neurologically healthy participants, gaining effective control of an EEG-BCI takes many distinct sessions (Pfurtscheller et al., 2003; Stieger et al., 2021). Without seeing tangible results within the first training sessions, it is likely that patients loose motivation to continue investing effort into trying to control the BCI. Another factor known to influence learning is the large variation in individual capability for motor imagery, contributing to the fact that 10-30% (Allison & Neuper, 2010; Lotte & Jeunet, 2015; Vidaurre & Blankertz, 2010) of users never achieve control over the BCI; a phenomenon historically referred to as BCI illiteracy but more recently coined BCI inefficiency (Thompson, 2019). For BCI to serve as a useful therapeutic tool for neurorehabilitation, solutions that allow users to achieve control within a shorter time frame, and that are effective across a wider range of motor imagery capabilities, are needed to secure the future of this technology. In review As the neural signals used to drive effectors in an EEG-BCI are heavily influenced by ongoing mental state, the individual capacity to generate high quality mental imagery of movement dictates how easily detectable the relevant motor signals will be (Chavarriaga et al., 2017; Marchesotti et al., 2016). Visual motor imagery, imagining observing yourself perform a movement, produces less pronounced scalp-recorded signals than kinaesthetic motor imagery where the feeling of movement is effectively experienced (Neuper et al., 2005). Practical and technical challenges with clinical implementation of BCI Kinaesthetic motor imagery produces more easily detectible neural signals, at scalp locations overlaying motor cortical brain regions, and additionally modulates corticospinal excitability Deleted: (Stinear et al., 2006) Formatted: Font colour: Red Formatted: Font colour: Red Deleted: (SAME TWO REFS AS BEFORE COLIN_BOTH IEEE)… Deleted: M measured via Motor Evoked Potentials (MEPs) in response to Transcranial Magnetic Stimulation (TMS) (Stinear et al., 2006). However, about half of participants find it difficult to perform kinaesthetic motor imagery (Seiler et al., 2017), and those are the lowest performers on BCI (Vuckovic & Osuagwu, 2013). In some circumstances, it may be more beneficial to request that the patient make attempts to execute movements, rather than simply imagine movement (Bai et al., 2020; Balasubramanian et al., 2018; Blokland et al., 2012). For motor imagery-based BCI, multimodal or multi-phase BCI approaches (Fazli et al., 2012; Leamy et al., 2011) may lead to better accuracy, as different neuroimaging modalities may be more sensitive than EEG to detect very weak motor signals (albeit, technologies such as MRI, fNIRS, MEG or TMS may be less practical for long-term practise of BCI). As an example of a potential multi-phase approach, using a BCI based not upon scalp signals but upon muscle responses to TMS over the motor cortex, control of on-screen feedback using motor imagery could be achieved within just two training sessions (Ruddy et al., 2018). Whereas EEG scalp signals associated with movement intentions have poor spatial resolution, TMS can be used to target the motor cortical representations for specific muscles, selectively providing feedback of excitability of corticospinal pathways by measuring the amplitude of motor evoked potentials (MEPs) recorded at the muscle with electromyography (EMG). When tested in a sample of stroke patients, most were capable of learning to increase the excitability of their corticospinal pathways with TMS neurofeedback, using only motor imagery (Liang et al., 2020). Using a multimodal approach it may be possible to train individuals to develop robust motor imagery strategies that optimally excite the targeted brain-muscle pathways using TMS neurofeedback within just two sessions, that then translates to better subsequent performance using functionally relevant signals on an EEG BCI that has better portability options. This hypothesis remains to be tested empirically, and it is notable to point out that the approach may only be applicable to patients who exhibit MEPs when stimulated. Practical and technical challenges with clinical implementation of BCI Approximately 13.4% (Stinear et al., 2017a) are deemed ‘MEP negative’, and those tend to be the most severely affected (Lundquist et al., 2021; Smith et al., 2019; Stinear et al., 2017b). Incorporating multimodality into BCI paradigms may also extend beyond the aforementioned suggestions concerning acquisition modalities. Multimodal feedback (ie. visual plus auditory or somatosensory) can also enhance the BCI learning experience (Sollfrank et al., 2016) and improve the quality of detectable brain signals (Sollfrank et al., 2015). In review In order for BCI for neurorehabilitation to become scalable, it needs to answer to the current requirements of healthcare providers. Namely, it should reduce rather than increase Deleted: Formatted: Font colour: Red Deleted: stroke w burden and need for expert supervision, and instead place high quality rehabilitation into the hands and home of the patient in a cost effective manner. Even if initial control of the EEG BCI can be achieved more quickly using multimodal or multiphase approaches, longer term use over weeks or months would still be required alongside the patient’s standard rehabilitative care in order to promote functional upper limb improvement. Current implementations of EEG- BCI are not well adapted for this purpose as they are cumbersome, require lengthy setup times with wet electrolyte-filled sensors, and a skilled operator to ensure sufficient signal quality, positioning of the headgear and execution of (often custom written and not user friendly) software. Recent technological advances in wireless, high impedance (dry electrode) EEG systems may enable better scalability. Using tablet-based software allowing real-time wireless streaming from a comfortable, wearable EEG cap with dry electrodes, signals of sufficient quality can be recorded even in home environments by elderly participants without assistance (McWilliams et al., 2021; Murphy et al., 2018a, 2018b, 2019). Advancing this new technology to additionally provide real-time feedback to the participant is a necessary next step towards home-based BCI that would allow extended training to be conducted in the weeks and months following brain injury. Further, it encourages the patient to feel in control of their own recovery process, rather than dependant on professionals or their caregivers. To date, existing implementations of wireless BCI for neurorehabilitation are at an early stage of technology readiness, with none reaching even small scale clinical trials (Baniqued et al., 2021). In review Challenges for elucidating mechanisms underlying BCI-induced functional improvements While it is encouraging that many studies are now including explicit measures to monitor training-induced changes to motor neural circuitry (ie. using neuroimaging), measures of the aforementioned unspecific effects are rarely included. Thus, their contribution cannot be evaluated with the currently available evidence. eview related simply to participating in an experiment). iii) Repetition-related effects and iv) Natural effects occurring during the intervention period such as cognitive development (in children/adolescents) or age-related cognitive decline in elderly participants. This list is, however, not exhaustive. Additional to the aforementioned mechanisms, BCI performance may also be influenced by effects from sustained exertion of effort or engagement, feelings of empowerment or competence from achieving control of the BCI, or general improvements in mood or enjoyment resulting from engaging with a challenging gamified task. Even with seemingly adequate control groups, it may be challenging to match aspects such as effort, attention, enthusiasm and enjoyment between those receiving real neurofeedback and those receiving pseudofeedback which may be less intrinsically motivating. While it is encouraging that many studies are now including explicit measures to monitor training-induced changes to motor neural circuitry (ie. using neuroimaging), measures of the aforementioned unspecific effects are rarely included. Thus, their contribution cannot be evaluated with the currently available evidence. eview While the presence of unspecific BCI effects makes it more difficult to draw conclusions on how motor improvement for neurorehabilitation is achieved, it leaves open the intriguing possibility that although BCI training is conducted in the motor domain (ie. using motor imagery), beneficial effects may not be exclusive to the motor system. For instance, it is conceivable that increased effortful focus on the BCI task over a sustained training period may lead to a top-down, generalised improvement in brain health, materialising as motor gains (measured by most studies) but also gains in other (eg. cognitive) domains, which are not routinely quantified in BCI studies. This may materialise in the form of improved executive function, memory, attention or processing speed. Such general improvements may be brought about by, for example, increased blood flow to the brain, enhanced neurochemical environment promoting plasticity induction, or simply increased traffic in neural circuitry sustaining healthy activity-induced myelination processes. Generalised (cross-domain) transfer from trained to untrained tasks is greatest when the trained task requires a high degree of attentional focus and cognitive flexibility (Bavelier et al., 2018; Green & Bavelier, 2008). Challenges for elucidating mechanisms underlying BCI-induced functional improvements Advocating for clinical use of BCI is difficult when the specific mechanisms underlying functional improvements remain largely unknown. In order to make justifiable predictions regarding whether a patient is likely to benefit from BCI training, clinicians need to know what aspects of neural function are being targeted. The vast heterogeneity of available BCI methods further complicates attempts to elucidate mechanisms, as it is likely that different approaches target different aspects of neural circuitry to bring about functional improvements. A key issue to shed light upon across all types of BCI for neurorehabilitation is the potential contribution of unspecific (ie. non-motor) mechanisms for recovery. Ros et al., (2020) name four non- specific factors that may contribute to overall BCI performance improvement. These include i) Factors unique to the neurofeedback environment (such as trainer-participant interaction in a neurotechnology context). ii) Factors that are common across interventions (such as benefits from engaging in a form of cognitive training, as well as psychosocial and placebo mechanisms Deleted: (Ros et al., 2020). These may include, for example, effects from sustained exertion of effort or engagement, feelings of empowerment or competence from achieving control of the BCI, or general improvements in mood or enjoyment resulting from engaging with a challenging gamified task. Deleted: (Ros et al., 2020). These may include, for example, effects from sustained exertion of effort or engagement, feelings of empowerment or competence from achieving control of the BCI, or general improvements in mood or enjoyment resulting from engaging with a challenging gamified task. Deleted: These may include, for example, effects Deleted: Formatted: Font colour: Red Deleted: (Bavelier et al., 2018; Green & Bavelier, 2008) w related simply to participating in an experiment). iii) Repetition-related effects and iv) Natural effects occurring during the intervention period such as cognitive development (in children/adolescents) or age-related cognitive decline in elderly participants. This list is, however, not exhaustive. Additional to the aforementioned mechanisms, BCI performance may also be influenced by effects from sustained exertion of effort or engagement, feelings of empowerment or competence from achieving control of the BCI, or general improvements in mood or enjoyment resulting from engaging with a challenging gamified task. Even with seemingly adequate control groups, it may be challenging to match aspects such as effort, attention, enthusiasm and enjoyment between those receiving real neurofeedback and those receiving pseudofeedback which may be less intrinsically motivating. Challenges for elucidating mechanisms underlying BCI-induced functional improvements In this regard, it is debatable whether the motor imagery BCI is primarily a motor task, or a cognitive task, making direction of transfer difficult to ascertain. Future work may focus on whether motor improvements arise as a result of transfer from improved cognitive function, or whether a portion of the specific motor improvements transfer to improve cognitive function. To test this empirically, design of future BCI studies should routinely include cognitive performance In revi measurements alongside motor function tests, with measures taken at multiple timepoints throughout the learning process. A small proportion of Randomised Controlled Trials (RCTs) investigating BCI for neurorehabilitation have made efforts to measure and/or discuss potential mechanisms that give rise to functional improvements. Of these, candidate neural changes that co-occur with improvement in motor function include enhanced desynchronisation of sensorimotor rhythms (ie. neural oscillations in the alpha 8-12Hz and beta 13-30Hz frequency range) over scalp locations corresponding to motor cortex (Buch et al., 2008; Li et al., 2014; Ono et al., 2015; Prasad et al., 2010), changes in functional connectivity (Biasiucci et al., 2018; Pichiorri et al., 2015; Varkuti et al., 2013; Wu et al., 2020), lateralisation of neural activity (Ramos- Murguialday et al., 2013) or changes to white matter microstructure (Hong et al., 2017; Song et al., 2015). Others have speculated that BCI works by mobilising the brain’s intrinsic learning mechanisms, adapting behaviour using classical and/or operant conditioning giving rise to neural adaptations (Mrachacz-Kersting et al., 2021). To date, there has not been a comprehensive account that successfully resolves the aforementioned different perspectives into a holistic mechanistic model encompassing the electrophysiological, haemodynamic, and neurochemical components. Multimodal investigations measuring BCI-related neural changes simultaneously in each of these different modalities (eg. EEG, fMRI, MR-spectroscopy) are warranted in order to understand how the neural elements interact to bring about functional motor improvements. A point to note is that in none of the above studies were non-motor, unspecific mechanisms tested, so their contribution to improving motor function remains unknown. Transfer of benefits to the non-motor domain were also not quantified, leaving open the possibility that improvement in motor function may be a result of general brain health improvement. In review Elucidating mechanisms of functional improvement from BCI is further complicated by the fact that brain injured patients have widely heterogenous lesions, and lesion size and location do not predict functional outcomes in a straightforward manner (Umarova et al., 2021). Challenges for elucidating mechanisms underlying BCI-induced functional improvements Even in patients with similar extents of impairment, lesion location influences performance of BCI decoding of movement intentions (López-Larraz et al., 2017), and the scalp detected signals are qualitatively different when comparing cortical vs subcortical lesions in particular (López-Larraz et al., 2019). This poses challenges for a ‘one size fits all’ approach to BCI for neurorehabilitation and stresses the importance of adaptive algorithms that do not make rigid a-priori assumptions regarding location or characteristics of scalp detected signals, but rather allow flexibility to detect idiosyncratic patterns of neural activity that could be used to drive the BCI in an individually tailored fashion. Outlook for future scalability and justification of BCI use clinically The field of BCI for neurorehabilitation has benefitted in recent years from collaborative efforts to standardise approaches using the best evidence-based technologies, and with recommendations for best practise in conducting research. For instance, the MT-BCI consortium is a multinational effort collecting the largest existing sample of BCI data across 9 countries, with the objective to deepen understanding of learning mechanisms in mental-task based BCI (MT-BCI), improve efficiency and reliability of MT-BCI and make them more useable for clinical and non-clinical applications (Jeunet et al., 2020). This ‘big data’ approach to BCI breaks away from the typical small scale studies that are characteristic in this field, and may facilitate more advanced analyses techniques such as machine learning. review A key challenge is to make BCI tasks more user friendly, providing motivating feedback in a style that the user finds useful (Kübler et al., 2014; Pillette et al., 2017). Both hardware and software must be simple to use for patients and caregivers alike, which may result in greater enthusiasm towards the technology (Käthner et al., 2017). Tasks should avoid being fixed and repetitive, but rather should have an adaptive nature allowing the user to clearly see progression through stages as performance improves (Jeunet et al., 2016). BCI approaches that focus on assistance with activities of daily living (Soekadar et al., 2016) (particularly bimanual tasks in stroke patients) during physiotherapy may foster motivation and generalization of skills towards everyday life (Soekadar et al., 2019). Additional to this, attempts to improve scientific rigour and reproducibility in neurofeedback research have established the CRED-nf framework for reporting of results, and recommendations for future design of studies (Ros et al., 2020). It is hoped that these collaborative efforts will improve understanding of BCI mechanisms by establishing a degree of standardisation of measurement, and ensuring that adequate experimental controls are in place. In revie Acknowledgements The authors would like to thank Alison Buick, Claire Donnellan and Nicole Wenderoth for helpful comments and discussions relating to this work. Conclusions The technological and practical scalability and clinical justifiability of BCI still pose challenges preventing widespread use for neurorehabilitation. Recent advances in knowledge and Formatted: Font colour: Red Deleted: illiteracy Formatted: Font colour: Red Deleted: illiteracy Deleted: Deleted: (Lövdén et al., 2010; Cremen & Carson, 2017) w technology provide opportunities to facilitate a change, provided that researchers and clinicians using BCI agree on standardisation of guidelines for protocols and shared efforts to uncover mechanisms. Addressing BCI inefficiency and speed of learning are priorities for the field, which may be improved by multimodal or multi-stage BCI approaches harnessing more sensitive neuroimaging technologies in the early learning stages, before transitioning to more practical, mobile implementations. Clarification of the neural mechanisms that give rise to improvement in motor function is an essential next step towards justifying clinical use of BCI. In particular, quantifying the unknown contribution of non-motor mechanisms to motor recovery remains elusive and calls for more stringent control conditions in experimental work. Measurement of additional neural (non-motor) systems and of performance on non-motor tasks is also essential to demonstrate specificity or transfer of the improvements across cognitive and motor domains. If the effects of motor imagery based BCI are found to generalise beyond the motor system, for instance to improve cognitive control or gait, the potential relevance of BCI is expanded presenting an intriguing opportunity for the field. Ultimately, if the benefits are further found to generalise beyond lab-based experimental settings to more ecologically valid aspects affecting quality of life such as competence and autonomy (Cremen & Carson, 2017; Lövdén et al., 2010), the intervention can truly be deemed as effective and worthwhile implementing clinically for neurorehabilitation. In review Funding Sources KLR and CS are supported by a grant from the Health Research Board of Ireland [HRB-EIA- 2019-003]. KR would also like to acknowledge grant funding relating to this work from Enterprise Ireland [SI20203045] and [CS20202099]. Conflicts of Interest The authors declare no conflicts of interest Author Contributions CS wrote the first draft and edited the final draft. KR and CS conceptualised the idea for the manuscript. KR, SS, NK and DB worked on subsequent drafts of the manuscript editing and providing feedback. All authors reviewed the final version and were involved in discussions throughout. REFERENCES Allison, B. Z., & Neuper, C. (2010). Could Anyone Use a BCI? In D. S. Tan & A. Nijholt (Eds.), Brain-Computer Interfaces: Applying our Minds to Human-Computer Interaction (pp. 35–54). Springer. https://doi.org/10.1007/978-1-84996-272-8_3 Bai, Z., Fong, K. N. K., Zhang, J. J., Chan, J., & Ting, K. H. (2020). Immediate and long- term effects of BCI-based rehabilitation of the upper extremity after stroke: A systematic review and meta-analysis. Journal of NeuroEngineering and Rehabilitation, 17(1), 57. https://doi.org/10.1186/s12984-020-00686-2 Balasubramanian, S., Garcia-Cossio, E., Birbaumer, N., Burdet, E., & Ramos-Murguialday, A. (2018). Is EMG a Viable Alternative to BCI for Detecting Movement Intention in Severe Stroke? IEEE Transactions on Biomedical Engineering, 65(12), 2790–2797. https://doi.org/10.1109/TBME.2018.2817688 Baniq ed P D E Stan er E C A ais M Ala mani A Jackson A E Mon Williams In revi Baniqued, P. D. E., Stanyer, E. C., Awais, M., Alazmani, A., Jackson, A. E., Mon-Williams, M. A., Mushtaq, F., & Holt, R. J. (2021). Brain–computer interface robotics for hand rehabilitation after stroke: A systematic review. Journal of NeuroEngineering and Rehabilitation, 18(1), 1–25. https://doi.org/10.1186/s12984-021-00820-8 rehabilitation after stroke: A systematic review. Journal of NeuroEngineering and Rehabilitation, 18(1), 1–25. https://doi.org/10.1186/s12984-021-00820-8 Bavelier, D., Bediou, B., & Green, C. S. (2018). Expertise and generalization: Lessons from action video games. Current Opinion in Behavioral Sciences, 20, 169–173. https://doi.org/10.1016/j.cobeha.2018.01.012 Biasiucci, A., Leeb, R., Iturrate, I., Perdikis, S., Al-Khodairy, A., Corbet, T., Schnider, A., Schmidlin, T., Zhang, H., Bassolino, M., Viceic, D., Vuadens, P., Guggisberg, A. G., & Millán, J. d R. (2018). Brain-actuated functional electrical stimulation elicits lasting arm motor recovery after stroke. Nature Communications, 9(1), 2421. https://doi.org/10.1038/s41467-018-04673-z view Biasiucci, A., Leeb, R., Iturrate, I., Perdikis, S., Al-Khodairy, A., Corbet, T., Schnider, A., Schmidlin, T., Zhang, H., Bassolino, M., Viceic, D., Vuadens, P., Guggisberg, A. G., & Millán, J. d R. (2018). Brain-actuated functional electrical stimulation elicits lasting w https://doi.org/10.1038/s41467-018-04673-z Blokland, Y., Vlek1, R., Karaman, B., Özin, F., Thijssen, D., Eijsvogels, T., Colier, W., revie Blokland, Y., Vlek1, R., Karaman, B., Özin, F., Thijssen, D., Eijsvogels, T., Colier, W., Floor-Westerdijk, M., Bruhn, J., & Farquhar, J. (2012). Author Contributions Detection of event-related desynchronization during attempted and imagined movements in tetraplegics for brain switch control. 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 3967–3969. https://doi.org/10.1109/EMBC.2012.6346835 In review desynchronization during attempted and imagined movements in tetraplegics for brain In rev https://doi.org/10.1109/EMBC.2012.6346835 Formatted: German https://doi.org/10.1109/EMBC.2012.6346835 Buch, E., Weber, C., Cohen, L. G., Braun, C., Dimyan, M. A., Ard, T., Mellinger, J., Caria, A., Soekadar, S., Fourkas, A., & Birbaumer, N. (2008). Think to move: A neuromagnetic brain-computer interface (BCI) system for chronic stroke. Stroke, 39(3), 910–917. https://doi.org/10.1161/STROKEAHA.107.505313 Buch, E., Weber, C., Cohen, L. G., Braun, C., Dimyan, M. A., Ard, T., Mellinger, J., Caria, neuromagnetic brain-computer interface (BCI) system for chronic stroke. Stroke, Buxbaum, L. J., Varghese, R., Stoll, H., & Winstein, C. J. (2020). Predictors of Arm Nonuse in Chronic Stroke: A Preliminary Investigation. Neurorehabilitation and Neural Repair, 34(6), 512–522. https://doi.org/10.1177/1545968320913554 Carvalho, R., Dias, N., & Cerqueira, J. J. (2019). Brain-machine interface of upper limb recovery in stroke patients rehabilitation: A systematic review. Physiotherapy Research International, 24(2). Scopus. https://doi.org/10.1002/pri.1764 Cervera, M. A., Soekadar, S. R., Ushiba, J., Millán, J. D. R., Liu, M., Birbaumer, N., & Garipelli, G. (2018). Brain-computer interfaces for post-stroke motor rehabilitation: A meta-analysis. Annals of Clinical and Translational Neurology, 5(5), 651–663. Scopus. https://doi.org/10.1002/acn3.544 Chavarriaga, R., Fried-Oken, M., Kleih, S., Lotte, F., & Scherer, R. (2017). Heading for new shores! Overcoming pitfalls in BCI design. Brain Computer Interfaces (Abingdon, England), 4(1–2), 60–73. https://doi.org/10.1080/2326263X.2016.1263916 view Corbet, T., Iturrate, I., Pereira, M., Perdikis, S., & Millán, J. del R. (2018). Sensory threshold review neuromuscular electrical stimulation fosters motor imagery performance. In revie NeuroImage, 176, 268–276. https://doi.org/10.1016/j.neuroimage.2018.04.005 In rev Coscia, M., Wessel, M. J., Chaudary, U., Millán, J. D. R., Micera, S., Guggisberg, A., Vuadens, P., Donoghue, J., Birbaumer, N., & Hummel, F. C. (2019). Neurotechnology-aided interventions for upper limb motor rehabilitation in severe chronic stroke. Brain, 142(8), 2182–2197. Scopus. https://doi.org/10.1093/brain/awz181 In r Coscia, M., Wessel, M. J., Chaudary, U., Millán, J. D. R., Micera, S., Guggisberg, A., In r chronic stroke. Brain, 142(8), 2182–2197. Scopus. Cremen, I. A., & Carson, R. G. (2017). Have Standard Tests of Cognitive Function Been Misappropriated in the Study of Cognitive Enhancement? Frontiers in Human Neuroscience, 11. https://doi.org/10.3389/fnhum.2017.00276 de Castro-Cros, M., Sebastian-Romagosa, M., Rodríguez-Serrano, J., Opisso, E., Ochoa, M., Ortner, R., Guger, C., & Tost, D. (2020). Effects of Gamification in BCI Functional Rehabilitation. Frontiers in Neuroscience, 14. https://doi.org/10.3389/fnins.2020.00882 https://doi.org/10.3389/fnins.2020.00882 Egorova, N., Dhollander Thijs, Khlif Mohamed Salah, Khan Wasim, Werden Emilio, & Brodtmann Amy. (2020). Pervasive White Matter Fiber Degeneration in Ischemic Stroke. Stroke, 51(5), 1507–1513. https://doi.org/10.1161/STROKEAHA.119.028143 Fazli, S., Mehnert, J., Steinbrink, J., Curio, G., Villringer, A., Müller, K.-R., & Blankertz, B. (2012). Formatted: German https://doi.org/10.1109/EMBC.2012.6346835 Enhanced performance by a hybrid NIRS–EEG brain computer interface. NeuroImage, 59(1), 519–529. https://doi.org/10.1016/j.neuroimage.2011.07.084 Foldes, S. T., Weber, D. J., & Collinger, J. L. (2015). MEG-based neurofeedback for hand rehabilitation. Journal of NeuroEngineering and Rehabilitation, 12(1), 85. https://doi.org/10.1186/s12984-015-0076-7 iew Egorova, N., Dhollander Thijs, Khlif Mohamed Salah, Khan Wasim, Werden Emilio, & Brodtmann Amy. (2020). Pervasive White Matter Fiber Degeneration in Ischemic Stroke. Stroke, 51(5), 1507–1513. https://doi.org/10.1161/STROKEAHA.119.028143 Fazli, S., Mehnert, J., Steinbrink, J., Curio, G., Villringer, A., Müller, K.-R., & Blankertz, B. Foldes, S. T., Weber, D. J., & Collinger, J. L. (2015). MEG-based neurofeedback for hand rehabilitation. Journal of NeuroEngineering and Rehabilitation, 12(1), 85. https://doi.org/10.1186/s12984-015-0076-7 iew Green, C. S., & Bavelier, D. (2008). Exercising your brain: A review of human brain plasticity and training-induced learning. Psychology and Aging, 23(4), 692–701. https://doi.org/10.1037/a0014345 In review Hatem, S. M., Saussez, G., Della Faille, M., Prist, V., Zhang, X., Dispa, D., & Bleyenheuft, Y. (2016). Rehabilitation of Motor Function after Stroke: A Multiple Systematic Review Focused on Techniques to Stimulate Upper Extremity Recovery. Frontiers in Human Neuroscience, 10, 442. https://doi.org/10.3389/fnhum.2016.00442 In rev Hong, X., Lu, Z. K., Teh, I., Nasrallah, F. A., Teo, W. P., Ang, K. K., Phua, K. S., Guan, C., Chew, E., & Chuang, K.-H. (2017). Brain plasticity following MI-BCI training combined with tDCS in a randomized trial in chronic subcortical stroke subjects: A preliminary study. Scientific Reports, 7(1), 9222. https://doi.org/10.1038/s41598-017- 08928-5 Jeunet, C., Benaroch, C., Cabestaing, F., Chavarriaga, R., Colamarino, E., Corsi, M.-C., Coyle, D., De Vico Fallani, F., Enriquez-Geppert, S., Figueirédo, P., Grosse-Wentrup, M., Kleih, S., Kober, S., Kübler, A., Lotte, F., Maby, E., Mattia, D., Mattout, J., Müller-Putz, G. R., … Wriessnegger, S. (2020, September 11). A user-centred approach to unlock the potential of non-invasive BCIs: An unprecedented international translational effort. https://www.chistera.eu/conference/2020/abstract/camille-jeunet-0 approach to unlock the potential of non-invasive BCIs: An unprecedented international translational effort. https://www.chistera.eu/conference/2020/abstract/camille-jeunet-0 Jeunet, C., Jahanpour, E., & Lotte, F. (2016). Why standard brain-computer interface (BCI) training protocols should be changed: An experimental study. Journal of Neural Engineering, 13(3), 036024. https://doi.org/10.1088/1741-2560/13/3/036024 Käthner, I., Halder, S., Hintermüller, C., Espinosa, A., Guger, C., Miralles, F., Vargiu, E., Dauwalder, S., Rafael-Palou, X., Solà, M., Daly, J. M., Armstrong, E., Martin, S., & Kübler, A. (2017). A Multifunctional Brain-Computer Interface Intended for Home Use: An Evaluation with Healthy Participants and Potential End Users with Dry and Gel-Based Electrodes. Frontiers in Neuroscience, 11. https://doi.org/10.1109/EMBC.2012.6346835 https://doi.org/10.3389/fnins.2017.00286 In review https://doi.org/10.3389/fnins.2017.00286 In re Kerous, B., Skola, F., & Liarokapis, F. (2018). EEG-based BCI and video games: A progress report. Virtual Reality, 22(2), 119–135. https://doi.org/10.1007/s10055-017-0328-x In rev Kerous, B., Skola, F., & Liarokapis, F. (2018). EEG-based BCI and video games: A progress report. Virtual Reality, 22(2), 119–135. https://doi.org/10.1007/s10055-017-0328-x Khan, M. A., Das, R., Iversen, H. K., & Puthusserypady, S. (2020). Review on motor imagery based BCI systems for upper limb post-stroke neurorehabilitation: From In rev Khan, M. A., Das, R., Iversen, H. K., & Puthusserypady, S. (2020). Review on motor imagery based BCI systems for upper limb post-stroke neurorehabilitation: From designing to application. Computers in Biology and Medicine, 123, 103843. https://doi.org/10.1016/j.compbiomed.2020.103843 In Kovyazina, M. S., Varako, N. A., Lyukmanov, R. K., Asiatskaya, G. A., Suponeva, N. A., & Trofimova, A. K. (2019). Neurofeedback in the Rehabilitation of Patients with Motor Disorders after Stroke. Human Physiology, 45(4), 444–451. Scopus. https://doi.org/10.1134/S0362119719040042 Kübler, A., Holz, E. M., Riccio, A., Zickler, C., Kaufmann, T., Kleih, S. C., Staiger-Sälzer, P., Desideri, L., Hoogerwerf, E.-J., & Mattia, D. (2014). The User-Centered Design as Novel Perspective for Evaluating the Usability of BCI-Controlled Applications. PLoS Novel Perspective for Evaluating the Usability of BCI-Controlled Applications. PLoS ONE 9(12) https://doi org/10 1371/journal pone 0112392 ONE, 9(12). https://doi.org/10.1371/journal.pone.0112392 Kwakkel, G., Veerbeek, J. M., van Wegen, E. E. H., & Wolf, S. L. (2015). Constraint- Induced Movement Therapy after Stroke. The Lancet. Neurology, 14(2), 224–234. https://doi.org/10.1016/S1474-4422(14)70160-7 Lawrence, E. S., Coshall Catherine, Dundas Ruth, Stewart Judy, Rudd Anthony G., Howard Robin, & Wolfe Charles D. A. (2001). Estimates of the Prevalence of Acute Stroke Impairments and Disability in a Multiethnic Population. Stroke, 32(6), 1279–1284. https://doi.org/10.1161/01.STR.32.6.1279 iew Leamy, D. J., Collins, R., & Ward, T. E. (2011). Combining fNIRS and EEG to Improve Motor Cortex Activity Classification during an Imagined Movement-Based Task. In review Leamy, D. J., Collins, R., & Ward, T. E. (2011). Combining fNIRS and EEG to Improve Motor Cortex Activity Classification during an Imagined Movement-Based Task. In D. D. Schmorrow & C. M. Fidopiastis (Eds.), Foundations of Augmented Cognition. Directing the Future of Adaptive Systems (pp. 177–185). Springer. https://doi.org/10.1007/978-3-642-21852-1_22 In review D. D. Schmorrow & C. M. Fidopiastis (Eds.), Foundations of Augmented Cognition. In revie Directing the Future of Adaptive Systems (pp. 177–185). Springer. In rev https://doi.org/10.1007/978-3-642-21852-1_22 In r Li, M., Liu, Y., Wu, Y., Liu, S., Jia, J., & Zhang, L. (2014). https://doi.org/10.1109/EMBC.2012.6346835 Neurophysiological substrates of stroke patients with motor imagery-based brain-computer interface training. International Journal of Neuroscience, 124(6), 403–415. https://doi.org/10.3109/00207454.2013.850082 In Liu, Y., Wu, Y., Liu, S., Jia, J., & Zhang, L. (2014). Neurophysiological substrates of In Formatted: German stroke patients with motor imagery-based brain-computer interface training. International Journal of Neuroscience, 124(6), 403–415. Liang, W., Xu, Y., Schmidt, J., Zhang, L., & Ruddy, K. L. (2020). Upregulating excitability of corticospinal pathways in stroke patients using TMS neurofeedback; A pilot study. NeuroImage : Clinical, 28. https://doi.org/10.1016/j.nicl.2020.102465 López-Larraz, Eduardo, Ray, A. M., Birbaumer, N., & Ramos-Murguialday, A. (2019). Sensorimotor rhythm modulation depends on resting-state oscillations and cortex integrity in severely paralyzed stroke patients. 2019 9th International IEEE/EMBS Conference on Neural Engineering (NER), 37–40. Conference on Neural Engineering (NER), 37–40. https://doi.org/10.1109/NER.2019.8717112 López-Larraz, Eduardo, Ray, A. M., Figueiredo, T. C., Bibián, C., Birbaumer, N., & Ramos- Murguialday, A. (2017). Stroke lesion location influences the decoding of movement intention from EEG. 2017 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 3065–3068. https://doi.org/10.1109/EMBC.2017.8037504 López-Larraz, Eduardo, Sarasola-Sanz, A., Irastorza-Landa, N., Birbaumer, N., & Ramos- Murguialday, A. (2018). Brain-machine interfaces for rehabilitation in stroke: A review. NeuroRehabilitation, 43(1), 77–97. Scopus. https://doi.org/10.3233/NRE- 172394 eview Lotte, F., & Jeunet, C. (2015). Towards improved BCI based on human learning principles. The 3rd International Winter Conference on Brain-Computer Interface, 1–4. https://doi.org/10.1109/IWW-BCI.2015.7073024 In revie Lövdén, M., Bäckman, L., Lindenberger, U., Schaefer, S., & Schmiedek, F. (2010). A theoretical framework for the study of adult cognitive plasticity. Psychological Bulletin, 136(4), 659–676. https://doi.org/10.1037/a0020080 In Lundquist, C. B., Nielsen, J. F., Arguissain, F. G., & Brunner, I. C. (2021). Accuracy of the Upper Limb Prediction Algorithm PREP2 Applied 2 Weeks Poststroke: A Prospective Longitudinal Study. Neurorehabilitation and Neural Repair, 35(1), 68– 78. https://doi.org/10.1177/1545968320971763 Majid, D. S. A., Lewis, C., & Aron, A. R. (2015). Training voluntary motor suppression with real-time feedback of motor evoked potentials. Journal of Neurophysiology, 113(9), 3446–3452. https://doi.org/10.1152/jn.00992.2014 Marchesotti, S., Bassolino, M., Serino, A., Bleuler, H., & Blanke, O. (2016). Quantifying the role of motor imagery in brain-machine interfaces. Scientific Reports, 6(1), 24076. https://doi.org/10.1038/srep24076 McConnell, A. C., Moioli, R. C., Brasil, F. L., Vallejo, M., Corne, D. W., Vargas, P. A., & Stokes, A. A. (2017). Robotic devices and brain-machine interfaces for hand rehabilitation post-stroke. Journal of Rehabilitation Medicine, 49(6), 449–460. Scopus. https://doi.org/10.2340/16501977-2229 McWilliams, E. C., Barbey, F. M., Dyer, J. D., Islam, Md. N., McGuinness, B., Murphy, B., Nolan, H., Passmore, P., Rueda-Delgado, L. M., & Buick, A. R. (2021). Feasibility of repeated assessment of cognitive function in older adults using a wireless, mobile, dry-EEG headset and tablet-based games. review Nolan, H., Passmore, P., Rueda-Delgado, L. M., & Buick, A. R. (2021). Feasibility of iew Monge-Pereira, E., Ibañez-Pereda, J., Alguacil-Diego, I. M., Serrano, J. I., Spottorno-Rubio, M. P., & Molina-Rueda, F. (2017). Use of Electroencephalography Brain-Computer Interface Systems as a Rehabilitative Approach for Upper Limb Function After a Stroke: A Systematic Review. PM and R, 9(9), 918–932. Scopus. https://doi.org/10.1016/j.pmrj.2017.04.016 In revie Monge-Pereira, E., Ibañez-Pereda, J., Alguacil-Diego, I. M., Serrano, J. I., Spottorno-Rubio, In revie M. P., & Molina-Rueda, F. (2017). Formatted: German https://doi.org/10.1109/NER.2019.8717112 Use of Electroencephalography Brain-Computer In rev Interface Systems as a Rehabilitative Approach for Upper Limb Function After a In r Stroke: A Systematic Review. PM and R, 9(9), 918–932. Scopus. In https://doi.org/10.1016/j.pmrj.2017.04.016 Mrachacz-Kersting, N., Ibáñez, J., & Farina, D. (2021). Towards a mechanistic approach for the development of non-invasive brain-computer interfaces for motor rehabilitation. The Journal of Physiology, 599(9), 2361–2374. https://doi.org/10.1113/JP281314 Murphy, B., Aleni, A., Belaoucha, B., Dyer, J., & Nolan, H. (2018). Quantifying cognitive aging and performance with at-home gamified mobile EEG. 2018 International Workshop on Pattern Recognition in Neuroimaging (PRNI), 1–4. the development of non invasive brain computer interfaces for motor rehabilitation. The Journal of Physiology, 599(9), 2361–2374. https://doi.org/10.1113/JP281314 Murphy, B., Aleni, A., Belaoucha, B., Dyer, J., & Nolan, H. (2018). Quantifying cognitive aging and performance with at-home gamified mobile EEG. 2018 International Workshop on Pattern Recognition in Neuroimaging (PRNI), 1–4. https://doi.org/10.1109/PRNI.2018.8423954 Murphy, B., Barbey, F., Buick, A. R., Dyer, J., Farina, F., McGuinness, B., Passmore, A. P., & Whelan, R. (2019). F3-03-03: Replicating lab electrophysiology with older users in the home, using gamified dry eeg. Alzheimer’s & Dementia, 15(7S_Part_16), P867– P867. https://doi.org/10.1016/j.jalz.2019.06.4606 Murphy, B., Buick, A. R., Dyer, J., Nolan, H., McGuinness, B., & Passmore, A. P. (2018). Murphy, B., Buick, A. R., Dyer, J., Nolan, H., McGuinness, B., & Passmore, A. P. (2018). P4-317: Measuring cognitive decline with home-based gamified mobile eeg. Alzheimer’s & Dementia, 14(7S_Part_30), P1579–P1579. https://doi.org/10.1016/j.jalz.2018.07.140 P4-317: Measuring cognitive decline with home-based gamified mobile eeg. Alzheimer’s & Dementia, 14(7S_Part_30), P1579–P1579. https://doi.org/10.1016/j.jalz.2018.07.140 Nakayama, H., Jørgensen, H. S., Raaschou, H. O., & Olsen, T. S. (1994). Recovery of upper extremity function in stroke patients: The Copenhagen Stroke Study. Archives of Physical Medicine and Rehabilitation, 75(4), 394–398. https://doi.org/10.1016/0003- 9993(94)90161-9 view Neuper, C., Scherer, R., Reiner, M., & Pfurtscheller, G. (2005). Imagery of motor actions: Differential effects of kinesthetic and visual–motor mode of imagery in single-trial EEG. Cognitive Brain Research, 25(3), 668–677. https://doi.org/10.1016/j.cogbrainres.2005.08.014 In review Ono, T., Tomita, Y., Inose, M., Ota, T., Kimura, A., Liu, M., & Ushiba, J. (2015). Multimodal Sensory Feedback Associated with Motor Attempts Alters BOLD Responses to Paralyzed Hand Movement in Chronic Stroke Patients. Brain Topography, 28(2), 340–351. https://doi.org/10.1007/s10548-014-0382-6 In Othmer, S. (2009). Neuromodulation technologies: An attempt at classification. In T. H. Budzynski, H. K. Budzynski, J. R. Evans, & A. Abarbanel (Eds.), Introduction to quantitative EEG and neurofeedback: Advanced theory and applications (2nd ed). Academic Press/Elsevier. Pfurtscheller, G., Neuper, C., Muller, G. https://doi.org/10.1109/NER.2019.8717112 R., Obermaier, B., Krausz, G., Schlogl, A., Scherer, R., Graimann, B., Keinrath, C., Skliris, D., Wortz, M., Supp, G., & Schrank, C. Neural Systems and Rehabilitation Engineering, 11(2), 1–4. Neural Systems and Rehabilitation Engineering, 11(2), 1–4. https://doi.org/10.1109/TNSRE.2003.814454 Pichiorri, F., Morone, G., Petti, M., Toppi, J., Pisotta, I., Molinari, M., Paolucci, S., Inghilleri, M., Astolfi, L., Cincotti, F., & Mattia, D. (2015). Brain-computer interface boosts motor imagery practice during stroke recovery: BCI and Motor Imagery. Annals of Neurology, 77(5), 851–865. https://doi.org/10.1002/ana.24390 lette, L., Jeunet, C., Mansencal, B., N ’Kambou, R., N ’kaoua, B., & Lotte, F. (2017, Pillette, L., Jeunet, C., Mansencal, B., N ’Kambou, R., N ’kaoua, B., & Lotte, F. (2017, September). PEANUT: Personalised Emotional Agent for Neurotechnology User- Training. 7th International BCI Conference. https://hal.archives-ouvertes.fr/hal- 01519480 view Prasad, G., Herman, P., Coyle, D., McDonough, S., & Crosbie, J. (2010). Applying a brain- computer interface to support motor imagery practice in people with stroke for upper limb recovery: A feasibility study. Journal of Neuroengineering and Rehabilitation, 7, 60. https://doi.org/10.1186/1743-0003-7-60 In review Raffin, E., & Hummel, F. C. (2018). Restoring Motor Functions After Stroke: Multiple In Raffin, E., & Hummel, F. C. (2018). Restoring Motor Functions After Stroke: Multiple Approaches and Opportunities. The Neuroscientist : A Review Journal Bringing Neurobiology, Neurology and Psychiatry, 24(4), 400–416. https://doi.org/10.1177/1073858417737486 In Approaches and Opportunities. The Neuroscientist : A Review Journal Bringing https://doi.org/10.1177/1073858417737486 Ramos-Murguialday, A., Broetz, D., Rea, M., Läer, L., Yilmaz, Ö., Brasil, F. L., Liberati, G., Curado, M. R., Garcia-Cossio, E., Vyziotis, A., Cho, W., Agostini, M., Soares, E., Soekadar, S., Caria, A., Cohen, L. G., & Birbaumer, N. (2013). Brain-machine interface in chronic stroke rehabilitation: A controlled study: BMI in Chronic Stroke. Annals of Neurology, 74(1), 100–108. https://doi.org/10.1002/ana.23879 Ramos-Murguialday, A., Broetz, D., Rea, M., Läer, L., Yilmaz, Ö., Brasil, F. L., Liberati, G., Curado, M. R., Garcia-Cossio, E., Vyziotis, A., Cho, W., Agostini, M., Soares, E., Soekadar, S., Caria, A., Cohen, L. G., & Birbaumer, N. (2013). Brain-machine interface in chronic stroke rehabilitation: A controlled study: BMI in Chronic Stroke. Annals of Neurology, 74(1), 100–108. https://doi.org/10.1002/ana.23879 Rathee, D., Cecotti, H., & Prasad, G. (2017). Single-trial effective brain connectivity patterns enhance discriminability of mental imagery tasks. Journal of Neural Engineering, 14(5), 056005. https://doi.org/10.1088/1741-2552/aa785c Remsik, A., Young, B., Vermilyea, R., Kiekhoefer, L., Abrams, J., Evander Elmore, S., Schultz, P., Nair, V., Edwards, D., Williams, J., & Prabhakaran, V. (2016). A review of the progression and future implications of brain-computer interface therapies for restoration of distal upper extremity motor function after stroke. Expert Review of Medical Devices, 13(5), 445–454. Scopus. https://doi.org/10.1080/17434440.2016.1174572 iew https://doi.org/10.1080/17434440.2016.1174572 Ros, T., Enriquez-Geppert, S., Zotev, V., Young, K. D., Wood, G., Whitfield-Gabrieli, S., Wan, F., Vuilleumier, P., Vialatte, F., Van De Ville, D., Todder, D., Surmeli, T., Sulzer, J. S., Strehl, U., Sterman, M. B., Steiner, N. J., Sorger, B., Soekadar, S. R., Sitaram, R., … Thibault, R. T. (2020). Consensus on the reporting and experimental design of clinical and cognitive-behavioural neurofeedback studies (CRED-nf checklist). Brain, 143(6), 1674–1685. https://doi.org/10.1093/brain/awaa009 In review Sulzer, J. S., Strehl, U., Sterman, M. B., Steiner, N. J., Sorger, B., Soekadar, S. R., In revie Sitaram, R., … Thibault, R. T. (2020). Consensus on the reporting and experimental In rev Ruddy, K., Balsters, J., Mantini, D., Liu, Q., Kassraian-Fard, P., Enz, N., Mihelj, E., Subhash Chander, B., Soekadar, S. R., & Wenderoth, N. (2018). Neural activity related to volitional regulation of cortical excitability. ELife, 7, e40843. https://doi.org/10.7554/eLife.40843 Seiler, B. D., Newman-Norlund, R. D., & Monsma, E. V. (2017). Inter-individual neural differences in movement imagery abilities. Psychology of Sport and Exercise, 30, 153–163. https://doi.org/10.1016/j.psychsport.2017.02.007 Si-Mohammed, H., Petit, J., Jeunet, C., Argelaguet, F., Spindler, F., Evain, A., Roussel, N., Casiez, G., & Lecuyer, A. (2018). https://doi.org/10.1177/1073858417737486 Towards BCI-Based Interfaces for Augmented Si-Mohammed, H., Petit, J., Jeunet, C., Argelaguet, F., Spindler, F., Evain, A., Roussel, N., Casiez, G., & Lecuyer, A. (2018). Towards BCI-Based Interfaces for Augmented Reality: Feasibility, Design and Evaluation. IEEE Transactions on Visualization and Computer Graphics, 26(3), 1608–1621. https://doi.org/10.1109/TVCG.2018.2873737 Smith, M.-C., Ackerley, S. J., Barber, P. A., Byblow, W. D., & Stinear, C. M. (2019). PREP2 Algorithm Predictions Are Correct at 2 Years Poststroke for Most Patients. Neurorehabilitation and Neural Repair, 33(8), 635–642. https://doi.org/10.1177/1545968319860481 Soekadar, S. R., Kohl, S. H., Mihara, M., & von Lühmann, A. (2021). Optical brain imaging and its application to neurofeedback. NeuroImage: Clinical, 30, 102577. https://doi.org/10.1016/j.nicl.2021.102577 Soekadar, S. R., Nann, M., Crea, S., Trigili, E., Gómez, C., Opisso, E., Cohen, L. G., Birbaumer, N., & Vitiello, N. (2019). Restoration of Finger and Arm Movements Using Hybrid Brain/Neural Assistive Technology in Everyday Life Environments. In C. Guger, N. Mrachacz-Kersting, & B. Z. Allison (Eds.), Brain-Computer Interface Research: A State-of-the-Art Summary 7 (pp. 53–61). Springer International Publishing. https://doi.org/10.1007/978-3-030-05668-1_5 Soekadar, S.R., Witkowski, M., Gómez, C., Opisso, E., Medina, J., Cortese, M., Cempini, M., Carrozza, M.C., Cohen, L.G., Birbaumer, N., Vitiello, N. (2016). Hybrid EEG/EOG-based brain/neural hand exoskeleton restores fully independent daily living activities after quadriplegia. Science Robotics, 1(1), eaag3296. https://doi: 10.1126/scirobotics.aag3296. Sollfrank, T., Hart, D., Goodsell, R., Foster, J., & Tan, T. (2015). 3D visualization of movements can amplify motor cortex activation during subsequent motor imagery. Frontiers in Human Neuroscience, 9. https://doi.org/10.3389/fnhum.2015.00463 Sollfrank, T., Ramsay, A., Perdikis, S., Williamson, J., Murray-Smith, R., Leeb, R., Millán, J. d. R., & Kübler, A. (2016). The effect of multimodal and enriched feedback on SMR- Formatted: German Formatted: Font colour: Red Formatted: Normal, Indent: Left: 0 cm, Hanging: 1.25 cm, Line spacing: Double In review Reality: Feasibility, Design and Evaluation. IEEE Transactions on Visualization and Computer Graphics, 26(3), 1608–1621. https://doi.org/10.1109/TVCG.2018.2873737 Smith, M.-C., Ackerley, S. J., Barber, P. A., Byblow, W. D., & Stinear, C. M. (2019). PREP2 Algorithm Predictions Are Correct at 2 Years Poststroke for Most Patients. Neurorehabilitation and Neural Repair, 33(8), 635–642. https://doi.org/10.1177/1545968319860481 Smith, M.-C., Ackerley, S. J., Barber, P. A., Byblow, W. D., & Stinear, C. M. (2019). PREP2 Algorithm Predictions Are Correct at 2 Years Poststroke for Most Patients Smith, M. C., Ackerley, S. J., Barber, P. A., Byblow, W. D., & Stinear, C. M. (2019). PREP2 Algorithm Predictions Are Correct at 2 Years Poststroke for Most Patients. Soekadar, S. R., Kohl, S. https://doi.org/10.1177/1073858417737486 https://doi.org/10.1161/STROKEAHA.116.016020 In review , , y , , y, , , , g , , , (2017a). Proportional Motor Recovery After Stroke. Stroke, 48(3), 795–798. In review (2017a). Proportional Motor Recovery After Stroke. Stroke, 48(3), 795–798. In revie Formatted: German https://doi.org/10.1177/1073858417737486 H., Mihara, M., & von Lühmann, A. (2021). Optical brain imaging and its application to neurofeedback. NeuroImage: Clinical, 30, 102577. https://doi.org/10.1016/j.nicl.2021.102577 iew Soekadar, S. R., Nann, M., Crea, S., Trigili, E., Gómez, C., Opisso, E., Cohen, L. G., Birbaumer, N., & Vitiello, N. (2019). Restoration of Finger and Arm Movements Using Hybrid Brain/Neural Assistive Technology in Everyday Life Environments. In C. Guger, N. Mrachacz-Kersting, & B. Z. Allison (Eds.), Brain-Computer Interface Research: A State-of-the-Art Summary 7 (pp. 53–61). Springer International Publishing. https://doi.org/10.1007/978-3-030-05668-1_5 In review Soekadar, S.R., Witkowski, M., Gómez, C., Opisso, E., Medina, J., Cortese, M., Cempini, M., Carrozza, M.C., Cohen, L.G., Birbaumer, N., Vitiello, N. (2016). Hybrid EEG/EOG-based brain/neural hand exoskeleton restores fully independent daily living activities after quadriplegia. Science Robotics, 1(1), eaag3296. https://doi: 10.1126/scirobotics.aag3296. Sollfrank, T., Hart, D., Goodsell, R., Foster, J., & Tan, T. (2015). 3D visualization of movements can amplify motor cortex activation during subsequent motor imagery. Frontiers in Human Neuroscience, 9. https://doi.org/10.3389/fnhum.2015.00463 Sollfrank, T., Ramsay, A., Perdikis, S., Williamson, J., Murray-Smith, R., Leeb, R., Millán, J. d. R., & Kübler, A. (2016). The effect of multimodal and enriched feedback on SMR- Formatted: Font colour: Red Formatted: Normal, Indent: Left: 0 cm, Hanging: 1.25 cm, Line spacing: Double BCI performance. Clinical Neurophysiology, 127(1), 490–498. https://doi.org/10.1016/j.clinph.2015.06.004 https://doi.org/10.1016/j.clinph.2015.06.004 Song, J., Nair, V. A., Young, B. M., Walton, L. M., Nigogosyan, Z., Remsik, A., Tyler, M. E., Farrar-Edwards, D., Caldera, K. E., Sattin, J. A., Williams, J. C., & Prabhakaran, V. (2015). DTI measures track and predict motor function outcomes in stroke rehabilitation utilizing BCI technology. Frontiers in Human Neuroscience, 9. https://doi.org/10.3389/fnhum.2015.00195 V. (2015). DTI measures track and predict motor function outcomes in stroke rehabilitation utilizing BCI technology. Frontiers in Human Neuroscience, 9. Stieger, J. R., Engel, S. A., & He, B. (2021). Continuous sensorimotor rhythm based brain computer interface learning in a large population. Scientific Data, 8(1), 98. i w Stieger, J. R., Engel, S. A., & He, B. (2021). Continuous sensorimotor rhythm based brain w computer interface learning in a large population. Scientific Data, 8(1), 98. ie https://doi.org/10.1038/s41597-021-00883-1 view https://doi.org/10.1038/s41597-021-00883-1 Stinear, C. M., Byblow, W. D., Ackerley, S. J., Smith, M.-C., Borges, V. M., & Barber, P. A. review Stinear, C. M., Byblow, W. D., Ackerley, S. J., Smith, M.-C., Borges, V. M., & Barber, P. A. (2017a). Proportional Motor Recovery After Stroke. Stroke, 48(3), 795–798. https://doi.org/10.1161/STROKEAHA.116.016020 In re Stinear, C. M., Byblow, W. D., Ackerley, S. J., Smith, M.-C., Borges, V. M., & Barber, P. A. (2017b). PREP2: A biomarker-based algorithm for predicting upper limb function after stroke. Annals of Clinical and Translational Neurology, 4(11), 811–820. https://doi.org/10.1002/acn3.488 In r Stinear, C. M., Byblow, W. D., Steyvers, M., Levin, O., & Swinnen, S. P. (2006). Kinesthetic, but not visual, motor imagery modulates corticomotor excitability. Experimental Brain Research, 168(1–2), 157–164. https://doi.org/10.1007/s00221- 005-0078-y Stinear, C. M., Lang, C. E., Zeiler, S., & Byblow, W. D. (2020). Advances and challenges in stroke rehabilitation. The Lancet Neurology, 19(4), 348–360. https://doi.org/10.1016/S1474-4422(19)30415-6 Formatted: German Tariq, M., Trivailo, P. M., & Simic, M. (2018). EEG-Based BCI Control Schemes for Lower- Limb Assistive-Robots. Frontiers in Human Neuroscience, 12. https://doi.org/10.3389/fnhum.2018.00312 Thibault, R. T., MacPherson, A., Lifshitz, M., Roth, R. R., & Raz, A. (2018). Neurofeedback with fMRI: A critical systematic review. NeuroImage, 172, 786–807. https://doi.org/10.1016/j.neuroimage.2017.12.071 https://doi.org/10.1016/j.neuroimage.2017.12.071 Thompson, M. C. (2019). Critiquing the Concept of BCI Illiteracy. Science and Engineering Thompson, M. C. (2019). Critiquing the Concept of BCI Illiteracy. Science and Engineering Ethics, 25(4), 1217–1233. https://doi.org/10.1007/s11948-018-0061-1 Ethics, 25(4), 1217–1233. https://doi.org/10.1007/s11948-018-0061-1 Umarova, R. M., Schumacher, L. V., Schmidt, C. S. M., Martin, M., Egger, K., Urbach, H., Hennig, J., Klöppel, S., & Kaller, C. P. (2021). Interaction between cognitive reserve and age moderates effect of lesion load on stroke outcome. Scientific Reports, 11(1), 4478. https://doi.org/10.1038/s41598-021-83927-1 In review 4478. https://doi.org/10.1038/s41598-021-83927-1 In re Varkuti, B., Guan, C., Pan, Y., Phua, K. S., Ang, K. K., Kuah, C. W. K., Chua, K., Ang, B. In rev Varkuti, B., Guan, C., Pan, Y., Phua, K. S., Ang, K. K., Kuah, C. W. K., Chua, K., Ang, B. T., Birbaumer, N., & Sitaram, R. (2013). Resting State Changes in Functional Connectivity Correlate With Movement Recovery for BCI and Robot-Assisted Upper- Extremity Training After Stroke. Neurorehabilitation and Neural Repair, 27(1), 53– 62. https://doi.org/10.1177/1545968312445910 In rev T., Birbaumer, N., & Sitaram, R. (2013). Resting State Changes in Functional In r T., Birbaumer, N., & Sitaram, R. (2013). Resting State Changes in Functional Connectivity Correlate With Movement Recovery for BCI and Robot-Assisted Upper- Extremity Training After Stroke. Neurorehabilitation and Neural Repair, 27(1), 53– 62 htt //d i /10 1177/1545968312445910 In Connectivity Correlate With Movement Recovery for BCI and Robot-Assisted Upper- In Extremity Training After Stroke. Neurorehabilitation and Neural Repair, 27(1), 53– 62. https://doi.org/10.1177/1545968312445910 Formatted: German 62. https://doi.org/10.1177/1545968312445910 Veerbeek, J. M., Wegen, E. van, Peppen, R. https://doi.org/10.1161/STROKEAHA.116.016020 In re van, Wees, P. J. van der, Hendriks, E., Rietberg, M., & Kwakkel, G. (2014). What Is the Evidence for Physical Therapy Poststroke? A Systematic Review and Meta-Analysis. PLOS ONE, 9(2), e87987. https://doi.org/10.1371/journal.pone.0087987 Vidaurre, C., & Blankertz, B. (2010). Towards a Cure for BCI Illiteracy. Brain Topography, 23(2), 194–198. https://doi.org/10.1007/s10548-009-0121-6 Vuckovic, A., & Osuagwu, B. A. (2013). Using a motor imagery questionnaire to estimate the performance of a Brain–Computer Interface based on object oriented motor imagery. Clinical Neurophysiology, 124(8), 1586–1595. https://doi.org/10.1016/j.clinph.2013.02.016 https://doi.org/10.1016/j.clinph.2013.02.016 Wierzgała, P., Zapała, D., Wojcik, G. M., & Masiak, J. (2018). Most Popular Signal Processing Methods in Motor-Imagery BCI: A Review and Meta-Analysis. Frontiers in Neuroinformatics, 12. https://doi.org/10.3389/fninf.2018.00078 Wu, Q., Yue, Z., Ge, Y., Ma, D., Yin, H., Zhao, H., Liu, G., Wang, J., Dou, W., & Pan, Y. (2020). Brain Functional Networks Study of Subacute Stroke Patients With Upper Limb Dysfunction After Comprehensive Rehabilitation Including BCI Training. Frontiers in Neurology, 10. https://doi.org/10.3389/fneur.2019.01419 iew Xing, Y., & Bai, Y. (2020). A Review of Exercise-Induced Neuroplasticity in Ischemic Stroke: Pathology and Mechanisms. Molecular Neurobiology, 57(10), 4218–4231. https://doi.org/10.1007/s12035-020-02021-1 In review
https://openalex.org/W4245441755	https://www.researchsquare.com/article/rs-65315/latest.pdf	English	null	A Method for Experimental Warming of Developing Tree Seeds With A Common Garden Demonstration of Seedling Responses	Research Square (Research Square)	2,020	cc-by	9,128	Conclusions The method was shown to reliably elevate the temperature surrounding seed cones and thereby facilitate experimental in-situ climate warming research on forest trees. The method was furthermore shown to influence plant traits that may affect seedling performance under climate warming. Research Page 1/23 Version of Record: A version of this preprint was published on January 6th, 2021. See the published version at https://doi.org/10.1186/s13007-020-00700-7. Page 2/23 Results We developed and validated a simple method for passively warming seeds as they develop in tree canopies to enable controlled study of climate warming on seedling performance. We quantified thermal effects of the cone-warming method across individual pine trees and stands by measuring the air temperature surrounding seed cones using thermal loggers and the temperature of seed cone tissue using thermocouples. We then investigated seedling phenotypes in relation to the warming method through a common garden study. We assessed seedling morphology, physiology, and mycorrhizal nodulation in response to experimental cone-warming in 20 seed-source-tree canopies on the San Francisco Peaks in northern Arizona, USA. The warming method increased air temperature surrounding developing seed cones by 2.1◦C, a plausible increase in mean air temperature by 2050 under current climate projections. Notable effect sizes of cone-warming were detected for seedling root length, shoot length, and diameter at root collar using Cohen’s Local f 2. Root length was most affected by cone- warming, however, effect sizes of cone-warming on root length and diameter at root collar became negligible after the first year of growth. Cone-warming had small but significant effects on mycorrhizal fungal richness and seedling multispectral near-infrared indices indicative of plant health. Background Forest dieback driven by rapid climate warming threatens ecosystems worldwide. The health of forested ecosystems depends on how tree species respond to warming during all life history stages. While it is known that seed development is temperature-sensitive, little is known about possible effects of climate warming on seed development and subsequent seedling performance. Exposure of seeds to high air temperatures may influence subsequent seedling performance negatively, though conversely, warming during seed development may aid acclimation of seedlings to subsequent thermal stress. Technical challenges associated with in-situ warming of developing tree seeds limit understanding of how tree species may respond to seed development in a warmer climate. Introduction Forest tree mortality related to global climate warming is occurring worldwide1. Reproductive processes involved in seed production in trees are affected both indirectly and directly by environmental perturbations including changes in temperature [2,3,4]. A lack of practical methods for warming the Page 3/23 Page 3/23 environment in which seeds develop, particularly for tree species, limits our understanding of the sensitivity of seed production to higher temperatures related to climate change. environment in which seeds develop, particularly for tree species, limits our understanding of the sensitivity of seed production to higher temperatures related to climate change. In situ experimental warming treatments employ either active or passive warming systems [5]. Passive warming systems do not require supplemental energy, and instead reduce the loss of emitted longwave radiation by sheltering surfaces from boundary layer turbulence [6]. The thermal effect of warming treatments must be quantified carefully to account for differences among experimentally warmed microsites [7], and care must be taken to shield thermal loggers from direct shortwave radiation in order to accurately estimate warming effects [8]. Past heat exposure may predispose plants to adaptive responses to future episodes of heat exposure (i.e. conditioning [9]). Conditioned responses may be attributable to altered hormones, nutrients, antibodies, small RNAs, and epigenomic changes to gene expression that may persist in a lineage across generations [10]. In some cases, conditioning affords organisms a more rapid adjustment to prevailing environmental conditions [11], which may be crucial during vulnerable early life stages of plants [12]. Tree life history stages from reproduction through seedling establishment are vulnerable to abiotic stress related to climate warming [13]. High temperatures and drought can limit seed production [2,14] and seedling establishment [15]. Temperature affects the production and quality of tree seeds in forests ranging from dry temperate [14] to subarctic regions [16]. Longer periods of seed development (e.g. > 2 years for many pine species) present more opportunities for suboptimal temperatures to reduce seed quality [4]. Heat experienced by parental plants and directly by seeds can reduce seed viability and seedling vigor [3,9], and maladaptively affect progeny bud burst phenology and cold hardiness [17]. While effects of elevated temperature during tree seed development have been studied with clones in temperature-controlled greenhouses [18] and by inferring temperature differences during seed development based on provenance climates [19], there is a dearth of knowledge of the consequences of warming during seed development in ecologically-realistic settings. Introduction For instance, a study by Carneros et al. (2017, [17]), which found differences in bud burst and cold hardiness of Norway spruce (Picea abies) grown at different temperatures, was conducted by producing genetic replicates by somatic embryogenesis under cold (18◦C) and warm (28◦C) greenhouse conditions. Such controlled studies stand to provide insights into the mechanisms by which seedlings may be affected by warming during seed development, but do not readily improve understanding of phenomena at the landscape level. A lack of seed warming studies conducted in ecological settings has hindered our ability to predict possible large- scale consequences of seed warming for forest function and species diversity. Phenotypic trait responses to environmental conditions vary both across species and intraspecific ecotypes, and are constrained by covariance among traits [20,21]. Accordingly, although less common, assessment of a broad range of plant traits can deepen insights into possible trait limitations and tradeoffs associated with plant responses to warming [21,9]. For instance, a common garden study of Douglas-fir (Pseudotsuga menziesii) found that combined measures of drought and cold stress tolerance revealed trait covariance in relation to coupled abiotic stressors, suggesting tradeoffs in stress tolerance Page 4/23 Page 4/23 mechanisms [22]. In response to heat exposure, plants have been shown to alter tissue allocation (e.g. the proportion of resources invested in root versus shoot growth [23]), alter microbial community assemblages and function [24], and reflect modified profiles of near-infrared electromagnetic radiation [25]. mechanisms [22]. In response to heat exposure, plants have been shown to alter tissue allocation (e.g. the proportion of resources invested in root versus shoot growth [23]), alter microbial community assemblages and function [24], and reflect modified profiles of near-infrared electromagnetic radiation [25]. We present a simple and effective method for in situ warming of seed cones during seed development using southwestern white pine (SWWP; Pinus strobiformis); a long-lived conifer found in a wide range of climatic conditions across the southwestern USA and western Mexico. The species is threatened by an exotic fungal pathogen [26], exhibits greater drought sensitivity than co-occurring ponderosa pine (P. ponderosae [27]), is sensitive to interspecific competition [28], and is expected to undergo extensive constriction and fragmentation of the species’ historical range in response to climate change [29]. Introduction This study addressed two objectives, including: (1) introduce and evaluate a method for warming seed cones during development, and (2) demonstrate the effect of the cone-warming method on SWWP seedlings grown in a common garden and assessed for changes in above- and below-ground traits (morphological, foliar spectra, and mycorrhizal fungal communities). To address objective (1), we developed a method for warming seed cones as they develop in tree canopies and evaluated the effect of the method by comparing temperatures achieved by the cone-warming treatment and control. We also assessed how well temperature data from ground-based weather stations and HOBO loggers in canopies estimated the temperature of seed cones during development. To address objective (2), we demonstrated the effect of our warming method by quantifying effect sizes of controlled cone-warming on above- and below-ground traits of P. strobiformis seedlings grown for four years in a common garden. We focused our common garden measures on three aspects of plant traits expected to influence plant performance as the climate warms: (1) plant morphology, (2) foliar spectra, and (3) mycorrhizal fungal colonization. We anticipate that our method for studying plant trait responses to cone-warming will help expedite discovery of heat-adapted seed sources. CONE WARMING TREATMENT AND TEMPERATURE MEASURES Over the course of three sequential growing seasons (2014, 2015, and 2016), cone-warming treatments were deployed in tree canopies to develop, evaluate, and refine the cone-warming method presented here. In each deployment year, SWWP seed cones were passively warmed during the periods of fertilization and seed maturation, i.e. the full final growing season in the 27-month seed cone production cycle [30]. During the 2014 deployment, bagging materials were compared for their ability to warm seed cones in canopies, methods were developed for quantifying the warming effect, and warmed and unwarmed (control) seeds were collected for use in a common garden demonstration of seedling responses. During the 2015 deployment, the best bagging material based on performance during the 2014 season was evaluated for the temperature effect achieved by the cone-warming treatment and control groups at five new stands. The 2016 deployment was conducted to compare the effect of the cone-warming treatment on the Page 5/23 Page 5/23 temperature of seed cone tissues and the air surrounding seed cones, and to determine whether seed cone temperatures could be reliably deduced from measures of air temperature in canopies and at the ground level. During the 2014 deployment (n = 20 trees in three stands throughout the San Francisco Peaks in northern Arizona), three to five controls and three to five cone-warming treatments were deployed in tree canopies. Each control and cone-warming treatment contained at least two seed cones. The cone-warming treatment in 2014 compared the efficacy of two materials for warming seeds: (1) a non-porous, insulative bag composed of translucent plastic bubble-wrap packaging material (Fig. 2A) inside of a low-airflow fine porous polyester pollination bag (Fig. 2B), and (2) a glassine bag. Bags were affixed to branches with Velcro tape. The warming effects of the two materials were not statistically different, and the bubble-wrap bagging material was preferred due to its greater durability. No bagging material was placed over control- group seed cones in 2014. Air temperature was measured inside and outside cone-warming treatments using HOBO loggers (ONSET© HOBO V2 TidbiT Temperature Logger, Part # UTBI-001), suspended from the middle of a 10 cm long segment of white 2.54 cm diameter PVC tubing to shade loggers from direct insolation (Fig. 3), and hung from a branch with PVC tubes positioned laterally. CONE WARMING TREATMENT AND TEMPERATURE MEASURES In one of the three stands studied in 2014, two trees were affixed with one HOBO inside treatment bags (n = 2) and one HOBO outside treatment bags to record ambient air temperature (n = 2). In 2015, cone-warming treatments and controls were deployed with HOBO loggers to quantify the effect of the treatment on air temperature at five additional stands (n = 1 treatment and n = 1 control per stand). The bubble-wrap material, which was found to be the most durable cone-warming bag type in the 2014 deployment, was the sole type of warming bag used in the 2015 deployment. Loss of treatment bags from branches during the 2014 deployment prompted us to use cable ties in 2015. Control and cone- warming treatment bags were loosely fitted around the cones, and bags were affixed to tree branches proximal to the cones by plastic cable ties placed over a ~5 cm segment of polyethylene foam pipe insulation used to increase the tree branch surface area affected by the cable tie (Fig. 1). Small branches and needles that spanned the pipe insulation barrier ensured channels for gas exchange. Whereas the 2014 deployment did not include a bag for the control, we included a control treatment bag from 2015 onward due to changing to the use of cable ties in order to ensure that the pressure that was exerted on branches was similar across cone-warming treatments and controls. The control treatment consisted of a high-airflow porous mesh nylon bag (Fig. 2C), while the cone-warming treatment consisted of the combined non-porous, insulative bubble-wrap packaging material (Fig. 2A) inside of a polyester pollination bag (Fig. 2B), as described above. Paired logged data (i.e. data from one cone-warming treatment and one control in a single tree) were retrieved from three of the five stands, whereas data from the fourth stand could only be retrieved from the control group and data from the fifth stand could only be retrieved from the cone-warming treatment due to loss of loggers during the course of the experiment (n treatment = 4, n control = 4). We conducted a final experiment during the 2016 growth season to assess whether increased air temperatures inside cone-warming bags also increased the temperature of cone tissues. CONE WARMING TREATMENT AND TEMPERATURE MEASURES In contrast, only Page 6/23 Page 6/23 the temperature of air surrounding seed cones was measured during the 2014 and 2015 deployments, and not the temperature of seed cones themselves. In late May of 2016, cone-warming treatments and controls and two types of sensors were deployed in three P. strobiformis tree canopies 110 m from a weather station at Hart Prairie Preserve near Flagstaff, Arizona (35°21'06.0"N, 111°44'05.0"W). This experiment enabled evaluation of the effect of the cone-warming treatment on the temperature of seed cones using thermocouples, and to determine whether canopy air temperatures (measured with HOBO loggers) or air temperatures near the ground (measured with a thermistor 1.5 m aboveground) could be used to reliably estimate seed cone temperatures. In the canopies of three pines, three cone-warming treatment replicates and three control replicates were deployed. Each replicate contained at least two seed cones. A thermocouple was inserted into one cone within each control and cone-warming treatment bag to evaluate the effect of the cone-warming treatment on seed cone tissue (n treatment = 3, n control = 3). Thermocouple wires were inserted approximately 2 cm deep into seed cones. Each treatment and control bag in each tree contained one HOBO to evaluate the effect of the cone-warming treatment on air temperature within the bag, except in one of the three trees which received one HOBO in a cone-warming treatment. We obtained n treatment = 6 and n control = 4 HOBO data streams. Thermocouples logged temperature at five-minute intervals, and HOBOs logged temperature at hourly intervals. Temperature data were recorded from July – September. CONE WARMING TREATMENT TEMPERATURE ANALYSES The effect of the cone-warming treatment on cone tissue temperature was determined by fitting a linear model with the warmed cone temperature as the response variable and control cone temperature as the independent variable. We also compared the influence of the cone-warming treatment on the air temperature inside bags by fitting a linear model to HOBO logger data from inside the warming bag as the response variable and data from control bags as the independent variable. For the analysis of thermocouple data, measurements from the three cone-warming treatments and three control cones were averaged at each time point, then aggregated to daytime (7am – 7pm) and nighttime (7pm – 7am) average values. For HOBO logger measurements, replicates were averaged for each tree (n = 2 for the control, n = 3 for the cone-warming treatment), then an average value determined for all trees (n = 3). Values from HOBO loggers were also aggregated to daytime and nighttime averages. To compare measurements from thermocouples and HOBO loggers, we fit a linear model with warmed cone tissue temperature as the response variable and inside-bag air temperature as the independent variable. The passive warming treatment is most effective when incoming shortwave radiation inputs are greatest, hence models were fit separately for temperature values logged during day and night to more accurately quantify the daytime warming effect. We also calculated standard differences between average maximum monthly temperatures recorded in cone-warming treatment and control groups across all deployment years. To calculate standard treatment differences, we first estimated average maximum daily temperature per measurement and treatment type (e.g. thermocouple measurement in cone- warming treatment versus control) across all replicate measurements per year, calculated an average monthly maximum temperature from daily average maximum temperatures, and then calculated differences between the control group and warming group values. Page 7/23 COMMON GARDEN EXPERIMENT Seeds collected at the end of the 2014 cone-warming deployment were used in the common garden experiment. Following cone collection, cones were bench-dried in a greenhouse and extracted seeds were weighed in five replicated sets of ten seeds to estimate an average seed mass. Seeds were sown in the greenhouse in early October 2014 with subsequent greenhouse transplanting on November 18, Seeds collected at the end of the 2014 cone-warming deployment were used in the common garden experiment. CONE WARMING TREATMENT TEMPERATURE ANALYSES Following cone collection, cones were bench-dried in a greenhouse and extracted seeds were weighed in five replicated sets of ten seeds to estimate an average seed mass. Seeds were sown in the greenhouse in early October 2014 with subsequent greenhouse transplanting on November 18, 2014. Seeds were sown into labeled SC10 container growth tubes (Stuewe & Sons, Inc.; 3.8 cm diameter × 21 cm deep, 164 mL volume) in a completely randomized design across populations, genetic families, and cone-warming treatments. Seedling emergence occurred between 1 - 6 weeks following sowing. Seedlings were grown in the greenhouse for five months under ambient daylight conditions plus high pressure sodium lights to achieve a consistent 15 hr. day : 9 hr. night photoperiod. Seedlings were watered every other day and fertilized twice a week with 20-20-20 NPK fertilizer. Irrigation and fertilizer solutions were brought within a pH range of 5.5 to 6.2 using food grade phosphoric acid. Seedlings were placed outside of the greenhouse, and fertilization was ceased one month before outplanting to prepare seedlings for field conditions. Seedlings were then watered to keep the soil medium consistently moist. Replicates of each seedling experimental group (population, family, and cone-warming treatment) were planted into 1.2m x 1.2m raised bed garden boxes constructed at the Arboretum at Flagstaff Southwest Experimental Garden Array site (35.1603° N, 111.7309° W). Soil medium in the boxes consisted of 50% Cornell soil mix (one-part sphagnum peat moss, one-part horticultural perlite, and one-part coarse vermiculite), and 50% volcanic cinders sourced from The Landscape Connection, Flagstaff. Just before planting, each raised garden bed was inoculated with one shovel-full of a mixture of soils gathered from all seed-source stands to include native soil microbes in the garden boxes. Eighty-one experimental seedlings were transplanted in a randomized design across both boxes in a 9 × 9 arrangement on June 6, 2015. Seedlings were spaced approximately 12 cm apart and competing vegetation was routinely removed to eliminate competition for light. Extra (i.e. non-experimental) seedlings were planted along box edges to buffer experimental seedlings from heat radiated by the sides of the raised-bed boxes during the day. Edge seedlings were clipped two years after planting to avoid unintended effects of belowground competition. An average of 8 seedlings were planted per each of the 20 seed source trees included in the common garden. CONE WARMING TREATMENT TEMPERATURE ANALYSES Between one and 18 seedlings remained per seed source tree after the first year of growth. Each garden box was hand-watered using a spray wand fitted to a hose to apply 3.79 L of water every 7-10 days between the months of April and November. Seedlings were grown for four summers until harvesting during the spring of the fifth season, on May 2, 2019. Traits measured in the common garden included (1) plant growth above-ground (measured annually) and below-ground (measured once during transplanting and once post-harvest), (2) multispectral and thermal indices via an unpiloted aircraft system (UAS) measured during the summer in 2017 and 2018, as in [31], and (3) morphotypic mycorrhizal nodulation (measured post-harvest in 2019). Plant growth traits included plant height measured as the distance from soil level to the top of the top- most bud on the central stem, diameter at root collar (DRC) measured as seedling stem diameter at soil- Page 8/23 level, full shoot length measured as the distance from root collar to the top-most bud, full root length measured during transplanting to raised-bed garden boxes before the first summer of growth, root and shoot dry-mass measured post-harvest, and dates of bud development. Calculated plant growth traits included mean annual height and DRC growth increments (mean change (∆) in measure each year for both height and DRC), root-to-shoot length and mass (root measure divided by shoot measure, completed for both length and mass-based measures), yearly slenderness (shoot length divided by DRC), and full ∆ height and ∆ DRC (final measure minus initial measure, divided by initial measure). Multispectral and thermal infrared sensors carried by UAS recorded spectra at one timepoint at midday on May 18, 2017 and again at midday on June 2, 2018. Near infra-red spectra were used to estimate seedling crown temperatures, corresponding leaf-to-air temperature differences, and spectral indices indicative of plant health including the normalized difference vegetation index (NDVI), green NDVI (GNDVI), normalized difference red edge index (NDRE), triangular greenness index (TGI), and green-red vegetation index (GRVI). Post-harvest and before roots were dried, mycorrhizal fungi on seedling roots were assessed to the level of morphotype to determine whether cone-warming affected mycorrhizal assemblages. Mycorrhizal assemblages can affect plant performance [32], and mycorrhizal fungal species richness can be estimated by assessing mycorrhizal morphotypes [33]. CONE WARMING TREATMENT TEMPERATURE ANALYSES Percent ectomycorrhizal fungal (EMF) colonization and EMF diversity were estimated on up to 100 root tips per seedling, noting (1) dead root tips, (2) live root tips, (3) dead EMF tips, and (4) live EMF tips. Each living EMF tip was assigned a morphotype designation based on color, texture, shape, and external hyphal characteristics following [33]. COMMON GARDEN STATISTICAL ANALYSES Multivariate and univariate models were used to investigate statistical relationships between the cone-warming treatment and response variables. Effect sizes of cone-warming on responses were then estimated as described below. All analyses were conducted in R (version 3.5.1, R Core Team 2018). Seedlings in the common garden demonstration from each type of cone-warming bag (glassine versus plastic bubble-wrap packaging material, both inside of a polyester pollination bag) were treated the same because there was no statistically significant difference between the effect of the two types of cone- warming bags on seedling traits. Multivariate models were built using both principal component analysis (PCA; via the function prcomp) and permutational multivariate analysis of variance (PERMANOVA; via the function adonis) separately for the following three categories of response variables: (1) plant growth traits including bud phenology, (2) foliar spectra, and (3) mycorrhizal assemblages. The first principal component of the PCA generated from variables, belonging to one response category at a time, was used as the response in linear mixed effect models. Next, PERMANOVAs were executed using Euclidean distance matrices composed of aggregated response variables for each of the three response categories (plant growth, spectra, and mycorrhizae), separately, specifying seed source tree as a random effect. Univariate linear mixed effect models were also fitted for all response variables, specifying seed source tree nested within stand, and raised-bed box (where models would allow), as random effects using Multivariate and univariate models were used to investigate statistical relationships between the cone-warming treatment and response variables. Effect sizes of cone-warming on responses were then estimated as described below. All analyses were conducted in R (version 3.5.1, R Core Team 2018). Page 9/23 Page 9/23 functions from the R package lme4. In both multivariate and univariate models, seed mass was tested for inclusion as a covariate via AIC comparisons. Models with the smallest AIC were favored, and when models competed with AIC values within 2 AIC units of the smallest AIC, the simplest model structure with the least predictors was selected for subsequent ANOVAs. Satterthwaite approximation of denominator degrees of freedom was specified for all omnibus F-tests of fixed effects as well as type III sum of square ANOVAs for models that included interactions between seed mass and warming treatment. Type II sum of square ANOVAs were specified for models that included seed mass as an added covariate. COMMON GARDEN STATISTICAL ANALYSES The magnitude of the variance explained by the cone-warming fixed effect was estimated by Cohen’s Local f 2, which is suitable for use with mixed models for which denominator degrees of freedom must be approximated, and is suitable for use with unbalanced experimental designs [34,35]. Input for the calculation of Cohen’s Local f 2 includes marginal R2 goodness-of-fit values both from models with and without the factor of interest, as follows: R2 with refers to the marginal coefficient of determination from a model containing a fixed factor of interest, and R2 without refers to the marginal coefficient of determination from the same model with the fixed factor of interest removed. For instance, in this study the warming treatment was present in the R2 with model and omitted from the R2 without model. Cohen’s Local f 2 effect sizes ≥ 0.02, ≥ 0.15, and ≥ 0.35 are respectively considered small, medium, and large [36,35]. Code and data related to this work are accessible through the Knowledge Network for Biocomplexity. R2 with refers to the marginal coefficient of determination from a model containing a fixed factor of interest, and R2 without refers to the marginal coefficient of determination from the same model with the fixed factor of interest removed. For instance, in this study the warming treatment was present in the R2 with model and omitted from the R2 without model. Cohen’s Local f 2 effect sizes ≥ 0.02, ≥ 0.15, and ≥ 0.35 are respectively considered small, medium, and large [36,35]. Code and data related to this work are accessible through the Knowledge Network for Biocomplexity. VERIFICATION OF CONE-WARMING METHOD Across all deployment years, temperature differences between the cone-warming treatment and control varied temporally, with the greatest increase in temperature due to the warming treatment recorded during the early summer (Table 1). During the 2016 growing season at Hart Prairie, thermocouples inside cones measured a statistically significant increase in daytime temperature of 0.9◦C in cone-warming treatments compared to controls (t1,64 = 45.4, p < 0.001; Fig. 4a). In the same experiment, HOBO loggers inside cone- warming bags measured a statistically significant increase in daytime mean air temperature of 2.1◦C in cone-warming treatments compared to controls (t1,64 = 37.3, p < 0.001; Fig. 4b). No significant cone- warming treatment effect was observed in data recorded at night (Figs. 4c and 4d). A linear model of within-cone temperature as a function of air temperature, with sensor type as an additive covariate, showed that daytime mean thermocouple measurements were 2.92◦C warmer than daytime HOBO Page 10/23 Page 10/23 measurements at the same air temperature. This standard offset value, which reflects temperature differences recorded by HOBO loggers and thermocouples inside cones, both inside cone-warming bags, allows us to predict cone interior temperature (as measured with thermocouples) based on air temperatures surrounding cones recorded by HOBO loggers inside cone-warming treatment bags. However, differences between temperatures measured by thermocouples inside cones and air temperatures measured by HOBO data loggers varied over the diurnal cycle and across months (Fig. 5). Temporal variation in temperature stability was also observed. For instance, during the North American monsoon, which affects the study region primarily during August and September, thermal variation increases coincident with increased daytime cloud cover, precipitation, and humidity (Fig. 5). Inconsistent differences between temperatures measured at the weather station and inside cone-warming treatments precluded calculation of a standard offset between those measurement locations. However, average daily temperatures recorded from a thermistor at 1.5 meters above the ground surface at the weather station were always lower than those measured in canopies (Fig. 5), likely due to the combined effect of decreased albedo and increased boundary layer imposed by the canopy. During the period of June 28 – September 9th, 2016, mean daily air temperatures measured at the weather station were on average 2.6◦C cooler than air temperatures recorded in control-treatments deployed in nearby canopies. Average daily maximum values for 2016 HOBO logger data at Hart Prairie in July, August, and September were 40.5◦C, 32.4◦C, and 30.4◦C, respectively. VERIFICATION OF CONE-WARMING METHOD Average daily maximum values for 2016 thermocouple data at Hart Prairie in July, August, and September were 36.0◦C, 34.7◦C, and 34.4◦C, respectively. Table 1. Mean monthly daytime temperature increases (°C) of cone-warming treatment over controls as measured by HOBO loggers. Table 1. Mean monthly daytime temperature increases (°C) of cone-warming treatment over controls as measured by HOBO loggers. May June July August September 2014 3.30 4.06 2.63 2.50 NA 2015 5.35 3.70 3.47 2.79 NA 2016 NA 1.30 3.03 1.81 1.19 COMMON GARDEN EXPERIMENT Cone-warming produced notable effect sizes (i.e. Cohen’s Local f 2 ≥ 0.02) for 1st year root length, 1st year root:shoot length, 1st year shoot length, 1st year stem length, 2nd year DRC, and total standard ∆ height (Fig. 6). Statistically significant interactions between cone-warming and seed mass showed that warming related to an increase of 0.3 cm in 1st year root length (X2 1 = 4.0, p = 0.045), an increase of mycorrhizal morphotype richness from 3 to 3.2 (X2 1 = 6.4, p = 0.01), and the PC1 composite of spectra measured in 2018 (X2 1 = 6.8, p = 0.01). PC1 of 2018 spectra explained 77.9% of variation in spectral data. TGI accounted for the largest eigenvalue of PC1, and positive values of PC1 corresponded to larger values of TGI. While individual spectral values did not vary significantly across warming treatments, NDVI values measured in 2017 (mean = 0.37, SD = 0.05) were on average greater than those measured in 2018 (mean = 0.29, SD = 0.09). First-year root length was the only statistically significant response to cone- Page 11/23 Page 11/23 warming that also showed a notable effect size (Fig. 6, Table 2). While significant interactions between seed mass and cone-warming influenced mean values of multiple response variables, cone-warming did not affect seed mass directly (mean warmed seed mass = 1.67 g, SD = 0.32; mean control seed mass = 1.63 g, SD = 0.32). Table 2. Summary statistics of response variables tested shown here with largest effect sizes at top and diminishing effect sizes toward bottom of table. (* appears beside p-values for models that also showed significant interactions between cone-warming treatment and seed mass.) Discussion It remains to be seen whether warmer future environmental conditions will influence forest regeneration through effects on the process of seed development. The method introduced here reliably increases mean air temperatures surrounding seed cones to an extent approximately matching the minimum mean air temperature rise expected in the USA by the year 2050 (assuming relative concentration pathways ≥4.5 [37]), and produces notable effects on seedling morphology. In our common garden demonstration, we found large effect sizes of cone-warming on morphological plant traits, but that those effects may be short-lived. We found a large effect size of cone-warming on root tissue allocation following eight months of growth in a greenhouse before seedlings were transplanted into common gardens. After four summers in the common gardens, however, the effect disappeared. Relatedly, the effect size of cone-warming on DRC, which is an indirect measure of belowground plant tissue allocation [38], also decreased with each year past the cone-warming treatment. Mortality of long-lived plant species is most likely to occur during the seedling stage soon after germination [39], and this is partly attributable to the small volume of seedling roots [40]. Thus, even short-lived increases in root tissue allocation in response to seed development under warmer conditions could improve seedling survival if warm conditions during seed development correlate with warm, dry conditions during early seedling growth. Multispectral indices, particularly NDVI and TGI, provide useful indicators of plant health [25], and mycorrhizal communities influence plant water relations [41], but it is unclear why either would vary based on cone-warming treatments unless warming induced alterations in developmental pathways that persist through the transition from the zygote stage to the sporophyte stage. The 2018 PC1 of multispectral indices was significantly reduced by cone-warming. TGI, which indicates leaf chlorophyll content [42], was the spectral index with the largest eigenvalue of PC1, and this indicates that cone- warming may have resulted in deleterious effects on leaf phytochemistry. However, we emphasize that the statistical effect size was negligible, and that while PCA1 of multispectral indices differed significantly across cone-warming versus control treatments, individual indices did not. While bud phenology did not respond to cone-warming in our study, Johnsen et al. (2005, [18]) found that bud phenology responded significantly to cone-warming in Norway spruce. This discrepancy highlights the need to evaluate the ecological relevance of notable trends detected in tightly controlled studies, such as that of Johnsen et al. Discussion (2005) across species and in natural settings, as the method introduced here enables. We did not compare relative humidity in warming versus control bags, which, all else being equal, would tend to decrease as temperature increases (Lawrence 2005). But we included as little mass as possible of non-cone plant tissue in warming bags to minimize accumulation of respired water vapor. We also do not expect that possible differences in light intensity between control and warming treatments would have influenced seed or seedling traits because pine seed-cones are composed of dead cells during the final summer of seed development (Dawson et al. 1997), which is the only period during which our warming treatments were deployed. Thus, it is unlikely that a biological mechanism capable of detecting and responding to altered photosynthetic photon flux density occurs in mature cones. Page 13/23 There is a dire need for a more complete understanding of how climate change will affect the reproductive biology of forest trees and subsequent performance of seedlings. The method presented here provides a means for addressing a crucial knowledge gap concerning the effect of elevated air temperature during seed development on seed and seedling traits that influence forest regeneration. There is no clear consensus regarding how seed traits may respond to warming, though it has been argued that seed trait responses to climate warming are likely to vary by species [13]. For instance, Dewan et al. (2018, [19]) reported reduced seed germination success of seeds from maternal trees grown under warmer conditions, while Alexander & Wulff (1985, [43]) reported reduced germination success and altered phenology for Plantago lanceolata seeds produced by F0 and F1 plants following warming of the F0 generation environment. In contrast to these examples of warming negatively influencing seeds or seedlings, Scotch thistle (Onopordum acanthium) was reported to show increased germination for seeds from warmer maternal environments [44]. Studies of Arabidopsis thaliana demonstrated that warming of parental plants during flowering and seed development resulted in the expression of traits in progeny that may confer greater survival and fitness under elevated temperatures, and this effect related to increased nitrogen content in seeds from warmed parent plants [45]. Discussion While there are various mechanisms by which temperature may directly affect seed development [3,4], the method presented here does not account for ways in which warming whole parental plants may affect seed and subsequent seedling performance, and this may represent a limitation of the method [46]. There is a dire need for a more complete understanding of how climate change will affect the reproductive biology of forest trees and subsequent performance of seedlings. The method presented here provides a means for addressing a crucial knowledge gap concerning the effect of elevated air temperature during seed development on seed and seedling traits that influence forest regeneration. Conclusion The accessible, inexpensive method presented here can be used to warm reproductive structures in situ for virtually any type of land-plant, including forest trees. The broad applicability and simplicity of this method should facilitate a better understanding of the consequences of seed development under a warming climate for seed and seedling performance in multiple species and climate futures (e.g. RCP4.5 - RCP8.5). We also anticipate that the method and demonstration presented here will lead to approaches for detecting genetic material with desirable climate-hardy traits. - We thank the Arboretum at Flagstaff, NAU’s Southwest Experimental Garden Array, The Nature Conservancy, Danny DePinte, and Paul Heinrich for technical assistance. - We thank the Arboretum at Flagstaff, NAU’s Southwest Experimental Garden Array, The Nature Conservancy, Danny DePinte, and Paul Heinrich for technical assistance. - We thank the Arboretum at Flagstaff, NAU’s Southwest Experimental Garden Array, The Nature - We thank the Arboretum at Flagstaff, NAUs Southwest Experimental Garden Array, The Nature Conservancy, Danny DePinte, and Paul Heinrich for technical assistance. Conservancy, Danny DePinte, and Paul Heinrich for technical assistance. Conservancy, Danny DePinte, and Paul Heinrich for technical assistance Declarations Ethics approval and consent to participate Ethics approval and consent to participate - NA Consent for publication - Dr. Amy Whipple gives consent for the use of Figure 1, in which she is featured. Availability of data and materials P 14/23 - Upon publication, datasets used in this manuscript will be made available through the online Knowledge Network for Biocomplexity. The corresponding author will also provide datasets upon request. Page 14/23 - Upon publication, datasets used in this manuscript will be made available through the online Knowledge Network for Biocomplexity. The corresponding author will also provide datasets upon request. Page 14/23 Competing interests Funding - This material is based upon work supported by the National Science Foundation under Grants No. EF- 1442597 and EF-1442456. Seed and cone bagging and collections, plus materials used in the common gardens were partially funded by the USDA Forest Service Gene Conservation Program, NAU Technology Research Initiative Fund, and the Southwest Experimental Garden Array. Competing interests - The authors declare that they have no competing interests - The authors declare that they have no competing interests Funding Authors' contributions Authors' contributions - ERVM collected, analyzed, and interpreted the data and wrote the manuscript. GP assisted with data analysis, composition of the methods section, and manuscript editing. LFR assisted with design of the study and assisted with data interpretation and manuscript composition. CGG collected NIR data and assisted with NIR data analysis and interpretation. JBH collected ectomycorrhizal data and assisted with data interpretation. HFC assisted in design of the study and data collection. All authors read, edited, and approved the final manuscript. JS and SP assisted with data collection and interpretation. KMW and AVW designed the study, grew the plants, and assisted with data collection and data interpretation. Acknowledgements Acknowledgements References 1. Allen CD, Breshears DD, McDowell NG. On underestimation of global vulnerability to tree mortality and forest die-off from hotter drought in the Anthropocene. Ecosphere. 2015;6(8):art129. 1. Allen CD, Breshears DD, McDowell NG. On underestimation of global vulnerability to tree mortality and forest die-off from hotter drought in the Anthropocene. Ecosphere. 2015;6(8):art129. 2. Redmond MD, Forcella F, Barger NN. Declines in pinyon pine cone production associated with regional warming. Ecosphere. 2012;3(12):art120. 3. Gruwez R, De Frenne P, De Schrijver A, Leroux O, Vangansbeke P, Verheyen K. Negative effects of temperature and atmospheric depositions on the seed viability of common juniper (Juniperus communis). Ann Bot. 2014 Feb 1;113(3):489–500. 4. Brown CD, Dufour‐Tremblay G, Jameson RG, Mamet SD, Trant AJ, Walker XJ, et al. Reproduction as a bottleneck to treeline advance across the circumarctic forest tundra ecotone. Ecography. 2019;42(1):137–47. 4. Brown CD, Dufour‐Tremblay G, Jameson RG, Mamet SD, Trant AJ, Walker XJ, et al. Reproduction as a bottleneck to treeline advance across the circumarctic forest tundra ecotone. Ecography. 2019;42(1):137–47. 5. Aronson EL, McNulty SG. Appropriate experimental ecosystem warming methods by ecosystem, objective, and practicality. Agricultural and Forest Meteorology. 2009 Nov 3;149(11):1791–9. 5. Aronson EL, McNulty SG. Appropriate experimental ecosystem warming methods by ecosystem, objective, and practicality. Agricultural and Forest Meteorology. 2009 Nov 3;149(11):1791–9. Page 15/23 6. Chung H, Muraoka H, Nakamura M, Han S, Muller O, Son Y. Experimental warming studies on tree species and forest ecosystems: a literature review. J Plant Res. 2013 Jul 1;126(4):447–60. 7. Germino MJ, Smith WK. Differences in microsite, plant form, and low-temperature photoinhibition in alpine plants. Arctic, Antarctic, and Alpine Research. 2000 Nov;32(4):388–96. 8. Terando AJ, Youngsteadt E, Meineke EK, Prado SG. Ad hoc instrumentation methods in ecological studies produce highly biased temperature measurements. Ecology and Evolution. 2017;7(23):9890– 904. 9. Wang X, Xin C, Cai J, Zhou Q, Dai T, Cao W, et al. Heat priming induces trans-generational tolerance to high temperature stress in wheat. Frontiers in Plant Science. 2016;7(April):1–12. 10. Kuijper B, Hoyle RB. When to rely on maternal effects and when on phenotypic plasticity? Evolution. 2015;69(4):950–68. 11. Agathokleous E, Kitao M, Harayama H, Calabrese EJ. Temperature-induced hormesis in plants. J For Res. 2019 Feb 1;30(1):13–20. 12. Kolb PF, Robberecht R. High temperature and drought stress effects on survival of Pinus ponderosa seedlings. Tree Physiology. 1996 Aug 1;16(8):665–72. 13. Cochrane A, Yates CJ, Hoyle GL, Nicotra AB. References should be considered? Agriculture. 2016 Jun 1;6(2):26. should be considered? Agriculture. 2016 Jun 1;6(2):26. 22. Bansal S, Harrington CA, Clair JBS. Tolerance to multiple climate stressors: a case study of Douglas- fir drought and cold hardiness. Ecology and Evolution. 2016;6(7):2074–83. 23. He J, Wolfe‐Bellin KS, Bazzaz FA. Leaf‐level physiology, biomass, and reproduction of Phytolacca americana under conditions of elevated CO2 and altered temperature regimes. International Journal of Plant Sciences. 2005 Jul 1;166(4):615–22. 24. Carrell AA, Kolton M, Glass JB, Pelletier DA, Warren MJ, Kostka JE, et al. Experimental warming alters the community composition, diversity, and N2 fixation activity of peat moss (Sphagnum fallax) microbiomes. Global Change Biology. 2019;25(9):2993–3004. 24. Carrell AA, Kolton M, Glass JB, Pelletier DA, Warren MJ, Kostka JE, et al. Experimental warming alters the community composition, diversity, and N2 fixation activity of peat moss (Sphagnum fallax) microbiomes. Global Change Biology. 2019;25(9):2993–3004. 25. Buters TM, Belton D, Cross AT. Multi-sensor UAV tracking of individual seedlings and seedling communities at millimetre accuracy. Drones. 2019 Dec;3(4):81. 25. Buters TM, Belton D, Cross AT. Multi-sensor UAV tracking of individual seedlings and seedling communities at millimetre accuracy. Drones. 2019 Dec;3(4):81. 26. Looney CE, Waring KM. Patterns of forest structure, competition and regeneration in southwestern white pine (Pinus strobiformis) forests. Forest Ecology and Management. 2012 Dec 15;286:159–70. 26. Looney CE, Waring KM. Patterns of forest structure, competition and regeneration in southwestern white pine (Pinus strobiformis) forests. Forest Ecology and Management. 2012 Dec 15;286:159–70. 27. Barton AM, Teeri JA. The ecology of elevational positions in plants: drought resistance in five montane pine species in southeastern Arizona. American Journal of Botany. 1993;80(1):15–25. 27. Barton AM, Teeri JA. The ecology of elevational positions in plants: drought resistance in five montane pine species in southeastern Arizona. American Journal of Botany. 1993;80(1):15–25. 28. Goodrich BA, Waring KM. Pinus strobiformis seedling growth in southwestern US mixed conifer forests in managed and non-managed stands. Forestry (Lond). 2017 May 1;90(3):393–403. 28. Goodrich BA, Waring KM. Pinus strobiformis seedling growth in southwestern US mixed conifer forests in managed and non-managed stands. Forestry (Lond). 2017 May 1;90(3):393–403. 29. Shirk AJ, Cushman SA, Waring KM, Wehenkel CA, Leal-Sáenz A, Toney C, et al. Southwestern white pine (Pinus strobiformis) species distribution models project a large range shift and contraction due to regional climatic changes. Forest Ecology and Management. 2018 Mar 1;411:176–86. 30. References Will among-population variation in seed traits improve the chance of species persistence under climate change? Global Ecology and Biogeography. 2015;24(1):12–24. 13. Cochrane A, Yates CJ, Hoyle GL, Nicotra AB. Will among-population variation in seed traits improve the chance of species persistence under climate change? Global Ecology and Biogeography. 2015;24(1):12–24. 14. Flores-Rentería L, Whipple AV, Benally GJ, Patterson A, Canyon B, Gehring CA. Higher temperature at lower elevation sites fails to promote acclimation or adaptation to heat stress during pollen germination. Front Plant Sci [Internet]. 2018 [cited 2020 Mar 28];9. Available from: https://www.frontiersin.org/articles/10.3389/fpls.2018.00536/full 15. Kozlowski TT. Physiological ecology of natural regeneration of harvested and disturbed forest stands: implications for forest management. Forest Ecology and Management. 2002 Mar 15;158(1):195–221. 16. Harju AM, Kärkkäinen K, Ruotsalainen S. Phenotypic and genetic variation in the seed maturity of Scots pine. Silvae Genetica. 1996;45(4):205-210. 16. Harju AM, Kärkkäinen K, Ruotsalainen S. Phenotypic and genetic variation in the seed maturity of Scots pine. Silvae Genetica. 1996;45(4):205-210. 17. Carneros E, Yakovlev I, Viejo M, Olsen JE, Fossdal CG. The epigenetic memory of temperature during embryogenesis modifies the expression of bud burst-related genes in Norway spruce epitypes. Planta. 2017 Sep 1;246(3):553–66. 17. Carneros E, Yakovlev I, Viejo M, Olsen JE, Fossdal CG. The epigenetic memory of temperature during embryogenesis modifies the expression of bud burst-related genes in Norway spruce epitypes. Planta. 2017 Sep 1;246(3):553–66. 18. Johnsen Ø, Dæhlen OG, Østreng G, Skrøppa T. Daylength and temperature during seed production interactively affect adaptive performance of Picea abies progenies. New Phytologist. 2005;168(3):589–96. 19. Dewan S, Vander Mijnsbrugge K, De Frenne P, Steenackers M, Michiels B, Verheyen K. Maternal temperature during seed maturation affects seed germination and timing of bud set in seedlings of European black poplar. Forest Ecology and Management. 2018 Feb 15;410:126–35. 20. Sambatti JBM, Caylor KK. When is breeding for drought tolerance optimal if drought is random? New Phytologist. 2007;175(1):70–80. 20. Sambatti JBM, Caylor KK. When is breeding for drought tolerance optimal if drought is random? New Phytologist. 2007;175(1):70–80. Page 16/23 21. Simova-Stoilova L, Vassileva V, Feller U. Selection and breeding of suitable crop genotypes for drought and heat periods in a changing climate: which morphological and physiological properties Page 16/23 21. Simova-Stoilova L, Vassileva V, Feller U. Selection and breeding of suitable crop genotypes for drought and heat periods in a changing climate: which morphological and physiological properties Page 16/23 should be considered? Agriculture. 2016 Jun 1;6(2):26. References Conifer Tree Seed in the Inland Mountain West Symposium, Intermountain Research Station (Ogden, Utah), Shearer RC. Proceedings, Conifer Tree Seed in the Inland Mountain West Symposium : Missoula, Montana, August 5-6, 1985 / [Internet]. Ogden, Utah : U.S. Dept. of Agriculture, Forest Service, Intermountain Research Station,; 1986 [cited 2020 Aug 23]. Available from: http://www.biodiversitylibrary.org/bibliography/100178 31. Garms CG, Flores-Renteria L, Waring KM, Whipple AV, Wing MG, Strimbu BM. Augmenting size models for Pinus strobiformis seedlings using dimensional estimates from unmanned aircraft systems. Can J For Res [Internet]. 2020 May 1 [cited 2020 May 29]; Available from: https://www.nrcresearchpress.com/doi/abs/10.1139/cjfr-2019-0325 32. Gehring CA, Sthultz CM, Flores-Rentería L, Whipple AV, Whitham TG. Tree genetics defines fungal partner communities that may confer drought tolerance. Proc Natl Acad Sci USA. 2017 Oct 17;114(42):11169–74. 33. Lamit LJ, Busby PE, Lau MK, Compson ZG, Wojtowicz T, Keith AR, et al. Tree genotype mediates covariance among communities from microbes to lichens and arthropods. Journal of Ecology. 2015;103(4):840–50. Page 17/23 34. Selya AS, Rose JS, Dierker LC, Hedeker D, Mermelstein RJ. A practical guide to calculating Cohen’s f2, a measure of local effect size, from PROC MIXED. Front Psychol [Internet]. 2012 [cited 2020 Apr 5];3. Available from: https://www.frontiersin.org/articles/10.3389/fpsyg.2012.00111/full 35. Denton EM, Smith BS, Hamerlynck EP, Sheley RL. Seedling defoliation and drought stress: variation in intensity and frequency affect performance and survival. Rangeland Ecology & Management. 2018 Jan;71(1):25–34. 36. Cohen J. A power primer. Psychological bulletin. 1992;112(1):155–159. 36. Cohen J. A power primer. Psychological bulletin. 1992;112(1):155–159. 37. Langner LL, Joyce LA, Wear DN, Prestemon JP, Coulson D, O’Dea CB. Future scenarios: A technical document supporting the USDA Forest Service 2020 RPA Assessment [Internet]. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station; 2020 May [cited 2020 Jun 7] p. RMRS-GTR-412. Report No.: RMRS-GTR-412. Available from: https://www.fs.usda.gov/treesearch/pubs/59976 38. Mexal JG, Landis TD. Target seedling concepts: height and diameter. In Proceedings, Western Forest Nursery Association. 1990 August; 13-17. 38. Mexal JG, Landis TD. Target seedling concepts: height and diameter. In Proceedings, Western Forest Nursery Association. 1990 August; 13-17. 39. Fenner M. Seedlings. New Phytologist. 1987;106(s1):35–47. 39. Fenner M. Seedlings. New Phytologist. 1987;106(s1):35–47. 40. Seidel H, Menzel A. Above-ground dimensions and acclimation explain variation in drought mortality of Scots pine seedlings from various provenances. Front Plant Sci [Internet]. 2016 [cited 2018 Nov 29];7. Available from: https://www.frontiersin.org/articles/10.3389/fpls.2016.01014/full 40. Seidel H, Menzel A. References Above-ground dimensions and acclimation explain variation in drought mortality of Scots pine seedlings from various provenances. Front Plant Sci [Internet]. 2016 [cited 2018 Nov 29];7. Available from: https://www.frontiersin.org/articles/10.3389/fpls.2016.01014/full 41. Hobbie EA, Colpaert JV. Nitrogen availability and mycorrhizal colonization influence water use efficiency and carbon isotope patterns in Pinus sylvestris. New Phytologist. 2004;164(3):515–25. 41. Hobbie EA, Colpaert JV. Nitrogen availability and mycorrhizal colonization influence water use efficiency and carbon isotope patterns in Pinus sylvestris. New Phytologist. 2004;164(3):515–25. 42. Hunt ER, Doraiswamy PC, McMurtrey JE, Daughtry CST, Perry EM, Akhmedov B. A visible band index for remote sensing leaf chlorophyll content at the canopy scale. International Journal of Applied Earth Observation and Geoinformation. 2013 Apr;21:103–12. 42. Hunt ER, Doraiswamy PC, McMurtrey JE, Daughtry CST, Perry EM, Akhmedov B. A visible band index for remote sensing leaf chlorophyll content at the canopy scale. International Journal of Applied Earth Observation and Geoinformation. 2013 Apr;21:103–12. 43. Alexander HM, Wulff RD. Experimental ecological genetics in plantago: x. the effects of maternal temperature on seed and seedling characters in P. lanceolata. The Journal of Ecology. 1985 Mar;73(1):271. 43. Alexander HM, Wulff RD. Experimental ecological genetics in plantago: x. the effects of maternal temperature on seed and seedling characters in P. lanceolata. The Journal of Ecology. 1985 Mar;73(1):271. 44. Qaderi MM, Cavers PB. Variation in germination response within Scotch thistle, Onopordum acanthium L., populations matured under greenhouse and field conditions. Écoscience. 2000 Jan 1;7(1):57–65. 44. Qaderi MM, Cavers PB. Variation in germination response within Scotch thistle, Onopordum acanthium L., populations matured under greenhouse and field conditions. Écoscience. 2000 Jan 1;7(1):57–65. 45. Blödner C, Goebel C, Feussner I, Gatz C, Polle A. Warm and cold parental reproductive environments affect seed properties, fitness, and cold responsiveness in Arabidopsis thaliana progenies. Plant, Cell & Environment. 2007;30(2):165–75. 45. Blödner C, Goebel C, Feussner I, Gatz C, Polle A. Warm and cold parental reproductive environments affect seed properties, fitness, and cold responsiveness in Arabidopsis thaliana progenies. Plant, Cell & Environment. 2007;30(2):165–75. 46. Vivas M, Wingfield MJ, Slippers B. Maternal effects should be considered in the establishment of forestry plantations. Forest Ecology and Management. 2020 Mar;460:117909. 46. Vivas M, Wingfield MJ, Slippers B. Maternal effects should be considered in the establishment of forestry plantations. Forest Ecology and Management. 2020 Mar;460:117909. Figures Page 18/23 Figure 1 Figure 1 Warming treatment and control bags shown deployed in a seed source P. strobiformis tree. Photo copyright: Aaron Ellison. copyright: Aaron Ellison. Figure 2 copyright: Aaron Ellison. Figure 2 Page 19/23 (Left) A. Interior insulative bubble-wrap bag deployed in seed warming treatment. (Middle) B. Exterior low- airflow pollination bag deployed in seed warming treatment. (Right) C. High-airflow mesh bag deployed in control treatment. Figure 3 HOBO loggers were shielded from direct insolation by suspending HOBOs inside PVC tube segments. Figure 3 HOBO loggers were shielded from direct insolation by suspending HOBOs inside PVC tube segments. Page 20/23 Page 20/23 Figure 4 Figure 4 Comparisons of mean daytime and nighttime control and warmed treatments of cone and bag temperatures from 2016 thermal data. (a) daytime thermocouple and (b) daytime HOBO measurements. (c) nighttime thermocoupleand (d) nighttime HOBO measurements. Filled black circles represent the mean daytime (7am -7pm) or nighttime value each day from one tree. Dashed lines represent theoretical 1:1 relationships between data from control and warmed groups. Solid lines are linear model fits to the regression of temperature data from warmed versus control treatments. Figure 4 Comparisons of mean daytime and nighttime control and warmed treatments of cone and bag temperatures from 2016 thermal data. (a) daytime thermocouple and (b) daytime HOBO measurements. (c) nighttime thermocoupleand (d) nighttime HOBO measurements. Filled black circles represent the mean daytime (7am -7pm) or nighttime value each day from one tree. Dashed lines represent theoretical 1:1 relationships between data from control and warmed groups. Solid lines are linear model fits to the regression of temperature data from warmed versus control treatments. Comparisons of mean daytime and nighttime control and warmed treatments of cone and bag temperatures from 2016 thermal data. (a) daytime thermocouple and (b) daytime HOBO measurements. (c) nighttime thermocoupleand (d) nighttime HOBO measurements. Filled black circles represent the mean daytime (7am -7pm) or nighttime value each day from one tree. Dashed lines represent theoretical 1:1 relationships between data from control and warmed groups. Solid lines are linear model fits to the regression of temperature data from warmed versus control treatments. Page 21/23 Figure 5 Hart Prairie temperatures for three-day periods in July, August and September 2016. The top three panels show data from thermocouples (measurements recorded every five minutes). The bottom three panels show data from HOBO loggers (measurements recorded once hourly). Black line shows data from the cone-warming treatment, dashed line shows data from the control, and red line is air temperature 1.5 m from the ground from a nearby weather station away from the forest overstory. Gray shading is the upper and lower bounds of the standard error range relative to the mean for each time point. Data represent averaged values across trees from which a standard error was calculated from averaged replicates within trees. Each line represents one data stream from a tree/treatment combination. Figure 5 Figure 5 Hart Prairie temperatures for three-day periods in July, August and September 2016. The top three panels show data from thermocouples (measurements recorded every five minutes). The bottom three panels show data from HOBO loggers (measurements recorded once hourly). Black line shows data from the cone-warming treatment, dashed line shows data from the control, and red line is air temperature 1.5 m from the ground from a nearby weather station away from the forest overstory. Gray shading is the upper and lower bounds of the standard error range relative to the mean for each time point. Data represent averaged values across trees from which a standard error was calculated from averaged replicates within trees. Each line represents one data stream from a tree/treatment combination. Page 22/23 Figure 6 Figure 6 Figure 6 Effect sizes of cone-warming upon 11 response variables. Bar heights below the black dotted horizontal line at y=0.02 are negligible effects. Bar heights between the black dotted and black solid horizontal lines are small but notable effects. Bar heights between the solid black horizontal line and solid red horizonal line are medium effects. Bar heights above the solid red horizontal line are large effects. 1st year of growth = 2015, 4th year of growth = 2018. 4th year growth measurements were taken during the spring prior to the beginning of the 5th growth year. Effect sizes of cone-warming upon 11 response variables. Bar heights below the black dotted horizontal line at y=0.02 are negligible effects. Bar heights between the black dotted and black solid horizontal lines are small but notable effects. Bar heights between the solid black horizontal line and solid red horizonal line are medium effects. Bar heights above the solid red horizontal line are large effects. 1st year of growth = 2015, 4th year of growth = 2018. 4th year growth measurements were taken during the spring prior to the beginning of the 5th growth year. Page 23/23 Page 23/23
https://openalex.org/W2007866389	https://www.animbiosci.org/upload/pdf/23-201.pdf	English	null	Porcine Xenotransplantation to Primates	Asian-Australasian journal of animal sciences	2,010	cc-by	6,553	INTRODUCTION reports suggest that pigs can be used as alternative to the shortage of human organs, if we produce appropriate and optimal pig strain without immune response to human (Vodicka et al, 2005; Puga Yung et al, 2009). However, many crucial answers on efficacy and safety will ultimately only be solved by well designed and controlled solid organ xenotransplantation trials on humans. Therefore, further research on the potential effects of crossing the species barrier in pig-to-primate model is essential before clinical application is acceptable (Ravelingien et al., 2004; Ekser et al., 2009). This review article provides an overview of the rejection in pig xenotransplantation to primates, and genetically modified and clonged pig in xenotransplantation. It also highlights major target organs in porcine xenotransplantation and virus infection in xenotransplantation. Transplantation is one of greatest challenging achievements and saves thousands of lives each year. It also improves the quality of patients who present each year with organ failure. However, the transplanted organs are very insufficiency owing to the overwhelming demands for donated organs. Xenotransplantation may be one of the best possible approaches to solving the severe shortage of human donors, which greatly limits progress in clinical transplantation. Therefore, xenotransplantation is currently a hot topic in bio-organ of biomedical research. Among many species, pigs are suitable animals due to easier animal husbandry, comparatively similar anatomical and physiological similarities to human organs (Hughes, 1986; Pereira-Sampaio et al, 2004; Vodicka et al, 2005). These Porcine Xenotransplantation to Primates* T. S. Min, H. J. Han1 and S. H. Park1,** National Research Foundation of Korea, Daejeon 305-350, Korea ABSTRACT : Xenotransplantation is a hot topic currently, since the demand for diverse organs is increasing in patients. Among many species, pigs are suitable animals for xenotranplantation as they share many anatomical and physiological characteristics with humans. This review article provides an overview of porcine xenotransplantation and the rejection of pig xenotransplants in primates, and use of genetically modified and cloned pigs in xenotransplantation. It also highlights major target organs in porcine xenotransplantation and virus infection in xenotransplantation. (Key Words : Xenotransplantation, Pig, Rejection, Virus Infection) 1535 1535 Asian-Aust. J. Anim. Sci. Vol. 23, No. 11 : 1535 - 1542 November 2010 Asian-Aust. J. Anim. Sci. Vol. 23, No. 11 : 1535 - 1542 November 2010 www.ajas.info * This work was supported by a grant (No. 20070401034006) from the BioGreen 21 Program of the Rural Development Administration, Korea. This work was also supported by the Grant of the Korean Ministry of Education, Science and Technology (The Regional Core Research Program/Biohousing Research Institute). This work was also supported by the Biohousing Research Center. The authors would like to acknowledge a graduate fellowship provided by the Ministry of Education and Human Resources Development through the Brain Korea 21 project in Korea. ** Corresponding Author : Soo Hyun Park. Tel: +82-62-530- 2832, Fax: +82-62-530-2809, E-mail: parksh@chonnam.ac.kr 1 Bio-therapy Human Resources Center, Department of Veterinary Physiology, College of Veterinary Medicine, Chonnam National University, Gwangju 500-757, Korea. Received August 20, 2010; Accepted September 29, 2010 REJECTION IN PORCINE XENOTRANSPLANTATION * This work was supported by a grant (No. 20070401034006) from the BioGreen 21 Program of the Rural Development Administration, Korea. This work was also supported by the Grant of the Korean Ministry of Education, Science and Technology (The Regional Core Research Program/Biohousing Research Institute). This work was also supported by the Biohousing Research Center. The authors would like to acknowledge a graduate fellowship provided by the Ministry of Education and Human Resources Development through the Brain Korea 21 project in Korea. ** Corresponding Author : Soo Hyun Park. Tel: +82-62-530- 2832, Fax: +82-62-530-2809, E-mail: parksh@chonnam.ac.kr 1 Bio-therapy Human Resources Center, Department of Veterinary Physiology, College of Veterinary Medicine, Chonnam National University, Gwangju 500-757, Korea. Received August 20, 2010; Accepted September 29, 2010 * This work was supported by a grant (No. 20070401034006) from the BioGreen 21 Program of the Rural Development Administration, Korea. This work was also supported by the Grant of the Korean Ministry of Education, Science and Technology (The Regional Core Research Program/Biohousing Research Institute). This work was also supported by the Biohousing Research Center. The authors would like to acknowledge a graduate fellowship provided by the Ministry of Education and Human Resources Development through the Brain Korea 21 project in Korea. The most profound barrier to pig-to-primate xenotransplantation is the rejection of the grafted organ by a cascade of immune mechanisms commonly referred to as hyperacute rejection (HAR), acute vascular rejection (AVR)/humoral xenograft rejection (AHXR), immune cell- mediated rejection, and chronic rejection. HAR leads to graft rejection within minutes, was considered as first barrier to pig-to-human xenotransplantation. Gala1, 3Gal (Gal), a particular carbohydrate, is majorly implicated in the development of hyperacute rejection of pig organs transplanted into humans and non-human primates (Good et Min et al. (2010) Asian-Aust. J. Anim. Sci. 23(11):1535-1542 1536 for xenotransplantation research. For germ-line gene transfer, retroviral vector (Cabot et al., 2001), lentiviral vectors (Hofmann et al., 2003; Whitelaw et al., 2004), sperm-mediated gene transfer (SMGT) (Lavitrano et al., 2006) and somatic cell nuclear transfer (SCNT) were used. Lentiviral gene transfer has been used in a variety of experiments to transduce cells with various transgenes owing to high efficiency (Sachs and Galli, 2009). SMGT has also been used with a nonviral episomal vector (Manzini et al., 2006). SCNT has become the optimal tool for generating pig animals from genetically engineered somatic cells. REJECTION IN PORCINE XENOTRANSPLANTATION Recently, the Cre site-specific DNA recombinase system, a powerful tool for manipulating DNA in vivo, has been used for porcine xenotransplantation (Prather et al., 2003; Lunney, 2007). A reporter pig strain containing the EGFP gene driven by the CMV promoter, was used, in which the EGFP gene is expressed only after Cre-mediated excision of loxP-flanked stop sequences. The EGFP will be conditionally expressed in the resulting embryos and adult pigs. The reporter pig strain is capable of generating recombinant animal tissue in which Cre- mediated excision events can be studied in vivo in a variety of experimental contexts (reviewed in Li et al. in 2009). al., 1992; Cooper et al., 1993), since αGal epitope is absent in humans, apes, and monkeys. Genetically engineered ‘‘knock-out’’ pigs that lack the a1,3-galactosyltransferase gene (GalT-KO) and thus do not express the Gal oligosaccharide have been used to improve the hyperacute rejection (Hisashi et al., 2008; Shimizu et al., 2008). Therapeutics for Gal also has several problems, since blocking of Gal attenuated hyperacute rejection but not AVR injury (Pierson, 2009). Classical complement pathways also are involved in the hyperacute rejection. Inefficient coagulation inhibitory pathways involving endothelial proteins such as thrombomodulin and tissue factor pathway inhibitor (TFPI) result in increase of thrombosis, although porcine endothelium exposed to human blood have somewhat similar mechanism compared to that of human endothelium exposed to human blood (Pierson et al., 2009). AVR/AHXR is defined as a rejection that begins within 24 h after transplantation and gradually destroys the graft (Platt, 1998). AVR is generally known to be initiated by xeno-reactive antibodies, including non-Gal antibodies and subsequent activation of the graft endothelium, the complement and the coagulation systems. It occurs within days by the humoral and cellular action of B lymphocytes and T lymphocytes. Antibodies targeting the CD 154/CD40 pathway were used to improve the rejection of T cell and B cell response to xenoantigen (Kawai et al., 2000; Kirk and Harlan, 2001). Direct xenorecognition and indirect xenorecognition are available routes for presentation of pig antigens to primate T cells (Pierson et al., 2009). CD40 and the B7 family proteins (CD80, CD86, and others), which are expressed on ‘‘antigen presenting cells’’, bind to CD154 and CD28 or CTLA-4, respectively, ligands which are expressed mainly on responding T cells. Therefore, blocking CD 28 or the F7 family can be a targeting for the rejection of porcine xenotransplantation into primates. REJECTION IN PORCINE XENOTRANSPLANTATION Engineered pigs are of great value for research and commercial applications and could serve as models for human disease (Brunetti et al., 2008). Porcine genetic engineered model is based upon the immune rejection abovementioned. Since HAR is mediated majorly by the antibody of Gala1, 3Gal (Gal), it is generally accepted that Gal KO is an important prophylactic genetic modification for xenotransplantation of vascularised organs. d'Apice and Cowan (2009) insisted that Gal KO pig is the basic “standard platform” pig on which other genetic modifications can be assembled. HAR may be overcome by using genetically modified donor pigs lacking functional GGTA1 expression. Competitive glycosylation strategies by transgenic expression of human alpha-1,2- fucosyltransferase (alpha-1,2-FT) or human beta-1,4-N- acetylglucosaminyltransferase III (GnT-III) were used to reduce alpha-1,3-Gal epitopes on porcine cells (reviewed in Klymiuk et al. in 2010). Transgenic pigs for human complement-regulatory proteins (CRPs) such as human decay-accelerating factor (hDAF), CD46 and CD59 have been characterized (Byrne et al., 1997; Cozzi et al., 1997; Diamond et al., 2001; Lee et al., 2010). Transgenic pigs for natural killer cell-mediated rejection were also established (Weiss et al., 2009). Recently triple transgenic (CD59/DAF/hTM) pigs are used to in pig xenotransplantation (Petersen et al., 2009). Therefore, the use of transgene or knock-out pig organs in pig-to-non- human primate xenotransplantation can prevent HAR, AVR, and immune-mediated rejection, eventually extending Several lines of evidence indicates that NK cells play an important role in rejection of porcine xenografts into human (Inverardi et al., 1994; Khalfoun et al., 2000). The initiating cause of failure of pig xenografts may be antibody-mediated injury to the endothelium, leading to the development of microvascular thrombosis. Therefore, potential contributing factors to the development of microvascular thrombosis are the presence of preformed anti-non-Gal antibodies, the development of very low levels of elicited antibodies to non-Gal antigens, natural killer cell or macrophage activity, and inherent coagulation dysregulation between pigs and primates (Ekser et al., 2009). insufficiency (Gewartowska and Olszewski, 2007). insufficiency (Gewartowska and Olszewski, 2007). xenograft survival. Porcine Liver transplantation The liver is the primary site of xenotransplantation. Liver transplantation offers several benefits for the treatment of patients with liver failure or acute liver failure and end-stage chronic liver disease. According to the review of Esker et al. (2009), during the past 13 years, 30,000 patients have died waiting for a liver transplantation. Ramirez et al. (2000) transplanted livers from wild-type pigs or pigs transgenic for the human complement- regulatory protein, human decay-accelerating factor (hDAF), into immunosuppressed baboons. Makowka et al. (1995) reported that there was an attempt to porcine liver xenotransplanation to human, eventually resulting in the death without success. Recently, Ekser et al. (2010) reported that after the transplantation of genetically engineered pig livers into baboons i) many parameters of hepatic function, including coagulation, were normal or near normal; ii) there was evidence for production of pig proteins, including coagulation factors; and iii) these appeared to function adequately in baboons although interspecies compatibility of such proteins remains to be confirmed. This report gives us good possibility that the liver xenotransplantation may be applicable to clinical usage in human in future. Some reports suggest that pig livers may be rejected less vigorously than other pig organs (Tuso et al., 1993; Tector et al., 2001). They observed that xenoperfused pig livers may function for up to 5 h, in contrast to porcine kidneys or hearts. According to the reports of Ramirez et al. (2000; 2001), 3 baboons transplanted with livers from unmodified pigs survived for <12 h with the response of HAR but two baboons transplanted with livers from pigs transgenic for hDAF survived for 4 and 8 days. These reports suggest that if HAR is abrogated (by the presence of hDAF), the porcine liver can maintain reasonable levels of coagulation factors and protein in the baboon for up to 8 days (Hara et al., 2008). Porcine hepatocyte transplantation also could be proposed as a method to support patients with liver Porcine cardiac xenotransplation Use of a pig heart as a bridge to allotransplantation could be a solution due to heart deficiency organ. Orthotopic pig-to-baboon heart transplantation is the accepted preclinical model for cardiac xenotransplantation in humans, although heterotopic thoracic pig-to-baboon heart transplantation was reported (Bauer et al., 2010). When the hearts from hdAF transgenic pigs were perfused with human blood, HAR was avoided and the hearts were relatively metablically and functionally stable (Smolenski et al., 2007). Hisashi et al. (2008) observed that no hyperacute rejection developed and one graft survived up to 6 months after transplantation. However, they also indicated that all GalT-KO heart grafts underwent graft failure with AHXR and/or chronic rejection. In recent report, the role of alpha- 1,3-galactosyltransferase (alpha-Gal) antigen in valve calcification by comparing alpha-Gal-positive and alpha- Gal-deficient (GT-KO) pig pericardium was examined to improve the use of heart valve in xenotransplantation (Lila et al., 2010). Qv et al. (2009) observed that complement is dysregulated in heart xenotransplanation in pig-to-primate models. These reports suggest that the rejection should be overcome in porcine heart xenotransplantation. Cooper et al. (2010) provided several guidances as preliminary requirements in animal models in porcine xenotransplantation. i) Heterotopically-placed pig heart grafts survive and function fairly consistently (eg, 7 of 10) for at least 6 months. ii) Orthotopically-placed pig heart grafts survive and function fairly consistently (eg, 7 of 10) for >3 months, with some primates surviving >6 months (Cooper et al., 2000). iii) Absence of life-threatening consumptive coagulopathy (Buhler et al., 2000; Lin et al., 2009; 2010). 4. Low incidence of immunosuppression- related complications, such as infection and malignancy (Teotia et al., 2005). They also insisted that patient selection must be considered. In the heart, the left ventricular pressure by telemetry proved to be the most valuable parameter to assess graft heart function in pig-to-primate models (Horvath et al., 2010). The lack of sufficient numbers of donor organs resulted in the deaths of the patients. Xenotransplantation using pig organs could resolve the shortage of suitable donor organs (Cooper et al., 2002). The initiating cause of failure of pig xenografts may be antibody-mediated injury to the endothelium, leading to the development of microvascular thrombosis. In this paper, we would like to briefly introduce the major organs of porcine xenotransplantation. PORCINE GENETIC ENGINEERED MODEL FOR XENOTRANSPLANTATION Techniques for porcine genetic modification are crucial Min et al. (2010) Asian-Aust. J. Anim. Sci. 23(11):1535-1542 1537 Porcine renal xenotransplantation Kirkeby and Mikkelsen (2008) reported the distribution of the alphaGal- and the non-alphaGal T-antigens in the pig kidney, suggesting the potential targets for rejection in pig- to-man xenotransplantation. Kidneys from normal pigs were almost rejected rapidly by antibody-mediated complement injury. In HAR, the recognition of pig antigens, predominantly Galα1,3Gal (Gal), by primate natural antibodies leads to complement activation, resulting in extensive intravascular coagulation and thrombosis, endothelial injury, interstitial hemorrhage and edema, and Min et al. (2010) Asian-Aust. J. Anim. Sci. 23(11):1535-1542 1538 infiltration of polymorphonuclear leukocytes into the tissues (Shimizu and Yamada, 2006). An attempt to eliminate hyperacute rejection with human and baboon antibody in GalT-KO kidney has been tried (Diswall et al., 2010). Chen et al. (2005) reported that acute rejection is associated with antibodies to non-Gal antigens in baboons using Gal-knockout pig kidneys. They also suggest that non-Gal antibodies are associated with the development of acute humoral xenograft rejection of hDAF transgenic porcine kidneys in baboons receiving anti-Gal antibody neutralization therapy (Chen et al., 2005). Renal endothelial heterogeneity accelerates AVR in pig-to-baboon xenotransplantation, compared to heart xenografts (Knosalla et al., 2009). With full immunosuppression, there was deposition of IgM, IgG, and complement, and neutrophil and macrophage infiltration, but there was minimal T and B cell infiltration, and no evidence of a T- cell-dependent elicited antibody response, extending the kidney xenograft to 3 days (Ezzelarab et al., 2009). Recently Lin et al. (2010) reported that prevention of recipient platelet activation may be crucial for successful pig-to-primate kidney transplantation. The challenges for the prevention of human and baboon blood immune responses in kidney transplantation have been attempted (Diswall et al., 2010). of porcine pancreatic islets xenograft into non-human primates was restricted to type I diabetes (van der Windt et al., 2009; Cooper and Casu, 2009), the usefulness in type II diabetes also remains to be attempted. Komoda et al. (2005) have demonstrated that islets from transgenic pigs expressing N-acetylglucosaminyltransferase-III showed prolonged survival after transplantation into cynomolgus monkeys. Pancreatic islet xenotransplantation High yields of pure and viable porcine islet cells to be used for microencapsulation are crucial for successful xenotransplantation and the reduction of the damage in porcine islet isolation should be considered (Stiegler et al., 2010). Islet transplantation into the portal vein is the current clinical practice. This method can induce the dysfunction in islet engraftment and survival owing to low oxygen tension, an active innate immune system, and the provocation of an inflammatory response, eventually resulting in the loss of many transplanted islets. Thus, subcutaneous transplantation has been recommended as an alternative choice (van der Windt et al., 2008). The major strategies used to reduce hyperacute hyperacute rejection such as depletion of anti-Gal antibodies and genetic engineering of swine to express human complement regulatory proteins to decrease complement deposition, might impact host defenses against viral infection (Meije et al., 2010). Increased levels of circulating PERV virus have not been detected in the GalT- KO swine or in immunosuppressed baboon recipients of GalT-KO grafts (Issa et al., 2007). To reduce the risk of PERV infection in xenograft recipients, diverse strategies are attempted, including use of nontransmitting swine or swine without active PERV loci as source animals, use of antiretroviral agents in recipients, viral vaccines, or the reduction of viral replication in vitro using RNA interference, various antibody therapies and amplification of antiviral restriction factors (Dieckhoff et al., 2007; Meije et al., 2010). In addition to PERV, porcine cytomegalovirus, herpesvirus, and hepatitis E virus also should be controlled since these viruses are common in immune-suppressed recipients of allotransplants (Mueller et al., 2004; Kamar et al., 2008). Adult and neonatal pig islets xeno-transplanted in immunosuppressive nonhuman primates survived for more than 6 months (Cardona et al., 2006; Hering et al., 2006). These reports are encouraging since porcine pancreatic xenograft can survive longer than other xenograft organs, suggesting the crucial clue in xenotransplantation. Several investigators alleged that xenotransplantaion of pig pancreatic primoria may be a candidate model as a therapeutics for both types 1 and 2 diabetes (Thomas et al., 1995; Rogers et al., 2006). Rogers et al. (2007) demonstrated that transplantation of embryonic day (E) 28 (E28) pig pancreatic primordia into the mesentery of STZ- diabetic rhesus macaques reduced insulin requirement, suggesting the availability for diabetes. Although the report PORCINE VIRUS INFECTION Substitution of human organs to pig organs raises concerns about the risks of transfer of infectious agents from the pig cells to xenotransplantation recipients. Among many viruses, porcine endogenous retroviruses (PERVs) are important candidate to these problems. A major concern in pig-to-human xenotransplantations is the potential risk of transmission of PERVs integrated in the pig genome. Careful selection and/or genetic-engineering of pig herds should aid in minimizing the risk of PERVs infection and/or pathogenicity. Two types of PERV are present in pigs, human-tropic PERV-A and PERV-B, which are both present in the genome of all pigs, and PERV-C, which is not ubiquitous and infects only pig cells (Denner et al., 2009). Yu et al. (2009) indicated that a short-term and long-term PERV infection of HEK-293 cells in vitro does not result in any significant changes in host cells as well as in PERV LTR sequence. CONCLUSION In this review, we described the basic concept of the Min et al. (2010) Asian-Aust. J. Anim. Sci. 23(11):1535-1542 1539 rejection in pig-to primate xenotransplantation. The rejection of porcine xenotransplantation should be solved before clinical application to human. We also discussed a number of genetic engineered pig models used in xenotransplantation, and several important organs in porcine xenotransplantation. Further research for genetic modification of the organ-source pig will be prosperous. These may be associated with the identification of pig non- Gal antigens that are targets for natural or elicited antibodies, natural killer cells, and/or macrophages. Although genetically-modified pigs have overcome the problem of HAR, coagulation dysregulation between species remains an important challenge. In addition, the risks of the approaches to cell function and to the health of the recipient should be carefully evaluated. It is important to emphasize that the escape from PERV is essential for the complete success of porcine xenotransplantatoin into non- human primates before human clinical trials. It is generally recognized that success of xentransplantation into non- human primate model should be preceded prior to human clinical application. Therefore, all efforts to solve many obstacles in porcine xenotransplantation should probably be performed. Long-term survival of neonatal porcine islets in nonhuman primates by targeting costimulation pathways. Nat. Med. 12:304-306. Chen, G., H. Qian, T. Starzl, H. Sun, B. Garcia, X. Wang, Y. Wise, Y. Liu, Y. Xiang, L. Copeman, W. Liu, A. Jevnikar, W. Wall, D. K. Cooper, N. Murase, Y. Dai, W. Wang, Y. Xiong, D. J. White and R. Zhong. 2005. Acute rejection is associated with antibodies to non-Gal antigens in baboons using Gal-knockout pig kidneys. Nat. Med. 11:1295-1298. Chen, G., H. Sun, H. Yang, D. Kubelik, B. Garcia, Y. Luo, Y. Xiang, A. Qian, L. Copeman, W. Liu, C. J. Cardella, W. Wang, Y. Xiong, W. Wall, D. J. White and R. Zhong. 2006. The role of anti-non-Gal antibodies in the development of acute humoral xenograft rejection of hDAF transgenic porcine kidneys in baboons receiving anti-Gal antibody neutralization therapy. Transplantation 81:273-283. Cooper, D. K. and A. Casu. 2009. The International Xenotransplantation Association consensus statement on conditions for undertaking clinical trials of porcine islet products in type 1 diabetes--chapter 4: Pre-clinical efficacy and complication data required to justify a clinical trial. Xenotransplantation 16:229-238. Cooper, D. K. and J. J. Teuteberg. 2010. Pig heart xenotransplantation as a bridge to allotransplantation. J. Heart Lung Transplant. 29:838-840. Cooper, D. REFERENCES Bauer, A., J. Postrach, M. Thormann, S. Blanck, C. Faber, B. Wintersperger, S. Michel, J. M. Abicht, F. Christ, C. Schmitz, M. Schmoeckel, B. Reichart and P. Brenner. 2010. First experience with heterotopic thoracic pig-to-baboon cardiac xenotransplantation. Xenotransplantation 17:243-249. pulmonary diseases. J. Heart Lung Transplant. 19:1125-11 Cooper, D. K., B. Gollackner and D. H. Sachs. 2002. Will the pig solve the transplantation backlog? Annu. Rev. Med. 53:133. Brunetti, D., I. Lagutina, S. Colleoni, R. Duchi, F. Calabrese, M. Seveso, E. Cozzi, G. Lazzari, F. Lucchini and C. Galli. 2008. Transgene expression of green fluorescent protein and germ line transmission in cloned pigs derived from in vitro transfected adult fibroblasts. Cloning Stem Cells 10:409-419. Cooper, D. K., E. Koren and R. Oriol. 1993. Genetically engineered pigs. Lancet 342: 682-683. Cozzi, E., A. W. Tucker, G. A. Langford, G. Pino-Chavez, L. Wright, M. J. O'Connell, V. J. Young, R. Lancaster, M. McLaughlin, K. Hunt, M. C. Bordin and D. J. White. 1997. Characterization of pigs transgenic for human decay- accelerating factor. Transplantation 64:1383-1392. Bühler, L., M. Basker, I. P. Alwayn, C. Goepfert, H. Kitamura, T. Kawai, S. Gojo, T. Kozlowski, E. L. Ierino, M. Awwad, D. H. Sachs, R. Sackstein, S. C. Robson and D. K. Cooper. 2000. Coagulation and thrombotic disorders associated with pig organ and hematopoietic cell transplantation in nonhuman primates. Transplantation 70:1323-1331. d'Apice, A. J. and P. J. Cowan. 2009. Xenotransplantation: the next generation of engineered animals. Transpl Immunol. 21:111- 115. Denner, J., H. J. Schuurman and C. Patience. 2009. The International Xenotransplantation Association consensus statement on conditions for undertaking clinical trials of porcine islet products in type 1 diabetes-chapter 5: Strategies to prevent transmission of porcine endogenous retroviruses. Xenotransplantation 16:239-248. Byrne, G. W., K. R. McCurry, M. J. Martin, S. M. McClellan, J. L. Platt and J. S. Logan. 1997. Transgenic pigs expressing human CD59 and decay-accelerating factor produce an intrinsic barrier to complement-mediated damage. Transplantation 63:149-155. Cabot, R. A., B. Kuhholzer, A. W. Chan, L. Lai, K. W. Park, K. Y. Chong, G. Schatten, C. N. Murphy, L. R. Abeydeera, B. N. Day and R. S. Prather. 2001. Transgenic pigs produced using in vitro matured oocytes infected with a retroviral vector. Anim. Biotechnol. 12:205-214. Diamond, L. E., C. M. Quinn, M. J. Martin, J. Lawson, J. L. Platt and J. S. Logan. 2001. A human CD46 transgenic pig model system for the study of discordant xenotransplantation. CONCLUSION K., A. Keogh, J. Brink, P. A. Corris, W. Klepetko, R. N. Pierson, M. Schmoeckel, R. Shirakura and L. Warner Stevenson L. 2000. Report of the Xenotransplantation Advisory Committee of the International Society for Heart and Lung Transplantation: the present status of xenotransplantation and its potential role in the treatment of end-stage cardiac and pulmonary diseases. J. Heart Lung Transplant. 19:1125-1165. REFERENCES Ezzelarab, M., B. Garcia, A. Azimzadeh, H. Sun, C. C. Lin, H. Hara, S. Kelishadi, T. Zhang, Y. J. Lin, H. C. Tai, R. Wagner, J. Thacker, N. Murase, K. McCurry, R. N. Barth, D. Ayares, R. N. Pierson 3rd and D. K. Cooper. 2009. The innate immune response and activation of coagulation in alpha1,3- galactosyltransferase gene-knockout xenograft recipients. Transplantation 87:805-812. Khalfoun, B., D. Barrat, H. Watier, M. C. Machet, B. Arbeille- Brassart, J. G. Riess, H. Salmon, Y. Gruel, P. Bardos and Y. Lebranchu. 2000. Development of an ex vivo model of pig kidney perfused with human lymphocytes. Analysis of xenogeneic cellular reactions. Surgery 128:447-457 Gewartowska, M. and W. L. Olszewski. 2007. Hepatocyte transplantation-biology and application. Ann. Transplant. 12:27-36. Kirk, A. D. and D. M. Harlan. 2001. Reply: Thromboembolic complications after treatment with monoclonal antibody against CD40 ligand. Nat. Med. 6:114. Good, A. H., D. K. Cooper, A. J. Malcolm, R. M. Ippolito, E. Koren, F. A. Neethling, Y. Ye, N. Zuhdi and L. R. Lamontagne. 1992. Identification of carbohydrate structures that bind human antiporcine antibodies: implications for discordant xenografting in humans. Transplant Proc. 24:559-562. Kirkeby, S. and H. B. Mikkelsen. 2008. Distribution of the alphaGal- and the non-alphaGal T-antigens in the pig kidney: potential targets for rejection in pig-to-man xenotransplantation. Immunol Cell Biol. 86:363-371. Hara, H., B. Gridelli, Y. J. Lin, A. Marcos and D. K. Cooper. 2008. Liver xenografts for the treatment of acute liver failure: clinical and experimental experience and remaining immunologic barriers. Liver Transpl. 14:425-434. Klymiuk, N., B. Aigner, G. Brem and E. Wolf. 2010. Genetic modification of pigs as organ donors for xenotransplantation. Mol. Reprod. Dev. 77:209-221. p Knosalla, C., K. Yazawa, A. Behdad, N. Bodyak, H. Shang, L. Bühler, S. Houser, B. Gollackner, A. Griesemer, I. Schmitt- Knosalla, H. J. Schuurman, M. Awwad, D. H. Sachs, D. K. Cooper, K. Yamada, A. Usheva and S. C. Robson. 2009. Renal and cardiac endothelial heterogeneity impact acute vascular rejection in pig-to-baboon xenotransplantation. Am. J. Transplant. 9:1006-1016. Hering, B. J., M. Wijkstrom, M. L. Graham, M. Hårdstedt, T. C. Aasheim, T. Jie, J. D. Ansite, M. Nakano, J. Cheng, W. Li, K. Moran, U. Christians, C. Finnegan, N. Kirchhof and H. J. Schuurman. 2006. Prolonged diabetes reversal after intraportal xenotransplantation of wild-type porcine islets in immunosuppressed nonhuman primates. Nat. Med. 12:301-303. Hisashi, Y., K. Yamada, K. Kuwaki, Y. L. Tseng, F. J. Dor, S. L. REFERENCES Transplantation 71:132-142. Dieckhoff, B., A. Karlas, A. Hofmann, W. A. Kues, B. Petersen, A. Pfeifer, H. Niemann, R. Kurth and J. Denner. 2007. Inhibition of porcine endogenous retroviruses (PERVs) in primary porcine cells by RNA interference using lentiviral vectors. Cardona, K., G. S. Korbutt, Z. Milas, J. Lyon, J. Cano, W. Jiang, H. Bello-Laborn, B. Hacquoil, E. Strobert, S. Gangappa, C. J. Weber, T. C. Pearson, R. V. Rajotte and C. P. Larsen. 2006. Min et al. (2010) Asian-Aust. J. Anim. Sci. 23(11):1535-1542 1540 Hughes, H. C. 1986. Swine in cardiovascular research. Lab. Anim. Sci. 36:348-350. Hughes, H. C. 1986. Swine in cardiovascular research. Lab. Anim. Sci. 36:348-350. Arch. Virol. 152:629-634. Arch. Virol. 152:629-634. Diswall, M., J. Angström, H. Karlsson, C. J. Phelps, D. Ayares, S. Teneberg and M. E. Breimer. 2010. Structural characterization of alpha1,3-galactosyltransferase knockout pig heart and kidney glycolipids and their reactivity with human and baboon antibodies. Xenotransplantation 17:48-60. Inverardi, L. and R. Pardi. 1994. Early events in cell-mediated recognition of vascularized xenografts: cooperative interactions between selected lymphocyte subsets and natural antibodies. Immunol. Rev. 141:71-93. Issa, N. C., R. A. Wilkinson, A. Griesemer, D. K. Cooper, K. Yamada, D. H. Sachs and J. A. Fishman. 2008. Absence of replication of porcine endogenous retrovirus and porcine lymphotropic herpesvirus type 1 with prolonged pig cell microchimerism after pig-to-baboon xenotransplantation. J. Virol. 82:12441-12448. Ekser, B., B. Gridelli, A. J. Tector and D. K. Cooper. 2009. Pig liver xenotransplantation as a bridge to allotransplantation: which patients might benefit? Transplantation 88:1041-1049. Ekser, B., G. J. Echeverri, A. C. Hassett, M. H. Yazer, C. Long, M. Meyer, M. Ezzelarab, C. C. Lin, H. Hara, D. J. van der Windt, E. M. Dons, C. Phelps, D. Ayares, D. K. Cooper and B. Gridelli. 2010. Hepatic function after genetically engineered pig liver transplantation in baboons. Transplantation (in press) Kamar, N., J. Selves, J. M. Mansuy, L. Ouezzani, J. M. Péron, J. Guitard, O. Cointault, L. Esposito, F. Abravanel, M. Danjoux, D. Durand, J. P. Vinel, J. Izopet and L. Rostaing. 2008. Hepatitis E virus and chronic hepatitis in organ-transplant recipients. N. Engl. J. Med. 358:811-817. Ekser, B., P. Rigotti, B. Gridelli and D. K. Cooper. 2009. Xenotransplantation of solid organs in the pig-to-primate model. Transpl. Immunol. 21:87-92. Kawai, T., D. Andrews, R. B. Colvin, D. H. Sachs and A. B. Cosimi. 2000. Thromboembolic complications after treatment with monoclonal antibody against CD40 ligand. Nat. Med. 6:114. REFERENCES Houser, S. C. Robson, H. J. Schuurman, D. K. Cooper, D. H. Sachs, R. B. Colvin and A. Shimizu. 2008. Rejection of cardiac xenografts transplanted from alpha1,3- galactosyltransferase gene-knockout (GalT-KO) pigs to baboons. Am. J. Transplant. 8:2516-2526. Komoda, H., S. Miyagawa, T. Omori, Y. Takahagi, H. Murakami, T. Shigehisa, T. Ito, H. Matsuda and R. Shirakura. 2005. Survival of adult islet grafts from transgenic pigs with N- acetylglucosaminyltransferase-III (GnT-III) in cynomolgus monkeys. Xenotransplantation 12:209-216. Hofmann, A., B. Kessler, S. Ewerling, M. Weppert, B. Vogg, H. Ludwig, M. Stojkovic, M. Boelhauve, G. Brem, E. Wolf and A. Pfeifer. 2003. Efficient transgenesis in farm animals by lentiviral vectors. EMBO Rep. 4:1054-1060. Lavitrano, M., M. Busnelli, M. G. Cerrito, R. Giovannoni, S. Manzini and A. Vargiolu. 2006. Sperm-mediated gene transfer. Reprod. Fertil. Dev. 18:19-23. Lee, H., B. Lee, Y. Kim, N. Paik and H. Rho. 2010. Characterization of transgenic pigs that express human decay accelerating factor and cell membrane-tethered human tissue factor pathway inhibitor. Reprod. Domest. Anim. (in press) Horvath, K. A., P. C. Corcoran, A. K. Singh, R. F. Hoyt, C. Carrier, M. L. 3rd Thomas and M. M. Mohiuddin. 2010. Left ventricular pressure measurement by telemetry is an effective means to evaluate transplanted heart function in experimental heterotopic cardiac xenotransplantation. Transplant. Proc. 42:2152-2155. Li, L., D. Pang, T. Wang, Z. Li, L. Chen, M. Zhang, N. Song, D. Nie, Z. Chen, L. Lai and H. Ouyang. 2009. Production of a reporter transgenic pig for monitoring Cre recombinase activity. 1541 Min et al. (2010) Asian-Aust. J. Anim. Sci. 23(11):1535-1542 Biochem. Biophys. Res. Commun. 382:232-235. Biochem. Biophys. Res. Commun. 382:232-235. Curr. Opin. Organ Transplant. 14:154-160. Lila, N., C. G. McGregor, S. Carpentier, J. Rancic, G. W. Byrne and A. Carpentier. 2010. Gal knockout pig pericardium: new source of material for heart valve bioprostheses. J. Heart Lung Transplant. 29:538-543. Qv, J., G. Chen, Z. Shen, G. Jiang, J. Ding, W. Gao, and C. Chen. 2009. Activation of complement in the discordant heart xenotransplantation of pig-to-monkey model and the impact of intrathymic inoculation of xenogeneic antigen combined with whole-body gamma-radiation. Transplant. Proc. 41:3905-3908. Lin, C. C., D. K. Cooper and A. Dorling. 2009. Coagulation dysregulation as a barrier to xenotransplantation in the primate. Transpl. Immunol. 21:75-80. Ramirez, P., J. Yelamos, P. Parrilla and R. Chavez. 2001. Hepatic xenotransplantation will benefit from strategies aimed to reduce complement activation. Liver Transpl. 7:562-563. Lin, C. C., M. REFERENCES Tönjes and J. A. Fishman. 2010. Retroviral restriction factors and infectious risk in xenotransplantation. Am. J. Transplant. 10:1511-1516. Mueller, N. J., C. Livingston, C. Knosalla, R. N. Barth, S. Yamamoto, B. Gollackner, F. J. Dor, L. Buhler, D. H. Sachs, K. Yamada, D. K. Cooper and J. A. Fishman. 2004. Activation of porcine cytomegalovirus, but not porcine lymphotropic herpesvirus, in pig-to-baboon xenotransplantation. J. Infect. Dis. 189:1628-1633. Rogers, S. A., F. Chen, M. R. Talcott, C. Faulkner, J. M. Thomas, M. Thevis and M. R. 2007. Hammerman. Long-term engraftment following transplantation of pig pancreatic primordia into non-immunosuppressed diabetic rhesus macaques. Xenotransplantation 14:591-602. Sachs, D. H. and C. Galli. 2009. Genetic manipulation in pigs. Curr. Opin. Organ Transplant. 14:148-153. Pereira-Sampaio, M. A., L. A. Favorito and F. J. Sampaio. 2004. Pig kidney: anatomical relationships between the intrarenal arteries and the kidney collecting system. Applied study for urological research and surgical training. J. Urol. 172:2077- 2081. Shimizu, A. and K. Yamada. 2006. Pathology of renal xenograft rejection in pig to non-human primate transplantation. Clin. Transplant. 20:46-52. Petersen, B., W. Ramackers, A. Tiede, A. Lucas-Hahn, D. Herrmann, B. Barg-Kues, W. Chuettler, L. Friedrich, R. Schwinzer, M. Winkler and H. Niemann. 2009. Pigs transgenic for human thrombomodulin have elevated production of activated protein C. Xenotransplantation 16:486-495. Shimizu, A., Y. Hisashi, K. Kuwaki, Y. L. Tseng, F. J. Dor, S. L. Houser, S. C. Robson, H. J. Schuurman, D. K. Cooper, D. H. Sachs, K. Yamada and R. B. Colvin. 2008. Thrombotic microangiopathy associated with humoral rejection of cardiac xenografts from alpha1,3-galactosyltransferase gene-knockout pigs in baboons. Am. J. Pathol. 172:1471-1481. Pierson, R. N. 3rd., A. Dorling, D. Ayares, M. A. Rees, J. D. Seebach, J. A. Fishman, B. J. Hering and D. K. Cooper. 2009. Current status of xenotransplantation and prospects for clinical application. Xenotransplantation 16:263-280. Smolenski, R. T., M. Forni, M. Maccherini, M. L. Bacci, E. M. Slominska, H. Wang, P. Fornasari, R. Giovannoni, F. Simeone, A. Zannoni, G. Frati, K. Suzuki, M. H. Yacoub and M. Lavitrano. 2007. Reduction of hyperacute rejection and protection of metabolism and function in hearts of human decay accelerating factor (hDAF)-expressing pigs. Cardiovasc. Res. 73:143-152. Pierson, R. N. III. 2009. Antibody-mediated xenograft injury: mechanisms and protective strategies. Transpl. Immunol. 21: 65-69. Platt, J. L. 1998. New directions for organ transplantation. Nature 392:11-17. Stiegler, P., V. Stadlbauer, F. Hackl, S. Schaffellner, F. Iberer, J. Greilberger, D. Strunk, S. Zelzer, C. REFERENCES Ezzelarab, R. Shapiro, B. Ekser, C. Long, H. Hara, G. Echeverri, C. Torres, H. Watanabe, D. Ayares, A. Dorling and D. K. Cooper. 2010. Recipient tissue factor expression is associated with consumptive coagulopathy in pig-to-primate kidney xenotransplantation. Am. J. Transplant. 10:1556-1568. Ramirez, P., R. Chavez, M. Majado, V. Munitiz, A. Muñoz, Q. Hernandez, C. G. Palenciano, G. Pino-Chavez, M. Loba, A. Minguela, J. Yelamos, M. R. Gago, A. S. Vizcaino, H. Asensi, M. G. Cayuela, B. Segura, F. Marin, A. Rubio, T. Fuente, R. Robles, F. S. Bueno, T. Sansano, F. Acosta, J. M. Rodriguez, F. Navarro, J. Cabezuelo, E. Cozzi, D. J. White, R. Y. Calne and P. Parrilla. 2000. Life-supporting human complement regulator decay accelerating factor transgenic pig liver xenograft maintains the metabolic function and coagulation in the nonhuman primate for up to 8 days. Transplantation 70:989- 998. Ramirez, P., R. Chavez, M. Majado, V. Munitiz, A. Muñoz, Q. Hernandez, C. G. Palenciano, G. Pino-Chavez, M. Loba, A. Minguela, J. Yelamos, M. R. Gago, A. S. Vizcaino, H. Asensi, M. G. Cayuela, B. Segura, F. Marin, A. Rubio, T. Fuente, R. Robles, F. S. Bueno, T. Sansano, F. Acosta, J. M. Rodriguez, F. Navarro, J. Cabezuelo, E. Cozzi, D. J. White, R. Y. Calne and P. Parrilla. 2000. Life-supporting human complement regulator d l ti f t t i i li ft Lunney, J. K. 2007. Advances in swine biomedical model genomics. Int. J. Biol. Sci. 3:179-184. Makowa, L., D. V. Cramer, A. Hoffman, M. Breda, L. Sher, G. Eiras-Hreha, P. J. Tuso, C. Yasunaga, C. A. Cosenza and G. D. Wu. 1995. The use of pig liver xenograft for temporary support of a patient with fulminant hepatic failure. Transplantation 59:1654-1659. Ravelingien, A., F. Mortier, E. Mortier, I. Kerremans and J. Braeckman. 2004. Proceeding with clinical trials of animal to human organ transplantation: a way out of the dilemma. J. Med. Ethics. 30:92-98. Manzini, S., A. Vargiolu, I. M. Stehle, M. L. Bacci, M. G. Cerrito, R. Giovannoni, A. Zannoni, M. R. Bianco, M. Forni, P. Donini, M. Papa, H. J. Lipps and M. Lavitrano. 2006. Genetically modified pigs produced with a nonviral episomal vector. Proc. Natl. Acad. Sci. USA. 103:17672-17677. Rogers, S. A., F. Chen, M. Talcott, H. Liapis and M. R. Hammerman. 2006. Glucose tolerance normalization following transplantation of pig pancreatic primordia into non- immunosuppressed diabetic ZDF rats. Transpl. Immunol. 16:176-184. Meije, Y., R. R. REFERENCES Lackner and K. Tscheliessnigg. 2010. Prevention of oxidative stress in porcine islet isolation. J. Artif. Organs 13:38-47. Prather, R. S., R. J. Hawley, D. B. Carter, L. Lai and J. L. Greenstein. 2003. Transgenic swine for biomedicine and agriculture. Theriogenology 59:115-123. Puga Yung, G., M. K. Schneider and J. D. Seebach. 2009. Immune responses to alpha1,3 galactosyltransferase knockout pigs. Tector, A. J., M. Berho, J. A. Fridell, A. DiCarlo, S. Liu, C. Soderland, J. S. Barkun, P. Metrakos and J. I. Tchervenkov. Min et al. (2010) Asian-Aust. J. Anim. Sci. 23(11):1535-1542 1542 2001. Rejection of pig liver xenografts in patients with liver failure: implications for xenotransplantation. Liver Transpl. 7:82-89. van der Windt, D. J., R. Bottino, A. Casu, N. Campanile, C. Smetanka, J. He, H. Murase, H. Hara, S. Ball, B. E. Loveland, D. Ayares, F. G. Lakkis, D. K. Cooper and M. Trucco. 2009. Long-term controlled normoglycemia in diabetic non-human primates after transplantation with hCD46 transgenic porcine islets. Am. J. Transplant. 9:2716-2726. Teotia, S. S., R. C. Walker, J. M. Schirmer, H. D. Tazelaar, M. G. Michaels, J. M. Risdahl, G. W. Byrne, J. S. Logan and C. G. McGregor. 2005. Prevention, detection, and management of early bacterial and fungal infections in a preclinical cardiac xenotransplantation model that achieves prolonged survival. Xenotransplantation 12:127-133. Vodicka, P., K. Jr. Smetana, B. Dvoránková, T. Emerick, Y. Z. Xu, J. Ourednik, V. Ourednik and J. Motlík. 2005. The miniature pig as an animal model in biomedical research. Ann. NY Acad. Sci. 1049:161-171. Thomas, F. T., T. Ljung, J. Henretta, K. Pittman and J. Thomas. 1995. Reversal of type II (NIDDM) diabetes by pancreas islet transplantation: an emerging new concept in pathophysiology of an enigmatic disease. Transplant. Proc. 27:3167-3169. Weiss, E. H., B. G. Lilienfeld, S. Müller, E. Müller, N. Herbach, B. Kessler, R. Wanke, R. Schwinzer, J. D. Seebach, E. Wolf and G. Brem. 2009. HLA-E/human beta2-microglobulin transgenic pigs: protection against xenogeneic human anti-pig natural killer cell cytotoxicity. Transplantation 87:35-43. Tuso, P. J., D. V. Cramer, C. Yasunaga, C. A. Cosenza, G. D. Wu and L. Makowka. 1993. Removal of natural human xenoantibodies to pig vascular endothelium by perfusion of blood through pig kidneys and livers. Transplantation 55:1375- 1378. Whitelaw, C. B., P. A. Radcliffe, W. A. Ritchie, A. Carlisle, F. M. Ellard, R. N. Pena, J. Rowe, A. J. Clark, T. J. King and K. A. Mitrophanous. 2004. REFERENCES Efficient generation of transgenic pigs using equine infectious anaemia virus (EIAV) derived vector. FEBS Lett. 571:233-236. van der Windt, D. J., G. J. Echeverri, J. N. Ijzermans and D. K. Cooper. 2008. The choice of anatomical site for islet transplantation. Cell Transplant. 17:1005-1014. Yu, P., P. Zhang, L. Zhang, S. F. Li, J. Q. Cheng, Y. R. Lu, Y. P. Li and H. Bu. 2009. Studies on long-term infection of human cells with porcine endogenous retrovirus. Acta Virol. 53:169- 174.
https://openalex.org/W4232641136	https://digitalcommons.uri.edu/cgi/viewcontent.cgi?article=1693&context=oa_diss	English	null	Asymptotic Stepwise Coupled Modes for Horizontally-Variable Acoustic Media	null	2,020	cc-by	33,853	Asymptotic Stepwise Coupled Modes for Horizontally-Variable Asymptotic Stepwise Coupled Modes for Horizontally-Variable Acoustic Media Acoustic Media Charles E. White University of Rhode Island, cewhite83@gmail.com Follow this and additional works at: https://digitalcommons.uri.edu/oa_diss Terms of Use All rights reserved under copyright. Recommended Citation Recommended Citation Recommended Citation Recommended Citation White, Charles E., "Asymptotic Stepwise Coupled Modes for Horizontally-Variable Acoustic Media" (2017). Open Access Dissertations. Paper 672. https://digitalcommons.uri.edu/oa_diss/672 This Dissertation is brought to you by the University of Rhode Island. It has been accepted for inclusion in Open Access Dissertations by an authorized administrator of DigitalCommons@URI. For more information, please contact digitalcommons-group@uri.edu. For permission to reuse copyrighted content, contact the author directly. ASYMPTOTIC STEPWISE COUPLED MODES FOR HORIZONTALLY-VARIABLE ACOUSTIC MEDIA BY CHARLES E. WHITE DOCTOR OF PHILOSOPHY DISSERTATION OF CHARLES E. WHITE OF CHARLES E. WHITE ABSTRACT Eﬃcient and accurate mathematical codes for the prediction of underwater sound propagation are a critical component of SONAR system development and operation. Shallow water provides a unique set of complications to the problem of acoustic propagation prediction due to range dependence of acoustic properties resulting from a high number of wave interactions with the surface and seaﬂoor and bathymetric variation. In this situation, the modes of vibration of the acoustic wave equation become coupled, with a transfer of energy between adjacent modes occurring upon traversing a horizontal change of environment. While mode meth- ods have been developed for solving this range-dependent problem with varying accuracy, the computational intensity of these solutions makes them unsuitable for use in applications where real-time or near real-time performance is desired. The goal of the research presented herein was to develop, implement, and verify an eﬃcient and rigorous coupled-mode solution for acoustic wave propagation in shallow water range-dependent environments. Particular interest was given to de- veloping a solution that maintains analytical integrity while executing in a time window that is useful for tactical applications. A theoretical framework involv- ing a range-expanded inner product for capturing the coupling between modes as they propagate through a horizontally-variable medium is presented. This frame- work includes a novel discretization of the range-dependent acoustic medium. A diﬀerence equations approach, which implements the inner product to account for non-adiabatic energy transfer between modes, is used to recursively compute reﬂection and transmission coeﬃcients throughout the discretized environment. Increased eﬃciency is gained in the method due to the ability to compute coupling via closed-form algebraic expressions and in the application of asymptotic analysis to simplify the transmission and reﬂection coeﬃcient solutions. Eﬃcient and accurate mathematical codes for the prediction of underwater sound propagation are a critical component of SONAR system development and operation. Shallow water provides a unique set of complications to the problem of acoustic propagation prediction due to range dependence of acoustic properties resulting from a high number of wave interactions with the surface and seaﬂoor and bathymetric variation. In this situation, the modes of vibration of the acoustic wave equation become coupled, with a transfer of energy between adjacent modes occurring upon traversing a horizontal change of environment. APPROVED: Dissertation Committee: ABSTRACT While mode meth- ods have been developed for solving this range-dependent problem with varying accuracy, the computational intensity of these solutions makes them unsuitable for use in applications where real-time or near real-time performance is desired. The goal of the research presented herein was to develop, implement, and verify an eﬃcient and rigorous coupled-mode solution for acoustic wave propagation in shallow water range-dependent environments. Particular interest was given to de- veloping a solution that maintains analytical integrity while executing in a time window that is useful for tactical applications. A theoretical framework involv- ing a range-expanded inner product for capturing the coupling between modes as they propagate through a horizontally-variable medium is presented. This frame- work includes a novel discretization of the range-dependent acoustic medium. A diﬀerence equations approach, which implements the inner product to account for non-adiabatic energy transfer between modes, is used to recursively compute reﬂection and transmission coeﬃcients throughout the discretized environment. Increased eﬃciency is gained in the method due to the ability to compute coupling via closed-form algebraic expressions and in the application of asymptotic analysis to simplify the transmission and reﬂection coeﬃcient solutions. ACKNOWLEDGMENTS A number of people are owed thanks for their contribution toward the com- pletion of this dissertation. First, the author would like to express his sincere gratitude towards Prof. Gopu Potty, Prof. James H. Miller, and Dr. Cathy Ann Clark, members of the Doctoral Program Committee. Their constant encourage- ment and presentation of challenging questions have greatly inﬂuenced the author’s growth as a scientist. In addition, Prof. Yang Shen and Dr. Steven Crocker are acknowledged for their feedback and service on the oral examining committees. The support of the Naval Undersea Warfare Center (NUWC) of Newport, Rhode Island is also acknowledged. In particular, thanks is expressed toward Mr. Neil Dubois, director of the NUWC In-house Laboratory Independent Research Program. Support given to the author by his supervisor, Dr. Stephen E. Turner, Head of the Undersea Modeling Branch of the Ranges, Engineering, & Analysis Department is also expressly acknowledged. The author wishes to express his thanks and appreciation to Mr. Roy Deaven- port for his input during the literature review process and to Dr. Russell Herman for multiple conversations concerning the solutions to partial diﬀerential equations. Dr. Georges Dossot is thanked for his assistance with generating the parabolic equation comparison data. Gratitude also goes to Dr. Marcia Isakson and Ms. Fiona Cheung for their time and eﬀort in providing the three-dimensional ﬁnite element data used in the method veriﬁcation process. Finally, the author would like to acknowledge Dr. Gustave A. Leibiger. While his passing did not permit direct contact with the author, his work in the area of underwater acoustic propagation prediction served as a guide and inspiration for much of the research performed in the context of this dissertation. iii TABLE OF CONTENTS ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . iii DEDICATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv TABLE OF CONTENTS . . . . . . . . . . . . . . . . . . . . . . . . . . v LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii CHAPTER 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Asymptotic Analysis . . . . . . . . . . . . . . . . . . . . . . . . 3 1.3 The Acoustic Wave Equation . . . . . . . . . . . . . . . . . . . . 6 1.4 The Acoustic Velocity Potential . . . . . . . . . . . . . . . . . . 12 1.5 The Helmholtz Equation . . . . . . . . . . . . . . . . . . . . . . 15 1.6 Source Representation . . . . . . . . . . . . . . . . . . . . . . . 16 1.7 General Solution: Green’s Function . . . . . . . . . . . . . . . . 20 List of References . . . . . . . . . . . . . DEDICATION This dissertation is dedicated to the Ancestors and to the Orisha, without which I would be lost, and to my spiritual godparent, Raven Morgaine. To my partner, Ian Strickland, my mother, Debbie Carter, my uncle, Jeﬀ Carter, and to my sisters at heart, Elizabeth McCullough and Erica Penardo, thank you for your constant love and support through this journey. To Dr. Gene A. Tagliarini, I thank you for inspiring me to enter the ﬁeld of academic research as an undergraduate. Your selﬂess passion for academic instruction and encouragement has impacted my life in a way that cannot be overstated. iv TABLE OF CONTENTS . . . 16 1.7 General Solution: Green’s Function . . . . . . . . . . . . . . . . 20 List of References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 2 Normal Mode Theory . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Asymptotic Analysis . . . . . . . . . . . . . . . . . . . . . . . . 3 1.3 The Acoustic Wave Equation . . . . . . . . . . . . . . . . . . . . 6 1.4 The Acoustic Velocity Potential . . . . . . . . . . . . . . . . . . 12 1.5 The Helmholtz Equation . . . . . . . . . . . . . . . . . . . . . . 15 1.6 Source Representation . . . . . . . . . . . . . . . . . . . . . . . 16 1.7 General Solution: Green’s Function . . . . . . . . . . . . . . . . 20 List of References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 2 Normal Mode Theory . . . . . . . . . . . . . . . . . . . . . . TABLE OF CONTENTS . . . . . . . . . . . . . . . . 22 2 Normal Mode Theory . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.2 Solution via Generalized Separation of Variables . . . . . . . . . 23 2.3 Wavenumber Decomposition and Evanescent Modes . . . . . . . 28 2.4 Modal Velocities . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 TABLE OF CONTENTS . . . . . . . . . . . . . . . . . . . . . . . . . . v LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii CHAPTER 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Asymptotic Analysis . . . . . . . . . . . . . . . . . . . . . . . . 3 1.3 The Acoustic Wave Equation . . . . . . . . . . . . . . . . . . . . 6 1.4 The Acoustic Velocity Potential . . . . . . . . . . . . . . . . . . 12 1.5 The Helmholtz Equation . . . . . . . . . . . . . . . . . . . . . . 15 1.6 Source Representation . . . . . . . . . . . . . . . . . . . . TABLE OF CONTENTS . . . 23 2.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.2 Solution via Generalized Separation of Variables . . . . . . . . . 23 2.3 Wavenumber Decomposition and Evanescent Modes . . . . . . . 28 2.4 Modal Velocities . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 v Page i 2.5 Solution via Cauchy’s Residue Theorem . . . . . . . . . . . . . . 33 2.6 Range-Dependent Normal Modes . . . . . . . . . . . . . . . . . 36 2.7 Previous Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 List of References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 3 Asymptotic Stepwise Coupled Modes . . . . . . . . . . . . . . . 44 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 3.2 The Range-Expanded Normal Mode Inner Product . . . . . . . 45 3.3 Discretization of the Range-Dependent Medium . . . . . . . . . 50 3.4 Asymptotic Identities . . . . . . . . . . . . . . . . . . . . . . . . 57 3.5 Asymptotic Solution . . . . . . . . . . . . . . . . . . . . . . . . 58 3.6 Cylindrical Wave Extension . . . . . . . . . . . . . . . . . . . . 65 3.7 Reformulation for Numerical Stability . . . . . . . . . . . . . . . TABLE OF CONTENTS 68 3.8 Mode Turnaround and Eigenvalue Lag Distance . . . . . . . . . 71 List of References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 4 Veriﬁcation of Theory . . . . . . . . . . . . . . . . . . . . . . . . . 77 4.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 4.2 Comparison to Benchmark Solution . . . . . . . . . . . . . . . . 78 4.3 Comparison to Parabolic Equation Solution . . . . . . . . . . . 86 4.4 Comparison to 3D Finite Element Solution . . . . . . . . . . . . 90 4.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 List of References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 List of References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 vi g APPENDIX Note on the Eﬀects of Range-Dependent Bathymetry on Modal Group Velocities . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 List of References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 BIBLIOGRAPHY . . . . . . . . . . . . . . . . TABLE OF CONTENTS . . . . . . . . . . . . . . . 104 APPENDIX vii LIST OF FIGURES Figure Page 3.1 Discretization of Horizontally-Variable Bottom Depth . . . . . . 53 3.1a Continuous Environment . . . . . . . . . . . . . . . . 53 3.1b Discretized Environment . . . . . . . . . . . . . . . . 53 3.2 Stepwise Discretization of Subenvironment . . . . . . . . . . . . 54 3.3 Energy Transport Through Discretized Environment . . . . . . 55 3.4 Eigenvalue Lag Distance . . . . . . . . . . . . . . . . . . . . . . 72 3.5 Eﬀect of Mode Turnaround on Transmission Coeﬃcients . . . . 73 3.5a Grazing Angle Turnover . . . . . . . . . . . . . . . . 73 3.5b Transmission Coeﬃcient Discontinuity . . . . . . . . 73 3.6 Artiﬁcial Peak Correction in Transmission Loss Estimate . . . . 75 3.6a Before Correction . . . . . . . . . . . . . . . . . . . . 75 3.6b After Correction . . . . . . . . . . . . . . . . . . . . 75 4.1 Comparison to Benchmark: Upslope . . . . . . . . . . . . . . . 79 4.1a Upslope Scenario . . . . . . . . . . . . . . . . . . . . 79 4.1b Transmission Loss Comparison . . . . . . . . . . . . 79 4.2 Comparison to Benchmark: Downslope . . . . . . . . . . . . . . 80 4.2a Downslope Scenario . . . . . . . . . . . . . . . . . . . 80 4.2b Transmission Loss Comparison . . . . . . . . . . . . 80 4.3 Comparison to Benchmark: Two-slope Uphill . . . . . . . . . . 81 4.3a Two-slope Uphill Scenario . . . . . . . . . . . . . . . 81 4.3b Transmission Loss Comparison . . . . . . . . . . . . LIST OF FIGURES 81 viii Figure Page 4.4 Comparison to Benchmark: Two-slope Downhill . . . . . . . . . 82 4.4a Two-slope Downhill Scenario . . . . . . . . . . . . . 82 4.4b Transmission Loss Comparison . . . . . . . . . . . . 82 4.5 Comparison to Benchmark: 2D Seamount . . . . . . . . . . . . 83 4.5a 2D Seamount Scenario . . . . . . . . . . . . . . . . . 83 4.5b Transmission Loss Comparison . . . . . . . . . . . . 83 4.6 Comparison to Benchmark: 2D Canyon . . . . . . . . . . . . . . 84 4.6a 2D Canyon Scenario . . . . . . . . . . . . . . . . . . 84 4.6b Transmission Loss Comparison . . . . . . . . . . . . 84 4.7 Comparison to Benchmark: General Multi-slope . . . . . . . . . 85 4.7a General Multi-slope Scenario . . . . . . . . . . . . . 85 4.7b Transmission Loss Comparison . . . . . . . . . . . . 85 4.8 Comparison to Parabolic Equation Solution . . . . . . . . . . . 88 4.8a High Frequency 2D Seamount . . . . . . . . . . . . . 88 4.8b ASCM vs. COUPLE . . . . . . . . . . . . . . . . . . 88 4.8c MMPE vs. COUPLE . . . . . . . . . . . . . . . . . . 88 4.8d MMPE vs. ASCM . . . . . . . . . . . . . . . . . . . 88 4.9 MMPE Pressure Field . . . . . . . . . . . . . . . . . . . . . . . 89 4.10 Comparison to 3D Finite Element Solution . . . . . . . . . . . . 91 4.10a COUPLE vs. FEM . . . . . . . . . . . . . . . . LIST OF FIGURES . . . 91 4.10b ASCM vs. FEM . . . . . . . . . . . . . . . . . . . . 91 A.1 Eﬀects of Variable Bottom Depth: Mode 1 . . . . . . . . . . . . 100 A.1a Group Velocity Pertubation . . . . . . . . . . . . . . 100 ix Figure Page A.1b Grazing Angle Perturbation . . . . . . . . . . . . . . 100 A.2 Eﬀects of Variable Bottom Depth: Mode 3 . . . . . . . . . . . . 101 A.2a Group Velocity Pertubation . . . . . . . . . . . . . . 101 A.2b Grazing Angle Perturbation . . . . . . . . . . . . . . 101 A.3 Eﬀects of Variable Bottom Depth: Mode 7 . . . . . . . . . . . . 102 A.3a Group Velocity Pertubation . . . . . . . . . . . . . . 102 A.3b Grazing Angle Perturbation . . . . . . . . . . . . . . 102 x x 1.1 Background Numerical models have been used as standard research tools in the ﬁeld of ocean acoustics since the mid-1970s. Technological leaps in the ﬁelds of com- puter science and computer engineering have made it possible for these models to handle complex nonlinear and inter-coupled phenomena. These models have be- come particularly useful in the discipline of acoustic propagation prediction, where spatially and temporally dynamic acoustic terrain necessitate the simulation of multiple simultaneous physical phenomena to determine how acoustic energy will propagate in the underwater environment. Despite advances in computer micro- processor technology, the use of multiphysics models (such as those used in acoustic propagation prediction) in real-time or near real-time applications still presents a signiﬁcant challenge due to the computational complexity of these models. Since the early days of World War II anti-submarine SONAR systems have played a key part in naval operations. Eﬃcient and accurate mathematical codes for the prediction of underwater sound propagation are a critical component of these systems. In deep water regimes, where the acoustic properties of the ocean are slowly varying with horizontal range, the pressure ﬁeld associated with acous- tic phenomena can be approximated by a sum of linearly independent normal modes. A number of eﬃcient and accurate models have been developed around this range-independent solution to acoustic propagation. In shallow water regimes the assumption of range independence fails in environments with rapidly-varying bathymetry and/or sound speed. In this situation, the modes of vibration of the acoustic wave equation become coupled, with a transfer of energy between adja- cent modes occurring upon traversing a horizontal change of environment. While 1 1 mode methods have been developed for solving this range-dependent problem with varying accuracy, the computational intensity of these solutions often makes them unsuitable for use in the warfare environment, where tactical constraints dictate the need for more eﬃcient codes. The goal of the research presented herein was to develop, implement, and ver- ify an eﬃcient and rigorous coupled-mode solution for acoustic wave propagation in range-dependent environments. Particular interest was given to developing a solution that maintains analytical integrity while executing in a time window that is tactically useful for warfare applications. In this dissertation a theoretical frame- work involving an inner product for capturing the coupling between modes as they propagate through a horizontally-variable medium is presented. 1.1 Background A diﬀerence equa- tions approach, which implements the inner product to account for non-adiabatic energy transfer between modes, is used to recursively compute reﬂection and trans- mission coeﬃcients at discrete interfaces separating stepwise regions throughout the medium. Increased eﬃciency is obtained in the presented method by writing the range- dependent eigenfunctions and eigenvalues associated with the acoustic modes of propagation as an expansion in range. This leads to the ability to estimate the inner product via a truncated Taylor series. Asymptotic analysis is then applied to de- velop a set of asymptotic identities involving the eigenfunctions and eigenvalues in adjacent regions, ultimately leading to simpliﬁed expressions for the reﬂection and transmission coeﬃcients. A novel discretization approach, involving the estimation of the general range-dependent environment via a series of simpler range-dependent subenvironments, results in closed-form expressions for computing non-adiabatic mode coupling. The ultimate payoﬀof this discretization method is the elimina- tion of the need to numerically compute inner product integrals at each interface 2 2 throughout the discretized environment. Elimination of the need to numerically solve the eigenvalue search problem is an additional computational beneﬁt of this approach. 1.2 Asymptotic Analysis Asymptotic analysis is a method of describing limiting behavior. This form of analysis has applications across many areas of science. In computer science, for example, asymptotic analysis is used to analyze the computational eﬃciency of speciﬁc computing algorithms. In applied mathematics asymptotic analysis is used to derive numerical methods for approximating mathematical functions and equations. Regarding this latter case, the following equivalence relation forms the basis of many such analyses: f ∼g f ∼g if and only if lim n→∞ f(n) g(n) = 1. (1.1) (1.1) This is an equivalence relation on the set of functions of the variable n deﬁned on some domain. The equivalence class of f then consists of all functions g which approximate f in the stated limit. In addition to the typical mathematical notation from real and complex anal- ysis and calculus, asymptotic analysis has its own family of notation collectively known as Bachmann-Landau notation. Of particular interest is big-O and little-o notation. In regards to the former, let f and g be two functions deﬁned on some subset of the real numbers. We write: f(x) = O [g(x)] as x →∞, (1.2) (1.2) if and only if there is a positive constant M such that for all suﬃciently large values of x, the absolute value of f(x) is at most M multiplied by the absolute if and only if there is a positive constant M such that for all suﬃciently large values of x, the absolute value of f(x) is at most M multiplied by the absolute 3 value of g(x) [1]. Mathematically that is: value of g(x) [1]. Mathematically that is: \|f(x)\| ≤M\|g(x)\| for all x ≥x0, (1.3) (1.3) where x0 is some large value that exists in the domain of f and g. This can be extended to describe the behavior of f near some real number a: where x0 is some large value that exists in the domain of f and g. This can be extended to describe the behavior of f near some real number a: f(x) = O [g(x)] as x →a, (1.4) (1.4) if and only if there exist positive numbers δ and M such that \|f(x)\| ≤M\|g(x)\| for all \|x −a\| ≤δ. (1.5) (1.5) Little-o notation, while similar to big-O, makes a stronger statement re- garding the relationship between two functions f and g. 1.2 Asymptotic Analysis Informally, if we write f(x) = o [g(x)] we are asserting that g(x) grows much faster than f(x) [1]. Math- ematically we deﬁne little-o as: f(x) = o [g(x)] as x →∞, (1.6) (1.6) if for every positive constant ϵ there exists a constant N such that \|f(x)\| ≤ϵ\|g(x)\| for all x ≥N. (1.7) (1.7) Comparing Equations (1.2) and (1.3) to Equations (1.6) and (1.7) we see that big-O notation only requires that the condition in Equation (1.3) be met for at least one constant M. Conversely, little-o notation requires that the inequality in Equation (1.7) hold for every positive constant ϵ, however small. An example of an asymptotic approximation that arises in acoustics is the asymptotic form of the Hankel function. Hankel functions of the ﬁrst and second kind arise as two linearly independent solutions to Bessel’s equation, which speciﬁes 4 the behavior of the range solutions to the separated Helmholtz equation. In their closed form these functions are written: the behavior of the range solutions to the separated Helmholtz equation. In their closed form these functions are written: H(1) 0 (kr) = J0(kr) + iY0(kr) (1.8) H(1) 0 (kr) = J0(kr) + iY0(kr) (1.8 (1.8) and H(2) 0 (kr) = J0(kr) −iY0(kr), (1.9) (1.9) where J0(kr) and Y0(kr) are Bessel functions of the ﬁrst and second kind, re- spectively, corresponding to Bessel’s equation of order 0. The Hankel functions in Equations (1.8) and (1.9) express outward- and inward-propagating cylindrical wave solutions to the cylindrical wave equation (or vice versa depending on the sign convention chosen for the time dependence of the acoustic source). Hankel functions of the ﬁrst and second kind each admit a contour integral representation. Namely: H(1) ν (z) = 1 π Z C1 e−iz sin ζ+iνζdζ (1.10) (1.10) and and H(2) ν (z) = 1 π Z C2 e−iz sin ζ+iνζdζ, (1.11) H(2) ν (z) = 1 π Z C2 e−iz sin ζ+iνζdζ, (1.11) (1.11) where the contours C1 and C2 exist in the complex ζ-plane. The contour C1 traces a path from A = −π + i∞to B = −i∞while C2 traces a path from A = −i∞to B = π + i∞. 1.2 Asymptotic Analysis Following reference [2], if we consider Re(z) > 0 one can apply the method of steepest descent to express the integrals in Equations (1.10) and (1.11) in the form of a series of inverse powers of z, where z is taken to be large. This leads to asymptotic forms of the Hankel functions of the ﬁrst and second kind. In the case of H(1) ν (z) we have H(1) ν (z) = O "r 2 πzei(z−νπ/2−(π/4)) # for \|z\| →∞. (1.12) (1.12) 5 If we now replace z in Equation (1.12) with kr and let kr →∞and ν = 0 (corresponding to the zeroth order Bessel’s equation) we arrive at an asymptotic form of the Hankel function in Equation (1.8) that is used frequently in acoustic propagation problems: H(1) 0 (kr) ∼ r 2 πkrei(kr−(π/4)). (1.13) (1.13) Similarly, for Equation (1.9) we obtain H(2) 0 (kr) ∼ r 2 πkre−i(kr−(π/4)). (1.14) (1.14) Here k = ω/c is the acoustic wavenumber, ω is the angular frequency of an acoustic source, c is the sound speed of the acoustic medium, and r is range from the source. The asymptotic expressions in Equations (1.13) and (1.14) are often used to model the radiation of outgoing and incoming energy, respectively, in the normal modes solution to the acoustic wave equation to be discussed in detail in later sections. 1.3 The Acoustic Wave Equation In mathematical physics, the acoustic wave equation is a second order par- tial diﬀerential equation (PDE) that governs the propagation of acoustic energy through a material medium. This equation describes the spatial and temporal evolution of acoustic pressure or particle velocity as a set of acoustic waves radi- ates through a particular environment. In underwater acoustics, the environment is speciﬁed through acoustic variables such as sound speed and density in water, as well as through boundary conditions at the surface and seaﬂoor. Except for the case of very large wave amplitudes, a linear wave equation is used to describe acoustic processes underwater, which is a suﬃcient approximation for the problems of interest here. As such, the derivation of the linear acoustic wave equation will now be treated. 6 Derivation of the linear acoustic wave equation involves manipulation of three basic hydrodynamic equations: conservation of mass, conservation of linear mo- mentum, and the thermodynamic equation of state for a ﬂuid. The ﬁrst two of these three equations are derived by applying Reynold’s Transport Theorem to mass and linear momentum, respectively, for an inﬁnitesimally small control vol- ume. Reynold’s Transport Theorem for a control volume states that the time derivative of a particular ﬂuid system property, B, equals the change of B within the control volume plus the ﬂux of B out of the control surface, minus the ﬂux of B into the control surface [3]. Mathematically, we write: d dt(Bsyst) = d dt Z CV βρ dV + Z CS βρ(v • n) dA, (1.15) (1.15) where β = dB/dm is the intensive value, or the amount of B per unit mass in the ﬂuid element, ρ is density, typically measured in kilograms per cubic meters and v is the ﬂuid particle velocity, a vector quantity measured in meters per second. CV represents the control volume, while CS represents the control surface enclosing the control volume, and n is the outward unit normal vector to the control surface. In the case of conservation of mass, the ﬂuid property of interest is the system mass, m. That is, B = m and β = dm/dm = 1 in Equation (1.15). Applying Reynold’s Transport Theorem to this property for an inﬁnitesimally small control volume and recognizing that the time derivative of the system mass is equal to zero (i.e. 1.3 The Acoustic Wave Equation mass conservation) leads to the following hydrodynamic equation for mass, also known as the continuity equation: ∂ρ ∂t + ∇• (ρv) = 0, (1.16) (1.16) where ∇is understood to be the gradient vector deﬁned by ∇= ∂ ∂x, ∂ ∂y, ∂ ∂z . (1.17) (1.17) 7 This equation requires no assumptions except that the density and velocity are continuum functions. That is, the ﬂow may be steady or unsteady, viscous or inviscid, compressible or incompressible. However, the equation does not allow for any source or sink singularities within the ﬂuid element [3]. For conservation of linear momentum, the ﬂuid property of interest becomes the product of the system mass times its velocity. That is, B = mv and β = dB/dm = v in Equation (1.15). Recognizing again that the system mass is constant, the left side of Equation (1.15) becomes the product of mass and ac- celeration. From Newton’s second law we recognize that this product is equal to the sum of all forces acting on the system; that is, the sum of all surface and body forces acting on the control volume at rest. In the case of body forces, we include the gravitational force on the ﬂuid ele- ment. The surface forces are more complicated and include the sum of hydrostatic pressure plus viscous stresses acting on the six ﬂuid element faces. Unlike velocity, which is a three-component vector, the stresses and strain rates corresponding to the surface forces on the ﬂuid element are nine-component tensors. Careful anal- ysis shows that it is not the stresses themselves, but their gradients, that cause a net force on the diﬀerential control volume. Substituting the sum of the body and surface forces into the left side of Reynold’s Transport Theorem for linear momentum leads to the following gen- eral expression for the conservation of linear momentum in a diﬀerential control volume: ρg −∇p + ∇• τij = ρDv Dt , (1.18) ρg −∇p + ∇• τij = ρDv Dt , (1.18) ρg −∇p + ∇• τij = ρDv Dt , (1.18) where D {·} = ∂{·} + v • ∇{·} (1.19) (1.18) where D Dt{·} = ∂ ∂t{·} + v • ∇{·} (1.19) (1.19) is known as the material or total derivative, g is the gravity force vector, and τij is 8 the viscous stress tensor acting on the ﬂuid element. 1.3 The Acoustic Wave Equation This expression holds for ar- bitrary body and surface forces and it states that: gravity force per unit volume+ pressure force per unit volume + viscous force per unit volume = density × acceleration. If we assume frictionless ﬂow and negligible gravitational force the conservation of linear momentum equation is given by: −∇p = ρ ∂v ∂t + v • ∇v , (1.20) (1.20) where p is pressure measured in Newtons per square meter, also referred to as pascals. This equation is known as Euler’s equation and it is the typical form of the conservation of linear momentum that is used in the derivation of the acoustic wave equation. Next we consider the state equation for a ﬂuid. In physics and thermody- namics, an equation of state is a relation between state variables of a thermal system. These variables include but are not limited to pressure, density, temper- ature, entropy, and enthalpy. Typically, knowing the values of two of these state variables allows us to ﬁx a third. That is, state relations typically express one thermodynamic variable as a function of two other thermodynamic variables. In the case of an ideal gas, we have an explicit formula for the equation of state known as the ideal gas law. In the case of ﬂuids, there is no comparable “ideal liquid law.” Instead, the state equation for liquids typically involves an empirical relationship between pressure, density and a third thermodynamic variable. In underwater acoustic applications, this third variable is taken to be entropy, S, and the general equation of state for our ﬂuid of interest is written as: p = p(ρ, S). (1.21) (1.21) Armed with Equations (1.16)-(1.21) we are now ready to begin derivation of the linear acoustic wave equation. This will be done following the notation of reference [2]. [2]. 9 In the absence of a sound wave the undisturbed ﬂuid variables of interest have the following values: ρ′ = ρ0 p′ = p0 v′ = v0 = 0 S′ = S0 (1.22) (1.22) where Q′ represents the strength of our acoustic source. Note that Q0 and v0 are zero since the ﬂuid is at rest (i.e. no acoustic wave is present). Now suppose that the acoustic source is turned on with source strength of Q1. 1.3 The Acoustic Wave Equation We assume this source to be a vibrating object whose amplitude and velocity of vibration are so small that they cause only a small perturbation of the undisturbed ﬂuid variables in Equation (1.22). Let these perturbations be denoted by ρ1, p1, v1, and S1. Note that the source strength, Q1, is of the same order of magnitude as the perturbed ﬂuid variables. Given this information, the total density, pressure, velocity, entropy, and source strength can be represented by the expressions: ρ′ = ρ0 + ρ1 p′ = p0 + p1 v′ = v0 + v1 = v1 S′ = S0 + S1 Q′ = Q0 + Q1 = Q1. (1.23) (1.23) Q′ = Q0 + Q1 = Q1. Substituting the acoustic variables in Equation (1.23) into Equation (1.16) and (1.20) and assuming that squares or higher order terms involving the (small) perturbations of the acoustic variables can be neglected leads to linearized forms of the conservation of mass and conservation of linear momentum equations. Lin- earizing the state equation requires a bit more work. We expand the state equation 10 about the equilibrium state of the ﬂuid, again neglecting second and higher order terms in the acoustic variables. This leads to the following linearized equations: ∂ρ1 ∂t + v1 • ∇ρ0 + ρ0∇• v1 = Q1 (1.24) ∂v1 ∂t = −1 ρ0 ∇p1 (1.25) p1 = ρ1c2. (1.26) ∂ρ1 ∂t + v1 • ∇ρ0 + ρ0∇• v1 = Q1 (1.24) (1.24) ∂v1 ∂t = −1 ρ0 ∇p1 (1.25) p1 = ρ1c2. (1.26) (1.25) (1.26) p1 = ρ1c2. (1.26) To derive a linear wave equation we ﬁrst take the divergence of the momentum equation in Equation (1.25), leading to the expression: ∂ ∂t∇• v1 + ∇• 1 ρ0 ∇p1 = 0. (1.27) (1.27) Next, we recall the constitutive equation for isentropic ﬂow, Next, we recall the constitutive equation for isentropic ﬂow, Next, we recall the constitutive equation for isentropic ﬂow, Dp′ Dt = c2Dρ′ Dt , (1.28) (1.28) where c is the speed of sound in our ﬂuid of interest. We simplify this equation by substituting the perturbed pressure, p′, and the perturbed density, ρ′, from Equation (1.23) into the equation and again assume that squares or higher order terms involving the perturbations of the acoustic variables can be neglected. 1.3 The Acoustic Wave Equation Doing so, we arrive at: ∂p1 ∂t + v1 • ∇p0 = c2 ∂ρ1 ∂t + v1 • ∇ρ0 . (1.29) (1.29) e now combine Equation (1.29) with Equation (1.24) to obtain the expression 1 c2 ∂p1 ∂t + v1 • ∇p0 + ρ0∇• v1 = Q1. (1.30) (1.30) From thermodynamic considerations one has that ∇p0/p0 ≪∇T0/T0, where T0 corresponds to the ambient temperature of the ﬂuid medium. As such, we have that the ∇p0 term is negligible, leading to the following: 1 c2 ∂p1 ∂t + ρ0∇• v1 ≈Q1. (1.31) (1.31) 11 We now diﬀerentiate Equation (1.31) with respect to time and use Equation (1.27) to eliminate ∇• v1, arriving at: 1 c2 ∂2p1 ∂t2 −ρ0∇• 1 ρ0 ∇p1 ≈∂Q1 ∂t . (1.32) (1.32) Going forward, subscripts on the acoustic variables will be dropped and it is understood that all variables refer to their perturbative quantities. In the case of density, we will consider the variation due to the acoustic disturbance to be negligible and ρ will refer to the density of the medium at rest. Taking this into account and rearranging terms in Equation (1.32), the following expression for the linear acoustic wave equation is obtained: ρ∇• 1 ρ∇p −1 c2 ∂2p ∂t2 = −∂Q ∂t . (1.33) (1.33) If we consider density to be constant in space and the source strength to be time independent we arrive at the standard form of the linear homogeneous wave equation for small-amplitude acoustic processes: ∇2p −1 c2 ∂2p ∂t2 = 0. (1.34) (1.34) where ∇2 is the Laplacian operator: where ∇2 is the Laplacian operator: where ∇2 is the Laplacian operator: where ∇2 is the Laplacian operator: ∇2{·} = ∂2 ∂x2{·} + ∂2 ∂y2{·} + ∂2 ∂z2{·}. (1.35) (1.35) 1.4 The Acoustic Velocity Potential 1.4 The Acoustic Velocity Potential Following a procedure similar to that used in deriving the wave equation for pressure, an acoustic wave equation for particle velocity may be obtained. To accomplish this, we will assume that we are a signiﬁcant distance from the source and that ∇ρ0/ρ0 ≪1. Doing so, Equation (1.24) becomes: ∂ρ1 ∂t = −ρ0∇• v1. (1.36) (1.36) 12 12 We now take the gradient of Equation (1.36) and the time derivative of Equation (1.25). Combining the two results using Equation (1.26) and dropping subscripts we arrive at a wave equation for particle velocity. This equation is: 1 ρ∇(ρc2∇• v) −∂2v ∂t2 = 0, (1.37) (1.37) where v is a function of the position vector r and time t. If the density of the acoustic medium is slowly varying, the vector equation in Equation (1.37) may be transformed into a scalar equation by deﬁning a velocity potential Φ(r, t) given by where v is a function of the position vector r and time t. If the density of the acoustic medium is slowly varying, the vector equation in Equation (1.37) may be transformed into a scalar equation by deﬁning a velocity potential Φ(r, t) given by v(r, t) = ∇Φ(r, t). (1.38) (1.38) If one substitutes Equation (1.38) into Equation (1.37) and assumes that ∇ρ = 0, then the following expression is obtained ∇(c2∇2Φ −∂2Φ ∂t2 ) = 0. (1.39) (1.39) This expression is satisﬁed if the velocity potential satisﬁes the scalar wave equation This expression is satisﬁed if the velocity potential satisﬁes the scalar wave equation ∇2Φ −1 c2 ∂2Φ ∂t2 = 0, (1.40) (1.40) which we note is identical to that for pressure in Equation (1.34). A direct relationship between velocity potential and pressure may be obtained through the introduction of a displacement potential, ψ, deﬁned by d(r, t) = ∇ψ(r, t), (1.41) (1.41) where d(r, t) is the particle displacement vector given by the kinematic relationship v = ∂d/∂t. Combining Equations (1.36), (1.26), and (1.41), we arrive at the relationship between pressure and the displacement potential p = −K∇2ψ, (1.42) (1.42) 13 13 where we recognize that K is the bulk modulus deﬁned by where we recognize that K is the bulk modulus deﬁned by (1.43) K = ρc2. (1.43) Equation (1.42) is the equivalent of Hooke’s law for an ideal, linearly elas- tic ﬂuid. 1.4 The Acoustic Velocity Potential Note, that the displacement potential is governed by a wave equation identical to that of pressure and the velocity potential: ∇2ψ −1 c2 ∂2ψ ∂t2 = 0. (1.44) (1.44) Combining Equations (1.42)-(1.44) results in the following alternative expression for the acoustic pressure ∂2ψ p = −ρ∂2ψ ∂t2 . (1.45) (1.45) Next, we assume a continuous wave time harmonic acoustic source, leading to a time dependence of the form e−iωt, where ω = 2πf is angular frequency of the source in radians per second and f is frequency in Hertz. Such a source may be represented by the expression: Q(r, t) = ˆQ(r, ω)e−iωt. (1.46) (1.46) This leads to the following expression for the acoustic pressure: to the following expression for the acoustic pressure: p(r, t) = ˆp(r, ω)e−iωt. (1.47) (1.47) An expression for the acoustic velocity potential is similarly obtained. This is: An expression for the acoustic velocity potential is similarly obtained. This is: Φ(r, t) = ˆΦ(r, ω)e−iωt. (1.48) (1.48) One can show that Φ = ∂ψ/∂t. Using this relationship, we ﬁnally arrive at the following relationship between spatially-dependent pressure and velocity potential for a time harmonic acoustic source: ˆp(r, ω) = iωρˆΦ(r, ω). (1.49) (1.49) 14 14 14 The relationship in Equation (1.49) will be utilized a number of times through- out the remainder of the presented work. While the proposed coupled-mode theory is developed in the context of velocity potentials, there will be times at which con- verting quantities of interest to pressure provides for a more intuitive discussion. 1.5 The Helmholtz Equation Time-independence of the coeﬃcients to the diﬀerential operators in Equa- tion (1.33) allows for the acoustic wave equation to be further simpliﬁed. This is accomplished through application of the following Fourier transform operator: F[f(t)] = ˆf(ω) = Z ∞ −∞ f(t)eiωt dt. (1.50) (1.50) The inverse of which is: The inverse of which is: F−1[ ˆf(ω)] = f(t) = 1 2π Z ∞ −∞ ˆf(ω)e−iωt dω. (1.51) (1.51) Application of the transform in Equation (1.50) to Equation (1.33) leads to what is known as the Helmholtz equation or reduced wave equation, expressed as: Application of the transform in Equation (1.50) to Equation (1.33) leads to what is known as the Helmholtz equation or reduced wave equation, expressed as: ρ∇• 1 ρ∇ˆp(r, ω) + k2(r)ˆp(r, ω) = iω ˆQ(r, ω), (1.52) (1.52) where k(r) is the medium wavenumber at radial frequency ω and sound velocity c(r), that is: k(r) = ω c(r). (1.53) (1.53) Recall that for a time harmonic acoustic source of a single frequency the time- dependent pressure may be written as in Equation (1.47). Under these conditions, Equation (1.52) is equivalently obtained by substituting the expressions on the right sides of Equations (1.46) and (1.47) into the wave equation in Equation (1.33) and carrying out the diﬀerentiation with respect to time. A Helmholtz equation similar to Equation (1.52) may be obtained for the velocity potential by utilizing the relationship given in Equation (1.49). Doing so 15 leads to the result: leads to the result: ∇• 1 ρ∇ρˆΦ(r, ω) + k2(r)ˆΦ(r, ω) = ˆQ(r, ω) ρ . (1.54) (1.54) Note that if one assumes constant density, the left sides of Equations (1.52) and (1.54) are identical. Due to the reduction in dimension, the PDE’s in Equations (1.52) and (1.54) are simpler to solve than the full wave equation. This simpliﬁcation is achieved, however, at the cost of having to evaluate the inverse Fourier transform given by Equation (1.51) to achieve the ﬁnal solution. 1.6 Source Representation Underwater sound due to natural or artiﬁcial phenomena is produced through forced mass injection [4]. For acoustic sources with complex geometries, mathe- matically representing this mass injection term can be diﬃcult. In the case of an omni-directional point source, an analytical expression for the forcing term may be derived by recognizing that the acoustic ﬁeld is produced by a small sphere inducing body forces on the ﬂuid medium through expansions and contractions. These body forces create small particle displacements in the ﬂuid that propagate out from the source as spherical acoustic waves. If we consider the medium to be inﬁnite (i.e. unbounded), then the acoustic ﬁeld will be spherically symmetric and we need only concern ourselves with the radial distance from the center of the source, r. Given this set of assumptions, the particle displacement ﬁeld in Equation (1.41) may be written as d(r, t) = ∂ψ(r, t) ∂r , (1.55) (1.55) where ψ(r, t) is again recognized as the displacement potential function. If we then deﬁne the radius of the sphere with the symbol a and apply a no-slip boundary 16 condition at the source-ﬂuid interface we recognize that condition at the source-ﬂuid interface we recognize that condition at the source-ﬂuid interface we recognize that condition at the source-ﬂuid interface we recognize th (1.56) d(a, t) = D(t), (1.56) where D(t) is surface displacement of the sphere at time t with dimensions of length. To simplify the discussion we will consider the density in the acoustic medium to be constant. Taking this into account, the Fourier-transformed displacement potential for our propagation scenario of interest satisﬁes the Helmholtz equation 1 r2 ∂ ∂rr2 ∂ ∂r + k2 ˆψ(r, ω) = ˆSψ(ω) ˆQ(r) (1.57) (1.57) with boundary condition with boundary condition with boundary condition ˆd(a, ω) = ˆD(ω). (1.58) (1.58) Here ˆQ(r) is representative of our point source and ˆSψ(ω) is a source strength term to be analytically derived. Note that the ﬁeld for this scenario depends solely on range from the source. As such, Equation (1.57) has been written in spherical coordinates for convenience. Considering ranges a signiﬁcant distance from the source (i.e. r ≫0) the displacement potential satisﬁes the homogeneous equivalent of Equation (1.57), the solution of which is a linear combination of incoming and outgoing waves given by the expressions: ˆψ(r, ω) =        (A/r)eikr outgoing wave (B/r)e−ikr incoming wave. 1.6 Source Representation (1.59) (1.59) Since the sphere is assumed to be the only source in the inﬁnite medium there are no incoming waves thus requiring that B = 0. Thus for the current case of interest ˆψ(r, ω) = Aeikr r , (1.60) (1.60) 17 with corresponding displacement ﬁeld with corresponding displacement ﬁeld ˆd(r, ω) = Aeikr ik r −1 r2 , (1.61) (1.61) where Equation (1.61) was obtained by applying the Fourier transform operator in Equation (1.50) to the displacement ﬁeld in Equation (1.55) and carrying out the diﬀerentiation with respect range, r. Our assumed model of a simple point source requires that the radius of the sphere be small compared to the acoustic wavelength. This necessitates that ka ≪1, in which case Equation (1.61) takes the form ˆd(a, ω) = Aeikaika −1 a2 ≈−A a2. (1.62) (1.62) Utilizing Equation (1.58), we then arrive at the following solution for the wave amplitude A: A = −a2 ˆD(ω). (1.63) (1.63) If we now deﬁne the source strength ˆSψ(ω) = 4πa2 ˆD(ω) as the volume- injection amplitude produced by the source at frequency ω, the following solution is obtained for the ﬁeld in the ﬂuid, ˆψ(r, ω) = −ˆSψ(ω)eikr 4πr, (1.64) (1.64) where the source strength ˆSψ(ω) has dimensions of cubic length, or volume. The ratio appearing in Equations (1.64) is known as the free-space Green’s function, which has general form gω(r, r0) = eik\|r−r0\| 4π\|r −r0\|, (1.65) (1.65) where r0 is the location of the source. The free-space Green’s function satisﬁes the following inhomogeneous Helmholtz equation, [∇2 + k2(r)]gω(r, r0) = −δ(r −r0), (1.66) (1.66) 18 which can be veriﬁed by integrating Equation (1.66) over a small volume containing the source point. Taking this into account, if one sets ˆSψ(ω) to unity in Equation (1.57) it is easily veriﬁed that ˆQ(r) = −δ(r), or in general ˆQ(r) = −δ(r −r0). (1.67) (1.67) A source strength for velocity potential, ˆSΦ(ω), may be obtained by recalling that Φ = ∂ψ/∂t. Applying the Fourier transform operator in Equation (1.50) to this relationship and taking into account Equation (1.64), we ﬁnd that ˆSΦ(ω) = −iω ˆSψ(ω) = −iω4πa2 ˆD(ω), (1.68) (1.68) where ˆSΦ(ω) has dimensions of cubic length over time. 1.6 Source Representation Taking this into account, a source is said to have “unit strength” when ˙m = 1, meaning that the source injects one unit of mass per unit of time into the acoustic medium. In SI units this corresponds to 1 kg s−1 of mass injection. Under these conditions ˆSp(ω) = iω and ˆSΦ(ω) = 1/ρ, as in Equations (1.52) and (1.54). 1.6 Source Representation This implies that for the wave equation [∇2 + k2]ˆΦ(r, ω) = ˆSΦ(ω) ˆQ(r), (1.69) (1.69) the following is a solution: the following is a solution: the following is a solution: ˆΦ(r, ω) = −ˆSΦ(ω)eikr 4πr. (1.70) (1.70) A source strength for acoustic pressure, ˆSp(ω), may be obtained in a similar fashion by recalling the relationship in Equation (1.49). This leads to the deﬁnition: ˆSp(ω) = ρω2 ˆSψ(ω) = ρω24πa2 ˆD(ω), (1.71) (1.71) where ˆSp(ω) has dimensions mass over squared time. This implies that for the wave equation [∇2 + k2]ˆp(r, ω) = ˆSp(ω) ˆQ(r), (1.72) (1.72) the following is a solution: the following is a solution: ˆp(r, ω) = −ˆSp(ω)eikr 4πr. (1.73) (1.73) 19 Recalling Equations (1.52) and (1.54), we can redeﬁne ˆSp(ω) and ˆSΦ(ω) as: Recalling Equations (1.52) and (1.54), we can redeﬁne ˆSp(ω) and ˆSΦ(ω) as: Recalling Equations (1.52) and (1.54), we can redeﬁne ˆSp(ω) and ˆSΦ(ω) as: ˆSp(ω) = iω ˙m, (1.74) (1.74) and ˆSΦ(ω) = 1 ρ ˙m, (1.75) where and and ˆSΦ(ω) = 1 ρ ˙m, (1.75) ˆSΦ(ω) = 1 ρ ˙m, (1.7 (1.75) where ˙m = −iρω4πa2 ˆD(ω). (1.76) (1.76) Given these deﬁnitions, we see that for a point source in a medium with non- constant density the Fourier-transformed complex pressure, ˆp(r, ω), satisﬁes the equation: ρ∇• 1 ρ∇ˆp(r, ω) + k2(r)ˆp(r, ω) = ˆSp(ω) ˆQ(r). (1.77) (1.77) Similarly, the Fourier-transformed velocity potential, ˆΦ(r, ω), satisﬁes ∇• 1 ρ∇ρˆΦ(r, ω) + k2(r)ˆΦ(r, ω) = ˆSΦ(ω) ˆQ(r). (1.78) (1.78) Note that ˙m in Equations (1.74) and (1.75) has dimensions of mass over time. As such, this term represents the amount of mass per unit time being injected into the acoustic medium by an omni-directional point source. Taking this into account, a source is said to have “unit strength” when ˙m = 1, meaning that the source injects one unit of mass per unit of time into the acoustic medium. In SI units this corresponds to 1 kg s−1 of mass injection. Under these conditions ˆSp(ω) = iω and ˆSΦ(ω) = 1/ρ, as in Equations (1.52) and (1.54). Note that ˙m in Equations (1.74) and (1.75) has dimensions of mass over time. As such, this term represents the amount of mass per unit time being injected into the acoustic medium by an omni-directional point source. 1.7 General Solution: Green’s Function The application of Green’s functions also allows for the development of a solution to the general problem of a bounded medium with general source. To solve this problem we consider the wave equation: [∇2 + k2(r)]ˆΦ(r, ω) = ˆF(r, ω), (1.79) (1.79) 20 where the acoustic medium occupies a volume V bounded by a surface S, with prescribed conditions at the boundary. Here ˆF(r, ω) is a general forcing term, which includes the source strength and has dimensions of inverse seconds. Next, we formulate a general solution to Equation (1.66) as a sum of the particular solution, gω(r, r0), and a homogeneous solution hω(r, r0), that is, Gω(r, r0) = gω(r, r0) + hω(r, r0). (1.80) (1.80) Recognizing that by deﬁnition, [∇2 + k2(r)]hω(r, r0) = 0, (1.81) (1.81) we ﬁnd that [∇2 + k2(r)]Gω(r, r0) = −δ(r −r0). (1.82) (1.82) Following a process outlined in reference [5], we now multiply Equation (1.79) by Gω(r, r0) and multiply Equation (1.82) by ˆΦ(r, ω). Subtracting the two expres- sions, interchanging r with r0, and integrating over volume we obtain Z V h Gω(r, r0)∇2 0 ˆΦ(r0, ω) −ˆΦ(r0, ω)∇2 0Gω(r, r0) i dV0 = Z V ˆΦ(r0, ω)δ(r −r0) dV0 + Z V ˆF(r0, ω)Gω(r, r0) dV0, (1.83) (1.83) where we have used the fact that the Green’s function is symmetric, i.e., Gω(r, r0) = Gω(r0, r). Application of Green’s theorem to the term on the left side of Equation (1.83) and the sifting property of the Dirac delta function to the right side of Equation (1.83) leads to ˆΦ(r, ω) = Z S " Gω(r, r0)∂ˆΦ(r0, ω) ∂n0 −ˆΦ(r0, ω)∂Gω(r, r0) ∂n0 # dS0 − Z V ˆF(r0, ω)Gω(r, r0) dV0, (1.84) (1.84) where n0 is the outward-pointing normal on the surface S. Equation (1.84) is the general solution to Equation (1.79) for a bounded acoustic medium with general where n0 is the outward-pointing normal on the surface S. Equation (1.84) is the general solution to Equation (1.79) for a bounded acoustic medium with general 21 forcing term. Usefulness of Equation (1.84), however, depends on the ability to solve the integral equation, which may be nontrivial. forcing term. Usefulness of Equation (1.84), however, depends on the ability to solve the integral equation, which may be nontrivial. CHAPTER 2 Normal Mode Theory 2.1 Background The method of normal modes provides a means of solving the wave equation that does not require computation of a volume integral. This method involves solving a depth-dependent equation that results in a decomposition of the ﬁeld into a set of modes of vibration, roughly similar to that of a vibrating string. The complete acoustic ﬁeld is constructed via a weighted sum of these modes, where the weights are informed by the value of the modes at the depth of the sound source. In this section two approaches leading to the normal modes solution are presented. The ﬁrst is based on a generalized application of the method of separation of variables, the derivation of which will be discussed in detail. The second approach, which involves the computation of a wavenumber contour integral via a sum of residues, will be discussed more generally as a means of connecting the normal modes method to the ﬁeld of geometric acoustics. List of References [1] N. G. de Brujin, Asymptotic Methods in Analysis. New York: Dover Publica- tions, 1981. [2] C. A. Boyles, Acoustic Waveguides: Applications to Oceanic Science. New York: Wiley-Interscience, 1984. [3] F. M. White, Fluid Mechanics, 6th ed. New York: McGraw-Hill, 2006. [4] F. B. Jensen, W. A. Kuperman, M. B. Porter, and H. Schmidt, Computational Ocean Acoustics, 1st ed. New York: Springer-Verlag, 2000. [5] G. V. Frisk, Ocean and Seabed Acoustics: A Theory of Wave Propagation. New Jersey: Prentice-Hall, 1994. 22 2.2 Solution via Generalized Separation of Variables We begin by recalling from Equation (1.77) that the Fourier-transformed com- plex pressure, ˆp(r, ω), satisﬁes the Helmholtz wave equation: ρ∇• 1 ρ∇ˆp(r, ω) + k2(r)ˆp(r, ω) = ˆSp(ω) ˆQ(r), (2.1) (2.1) where ˆSp(ω) = iω ˙m and ˙m = −iρω4πa2 ˆD(ω). If we now restrict ourselves to the case of a cylindrical waveguide with azimuthal symmetry, Equation (2.1) takes the form: where ˆSp(ω) = iω ˙m and ˙m = −iρω4πa2 ˆD(ω). If we now restrict ourselves to the case of a cylindrical waveguide with azimuthal symmetry, Equation (2.1) takes the form: 1 r ∂ ∂r r∂ˆp ∂r + ρ(z) ∂ ∂z 1 ρ(z) ∂ˆp ∂z + k2ˆp = −iω ˙mδ(r)δ(z −zs) 2πr , (2.2) (2.2) 23 where zs corresponds to the depth of the source. Here we have expanded the ﬁrst term on the left side of Equation (2.1) and recalled the deﬁnition of the Laplacian operator in cylindrical coordinates: ∇2{·} = 1 r ∂ ∂r r ∂ ∂r{·} + 1 r2 ∂2 ∂θ2{·} + ∂2 ∂z2{·}. (2.3) (2.3) Note that our assumption of azimuthal symmetry dictates that the derivative in- volving the angular term, θ, in Equation (2.3) vanishes. To investigate the solution to Equation (2.2), we apply the method of separa- tion of variables to its associated homogeneous (unforced) equation. In doing so, we seek a solution of the form ˆp(r, z) = ϕ(r)u(z). We substitute this proposed so- lution into the homogeneous diﬀerential equation and divide by ϕ(r)u(z), leading to the expression: 1 ϕ 1 r d dr rdϕ dr + 1 u ρ(z) d dz 1 ρ(z) du dz + ω2 c2(z)u = 0. (2.4) (2.4) We see that the ﬁrst and second terms in brackets in this expression are solely functions of r and z, respectively. The only way that this equation can be satisﬁed is if both terms are equal to a separation constant that we will denote by k2 r. That is: is: 1 u ρ(z) d dz 1 ρ(z) du dz + ω2 c2(z)u = −1 ϕ 1 r d dr rdϕ dr = k2 r. 2.2 Solution via Generalized Separation of Variables (2.5) (2.5) Concerning ourselves with the depth equation in Equation (2.5), bringing the separation constant over to the left side of the equation, and multiplying by u, we arrive at an ordinary diﬀerential equation for the depth solutions, u(z): ρ(z) d dz 1 ρ(z) du(z) dz + ω2 c2(z) −k2 r u(z) = 0. (2.6) (2.6) Assuming a pressure-release surface located at z = 0 and a perfectly rigid bottom located at z = H imposes the following boundary conditions on the solutions to 24 Equation (2.6): u(0) = 0 du(z) dz z=H = 0. (2.7) (2.7) Equation (2.6) combined with the boundary conditions in Equation (2.7) deﬁne a classical Sturm-Liouville eigenvalue problem. There are an inﬁnite number of solu- tions um(z) to Equation (2.6) and we refer to these solutions as the eigenfunctions of the problem. If we represent the diﬀerential operator in Equation (2.6) by L, then for each eigenfunction um(z) there exists a corresponding eigenvalue k2 rm such that: L [um(z)] = k2 rmum(z). (2.8) (2.8) Note also that the eigenfunctions or modes, um(z), have the property of orthogo- nality, which is stated mathematically as: Note also that the eigenfunctions or modes, um(z), have the property of orthogo- nality, which is stated mathematically as: Z H 0 ρ−1(z)um(z)un(z) dz = δnm. (2.9) (2.9) Moreover, the eigenfunctions form a complete set thus allowing us to represent an arbitrary function as a sum of the eigenfunctions or normal modes. This allows us to further specify the form of our proposed solution to the original wave equation. Namely, we have: ∞ ˆp(r, z) = ∞ X m=1 ϕm(r)um(z), (2.10) (2.10) where we have expanded ˆp(r, z) in the normal mode basis um(z), and ϕm(r) are our expansion coeﬃcients, which we propose to be strictly a function of range, r. where we have expanded ˆp(r, z) in the normal mode basis um(z), and ϕm(r) are our expansion coeﬃcients, which we propose to be strictly a function of range, r. where we have expanded ˆp(r, z) in the normal mode basis um(z), and ϕm(r) are our expansion coeﬃcients, which we propose to be strictly a function of range, r. 2.2 Solution via Generalized Separation of Variables If we substitute Equation (2.10) into Equation (2.2) and apply the property in Equation (2.8) we obtain the intermediate equation: If we substitute Equation (2.10) into Equation (2.2) and apply the property in Equation (2.8) we obtain the intermediate equation: ∞ X m=1 1 r d dr rdϕm(r) dr um(z) + k2 rmϕm(r)um(z) = −iω ˙mδ(r)δ(z −zs) 2πr . (2.11) (2.11) 25 Taking into account the property in Equation (2.9), we now multiply Equation (2.11) by ρ−1(z)un(z) and integrate from 0 to H resulting in the following range- separated equation: 1 r d dr rdϕn(r) dr + k2 rnϕn(r) = −iω ˙mδ(r)un(zs) 2πrρ(zs) . (2.12) (2.12) Equation (2.12) is a standard equation known as Bessel’s equation. To solve this equation, we deﬁne the Hankel transform: Equation (2.12) is a standard equation known as Bessel’s equation. To solve this equation, we deﬁne the Hankel transform: H[f(r)] = F(κ) = Z ∞ 0 f(r)J0(κr)r dr, (2.13) (2.13) the inverse of which is the inverse of which is H−1[F(κ)] = f(r) = Z ∞ 0 F(κ)J0(κr)κ dκ. (2.14) (2.14) Next we note the following property of the transform in Equation (2.13): H 1 r d dr rdf(r) dr = −κ2F(κ). (2.15) (2.15) Applying the operator in Equation (2.13) to Equation (2.12) and letting F(κ) = H[ϕn], we arrive at [−κ2 + k2 rn]F(κ) = −iω ˙mun(zs) 2πρ(zs) Z ∞ 0 δ(r)J0(κr) dr = −iω ˙mun(zs) 2πρ(zs) . (2.16) (2.16) Solving for F(κ), we have Solving for F(κ), we have F(κ) = iω ˙mun(zs) 2πρ(zs)[κ2 −k2 rn]. (2.17) (2.17) Next we apply the inverse Hankel transform in Equation (2.14) to F(κ) in Equation (2.17) and note the identity Next we apply the inverse Hankel transform in Equation (2.14) to F(κ) in Equation (2.17) and note the identity K0(µr) = Z ∞ 0 J0(κr) κ2 + µ2κ dκ, (2.18) (2.18) where K0 is the modiﬁed Bessel function of the second kind [1]. Letting µ = ikrn in Equation (2.18) leads to an intermediate solution to Equation (2.12): ϕn(r) = iω ˙mun(zs) 2πρ(zs) K0(ikrnr). (2.19) (2.19) 26 Finally, we note that Finally, we note that K0(ζ) =      1 2πiH(1) 0 (iζ), −π < arg ζ ≤π 2 −1 2πiH(2) 0 (−iζ), −π 2 < arg ζ ≤π. 2.2 Solution via Generalized Separation of Variables (2.29) Equation (2.29) will be extended in later sections to compute forward transmission loss in a range-dependent shallow water environment. TL(r, z) ≈−20 log10 1 ρ(zs) r 2π r ∞ X m=1 um(zs)um(z)eikrmr √krm . (2.29) Equation (2.29) will be extended in later sections to compute forward transmission loss in a range-dependent shallow water environment. (2.29) 2.2 Solution via Generalized Separation of Variables (2.20) (2.20) For outgoing radiation we choose H(1) 0 leading to the following solution to the range-separated equation in Equation (2.12): ϕn(r) = −ω ˙mun(zs) 4ρ(zs) H(1) 0 (krnr). (2.21) (2.21) Substituting the right side of Equation (2.21) into Equation (2.10) we arrive at the normal modes solution to Equation(2.2) for a point source in a cylindrical waveguide with azimuthal symmetry: Substituting the right side of Equation (2.21) into Equation (2.10) we arrive at the normal modes solution to Equation(2.2) for a point source in a cylindrical waveguide with azimuthal symmetry: ˆp(r, z) = −ω ˙m 4ρ(zs) ∞ X m=1 um(zs)um(z)H(1) 0 (krnr). (2.22) (2.22) By utilizing the relationship between complex pressure and the velocity potential in Equation (1.49) an equivalent solution for the velocity potential may now be easily obtained. Applying this relationship to Equation (2.22), we arrive at the solution for the velocity potential due to a point source in a cylindrical waveguide with azimuthal symmetry: ˆΦ(r, z) = i ˙m 4ρ(zs)ρ(z) ∞ X m=1 um(zs)um(z)H(1) 0 (krnr). (2.23) (2.23) Asymptotic forms of Equations (2.22) and (2.23) may be obtained by recalling the asymptotic form of H(1) 0 (krnr) given in Equation (1.13). This leads to the following approximate solutions: ˆp(r, z) ≈− ω ˙m √ 8πrρ(zs)e−iπ/4 ∞ X m=1 um(zs)um(z)eikrmr √krm (2.24) (2.24) and ˆΦ(r, z) ≈ i ˙m √ 8πrρ(zs)ρ(z)e−iπ/4 ∞ X m=1 um(zs)um(z)eikrmr √krm . (2.25) ˆΦ(r, z) ≈ i ˙m √ 8πrρ(zs)ρ(z)e−iπ/4 ∞ X m=1 um(zs)um(z)eikrmr √krm . (2.25) (2.25) 27 Given Equations (2.24) and (2.25), we will now deﬁne a forward transmission loss equation: TL(r, z) = −20 log10 ˆp(r, z) ˆp0(r = 1) = −20 log10 ˆΦ(r, z) ˆΦ0(r = 1) , (2.26) (2.26) where where where ˆp0(r) = iω ˙meik0r 4πr (2.27) ˆp0(r) = iω ˙meik0r 4πr (2.2 (2.27) and and ˆΦ0(r) = 1 ρ(z) ˙meik0r 4πr , (2.28) ˆΦ0(r) = 1 ρ(z) ˙meik0r 4πr , (2.28) (2.28) where k0 = ω/c0 and c0 is a reference sound speed. Here ˆp0(r) and ˆΦ0(r) represent reference values to be taken at 1 m from the source. Inserting either Equation (2.24) or (2.25) with its associated reference value into Equation (2.26), leads to the following approximate expression for forward transmission loss TL(r, z) ≈−20 log10 1 ρ(zs) r 2π r ∞ X m=1 um(zs)um(z)eikrmr √krm . 2.3 Wavenumber Decomposition and Evanescent Modes An intuitive discussion of the decomposition of wave phase in the normal modes solution can be had if we consider the case of an isovelocity ﬂat-bottom waveguide with constant density. In this case, the modes have the general form: um(z) = A sin(kzz) + B cos(kzz), (2.30) (2.30) where kz = rω c 2 −k2 r (2.31) (2.31) is the vertical wavenumber and kr is the horizontal wavenumber. Recall the pressure-release surface boundary condition in Equation (2.7). This condition im- plies that B = 0 in Equation (2.30). The rigid bottom boundary condition from is the vertical wavenumber and kr is the horizontal wavenumber. Recall the pressure-release surface boundary condition in Equation (2.7). This condition im- plies that B = 0 in Equation (2.30). The rigid bottom boundary condition from 28 Equation (2.7) suggests that either the coeﬃcient A must also be zero (i.e. the trivial solution) or that Equation (2.7) suggests that either the coeﬃcient A must also be zero (i.e. the trivial solution) or that kz = kzm = m −1 2 π H , m = 1, 2, ... . (2.32) (2.32) quation (2.32) further implies that kr must assume particular values, namely: kr = krm = sω c 2 − m −1 2 π H 2 , m = 1, 2, ... , (2.33) (2.33) where krm is recognized as the eigenvalue associated with mode m and eigenfunc- tion um. Combining Equations (2.31)-(2.33), we arrive at the expression: ω c = p k2 rm + k2 zm. (2.34) (2.34) We recognize this quantity as the magnitude (norm) of the vector quantity K, also known as the total wavenumber of the complex wave phase. If we let ˆx and ˆz be the unit base vectors along the x and z axes, respectively, we can write the total wavenumber in component form. For our two-dimensional problem of interest this is: We recognize this quantity as the magnitude (norm) of the vector quantity K, also known as the total wavenumber of the complex wave phase. If we let ˆx and ˆz be the unit base vectors along the x and z axes, respectively, we can write the total wavenumber in component form. For our two-dimensional problem of interest this is: is: K = krmˆx + kzmˆz. (2.35) (2.35) Recall the asymptotic form of the Hankel function in Equation (1.13). 2.3 Wavenumber Decomposition and Evanescent Modes It is easily veriﬁable from this equation that modes with positive imaginary eigenvalues are exponentially decaying in range. This allows us to partition the modes into propagating and evanescent (exponentially decaying/non-propagating) modes. For propagating modes (i.e. real krm) the following inequality holds: m < \|K\|H π + 1 2. (2.36) (2.36) Conversely, for evanescent modes (i.e. imaginary krm): m > \|K\|H π + 1 2. (2.37) (2.37) 29 This decomposition of the total wavenumber has led to an interesting fact with regards to the modes in our isovelocity ﬂat-bottom waveguide. Namely, the real eigenvalues and therefore the propagating modes have an upper bound of \|K\| = (ω/c). As the source frequency is reduced the eigenvalues on the real axis slide to the left and up the imaginary axis. For a suﬃciently low frequency, the ﬁrst mode (i.e. m = 1) will make this transition leaving no propagating modes in the problem. The frequency at which this occurs is known as the cutoﬀfrequency for the waveguide and in appealing to the inequality in Equation (2.37), it is found to be: fc = c 4H . (2.38) fc = c 4H . (2.38) Note that the depth of the waveguide also has an eﬀect on which modes can propagate. As the waveguide becomes more shallow H in the above inequalities becomes smaller in magnitude and lower order modes are cutoﬀ. As was the case with frequency, there is a critical waveguide depth at which the ﬁrst mode (i.e. m = 1) will become cutoﬀleaving no propagating modes. Appealing again to Equation (2.37) we see that this depth is: Hc = c 4f . (2.39) (2.39) The decomposition of the total wavenumber allows for the discussion of an- other modal concept, that of mode grazing angle. We deﬁne the grazing angle of a mode as θm = arcsin kzm k = arcsin ckzm ω , (2.40) (2.40) where θm corresponds to the angle that an equivalent ray associated with mode m would make with the bottom boundary of the waveguide. For a ﬂat-bottom waveguide kzm is constant for a particular mode m, as evident from Equation (2.31). If sound speed is also constant, the only quantity that can change in the 30 argument to the arcsine function in Equation (2.40) is the angular frequency of the source, ω. 2.3 Wavenumber Decomposition and Evanescent Modes Decreasing the frequency of the source will result in an increase in the argument in Equation (2.40) thus pushing the arcsine toward its maximum real value of 90◦. As previously discussed, decreasing the source frequency also pushes the modes in a waveguide toward their cutoﬀ. This means that the closer a mode is to its cutoﬀthe steeper its grazing angle, θm, will be. Once a mode is fully cutoﬀ(i.e. its eigenvalue is purely imaginary), the real part of its grazing angle reaches 90◦and the mode is said to have “turned around.” Steep-angle modes will be of particular interest in later sections of the presented work. 2.4 Modal Velocities Another useful concept is modal velocity. To understand this concept consider a traveling wave of amplitude A and a single frequency f. Such a wave may be represented by the sinusoid: A cos(kx −ωt). (2.41) (2.41) After some small unit of time dt this wave will have experienced ωt/2π = ft oscillations. Over this same time window, an initial peak of the wave will have propagated a distance dx away from the source to make room for this same number of oscillations (i.e. k dx = ω dt). Recognizing this relationship allows us to deﬁne the following: cp = dx dt = ω k , (2.42) (2.42) where cp is referred to as the phase velocity of the wave. The phase velocity represents the horizontal velocity of a phase point of the wave. It does not represent the speed of energy transport, which must be less than or equal to the speed of sound in the medium, c. where cp is referred to as the phase velocity of the wave. The phase velocity represents the horizontal velocity of a phase point of the wave. It does not represent the speed of energy transport, which must be less than or equal to the speed of sound in the medium, c. To obtain the speed of energy transport consider the combination of two waves 31 with slightly diﬀerent frequencies and wave lengths: with slightly diﬀerent frequencies and wave lengths: cos [(k −∆k)x −(ω −∆ω)t] + cos [(k + ∆k)x −(ω + ∆ω)t] = 2 cos(∆kx −∆ωt) cos(kx −ωt) (2.43) (2.43) The amplitude of the wave in Equation (2.43) is a sinusoidal envelope with phase speed ∆ω/∆k. In the limit, we refer to this speed as the group velocity cg = dω dk , (2.44) (2.44) which is the speed at which energy is transported by a packet of waves. The ratio between the speed of sound in the medium and the phase velocity of a wave propagating in the medium is known as the refractive index, n = c/cp = ck/ω. If one takes the derivative of ω = ck/n with respect to the wavenumber, k, we obtain d k d cg = dω dk = c n − ck n2 × dn dk = cp − ck n2 × dn dk . 2.4 Modal Velocities (2.45) (2.45) As is evident from Equation (2.45), the only way for the group velocity and phase velocity to be equal is for the refractive index to be constant. In this case, the phase and group velocities are independent of frequency and the medium is said to be non-dispersive. Otherwise, the medium is said to be dispersive, meaning that a wave will separate into its individual frequency components while propagating through the medium due to the diﬀerence between the phase and group velocities. Now we would like to connect the concept of phase and group velocity to the normal modes of propagation of an acoustic waveguide. At high frequencies all modes in an acoustic waveguide approach a propagating eigenvalue, which ap- proaches the medium wavenumber [2]. That is, krm →k as ω →∞. (2.46) (2.46) 32 Taking this into account and recalling Equation (2.42), we deﬁne the phase velocity of an individual mode m as Taking this into account and recalling Equation (2.42), we deﬁne the phase velocity of an individual mode m as cpm = ω krm . (2.47) (2.47) For steep-angle modes (i.e. those near cutoﬀ), the phase velocity approaches inﬁn- ity. Conversely, as the grazing angle of a mode becomes more horizontal (i.e. more propagating) the phase velocity approaches c, the speed of sound in the waveguide [3]. For steep-angle modes (i.e. those near cutoﬀ), the phase velocity approaches inﬁn- ity. Conversely, as the grazing angle of a mode becomes more horizontal (i.e. more propagating) the phase velocity approaches c, the speed of sound in the waveguide [3]. Similar to the case for phase velocity, we can deﬁne a modal group velocity by taking into account Equations (2.46) and (2.44) cgm = dω dkrm . (2.48) (2.48) The modal group velocity is the speed of energy transport of a particular mode. As the frequency of the acoustic source approaches the cutoﬀfrequency for the mode its group velocity will approach zero. Conversely, as the frequency of the source approaches inﬁnity, the group velocities of all modes in the waveguide will approach the speed of sound. Further investigation into the eﬀects of horizontally-variable waveguide depth on modal group velocity has been documented in Appendix A. 2.5 Solution via Cauchy’s Residue Theorem The Green’s function has the form: for z ≤zs: +( ) ( ) G(z, zs, k2 r) = u+(z)u−(zs) W(u+, u−, k2 r), (2.52) (2.52) 34 for z ≥zs: for z ≥zs: for z ≥zs: ( ) +( ) for z ≥zs: u−(z)u+(z ) G(z, zs, k2 r) = u−(z)u+(zs) W(u+, u−, k2 r), (2.53) (2.53) where W(u+, u−, k2 r) is the Wronskian of the two chosen solutions to the homoge- neous equation, deﬁned as: (2.54) W(u+, u−, k2 r) = u+(zs)u′−(zs) −u−(zs)u′+(zs), (2.54) where primes denote diﬀerentiation with respect to z. Note, the Wronskian is constant for a given value of kr. Next we observe that a normal mode will exist when the upward traveling and downward traveling waves superimpose to create a standing wave within the waveguide. This condition implies linear dependence of the two solutions u+(z) and u−(z), which will occur when W(u+, u−, k2 r) = 0. In general, the sum of our particular solution to the inhomogeneous equation with the general solution of the homogeneous equation (i.e. uh = u+(z) + u−(z)) will give us the total solution to the depth-separated equation. If we then combine this solution with the solution to the range-separated equation we can express the total solution to the wave equation in Equation (2.1) as a contour integral in wavenumber space. Thus we have: where primes denote diﬀerentiation with respect to z. Note, the Wronskian is constant for a given value of kr. Next we observe that a normal mode will exist when the upward traveling and downward traveling waves superimpose to create a standing wave within the waveguide. This condition implies linear dependence of the two solutions u+(z) and u−(z), which will occur when W(u+, u−, k2 r) = 0. In general, the sum of our particular solution to the inhomogeneous equation with the general solution of the homogeneous equation (i.e. uh = u+(z) + u−(z)) will give us the total solution to the depth-separated equation. If we then combine this solution with the solution to the range-separated equation we can express the total solution to the wave equation in Equation (2.1) as a contour integral in wavenumber space. Thus we have: ˆp(r, z) = I H(1) 0 (krr)G(z, zs, k2 r)kr dkr. 2.5 Solution via Cauchy’s Residue Theorem For a point source in a medium with well-deﬁned upper and lower boundaries that reﬂect the energy associated with an acoustic wave, we can deﬁne two solu- tions, u+(z) and u−(z), of the homogeneous depth-separated equation in Equation (2.6). Here u+(z) accounts for the sum of the upper traveling wave plus the wave reﬂected from the upper boundary. Conversely, u−(z) accounts for the lower trav- eling wave plus the wave reﬂected from the lower boundary. Mathematically these solutions are given by the expressions u+(z) = ei R z1 z kz dz + R1e−i R z1 z kz dz (2.49) (2.49) 33 and u−(z) = ei R z2 z kz dz + R2e−i R z2 z kz dz, (2.50) (2.50) where z2 > zs > z1 and zs is the depth of the source. The vertical wavenumber, kz, in these solutions is given by the expression in Equation (2.31), and R1 and R2 are reﬂection coeﬃcients at the upper and lower boundaries of the medium, respectively. The solutions in Equations (2.49) and (2.50) are known as Wentzel-Kramers- Brillouin, or WKB, solutions to the depth-separated wave equation. An alternative expression for the WKB solution is u(z) = 1 √η sin θe±ik0 R η sin θ dz, (2.51) (2.51) where c0 = ω/k0 is a reference sound speed at the source location, η = c0/c is the diﬀraction coeﬃcient due to depth-variable sound speed, and θ is the grazing angle of a ray associated with the solution u(z) to the depth-separated equation. When written in this form we see that the solution blows up when θ = 0. This is be- cause when θ is zero the equivalent ray associated with the solution u(z) becomes horizontal and consequently turns over. As such, we say that the WKB approxi- mation breaks down at turning points. Further details on the WKB solution and the discipline of geometric acoustics can be found in reference [2]. Recalling the method of variation of parameters, we can derive a particular solution to the inhomogeneous equation corresponding to Equation (2.6). This solution is of the form of an integral product of a piece-wise deﬁned Green’s function for our medium of interest and the forcing term on the right side of the depth- separated equation. 2.6 Range-Dependent Normal Modes (2.2) The introduction of range-dependence into the conditions of Equation (2.2) signiﬁcantly increases the diﬃculty of ﬁnding a solution for the acoustic ﬁeld. This is primarily due to the fact that the modes of vibration of the wave equation become a function of depth and range. This means that our assumed solution in Equation (2.10) takes the form ∞ ˆp(r, z) = ∞ X m=1 ϕm(r)um(r, z). (2.56) (2.56) Inserting this proposed solution into Equation (2.2) and following a procedure similar to that for the canonical normal modes derivation leads to the following set of coupled equations for the expansion coeﬃcients ϕm(r): 1 r d dr rdϕn dr + ∞ X m=1 2Bmn dϕm dr + ∞ X m=1 Amnϕm + k2 rn(r)ϕn = −iω ˙mδ(r)un(r, zs) ρ(zs)2πr , (2.57) (2.57) where Amn = Z 1 r ∂ ∂r r∂um ∂r un ρ dz (2.58) Amn = Z 1 r ∂ ∂r r∂um ∂r un ρ dz (2.58) (2.58) and Bmn = Z ∂um ∂r un ρ dz. (2.59) (2.59) The expressions for the coeﬃcients in Equations (2.58) and (2.59) involve modes of mixed index. This is due to the fact that in range-dependent environments energy is transferred between adjacent modes. This non-adiabatic energy transference, known as mode coupling, prevents the use of a classical normal modes approach to solving for the acoustic ﬁeld. The expressions for the coeﬃcients in Equations (2.58) and (2.59) involve modes of mixed index. This is due to the fact that in range-dependent environments energy is transferred between adjacent modes. This non-adiabatic energy transference, known as mode coupling, prevents the use of a classical normal modes approach to solving for the acoustic ﬁeld. 2.5 Solution via Cauchy’s Residue Theorem (2.55) (2.55) The previous statement regarding the conditions for the existence of normal modes implies that values of kr corresponding to normal modes represent singularities (poles) in the integrand of Equation (2.55). To evaluate this integral we apply Cauchy’s Residue Theorem [4] to express its solution as a sum of the individual residues corresponding to each normal mode. Further details on this normal modes approach can be found in reference [3]. 35 2.7 Previous Work A number of approaches have been developed for treating the range-dependent problem described in Section (2.6). One of the ﬁrst solutions to this problem was described by Pierce [5] in 1964. Known as the adiabatic assumption in mode cou- 36 pling theory, this approach assumes a nearly stratiﬁed acoustic medium such that coupling between modes may be neglected. The justiﬁcation for this simpliﬁcation is that, for suﬃciently slowly varying media, the terms involving the coupling co- eﬃcients Amn and Bmn in Equation (2.57) are small and may be neglected. Milder [6] presented a similar argument in 1969 based on the proposition that the equiv- alent rays associated with the normal modes of acoustic propagation are invariant in acoustic media with mild range-dependence. While this adiabatic approach to range-dependent acoustic propagation has been justiﬁed in a number of cases, there are still many other propagation scenarios for which the neglect of non-adiabatic mode coupling leads to signiﬁcant errors in the estimation of the acoustic ﬁeld. For example, Godin [7] showed that ignoring coupling between modes leads to a violation of conservation of energy and reciprocity principles in the presence of a sloping interface or a range-dependent rigid boundary. Most approaches to the range-dependent problem that account for mode cou- pling fall into one of three classes. In the ﬁrst class of solutions one attempts to directly solve the coupled equation given in Equation (2.57). Among this family of solutions is the approach described by Buckingham [8] in 1987. In this approach, the author transforms the acoustic wave equation into a cylindrical coordinate sys- tem where the z axis is horizontal, termed the “wedge coordinate system.” For the case of an isovelocity waveguide, it is then possible to derive a set of uncoupled “wedge modes,” which may be summed to approximate the acoustic ﬁeld. In 1991 Primack and Gilbert [9] extended the work of Buckingham by developing a compu- tational model based on this wedge modes approach. They showed that for realistic sound-speed proﬁles coupling between wedge modes must be included in the calcu- lation of the ﬁeld. Fawcett, Westwood, and Tindle [10] further extended the work of Primack and Gilbert by including leaky modes into the ﬁeld calculation and de- 37 veloping a least-squares approximation for computing mode coupling in the wedge coordinate system. 2.7 Previous Work While this wedge mode approach to range-dependent acoustic propagation shows promise for single-slope propagation scenarios, its applicability to general range-dependent media requires further development. Also belonging to this ﬁrst class of solutions to the range-dependent prob- lem is the work of Desaubies, Chiu, and Miller [11]. In their 1986 study these authors set out to quantify when it is appropriate to make the adiabatic approx- imation in the context of slow changes to the sound-speed structure of the ocean due to mesoscale eddies. In doing so, they present a simpliﬁcation to the coupled system given by Equations (2.57)-(2.59) based on an adiabatic expansion of the range-dependent wavenumber. It was ultimately determined in this study that the adiabatic solution fails for environments with range-dependent sound speed under the conditions of high frequency, high mode number, and long range. Note, the approach of this study assumes a ﬂat-bottom waveguide and therefore is limited to environments where sound speed is the dominant mechanism for range dependence. Chiu, Miller, and Lynch [12] also set out to solve the range-dependent problem via directly solving Equation (2.57) in 1996. In their approach a fourth-order ﬁnite dif- ference method is utilized for approximating the ﬁrst and second depth derivatives of the eigenfunctions associated with the modes of acoustic propagation. Fourth and ﬁfth-order Runge-Kutta algorithms are used to solve a coupled system of equa- tions governing complex envelopes of the mode amplitudes. The pressure ﬁeld is ultimately resolved via an inverse Fourier transform of the product of the source signal spectrum and a source-to-receiver ocean transfer function. Computational savings are achieved in this approach through computing the gradually varying complex envelopes of the modal amplitudes rather than the rapidly varying mode amplitudes themselves. Conversely, computational bottlenecks are found in the 38 method’s need to numerically solve the integrals associated with the coupling co- eﬃcients and in the iterative approach (i.e. Runge-Kutta) required to solve the coupled diﬀerential equation governing the modal envelopes. Other direct solu- tions to the range-dependent wave equation can be found in works by Boyles [13], Gillette [14], and Stotts [15]. The second class of solutions to the problem described in Section (2.6) includes approximate methods whereby assumptions are made that lead to simpliﬁcations in the range-dependent acoustic wave equation. 2.7 Previous Work These methods are often more computationally eﬃcient due to their simplicity; however, this gain in eﬃciency is often at the cost of accuracy in environments with severe range-dependence. The family of methods known as parabolic-equation methods belong to this class of approximate solutions. The parabolic-equation method was ﬁrst introduced into underwater acoustics in the early 1970s by Hardin and Tappert [16]. In these solutions, one begins with a Helmholtz equation of the form ∂2p ∂r2 + 1 r ∂p ∂r + ∂2p ∂z2 + k2 0n2p = 0, (2.60) (2.60) where n(r, z) = c0/c(r, z) is the index of refraction and c0 is a reference sound speed. A solution of the form p(r, z) = ψ(r, z)H(1) 0 (k0r) is assumed and inserted into Equation (2.60). Next, it is assumed that k0r ≫1 (i.e. the farﬁeld assumption), resulting in the diﬀerential equation: ∂2ψ ∂r2 + 2ik0 ∂ψ ∂r + ∂2ψ ∂z2 + k2 0(n2 −1)ψ = 0. (2.61) (2.61) From here, the following paraxial approximation (also known as the small angle approximation) is made: ∂2ψ ∂r2 ≪2ik0 ∂ψ ∂r . (2.62) (2.62) This ultimately leads to what is known as the parabolic equation of underwater This ultimately leads to what is known as the parabolic equation of underwater 39 acoustics: acoustics: 2ik0 ∂ψ ∂r + ∂2ψ ∂z2 + k2 0(n2 −1)ψ = 0. (2.63) (2.63) Methods in this family of solutions are then focused upon solving Equation (2.63). While the parabolic equation given in Equation (2.63) is signiﬁcantly easier to solve than the quadratic acoustic wave equation, this approach comes with some caveats. In addition to the farﬁeld assumption, the paraxial approximation in Equation (2.62) puts additional constraints on the propagation scenarios that may be considered. In particular, parabolic-equation methods require that back scatter- ing be negligible and that range-dependence in the acoustic medium be relatively weak. These methods also constrain the propagation scenario to angles less than or equal to about ten to ﬁfteen degrees oﬀthe horizontal and to high source fre- quencies. Methods in this family of solutions are then focused upon solving Equation (2.63). While the parabolic equation given in Equation (2.63) is signiﬁcantly easier to solve than the quadratic acoustic wave equation, this approach comes with some caveats. In addition to the farﬁeld assumption, the paraxial approximation in Equation (2.62) puts additional constraints on the propagation scenarios that may be considered. 2.7 Previous Work In particular, parabolic-equation methods require that back scatter- ing be negligible and that range-dependence in the acoustic medium be relatively weak. These methods also constrain the propagation scenario to angles less than or equal to about ten to ﬁfteen degrees oﬀthe horizontal and to high source fre- quencies. Tindle, O’Driscoll, and Higham [17] developed a method in 2000 that also be- longs to the class of approximate solutions to the range-dependent wave equation. In their approach the eﬀects of mode coupling due to range-dependent sound speed are accounted for through a novel application of perturbation theory. Mode cou- pling is represented via small perturbative terms that are added to the eigenfunc- tions and eigenvalues associated with the modes. These perturbed eigenfunctions and eigenvalues are then used in the canonical (i.e. range-independent) normal modes solution to approximate the acoustic ﬁeld. Tindle and Higham later ex- tended this approach to account for coupling due to range-dependent density [18] and internal waves [19]. Kampanis and Dougalis [20] developed an approximate method for solving the range-dependent wave equation based on the Galerkin ﬁ- nite element method in 1999. In 2005 Clark [21] extended a multipath expansion method for solving the Helmholtz wave equation, originally developed by Leibiger [22], to account for horizontal variations in bottom depth, bottom type, and sound 40 speed via a stationary phase approximation. A deep water variation of this method that maintains continuity through mode turning points was published by Clark and Smith in 2008 [23]. In the third class of solutions to the problem of range-dependent acoustic propagation the horizontally-variable acoustic environment is approximated via a series of range-independent steps. Within each of these steps a canonical normal modes solution such as the one described in Sections (2.2) and (2.5) may be ap- plied. The non-adiabatic transfer of energy between modes is then accounted for by applying continuity conditions at discrete interfaces separating the steps. This approach ﬁrst appears in a dissertation by Leibiger [22] in 1968. In this work, Leibiger develops an asymptotic theory for estimating transmission and reﬂection coeﬃcients at discrete interfaces throughout the range-dependent medium by writ- ing the eigenfunctions and eigenvalues as an expansion in range. This theory is developed in the context of near-surface propagation in an inﬁnitely deep ocean and therefore coupling eﬀects due to a sloping seaﬂoor are neglected. 2.7 Previous Work Moreover, the transmission and reﬂection coeﬃcient solutions developed in this work are lim- ited by the assumption that the wavefronts of the acoustic excitation are planar (i.e. the farﬁeld assumption). Evans [24] developed a more general stepwise cou- pled mode approach in 1983. This solution was implemented into a computational model known as COUPLE and is considered a benchmark solution for problems in range-dependent acoustic propagation. Despite its ability to provide an exact solution, the computational intensity required by the COUPLE solution makes it unsuitable for real-time or near real-time applications. Instead, it often serves as a benchmark for less computationally intensive approximate solutions [3]. 41 41 List of References List of References [1] G. N. Watson, A Treatise on the Theory of Bessel Functions, 2nd ed. New York: Cambridge University Press, 1995. [2] C. A. Boyles, Acoustic Waveguides: Applications to Oceanic Science. New York: Wiley-Interscience, 1984. [3] F. B. Jensen, W. A. Kuperman, M. B. Porter, and H. Schmidt, Computational Ocean Acoustics, 1st ed. New York: Springer-Verlag, 2000. [4] J. W. Brown and R. V. Churchill, Complex Variables and Applications, 7th ed. New York: McGraw-Hill, 2003. [5] A. D. Pierce, “Extension of the method of normal modes to sound propagation in an almost-stratiﬁed medium,” Journal of the Acoustical Society of America, vol. 37, no. 1, pp. 19–27, 1964. [6] D. M. Milder, “Ray and wave invariants for sofar channel propagation,” Jour- nal of the Acoustical Society of America, vol. 46, no. 5, pp. 1259–1263, 1969. [7] O. A. Godin, “A note on diﬀerential equations of coupled-mode propagation in ﬂuids,” Journal of the Acoustical Society of America, vol. 103, no. 1, pp. 159–168, 1998. [8] M. J. Buckingham, “Theory of three-dimensional acoustic propagation in a wedgelike ocean with a penetrable bottom,” Journal of the Acoustical Society of America, vol. 82, no. 1, pp. 198–210, 1987. [9] H. Primack and K. E. Gilbert, “A two-dimensional downslope propagation model based on coupled wedge modes,” Journal of the Acoustical Society of America, vol. 90, no. 6, pp. 3254–3262, 1991. [10] J. A. Fawcett, E. K. Westwood, and C. T. Tindle, “A simple coupled wedge mode propagation method,” Journal of the Acoustical Society of America, vol. 98, no. 3, pp. 1673–1681, 1995. [11] Y. Desaubies, C.-S. Chiu, and J. H. Miller, “Acoustic mode propagation in a range-dependent ocean,” Journal of the Acoustical Society of America, vol. 80, no. 4, pp. 1148–1160, 1986. [12] C.-S. Chiu, J. H. Miller, and J. F. Lynch, “Forward coupled-mode propagation modeling for coastal acoustic tomography,” Journal of the Acoustical Society of America, vol. 99, no. 2, pp. 793–802, 1996. [13] C. A. Boyles, “Coupled mode solution for a cylindrically symmetric oceanic waveguide with a range and depth dependent refractive index and a time vary- ing rough sea surface,” Journal of the Acoustical Society of America, vol. 73, no. 3, pp. 800–805, 1983. 42 [14] G. Gillette, “Coupled modes in a waveguide with a range-dependent rigid basement,” Journal of the Acoustical Society of America, vol. 95, no. 1, pp. 187–200, 1994. [15] S. A. CHAPTER 3 Asymptotic Stepwise Coupled Modes List of References Stotts, “Coupled-mode solutions in generalized ocean environments,” Journal of the Acoustical Society of America, vol. 111, no. 4, pp. 1623–1643, 2002. [16] R. H. Hardin and F. D. Tappert, “Applications of the split-step fourier method to the numerical solution of nonlinear and variable coeﬃcient wave equations,” SIAM Rev., vol. 15, p. 423, 1973. [17] C. T. Tindle, L. M. O’Driscoll, and C. J. Higham, “Coupled mode perturba- tion theory of range dependence,” Journal of the Acoustical Society of Amer- ica, vol. 108, no. 1, pp. 76–83, 2000. [18] C. J. Higham and C. T. Tindle, “Coupled perturbed modes over a sloping penetrable bottom,” Journal of the Acoustical Society of America, vol. 114, no. 6, pp. 3119–3124, 2003. [19] C. J. Higham and C. T. Tindle, “Coupled perturbed modes and internal solitary waves,” Journal of the Acoustical Society of America, vol. 113, no. 5, pp. 2515–2522, 2003. [20] N. A. Kampanis and V. A. Dougalis, “A ﬁnite element code for the numeri- cal solution of the Helmholtz equation in axially symmetric waveguides with interfaces,” Journal of Computational Acoustics, vol. 7, no. 2, pp. 83–110, 1999. [21] C. A. Clark, “Acoustic wave propagation in horizontally variable media,” IEEE Journal of Oceanic Engineering, vol. 30, no. 1, pp. 188–197, 2005. [22] G. A. Leibiger, “Wave propagation in an inhomogeneous medium with slow spatial variation,” Ph.D. dissertation, Stevens Institute of Technology, Castle Point, Hoboken, New Jersey, 1968. [23] C. A. Clark and K. B. Smith, “An eﬃcient normal mode solution to wave propagation prediction,” IEEE Journal of Oceanic Engineering, vol. 33, no. 4, pp. 462–476, 2008. [24] R. B. Evans, “A coupled mode solution for acoustic propagation in a waveg- uide with stepwise depth variations of a penetrable bottom,” Journal of the Acoustical Society of America, vol. 74, no. 1, pp. 188–195, 1983. 43 3.1 Introduction The remaining goal of this dissertation is to develop a non-adiabatic mode coupling solution suitable for implementation in a computational model for the prediction of acoustic energy propagation in range-dependent shallow water envi- ronments. Particular attention will be given to developing a solution that main- tains analytical integrity while oﬀering an increase in computational eﬃciency over currently existing solutions. This will be accomplished through a novel two-layer discretization of the environment, which leads to closed-form expressions for com- puting non-adiabatic mode coupling at discrete interfaces throughout the acoustic medium. This approach depends upon writing the range-dependent eigenfunctions and eigenvalues as an expansion in range. Doing so leads to the development of a range-expanded normal mode inner product for estimating the transfer of energy between modes upon encountering a horizontal change in environment. Further eﬃciency in the presented solution is gained through the extension and applica- tion of an asymptotic coupled mode theory originally developed for near-surface propagation in an inﬁnitely deep ocean, which is documented in reference [1]. We begin by assuming an environment in which the horizontal changes in the acoustic properties of the medium are small over a distance equal to the wavelength of the source frequency. A medium possessing this quality will be termed slowly varying. This assumption will be coupled with the assumption that the acoustic velocity potential may be expanded locally as a sum of normal modes, i.e.,: ˆΦ(x, y, z) = ∞ X n=1 ϕn(x, y)un(z; x, y), (3.1) (3.1) 44 where the un(z; x, y) are the solutions of the Sturm-Liouville problem d2u dz2 + ω2 c2(z; x, y) −k2 r u = 0, (3.2) (3.2) with boundary conditions with boundary conditions u(0) = 0 du(z) dz z=H = 0. (3.3) (3.3) Here, the range coordinates (x, y) are ﬁxed and the depth coordinate, z, is mea- sured positive-down. The form of the proposed solution in Equation (3.1) suggests that the propa- gation in a slowly varying medium is by modes, although the horizontally variable character of the medium serves to perturb the eigenvalues, kn, and the eigenfunc- tions, un(z; x, y), deﬁning these modes. Intuitive justiﬁcation for this approach is found in the fact that even though the individual modes experience this pertur- bation, the slowly variable nature of the medium should not have strong eﬀects on the energy transport through the medium. 3.1 Introduction This implies that such transport is essentially the same in character as that of the canonical normal modes problem. 3.2 The Range-Expanded Normal Mode Inner Product For simpliﬁcation of the following discussion we will turn our attention to the two-dimensional problem. In doing so, the acoustic ﬁeld will be considered a function of the spatial coordinates x and z, where these coordinates represent range and depth, respectively. The property of the slowly varying range-dependent medium to perturb the eigenvalues and eigenfunctions is then introduced into the analysis by assuming the validity, at least to second order, of the following representation: kn(x + ∆x) = kn(x) + k′ n(x)∆x + 1 2k′′ n(x)(∆x)2 + ... (3.4) un(z, x + ∆x) = un(z, x) + u′ n(z, x)∆x + 1 2u′′ n(z, x)(∆x)2 + ..., (3.5) (3.4) (3.5) 45 45 where primes denote diﬀerentiation with respect to x. Thus Equations (3.4) and (3.5) provide expressions for the perturbed eigenvalues and eigenfunctions, respec- tively, upon taking a single range step of size ∆x. It is next noted that due to the orthonormality of the original eigenfunctions, un(z, x), the perturbed eigenfunc- tions, un(z, x + ∆x), may be expressed as: un(z, x + ∆x) = N X m=1 ⟨un(z, x + ∆x), um(z, x)⟩um(z, x), (3.6) un(z, x + ∆x) = N X m=1 ⟨un(z, x + ∆x), um(z, x)⟩um(z, x), n = 1, 2, ..., N (3.6) (3.6) n = 1, 2, ..., N n = 1, 2, ..., N where ⟨f(z, x), g(z, x)⟩denotes the inner product where ⟨f(z, x), g(z, x)⟩denotes the inner product ⟨f(z, x), g(z, x)⟩= Z H(x) 0 f(z, x)g(z, x) dz, (3.7) (3.7) and H(x) is the bottom depth from the surface at range x, which may be either ﬁnite or inﬁnite. In proceeding from range x to range x + ∆x, this transformation imposes a rotation of the basis vectors of the appropriate function space, or domain of the diﬀerential operator from Equation (3.2). The inner product in Equation (3.6) then describes the amount of energy transferred from each of the modes m to mode n upon taking this step in range. Geometrically, this inner product can be interpreted as the projection of mode n at range x+∆x onto each of the modes m at range x. 3.2 The Range-Expanded Normal Mode Inner Product Substituting Equation (3.5) into the inner product in Equation (3.6), we arrive at the following range-expanded normal mode inner product describing the coupling between modes m and n: ⟨un(z, x + ∆x),um(z, x)⟩ ⟨un(z, x + ∆x),um(z, x)⟩ ≈⟨un(z, x), um(z, x)⟩+ ⟨u′ n(z, x), um(z, x)⟩∆x + 1 2⟨u′′ n(z, x), um(z, x)⟩(∆x)2 + ... . (3.8) (3.8) There are three important properties of the above range-expanded normal mode inner product that will be utilized in further derivation of the presented 46 theory. The ﬁrst two of these properties relate the value of the inner product for coupling due to forward propagating acoustic energy to that for backwards propa- gation. We will derive these properties by ﬁrst recalling the deﬁnition of the Taylor series for a real or complex-valued function f(t) that is inﬁnitely diﬀerentiable at a real or complex number, a: f(t) ≈f(a) + f ′(a) 1! (t −a) + f ′′(a) 2! (t −a)2 + f ′′′(a) 3! (t −a)3 + ... . (3.9) (3.9) For the case of forward coupling, let t = (x+∆x) and a = x in Equation (3.9) and let the function f be the eigenfunctions, un, associated with the modes of vibration of the acoustic medium. In doing so, we see that un(z, x + ∆x) is equal to the expression given in Equation (3.5), which when inserted into the inner product in Equation (3.6) leads to the expression in Equation (3.8) for the forward range-expanded normal mode inner product. Following a similar procedure for the case of backwards coupling, we let t = (x −∆x) and a = x in Equation (3.9) and again let f correspond to the eigenfunctions, un. In doing so, we arrive at an expression for the backwards range-expanded normal mode inner product: ⟨un(z, x −∆x),um(z, x)⟩ ⟨un(z, x −∆x),um(z, x)⟩ (3.10) ≈⟨un(z, x), um(z, x)⟩−⟨u′ n(z, x), um(z, x)⟩∆x + 1 2⟨u′′ n(z, x), um(z, x)⟩(−∆x)2 + ... . (3.10) If we now restrict ourselves to the case of m ̸= n, we see that the ﬁrst term in both Equations (3.8) and (3.10) cancel due to the orthogonality of the eigenfunc- tions. We will further specify that ∆x is small and therefore second and higher order terms in these expansions are negligible. 3.2 The Range-Expanded Normal Mode Inner Product Under these assumptions we see that ⟨un(z, x + ∆x), um(z, x)⟩≈⟨u′ n(z, x), um(z, x)⟩∆x m ̸= n (3.11) ⟨un(z, x + ∆x), um(z, x)⟩≈⟨u′ n(z, x), um(z, x)⟩∆x ̸ (3.11) ⟨un(z, x + ∆x), um(z, x)⟩≈⟨u′ n(z, x), um(z, x)⟩∆x (3.11) m ̸= n 47 47 and ⟨un(z, x −∆x), um(z, x)⟩≈−⟨u′ n(z, x), um(z, x)⟩∆x. (3 12) (3.12) m ̸= n m ̸= n Taking into account Equations (3.11) and (3.12), we ﬁnally arrive at the ﬁrst desired property: Taking into account Equations (3.11) and (3.12), we ﬁnally arrive at the ﬁrst desired property: ⟨un(z, x + ∆x), um(z, x)⟩≈−⟨un(z, x −∆x), um(z, x)⟩. m ̸= n (3.13) (3.13) m ̸= n m ̸= n m ̸= n This property states that for modes of mixed index (i.e. cross coupling) the forward and backward inner product diﬀer by a sign. A similar procedure can be applied to the case m = n, leading to the second desired property between the forward and backward inner products: (3.14) ⟨un(z, x + ∆x), un(z, x)⟩≈−⟨un(z, x −∆x), un(z, x)⟩+ 2. (3.14) In deriving the third property of interest we must ﬁrst recall the Leibniz Integral Rule, which provides the value of the derivative of an integral whose limits of integration are a function of the variable of diﬀerentiation. The formal statement of this rule says that if a function f(z, x) exists such that its partial derivative with respect to the variable x exists and is continuous then, ∂ ∂x Z b(x) a(x) f(z, x) dz ! = f(b(x), x) • b′(x) −f(a(x), x) • a′(x) + Z b(x) a(x) ∂ ∂xf(z, x) dz. (3.15) (3.15) In applying this rule to acoustic propagation via modes in a range-dependent waveguide we note that a(x) = 0 and b(x) = H(x), where we recall that H(x) is the range-dependent bottom depth function. Next, we let the function in Equation (3.15) be u2 n(z, x), such that the integral on the left side of this equation is the inner product ⟨un, un⟩. In doing so, we recognize that this integral evaluates to unity due to orthogonality of the eigenfunctions and therefore the entire term vanishes. 3.2 The Range-Expanded Normal Mode Inner Product 48 Applying the product rule for diﬀerentiation to the integrand on the right side of Equation (3.15) and bringing the result over to the left side leads to the desired property: ⟨u′ n(z, x), un(z, x)⟩= −1 2u2 n(H(x), x) · H(x). (3.16) (3.16) Note that in reference [1] the left side of Equation (3.16) is stated to be equal to zero. This is due to the fact that variation in bottom depth with range can be neglected when considering near-surface propagation in a deep ocean waveguide. The eﬀect of this assumption is most easily illustrated if we consider the simpliﬁed case of an isovelocity acoustic wedge with pressure-release surface and rigid bottom. For this case, the range-dependent eigenfunctions can be written as un(z, x) = s 2ρ H(x) sin n −1 2 π z H(x) , (3.17) (3.17) where density, ρ, is constant. The corresponding range-dependent eigenvalues in this environment are given by the expression kn(x) = sω c 2 − n −1 2 2 π H(x) 2 . (3.18) (3.18) A closed-form expression can also be written for the range-dependent bottom depth function in the wedge waveguide. Namely, A closed-form expression can also be written for the range-dependent bottom depth function in the wedge waveguide. Namely, H(x) = H0 + x tan β, (3.19) (3.19) where H0 is the initial bottom depth at the source and β is the bottom slope angle. Inserting Equation (3.17) and (3.19) into right side of Equation (3.16) leads to ⟨u′ n(z, x), un(z, x)⟩= − 1 H(x) tan β. (3.20) (3.20) From this form of the property we can see that for a ﬂat bottom (i.e. β = 0) the right side of the equation vanishes. The implication of this property for small values of ∆x is that the inner product in Equation (3.8) is approximately unity From this form of the property we can see that for a ﬂat bottom (i.e. β = 0) the right side of the equation vanishes. The implication of this property for small values of ∆x is that the inner product in Equation (3.8) is approximately unity 49 for the case m = n (i.e. direct coupling). This is due to the fact that all terms on the right side of Equation (3.8) except for the ﬁrst are approximately zero. If, however, the bottom depth is range-dependent (i.e. 3.2 The Range-Expanded Normal Mode Inner Product β ̸= 0) the term in Equation (3.16) must be included when evaluating the inner product for direct coupling. It is important to note that the range-expanded normal mode inner product in Equation (3.8) describes the coupling between modes due to horizontal changes in the environment encountered when taking a single step in range. This inner product cannot be used to compute modes at successive range steps via Equation (3.6) as the condition of orthogonality is lost after successive steps in range due to estimation error accumulating in the approximated modes. Moreover, use of the inner product in Equation (3.8) to estimate mode coupling at range x+∆x requires prior knowledge of the amplitudes and shapes of the orthonormal eigenfunctions corresponding to the modes at range x. Taking this into account, a method is desired for resolving the mode amplitudes and shapes, which implements the inner product in Equation (3.8) to account for non-adiabatic mode coupling at discrete steps through the medium. Furthermore, it is desirable that this method maintains analytical rigor and computational eﬃciency so as to remain compatible with real- time or near real-time applications. In the next section a discretization technique for the range-dependent medium is developed that provides a way forward. 3.3 Discretization of the Range-Dependent Medium As previously stated, implementation of the inner product in Equation (3.8) for estimating non-adiabatic mode coupling requires prior knowledge of the am- plitudes and shapes of the normal modes associated with the range-dependent acoustic medium. Once the modes have been resolved, we then need a means for implementing the inner product to account for the energy transport down range as a set of acoustic waves propagates through the medium. To address the resolu- 50 tion of the modes a novel discretization approach is proposed whereby the general range-dependent environment is approximated by a series of connected simpler range-dependent subenvironments. These subenvironments are chosen such that their horizontal variability is simple enough to be captured within closed-form ex- pressions for the eigenfunctions and eigenvalues corresponding to the modes that exist within each environment. As an example, consider a hypothetical environment with constant sound speed and general range-dependent bottom depth, as depicted in Figure (3.1a). For simplicity of discussion we will further specify that this environment have a pressure-release surface and rigid bottom. Applying the proposed discretization technique, this environment may be approximated by a series of acoustic wedges as depicted in Figure(3.1). As discussed in Section (3.2), the range-dependent eigen- functions for the isovelocity acoustic wedge with pressure-release surface and rigid bottom are given by the expression in Equation (3.17) and the range-dependent eigenvalues are given by Equation (3.18). Use of these expressions allows for the range-expanded normal mode inner product in Equation (3.8) to be computed in closed form for each of the acoustic wedges used to approximate the general en- vironment. The terms of the range-expanded normal mode inner product for the acoustic wedge are given by the following closed-form expressions: For m = n: ⟨un(z, x), un(z, x)⟩= 1 (3.21) ⟨u′ n(z, x), un(z, x)⟩= −1 H tan β (3.22) ⟨u′′ n(z, x), un(z, x)⟩= 7 4 −1 3 n −1 2 2 π2 ! H−2 tan2 β, (3.23) For m = n: (3.21) H ⟨u′′ n(z, x), un(z, x)⟩= 7 4 −1 3 n −1 2 2 π2 ! 3.3 Discretization of the Range-Dependent Medium H−2 tan2 β, (3.23 For m ̸= n: For m ̸= n: 51 ⟨un(z, x), um(z, x)⟩= 0 (3.24) ⟨u′ n(z, x), um(z, x)⟩= (−1)(m+n) (2n −1)2 2H(m −n)(m + n −1) tan β (3.25) ⟨u′′ n(z, x), um(z, x)⟩= (−1)(m+n−1)(2n −1)2 2H2 (3.26) × 3 (m −n)(m + n −1) + (m + n −1)2 + (m −n)2 (m −n)2(m + n −1)2 × tan2 β. (3.24) Additionally, each of the subenvironments used in this discretization are fully speciﬁed by the parameter β, the bottom slope angle, once the initial bottom depth at the acoustic source is known. This is evident through observance of the expres- sion for the bottom depth function, H(x), in Equation (3.19). Application of this discretization approach to more general environments necessitates thought as one must choose the subenvironments such that all mechanisms for range-dependence are captured in closed-form expressions for the eigenfunctions and eigenvalues cor- responding to the modes of each environment. Once this is accomplished, however, signiﬁcant computational gains are achieved as mode coupling is calculated via closed-form algebraic expressions. This diﬀers from direct approaches to solving the range-dependent wave equation, which require numerically solving the inte- grals in Equations (2.58) and (2.59) to obtain coupling coeﬃcients. An additional beneﬁt of this approach is the elimination of the need to resolve the eigenval- ues numerically, a computationally expensive problem that exists in many other solutions. We now desire a method for incorporating the closed-form solutions for the range-expanded normal mode inner product obtained from this discretization ap- proach into a calculation of energy transport through the range-dependent envi- ronment due to an acoustic excitation. To accomplish this we let any one of the subenvironments previously discussed (e.g. the wedge subenvironments in Fig- 52 Range c(z) = c ρ(z) = ρ H(x) Range x z Depth (a) c(z) = c ρ(z) = ρ Hδ(x) Range x z Depth β1 β2 β3 β4 β5 β6 β7 β8 β9 δ = 1; 2; :::; 9 (b) Figure 3.1: (a) Environment with general range-dependent bottom depth (b) Ap- proximation of the environment in (a) via a series of connected wedge subenvi- ronments. Each subenvironment is deﬁned by the parameter βδ, its bottom slope angle. Eigenfunctions and eigenvalues for each subenvironment are given in closed- form by Equations (3.17) and (3.18). 3.3 Discretization of the Range-Dependent Medium (b) Figure 3.1: (a) Environment with general range-dependent bottom depth (b) Ap- proximation of the environment in (a) via a series of connected wedge subenvi- ronments. Each subenvironment is deﬁned by the parameter βδ, its bottom slope angle. Eigenfunctions and eigenvalues for each subenvironment are given in closed- form by Equations (3.17) and (3.18). 53 ure(3.1)) occupy the range interval [α1, α2]. We then subdivide this interval into a number of subintervals deﬁned by the boundary points x = xν, ν = 0, 1, 2, ..., τ and consider a general range-dependent function fn(x) that describes the environ- ment and is deﬁned on this range. Next, we shall approximate the function fn(x) by a series of step functions fnν = fn(xν) as depicted in Figure (3.2). The range subintervals corresponding to constant values of the function fn(x) will henceforth be referred to as regions. Note that it is always possible to relate the value of fn ... ... ... ... fn1 fn;ν−1 fn;τ−1 x1 = α1 x2 xν−1 xν xτ = α2 xτ−1 Figure 3.2: Stepwise discretization of subenvironment Figure 3.2: Stepwise discretization of subenvironment in one region to that of an adjacent region via an expansion in range, i.e.: in one region to that of an adjacent region via an expansion in range, i.e.: fn,ν+1 = fnν + f ′ nν∆x + 1 2f ′′ nν(∆x)2 + ... . (3.27) (3.27) To illustrate the method we consider the case of two propagating modes and note that the results can be generalized to N modes. We will denote the coeﬃcients of reﬂection at the interface located at x = xν by An,ν−1 and the coeﬃcients of transmission by Bnν, where n = 1, 2 is the mode index. Given this notation, the incident potential function at a subinterval interface x = xν is then: ˆΦν−1(x, z) = B1,ν−1eik1,ν−1xu1,ν−1(z) + B2,ν−1eik2,ν−1xu2,ν−1(z). (3.28) ˆΦν−1(x, z) = B1,ν−1eik1,ν−1xu1,ν−1(z) + B2,ν−1eik2,ν−1xu2,ν−1(z). 3.3 Discretization of the Range-Dependent Medium (3.28) (3.28) 54 This function will give rise to a transmitted set of waves ˆΦν(x, z) in the region xν ≤x < xν+1, and a reﬂected set, ˆΦ∗ ν−1(x, z), for x < xν traveling to the left: This function will give rise to a transmitted set of waves ˆΦν(x, z) in the region xν ≤x < xν+1, and a reﬂected set, ˆΦ∗ ν−1(x, z), for x < xν traveling to the left: ˆΦν(x, z) = B1νeik1νxu1ν(z) + B2νeik2νxu2ν(z) (3.29) (3.29) ˆΦ∗ ν−1(x, z) = A1,ν−1e−ik1,ν−1xu1,ν−1(z) + A2,ν−1e−ik2,ν−1xu2,ν−1(z). (3.30) ˆΦ∗ ν−1(x, z) = A1,ν−1e−ik1,ν−1xu1,ν−1(z) + A2,ν−1e−ik2,ν−1xu2,ν−1(z). (3.30) (3.30) Note, to simplify the discussion it has been assumed that the ﬁrst region of interest is suﬃciently far away from the source of the acoustic excitation such that the wavefronts are planar. The constants Bnν and Anν of our incident, reﬂected, and transmitted waves are to be determined from the following continuity conditions: ˆΦν−1(xν, z) + ˆΦ∗ ν−1(xν, z) = ˆΦν(xν, z) (3.31) ˆΦν−1(xν, z) + ˆΦ∗ ν−1(xν, z) = ˆΦν(xν, z) (3.31) ˆΦ′ ν−1(xν, z) + ˆΦ∗′ ν−1(xν, z) = ˆΦ′ ν(xν, z), (3.32) (3.31) ˆΦ′ ν−1(xν, z) + ˆΦ∗′ ν−1(xν, z) = ˆΦ′ ν(xν, z), (3.32) (3.32) where Equation (3.31) is continuity of the total potential and Equation (3.32) is continuity of the normal derivative, each at the interface located at x = xν. Figure (3.3) depicts this stepwise discretization approach, including an illustration of energy transport through the discretized waveguide. Recalling Equation (3.6), ^Φ0(x; z) ^Φ∗ 0(x; z) ^Φ1(x; z) ^Φν−1(x; z) ^Φ∗ ν−1(x; z) ^Φν(x; z) ... ... ... ... fn1 fn;ν−1 fn;τ−1 x1 = α1 x2 xν−1 xν xτ = α2 ^Φτ−1(x; z) xτ−1 Figure 3.3: Energy transport through the discretized environment Figure 3.3: Energy transport through the discretized environment we see that the relation between eigenfunctions in subinterval ν −1 and ν is given 55 by: u1,ν−1 = ⟨u1,ν−1, u1ν⟩u1ν + ⟨u1,ν−1, u2ν⟩u2ν (3.33) (3.33) (3.34) u2,ν−1 = ⟨u2,ν−1, u1ν⟩u1ν + ⟨u2,ν−1, u2ν⟩u2ν. 3.3 Discretization of the Range-Dependent Medium (3.34) Substituting these expressions into the continuity conditions and equating coeﬃ- cients of u1ν and u2ν yields a coupled set of diﬀerence equations for the coeﬃcients of the reﬂected and transmitted waves: B1νeik1νxν = B1,ν−1eik1,ν−1xν + A1,ν−1e−ik1,ν−1xν ⟨u1,ν−1, u1ν⟩ (3.35) + B2,ν−1eik2,ν−1xν + A2,ν−1e−ik2,ν−1xν ⟨u2,ν−1, u1ν⟩ B1νeik1νxν = B1,ν−1eik1,ν−1xν + A1,ν−1e−ik1,ν−1xν ⟨u1,ν−1, u1ν⟩ (3.35) + B2,ν−1eik2,ν−1xν + A2,ν−1e−ik2,ν−1xν ⟨u2,ν−1, u1ν⟩ k1νB1νeik1νxν = k1,ν−1 B1,ν−1eik1,ν−1xν −A1,ν−1e−ik1,ν−1xν ⟨u1,ν−1, u1ν⟩ (3.36) + k2,ν−1 B2,ν−1eik2,ν−1xν −A2,ν−1e−ik2,ν−1xν ⟨u2,ν−1, u1ν⟩ B2νeik2νxν = B1,ν−1eik1,ν−1xν + A1,ν−1e−ik1,ν−1xν ⟨u1,ν−1, u2ν⟩ (3.37) + B2,ν−1eik2,ν−1xν + A2,ν−1e−ik2,ν−1xν ⟨u2,ν−1, u2ν⟩ k2νB2νeik2νxν = k1,ν−1 B1,ν−1eik1,ν−1xν −A1,ν−1e−ik1,ν−1xν ⟨u1,ν−1, u2ν⟩ (3.38) + k2,ν−1 B2,ν−1eik2,ν−1xν −A2,ν−1e−ik2,ν−1xν ⟨u2,ν−1, u2ν⟩. (3.35) νB1νeik1νxν = k1,ν−1 B1,ν−1eik1,ν−1xν −A1,ν−1e−ik1,ν−1xν ⟨u1,ν−1, u1ν⟩ (3.36) + k2,ν−1 B2,ν−1eik2,ν−1xν −A2,ν−1e−ik2,ν−1xν ⟨u2,ν−1, u1ν⟩ B2νeik2νxν = B1,ν−1eik1,ν−1xν + A1,ν−1e−ik1,ν−1xν ⟨u1,ν−1, u2ν⟩ (3.37) + B2,ν−1eik2,ν−1xν + A2,ν−1e−ik2,ν−1xν ⟨u2,ν−1, u2ν⟩ k2νB2νeik2νxν = k1,ν−1 B1,ν−1eik1,ν−1xν −A1,ν−1e−ik1,ν−1xν ⟨u1,ν−1, u2ν⟩ (3.38) + k2,ν−1 B2,ν−1eik2,ν−1xν −A2,ν−1e−ik2,ν−1xν ⟨u2,ν−1, u2ν⟩. (3.37) Here, Equations (3.35) and (3.36) were obtained by equating the coeﬃcients of u1ν, while Equations (3.37) and (3.38) were obtained by equating the coeﬃcients of u2ν. The coupled system consisting of Equations (3.35)-(3.38) provides a path toward a solution for energy transport through the discretized range-dependent acoustic environment for which closed-form expressions for the range-expanded normal mode inner product may be used to compute mode coupling. This system is, however, complicated and does not provide a direct means for implementing the desired solution into a computational model for acoustic propagation prediction. Fortunately, simpliﬁcation of this system of equations is possible by applying the 56 56 previously derived properties of the range-expanded normal mode inner product and making use of a set of asymptotic identities to be developed in the next section. 3.4 Asymptotic Identities Recalling the range-expansion representation in Equations (3.4) and (3.5), a set of asymptotic identities relating the eigenvalues and eigenfunctions in adjacent regions in the discretized medium may be developed. To derive these identities we let kn,ν−1 and un,ν−1 correspond to the eigenvalue and eigenfunction associated with mode n at range x, which exists in region ν −1. Similarly, we let knν and unν correspond to the eigenvalue and eigenfunction associated with mode n at range x+∆x, which exists in region ν. Given this notation, we ﬁrst consider the relation knν +kn,ν−1. Applying the range expansion in Equations (3.4) to the term knν and dropping terms higher than second order we obtain the expression: knν + kn,ν−1 ≈2kn(x) + k′ n(x)∆xν + 1 2k′′ n(x)(∆xν)2. (3.39) (3.39) Following a similar procedure, the relations knν −kn,ν−1 ≈k′ n(x)∆xν + 1 2k′′ n(x)(∆xν)2 (3.40) (3.40) and kn,ν−1 −knν ≈−k′ n(x)∆xν + 1 2k′′ n(x)(∆xν)2 (3.41) (3.41) may also be obtained. may also be obtained. may also be obtained. Next, we apply the range expansion in Equation (3.5) to the term unν to obtain the following two expressions for the range-expanded normal mode inner product: ⟨unν, um,ν−1⟩≈⟨u′ n(z, x), um(z, x)⟩∆xν + 1 2⟨u′′ n(z, x), um(z, x)⟩(∆xν)2, m ̸= n (3.42) (3.42) and ⟨unν, un,ν−1⟩≈1 + ⟨u′ n(z, x), un(z, x)⟩∆xν + 1 2⟨u′′ n(z, x), un(z, x)⟩(∆xν)2. (3.43) (3.43) Applying the Bachmann-Landau notation developed in Section (1.2), the following asymptotic identities may now be established Applying the Bachmann-Landau notation developed in Section (1.2), the following asymptotic identities may now be established knν + kn,ν−1 = O [1] knν −kn,ν−1 = O [∆xν] kn,ν−1 −knν = O [∆xν] ⟨unν, um,ν−1⟩= O [∆xν] ⟨unν, un,ν−1⟩= 1 + O [∆xν]                            m ̸= n. (3.44) kn,ν−1 −knν = O [∆xν] ⟨unν, um,ν−1⟩= O [∆xν] ⟨unν, un,ν−1⟩= 1 + O [∆xν]                m ̸= n. (3.44) (3.44) These identities will be utilized in simplifying the system of coupled diﬀerence equations described by Equations (3.35)-(3.38). 3.5 Asymptotic Solution In deriving a solution to the system in Equations (3.35)-(3.38), we will ﬁrst derive expressions for the reﬂection coeﬃcients A1,ν−1 and A2,ν−1 that occur at the interface x = xν of our two-mode example. We wish to obtain expressions for these coeﬃcients in terms of the transmission coeﬃcients B1,ν−1 and B2,ν−1 as this will allow for a recursive means of computing the coeﬃcients in any region given knowledge of the initial transmission coeﬃcients in the ﬁrst region. To solve for the reﬂection coeﬃcients we begin by multiplying Equation (3.35) by k1ν and subtracting Equation (3.36) from the resulting expression. Similarly, we multiply Equation (3.37) by k2ν and subtract from it Equation (3.38). These operations 58 result in two intermediate equations: result in two intermediate equations: result in two intermediate equations: A1,ν−1e−ik1,ν−1xν(k1ν + k1,ν−1) + B1,ν−1eik1,ν−1xν(k1ν −k1,ν−1) ⟨u1,ν−1, u1ν⟩ + A2,ν−1e−ik2,ν−1xν(k1ν + k2,ν−1) + B2,ν−1eik2,ν−1xν(k1ν −k2,ν−1) ⟨u2,ν−1, u1ν⟩ = 0 (3.45) ( ) A1,ν−1e−ik1,ν−1xν(k2ν + k1,ν−1) + B1,ν−1eik1,ν−1xν(k2ν −k1,ν−1) ⟨u1,ν−1, u2ν⟩ + A2,ν−1e−ik2,ν−1xν(k2ν + k2,ν−1) + B2,ν−1eik2,ν−1xν(k2ν −k2,ν−1) ⟨u2,ν−1, u2ν⟩ = 0. (3.46) To solve for A1,ν−1 we eliminate A2,ν−1 from these expressions by mul- tiplying Equation (3.45) by (k2ν + k2,ν−1)⟨u2,ν−1, u2ν⟩and Equation (3.46) by (k1ν + k2,ν−1)⟨u2,ν−1, u1ν⟩and subtracting the two resulting expressions. This leads to the expression: A1,ν−1e−ik1,ν−1xν (k1ν + k1,ν−1)(k2ν + k2,ν−1)⟨u1,ν−1, u1ν⟩⟨u2,ν−1, u2ν⟩ −(k2ν + k1,ν−1)(k1ν + k2,ν−1)⟨u1,ν−1, u2ν⟩⟨u2,ν−1, u1ν⟩ +B1,ν−1eik1,ν−1xν (k1ν −k1,ν−1)(k2ν + k2,ν−1)⟨u1,ν−1, u1ν⟩⟨u2,ν−1, u2ν⟩ −(k2ν −k1,ν−1)(k1ν + k2,ν−1)⟨u1,ν−1, u2ν⟩⟨u2,ν−1, u1ν⟩ +B2,ν−1eik2,ν−1xν (k1ν −k2,ν−1)(k2ν + k2,ν−1)⟨u2,ν−1, u1ν⟩⟨u2,ν−1, u2ν⟩ −(k2ν −k2,ν−1)(k1ν + k2,ν−1)⟨u2,ν−1, u2ν⟩⟨u2,ν−1, u1ν⟩ (3.47) (3.47) = 0. To simplify Equation (3.47) we apply the asymptotic identities given in Equation (3.44) and recall that ∆xν is small and therefore it is reasonable to eliminate terms of O [(∆x)2] and smaller. In doing so and dividing by (k2ν + k2,ν−1), we arrive at 59 the following simpliﬁed equation: the following simpliﬁed equation: the following simpliﬁed equation: A1,ν−1e−ik1,ν−1xν(k1ν + k1,ν−1)⟨u1,ν−1, u1ν⟩⟨u2,ν−1, u2ν⟩ + B1,ν−1eik1,ν−1xν(k1ν −k1,ν−1)⟨u1,ν−1, u1ν⟩⟨u2,ν−1, u2ν⟩ + B2,ν−1eik2,ν−1xν(k1ν −k2,ν−1)⟨u2,ν−1, u1ν⟩⟨u2,ν−1, u2ν⟩= 0. (3.48) + B1,ν−1eik1,ν−1xν(k1ν −k1,ν−1)⟨u1,ν−1, u1ν⟩⟨u2,ν−1, u2ν⟩ (3.48) (3.48) Next, we recognize that Next, we recognize that Next, we recognize that k1ν + k1,ν−1 = 2k1,ν−1 + (k1ν −k1,ν−1) = 2k1,ν−1 + O [∆xν] (3.49) (3.49) and k1ν −k2,ν−1 = (k1,ν−1 −k2,ν−1) + (k1ν −k1,ν−1) = (k1,ν−1 −k2,ν−1) + O [∆xν] . 3.5 Asymptotic Solution (3.50) (3.50) = (k1,ν−1 −k2,ν−1) + O [∆xν] . If we now substitute into Equation (3.48) the expressions on the right side of Equations (3.49) and (3.50) and again drop terms of O [(∆x)2] and smaller from the resulting expression we arrive at the following asymptotic solution for the reﬂection coeﬃcient A1,ν−1: A1,ν−1 = −B1,ν−1 (k1ν −k1,ν−1) 2k1,ν−1 e2ik1,ν−1xν −B2,ν−1 (k1,ν−1 −k2,ν−1) 2k1,ν−1 ei(k1,ν−1+k2,ν−1)xν ⟨u2,ν−1, u1ν⟩ ⟨u1,ν−1, u1ν⟩. (3.51) (3.51) To solve for A2,ν−1 we follow a similar procedure. We eliminate A1,ν−1 from Equations (3.45) and (3.46) by multiplying Equation (3.45) by (k2ν + k1,ν−1)⟨u1,ν−1, u2ν⟩and Equation (3.46) by (k1ν + k1,ν−1)⟨u1,ν−1, u1ν⟩and 60 60 subtracting the two resulting expressions. This leads to: A2,ν−1e−ik2,ν−1xν (k1ν + k2,ν−1)(k2ν + k1,ν−1)⟨u2,ν−1, u1ν⟩⟨u1,ν−1, u2ν⟩ −(k2ν + k2,ν−1)(k1ν + k1,ν−1)⟨u2,ν−1, u2ν⟩⟨u1,ν−1, u1ν⟩ +B1,ν−1eik1,ν−1xν (k1ν −k1,ν−1)(k2ν + k1,ν−1)⟨u1,ν−1, u1ν⟩⟨u1,ν−1, u2ν⟩ −(k2ν −k1,ν−1)(k1ν + k1,ν−1)⟨u1,ν−1, u2ν⟩⟨u1,ν−1, u1ν⟩ +B2,ν−1eik2,ν−1xν (k1ν −k2,ν−1)(k2ν + k1,ν−1)⟨u2,ν−1, u1ν⟩⟨u1,ν−1, u2ν⟩ −(k2ν −k2,ν−1)(k1ν + k1,ν−1)⟨u2,ν−1, u2ν⟩⟨u1,ν−1, u1ν⟩ (3.52) (3.52) = 0. = 0. Once again, we utilize the asymptotic identities given in Equation (3.44) to simplify the above expression by eliminating terms of O [(∆x)2] and smaller. After dividing by (k1ν + k1,ν−1), this simpliﬁcation leads to the expression Once again, we utilize the asymptotic identities given in Equation (3.44) to simplify the above expression by eliminating terms of O [(∆x)2] and smaller. After dividing by (k1ν + k1,ν−1), this simpliﬁcation leads to the expression −A2,ν−1e−ik2,ν−1xν(k2ν + k2,ν−1)⟨u2,ν−1, u2ν⟩⟨u1,ν−1, u1ν⟩ −B1,ν−1eik1,ν−1xν(k2ν −k1,ν−1)⟨u1,ν−1, u2ν⟩⟨u1,ν−1, u1ν⟩ −B2,ν−1eik2,ν−1xν(k2ν −k2,ν−1)⟨u2,ν−1, u2ν⟩⟨u1,ν−1, u1ν⟩= 0. (3.53) (3.53) Using similar arguments as those used in Equations (3.49) and (3.50) we can re- place k2ν + k2,ν−1 in Equation (3.53) with 2k2,ν−1 + O [∆xν] and k2ν −k1,ν−1 with (k2,ν−1 −k1,ν−1) + O [∆xν]. Eliminating any remaining terms that are smaller than O [∆x] leads to the following asymptotic solution for A2,ν−1: A2,ν−1 = −B1,ν−1 (k2,ν−1 −k1,ν−1) 2k2,ν−1 ei(k1,ν−1+k2,ν−1)xν ⟨u1,ν−1, u2ν⟩ ⟨u2,ν−1, u2ν⟩ −B2,ν−1 (k2ν −k2,ν−1) 2k2,ν−1 e2ik2,ν−1xν. (3.54) (3.54) To derive expressions for the transmission coeﬃcients B1ν and B2ν for our two mode example in terms of the coeﬃcients B1,ν−1 and B2,ν−1 we follow a procedure 61 similar to that for the reﬂection coeﬃcients. First we multiply Equation (3.35) by k1,ν−1 and add Equation (3.36) to the resulting expression. Similarly, we multiply Equation (3.37) by k2,ν−1 and add to it Equation (3.38). 3.5 Asymptotic Solution Doing so, we arrive at the intermediate equations: B1νeik1νxν(k1,ν−1 + k1ν) = 2k1,ν−1B1,ν−1eik1,ν−1xν⟨u1,ν−1, u1ν⟩ + A2,ν−1e−ik2,ν−1xν(k1,ν−1 −k2,ν−1) + B2,ν−1eik2,ν−1xν(k1,ν−1 + k2,ν−1) ⟨u2,ν−1, u1ν⟩ (3.55) (3.55) + B2,ν−1eik2,ν−1xν(k1,ν−1 + k2,ν−1) ⟨u2,ν−1, u1ν⟩ and B2νeik2νxν(k2,ν−1 + k2ν) = B1,ν−1eik1,ν−1xν(k2,ν−1 + k1,ν−1) + A1,ν−1e−ik1,ν−1xν(k2,ν−1 −k1,ν−1) ⟨u1,ν−1, u2ν⟩ + 2k2,ν−1B2,ν−1eik2,ν−1xν⟨u2,ν−1, u2ν⟩. (3.56) B2νeik2νxν(k2,ν−1 + k2ν) = B1,ν−1eik1,ν−1xν(k2,ν−1 + k1,ν−1) + A1,ν−1e−ik1,ν−1xν(k2,ν−1 −k1,ν−1) ⟨u1,ν−1, u2ν⟩ + 2k2,ν−1B2,ν−1eik2,ν−1xν⟨u2,ν−1, u2ν⟩. (3.56) + 2k2,ν−1B2,ν−1eik2,ν−1xν⟨u2,ν−1, u2ν⟩. Next we substitute the known values of A1,ν−1 and A2,ν−1 from Equations (3.51) and (3.54) into Equations (3.55) and (3.56). Doing so and solving Equations (3.55) for B1ν we arrive at the expression: B1ν = B1,ν−1 eik1,ν−1xν eik1νxν(k1,ν−1 + k1ν) × 2k1,ν−1⟨u1,ν−1, u1ν⟩ −(k2,ν−1 −k1,ν−1)(k1,ν−1 −k2,ν−1) 2k2,ν−1 ⟨u1,ν−1, u2ν⟩ ⟨u2,ν−1, u2ν⟩⟨u2,ν−1, u1ν⟩ + B2,ν−1 eik2,ν−1xν eik1νxν(k1,ν−1 + k1ν) × (k1,ν−1 + k2,ν−1) −(k2ν −k2,ν−1)(k1,ν−1 −k2,ν−1) 2k2,ν−1 ⟨u2,ν−1, u1ν⟩. (3.57) B1ν = B1,ν−1 eik1,ν−1xν eik1νxν(k1,ν−1 + k1ν) × 2k1,ν−1⟨u1,ν−1, u1ν⟩ (3.57) We simplify Equation (3.57) by performing an order of magnitude analysis of its terms using the asymptotic identities given in Equation (3.44) and eliminating terms of O [(∆x)2] and smaller. This simpliﬁcation yields the following asymptotic 62 solution for B1ν: B1ν =B1,ν−1 2k1,ν−1 (k1,ν−1 + k1ν)ei(k1,ν−1−k1ν)xν⟨u1,ν−1, u1ν⟩ + B2,ν−1 (k2,ν−1 + k1,ν−1) 2k1,ν−1 ei(k2,ν−1−k1ν)xν⟨u2,ν−1, u1ν⟩. (3.58) (3.58) To obtain a solution for B2ν we follow a similar procedure. We begin by solving Equation (3.56) for B2ν, where A1,ν−1 and A2,ν−1 and are replaced with their known values. This results in the expression: eik1,ν−1xν B2ν = B1,ν−1 eik1,ν−1xν eik2νxν(k2,ν−1 + k2ν) × (k2,ν−1 + k1,ν−1) −(k1ν −k1,ν−1)(k2,ν−1 −k1,ν−1) 2k1,ν−1 ⟨u1,ν−1, u2ν⟩ + B2,ν−1 eik2,ν−1xν eik2νxν(k2,ν−1 + k2ν) × 2k2,ν−1⟨u2,ν−1, u2ν⟩ −(k1,ν−1 −k2,ν−1)(k2,ν−1 −k1,ν−1) 2k1,ν−1 ⟨u2,ν−1, u1ν⟩ ⟨u1,ν−1, u1ν⟩⟨u1,ν−1, u2ν⟩ . B2ν = B1,ν−1 eik1,ν−1xν eik2νxν(k2,ν−1 + k2ν) × (k2,ν−1 + k1,ν−1) −(k1ν −k1,ν−1)(k2,ν−1 −k1,ν−1) 2k1,ν−1 ⟨u1,ν−1, u2ν⟩ k (3.59) Yet again utilizing the asymptotic identities from Equation (3.44), we eliminate terms of O [(∆x)2] and smaller and arrive at the following asymptotic solution for B2 : B2ν: B2ν: ν =B1,ν−1 (k1,ν−1 + k2,ν−1) 2k2,ν−1 ei(k1,ν−1−k2ν)xν⟨u1,ν−1, u2ν⟩ + B2,ν−1 2k2,ν−1 (k2,ν−1 + k2ν)ei(k2,ν−1−k2ν)xν⟨u2,ν−1, u2ν⟩. (3.60) (3.60) The asymptotic expressions for the reﬂection and transmission coeﬃcients admit a convenient matrix formulation. 3.5 Asymptotic Solution To express the matrix form of these solu- tions we begin by deﬁning a reﬂection coeﬃcient matrix, Rν, and a transmission coeﬃcient matrix, Tν: κRν 12 ei(k1,ν−1+k2,ν−1)xν ⟨u2,ν−1, u1ν⟩ ⟨u1,ν−1, u1ν⟩ κRν 22 e2ik2,ν−1xν       , Rν =       κRν 11 e2ik1,ν−1xν κRν 12 ei(k1,ν−1+k2,ν−1)xν ⟨u2,ν−1, u1ν⟩ ⟨u1,ν−1, u1ν⟩ κRν 21 ei(k1,ν−1+k2,ν−1)xν ⟨u1,ν−1, u2ν⟩ ⟨u2,ν−1, u2ν⟩ κRν 22 e2ik2,ν−1xν       , (3.61) (3.61) 63 where κRν mn =            −(kmν −km,ν−1) 2km,ν−1 m = n −(km,ν−1 −kn,ν−1) 2km,ν−1 m ̸= n (3.62) where (3.62) and Tν =       κTν 11ei(k1,ν−1−k1ν)xν⟨u1,ν−1, u1ν⟩ κTν 12ei(k2,ν−1−k1ν)xν⟨u2,ν−1, u1ν⟩ κTν 21ei(k1,ν−1−k2ν)xν⟨u1,ν−1, u2ν⟩ κTν 22ei(k2,ν−1−k2ν)xν⟨u2,ν−1, u2ν⟩       , (3.63) and (3.63) where  2k where κTν mn =            2km,ν−1 (km,ν−1 + kmν) m = n (km,ν−1 + kn,ν−1) 2km,ν−1 m ̸= n. (3.64) (3.64) This allows one to write the following recursive matrix expressions for the reﬂection and transmission coeﬃcients: A1,ν−1 A2,ν−1 = Rν B1,ν−1 B2,ν−1 = Rν ν−1 Y σ=1 Tσ B10 B20 (3.65) (3.65) and B1ν B2ν = Tν B1,ν−1 B2,ν−1 = νY σ=1 Tσ B10 B20 , (3.66) (3.66) where B10 and B20 are the transmission coeﬃcients corresponding to modes 1 and 2, respectively, in the region containing the acoustic source. If acoustic absorption is considered negligible in this region both of these coeﬃcients are equal to unity. The recursive formulae in Equations (3.65) and (3.66) represent the principle result of the presented theory. These expressions provide a means of mathemati- cally modeling energy transport through a range-dependent shallow water acoustic medium where non-adiabatic coupling between modes is estimated via the range- expanded normal mode inner product discussed in Section (3.2). Moreover, this 64 diﬀerence equations solution is easily implemented into a discrete computing de- vice thus making it suitable for approximating non-adiabatic mode coupling in a computational model for acoustic propagation prediction. When paired with the discretization approach discussed in Section (3.3), this solution has the potential to provide signiﬁcant computational savings over currently existing coupled mode solutions. 3.5 Asymptotic Solution In particular, the ability to compute mode coupling via closed-form al- gebraic expressions oﬀers signiﬁcant savings as it avoids the need for numerically solving an inner product integral at each of the interfaces separating the horizontal regions throughout the medium. The elimination of the eigenvalue search problem is an additional beneﬁt of this asymptotic coupled mode solution, as mentioned previously. The remaining goal is to extend this solution to study a particular problem of interest: mode coupling in shallow water due to horizontally-variable bottom depth. To accomplish this we will ﬁrst adapt the solution to allow for non-planar wavefronts due to short range (i.e. near-source) propagation. This will allow for the formulation of a transmission loss equation for propagation in a cylindrical waveguide. Following this, adjustments will be made to the solution to ensure numerical stability. Finally, transmission loss estimates produced via this asymp- totic stepwise coupled mode solution will be compared to benchmark solutions for several cases of interest. 3.6 Cylindrical Wave Extension In many propagation scenarios of interest the plane wave assumption made in Section (3.3) is insuﬃcient. This is due to the fact that at short distances from the acoustic source the wavefronts of an omnidirectional point source are typically spherical or cylindrical depending on the nature of the waveguide. For the case of upslope propagation in shallow water this is particularly true as the constant 65 decrease in waveguide depth causes the modes of vibration of the acoustic medium to become rapidly cutoﬀ, restricting the propagation scenario to shorter ranges. To extend the solution in Section (3.5) to these scenarios one must reformulate the expressions for the incident, reﬂected, and transmitted waves in Equations (3.28)-(3.30) to accommodate non-planar wavefronts. Going forward, we will assume that acoustic propagation is occurring in a shallow water cylindrical waveguide with azimuthal symmetry. The wave equation governing the propagation of acoustic energy in this scenario was discussed in Section (2.2), and the coupled equation governing energy transport with range was discussed in Section (2.6). Applying the discretization technique of Section (3.3) to this waveguide, the incident potential function representing the set of waves arriving at an interface located at range r = rν due to an acoustic excitation created by a source located at r = 0 is given by the expression ˆΦν−1(r, z) = i ˙m √ 8πrρ(z)e−iπ/4 ( B1,ν−1eik1,ν−1r u1,ν−1(zs) ρ(zs) p k1,ν−1 u1,ν−1(z) + B2,ν−1eik2,ν−1r u2,ν−1(zs) ρ(zs) p k2,ν−1 u2,ν−1(z) ) , (3.67) (3.67) where the asymptotic solution in Equation (2.25) of Section (2.2) has been used to inform the expression in Equation (3.67). As was the case for the original derivation, we will limit ourselves to two modes to simplify the discussion but note that the method is, again, generalizable to N modes. As in the plane wave example, the incident set of waves given by Equation (3.67) will give rise to a set of transmitted waves ˆΦν in the region rν ≤r < rν+1, and a reﬂected set, ˆΦ∗ ν−1, for r < rν traveling to the left. 3.6 Cylindrical Wave Extension These waves are described by the expressions ˆΦν(r, z) = i ˙m √ 8πrρ(z)e−iπ/4 ( B1νeik1νr u1ν(zs) ρ(zs)√k1ν u1ν(z) + B2νeik2νr u2ν(zs) ρ(zs)√k2ν u2ν(z) ) (3.68) ˆΦν(r, z) = i ˙m √ 8πrρ(z)e−iπ/4 ( B1νeik1νr u1ν(zs) ρ(zs)√k1ν u1ν(z) + B2νeik2νr u2ν(zs) ρ(zs)√k2ν u2ν(z) ) (3.68) (3.68) 66 and ˆΦ∗ ν−1(r, z) = − i ˙m √ 8πrρ(z)eiπ/4 ( A1,ν−1e−ik1,ν−1r u1,ν−1(zs) ρ(zs) p k1,ν−1 u1,ν−1(z) + A2,ν−1e−ik2,ν−1r u2,ν−1(zs) ρ(zs) p k2,ν−1 u2,ν−1(z) ) (3.69) (3.69) As was the case in Section (3.3), application of the continuity conditions given by Equations (3.31) and (3.32) along with the relationship given by Equations (3.33) and (3.34) leads to a coupled set of diﬀerence equations for the constants Bnν and Anν of our incident, reﬂected, and transmitted waves. This system is simpliﬁed following a procedure identical to that of Section (3.5), leading to the following reﬂection and transmission coeﬃcient matrices for our discretized cylin- drical waveguide: Rν =     κRν 11 i e2ik1,ν−1rν κRν 12 i ei(k1,ν−1+k2,ν−1)rν ⟨u2,ν−1, u1ν⟩ ⟨u1,ν−1, u1ν⟩γRν 12 κRν 21 i ei(k1,ν−1+k2,ν−1)rν ⟨u1,ν−1, u2ν⟩ ⟨u2,ν−1, u2ν⟩γRν 21 κRν 22 i e2ik2,ν−1rν    , (3.70) h (3.70) where where γRν mn = un,ν−1(zs) um,ν−1(zs) p km,ν−1 p kn,ν−1 , m ̸= n (3.71) γRν mn = un,ν−1(zs) um,ν−1(zs) p km,ν−1 p kn,ν−1 , m ̸= n (3.71) and and Tν =       κTν 11ei(k1,ν−1−k1ν)rν⟨u1,ν−1, u1ν⟩γTν 11 κTν 12ei(k2,ν−1−k1ν)rν⟨u2,ν−1, u1ν⟩γTν 12 κTν 21ei(k1,ν−1−k2ν)rν⟨u1,ν−1, u2ν⟩γTν 21 κTν 22ei(k2,ν−1−k2ν)rν⟨u2,ν−1, u2ν⟩γTν 22       , (3.72) (3.72) (3.72) (3.72) (3.72) (3.72) where  (3.72) h where where γTν mn =              um,ν−1(zs) umν(zs) √kmν p km,ν−1 m = n un,ν−1(zs) umν(zs) √kmν p kn,ν−1 m ̸= n. (3.73) (3.73) 67 Here, the terms κRν mn and κTν mn are given by Equations (3.62) and (3.64), respec- tively. Similar to the plane wave case, these matrices can be incorporated into the recursive formulae given by Equations (3.65) and (3.66) to compute the reﬂection and transmission coeﬃcients at an interface corresponding to any region in the discretized medium given knowledge of the transmission coeﬃcients for the region containing the acoustic source. The reformulation of the transmission coeﬃcients via the matrix in Equation (3.72) allows for the deﬁnition of a forward transmission loss equation for the discretized range-dependent cylindrical waveguide with azimuthal symmetry. 3.6 Cylindrical Wave Extension This transmission loss equation is given by the expression: TL(r, z) = −20 log10 r 2π r N X m=1 Bmνumν(z) umν(zs) ρ(zs)√kmν eikmνr ; r ∈[rν, rν+1), (3.74) (3.74) where the terms Bmν are computed recursively by generalizing the results in Equa- tions (3.66) and (3.72) to N modes. Equation (3.74) accounts for losses due to cylindrical spreading in the waveguide as well as non-adiabatic mode coupling via the transmission coeﬃcients Bmν. While this equation is correct relative to the physical assumptions made, we will see in the next section that adjustments are necessary to ensure numerical stability of the solution for all ranges and depths. 3.7 Reformulation for Numerical Stability When implementing the transmission loss solution given by Equation (3.74) via Equations (3.66) and (3.72) into a discrete computing environment there are two terms that require special attention to maintain numerical stability. The ﬁrst is the exponential term that appears in the entries of the transmission coeﬃcient matrix in Equation (3.72). The numerical instability in this term becomes obvious if one rewrites the term as: ei(kn,ν−1−kmν)rν = eikn,ν−1rν e−ikmνrν . (3.75) (3.75) 68 When the mode of index m in the above expression transitions from propagating to evanescent, its associated eigenvalue, kmν, becomes purely imaginary. Under these conditions, the exponential on the left side of this expression becomes numerically unstable at large values of rν due to the fact that the term in brackets on the right grows without bound. The instability in this term can be removed if we distribute the exponential term from the forward transmission loss equation in Equation (3.74) inside the expression for the transmission coeﬃcient Bmν given by Equations (3.66) and (3.72). Doing so, we see that the exponential term in the entries of the transmission coeﬃcient matrix become eikn,ν−1rν e−ikmνrν × eikmνr = eikn,ν−1rν eikmν(r−rν) . (3.76) (3.76) Since rν corresponds to the range at the beginning of the horizontal region ν and r corresponds to ranges within said region, we have that (r −rν) ≥0. As such, the exponential term in Equation (3.76) is numerically stable since the term in brackets on the right side of this equation decays exponentially with increasing range from the interface located at r = rν. The second numerically unstable term with which we must concern ourselves is the eigenfunction ratio that appears in the entries of the matrix in Equation (3.72). This term is written un,ν−1(zs) umν(zs) √kmν p kn,ν−1 . (3.77) (3.77) It is clear from Equation (3.77) that for certain source depths (i.e. those cor- responding to the zeros of the eigenfunction umν) the above term will become undeﬁned due to division by zero. To remove this instability we again turn our attention to the forward transmission loss equation in Equation (3.74). If we dis- tribute the term umν(zs)/ρ(zs)√kmν from this equation inside the expression for the transmission coeﬃcient Bmν given by Equations (3.66) and (3.72), the term in It is clear from Equation (3.77) that for certain source depths (i.e. 3.7 Reformulation for Numerical Stability those cor- responding to the zeros of the eigenfunction umν) the above term will become undeﬁned due to division by zero. To remove this instability we again turn our attention to the forward transmission loss equation in Equation (3.74). If we dis- tribute the term umν(zs)/ρ(zs)√kmν from this equation inside the expression for the transmission coeﬃcient Bmν given by Equations (3.66) and (3.72), the term in 69 Equation (3.77) becomes: Equation (3.77) becomes: Equation (3.77) becomes: Equation (3.77) becomes: un,ν−1(zs) umν(zs) √kmν p kn,ν−1 × umν(zs) ρ(zs)√kmν = un,ν−1(zs) ρ(zs) p kn,ν−1 . (3.78) (3.78) It is apparent from Equation (3.78) that this term is now stable for all source depths. Taking these results into account, we now wish to deﬁne a new set of expres- sions for the numerically stable reformulation of the presented asymptotic stepwise coupled mode solution. We will begin by deﬁning the following expression corre- sponding to the initial transmission coeﬃcients existing in the horizontal partition containing the acoustic source: BS m0 = Bm0eikm0r0 um0(zs) ρ(zs)√km0 , (3.79) (3.79) where Bm0 are the initial transmission coeﬃcients appearing in Equation (3.66). Given this deﬁnition, the recursive matrix solution for the numerically stable trans- mission coeﬃcients is given by the expression     BS 1ν BS 2ν    = T S ν     BS 1,ν−1 BS 2,ν−1    = νY σ=1 T S σ     BS 10 BS 20    , (3.80) (3.80) where T S ν =       κTν 11eik1ν(r−rν)⟨u1,ν−1, u1ν⟩ κTν 12eik1ν(r−rν)⟨u2,ν−1, u1ν⟩ κTν 21eik2ν(r−rν)⟨u1,ν−1, u2ν⟩ κTν 22eik2ν(r−rν)⟨u2,ν−1, u2ν⟩       . (3.81) (3.81) Finally, we may deﬁne the following numerically stable equivalent of the forward transmission loss equation in Equation (3.74): TLS(r, z) = −20 log10 r 2π r N X m=1 BS mνumν(z) ; r ∈[rν, rν+1), (3.82) (3.82) 70 where the terms BS mν in the summation in Equation (3.82) are obtained by gener- alizing the results of Equations (3.79)-(3.81) to N modes. Going forward we wish to verify this result for several range-dependent bottom depth problems of interest. 3.8 Mode Turnaround and Eigenvalue Lag Distance An additional numerical issue that requires special treatment when imple- menting the presented solution is that of mode turnaround. Mode turnaround occurs when a mode’s grazing angle begins to approach 90◦in the vicinity of its cutoﬀdepth. Due to the stepwise nature of the approach presented here, this phe- nomenon creates a discontinuity in the transmission coeﬃcients computed via the matrix in Equation (3.72). As discussed in Section (2.3), when a mode becomes fully cutoﬀits eigenvalue becomes purely imaginary and thus its real part is zero. To understand how this aﬀects the transmission coeﬃcient calculation we turn our attention to the terms κTν nn, which appear as multipliers in the diagonal terms of the matrix in Equation (3.72). Recalling Equation (3.64), these terms are given by the expression: 2k κTν nn = 2kn,ν−1 (kn,ν−1 + knν). (3.83) (3.83) If we consider the case of upslope propagation, we see that the term knν in this expression will become imaginary in the vicinity of mode n’s cutoﬀdepth prior to the term kn,ν−1, as the latter term lags the former by a distance of ∆rν−1. This lag distance between the eigenvalues of the adjacent regions ν −1 and ν will create a temporary peak in the term κTν nn in the vicinity of the range at which mode n turns around. This peak will, in turn, create a discontinuity in the transmission coeﬃcient associated with mode n, as evident from the matrix in Equation (3.72). Figure (3.4) provides a depiction of the lag distance between the eigenvalues kn,ν−1 and knν in the vicinity of a mode turning around in an upslope propagation sce- nario. As can be seen in this ﬁgure, the real part of the eigenvalue knν becomes 71 71 zero a short distance before that of the eigenvalue kn,ν−1. In Figure (3.5a) a plot is provided of a mode’s grazing angle approaching 90◦as it becomes cutoﬀin an upsloping waveguide. The discontinuity created in the transmission coeﬃcient as- sociated with this mode is then depicted in Figure (3.5b). The result of the Range (km) 2.84 2.845 2.85 2.855 2.86 2.865 2.87 2.875 2.88 Re(kn) 0.000 0.002 0.004 0.006 0.008 0.010 0.012 0.014 0.016 0.018 Eigenvalue lag distance kn,ν−1 knν Figure 3.4: Depiction of eigenvalue lag distance in the vicinity of a mode’s turnaround point in an upslope propagation scenario. 3.8 Mode Turnaround and Eigenvalue Lag Distance As can be seen, the real part of the eigenvalue knν becomes zero before that of the eigenvalue kn,ν−1 due to the small distnace, ∆rν−1, separating the respective regions in which the two eigenvalues exist. Eigenvalue lag distance Range (km) 2.84 2.845 2.85 2.855 2.86 2.865 2.87 2.875 2.88 Re(kn) 0.000 0.002 0.004 0.006 0.008 0.010 0.012 0.014 0.016 0.018 Eigenvalue lag distance kn,ν−1 knν Figure 3.4: Depiction of eigenvalue lag distance in the vicinity of a mode’s turnaround point in an upslope propagation scenario. As can be seen, the real part of the eigenvalue knν becomes zero before that of the eigenvalue kn,ν−1 due to the small distnace, ∆rν−1, separating the respective regions in which the two eigenvalues exist. discontinuity in the transmission coeﬃcients as depicted in Figure (3.5b) is that artiﬁcial peaks will appear in the transmission loss curves computed via Equation (3.82) in the vicinity of mode cutoﬀdepths. There are a number of viable ap- proaches for removing these artiﬁcial peaks. 3.8 Mode Turnaround and Eigenvalue Lag Distance A simple approach is to maintain a 72 Range (km) 2.84 2.845 2.85 2.855 2.86 2.865 2.87 2.875 2.88 θgraz (deg) 80 81 82 83 84 85 86 87 88 89 90 Mode Grazing Angle Turnover ( ) Mode Grazing Angle Turnover Range (km) 2.84 2.845 2.85 2.855 2.86 2.865 2.87 2.875 2.88 θgraz (deg) 80 81 82 83 84 85 86 87 88 89 90 Mode Grazing Angle Turnover (a) Range (km) 2.84 2.845 2.85 2.855 2.86 2.865 2.87 2.875 2.88 Re(Bnν) 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Transmission Coefficient Discontinuity (b) gure 3.5: (a) Depiction of a mode’s grazing angle approaching 90◦near its cuto an upslope propagation scenario (b) Discontinuity in the transmission coeﬃcien sociated with a cutoﬀmode 73 (a) Range (km) 2.84 2.845 2.85 2.855 2.86 2.865 2.87 2.875 2.88 Re(Bnν) 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Transmission Coefficient Discontinuity (b) Figure 3.5: (a) Depiction of a mode’s grazing angle approaching 90◦near its cutoﬀ in an upslope propagation scenario (b) Discontinuity in the transmission coeﬃcient associated with a cutoﬀmode 73 Range (km) 2.84 2.845 2.85 2.855 2.86 2.865 2.87 2.875 2.88 Re(Bnν) 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Transmission Coefficient Discontinuity (b) (b) Figure 3.5: (a) Depiction of a mode’s grazing angle approaching 90◦near its cutoﬀ in an upslope propagation scenario (b) Discontinuity in the transmission coeﬃcient associated with a cutoﬀmode 73 73 running average of the term κTν nn for each mode. These average values of κTν nn may then be used as surrogate multipliers in the diagonal terms of the matrix given by Equation (3.72) for range intervals containing mode cutoﬀdepths. In Figure (3.6a) a transmission loss curve produced using Equation (3.82) is provided for an upslope propagation scenario. Distinct peaks can be seen in this ﬁgure as higher order modes are cutoﬀdue to the rapid decrease in waveguide depth. Figure (3.6b) shows a corrected version of the transmission loss curve for this propagation sce- nario where the peaks have been removed using the previously suggested moving average method. 3.8 Mode Turnaround and Eigenvalue Lag Distance 74 Range (km) 0.5 1 1.5 2 2.5 3 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 Upslope Propagation (Uncorrected) Upslope Propagation (Uncorrected) Upslope Propagation (Uncorrected) Upslope Propagation (Uncorrected) Range (km) 0.5 1 1.5 2 2.5 3 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 Upslope Propagation (Uncorrected) (a) Range (km) 0.5 1 1.5 2 2.5 3 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 Upslope Propagation (Corrected) (b) gure 3.6: (a) Depiction of artiﬁcial peaks appearing in the transmission imate for an upslope propagation scenario due to mode cutoﬀ(b) Remov iﬁcial peaks in the transmission loss curve via a moving average applied to m κTν nn of the matrix in Equation (3.72) in the vicinity of mode cutoﬀdept 75 Range (km) 0.5 1 1.5 2 2.5 3 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 Upslope Propagation (Corrected) (b) Upslope Propagation (Corrected) Upslope Propagation (Corrected) (b) Figure 3.6: (a) Depiction of artiﬁcial peaks appearing in the transmission loss estimate for an upslope propagation scenario due to mode cutoﬀ(b) Removal of artiﬁcial peaks in the transmission loss curve via a moving average applied to the term κTν nn of the matrix in Equation (3.72) in the vicinity of mode cutoﬀdepths 75 List of References List of References [1] G. A. Leibiger, “Wave propagation in an inhomogeneous medium with slow spatial variation,” Ph.D. dissertation, Stevens Institute of Technology, Castle Point, Hoboken, New Jersey, 1968. 76 4.1 Background To verify the eﬃcacy of the presented theory a particular problem of interest was chosen: modeling mode coupling due to horizontally-variable bottom depth in a rigid-bottom waveguide. As discussed in Section (3.3), an environment with general range-dependent bottom depth may be approximated by a series of con- nected wedge subenvironments. Moreover, each wedge subenvironment is purely speciﬁed by an initial bottom depth and a bottom slope angle. Taking this into account, a set of test cases was constructed that spans all possible combinations of two-slope propagation environments. It is proposed that the bottom depth for any environment may be approximated by a combination of these two-slope scenarios. To support this claim, one test case was constructed that consists of a bottom depth proﬁle approximated by seven slopes. For all scenarios a sound speed in water of 1500 m s−1 and a water column density of 1000 kg m−3 were chosen. To run these scenarios, a normal modes prop- agation model was developed in MATLAB [1] that implements the discretization technique described in Section (3.3). Output from this model is a set of trans- mission loss curves produced using the forward transmission loss equation given in Equation (3.82). Non-adiabatic mode coupling is accounted for in the model through the recursive computation of transmission coeﬃcients at discrete inter- faces throughout the medium using Equations (3.80). Going forward, this model will be referred to as the Asymptotic Stepwise Coupled Mode (ASCM) model. 77 77 4.2 Comparison to Benchmark Solution To verify the transmission loss estimates generated by the ASCM model, the same propagation scenarios were modeled using COUPLE, a benchmark solution for range-dependent problems developed by Evans [2], which was ﬁrst discussed in Section (2.7). To ensure that the rigid-bottom boundary condition was being correctly represented in COUPLE, a test was performed whereby a rigid ﬂat- bottom propagation scenario described by Jensen, Kuperman, Porter & Schmidt [3] was modeled repeatedly while increasing the sound speed and density in a horizontal layer representing the rigid basement. The sound speed and density for this basement layer were considered suﬃcient once no detectable diﬀerence existed between the transmission loss curve generated by COUPLE and that given in reference [3]. This resulted in the basement layer having a sound speed of 13 000 m s−1 and a density of 3500 kg m−3. In what follows, two ﬁgures are provided for each propagation scenario mod- eled using the ASCM and COUPLE solutions. The ﬁrst ﬁgure provides a plot of the bottom depth over range for each scenario. In this ﬁgure, the source and receiver depths are also depicted using a circle and star, respectively. The sec- ond ﬁgure provides a comparison of the transmission loss curves produced by the ASCM and COUPLE models for each scenario. Above each of the ﬁgures addi- tional information about the propagation scenario is provided. This information includes the source frequency, f, the bottom slope angle, β, the ﬁxed point (source) depth, Fxd, and the moving point (receiver) depth, Mvd. 4.2 Comparison to Benchmark Solution 78 Range (km) 0 0.5 1 1.5 2 2.5 3 Depth (m) -200 -180 -160 -140 -120 -100 -80 -60 -40 -20 0 f = 25 Hz, β = -2.50◦, Fxd = 60.00 m, Mvd = 40.00 m (a) Range (km) 0 0.5 1 1.5 2 2.5 3 Depth (m) -200 -180 -160 -140 -120 -100 -80 -60 -40 -20 0 f = 25 Hz, β = -2.50◦, Fxd = 60.00 m, Mvd = 40.00 m (a) Range (km) 0.5 1 1.5 2 2.5 3 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 f = 25 Hz, β = -2.50◦, Fxd = 60.00 m, Mvd = 40.00 m COUPLE ASCM (b) gure 4.1: (a) Upslope propagation scenario (b) Comparison of transmission l timates produced by the ASCM and COUPLE propagation models for the s ario depicted in (a) 79 Range (km) (a) Range (km) 0.5 1 1.5 2 2.5 3 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 f = 25 Hz, β = -2.50◦, Fxd = 60.00 m, Mvd = 40.00 m COUPLE ASCM (b) Figure 4.1: (a) Upslope propagation scenario (b) Comparison of transmission loss estimates produced by the ASCM and COUPLE propagation models for the sce- nario depicted in (a) 79 f = 25 Hz, β = -2.50◦, Fxd = 60.00 m, Mvd = 40.00 m Range (km) 0.5 1 1.5 2 2.5 3 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 f = 25 Hz, β = -2.50◦, Fxd = 60.00 m, Mvd = 40.00 m COUPLE ASCM (b) (b) Figure 4.1: (a) Upslope propagation scenario (b) Comparison of transmission loss estimates produced by the ASCM and COUPLE propagation models for the sce- nario depicted in (a) 79 79 Range (km) 0 0.5 1 1.5 2 2.5 3 3.5 4 Depth (m) -350 -300 -250 -200 -150 -100 -50 0 f = 25 Hz, β = 2.50◦, Fxd = 60.00 m, Mvd = 160.00 m (a) f = 25 Hz, β = 2.50◦, Fxd = 60.00 m, Mvd = 160.00 m Range (km) 0 0.5 1 1.5 2 2.5 3 3.5 4 Depth (m) -350 -300 -250 -200 -150 -100 -50 0 f = 25 Hz, β = 2.50◦, Fxd = 60.00 m, Mvd = 160.00 m (a) Range (km) 0.5 1 1.5 2 2.5 3 3.5 4 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 f = 25 Hz, β = 2.50◦, Fxd = 60.00 m, Mvd = 160.00 m COUPLE ASCM (b) re 4.2: (a) Downslope propagation scenario (b) Comparison of transmiss estimates produced by the ASCM and COUPLE propagation models for ario depicted in (a) 80 Range (km) 0 0.5 1 1.5 2 2.5 3 3.5 4 -350 (a) Range (km) 0.5 1 1.5 2 2.5 3 3.5 4 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 f = 25 Hz, β = 2.50◦, Fxd = 60.00 m, Mvd = 160.00 m COUPLE ASCM (b) Figure 4.2: (a) Downslope propagation scenario (b) Comparison of transmission loss estimates produced by the ASCM and COUPLE propagation models for the scenario depicted in (a) 80 Range (km) 0.5 1 1.5 2 2.5 3 3.5 4 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 f = 25 Hz, β = 2.50◦, Fxd = 60.00 m, Mvd = 160.00 m COUPLE ASCM (b) f = 25 Hz, β = 2.50◦, Fxd = 60.00 m, Mvd = 160.00 m (b) Figure 4.2: (a) Downslope propagation scenario (b) Comparison of transmission loss estimates produced by the ASCM and COUPLE propagation models for the scenario depicted in (a) 80 80 Range (km) 0 1 2 3 4 5 6 Depth (m) -200 -180 -160 -140 -120 -100 -80 -60 -40 -20 0 f = 25 Hz, β1 = -2.00◦, β2 = -0.50◦, Fxd = 60.00 m, Mvd = 30.00 m (a) Range (km) 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 f = 25 Hz, β1 = -2.00◦, β2 = -0.50◦, Fxd = 60.00 m, Mvd = 30.00 m COUPLE ASCM (b) igure 4.3: (a) Two-slope uphill propagation scenario (b) Comparison of transm on loss estimates produced by the ASCM and COUPLE propagation models he scenario depicted in (a) 81 Range (km) 0 1 2 3 4 5 6 Depth (m) -200 -180 -160 -140 -120 -100 -80 -60 -40 -20 0 f = 25 Hz, β1 = -2.00◦, β2 = -0.50◦, Fxd = 60.00 m, Mvd = 30.00 m (a) Range (km) 0 1 2 3 4 5 6 -200 (a) Range (km) 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 f = 25 Hz, β1 = -2.00◦, β2 = -0.50◦, Fxd = 60.00 m, Mvd = 30.00 m COUPLE ASCM (b) Figure 4.3: (a) Two-slope uphill propagation scenario (b) Comparison of transmis- sion loss estimates produced by the ASCM and COUPLE propagation models for the scenario depicted in (a) 81 Range (km) 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 f = 25 Hz, β1 = -2.00◦, β2 = -0.50◦, Fxd = 60.00 m, Mvd = 30.00 m COUPLE ASCM (b) (b) Figure 4.3: (a) Two-slope uphill propagation scenario (b) Comparison of transmis- sion loss estimates produced by the ASCM and COUPLE propagation models for the scenario depicted in (a) 81 81 Range (km) 0 1 2 3 4 5 6 Depth (m) -400 -350 -300 -250 -200 -150 -100 -50 0 f = 25 Hz, β1 = 0.50◦, β2 = 2.50◦, Fxd = 100.00 m, Mvd = 170.00 m (a) Range (km) 0 1 2 3 4 5 6 Depth (m) -400 -350 -300 -250 -200 -150 -100 -50 0 f = 25 Hz, β1 = 0.50◦, β2 = 2.50◦, Fxd = 100.00 m, Mvd = 170.00 m (a) Range (km) 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 f = 25 Hz, β1 = 0.50◦, β2 = 2.50◦, Fxd = 100.00 m, Mvd = 170.00 m COUPLE ASCM (b) re 4.4: (a) Two-slope downhill propagation scenario (b) Comparison of tra ion loss estimates produced by the ASCM and COUPLE propagation mod he scenario depicted in (a) 82 Range (km) 0 1 2 3 4 5 6 -400 -350 -300 (a) Range (km) 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 f = 25 Hz, β1 = 0.50◦, β2 = 2.50◦, Fxd = 100.00 m, Mvd = 170.00 m COUPLE ASCM (b) Figure 4.4: (a) Two-slope downhill propagation scenario (b) Comparison of trans- mission loss estimates produced by the ASCM and COUPLE propagation models for the scenario depicted in (a) 82 Range (km) 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 f = 25 Hz, β1 = 0.50◦, β2 = 2.50◦, Fxd = 100.00 m, Mvd = 170.00 m COUPLE ASCM (b) (b) Figure 4.4: (a) Two-slope downhill propagation scenario (b) Comparison of trans- mission loss estimates produced by the ASCM and COUPLE propagation models for the scenario depicted in (a) 82 82 Range (km) 0 1 2 3 4 5 6 Depth (m) -200 -180 -160 -140 -120 -100 -80 -60 -40 -20 0 f = 25 Hz, β1 = -2.50◦, β2 = 2.50◦, Fxd = 60.00 m, Mvd = 40.00 m (a) Range (km) 0 1 2 3 4 5 6 Depth (m) -200 -180 -160 -140 -120 -100 -80 -60 -40 -20 0 f = 25 Hz, β1 = -2.50◦, β2 = 2.50◦, Fxd = 60.00 m, Mvd = 40.00 m (a) Range (km) 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 f = 25 Hz, β1 = -2.50◦, β2 = 2.50◦, Fxd = 60.00 m, Mvd = 40.00 m COUPLE ASCM (b) igure 4.5: (a) 2D seamount propagation scenario (b) Comparison of transmissi ss estimates produced by the ASCM and COUPLE propagation models for t enario depicted in (a) 83 Range (km) 0 1 2 3 4 5 6 -200 (a) Range (km) 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 f = 25 Hz, β1 = -2.50◦, β2 = 2.50◦, Fxd = 60.00 m, Mvd = 40.00 m COUPLE ASCM (b) Figure 4.5: (a) 2D seamount propagation scenario (b) Comparison of transmission loss estimates produced by the ASCM and COUPLE propagation models for the scenario depicted in (a) 83 Range (km) 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 f = 25 Hz, β1 = -2.50◦, β2 = 2.50◦, Fxd = 60.00 m, Mvd = 40.00 m COUPLE ASCM (b) (b) Figure 4.5: (a) 2D seamount propagation scenario (b) Comparison of transmission loss estimates produced by the ASCM and COUPLE propagation models for the scenario depicted in (a) 83 83 f = 25 Hz, β1 = 2.50◦, β2 = -2.50◦, Fxd = 100.00 m, Mvd = 150.00 m Range (km) 0 1 2 3 4 5 6 Depth (m) -300 -250 -200 -150 -100 -50 0 f = 25 Hz, β1 = 2.50◦, β2 = -2.50◦, Fxd = 100.00 m, Mvd = 150.00 m (a) Range (km) 0 1 2 3 4 5 6 Depth (m) -300 -250 -200 -150 -100 -50 0 f = 25 Hz, β1 = 2.50◦, β2 = -2.50◦, Fxd = 100.00 m, Mvd = 150.00 m (a) Range (km) 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 f = 25 Hz, β1 = 2.50◦, β2 = -2.50◦, Fxd = 100.00 m, Mvd = 150.00 m COUPLE ASCM (b) igure 4.6: (a) 2D canyon propagation scenario (b) Comparison of transmissi ss estimates produced by the ASCM and COUPLE propagation models for t cenario depicted in (a) 84 Range (km) 0 1 2 3 4 5 6 -300 -250 (a) Range (km) 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 f = 25 Hz, β1 = 2.50◦, β2 = -2.50◦, Fxd = 100.00 m, Mvd = 150.00 m COUPLE ASCM (b) Figure 4.6: (a) 2D canyon propagation scenario (b) Comparison of transmission loss estimates produced by the ASCM and COUPLE propagation models for the scenario depicted in (a) 84 Range (km) 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 f = 25 Hz, β1 = 2.50◦, β2 = -2.50◦, Fxd = 100.00 m, Mvd = 150.00 m COUPLE ASCM (b) (b) Figure 4.6: (a) 2D canyon propagation scenario (b) Comparison of transmission loss estimates produced by the ASCM and COUPLE propagation models for the scenario depicted in (a) 84 84 Range (km) 0 1 2 3 4 5 6 Depth (m) -200 -180 -160 -140 -120 -100 -80 -60 -40 -20 0 Multi-slope: f = 25 Hz, Fxd = 60.00 m, Mvd = 40.00 m (a) Range (km) 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 Multi-slope: f = 25 Hz, Fxd = 60.00 m, Mvd = 40.00 m COUPLE ASCM (b) gure 4.7: (a) General multi-slope propagation scenario (b) Comparison of tra ission loss estimates produced by the ASCM and COUPLE propagation mod r the scenario depicted in (a) 85 Multi-slope: f = 25 Hz, Fxd = 60.00 m, Mvd = 40.00 m Range (km) 0 1 2 3 4 5 6 Depth (m) -200 -180 -160 -140 -120 -100 -80 -60 -40 -20 0 Multi-slope: f = 25 Hz, Fxd = 60.00 m, Mvd = 40.00 m (a) Range (km) 0 1 2 3 4 5 6 -200 -180 -160 -140 (a) Range (km) 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 Multi-slope: f = 25 Hz, Fxd = 60.00 m, Mvd = 40.00 m COUPLE ASCM (b) Figure 4.7: (a) General multi-slope propagation scenario (b) Comparison of trans- mission loss estimates produced by the ASCM and COUPLE propagation models for the scenario depicted in (a) 85 Range (km) 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 TL (dB) 20 30 40 50 60 70 80 90 100 110 120 Multi-slope: f = 25 Hz, Fxd = 60.00 m, Mvd = 40.00 m COUPLE ASCM (b) (b) Figure 4.7: (a) General multi-slope propagation scenario (b) Comparison of trans- mission loss estimates produced by the ASCM and COUPLE propagation models for the scenario depicted in (a) 85 85 4.3 Comparison to Parabolic Equation Solution While the strong peak-to-peak agreement achieved with the benchmark so- lution is suﬃcient for verifying the feasibility of the proposed coupled-mode ap- proach, comparisons with solutions not based on normal mode theory were also pursued. One such comparison was made with the Monterey-Miami Parabolic Equation (MMPE) model developed by Smith [4]. The MMPE solution uses a split-step Fourier algorithm to solve the parabolic wave equation. This algorithm involves decomposing the acoustic ﬁeld into a slowly modulating envelope function, known as the PE ﬁeld function, and a phase term, which oscillates at the acoustic frequency of the source. The use of recurrent Fast Fourier Transforms in the MMPE solution necessi- tates approximating the transition over the water-sediment interface with a smooth function. As such, the MMPE model did not allow for the representation of a perfectly-rigid bottom boundary condition as used in the test cases described in Section (4.2). To get around this limitation the source frequency was increased from 25 Hz to 200 Hz, which greatly reduced, but did not eliminate, the loss oﬀ acoustic energy to the sub-basement. Despite this diﬀerence in bottom boundary conditions, appreciable agreement was obtained between the benchmark (COU- PLE), ASCM, and MMPE solutions for this scenario. Figures (4.8a)-(4.8d) pro- vide a three-way comparison of the transmission loss estimates produced by the three solutions for the 2D seamount test case with a 200 Hz source. Figure (4.9) provides a plot of the MMPE-predicted complex pressure ﬁeld, which shows energy losses in the MMPE pseudo-rigid basement. 86 Range (km) 0 1 2 3 4 5 6 Depth (m) -200 -180 -160 -140 -120 -100 -80 -60 -40 -20 0 f = 200 Hz, β1 = -2.50◦, β2 = 2.50◦, Fxd = 60.00 m, Mvd = 40.00 m (a) High Frequency 2D Seamount Propagation Scenario Range (km) 0 1 2 3 4 5 6 TL (dB) 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 f = 200 Hz, β1 = -2.50◦, β2 = 2.50◦, Fxd = 60.00 m, Mvd = 40.00 m COUPLE ASCM (b) ASCM vs. 4.3 Comparison to Parabolic Equation Solution COUPLE Transmission Loss Comparison Range (km) 0 1 2 3 4 5 6 Depth (m) -200 -180 -160 -140 -120 -100 -80 -60 -40 -20 0 f = 200 Hz, β1 = -2.50◦, β2 = 2.50◦, Fxd = 60.00 m, Mvd = 40.00 m (a) High Frequency 2D Seamount Propagation Scenario (a) High Frequency 2D Seamount Propagation Scenario Range (km) 0 1 2 3 4 5 6 TL (dB) 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 f = 200 Hz, β1 = -2.50◦, β2 = 2.50◦, Fxd = 60.00 m, Mvd = 40.00 m COUPLE ASCM ( ) A (b) ASCM vs. COUPLE Transmission Loss Comparison 87 Range (km) 0 1 2 3 4 5 6 TL (dB) 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 f = 200 Hz, β1 = -2.50◦, β2 = 2.50◦, Fxd = 60.00 m, Mvd = 40.00 m COUPLE MMPE (c) MMPE vs. COUPLE Transmission Loss Comparison Range (km) 0 1 2 3 4 5 6 TL (dB) 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 f = 200 Hz, β1 = -2.50◦, β2 = 2.50◦, Fxd = 60.00 m, Mvd = 40.00 m ASCM MMPE (d) MMPE vs. ASCM Transmission Loss Comparison e 4.8: Comparison of transmission loss estimates between COUPLE, ASC MMPE propagation models for the high frequency 2D seamount scenario Range (km) 0 1 2 3 4 5 6 TL (dB) 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 f = 200 Hz, β1 = -2.50◦, β2 = 2.50◦, Fxd = 60.00 m, Mvd = 40.00 m COUPLE MMPE (c) MMPE vs. COUPLE Transmission Loss Comparison (c) MMPE vs. COUPLE Transmission Loss Comparison Range (km) 0 1 2 3 4 5 6 TL (dB) 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 f = 200 Hz, β1 = -2.50◦, β2 = 2.50◦, Fxd = 60.00 m, Mvd = 40.00 m ASCM MMPE (d) MMPE vs. 4.3 Comparison to Parabolic Equation Solution ASCM Transmission Loss Comparison Figure 4.8: Comparison of transmission loss estimates between COUPLE, ASCM, and MMPE propagation models for the high frequency 2D seamount scenario Range (km) 0 1 2 3 4 5 6 TL (dB) 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 f = 200 Hz, β1 = -2.50◦, β2 = 2.50◦, Fxd = 60.00 m, Mvd = 40.00 m ASCM MMPE f = 200 Hz, β1 = -2.50◦, β2 = 2.50◦, Fxd = 60.00 m, Mvd = 40.00 m (d) MMPE vs. ASCM Transmission Loss Comparison Figure 4.8: Comparison of transmission loss estimates between COUPLE, ASCM, and MMPE propagation models for the high frequency 2D seamount scenario Figure 4.8: Comparison of transmission loss estimates between COUPLE, ASCM, and MMPE propagation models for the high frequency 2D seamount scenario Figure 4.8: Comparison of transmission loss estimates between COUPLE, ASCM, and MMPE propagation models for the high frequency 2D seamount scenario 88 Transmission Loss (dB re 1m) for Pressure Range (km) 1 2 3 4 5 6 Depth (m) 0 100 200 300 400 500 600 10 20 30 40 50 60 70 80 90 100 Fi 4 9 MMPE di d l ﬁld h i l i h d Transmission Loss (dB re 1m) for Pressure Range (km) 1 2 3 4 5 6 Depth (m) 0 100 200 300 400 500 600 10 20 30 40 50 60 70 80 90 100 Figure 4.9: MMPE-predicted complex pressure ﬁeld showing losses in the pseudo- rigid basement Transmission Loss (dB re 1m) for Pressure Figure 4.9: MMPE-predicted complex pressure ﬁeld showing losses in the pseudo- rigid basement 89 4.4 Comparison to 3D Finite Element Solution As a ﬁnal means of veriﬁcation of the proposed coupled-mode solution, a comparison was made to a three-dimensional Finite Element Model (FEM) con- structed by the Applied Research Laboratories of the University of Texas at Austin (ARL-UT). This model utilizes out-of-plane wavenumber decomposition techniques, which involve a series of two-dimensional models to calculate the fully three-dimensional acoustic ﬁeld for longitudinally invariant environments [5, 6]. The geometry of the FEM-modeled environment consisted of an underwater sea ridge, which was constructed by extending the two-dimensional seamount scenario depicted in Figure (4.5) inﬁnitely in a second horizontal dimension. As such, com- parison to the FEM was achieved by taking a two-dimensional slice in-plane with the acoustic source. Figures (4.10a)-(4.10b) provide a comparison of transmis- sion loss estimates produced by the two-dimensional (i.e. COUPLE and ASCM) solutions and the three-dimensional FEM. Strong agreement was found between all three propagation solutions. The ASCM and FEM solutions showed a stronger peak-to-peak agreement in the vicin- ity of mode cutoﬀlocations. Conversely, the ASCM and COUPLE solutions showed stronger agreement at end range. Based on communication with investigators at ARL-UT, this disagreement at +4.5 km was suggested to be due to a diﬀerence in the predicted mode 1 energy on the downslope half of the ridge. Diﬀerences between the solutions may also be attributed to three-dimensional modal diﬀrac- tion and the backwards traveling ﬁeld, neither of which were included in the two- dimensional solutions. Despite these diﬀerences, the presented results suggest that the ASCM solution provides an accurate estimate of the acoustic ﬁeld in three- dimensional environments with signiﬁcant bathymetric variation. 90 Range (km) 0 1 2 3 4 5 6 TL (dB) 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 f = 25 Hz, β1 = -2.50◦, β2 = 2.50◦, Fxd = 60.00 m, Mvd = 40.00 m FEM COUPLE Range (km) 0 1 2 3 4 5 6 TL (dB) 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 f = 25 Hz, β1 = -2.50◦, β2 = 2.50◦, Fxd = 60.00 m, Mvd = 40.00 m FEM COUPLE (a) COUPLE vs. 4.4 Comparison to 3D Finite Element Solution FEM Transmission Loss Comparison Range (km) 0 1 2 3 4 5 6 TL (dB) 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 f = 25 Hz, β1 = -2.50◦, β2 = 2.50◦, Fxd = 60.00 m, Mvd = 40.00 m FEM ASCM (b) ASCM vs. FEM Transmission Loss Comparison e 4.10: Comparison of 2D and 3D solutions for in-plane transmission a three-dimensional underwater sea ridge (a) COUPLE vs. FEM Transmission Loss Comparison Range (km) 0 1 2 3 4 5 6 TL (dB) 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 f = 25 Hz, β1 = -2.50◦, β2 = 2.50◦, Fxd = 60.00 m, Mvd = 40.00 m FEM ASCM (b) ASCM vs. FEM Transmission Loss Comparison Figure 4.10: Comparison of 2D and 3D solutions for in-plane transmission loss over a three-dimensional underwater sea ridge Range (km) 0 1 2 3 4 5 6 TL (dB) 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 f = 25 Hz, β1 = -2.50◦, β2 = 2.50◦, Fxd = 60.00 m, Mvd = 40.00 m FEM ASCM f = 25 Hz, β1 = -2.50◦, β2 = 2.50◦, Fxd = 60.00 m, Mvd = 40.00 m (b) ASCM vs. FEM Transmission Loss Comparison Figure 4.10: Comparison of 2D and 3D solutions for in-plane transmission loss over a three-dimensional underwater sea ridge Figure 4.10: Comparison of 2D and 3D solutions for in-plane transmission loss over a three-dimensional underwater sea ridge 91 4.5 Discussion Strong agreement was found between the transmission loss curves produced by the ASCM and COUPLE models for all seven cases depicted in Figures (4.1)- (4.7). Small discrepancies can be found between the two solutions for the upslope propagation scenario depicted in Figure (4.1). These discrepancies correspond to areas where the transmission coeﬃcients in the ASCM calculation have been corrected for mode turnaround points as discussed in Section (3.8). This same disagreement is found in the ﬁrst half of the transmission loss curve for the 2D seamount scenario depicted in Figure (4.5). This is due to the fact that the scenario in Figure (4.5) was created by replicating and reﬂecting the bottom slope for the upslope scenario. Some disagreement between the two models is also found in the vicinity of the valley of the 2D canyon case depicted in Figure (4.6). This disagreement is again contributed to the mode turnaround issue discussed in Section (3.8). In the case of the 2D canyon, severe mode stripping occurs from 2.5 km to 3.5 km. Over this range interval higher order modes transition into a propagating state as acoustic energy propagates downslope and the increased waveguide depth moves away from the cutoﬀdepths of these modes. These newly propagating modes are, however, quickly cutoﬀagain as the energy radiates out of the valley of the canyon and up the second slope. The results depicted in Figure(4.7) for the general multi-slope propagation scenario oﬀers particular promise for the practical application of the presented theory. In this scenario strong agreement is found between the ASCM and COU- PLE solutions for a highly variable bottom depth proﬁle over a large range interval. These results support the previous conjecture that any bottom depth proﬁle may be approximated by a series of connected wedge subenvironments. Moreover, the 92 asymptotic solution (i.e. ASCM) is seen to perform similarly to the full-integral benchmark solution (i.e. COUPLE) for this more complicated environment. This would suggest that the ASCM solution has the potential to oﬀer comparable accu- racy to the benchmark solution at less computational cost. Strong agreement with the Parabolic Equation and three-dimensional Finite Element Model solutions dis- cussed in Sections (4.3) and (4.4) further support the eﬃcacy of the presented asymptotic stepwise coupled-mode approach. List of References [1] MATLAB, version 8.4.0 (R2014b). Natick, Massachusetts: The MathWorks Inc., 2014. [2] R. B. Evans, “A coupled mode solution for acoustic propagation in a waveg- uide with stepwise depth variations of a penetrable bottom,” Journal of the Acoustical Society of America, vol. 74, no. 1, pp. 188–195, 1983. [3] F. B. Jensen, W. A. Kuperman, M. B. Porter, and H. Schmidt, Computational Ocean Acoustics, 1st ed. New York: Springer-Verlag, 2000. [4] K. B. Smith, “Convergence, stability, and variability of shallow water acoustic predictions using a split-step fourier parabolic equation model,” Journal of Computational Acoustics, vol. 9, no. 1, pp. 243–285, 2001. [5] M. J. Isakson and N. P. Chotiros, “Finite element modeling of reverberation and transmission loss in shallow water waveguides with rough boundaries,” Journal of the Acoustical Society of America, vol. 129, no. 3, pp. 1273–1279, 2011. [6] M. J. Isakson, B. Goldsberry, and N. P. Chotiros, “A three-dimensional, longitudinally-invariant ﬁnite element model for acoustic propagation in shal- low water waveguides,” Journal of the Acoustical Society of America, vol. 136, no. 3, pp. EL206–EL211, 2014. 93 Conclusion An asymptotic theory for estimating non-adiabatic acoustic mode coupling in horizontally-variable shallow water environments has been presented. This theory is based on a novel discretization of the range-dependent acoustic medium, which allows for the computation of mode coupling via a range-expanded normal mode inner product. Energy transfer due to an acoustic excitation is accounted for via a diﬀerence equations approach that provides asymptotic solutions for transmission and reﬂection coeﬃcients at interfaces throughout the discretized acoustic environ- ment. This approach is unique in that coupling across these interfaces is calculated via closed-form algebraic expressions rather than inner product integrals. More- over, the novel discretization approach associated with this method eliminates the eigenvalue search problem. A computational model was developed in MATLAB [1] that implements this asymptotic stepwise coupled mode theory. Transmission loss curves produced by this model for several test cases involving environments with range-dependent bot- tom depth were compared to those produced by a benchmark model for range- dependent problems developed by Evans [2]. Strong agreement was found between the two models for all cases considered. For cases in which discrepancies existed between the two solutions, these diﬀerences were attributed to an eigenvalue lag distance that is inherent to the stepwise approach of the asymptotic solution in the vicinity of mode turnaround points, as discussed in Section (3.8). Additional ver- iﬁcation of the proposed theory was obtained through comparison to a Parabolic Equation solution in Sections (4.3) and to a three-dimensional Finite Element Model in Section (4.4). 94 As discussed in Section (3.3), application of the discretization technique as- sociated with the presented theory to general environments necessitates thought. This is due to the fact that one must construct a set of subenvironments that capture the range-dependence of the general environment via closed-form expres- sions for the eigenfunctions and eigenvalues associated with the acoustic modes of vibration of the medium. As shown in Section (3.3), range-dependence due to horizontally-variable bottom depth may be approximated via a series of acoustic wedges for which the desired closed-form expressions for the eigenfunctions and eigenvalues have been stated. It is plausible that range-dependence due to sound speed could be captured in a similar fashion by constraining each wedge suben- vironment to ranges of constant gradient in the sound speed proﬁle. Conclusion This would, however, require further analysis including re-derivation of the closed-form expres- sions for the terms of the range-expanded normal mode inner product given by Equations (3.21)-(3.26). The inclusion of range-dependent sound speed into the current model would allow for the modeling of more realistic environments and the ability to make meaningful runtime comparisons with other solutions. Despite the omission of these types of comparisons in the present work, it is the opinion of the author that appreciable gains in runtime performance will be achieved due to the ability to compute mode coupling via the evaluation of closed-form algebraic expressions instead of inner product integrals. This is particularly evident if one considers the number of range-independent steps necessary to accurately approximate a general range-dependent environment. As non-adiabatic mode coupling must be accounted for at the interfaces separating each of these steps, the multiplicity of the afore- mentioned coupling calculation is signiﬁcant. There is room for improvement in the handling of the mode turnaround issue 95 discussed in Section (3.8). While the proposed moving average solution success- fully eliminates the artiﬁcial peaks introduced into the transmission loss curves associated with this issue, the largest discrepancies between the asymptotic and benchmark models occurred at ranges associated with mode cutoﬀ. One potential improvement on this issue may be to simply apply a more sophisticated smoothing ﬁlter to the eigenvalue value ratio appearing in the transmission coeﬃcient matrix in Equation (3.81). Another approach would be to develop an adaptive range-step algorithm that reduces range step size in the propagation code in the neighborhood of mode cutoﬀdepths. This would eﬀectively reduce the eigenvalue lag distance discussed in Section (3.8) and therefore potentially reduce or eliminate the discon- tinuities in the transmission coeﬃcients associated with mode turnaround. [2] R. B. Evans, “A coupled mode solution for acoustic propagation in a waveg- uide with stepwise depth variations of a penetrable bottom,” Journal of the Acoustical Society of America, vol. 74, no. 1, pp. 188–195, 1983. List of References [1] MATLAB, version 8.4.0 (R2014b). Natick, Massachusetts: The MathWorks Inc., 2014. [2] R. B. Evans, “A coupled mode solution for acoustic propagation in a waveg- uide with stepwise depth variations of a penetrable bottom,” Journal of the Acoustical Society of America, vol. 74, no. 1, pp. 188–195, 1983. [2] R. B. Evans, “A coupled mode solution for acoustic propagation in a waveg- uide with stepwise depth variations of a penetrable bottom,” Journal of the Acoustical Society of America, vol. 74, no. 1, pp. 188–195, 1983. 96 APPENDIX APPENDIX Note on the Eﬀects of Range-Dependent Bathymetry on Modal Group Velocities Note on the Eﬀects of Range-Dependent Bathymetry on Modal Group Velocities Mode coupling is a process that aﬀects the amplitudes of the eigenfunctions associated with the normal modes of propagation of an acoustic waveguide. As such, it does not have a direct impact on modal phase or group velocities, which are functions of the eigenvalues rather than the eigenfunctions. This said, both mode coupling and perturbations in modal velocities originate due to environmen- tal range-dependence. As such, these two physical phenomena share a common source and are equally important in the understanding of wave propagation in horizontally-variable acoustic media. While conducting the research and development of the presented coupled- mode theory and associated model a numerical study was performed on the eﬀects of horizontally-variable bottom depth on modal group velocities. To perform this study, numerical models were constructed for three isovelocity waveguides: a ﬂat- bottom waveguide, an acoustic wedge with a bottom slope angle of −1◦, and an acoustic wedge with a bottom slope angle of −2◦. The sound speed in each waveguide was chosen to be 1500 m s−1 and the maximum waveguide depth was chosen to be 200 m. To assess the relationship between frequency and group speed an evenly-spaced frequency band from 40 Hz to 200 Hz was generated using a 5 Hz frequency spacing. Group velocities were computed for all modes and all frequencies at 1 m range increments across a distance of 3 km in each waveguide. The mean of these samples was then calculated, providing one range-averaged group velocity per mode, per frequency for each of the three waveguides. Note that for a given mode in the wedge waveguides, range increments at which the mode became cutoﬀand thus its 97 eigenvalue became purely imaginary were not included in the averaging process. Figures (A.1)-(A.3) provide depictions of the range-averaged group velocities and grazing angles for modes 1, 3, and 7 for the three sample waveguides. As is evident from the ﬁgures, increasing the bottom slope (i.e. reducing the waveguide depth with range) results in a reduction in modal group speeds across all frequencies. This “slowing down” of the modal envelope is most pronounced at lower frequencies and smallest at higher frequencies. This phenomenon can also be connected to mode grazing angle. That is, increasing the bottom slope of the waveguide increases the mode grazing angles at all frequencies. Note on the Eﬀects of Range-Dependent Bathymetry on Modal Group Velocities As is the case for group velocity, the perturbation in mode grazing angle is most pronounced at lower frequencies and least pronounced at higher frequencies. If one considers the relationship between frequency, waveguide depth, and mode cutoﬀ, the eﬀects illustrated in Figures (A.1)-(A.3) suggest that the closer a mode comes to its cutoﬀ conditions (i.e. steep-angle and imaginary eigenvalue) the more “slowed down” it becomes and therefore the greater the perturbation in its group speed relative to the ﬂat-bottom case. The ﬁndings of this experiment are particularly signiﬁcant for shallow wa- ter applications of mode theory in which modal velocities are considered at low frequencies. For example, models involving group velocity are often used in geoa- coustic inversion applications to estimate sediment properties. In these applica- tions it is desirable to excite the normal modes of the acoustic waveguide at steep angles (i.e. at low frequencies near modal cutoﬀs) to maximize sediment pene- tration. For ﬂat or nearly ﬂat waveguides it has been shown that the eﬀects of horizontally-variable sound speed on geoacoustic inversion is negligible as the per- turbation in modal group velocities due to this variability is most pronounced at higher frequencies [1]. As such, the assumption of range-independence is often 98 made in these applications. The results shown here suggest that if the waveguide has signiﬁcant bathymetric variability, the perturbation in modal group velocity is most pronounced in the vicinity of mode cutoﬀand therefore the assumption of range-independence could introduce considerable error into the inversion process. Note on the Eﬀects of Range-Dependent Bathymetry on Modal Group Velocities 99 Group velocity (m/s) 1497 1497.5 1498 1498.5 1499 1499.5 1500 Frequency (Hz) 40 60 80 100 120 140 160 180 200 Mode 1 Group Velocity β = 0◦ β = −1◦ β = −2◦ Mode 1 Group Velocity Group velocity (m/s) 1497 1497.5 1498 1498.5 1499 1499.5 1500 Frequency (Hz) 40 60 80 100 120 140 160 180 200 Mode 1 Group Velocity β = 0◦ β = −1◦ β = −2◦ (a) Grazing angle (degrees) 0.5 1 1.5 2 2.5 3 3.5 4 Frequency (Hz) 40 60 80 100 120 140 160 180 200 Mode 1 Grazing Angle β = 0◦ β = −1◦ β = −2◦ (b) igure A.1: Comparison of mode 1 (a) group velocity and (b) grazing angle acr hree waveguides with varied bottom slope angle Grazing angle (degrees) 0.5 1 1.5 2 2.5 3 3.5 4 Frequency (Hz) 40 60 80 100 120 140 160 180 200 g g β = 0◦ β = −1◦ β = −2◦ (b) (b) Figure A.1: Comparison of mode 1 (a) group velocity and (b) grazing angle across three waveguides with varied bottom slope angle 100 Group velocity (m/s) 1420 1430 1440 1450 1460 1470 1480 1490 1500 Frequency (Hz) 40 60 80 100 120 140 160 180 200 Mode 3 Group Velocity β = 0◦ β = −1◦ β = −2◦ (a) Grazing angle (degrees) 2 4 6 8 10 12 14 16 18 20 Frequency (Hz) 40 60 80 100 120 140 160 180 200 Mode 3 Grazing Angle β = 0◦ β = −1◦ β = −2◦ (b) igure A.2: Comparison of mode 3 (a) group velocity and (b) grazing angle acr hree waveguides with varied bottom slope angle Group velocity (m/s) 1420 1430 1440 1450 1460 1470 1480 1490 1500 Frequency (Hz) 40 60 80 100 120 140 160 180 200 Mode 3 Group Velocity β = 0◦ β = −1◦ β = −2◦ Mode 3 Group Velocity Grazing angle (degrees) 2 4 6 8 10 12 14 16 18 20 Frequency (Hz) 40 60 80 100 120 140 160 180 200 Mode 3 Grazing Angle β = 0◦ β = −1◦ β = −2◦ (b) (b) Figure A.2: Comparison of mode 3 (a) group velocity and (b) grazing angle across three waveguides with varied bottom slope angle 101 Group velocity (m/s) 1000 1050 1100 1150 1200 1250 1300 1350 1400 1450 1500 Frequency (Hz) 40 60 80 100 120 140 160 180 200 Mode 7 Group Velocity β = 0◦ β = −1◦ β = −2◦ (a) Grazing angle (degrees) 5 10 15 20 25 30 35 40 45 50 55 Frequency (Hz) 40 60 80 100 120 140 160 180 200 Mode 7 Grazing Angle β = 0◦ β = −1◦ β = −2◦ (b) igure A.3: Comparison of mode 7 (a) group velocity and (b) grazing angle acr hree waveguides with varied bottom slope angle Group velocity (m/s) 1000 1050 1100 1150 1200 1250 1300 1350 1400 1450 1500 Frequency (Hz) 40 60 80 100 120 140 160 180 200 Mode 7 Group Velocity β = 0◦ β = −1◦ β = −2◦ Mode 7 Group Velocity Grazing angle (degrees) 5 10 15 20 25 30 35 40 45 50 55 Frequency (Hz) 40 60 80 100 120 140 160 180 200 Mode 7 Grazing Angle β = 0◦ β = −1◦ β = −2◦ (b) (b) Figure A.3: Comparison of mode 7 (a) group velocity and (b) grazing angle across three waveguides with varied bottom slope angle 102 List of References [1] G. R. Potty, J. H. Miller, J. F. Lynch, and K. B. Smith, “Tomographic inversion for sediment parameters in shallow water,” Journal of the Acoustical Society of America, vol. 108, no. 3, pp. 973–986, 2000. List of References List of References List of References [1] G. R. Potty, J. H. Miller, J. F. Lynch, and K. B. Smith, “Tomographic inversion for sediment parameters in shallow water,” Journal of the Acoustical Society of America, vol. 108, no. 3, pp. 973–986, 2000. [1] G. R. Potty, J. H. Miller, J. F. Lynch, and K. B. Smith, “Tomographic inversion for sediment parameters in shallow water,” Journal of the Acoustical Society of America, vol. 108, no. 3, pp. 973–986, 2000. 103 103 BIBLIOGRAPHY Boyles, C. A., “Coupled mode solution for a cylindrically symmetric oceanic waveg- uide with a range and depth dependent refractive index and a time varying rough sea surface,” Journal of the Acoustical Society of America, vol. 73, no. 3, pp. 800–805, 1983. Boyles, C. A., Acoustic Waveguides: Applications to Oceanic Science. New York: Wiley-Interscience, 1984. Brown, J. W. and Churchill, R. V., Complex Variables and Applications, 7th ed. New York: McGraw-Hill, 2003. Buckingham, M. J., “Theory of three-dimensional acoustic propagation in a wedge- like ocean with a penetrable bottom,” Journal of the Acoustical Society of America, vol. 82, no. 1, pp. 198–210, 1987. Chiu, C.-S., Miller, J. H., and Lynch, J. F., “Forward coupled-mode propagation modeling for coastal acoustic tomography,” Journal of the Acoustical Society of America, vol. 99, no. 2, pp. 793–802, 1996. Clark, C. A., “Acoustic wave propagation in horizontally variable media,” IEEE Journal of Oceanic Engineering, vol. 30, no. 1, pp. 188–197, 2005. Clark, C. A. and Smith, K. B., “An eﬃcient normal mode solution to wave prop- agation prediction,” IEEE Journal of Oceanic Engineering, vol. 33, no. 4, pp. 462–476, 2008. de Brujin, N. G., Asymptotic Methods in Analysis. New York: Dover Publications, 1981. Desaubies, Y., Chiu, C.-S., and Miller, J. H., “Acoustic mode propagation in a range-dependent ocean,” Journal of the Acoustical Society of America, vol. 80, no. 4, pp. 1148–1160, 1986. Evans, R. B., “A coupled mode solution for acoustic propagation in a waveguide with stepwise depth variations of a penetrable bottom,” Journal of the Acous- tical Society of America, vol. 74, no. 1, pp. 188–195, 1983. Fawcett, J. A., Westwood, E. K., and Tindle, C. T., “A simple coupled wedge mode propagation method,” Journal of the Acoustical Society of America, vol. 98, no. 3, pp. 1673–1681, 1995. Frisk, G. V., Ocean and Seabed Acoustics: A Theory of Wave Propagation. New Jersey: Prentice-Hall, 1994. 104 Gillette, G., “Coupled modes in a waveguide with a range-dependent rigid base- ment,” Journal of the Acoustical Society of America, vol. 95, no. 1, pp. 187– 200, 1994. Godin, O. A., “A note on diﬀerential equations of coupled-mode propagation in ﬂuids,” Journal of the Acoustical Society of America, vol. 103, no. 1, pp. 159– 168, 1998. Hardin, R. H. and Tappert, F. D., “Applications of the split-step fourier method to the numerical solution of nonlinear and variable coeﬃcient wave equations,” SIAM Rev., vol. 15, p. 423, 1973. Higham, C. BIBLIOGRAPHY J. and Tindle, C. T., “Coupled perturbed modes and internal solitary waves,” Journal of the Acoustical Society of America, vol. 113, no. 5, pp. 2515–2522, 2003. Higham, C. J. and Tindle, C. T., “Coupled perturbed modes over a sloping pene- trable bottom,” Journal of the Acoustical Society of America, vol. 114, no. 6, pp. 3119–3124, 2003. Isakson, M. J. and Chotiros, N. P., “Finite element modeling of reverberation and transmission loss in shallow water waveguides with rough boundaries,” Journal of the Acoustical Society of America, vol. 129, no. 3, pp. 1273–1279, 2011. Isakson, M. J., Goldsberry, B., and Chotiros, N. P., “A three-dimensional, longitudinally-invariant ﬁnite element model for acoustic propagation in shal- low water waveguides,” Journal of the Acoustical Society of America, vol. 136, no. 3, pp. EL206–EL211, 2014. Jensen, F. B., Kuperman, W. A., Porter, M. B., and Schmidt, H., Computational Ocean Acoustics, 1st ed. New York: Springer-Verlag, 2000. Kampanis, N. A. and Dougalis, V. A., “A ﬁnite element code for the numerical solution of the Helmholtz equation in axially symmetric waveguides with in- terfaces,” Journal of Computational Acoustics, vol. 7, no. 2, pp. 83–110, 1999. Leibiger, G. A., “Wave propagation in an inhomogeneous medium with slow spatial variation,” Ph.D. dissertation, Stevens Institute of Technology, Castle Point, Hoboken, New Jersey, 1968. MATLAB, version 8.4.0 (R2014b). Natick, Massachusetts: The MathWorks Inc., 2014. Milder, D. M., “Ray and wave invariants for sofar channel propagation,” Journal of the Acoustical Society of America, vol. 46, no. 5, pp. 1259–1263, 1969. 105 Pierce, A. D., “Extension of the method of normal modes to sound propagation in an almost-stratiﬁed medium,” Journal of the Acoustical Society of America, vol. 37, no. 1, pp. 19–27, 1964. Potty, G. R., Miller, J. H., Lynch, J. F., and Smith, K. B., “Tomographic inversion for sediment parameters in shallow water,” Journal of the Acoustical Society of America, vol. 108, no. 3, pp. 973–986, 2000. Primack, H. and Gilbert, K. E., “A two-dimensional downslope propagation model based on coupled wedge modes,” Journal of the Acoustical Society of America, vol. 90, no. 6, pp. 3254–3262, 1991. Smith, K. B., “Convergence, stability, and variability of shallow water acoustic predictions using a split-step fourier parabolic equation model,” Journal of Computational Acoustics, vol. 9, no. 1, pp. 243–285, 2001. Stotts, S. A., “Coupled-mode solutions in generalized ocean environments,” Jour- nal of the Acoustical Society of America, vol. 111, no. 4, pp. 1623–1643, 2002. Tindle, C. BIBLIOGRAPHY T., O’Driscoll, L. M., and Higham, C. J., “Coupled mode perturbation theory of range dependence,” Journal of the Acoustical Society of America, vol. 108, no. 1, pp. 76–83, 2000. Watson, G. N., A Treatise on the Theory of Bessel Functions, 2nd ed. New York: Cambridge University Press, 1995. White, F. M., Fluid Mechanics, 6th ed. New York: McGraw-Hill, 2006. 106
https://openalex.org/W4313573699	https://www.jidhealth.com/index.php/jidhealth/article/download/259/155	English	null	Artificial intelligence-enabled rapid and symptom-based medication recommendation system (COV-MED) for the COVID-19 patients	Journal of ideas in health :	2,022	cc-by	8,153	Artificial intelligence-enabled rapid and symptom-based medication recommendation system (COV-MED) for the COVID-19 patients Abstract In a general COVID-19 population in Cox’s Bazar, Bangladesh, we developed a medication recommendation system based on clinical information from the electronic medical record (EMR). Our goal was also to enable deep learning (DL) strategies to quickly assist physicians and COVID-19 patients by recommending necessary medications. The general demographic data, clinical symptoms, basic clinical tests, and drug information of 8953 patients were used to create a dataset. The learning model in this COVID-MED model was created using Keras (an open-source artificial neural network library) to solve regression problems. In this study, a sequential model was adopted. In order to improve the prediction capability and achieve global minima quickly and smoothly, the COVID-MED model incorporates an adaptive optimizer dubbed Adam. The model calculated a mean absolute error of 0.0037, a mean squared error of 0.000035, and a root mean squared error of 0.0059. The model predicts the output medications, such as injections or other oral medications, with around 99% accuracy. These findings show that medication can be predicted using information from the EMR. Similar models allow for patient-specific decision support to help prevent medication errors in diseases other than COVID-19. Keywords: COVID-19, Electronic Medical Records, Artificial Neural Networks, Machine Learning, Deep Learning, Artificial Intelligence, Bangladesh D-19, Electronic Medical Records, Artificial Neural Networks, Machine Learning, Deep Learning, © The Author(s). 2022 This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (https://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Journal of Ideas in Health https://doi.org/10.47108/jidhealth.Vol5.Iss4.259 https://doi.org/10.47108/jidhealth.Vol5.Iss4.259 Artificial intelligence-enabled rapid and symptom-based medication recommendation system (COV-MED) for the COVID-19 patients Amit Shaha Surja1, Md. Tariqul Islam Limon2, Md. Janibul Alam Soeb3, Md. Shoaib Arifin4, Md. Meftaul Islam5, Md. Shahidullah Kayshar6, Md. Amirul Islam3, Md. Mizanur Rahman3, Md Abdul Malek2, Faroque Md Mohsin2, Mohammed Shah Jahan7, Anupam Barua8, Tanjima Binte Tofaz9, Irin Sultana2, Md. Fahad Jubayer6* jubayer.fet@sau.ac.bd 2Department of Food Engineering and Technology, Faculty of Agricultural Engineering and Technology, Sylhet Agricultural University, Sylhet-3100, Bangladesh. Full list of author information is available at the end of the article Background machine learning (ML) models improve, there is a growing belief that AI will change medical practice, redefining clinicians' responsibilities in the process [1,2,3]. The pandemic coronavirus disease (COVID-19) started affecting the global population in 2020 and is continuing to rise at a rapid pace with constantly changing variants. Though a handful of vaccines have come as of today, there is no permanent remedy yet. People are being hospitalized daily, and many are joining the death rally as a result of COVID-19. Due to the difficulties and sufferings of this novel disease, health professionals, patients, and their families have been compelled to make challenging and crucial decisions quickly and with minimal information. Given the priority of these situations, followed by the growing number of cases, an accurate and reliable tool based on existing healthcare resources is greatly warranted [4]. Advanced technologies such as ML and AI could better understand patient subgroups, drive clinical decision-making, and improve operational and patient-centered outcomes [4]. In consequence, ML algorithms could combine and analyze large-scale data Global healthcare and biomedical research are progressively changing with artificial intelligence (AI) advancement. In fact, medicine was identified before as one of the most potential and promising AI application domains. The application of AI is being extensively used in areas previously regarded as solely the domain of human expertise [1]. Recently, the AI revolution has been fueled largely by the successful application of deep learning systems, which entails training an artificial neural network (ANN) with many layers on massive datasets to large sources of labeled data [1]. AI systems outperform physicians in numerous diagnostic tasks, can estimate patient prognosis better than clinicians and can aid in surgical treatments as well As than clinicians and can aid in surgical treatments as well. As © The Author(s). 2022 This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Public Domain Dedication waiver (https://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless stated. Journal of Ideas in Health of cases, an accurate and reliable tool based on healthcare resources is greatly warranted [4]. Background A technologies such as ML and AI could better understan subgroups, drive clinical decision-making, and operational and patient-centered outcomes [4]. In cons ML algorithms could combine and analyze large-sc ___________________________________________________ Correspondence: jubayer.fet@sau.ac.bd 2Department of Food Engineering and Technology, Faculty of Agricultural Engineering and Technology, Sylhet Agricultural University, Sylhet-3100, Bangladesh. Full list of author information is available at the end of the article © The Author(s). 2022 This article is d (http://creativecommons.org/licenses/by/4.0/), which p credit to the original author(s) and the source, provide Public Domain Dedication waiver (https://creativecom stated. Journal of Ideas in Health ___________________________________________________ Correspondence: jubayer.fet@sau.ac.bd 2Department of Food Engineering and Technology, Faculty of Agricultural Engineering and Technology, Sylhet Agricultural University, Sylhet-3100, Bangladesh. Full list of author information is available at the end of the article Full list of author information is available at the end of the article Surja AS, et al., Journal of Ideas in Health (2022); 5(4):776-785 777 from COVID-19 patients to better comprehend more successful therapeutic approaches and potentially identify the most vulnerable groups based on personalized, genetic, and physiological traits. The substantial amount of COVID-19 treatment data in hospitals worldwide requires the use of advanced ML and DL methods to analyze it to discover and predict new therapeutic options that could provide better care for each patient and contribute to local hospital arrangements and operations [5]. In addition, AI can answer different queries of the patient in the absence of doctors [6]. Aiding the appropriate medications in emergency conditions is critical in fighting this global battle [7]. Projecting the proper medications for the desired patients might alleviate the enormous strain placed on doctors and nurses. In contrast, ML and AI will save time and also provide patients with accurate medication administration instructions. Besides, it is also helpful if lockdowns were implemented and people were required to stay at home and want to take essential precautions and medications against coronavirus symptoms. Electronic medical records (EMRs), which store information about health conditions, diagnoses, and treatment methods, are excellent resources for optimizing clinical care through alerts, reminders, recommendations, and other forms of clinical decision support [8]. Patients with the same disease always have inter-individual symptoms, medications, treatments, and health conditions in common [8]. In this present work, we used EMRs to attain the objectives. Background Though the system requires supervisory confirmation, the model's amazing accuracy ensures dependence on the system and can get essential approvals to be introduced in practical sectors. The significance of this initiative demonstrates AI advancements in medical sciences as well as humanitarian relief. Considering all the aforementioned issues, the current research work was aimed: from COVID-19 patients to better comprehend more successful therapeutic approaches and potentially identify the most vulnerable groups based on personalized, genetic, and physiological traits. The substantial amount of COVID-19 treatment data in hospitals worldwide requires the use of advanced ML and DL methods to analyze it to discover and predict new therapeutic options that could provide better care for each patient and contribute to local hospital arrangements and operations [5]. In addition, AI can answer different queries of the patient in the absence of doctors [6]. Aiding the appropriate medications in emergency conditions is critical in fighting this global battle [7]. Projecting the proper medications for the desired patients might alleviate the enormous strain placed on doctors and nurses. In contrast, ML and AI will save time and also provide patients with accurate medication administration instructions. Besides, it is also helpful if lockdowns were implemented and people were required to stay at home and want to take essential precautions and medications against coronavirus symptoms. Electronic medical records (EMRs), which store information about health conditions, diagnoses, and treatment methods, are excellent resources for optimizing clinical care through alerts, reminders, recommendations, and other forms of clinical decision support [8]. Patients with the same disease always have inter-individual symptoms, medications, treatments, and health conditions in common [8]. In this present work, we used EMRs to attain the a) To develop an AI-based model for the prediction and recommendation of medications for COVID-19 patients based on the input data using the ANN, and b) To build a robust prediction model (COVID-MED), enabling deep learning (DL) strategies that can help the physicians as well as the COVID-19 patients quickly by recommending necessary medications. Study design A structured dataset including necessary features connected to the article objectives is required to materialize the idea of suggesting the necessary medications based on the symptoms of COVID-19-affected individuals. Following that, the ANN architecture, ML & DL approach will take over the responsibility for developing a model that will be supervised by the available dataset and predict the required output feature, which is nothing more than the drugs for the affected patients. A basic ML perceptron is illustrated in Figure 1. Figure 1: Basic ML Perceptron Data Collection and Preparation Techniques for COVID- MED Model Data are the characteristics or information, usually numerical, that are collected through observation [9]. The first and most important stage in constructing a machine learning or deep learning model is to preprocess data, which starts the process. The data used in this COVID-MED model has undergone several significant steps known as preprocessing in machine learning and neural networks. Figure 2 illustrates the data preprocessing stages, beginning with data acquisition and ending with processed and extracted features that are ready to be input into the model. Detailed information about the preprocessing steps that have been carried out in this proposed model is provided below. Data Collection and Preparation Techniques for COVID- Data Collection and Preparation Techniques for COVID- MED Model Data Collection and Preparation Techniques for COVID- MED Model Data are the characteristics or information, usually numerical, that are collected through observation [9]. The first and most important stage in constructing a machine learning or deep learning model is to preprocess data, which starts the process. The data used in this COVID-MED model has undergone Dataset Acquisition and after the hospitalization, a variety of basic clinical tests, as well as therapies and drugs used throughout their hospital stay. A questionnaire and clinical records were used to gather and organize data. Figure 3 represents the COVID-MED dataset that is comprised of various features. From the criteria mentioned above, medication ('injection and oral’) is supposed to be predicted by the model using the other features as the supervisory input features. On the other hand, Table 1 shows the variables that were used in the analysis. To keep up with the ever-improving deep learning models, a suitable dataset needs to obtain initially. In this article, a dataset was obtained from the previous record of 8953 patients who were admitted to the 1) specialized COVID unit in the Cox’s Bazar Medical College and Hospital, Bangladesh, 2) Sadar Isolation Center, Cox’s Bazar, Bangladesh, and 3) Ramu Isolation Center, Cox’s Bazar, Bangladesh. The data was collected between October 2020 and October 2021. Surja AS, et al., Journal of Ideas in Health (2022); 5(4):776-785 778 Data Collection and Preparation Techniques for COVID- MED Model Data Collection and Preparation Techniques for COVID- MED Model several significant steps known as preprocessing in machine learning and neural networks. Figure 2 illustrates the data preprocessing stages, beginning with data acquisition and ending with processed and extracted features that are ready to be input into the model. Detailed information about the preprocessing steps that have been carried out in this proposed model is provided below. Data are the characteristics or information, usually numerical, that are collected through observation [9]. The first and most important stage in constructing a machine learning or deep learning model is to preprocess data, which starts the process. The data used in this COVID-MED model has undergone Figure 2: Flowchart for data preprocessing Figure 2: Flowchart for data preprocessing Data Collection and Preparation Techniques for COVID- MED Model several significant steps known as preprocessing in machine learning and neural networks. Figure 2 illustrates the data preprocessing stages, beginning with data acquisition and ending with processed and extracted features that are ready to be input into the model. Detailed information about the preprocessing steps that have been carried out in this proposed model is provided below. Data are the characteristics or information, usually numerical, that are collected through observation [9]. The first and most important stage in constructing a machine learning or deep learning model is to preprocess data, which starts the process. The data used in this COVID-MED model has undergone Data are the characteristics or information, usually numerical, that are collected through observation [9]. The first and most important stage in constructing a machine learning or deep learning model is to preprocess data, which starts the process. The data used in this COVID-MED model has undergone Surja AS, et al., Journal of Ideas in Health (2022); 5(4):776-785 Figure 3: COVID-MED Dataset Dataset Acquisition This dataset included general demographic data, different symptoms before Hb ESR TC N L PLT Lowest O2 level Treatment (medicine & injection) 25 M 71 172.72 No No No 3 Symptometic Moderate O 101 No 16.5 32 5700 71 25 230000 0.95 A 40 M 76 170.18 No No No 7 Symptometic Moderate I 104 No 16 33 5500 70 25 430000 0.95 A 52 M 68 170.18 Yes No Diabetic 7 Symptometic Moderate L 101 No 13.2 101 4220 62 32 185000 0.95 A 52 M 71 165.1 Yes No No 7 Symptometic Moderate N 101 No 15.2 35 5100 68 30 220000 0.96 A 57 M 74 170.18 No No Diabetic 7 Symptometic Moderate A 101 No 10.01 118 7000 65 62 288000 0.95 B 65 M 76 170.18 Yes Yes Hbs/Ag (+ve) 5 Symptometic Mild A 100 Yes 13 47 5500 67 26 206000 0.97 C 30 F 56 157.48 No No No 10 Symptometic Moderate A 104 No 11.8 80 13000 80 16 250000 0.96 A 35 F 50 157.48 No No No 14 Symptometic Moderate A 102 No 12.4 78 13500 85 15 220000 0.97 D 50 F 64 157.48 No 1st dose No 10 Symptometic Moderate A 104 Yes 12.1 43 5300 68 18 185000 0.95 D 35 F 51 160.02 No No No 5 Symptometic Moderate A 102 No 12.1 60 4700 60 18 386000 0.90 C 56 F 52 154.94 No Yes HTN 7 Symptometic Moderate I 102 No 10.8 83 4000 54 38 300000 0.93 B 59 F 73 160.02 No No HTN, Asthma 3 Symptometic Severe S 104 Yes 12.1 43 5300 68 18 185000 0.94 A 55 F 69 160.02 No No HTN 3 Symptometic Moderate C 101 No 12 22 5600 58 30 180000 0.90 E 56 F 57 157.48 No No Asthma 7 Symptometic Moderate B 102 No 13.4 17 5600 55 37 180000 0.90 A 65 F 68 157.48 No Yes Asthma 8 Symptometic Moderate A 102 No 13 47 1300 78 19 298000 0.96 F 60 M 63 170.18 No No No 8 Symptometic Moderate C 102 No 13.5 88 4500 66 30 245000 0.95 G 62 M 64 167.64 No No No 6 Symptometic Moderate L 102 No 11 81 11800 80 16 223500 0.98 A 32 M 50 170.18 No No Hbs/Ag (+ve) 2 Symptometic Mild S 102 No 12.6 78 5800 64 32 400000 0.99 H 54 M 66 170.18 No Yes HTN 7 Symptometic Moderate A 102 No 12 83 4800 72 28 400000 0.95 D 31 M 58 162.56 No No No 7 Symptometic Moderate B 102 No 15.1 10 7500 76 20 300000 0.98 A Complications after admission Clinical tests Associated diseases Suffered (Days) Symptometic/A symptometic Severity Symptoms Highest Temp (°F) Age Gender Weight Height (cm) Smoker Vaccination Surja AS, et al., Journal of Ideas in Health (2022); 5(4):776-785 779 779 Table 1. Identifying and Handling the Missing Values After importing the dataset, it is pivotal to identify and correctly handle the missing values to make the model output efficient because missing values hampers the model's efficiency by creating a void in the calculations. As the data-obtaining process from the hospitals was successive in nature, there were hardly any missing values in the whole dataset. However, the dataset was handled and scrutinized for these kinds of error detection. Finally, to cross-check the null values following commands were run, which provided the satisfactory statement that the dataset was continuous. Importing the Dataset At this step, the obtained dataset is imported for the deep learning model to be constructed. Data is typically saved in a csv or excel file format that is easily imported by python libraries. Dataset Acquisition import numpy as np import pandas as pd import matplotlib.pyplot as plt The three core predetermined libraries used for the preprocessing tasks are NumPy, Pandas, and Matplotlib. The three core predetermined libraries used for the preprocessing tasks are NumPy, Pandas, and Matplotlib. Importing the Dataset At this step, the obtained dataset is imported for the deep learning model to be constructed. Data is typically saved in csv or excel file format that is easily imported by python libraries. From Google. colab import drive drive.mount(‘/content/drive’) data=pd.readexcel(‘drive/My Drive/Injections_Manipulated.xlsx’) These lines of commands were implemented to import the dataset through the python library using Google Colaboratory notebook. Dataset Acquisition Characteristics of the acquired baseline and clinical data Variable All (n=8953) Age 47.2 ± 11.4 Gender Male=5211, Female=3742 Weight (kg) 66.5 ± 6.8 Height (cm) 163.8 ± 5.6 Smoker Male=4123, Female=22, Non- smoker=4808 Vaccination status Yes=2711, No=6242 Suffered with symptoms before admitting into the hospital (days) 6.8 ± 2.8 Symptomatic/ Asymptomatic Symptomatic= 8112, Asymptomatic=841 Severity Severe=328, Mild=2547, Not severe=6078 Highest body temperature (F) 101.6 ± 1.6 Hemoglobin (Hb), g/dL 12.4 ± 2.2 Erythrocyte sedimentation rate (ESR), mm/hr 59.5 ± 24.5 Total count (TC), wbc/mcL 7065 ± 2057 Neutrophils (N), cells/mcL 66.5 ± 10.2 Lymphocytes, % 27.9 ± 9.2 Platelet count (PLT) 248700 ± 55920 Lowest oxygen level during hospital staying. 0.94 ± 0.04 Clinical symptoms included (symptoms were classified as shown in Figure 3) Fever, cough, cough with sputum, weakness, sore throat, diarrhea, and shortness of breath (SOB) Associated diseases included Diabetes, asthma, hypertension, hypotension, cardiovascular diseases (CVD), arthritis, anemia, chronic obstructive pulmonary disease (COPD), hypothyroidism, GOUT, pulmonary tuberculosis, acute kidney injury (AKI), benign prostatic hyperplasia (BPH), hepatitis B, and acute pyelonephritis. Injections used for treatment Low molecular weight heparins, steroids, antivirals, antibiotics, insulin, loop diuretics, and anti-emetics. Medication (oral) used for treatment Leukotriene receptor antagonists (LTRAs), antiulcerants, non-steroidal anti-inflammatory drugs (NSAIDs), antihistamines, antivirals, antibiotics, anxiolytic, vitamin supplements, cough suppracents, relaxants, antihypertensives, calcium antagonists, and anti-diabetic. Importing Crucial Libraries Python now offers a framework that provides the most comprehensive collection of chosen libraries by prominent data scientists. As a result, the platform has served as the foundation for this evolutionary effort, and numerous library functions have been built to enable the COVID-MED model to be executed. Table 1. Characteristics of the acquired baseline and clinical Importing Crucial Libraries Python now offers a framework that provides the most comprehensive collection of chosen libraries by prominent data scientists. As a result, the platform has served as the foundation for this evolutionary effort, and numerous library functions have been built to enable the COVID-MED model to be executed. import numpy as np import pandas as pd import matplotlib.pyplot as plt The three core predetermined libraries used for the preprocessing tasks are NumPy, Pandas, and Matplotlib. import numpy as np import pandas as pd import matplotlib.pyplot as plt The three core predetermined libraries used for the preprocessing tasks are NumPy, Pandas, and Matplotlib. Dataset Splitting Dataset Splitting Feature Selection The process of automatically or manually picking characteristics that contribute the most to forecasting the variable or output the model is interested in is known as feature selection. Prominent techniques which have been incorporated for selecting features in this COVID-MED model are univariate selection and correlation with heatmap. Univariate selection measures the compatibility rank of the input features compared with the output feature. As a regression problem is being analyzed in this model, therefore "f_regression" score function is introduced to determine the best features. It works on selecting the best features according to the data points that fit the best line among all the features. Among the total 21 features, 19 prominent features are chosen using univariate selection, which ranks the features based on their importance to the output. On the other hand, correlation measures the relation among the features. Correlation states how the features are related to each other or the target variable. Correlation can be positive (an increase in one value of a feature increases the value of the target variable) or negative (an increase in one value of the feature decreases the value of the target variable). There are various correlation techniques, such as Pearson, Kendall, Spearman, etc. In this problem, the Pearson correlation technique is being used. The least correlated features among the inputs and the highly correlated features with the target output are chosen to be fed into the model. Heatmap easily identifies which features are most related to the target variable. Following that, the correlations between the features were determined, and lastly, the selection criteria chose 15 unique and essential features to train the model. A heatmap of correlated features using the seaborn library is represented in Figure 4. ML & DL models must split datasets into two separate sections, such as training and test sets, to ensure the dataset's relevant training and testing phase. The suggested model maintains a 70:30 ratio to train and test the data. The proposed model incorporated the train_test_split library to execute the task. Surja AS, et al., Journal of Ideas in Health (2022); 5(4):776-785 Surja AS, et al., Journal of Ideas in Health (2022); 5(4):776-785 780 780 Figure 4: Correlation heat map between Input Features Encoding the Categorical Data Neural network and deep learning models are fundamentally based on mathematical equations. In the dataset cited above, there are several categorical variables such as ‘Associated Diseases’, ‘Symptoms’, ‘Severity’, ‘Medications’ and so on. So, the categorical features need to be encoded in a numerical format for being able to be analyzed by the COVID-MED model [10]. Here is how some categorical data have been encoded in the COVID-MED model. ‘x’ is represented as categorical input features. From sklearn.preprocessing import LabelEncoder lbl = LabelEncoder() data[‘x’] = lbl.fit_transform(data[‘x’]) Here, the categorical features were encoded in corresponding numerical formats by the aforementioned lines of codes. It simply converted the data into a numerical form which is easily detectable by the neural network model. Surja AS, et al., Journal of Ideas in Health (2022); 5(4):776-785 781 Proposed COVID-MED Model (ReLu). The ReLu activation function has been incorporated into this SDL model to commence the learning behavior. The main rationale for utilizing it is that it eliminates problems such as vanishing gradients and exploding gradients produced by sigmoid or hyperbolic activation functions. The defining curve concludes that ReLu ranges between 0 and the maximum value of inputs. That is, if the input value is negative, it returns 0; otherwise, if it is positive, it returns the same positive integer. As a result, in this COVID-MED model, the derivative for this range will always be 1. Negatives, on the other hand, will yield a zero derivative. The derivatives in this method do not vary between 0 and 0.25, thus rescuing from the difficulties of vanishing gradients and exploding gradients [13]. Furthermore, the COVID-MED model has integrated an adaptive optimizer dubbed Adam in order to improve the prediction capability and attain global minima quickly and smoothly. It gradually improves competence by lowering losses and estimating the change in weights and learning rates with gradient position, resulting in accurate output forecasts. Deep Learning models are based on neural networks, whereas Keras is one of the most popular and user-friendly neural network libraries written in Python. In this proposed COVID- MED model, the learning model has been developed by using Keras to solve a regression problem. Two renowned ways to build and develop Keras models are sequential and functional. The sequential API consists of creating a model layer-by-layer for the majority of problems. On the contrary, the functional API enables the creation of models with significantly greater flexibility, allowing models to readily decide where layers relate to more than only the previous and next levels [12]. In this study, a Sequential model was adopted. It is made up of a single dense input layer with 64 neurons. The input layer is followed by three hidden layers with 32, 16, and 8 neurons, respectively, with multiple neural connections. Finally, they all deducted in predicting the outputs in a single neuron. Notably, the activation function utilized in each neural connection in this model is referred to as the rectified linear unit Figure 5: Neural Network Architecture for COVID-MED model Figure 5: Neural Network Architecture for COVID-MED model Figure 5: Neural Network Architecture for COVID-MED model Feature Scaling Feature scaling allows for comparison on common ground in order to narrow the range of variables. Deep learning makes it possible to apply a variety of scaling approaches. The 'Normalization' technique to scale the imported dataset has been used in this research work. 𝑋′ = 𝑋−min (𝑋) max( )−min (𝑋) (1) Where, X^'=New Value, X=Old Value [11] 𝑋′ = 𝑋−min (𝑋) max( )−min (𝑋) (1) (1) Where, X^'=New Value, X=Old Value [11] Where, Where, Where, X=Old Value [11] [11] X=Old Value By normalizing the dataset, all the data points were renovated to a format that ranges from 0 to 1. It is much more convenient and efficient to deal with the numbers for a model which ranges in a minimalistic scale. Dealing with large digits was cumbersome and might make the model slow to perform. As far as it does not change the original information, it is logical to scale the dataset to obtain maximum efficiency. Surja AS, et al., Journal of Ideas in Health (2022); 5(4):776-785 end Preprocess input features as 2.1 Results Figure 7: Expected vs predicted plot for COVID-MED To summarize what has been done so far, it can be said that the data was collected from 8953 patients across three hospitals where the mean age of subjects was 47.2 ± 11.4 years. The majority of the patients were male, and the majority of them did smoke. About 30% received the first or both doses of the COVID-19 vaccine. When they arrived at the hospital, most patients (about 90%) had some symptoms (Table 1). The clinical conditions of 328 patients have deteriorated during their hospital stay. Most of the patients have one or multiple associated diseases. Table 1 also includes a list of the medications used to treat these patients during their stays at these three hospitals. To validate the proposed COVID-MED model, a dataset comprising the affected patients' age, gender, height, weight, smoking habits, prevalent symptoms, associated diseases, different clinical test reports, medications used for treatments during the hospital stay, and so on was employed (Figure 3, Table 1). After the data was imported and fed into the suggested ML & DL-based model, it went through a series of preprocessing processes to help the prediction technique be more precise. Additionally, The COVID-MED model chooses the most significant features based on their relevance to the output feature, correlation with other features, and within the feature itself to produce a reliable forecast of the output medications. Soon after the data was prepared, a neural network model consisting of a single dense input layer with 64 neurons, followed by three distinct hidden layers with varying combinations of neural connections, was set up to predict the medications given to COVID-19 patients. Finally, the architecture deducts in predicting the output medication in a single neuron. Some outstanding results were discovered when the COVID-MED model was tested after it had been successfully implemented and trained. Parameters chosen for assessing the results are Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and R2 Score, where they resemble their usual expressions. A detailed elaboration of their executive computations can be found in Table 2. Figure 7: Expected vs predicted plot for COVID-MED The model accuracy can be judged by plotting the actual values and the model-predicted values in a similar graph. Preprocess input features as 2.1 - number of samples: 1896, 70% for Training and 30% for Test; - number of samples: 1896, 70% for Training and 30% for Test; Evaluate and test the neural network Predict outputs and calculate errors To delve deep into the characteristics and architecture of the proposed model, the following few lines illustrate the whole scenario of what has been implemented and how. The lines describe the model formulation and glimpse its possible outcomes. - Training Set: train_x; - Label of training data: train_y; Architectural network of COVID-MED - 64 input neurons, 3 hidden layers of 32, 16, and 8 neurons - 1 output layer import TensorFlow as tf import DateTime, os #from tensorflow.keras.callbacks import TensorBoard #tensorboard = TensorBoard(log_dir='logs/ {}') model = keras.models.Sequential() model.add(keras.layers.Dense(64, activation='relu', input_shape=(16,))) model.add(keras.layers.Dense(32, activation='relu')) model.add(keras.layers.Dense(16, activation='relu')) model.add(keras.layers.Dense(8, activation='relu')) model.add(keras.layers.Dense(1)) model.compile(optimizer= 'Adam', loss='mean_squared_error') for i=1: numofepochs % Returns a net structure with updated weights Surja AS, et al., Journal of Ideas in Health (2022); 5(4):776-785 782 782 Results As illustrated in Figure 7, the proposed COVID-MED model returned a 0.99 R2 score while offering the best fit curve of predicted data covering more than 99% of the actual data. That means the COVID-MED model anticipates the output medications, such as injections or other oral meds, with around 99% accuracy. Table 3: Error & accuracy output from COVID-MED model MAE MSE RMSE R2 Score 0.0037 0.000035 0.0059 99.995 Table 3 summarizes the errors and the accuracy outcomes of the COVID-MED model that has been developed. Here, standard statistical formulations have been introduced to calculate the error functions like MAE, MSE, and RMSE. Similarly, the R2 score represents the relevance of the output medications with the originally fitted curve [11]. On the other hand, Figure 8 represents the epoch vs. loss curve, where the mitigation of the loss function can be visualized with the increasing number of epochs. Once the training phase is complete, the model begins to predict the output neuron by neuron. The neurons and the loss function assess the error in their initial prediction. The adaptive nature of the NN architecture optimization eventually reduces it. The image below depicts the gradual process of locating the global minima, or target output. Discussion similar work has been done on the medication prediction model using the local–global memory neural network (LGMNN) and EMRs [8]. According to Song et al. [8], the developed model can improve prediction performance for any given disease in an unsupervised manner. In another study, a model was developed that could eventually help with patient-specific decision support to reduce the time spent placing orders or detecting unusual orders to prevent medication errors [25]. Inpatient medication predictions may have clinical decision support applications, despite the fact that they are outcome-agnostic. Recent studies have also used AI to ensure medication safety, particularly device-based administration, with minimal potential overhead for patients and healthcare professionals [26]. With so much diversity in modern medical research, we believe our system could have significant clinical implications. We hope that this system will be able to provide clinicians with continuous feedback. Based on the feedback from our system, health professionals can then make a clinical decision and cross-check the medication to avoid unintended consequences. According to the best of our knowledge, this is the first study to use ANN to recommend appropriate medications for COVID- 19 patients. The model will recommend a medication after the user enters the necessary information. Although this model was developed for COVID-19 patients, physicians and other relevant stakeholders can apply it to any disease. It could be a significant advantage of this model. The critical situation of COVID-19 requires mobilization and the saving of medical, logistical, and human resources, and AI can not only facilitate this but also save time at a time when even one hour saved could result in the saving of lives in all locations where people are dying in COVID-19 [14]. According to the findings of a recent study, AI can answer various patient queries without the presence of doctors [6]. Following the emergence of COVID- 19, various AI-based techniques are being used for drug repurposing and repositioning [15], designing novel anti- COVID compounds [16], vaccine design [17], and designing associated protein structures [18]. Many researchers believe that using AI in healthcare will result in error-free decisions once the input is provided [19]. Our proposed model does have some limitations. One major limitation of this study is the small sample size. Despite the promising results of using the AI model to screen patients for COVID-19, more data is needed to test the AI model's generalizability to other patient populations. Discussion Moreover, this study only used data from one city, limiting its generalizability. Our research focused on the prediction of medication compounds rather than dosages. Due to differences in patient populations, prescribing practices, and health information storage in EHRs, the model, would most likely need to be retrained for use in different settings. The model was evaluated using only retrospective data. Models are likely to need to be retrained with new data regularly to prevent model performance degradation due to temporal changes in prescribing practices, such as adding new medications to formularies. A recent study used a computational methodology to calculate the global impact on the number of fatalities [20], and others used the SEIR method to forecast illness burden with a focus on South Asian countries (Bangladesh, India, and Pakistan) [21]. Furthermore, several researchers concentrated on the deep neural network (DNN) and convolutional neural network (CNN) models for behavioral analysis based on heterogeneous health data gathered in social media [22]. Notably, some research has employed various mathematical models to anticipate disease transmission, predict the number of incidents, deal with healthcare lacking data, and healthcare facilities in combating COVID-19 spread. As a result, choosing the right model and parameter values is critical for the forecasting model. However, a few models have been employed exclusively for countries with a high number of instances, such as China, Italy, Spain, the United Kingdom, Germany, and the United States. Motivated by the theories, this research proposes a DL-based regression model for predicting the prescriptions of COVID-19 patients. Patient's age, exposure to SARS-CoV-2, fever, cough, and cough with sputum, and white blood cell counts were significant features associated with SARS-CoV-2 status [23]. In this current study, these features are being used from acquired EMRs to develop the COVID-MED model. To the best of our knowledge, this is the first study to use an AI system to recommend appropriate medicine for COVID-19 patients. This model will hopefully be helpful in three possible ways: 1) doctors will be able to use this in an emergency situation where quick medication is required; 2) physicians will also be able to double-check the medications they have prescribed; 3) In some cases if this model is applied to other diseases, patients will be able to choose their own medication. The best part is that this will also be useful for any specific disease with an approved list of medicines. Table 2. Accuracy & Error Measuring Parameters used in COV-MED Table 2. Accuracy & Error Measuring Parameters used in COV-MED 𝑀𝐴𝐸 = ∑\|𝑦𝑖−𝑦𝑝𝑟𝑒𝑑\| 𝑛 𝑛 𝑖=1 (1) 𝑀𝑆𝐸 = ∑(𝑦𝑖−𝑦𝑝𝑟𝑒𝑑) 2 𝑛 𝑛 𝑖=1 (2) 𝑅𝑀𝑆𝐸 = √∑(𝑦𝑖−𝑦𝑝𝑟𝑒𝑑) 2 𝑛 𝑛 𝑖=1 (3) 𝑅2 = 1 − ∑(𝑦𝑖−𝑦𝑝𝑟𝑒𝑑)2 ∑(𝑦𝑖−𝑦𝑚𝑒𝑎𝑛)2 (4) Table 2. Accuracy & Error Measuring Parameters used in COV-MED Table 2. Accuracy & Error Measuring Parameters used in COV-MED 𝑀𝐴𝐸 = ∑\|𝑦𝑖−𝑦𝑝𝑟𝑒𝑑\| 𝑛 𝑛 𝑖=1 (1) 𝑀𝑆𝐸 = ∑(𝑦𝑖−𝑦𝑝𝑟𝑒𝑑) 2 𝑛 𝑛 𝑖=1 (2) 𝑅𝑀𝑆𝐸 = √∑(𝑦𝑖−𝑦𝑝𝑟𝑒𝑑) 2 𝑛 𝑛 𝑖=1 (3) 𝑅2 = 1 − ∑(𝑦𝑖−𝑦𝑝𝑟𝑒𝑑)2 ∑(𝑦𝑖−𝑦𝑚𝑒𝑎𝑛)2 (4) Finally, it has provided a value for mean absolute error (MAE) of 0.0037, means squared error (MSE) of 0.000035, and 0.0059 roots mean squared error (RMSE). Figure 8: Epoch vs loss curve for COVID-MED Figure 8: Epoch vs loss curve for COVID-MED Figure 8: Epoch vs loss curve for COVID-MED The curve depicted in Figure 8 shows that, despite some slight unusual rises, the loss function is settled very close to zero (0.000035). This also demonstrates how accurately the model can anticipate or generate outputs. The curve depicted in Figure 8 shows that, despite some slight unusual rises, the loss function is settled very close to zero (0.000035). This also demonstrates how accurately the model can anticipate or generate outputs. Finally, it has provided a value for mean absolute error (MAE) of 0.0037, means squared error (MSE) of 0.000035, and 0.0059 roots mean squared error (RMSE). Surja AS, et al., Journal of Ideas in Health (2022); 5(4):776-785 783 783 Discussion This technology will increase healthcare professionals' efficiency and improve healthcare quality at a lower cost. According to some research findings, AI may help doctors make more accurate diagnoses [24]. Machine-learning (ML) models can be trained using EHR data to predict medication orders for hospitalized patients in a patient-specific and time-specific manner [25]. We found a Availability of data and materials Availability of data and materials 5Department of Agricultural Chemistry, Sher-E-Bangla Agricultural University, Dhaka-1207, Bangladesh 5Department of Agricultural Chemistry, Sher-E-Bangla Agricultural University, Dhaka-1207, Bangladesh De-identified data supporting the study's findings are available from the authors upon reasonable request and with a signed data use agreement. By signing the agreement, the requester agrees that the data 1) will be used only for research purposes and will not be used for any product-related effort and 2) will not be shared with a third party. You can use the corresponding author email for further contact (fahadbau21@hotmail.com) 6Department of Food Engineering and Technology, Faculty of Agricultural Engineering and Technology, Sylhet Agricultural University, Sylhet-3100, Bangladesh 6Department of Food Engineering and Technology, Faculty of Agricultural Engineering and Technology, Sylhet Agricultural University, Sylhet-3100, Bangladesh 7Department of Clinical Tropical Medicine, Cox’s Bazar Medical College and Hospital, Cox’s Bazar-4700, Bangladesh 7Department of Clinical Tropical Medicine, Cox’s Bazar Medical College and Hospital, Cox’s Bazar-4700, Bangladesh 8Department of Medicine, Cox’s Bazar Medical College and Hospital, Cox’s Bazar-4700, Bangladesh 8Department of Medicine, Cox’s Bazar Medical College and Hospital, Cox’s Bazar-4700, Bangladesh Authors’ contributions Md. Fahad Jubayer was the principal investigator of this manuscript and approved the final manuscript. Amit Shaha Surja, Md. Tariqul Islam Limon, Md. Janibul Alam Soeb, and Md. Fahad Jubayer was involved in conceptualization, data analysis, model development, writing, reviewing, and editing. Md. Shoaib Arifin, Md. Meftaul Islam, Md. Shahidullah Kayshar, Md. Amirul Islam and Md. Mizanur Rahman performed the data analysis, writing, reviewing, and editing. Md Abdul Malek, Faroque Md Mohsin, Mohammed Shah Jahan, Anupam Barua, Tanjima Binte Topaz, and Irin Sultana were involved in data collection, field study, and writing. 9Medical Officer, BRAC, Bangladesh. References 1. Yu KH, Beam AL, Kohane IS. Artificial intelligence in healthcare. Nature Biomedical Engineering. 2018;2(10):719-31. https://doi.org/10.1038/s41551-018- 0305-z. 2. Gulshan V, Peng L, Coram M, Stumpe MC, Wu D, Narayanaswamy A, Venugopalan S, Widner K, Madams T, Cuadros J, Kim R. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. Jama. 2016;316(22):2402-10. http://doi.org/10.1001/jama.2016.17216 2. Gulshan V, Peng L, Coram M, Stumpe MC, Wu D, Narayanaswamy A, Venugopalan S, Widner K, Madams T, Cuadros J, Kim R. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. Jama. 2016;316(22):2402-10. Author Details Author Details 1Department of Electrical & Electronic Engineering, Sylhet Engineering College, Sylhet-3100, Bangladesh 2Directorate General of Health Services, Ministry of Health and Family Welfare, Bangladesh. 3Department of Farm Power and Machinery, Faculty of Agricultural Engineering and Technology, Sylhet Agricultural University, Sylhet-3100, Bangladesh 4Department of Agricultural Chemistry, Bangladesh Agricultural University, Mymensingh-2202, Bangladesh 5Department of Agricultural Chemistry, Sher-E-Bangla Agricultural University, Dhaka-1207, Bangladesh 6Department of Food Engineering and Technology, Faculty of Agricultural Engineering and Technology, Sylhet Agricultural University, Sylhet-3100, Bangladesh 7Department of Clinical Tropical Medicine, Cox’s Bazar Medical College and Hospital, Cox’s Bazar-4700, Bangladesh 8Department of Medicine, Cox’s Bazar Medical College and Hospital, Cox’s Bazar-4700, Bangladesh 9Medical Officer, BRAC, Bangladesh. Declaration Acknowledgment None. 1Department of Electrical & Electronic Engineering, Sylhet Engineering College, Sylhet-3100, Bangladesh 1Department of Electrical & Electronic Engineering, Sylhet Engineering College, Sylhet-3100, Bangladesh 1Department of Electrical & Electronic Engineering, Sylhet Engineering College, Sylhet-3100, Bangladesh 2Directorate General of Health Services, Ministry of Health and Family Welfare, Bangladesh. 2Directorate General of Health Services, Ministry of Health and Family Welfare, Bangladesh. 2Directorate General of Health Services, Ministry of Health and Family Welfare, Bangladesh. Funding 3Department of Farm Power and Machinery, Faculty of Agricultural Engineering and Technology, Sylhet Agricultural University, Sylhet-3100, Bangladesh 3Department of Farm Power and Machinery, Faculty of Agricultural Engineering and Technology, Sylhet Agricultural University, Sylhet-3100, Bangladesh The authors received no financial support for their research, authorship, and/or publication of this article. 4Department of Agricultural Chemistry, Bangladesh Agricultural University, Mymensingh-2202, Bangladesh 4Department of Agricultural Chemistry, Bangladesh Agricultural University, Mymensingh-2202, Bangladesh Conclusion Finally, this paper summarizes the implementation of a Deep Learning-based Sequential Model (COV-MED) for recommending medications to COVID-19 patients using specific clinical input features. It represents a logical sequence of data preprocessing and feature selection steps that lead to the prediction of medication outputs. The results of this exploratory work are promising and highly accurate, as the medications were predicted so fluently and accurately. As a result, it can be implemented in practical field applications, which should aid in making important decisions and making necessary arrangements in times of emergency to save people from this deadly virus and help humanity. Furthermore, the model can be used for other types of prediction strategies, such as predicting medications for other diseases, forecasting load variables, solar radiations, and so on. Therefore, the significance of this model and research work extends beyond the medical sector and into the energy, health, and communication sectors, respectively. In future research, we hope to see if our model can help with better real-time detection and alerting for medication errors. Abbreviation EMR: Electronic Medical Record; DL: Deep Learning; COVID-19: Coronavirus Disease of 2019; AI: Artificial Intelligence; ANN: Artificial Neural Network; ML: Machine Surja AS, et al., Journal of Ideas in Health (2022); 5(4):776-785 784 784 Learning; API: Application Programming Interface; ReLu: Rectified Linear Unit; SDL: Specification and Description Language; MAE: Mean Absolute Error; MSE: Mean Squared Error; RMSE: Root Mean Squared Error; DNN: Deep Neural Network; CNN: Convolutional Neural Network; SEIR: Susceptible-Exposed-Infectious-Recovered; SARS-CoV-2: Severe Acute Respiratory Syndrome Coronavirus 2; LGMNN: Local–Global Memory Neural Network; EHR: Electronic Health Record. (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article unless otherwise stated (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article unless otherwise stated. Ethical approval and consent to participate We conducted the research following the Declaration of Helsinki. The study was checked and approved by the institutional review board of [Cox’s Bazar Medical College and Hospital, Bangladesh] (2022). For this study, which evaluated de-identified data and involved no potential risk to patients, the institutional review boards waived the requirement to obtain written informed consent. No link between the patients and the researchers was made available to avoid any potential breach of confidentiality. 3. Jha S, Topol EJ. Adapting to artificial intelligence: radiologists and pathologists as information specialists. Jama.2016;316(22):2353-4. https://doi.org/10.1001/jama.2016.17438. 4. Debnath S, Barnaby DP, Coppa K, Makhnevich A, Kim EJ, Chatterjee S, Tóth V, Levy TJ, Paradis MD, Cohen SL, Hirsch JS. Machine learning to assist clinical decision- making during the COVID-19 pandemic. Bioelectronic medicine. 2020;6(1):1-8. https://doi.org/10.1186/s42234- 020-00050-8 Consent for publication Not applicable Competing interest The authors declare that they have no competing interests. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License Consent for publication Not applicable Consent for publication Not applicable 5. Alimadadi A, Aryal S, Manandhar I, Munroe PB, Joe B, Cheng X. Artificial intelligence and machine learning to fight COVID-19. Physiological Genomics. 2020;52(4):200-2. https://doi.org/10.1152/physiolgenomics.00029.2020 5. Alimadadi A, Aryal S, Manandhar I, Munroe PB, Joe B, Cheng X. Artificial intelligence and machine learning to fight COVID-19. Physiological Genomics. 2020;52(4):200-2. https://doi.org/10.1152/physiolgenomics.00029.2020 Competing interest Competing interest p g The authors declare that they have no competing interests. https://doi.org/10.1152/physiolgenomics.00029.2020 Open Access Surja AS, et al., Journal of Ideas in Health (2022); 5(4):776-785 785 2019;9(6):231-37. https://doi.org/10.1016/j.cmrp.2019.11.005 19. Murdoch TB, Detsky AS. The inevitable application of big data to health care. Jama. 2012;309(13):1351-52. https://doi.org/10.1001/jama.2013.393 7. Alafif T, Tehame AM, Bajaba S, Barnawi A, Zia S. Machine and deep learning towards COVID-19 diagnosis and treatment: survey, challenges, and future directions. International Journal of Environmental Research and PublicHealth.2021;18(3):1117. https://doi.org/10.3390/ijerph18031117. 20. Fokas AS, Cuevas-Maraver J, Kevrekidis PG. A quantitative framework for exploring exit strategies from the COVID-19 lockdown. Chaos, Solitons & Fractals. 2020;140:110244. https://doi.org/10.1016/j.chaos.2020.110244 https://doi.org/10.3390/ijerph18031117. 8. Song J, Wang Y, Tang S, Zhang Y, Chen Z, Zhang Z, Zhang T, Wu F. Local–Global Memory Neural Network for Medication Prediction. IEEE Transactions on Neural Networks and Learning Systems. 2020;32(4):1723-36. https://doi.org/10.1109/TNNLS.2020.2989364 21. Paul A, Chatterjee S, Bairagi N. Prediction on Covid-19 epidemic for different countries: Focusing on South Asia under various precautionary measures. Medrxiv. 2020. https://doi.org/10.1101/2020.04.08.20055095. 22. Zhou X, Liang W, Kevin I, Wang K, Shimizu S. Multi- modality behavioral influence analysis for personalized recommendations in health social media environment. IEEE Transactions on Computational Social Systems. 2019;6(5):888-97. https://doi.org/10.1109/TCSS.2019.2918285 9. OECD. OECD Glossary of Statistical Terms. 2008. p. 119. ISBN 978-92-64-025561. 10. Fang L, Karakiulakis G, Roth M. Are patients with hypertension and diabetes mellitus at increased risk for COVID-19 infection?. The Lancet Respiratory Medicine. 2020;8(4):e21. https://doi.org/10.1016/S2213- 2600(20)30116-8 23. Mei X, Lee HC, Diao KY, Huang M, Lin B, Liu C, Xie Z, Ma Y, Robson PM, Chung M, Bernheim A. Artificial intelligence–enabled rapid diagnosis of patients with COVID-19. Nature Medicine. 2020;26(8):1224-28. https://doi.org/10.1038/s41591-020-0931-3 11. Dodge Y. The Oxford Dictionary of Statistical Terms, 2003. OUP. ISBN 0-19-920613-9 (entry for normalization of scores). 12. Eijaz Allibhai (Medium, India). Building A Deep Learning Model using Keras. 2018. Available in: https://towardsdatascience.com/building-a-deep-learning- model-using-keras-1548ca149d37, (accessed 18 February 2022). 24. Jiang F, Jiang Y, Zhi H, Dong Y, Li H, Ma S, Wang Y, Dong Q, Shen H, Wang Y. Artificial intelligence in healthcare: past, present and future. Stroke and Vascular Neurology. 2017;2(4). https://dx.doi.org/10.1136/svn- 2017-000101 25. Rough K, Dai AM, Zhang K, Xue Y, Vardoulakis LM, Cui C, Butte AJ, Howell MD, Rajkomar A. Predicting inpatient medication orders from electronic health record data. Clinical Pharmacology & Therapeutics. 2020;108(1):145-54. https://doi.org/10.1002/cpt.1826 13. Da K. A method for stochastic optimization. arXiv preprint. 2014; arXiv:1412.6980. https://doi.org/10.48550/arXiv.1412.6980 14. Jamshidi M, Lalbakhsh A, Talla J, Peroutka Z, Hadjilooei F, Lalbakhsh P, Jamshidi M, La Spada L, Mirmozafari M, Dehghani M, Sabet A. Open Access Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License 6. Haleem A, Javaid M, Khan IH. Current status and applications of artificial intelligence (AI) in medical field: An overview. Current Medicine Research and Practice. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License Surja AS, et al., Journal of Ideas in Health (2022); 5(4):776-785 Surja AS, et al., Journal of Ideas in Health (2022); 5(4):776-785 Artificial intelligence and COVID- 19: deep learning approaches for diagnosis and treatment. IEEE Access. 2020; 8:109581-95. https://doi.org/10.1109/ACCESS.2020.3001973 26. Zhao, M., Hoti, K., Wang, H., Raghu, A., & Katabi, D. Assessment of medication self-administration using artificial intelligence. Nature Medicine. 2021;27(4):727- 35. https://doi.org/10.1038/s41591-021-01273-1 15. Ge Y, Tian T, Huang S, Wan F, Li J, Li S, Yang H, Hong L, Wu N, Yuan E, Cheng L. A data-driven drug repositioning framework discovered a potential therapeutic agent targeting COVID-19. BioRxiv. 2020. https://doi.org/10.1101/2020.03.11.986836 16. Zhavoronkov A, Zagribelnyy B, Zhebrak A, Aladinskiy V, Terentiev V, Vanhaelen Q, Bezrukov DS, Polykovskiy D, Shayakhmetov R, Filimonov A, Bishop M. Potential non- covalent SARS-CoV-2 3C-like protease inhibitors designed using generative deep learning approaches and reviewed by human medicinal chemist in virtual reality. Chemrxiv. 2020. https://doi.org/10.26434/chemrxiv.12301457.v1 17. Ong E, Wong MU, Huffman A, He Y. COVID-19 coronavirus vaccine design using reverse vaccinology and machine learning. Frontiers in Immunology. 2020;1581. https://doi.org/10.3389/fimmu.2020.01581 18. Senior AW, Evans R, Jumper J, Kirkpatrick J, Sifre L, Green T, Qin C, Žídek A, Nelson AW, Bridgland A, Penedones H. Improved protein structure prediction using potentials from deep learning. Nature. 2020;577(7792):706-10. https://doi.org/10.1038/s41586- 019-1923-7
https://openalex.org/W2971932832	https://www.zora.uzh.ch/id/eprint/175232/1/PhysRevLett.123.102001.pdf	English	null	Triple Differential Dijet Cross Section at the LHC	Physical review letters	2,019	cc-by	6,088	Zurich Open Repository and Archive University of Zurich University Library Strickhofstrasse 39 CH-8057 Zurich www.zora.uzh.ch Year: 2019 Published by the American Physical Society under the terms of the Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article’s title, journal citation, and DOI. Funded by SCOAP3. DOI: 10.1103/PhysRevLett.123.102001 Jet production in proton-proton collisions results pre- dominantly from hard interactions, when two partons from the incident hadrons undergo a hard pointlike interaction and scatter at relatively large angles. As such, it is directly sensitive to the dynamics of the pointlike strong-interaction partonic cross section and to the nonperturbative descrip- tion of the internal proton structure encoded in the parton distribution functions (PDFs). It is the combined interplay between the parton-parton scattering matrix elements and the parton luminosities that determines the shape of the dijet cross section. For this reason, to fully exploit the wealth of available data from the LHC it is important to have a reliable and accurate theoretical prediction for the dijet cross section. More detailed information about the shape of the dijet cross section can be obtained when the latter is determined in triple-differential form [1], over a wide range in the variables that fully describe the two jet events, namely, the transverse momentum pT and rapidities y of the two leading jets in the event. Triply differential measurements of dijet production have so far been per- formed by the CDF experiment [2] at the Tevatron and by the CMS experiment [3] at the LHC. manifest the IR safety of the observable, and that directly map the measured cross section to surfaces on the ðx; Q2Þ plane where the PDFs are determined. In this Letter, we calculate the triple-differential dijet cross section as a function of the following three kinemati- cal variables: the average transverse momentum pT;avg ¼ ðpT;1 þ pT;2Þ=2 of the two leading jets, half of their rapidity separation y ¼ jy1 −y2j=2, and the boost of the dijet system yb ¼ jy1 þ y2j=2. Our results are presented below as six distributions and compared with CMS 8 TeV data [3]. manifest the IR safety of the observable, and that directly map the measured cross section to surfaces on the ðx; Q2Þ plane where the PDFs are determined. In this Letter, we calculate the triple-differential dijet cross section as a function of the following three kinemati- cal variables: the average transverse momentum pT;avg ¼ ðpT;1 þ pT;2Þ=2 of the two leading jets, half of their rapidity separation y ¼ jy1 −y2j=2, and the boost of the dijet system yb ¼ jy1 þ y2j=2. Our results are presented below as six distributions and compared with CMS 8 TeV data [3]. Triple Differential Dijet Cross Section at the LHC Gehrmann-De Ridder, A ; Gehrmann, T ; Glover, E W N ; Huss, A ; Pires, J Posted at the Zurich Open Repository and Archive, University of Zurich ZORA URL: https://doi.org/10.5167/uzh-175232 Journal Article Published Version The following work is licensed under a Creative Commons: Attribution 4. Posted at the Zurich Open Repository and Archive, University of Zurich ZORA URL: https://doi.org/10.5167/uzh-175232 Journal Article Published Version Originally published at: Gehrmann-De Ridder, A; Gehrmann, T; Glover, E W N; Huss, A; Pires, J (2019). Triple Differential Dijet Cross Section at the LHC. Physical Review Letters, 123(10):102001. DOI: https://doi.org/10.1103/physrevlett.123.102001 PHYSICAL REVIEW LETTERS 123, 102001 (2019) (Received 1 July 2019; published 6 September 2019) The measurement of the triple-differential dijet production cross section as a function of the average transverse momentum pT;avg, half the rapidity separation y, and the boost yb of the two leading jets in the event enables a kinematical scan of the underlying parton momentum distributions. We compute for the first time the second-order perturbative QCD corrections to this triple-differential dijet cross section, at leading color in all partonic channels, thereby enabling precision studies with LHC dijet data. A detailed comparison with experimental CMS 8 TeV data is performed, demonstrating how the shape of this differential cross section probes the parton densities in different kinematical ranges. DOI: 10.1103/PhysRevLett.123.102001 Triple Differential Dijet Cross Section at the LHC A. Gehrmann-De Ridder ,1,2 T. Gehrmann ,2 E. W. N. Glover ,3 A. Huss,4 and J. Pires 5,6 1Institute for Theoretical Physics, ETH, CH-8093 Zürich, Switzerland 2Department of Physics, Universität Zürich, Winterthurerstrasse 190, CH-8057 Zürich, Switzerland 3Institute for Particle Physics Phenomenology, University of Durham, Durham DH1 3LE, United Kingdom 4Theoretical Physics Department, CERN, 1211 Geneva 23, Switzerland 5CFTP, Instituto Superior T´ecnico, Universidade de Lisboa, P-1049-001 Lisboa, Portugal 6LIP, Avenida Professor Gama Pinto 2, P-1649-003 Lisboa, Portugal DOI: 10.1103/PhysRevLett.123.102001 Our calculation is performed in the NNLOJET framework, employing the antenna subtraction method [10–12] to remove all unphysical infrared singularities from the matrix elements, which we take at leading color in all partonic subprocesses at NNLO, while keeping the full color dependence at lower orders. The same setup was used for the calculation of the NNLO corrections to inclusive jet [13,14] and dijet production [15]. We use the MMHT2014 NNLO parton distribution functions [16] with αsðMZÞ ¼ 0.118 for all predictions at LO, NLO, and NNLO to emphasize the role of the perturbative corrections at each successive order. The CMS 8 TeV measurement [3] is performed in six different regions in ðy; ybÞ, each corresponding to different types of event topologies and probing different aspects of the partonic structure of the colliding protons. In Fig. 1, a density plot in the (x1, x2) plane of the triple-differential cross section for the six event topologies considered in the CMS study is shown. The CMS analysis [3] also performs a detailed study of the constraints on PDFs that can be derived from the measurement data. These turn out to be inherently limited by the precision of the theoretical description of the underlying hard scattering processes available. The combined nonperturbative (NP) contributions from hadronization and the underlying event, modeled through multiple parton interactions, are not included in the predictions at the parton level, but have been derived from parton shower predictions at NLO in Ref. [3]. The corresponding NP effects have been found to be at most 10% for the lowest pT;avg bins and negligible above 1 TeV. A recent study [17] has shown that for the R ¼ 0.7 cone The theoretical predictions for the jet cross section are obtained in perturbative QCD, as a convolution of the parton distribution functions for the incoming particles and the parton-parton hard scattering cross section. The pre- vious state of the art, as used in Ref. DOI: 10.1103/PhysRevLett.123.102001 [3], were predictions including next-to-leading order (NLO) QCD [1,4–6] and electroweak (EWK) effects[7–9] corrections for the 3 − 10 2 − 10 1 − 10 1 1 x 3 − 10 2 − 10 1 − 10 1 2 x NNLOJET < 1 0 < y* < 1 b 0 < y (pb) σ 3 − 10 2 − 10 1 − 10 1 10 2 10 3 10 4 10 5 10 < 3 − 10 2 − 10 1 − 10 1 1 x 3 − 10 2 − 10 1 − 10 1 2 x NNLOJET < 1 1 < y* < 2 b 0 < y (pb) σ 3 − 10 2 − 10 1 − 10 1 10 2 10 3 10 4 10 5 10 < 3 − 10 2 − 10 1 − 10 1 1 x 3 − 10 2 − 10 1 − 10 1 2 x NNLOJET < 1 2 < y* < 3 b 0 < y (pb) σ 3 − 10 2 − 10 1 − 10 1 10 2 10 3 10 4 10 5 10 < 3 − 10 2 − 10 1 − 10 1 1 x 3 − 10 2 − 10 1 − 10 1 2 x NNLOJET < 2 0 < y* < 1 b 1 < y (pb) σ 3 − 10 2 − 10 1 − 10 1 10 2 10 3 10 4 10 5 10 < 3 − 10 2 − 10 1 − 10 1 1 x 3 − 10 2 − 10 1 − 10 1 2 x NNLOJET < 2 1 < y* < 2 b 1 < y (pb) σ 3 − 10 2 − 10 1 − 10 1 10 2 10 3 10 4 10 5 10 < 3 − 10 2 − 10 1 − 10 1 1 x 3 − 10 2 − 10 1 − 10 1 2 x NNLOJET < 3 0 < y* < 1 b 2 < y (pb) σ 3 − 10 2 − 10 1 − 10 1 10 2 10 3 10 4 10 5 10 < FIG. 1. DOI: 10.1103/PhysRevLett.123.102001 Using the kinematical variables defining the triple differential dijet cross section, the longitudinal momentum fractions of the incoming partons can, for the Born process of back-to-back jets, be written using momentum conservation, x1 ¼ pTﬃﬃﬃs p ðeþy1 þ eþy2Þ ¼ 2pT;avg ﬃﬃﬃs p eyb coshðyÞ; x2 ¼ pTﬃﬃﬃs p ðe−y1 þ e−y2Þ ¼ 2pT;avg ﬃﬃﬃs p e∓yb coshðyÞ: ð1Þ ð1Þ To use the full potential of the LHC data to achieve the ultimate experimental and theoretical precision in the way the structure of the proton is accessed, it is convenient to make a clever choice of variables in the dijet process, which From Eq. (1) it is clear that for small values of yb the dijet data probe the configuration x1 ≈x2, with the x values determined by the pT;avg of the jets. Moreover, it can be seen that a variation in y at different pT;avg, probe approximately the same x at different Q2. On the other hand, a variation in the dijet cross section in yb is an additional handle that allows us to depart from the x1 ≈x2 region and probe the scattering of a high-x parton off a low-x parton. In this respect, it is noteworthy that the 02001-1 Published by the American Physical Society 102001-1 Published by the American Physical Society 0031-9007=19=123(10)=102001(5) Published by the American Physical Society PHYSICAL REVIEW LETTERS 123, 102001 (2019) dijet production cross section. At this level of accuracy, scale uncertainties and missing higher order corrections in the theoretical calculation significantly limit the achievable precision in the determination of the shape and normali- zation of the triple differential cross section. To improve the perturbative QCD description of this process, we present in this Letter for the first time a computation of the NNLO corrections to the triple-differential dijet cross section at the LHC. parton-parton subprocess matrix elements are independent of yb. As result, for a fixed y slice, the only variation of the cross section comes from the variation of the parton densities as yb varies over the allowed kinematic range. The triple-differential dijet observable is therefore ideal to study PDFs, because the choice of variables in the dijet system directly maps to surfaces on the ðx; Q2Þ plane where the PDFs are determined, in event topologies that are sensitive to the PDFs, while being insensitive to the matrix elements. DOI: 10.1103/PhysRevLett.123.102001 Allowed kinematical regions at LO in the triple differential dijet inclusive cross section (in pb) at ﬃﬃﬃs p ¼ 8 TeV in the partonic fraction x1, x2 plane for the jet pT cuts of the CMS measurement. 3 − 10 2 − 10 1 − 10 1 1 x 3 − 10 2 − 10 1 − 10 1 2 x NNLOJET < 1 1 < y* < 2 b 0 < y (pb) σ 3 − 10 2 − 10 1 − 10 1 10 2 10 3 10 4 10 5 10 < 3 − 10 2 − 10 1 − 10 1 1 x 3 − 10 2 − 10 1 − 10 1 2 x NNLOJET < 1 0 < y* < 1 b 0 < y (pb) σ 3 − 10 2 − 10 1 − 10 1 10 2 10 3 10 4 10 5 10 < 3 − 10 2 − 10 1 − 10 1 1 x 3 − 10 2 − 10 1 − 10 1 2 x NNLOJET < 1 2 < y* < 3 b 0 < y (pb) σ 3 − 10 2 − 10 1 − 10 1 10 2 10 3 10 4 10 5 10 < 1 3 − 10 2 − 10 1 − 10 1 1 x 3 − 10 2 − 10 1 − 10 1 2 x NNLOJET < 2 0 < y* < 1 b 1 < y (pb) σ 3 − 10 2 − 10 1 − 10 1 10 2 10 3 10 4 10 5 10 < 1 3 − 10 2 − 10 1 − 10 1 1 x 3 − 10 2 − 10 1 − 10 1 2 x NNLOJET < 2 1 < y* < 2 b 1 < y (pb) σ 3 − 10 2 − 10 1 − 10 1 10 2 10 3 10 4 10 5 10 < 3 − 10 2 − 10 1 − 10 1 1 x 3 − 10 2 − 10 1 − 10 1 2 x NNLOJET < 3 0 < y* < 1 b 2 < y (pb) σ 3 − 10 2 − 10 1 − 10 1 10 2 10 3 10 4 10 5 10 < FIG. 1. PHYSICAL REVIEW LETTERS 123, 102001 (2019) (GeV) T,avg p 4 − 10 3 − 10 2 − 10 1 − 10 1 10 2 10 3 10 4 10 5 10 6 10 7 10 8 10 (pb/GeV) dy* b dy T,avg dp σ 3 d 200 300 500 1000 2000 (R=0.7) t anti-k MMHT2014_nnlo ) jj =m μ EWK ( ⊗ NP ⊗ NNLO NNLOJET 8 TeV ) 2 < 1 0 < y* < 1 (x10 b 0 < y ) 2 < 1 1 < y* < 2 (x10 b 0 < y ) 1 < 1 2 < y* < 3 (x10 b 0 < y ) 1 < 2 0 < y* < 1 (x10 b 1 < y ) 1 < 2 1 < y* < 2 (x10 b 1 < y ) 0 < 3 0 < y* < 1 (x10 b 2 < y -1 = 19.7 fb int CMS L FIG. 2. The triple differential dijet cross section as a function of pT;avg for the six bins of y, yb at NNLO with central scale choice μ ¼ mjj, compared to CMS 8 TeV 19.7 fb−1 data. (GeV) T,avg p 4 − 10 3 − 10 2 − 10 1 − 10 1 10 2 10 3 10 4 10 5 10 6 10 7 10 8 10 (pb/GeV) dy* b dy T,avg dp σ 3 d 200 300 500 1000 2000 (R=0.7) t anti-k MMHT2014_nnlo ) jj =m μ EWK ( ⊗ NP ⊗ NNLO NNLOJET 8 TeV ) 2 < 1 0 < y* < 1 (x10 b 0 < y ) 2 < 1 1 < y* < 2 (x10 b 0 < y ) 1 < 1 2 < y* < 3 (x10 b 0 < y ) 1 < 2 0 < y* < 1 (x10 b 1 < y ) 1 < 2 1 < y* < 2 (x10 b 1 < y ) 0 < 3 0 < y* < 1 (x10 b 2 < y -1 = 19.7 fb int CMS L FIG. 3. NLO/LO (blue) and NNLO/NLO (red) K factors triple differential in pT;avg, y, and yb. Bands represent the scale variation of the numerator. NNLO PDFs are used for all predictions. FIG. 2. PHYSICAL REVIEW LETTERS 123, 102001 (2019) (GeV) T,avg p 4 − 10 3 − 10 2 − 10 1 − 10 1 10 2 10 3 10 4 10 5 10 6 10 7 10 8 10 (pb/GeV) dy* b dy T,avg dp σ 3 d 200 300 500 1000 2000 (R=0.7) t anti-k MMHT2014_nnlo ) jj =m μ EWK ( ⊗ NP ⊗ NNLO NNLOJET 8 TeV ) 2 < 1 0 < y* < 1 (x10 b 0 < y ) 2 < 1 1 < y* < 2 (x10 b 0 < y ) 1 < 1 2 < y* < 3 (x10 b 0 < y ) 1 < 2 0 < y* < 1 (x10 b 1 < y ) 1 < 2 1 < y* < 2 (x10 b 1 < y ) 0 < 3 0 < y* < 1 (x10 b 2 < y -1 = 19.7 fb int CMS L FIG. 2. The triple differential dijet cross section as a function of pT;avg for the six bins of y, yb at NNLO with central scale choice μ ¼ mjj, compared to CMS 8 TeV 19.7 fb−1 data. 1 1.5 2 200 300 500 1000 2000 < 3 0 < y* < 1 b 2 < y 1 1.5 2 < 2 1 < y* < 2 b 1 < y 1 1.5 2 < 2 0 < y* < 1 b 1 < y 1 1.5 2 b < 1 2 < y* < 3 0 < y 1 1.5 2 b < 1 1 < y < 2 0 < y (GeV) T,avg p FIG. 3. NLO/LO (blue) and NNLO/NLO (red) K factors triple differential in pT;avg, y, and yb. Bands represent the scale variation of the numerator. NNLO PDFs are used for all predictions. PHYSICAL REVIEW LETTERS 123, 102001 (2019) 2, we present the absolute dijet triple-differential cross section at NNLO ⊗NP ⊗EWK as a function of pT;avg, y and yb, which we compare to LHC data. The fiducial cuts of the measurement include all events with at least two jets with absolute rapidity up to jyj ≤5.0 recorded in the CMS 8 TeV 19.7 fb−1 data, where the two jets leading in pT are required to have pT ≥50 GeV and jyj ≤3.0. The jet reconstruction in both experimental analysis and theory prediction uses the anti-kT jet algorithm [18] with radius parameter R ¼ 0.7. We observe an excellent agreement in the description of the data by the NNLO prediction over 7 orders of magnitude. 1 1.5 2 200 300 500 1000 2000 < 3 0 < y* < 1 b 2 < y 1 1.5 2 < 2 1 < y* < 2 b 1 < y 1 1.5 2 < 2 0 < y* < 1 b 1 < y 1 1.5 2 b < 1 2 < y* < 3 0 < y 1 1.5 2 b < 1 1 < y* < 2 0 < y 1 1.5 2 ) jj =m F μ = R μ jets, R=0.7, ( T CMS, 8 TeV, anti-k NNLOJET b < 1 0 < y* < 1 0 < y NLO/LO NNLO/NLO (GeV) T,avg p 1 1.5 2 200 300 500 1000 2000 < 3 0 < y* < 1 b 2 < y 1 1.5 2 < 2 1 < y* < 2 b 1 < y 1 1.5 2 < 2 0 < y* < 1 b 1 < y 1 1.5 2 b < 1 2 < y* < 3 0 < y 1 1.5 2 b < 1 1 < y* < 2 0 < y 1 1.5 2 ) jj =m F μ = R μ jets, R=0.7, ( T CMS, 8 TeV, anti-k NNLOJET b < 1 0 < y* < 1 0 < y NLO/LO NNLO/NLO (GeV) T,avg p FIG. 3. NLO/LO (blue) and NNLO/NLO (red) K factors triple differential in pT;avg, y, and yb. Bands represent the scale variation of the numerator. NNLO PDFs are used for all predictions. g p y NNLO prediction over 7 orders of magnitude. PHYSICAL REVIEW LETTERS 123, 102001 (2019) The fiducial cuts of the measurement include all events with at least two jets with absolute rapidity up to jyj ≤5.0 recorded in the CMS 8 TeV 19.7 fb−1 data, where the two jets leading in pT are required to have pT ≥50 GeV and jyj ≤3.0. The jet reconstruction in both experimental analysis and theory prediction uses the anti-kT jet algorithm [18] with radius parameter R ¼ 0.7. We observe an excellent agreement in the description of the data by the NNLO prediction over 7 orders of magnitude. For the central yb slice, and at small y, we observe NNLO effects that range between a few percent at low pT;avg., rising to 15% at large pT;avg (first panel). For larger y slices, i.e., smaller scattering angles, the NNLO effects increase reaching up to 20% across the entire pT;avg range in the 2 < y < 3 rapidity slice. On the other hand, for the longitudinally boosted topologies, i.e., at larger values of yb, the NNLO effects are smaller (as can be seen in the lower three panels in Fig. 3). They range between a few percent at low pT;avg and increase slightly to 10%, for the smaller scattering angle topologies, typically found at large y. At any given fixed order in perturbation theory, the predictions retain some dependence on the unphysical renormalization and factorization scales. An assessment of the scale uncertainty of the calculation at NLO and NNLO is obtained from independent variations of the renormalization or factorization scales by a factor of 2 around an arbitrary central scale choice. For the process at hand, the production of dijet events at high pT have as a natural scale (see Ref. [15] for a detailed study) the mass of the dijet system mjj ¼ ðpj1 þ pj2Þ2, which closely approx- imates the four-momentum transferred in the interaction. We also observe a good convergence of the perturbative expansion with NNLO effects smaller in magnitude with respect to the NLO correction at the previous order, demonstrating that the variables chosen to describe the measurement are stable and infrared safe, and allow us to reliably predict the triply differential cross section over the whole kinematical range in perturbation theory. In this respect, we note that the NNLO correction changes both the shape and normalization of the NLO result, and can therefore have a significant impact towards obtaining better In Fig. DOI: 10.1103/PhysRevLett.123.102001 Allowed kinematical regions at LO in the triple differential dijet inclusive cross section (in pb) at ﬃﬃﬃs p ¼ 8 TeV in the partonic fraction x1, x2 plane for the jet pT cuts of the CMS measurement. 102001-2 PHYSICAL REVIEW LETTERS 123, 102001 (2019) PHYSICAL REVIEW LETTERS 123, 102001 (2019) size, there is an excellent agreement for the parton-level cross section between fixed-order and NLO-matched results. For this reason, we will take into acount the NP effects obtained in Ref. [3] as a multiplicative factor in each bin of the parton-level NNLO cross section, labeling the results as NNLO ⊗NP. An assessment of the impact of the newly computed NNLO contribution can be seen in Fig. 3, where we show explicitly the ratio between the NNLO prediction and the NLO result (in red), together with the ratio between the NLO cross section and the LO result (in blue). The size of the NNLO corrections varies significantly as a function of pT;avg: and the applied cuts on y and yb. The contribution from EWK effects from virtual exchanges of massive W and Z bosons have been computed in Ref. [7]. These are smaller than 3% below 1 TeV and reach 8% for the highest pT;avg. Using the results from Ref. [7], EWK corrections are applied multiplicatively to the QCD calculation for the central scale choice and we label the corresponding prediction NNLO ⊗NP ⊗EWK. The contribution from EWK effects from virtual exchanges of massive W and Z bosons have been computed in Ref. [7]. These are smaller than 3% below 1 TeV and reach 8% for the highest pT;avg. Using the results from Ref. [7], EWK corrections are applied multiplicatively to the QCD calculation for the central scale choice and we label the corresponding prediction NNLO ⊗NP ⊗EWK. At any given fixed order in perturbation theory, the predictions retain some dependence on the unphysical renormalization and factorization scales. An assessment of the scale uncertainty of the calculation at NLO and NNLO is obtained from independent variations of the renormalization or factorization scales by a factor of 2 around an arbitrary central scale choice. For the process at hand, the production of dijet events at high pT have as a natural scale (see Ref. [15] for a detailed study) the mass of the dijet system mjj ¼ ðpj1 þ pj2Þ2, which closely approx- imates the four-momentum transferred in the interaction. In Fig. 2, we present the absolute dijet triple-differential cross section at NNLO ⊗NP ⊗EWK as a function of pT;avg, y and yb, which we compare to LHC data. PHYSICAL REVIEW LETTERS 123, 102001 (2019) 4. The NLO (blue) and NNLO (green) theory predictions and CMS data normalized to the NLO central value. Parton-level predictions corrected for nonperturbative (NP) effects and combined NP and electroweak effects (EW) effects, implemented as a multiplicative factor to the NNLO result, are shown in pink and red. The shaded bands shown for the NLO and the NNLO ⊗NP ⊗ EWK predictions represent the variation of the theoretical scales in the numerator by factors of 0.5 and 2. The uncertainty in the data is the total experimental uncertainty, including systematic and statistical uncertainties added in quadrature. constraints on parton distribution functions using data from the triple differential dijet measurement. comparable or lower than experimental errors, while better explaining kinematical shapes. Our newly derived results will enable the usage of dijet measurements in precision studies of the partonic structure of the colliding hadrons. Owing to the large dynamical range covered by the CMS measurement, a quantitative comparison between data and theory is best performed by taking ratios, and we use the NLO QCD parton-level prediction as a reference value. The resulting ratios to the CMS data [3] and to the NNLO predictions at parton level, and with NP and EWK corrections applied are displayed in Fig. 4. The results show a significant improvement in the description of the data at NNLO with respect to NLO, for the entire kinematical range of the measurement, in particular for the central yb slice (top three panels). The authors thank Xuan Chen, James Currie, Marius Höfer, Imre Majer, Jan Niehues, Duncan Walker, and James Whitehead for useful discussions and their many contribu- tions to the NNLOJET code. This research was supported in part by the UK Science and Technology Facilities Council under Contract No. ST/G000905/1, by the Swiss National Science Foundation (SNF) under Contracts No. 200021- 172478 and No. 200020-175595, by the Research Executive Agency (REA) of the European Union through the ERC Consolidator Grant HICCUP (614577) and the ERC Advanced Grant MC@NNLO (340983), by the Fundação para a Ciência e Tecnologia (FCT-Portugal), under Projects No. UID/FIS/00777/2019, No. CERN/FIS-PAR/0022/2017, and COST Action CA16201 PARTICLEFACE. J. P. gratefully acknowledges the hospitality andfinancial support of the CERN theory group. In the lower three panels of Fig. PHYSICAL REVIEW LETTERS 123, 102001 (2019) PHYSICAL REVIEW LETTERS 123, 102001 (2019) (GeV) T,avg p 0.8 1 1.2 1.4 1.6 1.8 Ratio to NLO 200 300 500 1000 2000 NNLOJET 8 TeV (R=0.7) t anti-k < 1 0 < y* < 1 b 0 < y MMHT2014_nnlo -1 = 19.7 fb int CMS L ) jj =m μ NLO ( ) jj =m μ NP ( ⊗ NNLO ) jj =m μ NNLO ( ) jj =m μ EWK ( ⊗ NP ⊗ NNLO (GeV) T,avg p 0.8 1 1.2 1.4 1.6 1.8 Ratio to NLO 200 300 500 1000 2000 NNLOJET 8 TeV (R=0.7) t anti-k < 1 1 < y* < 2 b 0 < y MMHT2014_nnlo -1 = 19.7 fb int CMS L ) jj =m μ NLO ( ) jj =m μ NP ( ⊗ NNLO ) jj =m μ NNLO ( ) jj =m μ EWK ( ⊗ NP ⊗ NNLO (GeV) T,avg p 0.8 1 1.2 1.4 1.6 1.8 Ratio to NLO 200 300 500 1000 2000 NNLOJET 8 TeV (R=0.7) t anti-k < 1 2 < y* < 3 b 0 < y MMHT2014_nnlo -1 = 19.7 fb int CMS L ) jj =m μ NLO ( ) jj =m μ NP ( ⊗ NNLO ) jj =m μ NNLO ( ) jj =m μ EWK ( ⊗ NP ⊗ NNLO (GeV) T,avg p 0.8 1 1.2 1.4 1.6 1.8 Ratio to NLO 200 300 500 1000 2000 NNLOJET 8 TeV (R=0.7) t anti-k < 2 0 < y* < 1 b 1 < y MMHT2014_nnlo -1 = 19.7 fb int CMS L ) jj =m μ NLO ( ) jj =m μ NP ( ⊗ NNLO ) jj =m μ NNLO ( ) jj =m μ EWK ( ⊗ NP ⊗ NNLO (GeV) T,avg p 0.8 1 1.2 1.4 1.6 1.8 Ratio to NLO 200 300 500 1000 2000 NNLOJET 8 TeV (R=0.7) t anti-k < 2 1 < y* < 2 b 1 < y MMHT2014_nnlo -1 = 19.7 fb int CMS L ) jj =m μ NLO ( ) jj =m μ NP ( ⊗ NNLO ) jj =m μ NNLO ( ) jj =m μ EWK ( ⊗ NP ⊗ NNLO (GeV) T,avg p 0.8 1 1.2 1.4 1.6 1.8 Ratio to NLO 200 300 500 1000 2000 NNLOJET 8 TeV (R=0.7) t anti-k < 3 0 < y* < 1 b 2 < y MMHT2014_nnlo -1 = 19.7 fb int CMS L ) jj =m μ NLO ( ) jj =m μ NP ( ⊗ NNLO ) jj =m μ NNLO ( ) jj =m μ EWK ( ⊗ NP ⊗ NNLO FIG. PHYSICAL REVIEW LETTERS 123, 102001 (2019) PHYSICAL REVIEW LETTERS 123, 102001 (2019) (GeV) T,avg p 0.8 1 1.2 1.4 1.6 1.8 Ratio to NLO 200 300 500 1000 2000 NNLOJET 8 TeV (R=0.7) t anti-k < 1 1 < y* < 2 b 0 < y MMHT2014_nnlo -1 = 19.7 fb int CMS L ) jj =m μ NLO ( ) jj =m μ NP ( ⊗ NNLO ) jj =m μ NNLO ( ) jj =m μ EWK ( ⊗ NP ⊗ NNLO (GeV) T,avg p 0.8 1 1.2 1.4 1.6 1.8 Ratio to NLO 200 300 500 1000 2000 NNLOJET 8 TeV (R=0.7) t anti-k < 1 0 < y* < 1 b 0 < y MMHT2014_nnlo -1 = 19.7 fb int CMS L ) jj =m μ NLO ( ) jj =m μ NP ( ⊗ NNLO ) jj =m μ NNLO ( ) jj =m μ EWK ( ⊗ NP ⊗ NNLO (GeV) T,avg p 0.8 1 1.2 1.4 1.6 1.8 Ratio to NLO 200 300 500 1000 2000 NNLOJET 8 TeV (R=0.7) t anti-k < 2 0 < y* < 1 b 1 < y MMHT2014_nnlo -1 = 19.7 fb int CMS L ) jj =m μ NLO ( ) jj =m μ NP ( ⊗ NNLO ) jj =m μ NNLO ( ) jj =m μ EWK ( ⊗ NP ⊗ NNLO (GeV) T,avg p 0.8 1 1.2 1.4 1.6 1.8 Ratio to NLO 200 300 500 1000 2000 NNLOJET 8 TeV (R=0.7) t anti-k < 1 0 < y* < 1 b 0 < y MMHT2014_nnlo -1 = 19.7 fb int CMS L ) jj =m μ NLO ( ) jj =m μ NP ( ⊗ NNLO ) jj =m μ NNLO ( ) jj =m μ EWK ( ⊗ NP ⊗ NNLO (GeV) T,avg p 0.8 1 1.2 1.4 1.6 1.8 Ratio to NLO 200 300 500 1000 2000 NNLOJET 8 TeV (R=0.7) t anti-k < 1 2 < y* < 3 b 0 < y MMHT2014_nnlo -1 = 19.7 fb int CMS L ) jj =m μ NLO ( ) jj =m μ NP ( ⊗ NNLO ) jj =m μ NNLO ( ) jj =m μ EWK ( ⊗ NP ⊗ NNLO Ratio to NLO (GeV) T,avg p 0.8 1 1.2 1.4 1.6 1.8 Ratio to NLO 200 300 500 1000 2000 NNLOJET 8 TeV (R=0.7) t anti-k < 2 1 < y* < 2 b 1 < y MMHT2014_nnlo -1 = 19.7 fb int CMS L ) jj =m μ NLO ( ) jj =m μ NP ( ⊗ NNLO ) jj =m μ NNLO ( ) jj =m μ EWK ( ⊗ NP ⊗ NNLO (GeV) T,avg p 0.8 1 1.2 1.4 1.6 1.8 Ratio to NLO 200 300 500 1000 2000 NNLOJET 8 TeV (R=0.7) t anti-k < 3 0 < y* < 1 b 2 < y MMHT2014_nnlo -1 = 19.7 fb int CMS L ) jj =m μ NLO ( ) jj =m μ NP ( ⊗ NNLO ) jj =m μ NNLO ( ) jj =m μ EWK ( ⊗ NP ⊗ NNLO Ratio to NLO Ratio to NLO FIG. [1] W. T. Giele, E. W. N. Glover, and D. A. Kosower, Phys. Rev. Lett. 73, 2019 (1994). [2] T. Affolder et al. (CDF Collaboration), Phys. Rev. D 64, 012001 (2001); 65, 039902(E) (2002). [3] A. M. Sirunyan et al. (CMS Collaboration), Eur. Phys. J. C 77, 746 (2017). PHYSICAL REVIEW LETTERS 123, 102001 (2019) 4. The NLO (blue) and NNLO (green) theory predictions and CMS data normalized to the NLO central value. Parton-level predictions corrected for nonperturbative (NP) effects and combined NP and electroweak effects (EW) effects, implemented as a multiplicative factor to the NNLO result, are shown in pink and red. The shaded bands shown for the NLO and the NNLO ⊗NP ⊗ EWK predictions represent the variation of the theoretical scales in the numerator by factors of 0.5 and 2. The uncertainty in the data is the total experimental uncertainty, including systematic and statistical uncertainties added in quadrature. PHYSICAL REVIEW LETTERS 123, 102001 (2019) The triple differential dijet cross section as a function of pT;avg for the six bins of y, yb at NNLO with central scale choice μ ¼ mjj, compared to CMS 8 TeV 19.7 fb−1 data. 102001-3 102001-3 PHYSICAL REVIEW LETTERS 123, 102001 (2019) PHYSICAL REVIEW LETTERS 123, 102001 (2019) 4, which correspond to the highly boosted and PDF-sensitive kinematical bins, we observe that the data tends to be below the central value of the MMHT2014 PDF set even at NNLO, in particular at large values of pT;avg:. In this region, which is sensitive to the scattering of large-x parton on a low-x parton (see Fig. 1), the PDFs suffer from large uncertainties. Our results at NNLO suggest that the understanding of the high- x behavior of the PDFs can be improved upon by including measurements of triple differential dijet distributions in future global PDF determinations. In this Letter, we computed the second-order QCD corrections to the triple-differential dijet production cross section at hadron colliders. Our results substantially improve the theoretical description of this benchmark observable, with theory uncertainties now being 102001-4 PHYSICAL REVIEW LETTERS 123, 102001 (2019) PHYSICAL REVIEW LETTERS 123, 102001 (2019) [12] J. Currie, E. W. N. Glover, and S. Wells, J. High Energy Phys. 04 (2013) 066. [4] S. D. Ellis, Z. Kunszt, and D. E. Soper, Phys. Rev. Lett. 69, 1496 (1992). ( ) [5] Z. Nagy, Phys. Rev. Lett. 88, 122003 (2002). y [13] J. Currie, E. W. N. Glover, and J. Pires, Phys. Rev. Lett. 118, 072002 (2017). [6] S. Alioli, K. Hamilton, P. Nason, C. Oleari, and E. Re, J. High Energy Phys. 04 (2011) 081. [14] J. Currie, A. Gehrmann-De Ridder, T. Gehrmann, E. W. N. Glover, A. Huss, and J. Pires, J. High Energy Phys. 10 (2018) 155. [7] S. Dittmaier, A. Huss, and C. Speckner, J. High Energy Phys. 11 (2012) 095. [15] J. Currie, A. Gehrmann-De Ridder, T. Gehrmann, E. W. N. Glover, A. Huss, and J. Pires, Phys. Rev. Lett. 119, 152001 (2017). [8] J. M. Campbell, D. Wackeroth, and J. Zhou, Phys. Rev. D 94, 093009 (2016). [9] R. Frederix, S. Frixione, V. Hirschi, D. Pagani, H.-S. Shao, and M. Zaro, J. High Energy Phys. 04 (2017) 076. [16] L. A. Harland-Lang, A. D. Martin, P. Motylinski, and R. S. Thorne, Eur. Phys. J. C 75, 204 (2015). [10] A. Gehrmann-De Ridder, T. Gehrmann, and E. W. N. Glover, J. High Energy Phys. 09 (2005) 056. [17] J. Bellm et al., arXiv:1903.12563. [11] A. Daleo, T. Gehrmann, and D. Maitre, J. High Energy Phys. 04 (2007) 016. [18] M. Cacciari, G. P. Salam, and G. Soyez, J. High Energy Phys. 04 (2008) 063. 102001-5
https://openalex.org/W4214764399	https://www.frontiersin.org/articles/10.3389/fpls.2022.834416/pdf	English	null	Ancient Artworks and Crocus Genetics Both Support Saffron’s Origin in Early Greece	Frontiers in plant science	2,022	cc-by	9,514	Correspondence: Tony Heitkam tony.heitkam@tu-dresden.de Correspondence: Tony Heitkam tony.heitkam@tu-dresden.de Tony Heitkam tony.heitkam@tu-dresden.de †ORCID: Seyyedeh-Sanam Kazemi-Shahandashti orcid.org/0000-0001-8847-2720 Ludwig Mann orcid.org/0000-0001-8756-8106 Abdullah El-nagish orcid.org/0000-0002-5704-5475 Dörte Harpke orcid.org/0000-0003-1667-2912 Zahra Nemati orcid.org/0000-0002-4944-314X Björn Usadel orcid.org/0000-0003-0921-8041 Tony Heitkam orcid.org/0000-0003-0168-8428 ‡These authors have contributed equally to this work ‡These authors have contributed equally to this work ‡These authors have contributed equally to this work Keywords: saffron, Crocus sativus, #ArtGenetics, Genome, cytogenetics, historical art, Minoan frescoes Specialty section: This article was submitted to Plant Systematics and Evolution, a section of the journal Frontiers in Plant Science Specialty section: This article was submitted to Plant Systematics and Evolution, a section of the journal Frontiers in Plant Science REVIEW published: 25 February 2022 doi: 10.3389/fpls.2022.834416 REVIEW INTRODUCTION From Greece to Iran, from paintings and dyes to spice and perfumes, saffron’s flavor and sparkly yellow color has made its trail in human history. As saffron crocus (Crocus sativus L.) can be traced in artworks across history, its appearance in various arts and crafts gives insights into saffron’s early geographic distribution and finally into its origin.h Received: 13 December 2021 Accepted: 31 January 2022 Published: 25 February 2022 fi The Mediterranean is considered as the emergence site of many diploid Crocus species (Maw, 1886; Mathew, 1982), with Greece and Turkey possessing the highest number of Crocus taxa. Investigating ancient texts and artworks in these regions can help finding more information on early saffron domestication.if Keywords: saffron, Crocus sativus, #ArtGenetics, Genome, cytogenetics, historical art, Minoan frescoes Edited by: Ive De Smet, Vlaams Instituut voor Biotechnologie, Belgium Edited by: Ive De Smet, Vlaams Instituut voor Biotechnologie, Belgium Seyyedeh-Sanam Kazemi-Shahandashti 1,2†‡, Ludwig Mann 3†‡, Abdullah El-nagish 3,4†‡, Dörte Harpke 5†, Zahra Nemati 5†, Björn Usadel 1,2† and Tony Heitkam 3† Seyyedeh-Sanam Kazemi-Shahandashti 1,2†‡, Ludwig Mann 3†‡, Abdullah El-nagish 3,4†‡, Dörte Harpke 5†, Zahra Nemati 5†, Björn Usadel 1,2† and Tony Heitkam 3† Reviewed by: Irwin Goldman, University of Wisconsin-Madison, United States Ludmila Tsatsenko, Kuban State Agrarian University named after I.T. Trubilin, Russia 1 Institute of Biological Data Science, Heinrich-Heine Universität Düsseldorf, Düsseldorf, Germany, 2 IBG-4 Bioinformatics, Forschungszentrum Jülich, Jülich, Germany, 3 Faculty of Biology, Institute of Botany, Technische Universität Dresden, Dresden, Germany, 4 Botany and Microbiology Department, Faculty of Science, Sohag University, Sohag, Egypt, 5 Leibniz Institute of Plant Genetics and Crop Plant Research (IPK), Gatersleben, Germany Saffron crocus (Crocus sativus) is a male-sterile, triploid flower crop, and source of the spice and colorant saffron. For over three millennia, it was cultivated across the Mediterranean, including ancient Greece, Persia, and other cultures, later spreading all over the world. Despite saffron crocus’ early omnipresence, its origin has been the matter of a century-old debate, in terms of area and time as well as parental species contribution. While remnants of the ancient arts, crafts, and texts still provide hints on its origin, modern genetics has the potential to efficiently follow these leads, thus shedding light on new possible lines of descent. In this review, we follow ancient arts and recent genetics to trace the evolutionary origin of saffron crocus. We focus on the place and time of saffron domestication and cultivation, and address its presumed autopolyploid origin involving cytotypes of wild Crocus cartwrightianus. Both ancient arts from Greece, Iran, and Mesopotamia as well as recent cytogenetic and comparative next-generation sequencing approaches point to saffron’s emergence and domestication in ancient Greece, showing how both disciplines converge in tracing its origin. EARLY EVIDENCE OF SAFFRON’S ORIGIN STEMS FROM ANCIENT ARTWORKS Evidence for the use and cultivation of saffron have been found across different cultures and times. Here, we will follow saffron’s traces across ancient artworks in a chronological manner.h g The oldest evidence of crocus usage through mankind was detected in a 50,000-year-old depiction of beasts in a cave in today’s Iraq, where saffron-related pigments (including crocin) have been used (Caiola and Canini, 2010). Wild-grown “historic saffron” (“historic saffron”: any crocus species with at least partially similar traits as today’s cultivated saffron and used by humans for this particular trait) was reported to have been used by Sumerians (ca. 4100–1750 BCE) as part of their remedies and medical potions (Willard, 2002). Also, royals of old Assyria and Babylonia (2nd millennium BCE) used “historic saffron” as a treatment for several diseases and had their courts filled with textiles dyed with “historic saffron” or later possibly cultivated saffron, which were supplied by Phoenician traders (Basker and Negbi, 1983; Mousavi and Bathaie, 2011; Dewan, 2015).hf While the use of crocus-based pigments can be traced back about 50,000 years ago to prehistoric cave paintings in northwestern Iran (today’s Iraq; Humphries, 1996; Figures 1A,B), early signs of cultivation and domestication were found later, at about 1700 before the common era (BCE), during the time of the Minoan civilization in Crete (Deo, 2003; Dewan, 2015). As saffron’s high medicinal value and antioxidant ability were recognized, its commercial value as a spice increased over the next eras, leading to its spread across the Mediterranean (Abrishami, 2004; Caiola and Canini, 2010).f Although saffron is in high demand as spice and colorant these days, its cultivation still relies on traditional knowledge and has remained largely untouched by industrialization. This is rooted in saffron’s genetic constitution as a male-sterile triploid, which can only be propagated vegetatively. Currently, any alteration in saffron production also requires change in its vegetative propagation and quality control (Caiola and Canini, 2010). In contrast, sexual propagation of saffron would result in new, desirable traits, but would involve the selection of parental genotypes. To inform the required selection process, knowledge of saffron’s genetic and geographic origin is crucial. To fill this gap, over the last decade, several studies have attempted to resolve saffron’s domestication and to clarify its auto- or allopolyploid origin. Citation: The major difference here is that one type provides scientific evidence for the time scale, whereas the other type leaves room for interpretation and speculation. Hence, the visual analysis depends on multiple aspects like placing the art in the socio-historical context, knowledge on artistic techniques as well as possible restoration approaches. While the use of crocus-based pigments can be traced back about 50,000 years ago to prehistoric cave paintings in northwestern Iran (today’s Iraq; Humphries, 1996; Figures 1A,B), early signs of cultivation and domestication were found later, at about 1700 before the common era (BCE), during the time of the Minoan civilization in Crete (Deo, 2003; Dewan, 2015). As saffron’s high medicinal value and antioxidant ability were recognized, its commercial value as a spice increased over the next eras, leading to its spread across the Mediterranean (Abrishami, 2004; Caiola and Canini, 2010). Regarding ancient arts, images can be categorized into two types: On one hand, those that are painted using pigments derived from crocus and on the other hand, those that depict saffron crocus flowers. The first type of image can give us information about the probable timeline of saffron use, whereas the second may give us information about its origin as it provides access to the morphological aspects of the crocus species used at the time. The major difference here is that one type provides scientific evidence for the time scale, whereas the other type leaves room for interpretation and speculation. Hence, the visual analysis depends on multiple aspects like placing the art in the socio-historical context, knowledge on artistic techniques as well as possible restoration approaches. Frontiers in Plant Science \| www.frontiersin.org Citation: Kazemi-Shahandashti S-S, Mann L, El-nagish A, Harpke D, Nemati Z, Usadel B and Heitkam T (2022) Ancient Artworks and Crocus Genetics Both Support Saffron’s Origin in Early Greece. Front. Plant Sci. 13:834416. doi: 10.3389/fpls.2022.834416 f Regarding ancient texts, the first use of the word “saffron” dates back to the 12th century to the old French term safran that consecutively originated from the Latin safranum, the Arabic za’farān, and the Persian zarparan with the meaning “gold strung” (Asbaghi, 1988). As folk taxonomies were used to differentiate plant species prior to the formalization of the February 2022 \| Volume 13 \| Article 834416 1 Frontiers in Plant Science \| www.frontiersin.org Tracing Saffron: Artworks and Genetics Kazemi Shahandashti et al. Linnaean system, different species of purple, autumn-flowering crocuses were not distinguished before the 18th century (Day, 2011a). Hence, at the time, most crocuses were considered as one and we cannot rely on texts for clear species information. Nevertheless, Caiola and Canini (2010) suggested that Crocus cartwrightianus, Crocus thomasii, or Crocus hadriaticus were the most likely species that were cited by ancient historians. context with ancient arts. Especially as no ancient saffron samples have been uncovered from the Aegean time—that may potentially harbor aDNA—the ancient artworks are currently the only waypoints that may guide modern genomics efforts to fill in the remaining holes in saffron’s history: We go through time and explain how and where saffron use (and the use of other Crocus species) was described or depicted and compare the results with current genomics efforts. We put on a wider lens and draw a comprehensive picture of the current view on saffron’s history, as well as lay out what gaps are still open to be addressed in the future. cartwrightianus, Crocus thomasii, or Crocus hadriaticus were the most likely species that were cited by ancient historians. Regarding ancient arts, images can be categorized into two types: On one hand, those that are painted using pigments derived from crocus and on the other hand, those that depict saffron crocus flowers. The first type of image can give us information about the probable timeline of saffron use, whereas the second may give us information about its origin as it provides access to the morphological aspects of the crocus species used at the time. EARLY EVIDENCE OF SAFFRON’S ORIGIN STEMS FROM ANCIENT ARTWORKS The most prominent pictures of potentially cultivated saffron in historic artworks were created during the Aegean Bronze Age, particularly influenced by the Minoan civilization, dating back to 1700–1500 BCE (Deo, 2003; Dewan, 2015). Important excavation sites are located on the islands of Crete and Thera (today’s Santorini) both belonging to today’s Greece. For this review, we will mainly focus on the crocus depictions in ancient frescoes, but will also address ceramics and cloth found in Akrotiri (Thera) and Knossos (Crete; Warren, 2005).h y g Here, we illustrate how the insights gained from ancient artworks add to our knowledge of saffron’s origin, how they compare with the recent genetic approaches, and how they can steer further research on saffron evolution, domestication, and breeding. For this, we build on our previous work (Schmidt et al., 2019), in which we have used cytogenetics to derive the genetic origin of saffron from cytotypes of C. cartwrightianus, thereby excluding other Crocus species as parental contributors to its triploidy. In the related paper of Nemati et al. (2019), genotyping-by-sequencing (GBS) has been used to independently reach the same conclusion and to trace the genetic origin of saffron’s triploidy back to Greece. Now, we have combined forces to put the results of saffron’s genetic origin into the broader The most relevant representations for our review show the gathering of saffron (Figures 1C,D) and its usage in cultural rituals on multiple frescoes. These were found in remains of historical buildings (“Xeste 3,” “West House,” and others) in Akrotiri. The oldest saffron-depicting fresco (“The Saffron Gatherers,” sometimes referred to as “Blue Monkeys/Boys” fresco); however, was found in the so-called House of Frescoes in Knossos and was dated to approximately 1700 BCE. With the amount and prominence of saffron flowers depicted, these frescoes clearly demonstrate the importance of saffron around 1700–1500 BCE in the Aegean Bronze Age (Dewan, 2015). 1700–1500 BCE in the Aegean Bronze Age (Dewan, 2015). The crocuses depicted on the frescoes in Xeste 3 show distinctive similarities with modern cultivated saffron (C. sativus) February 2022 \| Volume 13 \| Article 834416 2 Tracing Saffron: Artworks and Genetics Kazemi-Shahandashti et al. h h l d l d h d l h ( b ) FIGURE 1 \| Cultivation of saffron (Crocus sativus L.) as traced by historic arts and modern genetics. (A) The timeline and map show four exemplary time-points (1–4) and their corresponding geographic locations. EARLY EVIDENCE OF SAFFRON’S ORIGIN STEMS FROM ANCIENT ARTWORKS (B) The earliest evidence of “historic saffron” (any crocus species with at least partially similar traits as today’s cultivated saffron and used by humans for this particular trait) use was found in pigments from cave wall paintings. These contained crocin (structure shown) that was derived from Crocus flowers. These cave paintings were found in today’s north western Iraq and were created about 50,000 years ago (tya). (C,D) The first signs of saffron cultivation were found in Aegean Bronze Age frescoes from Crete and Santorini, dating back to 1700 to 1500 before the common era (BCE). Shown here is a part of “The Saffron Gatherers” (C) and “Adorants” frescos (D), both found in Xeste 3 (Akrotiri, Santorini) from around 1600 BCE. (E) The first mentioning of saffron crocus propagation through corms was documented in Historia Plantarum by Theophrastus around 300 BCE. Shown here is the front cover of the Latin translation (1644). (F,G) Recent studies investigating the origin of saffron crocus using modern genetic approaches [multicolor FISH (F) and genotyping-by-sequencing GBS (G)] also point to early Greec as the place of origin, with Attica as the most likely region of origin. Countries that are major saffron-producers today are colored in red and purple on the map (Rukšāns, 2017). [Image sources: (A) The map was generated with Natural Earth (https://www.naturalearthdata.com/); (B) Crocin structure was made with ChemSketch; (C,D) “Th Saffron Gatherers” and “Adorants” pictures from Doumas (1992) p. 152/Figure 116, p. 136/Figure 100 (contrast and brightness enhanced, layered); (E) Historia Plantaru belongs to the public domain, image provided by Andrew Dalby (Wiki commons); (F) FISH image reused from Schmidt et al. (2019), Figure 2Q with permission from Joh Wiley and Sons; (G) Phylogenetic tree and cluster analysis reused from Nemati et al. (2019) under CC-BY]. The timeline is not to scale. FIGURE 1 \| Cultivation of saffron (Crocus sativus L.) as traced by historic arts and modern genetics. (A) The timeline and map show four exemplary time-points (1–4) and their corresponding geographic locations. (B) The earliest evidence of “historic saffron” (any crocus species with at least partially similar traits as today’s cultivated saffron and used by humans for this particular trait) use was found in pigments from cave wall paintings. These contained crocin (structure shown) that was derived from Crocus flowers. These cave paintings were found in today’s north western Iraq and were created about 50,000 years ago (tya). EARLY EVIDENCE OF SAFFRON’S ORIGIN STEMS FROM ANCIENT ARTWORKS It is particularly striking that these representations seem to show remarkable differences compared to most of the Crocus species natively occurring on Crete and the surrounding islands today, such as Crocus boryi, Crocus oreocreticus, Crocus sieberi, Crocus tournefortii, Crocus laevigatus (Greece mainland and Santorini), Crocus ruksansii (Karpathos), Crocus rhodensis (Rhodes), Crocus mazziaricus, Crocus olivieri, and Crocus pallasii (Greece to Turkey; Rukšāns, 2017). The only exception here is C. cartwrightianus (Crete to Greece mainland), which itself shows a huge variety of differing phenotypes and is also sometimes referred to as wild saffron (Larsen et al., 2015; Rukšāns, 2017; Nemati et al., 2019; Schmidt et al., 2019). Some of the C. cartwrightianus phenotypes found today look very similar to those depicted in the ancient frescoes. Hence, archeologists and (paleo-)botanists classified the flowers depicted in the frescoes as either C. sativus or C. cartwrightianus (Möbius, 1933; Warren, 2000; Day, 2011a). foot (Marinatos, 2015). While both is possible, we argue that this could indeed symbolize the human-led saffron domestication efforts in the Aegean Bronze Age. In the later Minoan ages, frescoes of saffron became less common, but it continued to be depicted on remains found of all kinds of ceramics and clothing (Ferrence and Bendersky, 2004; Day, 2011a; Dewan, 2015). Despite the most reliable evidence of the cultivation of saffron being found in the Aegean Bronze Age culture, saffron artworks have also been found around the same time in other countries, such as Egypt and Persia. In ancient Egypt, Theban tombs of five high officials whose career spanned the time from Queen Hatshepsut through the early reign of Amenophis III (New Kingdom; c. 1480–1380 BCE) depict men carrying textiles to be presented as a tribute. Style of ceilings and wall- paintings in these tombs is Aegean and reflects artistic motifs derived from saffron-colored textiles (Wachsmann and Wachsmann 1987; Barber 1991; Dewan 2015). The individuals appearing in these paintings are dressed in a stylized version of Aegean Bronze Age costumes that resemble those in the Procession Fresco from Knossos (Rehak 1996). The adjacent hieroglyphic inscriptions which appear above those men geographically identifies them as Aegean natives, the Keftiu, which is the Egyptian name for Crete, and describes them as people living in the “Islands in the midst of the sea” (Wachsmann and Wachsmann, 1987; Barber, 1991; Panagiotopoulos, 2002; Matić, 2019). EARLY EVIDENCE OF SAFFRON’S ORIGIN STEMS FROM ANCIENT ARTWORKS Although no record of saffron cultivation by Ancient Egyptians can be found, detailed historic information of saffron usage for several medical purposes is mentioned in the Papyrus Ebers (1550 BCE; Toellner, 2000; Mousavi and Bathaie, 2011). Moreover, the rareness of saffron-dyed Minoan textiles in Egypt reflects why they were considered ideal as elite gifts to Egyptian royals and support the notion that saffron products was mainly imported from Crete which was home to a thriving textile industry (Feldman, 2002; Mousavi and Bathaie, 2011; Dewan, 2015). No remains of bulbs or other parts of the actual plants have been found in the various excavation sites making it impossible to verify which species was depicted on the artistic works (Day, 2013). The chances to find any remains in the future seem very low considering that volcanic activities (“Minoan eruption”) destroyed major parts of the island Thera (Friedrich, 2009). During this geographical event also the surrounding islands including Crete have been influenced by earthquakes and tsunamis (Friedrich, 2009).h The importance of crocus use as highlighted on the frescoes in Xeste 3 allow for some speculation: The depicted crocus plants are often planted in clumps that could be interpreted as a sort of cultivation (Morgan, 1988; Figure 1C: “The Saffron Gatherers”). Moreover, the previously mentioned Linear B tablets also include information on large amounts of traded saffron, with many dedicated fabric-processing tools found in Akrotiri (Day, 2011a). Additionally, its theorized, cosmetic use for the coloring of lips and ears during festivities was also depicted in the “Young Priestess” fresco (West House, Akrotiri; Ferrence and Bendersky, 2004; Day, 2011a; Dewan, 2015). This suggests that saffron was economically important, already forming a specialized branch of industry. Likely, this also implies that saffron collection from the wild was insufficient to meet the high demands for saffron as spice, dye, and cosmetic product, but that saffron was instead cultivated in gardens or fields. Historic data indicate that Persian “historic saffron” (possibly C. haussknechtii) was cultivated in Derbena, Khorasan by the 10th century BCE and also during the Median Kingdom (708–550 BCE) near Zagros and Alvand mountains, suggesting that ancient Persians were one of the early nations who cultivated saffron instead of gathering wild crocus flowers (Dadkhah et al., 2003; Caiola and Canini, 2010; Mousavi and Bathaie, 2011). EARLY EVIDENCE OF SAFFRON’S ORIGIN STEMS FROM ANCIENT ARTWORKS (C,D) The first signs of saffron cultivation were found in Aegean Bronze Age frescoes from Crete and Santorini, dating back to 1700 to 1500 before the common era (BCE). Shown here is a part of “The Saffron Gatherers” (C) and “Adorants” frescos (D), both found in Xeste 3 (Akrotiri, Santorini) from around 1600 BCE. (E) The first mentioning of saffron crocus propagation through corms was documented in Historia Plantarum by Theophrastus around 300 BCE. Shown here is the front cover of the Latin translation (1644). (F,G) Recent studies investigating the origin of saffron crocus using modern genetic approaches [multicolor FISH (F) and genotyping-by-sequencing GBS (G)] also point to early Greece as the place of origin, with Attica as the most likely region of origin. Countries that are major saffron-producers today are colored in red and purple on the map (Rukšāns, 2017). [Image sources: (A) The map was generated with Natural Earth (https://www.naturalearthdata.com/); (B) Crocin structure was made with ChemSketch; (C,D) “The Saffron Gatherers” and “Adorants” pictures from Doumas (1992) p. 152/Figure 116, p. 136/Figure 100 (contrast and brightness enhanced, layered); (E) Historia Plantarum belongs to the public domain, image provided by Andrew Dalby (Wiki commons); (F) FISH image reused from Schmidt et al. (2019), Figure 2Q with permission from John Wiley and Sons; (G) Phylogenetic tree and cluster analysis reused from Nemati et al. (2019) under CC-BY]. The timeline is not to scale. such as their long and intensively red stigmas overtopping the dark violet petals. These visual characteristics are also found on other artistic representations of crocus flowers on ceramics and cloth (Day, 2011b; Dewan, 2015; Marinatos, 2015). Typical for this time was the syllabic script Linear B (Shelmerdine, 1998). The ideogram (script character) representing saffron February 2022 \| Volume 13 \| Article 834416 Frontiers in Plant Science \| www.frontiersin.org 3 Tracing Saffron: Artworks and Genetics Kazemi-Shahandashti et al. shows very long stigmas overtopping three pictured petals (Day, 2011a; Dewan, 2015) matching all afore-mentioned representations as well as today’s cultivated saffron. This leads to the assumption that most likely all depicted phenotypes across Akrotiri and other excavation sites belong to only a single Crocus species. Frontiers in Plant Science \| www.frontiersin.org The Debated Origin of Saffron Crocusf The Debated Origin of Saffron Crocus Regarding saffron crocus’ emergence by triploidization, a major question targets the nature of its triploidy. Various types of polyploidy are commonly observed in plants, including, for example, allopolyploids, segmental allopolyploids, and autopolyploids (Figure 2). Whereas allopolyploidy (at least two parental species contribute to the polyploid state) and autopolyploidy (only one parental species contributes to the polyploid state) denote extreme cases along a gradual spectrum, segmental allopolyploids are considered as an intermediate form between true autopolyploidy and allopolyploidy (Stebbins, 1947, 1950; Sattler et al., 2016; Mason and Wendel, 2020). For saffron crocus, two main scenarios were put forward to explain the origin of saffron’s triploidy (see also Figure 2): I. Allotriploidy, originating by hybridization between individuals of different species (Frello and Heslop-Harrison, 2000; Alavi-Kia et al., 2008; Petersen et al., 2008; Tsaftaris et al., 2011; Gismondi et al., 2013; Erol et al., 2014; Alsayied et al., 2015); and EARLY EVIDENCE OF SAFFRON’S ORIGIN STEMS FROM ANCIENT ARTWORKS The main applications of saffron in Persian art were as a dye in royal carpets and funeral shrouds as well as in paper colorings and Persian miniature paintings to prevent the corrosive effect of the verdigris pigments (Willard, 2002; Barkeshli, 2016; Dehboneh et al., 2019). fi One of the frescoes in Xeste 3 termed the “Adorants” fresco (Figure 1D) clearly shows a crocus flower harboring all typical saffron crocus traits with the bright orange/red stigmas highlighted. The woman sitting has a cut in her foot with a saffron flower depicted below. There are two interpretations on the meaning of this fresco discussed in the literature. One is that the blood colors the saffron’s stigmas red which might be seen as a sign of reaching female adulthood including menstruation (Rehak, 2004; Dewan, 2015). Others argue that the women is a goddess and the flower grows from her wounded Despite these many occurrences across artworks of the ancient world, the first written evidence of cultivated saffron (C. sativus) is only found around 350–300 BCE in the Historia Plantarum (Theophrastus, n.d.; Figure 1E), where the saffron- specific propagation through corms is described in detail. Hence, understanding and interpreting the ancient arts can push this time limit back for over a millennium, assisting us in formulating theories, and enabling modern saffron genetics to follow and verify these leads. February 2022 \| Volume 13 \| Article 834416 4 Tracing Saffron: Artworks and Genetics Kazemi-Shahandashti et al. GENETICS PINPOINTS THE ORIGIN OF SAFFRON CROCUS TO ANCIENT GREECE such as simple sequence repeats (SSRs), EST-derived SSRs, and amplified fragment length polymorphisms (AFLPs; Fluch et al., 2010; Siracusa et al., 2013). Hence, it is generally accepted that triploid saffron crocus emerged only once and was then distributed across the world (Rubio-Moraga et al., 2009; Fluch et al., 2010; Siracusa et al., 2013; Babaei et al., 2014; Alsayied et al., 2015; Nemati et al., 2019). However, a recent study by Busconi et al. (2021) reported a surprising amount of genetic differences (single- nucleotide polymorphisms, SNPs). This was detected by a genome- wide approach, in which primarily the global DNA methylation across five different saffron accessions was analyzed and a high epigenetic variability has been shown. These findings indicate that saffron has a higher genetic variability than previously assumed or detected. Moreover, the study confirmed the previously suggested epigenetic variation within accessions according to their geographic origin. Using canonical and methylation-sensitive AFLP (MS-AFLP) markers, 112 accessions from the World Saffron and Crocus Collection were compared before, yielding only low genetic but high epigenetic variability (Busconi et al., 2015). After prolonged co-cultivation of different saffron accessions in the same environment, their DNA methylation profiles converged, suggesting that saffron epi-genotypes are a result of adaptation to the environment (Busconi et al., 2018). This suggests that epigenetics, especially DNA methylation are likely the main cause for phenotypic variability within saffron accessions with genetic modification playing a yet not fully resolved role. A major key point to overcome the current lack of knowledge will be the establishment of a reference genome sequence. Although saffron’s origin has been under debate for over a century, with a large body of work attempting to clarify its genetic and geographic origin (reviewed in Koocheki and Khajeh-Hosseini, 2020), the history of saffron domestication remained unresolved until recently. Instead of chromosome pairs, saffron has eight chromosome triplets with a chromosome configuration of 2n = 3x = 24 and a triploid genome size of 10.5 Gb (Chichiriccò, 1984; Brandizzi and Caiola, 1998; Agayev, 2002; Schmidt et al., 2019). Saffron’s triploidy is the cause for many genetic peculiarities of saffron. Saffron Is Infertile and Can Only Be Propagated Vegetativelyhf The triploidy of saffron crocus results in a disturbed meiosis leading to at least partial sterility. As result of erroneous chromosome pairing, meiosis progresses only incompletely and yields abnormal pollen (Chichiriccò, 1984; Rashed-Mohassel, 2020). In consequence, cross-fertilization between C. sativus and other species is limited (Caiola, 1999; Caiola et al., 2000). Nevertheless, as the C. sativus pollen tube is incapable to penetrate the C. sativus ovule, a self- incompatible species was suggested as the most likely progenitor (Chichiriccò and Caiola, 1986). Although pollen infertility is higher than the ovule sterility (Chichiriccò, 1984), irregular chromosome arrays also occur in megaspores, making them genetically unbalanced and infertile (Caiola and Canini, 2010). Along the same lines, haploid gametes with a whole set of chromosomes only infrequently form in triploids; another cause of saffron’s sterility. A genetic regulation of chromosome segregation as was observed for hexaploid bread wheat (Riley and Chapman, 1958; Sears, 1976; Koo et al., 2017) or postulated for pentaploid dogroses (Täckholm, 1920; Herklotz and Ritz, 2017) is either absent or yet undiscovered. Similarly, there have been no reports of hexaploid saffron that may have emerged from triploid saffron as it was occasionally observed for other triploid plant species that used hexaploidization as a route to regain fertility (Husband, 2004). Therefore, the vegetative propagation by daughter corms is considered as the only way of saffron reproduction (Fernández, 2004; Gresta et al., 2008; Nehvi et al., 2010). Nevertheless, the multiplication rate of daughter corms reduces saffron productivity, rendering high-quality propagation material crucially important (Renau-Morata et al., 2013). Absence of Genetic Variation Among Saffron Accessions Suggests That Saffron Only Originated Onceff How this triploidy emerged has been debated for over a century, with two main hypotheses being brought forward: On the one hand, saffron may have originated with genome contributions from a single species, likely Crocus carthwrightianus (autotriploid scenario). On the other hand, at least two species may have contributed to the genome of saffron crocus, with one likely being C. cartwrightianus (allotriploid scenario). Both scenarios represent extreme cases along a gradual spectrum from auto- toward allopolyploidy, as indicated by the gradual arrangement in the figure. The current position of saffron supported by several recent studies tends to be more on the autotriploid side of the spectrum of polyploidy (Nemati et al., 2019; Schmidt et al., 2019). The capital letters represent sub-genomes from different species, with different cytotypes being represented by different typefaces. The representation of the spectrum of polyploidy using letter triplets only serves as symbolic visualization and may not represent all possible scenarios. the most probable ancestors (Brighton, 1977; Mathew, 1999). Intriguingly, Agayev (2002) and Agayev et al. (2010) showed that the triplet 5 in C. sativus contains heteromorphic chromosomes, already an indicator that at least two different genotypes have been involved in saffron’s domestication history. Considering allotriploidy as the origin of saffron prompted scientists to focus on its probable progenitors: Most consider C. cartwrightianus as the most probable donor of the diploid genome and either C. thomasii, C. pallasii, or C. haussknechtii, as possible second parent (Frello and Heslop-Harrison, 2000; Alavi-Kia et al., 2008; Petersen et al., 2008; Tsaftaris et al., 2011; Gismondi et al., 2013; Erol et al., 2014; Alsayied et al., 2015). cartwrightianus, also called wild saffron, often has shorter and lighter stigmas with a less prominent aroma (Mathew, 1999; Negbi and Negbi, 2002; Jacobsen and Ørgaard, 2004). Its stigmas have served as a “historic saffron” source even before the domestication and trait enhancements of C. sativus. The close relationship between C. sativus and C. cartwrightianus has been confirmed by multiple observations, such as their similar phenotypes (Mathew, 1977; Caiola et al., 2010), similar pigment composition (Nørbæk et al., 2002), as well as similar genome size and base composition (Brandizzi and Caiola, 1998; Schmidt et al., 2019). Due to these many similarities, it was already suggested four decades ago that C. cartwrightianus is one of the most likely progenitors of saffron crocus (Mathew, 1982). However, the contribution of additional species remained under debate. Absence of Genetic Variation Among Saffron Accessions Suggests That Saffron Only Originated Onceff y g As saffron is effectively sterile, it cannot generate genetic variation through recombination during sexual reproduction (Fernández, 2004). Intriguingly, phenotypic differences are still accumulating in today’s saffron accessions. After clonal selection, Agayev et al. (2009) could for instance select a population of corms that was characterized by very different phenotypes. The majority of studies only detected insignificant amounts of genetic variability among saffron accessions, if at all, using molecular marker technologies, II. Autotriploidy, originating by hybridization between individuals of the same species (Mathew, 1999; Negbi and Negbi, 2002; Jacobsen and Ørgaard, 2004; Nemati et al., 2019; Schmidt et al., 2019). Regardless of the scenario, at least one parent is generally assumed to be C. cartwrightianus, a species occurring in southeastern mainland Greece and on Aegean islands. Crocus February 2022 \| Volume 13 \| Article 834416 Frontiers in Plant Science \| www.frontiersin.org 5 Tracing Saffron: Artworks and Genetics Kazemi-Shahandashti et al. FIGURE 2 \| Scenarios for the auto- or allotriploid origin of saffron crocus (Crocus sativus L.). Saffron crocus is triploid and sterile and, hence, can only be propagated vegetatively through corms. How this triploidy emerged has been debated for over a century, with two main hypotheses being brought forward: On the one hand, saffron may have originated with genome contributions from a single species, likely Crocus carthwrightianus (autotriploid scenario). On the other hand, at least two species may have contributed to the genome of saffron crocus, with one likely being C. cartwrightianus (allotriploid scenario). Both scenarios represent extreme cases along a gradual spectrum from auto- toward allopolyploidy, as indicated by the gradual arrangement in the figure. The current position of saffron supported by several recent studies tends to be more on the autotriploid side of the spectrum of polyploidy (Nemati et al., 2019; Schmidt et al., 2019). The capital letters represent sub-genomes from different species, with different cytotypes being represented by different typefaces. The representation of the spectrum of polyploidy using letter triplets only serves as symbolic visualization and may not represent all possible scenarios. FIGURE 2 \| Scenarios for the auto- or allotriploid origin of saffron crocus (Crocus sativus L.). Saffron crocus is triploid and sterile and, hence, can only be propagated vegetatively through corms. Frontiers in Plant Science \| www.frontiersin.org Absence of Genetic Variation Among Saffron Accessions Suggests That Saffron Only Originated Onceff Building on what we can learn from the ancient arts, further genomics approaches will likely result in clarifying the impact of early saffron domestication on its genome and also gives us a better outlook on the genetic and epigenetic basis of saffron’s quality traits that can be eventually implemented in the improvement of saffron quality and production. Absence of Genetic Variation Among Saffron Accessions Suggests That Saffron Only Originated Onceff sativus as a hybrid of heterogeneous C. cartwrightianus cytotypes (Figure 1F). This was independently verified by Nemati et al. (2019), whose GBS experiments suggested that saffron crocus evolved in Attica as a hybrid of two different genotypes of C. cartwrightianus (Figure 1G). Both studies concluded that the natural variation among different cyto- and genotypes of C. cartwrightianus was sufficient to explain the high level of heterozygosity within C. sativus (Figure 2). Over 99% of saffron’s GBS alleles were found in C. cartwrightianus, with almost 98% present in accessions found in the region of Attica (Nemati et al., 2019). However, as allele composition is constantly reorganized by genetic recombination, it is unlikely to identify the exact allele combination characteristic for triploid C. sativus (Nemati et al., 2019). Concluding, the most recent genetic studies pinpoint the origin of cultivated saffron to the region of Attica (Greece).f Nevertheless, some mysteries of saffron’s past still remain: None of the studies could yet explain the origin of the heteromorphic chromosome in triplet 5. We showed before that this chromosome has a different genomic makeup and speculate that this unusual chromosome morphology has been the result of an introgression prior to saffron’s triploidization (Schmidt et al., 2019). In any case, we are just beginning to understand the genetics and epigenetics of triploid saffron crocus that underlie its sterility, clonal growth, and phenotypic variability. Especially the identification of saffron’s parentage as well as the understanding of the epigenetic impact on phenotypes may open the way toward saffron breeding. As discussed above, the phenotypic differences among saffron accessions likely originate from mostly epigenetic variability and do not point to multiple independent origins. In the future, comparative epigenomics will give a better understanding of saffron’s genetic and epigenetic composition, and offer new insights on its parental species contribution. Nevertheless, the generation of a full genome assembly will face considerable challenges, including saffron’s triploidy, the high level of genome heterozygosity, its large genome, and its high percentage of repetitive DNA. Similarly, the stability of saffron’s divergent phenotypes in consecutive growing seasons indicates that epigenetics may play a large role in phenotype breeding. COMBINING ART AND GENETICS: WHAT WE CAN LEARN FROM THE ARTS TO INFORM GENETIC APPROACHES Saffron crocus is the source of the most expensive spice in the world but is—unlike most other cash crops—inaccessible to classical breeding approaches. Investigating saffron crocus’ depictions in ancient arts and combining it with the current genetics can lead to insights into its early cultivation and geographical origin, opening up new breeding avenues for better equipped saffron varieties. Comparing ancient arts and modern genetics gave us symmetric results: Historians cited C. sativus, C. cartwrightianus, or C. haussknechtii as the most likely species that are shown in ancient arts, including crafts from various origins (Iran, Greece, Egypt, etc.). Importantly, flowers depicted in Aegean Bronze Age frescoes were classified by archeologists and (paleo-) botanists as either C. sativus or C. cartwrightianus, showing that at least one of those species occurred in the area of today’s Greece, providing early signs of cultivation and economic relevance. For example, the “Adorants” fresco from Xeste 3 Absence of Genetic Variation Among Saffron Accessions Suggests That Saffron Only Originated Onceff Contrary to this hypothesis, the autotriploidy (scenario II) of saffron crocus is also supported by a range of observations: Especially the high similarity between C. sativus and C. cartwrightianus suggests that C. cartwrightianus was the sole saffron progenitor, contributing all three chromosome sets p Favoring allotriploidy (scenario I), Caiola (1999) suggested the out-breeding of crocus diploids, as the origin of C. sativus. C. cartwrightianus, and C. thomasii were then introduced as February 2022 \| Volume 13 \| Article 834416 6 Tracing Saffron: Artworks and Genetics Kazemi-Shahandashti et al. depicts crocus flowers in great detail, highlighting the typical dark-red stigmas overtopping the flower (Figure 1D), a trait that is fixed in C. sativus, but can also occur in the more variable C. cartwrightianus. Together with the clumps of flowers seen in “The Saffron Gatherers” (Figure 1C) the frescos could even be interpreted as an artistic rendition of the human influence on saffron’s domestication. Meanwhile, genetic research has provided a large body of evidence for saffron’s autotriploid origin from two different, in Greece naturally occurring, cytotypes of C. cartwrightianus using GBS and multicolor cytogenetics (Nemati et al., 2019; Schmidt et al., 2019). Hence, despite 90% of saffron being grown in Iran (Monks, 2015; Koocheki and Khajeh-Hosseini, 2020), most evidence points to its origin in early Greece. The Aegean Bronze Age artworks point to the islands of Crete and Santorini (Thera), where the most prominent frescoes can be found. Indeed, the latest GBS analysis corroborates the art-inspired focus on the Aegean area, but allows a greater precision: This genetics study shifts our attention slightly north to the region of Attica, as its modern C. cartwrightianus accessions were shown to contain 98% of all saffron alleles (Nemati et al., 2019). Ideally, to follow up these leads, plant remains from the ancient Aegean Bronze Age would be needed to confirm these hypotheses—for species identification and specifically to serve as a template for comparative ancient DNA genomics. Fossil, microfossil, herbarium material or otherwise preserved plant remains have already led to the understanding of the genetic history of several crop species (van Heerwaarden et al., 2011; Scott et al., 2019). Nevertheless, so far, no ancient crocus remains were reported that geneticists could tackle. (Mathew, 1999; Negbi and Negbi, 2002; Jacobsen and Ørgaard, 2004). In 2019, two complementary studies added evidence that supported saffron’s autotriploid origin: Schmidt et al. (2019) used a multi-color cytogenetics approach to identify C. Frontiers in Plant Science \| www.frontiersin.org REFERENCES Botany meets archaeology: people and plants in the past. J. Exp. Bot. 64, 5805–5816. doi: 10.1093/jxb/ert068 Asbaghi, A. (1988). Persische Lehnwörter im Arabischen. Wiesbaden: Harrassowitz Verlag. Dehboneh, S. R., Samanian, K., and Afsharpour, M. (2019). Study of light fastness properties of saffron used in paper dyeing according to historical treatises. J. Color Sci. Technol. 13, 169–190. Babaei, S., Talebi, M., Bahar, M., and Zeinali, H. (2014). Analysis of genetic diversity among saffron (Crocus sativus) accessions from different regions of Iran as revealed by SRAP markers. Sci. Hortic. 171, 27–31. doi: 10.1016/j. scienta.2014.03.033 Deo, B. (2003). Growing Saffron—The World’s Most Expensive Spice. New Zealand: Crop & Food Research.l Barber, E. J. W. (1991). Prehistoric Textiles: The Development of Cloth in the Neolithic and Bronze Ages With Special Reference to the Aegean. New Jersey, United States: Princeton University Press. Dewan, R. (2015). Bronze age flower power: the Minoan use and social significance of saffron and crocus flowers. Chronica 5, 42–55.hh gifl Doumas, C. (1992). The Wall-Paintings of Thera. Athens: Thera Foundation.h Barkeshli, M. (2016). Historical Persian recipes for paper dyes. Restaur. Int. J. Preserv. Libr. Arch. Mater. 37, 49–89. doi: 10.1515/res-2015-0012f Erol, O., Kaya, H. B., Sik, L., Tuna, M., Can, L., and Tanyolac, B. (2014). The genus Crocus, series Crocus (Iridaceae) in Turkey and 2 East Aegean islands: a genetic approach. Turk. J. Biol. 38, 48–62. doi: 10.3906/biy-1305-14 Basker, D., and Negbi, M. (1983). Uses of saffron. Econ. Bot. 37, 228–236. doi: 10.1007/BF02858789 Feldman, M. H. (2002). Luxurious forms: redefining a Mediterranean “international style”, 1400-1200 BCE. Art Bull. 84, 6–29. doi: 10.1080/00043079.2002.10787009f Brandizzi, F., and Caiola, M. G. (1998). Flow cytometric analysis of nuclear DNA in Crocus sativus and allies (Iridaceae). Plant Syst. Evol. 211, 149–154. doi: 10.1007/BF00985356 Fernández, J. A. (2004). “Biology, biotechnology and biomedicine of saffron,” in Recent Research Developments in Plant Science. ed. S. G. Pandalai (Trivandrum, India:: Research Signpost), 127–159.hf Brighton, C. (1977). Cytology of Crocus sativus and its allies (Iridaceae). Plant Syst. Evol. 128, 137–157. doi: 10.1007/BF00984551 Busconi, M., Colli, L., Sánchez, R. A., Santaella, M., de-Los-Mozos Pascual, M., Santana, O., et al. (2015). AFLP and MS-AFLP analysis of the variation within saffron crocus (Crocus sativus L.) germplasm. PLoS One 10:0123434. doi: 10.1371/journal.pone.0123434 Ferrence, S. C., and Bendersky, G. (2004). Therapy with saffron and the goddess at Thera. Perspect. Biol. Med. 47, 199–226. doi: 10.1353/pbm.2004.0026 Fluch, S., Hohl, K., Stierschneider, M., Kopecky, D., and Kaar, B. REFERENCES Caiola, M. G., and Canini, A. (2010). “Looking for saffron’s (Crocus sativus L.) parents,” in Saffron. ed. A. M. Husaini (UK: Global Science Books), 1–14. f Caiola, M. G., Di Somma, D., and Lauretti, P. (2000). Comparative study of pollen and pistil in Crocus sativus L. (Iridaceae) and allied species. Ann. Bot. 58, 73–82. doi: 10.4462/annbotrm-9065 Abrishami, M.H. (2004). Saffron, From Yesterday Till Today, an Encyclopaedia of Its Production, Trade and Use. Tehran: Amirkabir. Agayev, Y. M. (2002). New features in karyotype structure and origin of saffron, Crocus sativus L. Cytologia 67, 245–252. doi: 10.1508/cytologia.67.245 Caiola, M. G., Leonardi, D., and Canini, A. (2010). Seed structure in Crocus sativus L., C. cartwrightianus herb., C. thomasii ten., and C. hadriaticus herb. at SEM. Plant Syst. Evol. 285, 111–120. doi: 10.1007/ s00606-009-0261-x Agayev, Y. M., Fernández, J. A., and Zarifi, E. (2009). Clonal selection of saffron (Crocus sativus L.): The first optimistic experimental results. Euphytica 169, 81–99. doi: 10.1007/s10681-009-9946-zi Chichiriccò, G. (1984). Karyotype and meiotic behaviour of the triploid Crocus sativus L. Caryologia 37, 233–239. doi: 10.1080/00087114.1984.10797702 Agayev, Y.M., Zarifi, E., and Fernández, J.-A. (2010). A study of karyotypes in the Crocus sativus L. aggregate and origin of cultivated saffron. in III International Symposium on Saffron: Forthcoming Challenges in Cultivation, Research and Economics 31 January 2010. Acta Horticulturae. 47–54. Chichiriccò, G., and Caiola, M. G. (1986). Crocus sativus pollen germination and pollen tube growth in vitro and after intraspecific and interspecific pollination. Can. J. Bot. 64, 2774–2777. doi: 10.1139/b86-369f f esearch and Economics 31 January 2010. Acta Horticulturae. 4 Alavi-Kia, S., Mohammadi, S., Aharizad, S., and Moghaddam, M. (2008). Analysis of genetic diversity and phylogenetic relationships in Crocus genus of Iran using inter-retrotransposon amplified polymorphism. Biotechnol. Biotechnol. Equip. 22, 795–800. doi: 10.1080/13102818.2008.10817555 Dadkhah, M.R., Ehtesham, M., and Fekrat, H. (2003). Iranian Saffron an Unknown Jewel. Tehran: Shahr Ashoob Publication. 1–20.f Day, J. (2011a). Counting threads: saffron in Aegean bronze age writing and society. Oxf. J. Archaeol. 30, 369–391. doi: 10.1111/j.1468-0092.2011.00372.x Alsayied, N. F., Fernández, J. A., Schwarzacher, T., and Heslop-Harrison, J. S. (2015). Diversity and relationships of Crocus sativus and its relatives analysed by inter-retroelement amplified polymorphism (IRAP). Ann. Bot. 116, 359–368. doi: 10.1093/aob/mcv103 Day, J. (2011b). Crocus in context: a diachronic survey of the Crocus motif in the Aegean bronze age. Hesperia J. Am. Sch. Class. Stud. Athens 80, 337–379. doi: 10.2972/hesperia.80.3.0337 Day, J. (2013). AUTHOR CONTRIBUTIONS S-SK-S, LM, and AE wrote the manuscript, created the figures, and contributed equally. TH initiated, outlined, and coordinated this endeavor. TH, BU, ZN, and DH guided the project intellectually and contributed to writing in their respective fields of expertise. All authors contributed to the article and approved the submitted version. February 2022 \| Volume 13 \| Article 834416 Frontiers in Plant Science \| www.frontiersin.org 7 Tracing Saffron: Artworks and Genetics Kazemi-Shahandashti et al. FUNDING We acknowledge funding from the Deutsche Forschungs gemeinschaft (DFG 433081887), awarded to BU (98/21-1), TH (HE 7194/2-1), and Frank Blattner (BL 462/19-1) as well as from the Egyptian Ministry of Higher Education awarded to AE (Call 2019-2020). We also thank the Open Access Funding by the Publication Fund of the TU Dresden and Sächsische Landesbibliothek – Staats- und Universitätsbibliothek (SLUB). The authors are grateful to Rachel Dewan for providing guidance regarding the interpretation of the Aegan Art, the appropriate use of related terminology, and for proofreading the manuscript. Similarly, we thank Frank Blattner for reading and discussing the manuscript, and Maximilian H.-W. Schmidt for intellectual guidance on saffron genomics. Lastly, we thank Annaliese Mason for discussions and input on polyploidy and Figure 2. We acknowledge funding from the Deutsche Forschungs gemeinschaft (DFG 433081887), awarded to BU (98/21-1), TH (HE 7194/2-1), and Frank Blattner (BL 462/19-1) as well as from the Egyptian Ministry of Higher Education awarded to AE (Call 2019-2020). We also thank the Open Access Funding by the Publication Fund of the TU Dresden and Sächsische Landesbibliothek – Staats- und Universitätsbibliothek (SLUB). Caiola, M. G., and Canini, A. (2010). “Looking for saffron’s (Crocus sativus L.) parents,” in Saffron. ed. A. M. Husaini (UK: Global Science Books), 1–14. REFERENCES Khajeh-Hosseini (Sawston, United Kingdom: Woodhead Publishing), 37–57. Humphries, J. (1996). The Essential Saffron Companion. London, United Kingdom: Grub Street Publishing.h Rehak, P. (1996). Aegean breechcloths, kilts, and the Keftiu paintings. Am. J. Archaeol. 100, 35–51. doi: 10.2307/506296 Husband, B. C. (2004). The role of triploid hybrids in the evolutionary dynamics of mixed-ploidy populations. Biol. J. Linn. Soc. 82, 537–546. doi: 10.1111/j. 1095-8312.2004.00339.x Rehak, P. (2004). “Crocus costumes in Aegean art” in XAPIS: Essays in Honor of Sara A. Immerwahr. ed. A. P. Chapin (Athens: American School of Classical Studies at Athens), 85–100. Jacobsen, N., and Ørgaard, M. (2004). Crocus cartwrightianus on the Attica peninsula. in I International Symposium on Saffron Biology and Biotechnology 650 31 May 2004. Acta Horticulturae. 65–69. Renau-Morata, B., Moyá, L., Nebauer, S. G., Seguí-Simarro, J. M., Parra-Vega, V., Gómez, M. D., et al. (2013). The use of corms produced under storage at low temperatures as a source of explants for the in vitro propagation of saffron reduces contamination levels and increases multiplication rates. Ind. Crop. Prod. 46, 97–104. doi: 10.1016/j.indcrop.2013.01.013 f 650 31 May 2004. Acta Horticulturae. 65–69. Koo, D.-H., Liu, W., Friebe, B., and Gill, B. S. (2017). Homoeologous recombination in the presence of Ph1 gene in wheat. Chromosoma 126, 531–540. doi: 10.1007/s00412-016-0622-5f f Crop. Prod. 46, 97–104. doi: 10.1016/j.indcrop.2013.01.013 Koocheki, A., and Khajeh-Hosseini, M. (2020). Saffron: Science, Technology and Health. Duxford, United Kingdom: Woodhead Publishing. Riley, R., and Chapman, V. (1958). Genetic control of the cytologically diploid behaviour of hexaploid wheat. Nature 182, 713–715. doi: 10.1038/182713a0 Rubio-Moraga, A., Castillo-López, R., Gómez-Gómez, L., and Ahrazem, O. (2009). Saffron is a monomorphic species as revealed by RAPD, ISSR and microsatellite analyses. BMC. Res. Notes 2:189. doi: 10.1186/1756-0500-2-189h Larsen, B., Orabi, J., Pedersen, C., and Ørgaard, M. (2015). Large intraspecific genetic variation within the saffron-crocus group (Crocus L., series Crocus; Iridaceae). Plant Syst. Evol. 301, 425–437. doi: 10.1007/ s00606-014-1083-z Rukšāns, J. (2017). The World of Crocuses. Riga, Latvija: Latvian Academy of Sciences.h Marinatos, N. (2015). Akrotiri, Thera and the East Mediterranean. Athens: Militos Editions. Sattler, M. C., Carvalho, C. R., and Clarindo, W. R. (2016). The polyploidy and its key role in plant breeding. Planta 243, 281–296. doi: 10.1007/ s00425-015-2450-x Mason, A. S., and Wendel, J. F. (2020). Homoeologous exchanges, segmental allopolyploidy, and polyploid genome evolution. Front. Genet. 11:1014. doi: 10.3389/fgene.2020.01014 Schmidt, T., Heitkam, T., Liedtke, S., Schubert, V., and Menzel, G. (2019). REFERENCES doi: 10.1016/S0065-2660(08)60490-3 S bb G L ( ) V d E l Pl O f d O f d Mousavi, S. Z., and Bathaie, S. Z. (2011). Historical uses of saffron: identifying potential new avenues for modern research. Avicenna J. Phytomed. 1, 57–66. doi: 10.22038/AJP.2011.122f Stebbins, G. L. (1947). Types of polyploids: their classification and significance. Adv. Genet. 1, 403–429. doi: 10.1016/S0065-2660(08)60490-3 Negbi, M., and Negbi, O. (2002). “Saffron domestication in Bronze age crete,” in World Islands in Prehistory: International Insular Investigations. V Deia International Conference of Prehistory. eds. W. H. Waldern and J. A. Ensayat (Oxford, United Kingdom: British Archaeological Reports Publishing). Stebbins, G. L. (1950). Variation and Evolution in Plants. Oxford: Oxford University Press. Täckholm, G. (1920). Preliminary note on the cytology of the genus Rosa. Sven. Bot. Tidskr. 14, 300–311.h Nehvi, F. A., Lone, A. A., Khan, M. A., and Maqhdoomi, M. I. (2010). Comparative study on effect of nutrient management on growth and yield of saffron under temperate conditions of Kashmir. in III International Symposium on Saffron: Forthcoming Challenges in Cultivation, Research and Economics 850. Acta Horticulturae 31 January 2010. 165–170.f Theophrastus, (n.d.). Historia Plantarum. Translation and comments by J B d A d N h l d 1644 Theophrastus, (n.d.). Historia Plantarum. Translation and comments by J. Bodaeus. Amsterdam, Netherlands, 1644. Theophrastus, (n.d.). Historia Plantarum. Translation and comments by J. Bodaeus. Amsterdam, Netherlands, 1644. Toellner, R. (2000). Illustrierte Geschichte der Medizin. Augsburg: Bechtermünz. h p ( ) y J. Bodaeus. Amsterdam, Netherlands, 1644. Toellner R (2000) Illustrierte Geschichte der Medizin Augsburg: Bechtermünz J. Bodaeus. Amsterdam, Netherlands, 1644. Toellner, R. (2000). Illustrierte Geschichte der Medizin. Augsburg: Bechtermünz.t J Toellner, R. (2000). Illustrierte Geschichte der Medizin. Augsburg: Bechtermünz.t Tsaftaris, A., Pasentsis, K., Makris, A., Darzentas, N., Polidoros, A., Kalivas, A., et al. (2011). The study of the E-class SEPALLATA3-like MADS-box genes in wild-type and mutant flowers of cultivated saffron crocus (Crocus sativus L.) and its putative progenitors. J. Plant Physiol. 168, 1675–1684. doi: 10.1016/j. jplph.2011.03.015 Nemati, Z., Harpke, D., Gemicioglu, A., Kerndorff, H., and Blattner, F. R. (2019). Saffron (Crocus sativus) is an autotriploid that evolved in Attica (Greece) from wild Crocus cartwrightianus. Mol. Phylogenet. Evol. 136, 14–20. doi: 10.1016/j.ympev.2019.03.022 Nørbæk, R., Brandt, K., Nielsen, J. K., Ørgaard, M., and Jacobsen, N. (2002). Flower pigment composition of Crocus species and cultivars used for a chemotaxonomic investigation. Biochem. Syst. Ecol. 30, 763–791. REFERENCES Adding color to a century-old enigma: multi-color chromosome identification unravels the autotriploid nature of saffron (Crocus sativus) as a hybrid of wild Crocus cartwrightianus cytotypes. New Phytol. 222, 1965–1980. doi: 10.1111/nph.15715 Mathew, B. (1977). Crocus sativus and its allies (Iridaceae). Plant Syst. Evol. 128, 89–103. doi: 10.1007/BF00985174h Mathew, B. (1982). The Crocus: A Revision of the Genus Crocus (Iridaceae). Oregon: Timber Press, Incorporated, Portland. Mathew, B. (1999). “Botany, taxonomy, and cytology of Crocus sativus L. and its allies,” in Saffron: Crocus sativus L. ed. M. Negbi (London: CRC Press), 19–30. Scott, M. F., Botigué, L. R., Brace, S., Stevens, C. J., Mullin, V. E., Stevenson, A., et al. (2019). A 3,000-year-old Egyptian emmer wheat genome reveals dispersal and domestication history. Nat. Plants 5, 1120–1128. doi: 10.1038/ s41477-019-0534-5 Matić, U. (2019). Memories into images: Aegean and Aegean-like objects in new kingdom Egyptian Theban tombs. Camb. Archaeol. J. 29, 653–669. doi: 10.1017/S095977431900026X Sears, E. R. (1976). Genetic control of chromosome pairing in wheat. Annu. Rev. Genet. 10, 31–51. doi: 10.1146/annurev.ge.10.120176.000335 Maw, G. (1886). A Monograph of the Genus Crocus. Soho Square, London: Dulau and Co.l Shelmerdine, C. W. (1998). “Where do we go from here? And how can the linear B tablets help us get there?” in The Aegean and the Orient in the Second Millenium. Proceedings of the 50th Anniversary Symposium, University of Cinccinnati, Department of Classics, Cincinnati, 18-20 April 1997 (Aegaeum 18). eds. E. H. Cline and D. Harris-Cline (Liège and Austin), 291–299. Möbius, M. (1933). Pflanzenbilder Der Minoischen Kunst in Botanischer Betrachtung Berlin, Germany: De Gruyter.hh Monks, K. (2015). Iran’s Homegrown Treasure: The Spice That Costs More Than Gold. CNN.hh h Morgan, L. (1988). The Miniature Wall Paintings from Thera: A Study in Aegean Culture and Iconography. Cambridge: University Press.f Siracusa, L., Gresta, F., Avola, G., Albertini, E., Raggi, L., Marconi, G., et al. (2013). Agronomic, chemical and genetic variability of saffron (Crocus sativus L.) of different origin by LC-UV#vis-DAD and AFLP analyses. Genet. Resour. Crop. Evol. 60, 711–721. doi: 10.1007/s10722-012-9868-9ii Siracusa, L., Gresta, F., Avola, G., Albertini, E., Raggi, L., Marconi, G., et al. (2013). Agronomic, chemical and genetic variability of saffron (Crocus sativus L.) of different origin by LC-UV#vis-DAD and AFLP analyses. Genet. Resour. Crop. Evol. 60, 711–721. doi: 10.1007/s10722-012-9868-9 Stebbins, G. L. (1947). Types of polyploids: their classification and significance. Adv. Genet. 1, 403–429. REFERENCES (2010). Crocus sativus L. molecular evidence on its clonal origin. in III International Symposium on Saffron: Forthcoming Challenges in Cultivation, Research and Economics 31 January 2010 850, 41–46. Busconi, M., Soffritti, G., Stagnati, L., Marocco, A., Martínez, J. M., de Los Mozos Pascual, M., et al. (2018). Epigenetic stability in saffron (Crocus sativus L.) accessions during four consecutive years of cultivation and vegetative propagation under open field conditions. Plant Sci. 277, 1–10. doi: 10.1016/j.plantsci.2018.09.005 Frello, S., and Heslop-Harrison, J. (2000). Repetitive DNA sequences in Crocus vernus hill (Iridaceae): the genomic organization and distribution of dispersed elements in the genus Crocus and its allies. Genome 43, 902–909. doi: 10.1139/g00-044 Busconi, M., Wischnitzki, E., Del Corvo, M., Colli, L., Soffritti, G., Stagnati, L., et al. (2021). Epigenetic variability among saffron crocus (Crocus sativus L.) accessions characterized by different phenotypes. Front. Plant Sci. 12:642631. doi: 10.3389/fpls.2021.642631ff Friedrich, W. L. (2009). Santorini: Volcano, Natural History, Mythology. Aarhus, Denmark: ISD, LLC. Gismondi, A., Fanali, F., Labarga, J. M. M., Caiola, M. G., and Canini, A. (2013). Crocus sativus L. genomics and different DNA barcode applications. Plant Syst. Evol. 299, 1859–1863. doi: 10.1007/s00606-013-0841-7 Caiola, M. G. (1999). “Reproduction biology of saffron and its allies,” in Saffron: Crocus sativus L. ed. M. Negbi (London: CRC Press), 40–53. February 2022 \| Volume 13 \| Article 834416 Frontiers in Plant Science \| www.frontiersin.org 8 Kazemi-Shahandashti et al. Tracing Saffron: Artworks and Genetics Gresta, F., Lombardo, G. M., Siracusa, L., and Ruberto, G. (2008). Effect of mother corm dimension and sowing time on stigma yield, daughter corms and qualitative aspects of saffron (Crocus sativus L.) in a mediterranean environment. J. Sci. Food Agric. 88, 1144–1150. doi: 10.1002/ jsfa.3177 Panagiotopoulos, D. (2002). Keftiu in context: Theban tomb-paintings as a historical source. Oxf. J. Archaeol. 20, 263–283. doi: 10.1111/ 1468-0092.00134 Petersen, G., Seberg, O., Thorsøe, S., Jørgensen, T., and Mathew, B. (2008). A phylogeny of the genus Crocus (Iridaceae) based on sequence data from five plastid regions. Taxon 57, 487–499. doi: 10.2307/25066017f Herklotz, V., and Ritz, C. M. (2017). Multiple and asymmetrical origin of polyploid dog rose hybrids (Rosa L. sect. Caninae (DC.) Ser.) involving unreduced gametes. Ann. Bot. 120, 209–220. doi: 10.1093/aob/ mcw217 Rashed-Mohassel, M.-H. (2020). “Evolution and botany of saffron (Crocus sativus L.) and allied species,” in Saffron: Science, Technology and Health. eds. A. Koocheki and M. Khajeh-Hosseini (Sawston, United Kingdom: Woodhead Publishing), 37–57. f eds. A. Koocheki and M. Conflict of Interest: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Warren, P. (2000). “From naturalism to essentialism in Theran and Minoan art,” in The Wall Paintings of Thera. ed. S. Sherratt (Piraeus, Greece: Thera Foundation), 364–380. Wachsmann, S., and Wachsmann, A. (1987). Aegeans in the Theban Tombs. Leuven, Belgium: Peeters Publishers.h Wachsmann, S., and Wachsmann, A. (1987). Aegeans in the Theban Tombs. Leuven, Belgium: Peeters Publishers. Warren, P. (2000). “From naturalism to essentialism in Theran and Minoan art,” in The Wall Paintings of Thera. ed. S. Sherratt (Piraeus, Greece: Thera Foundation), 364–380. Warren, P. (2005). Flowers for the goddess? New fragments of wall paintings from Knossos. Br Sch Athens Stud 13, 131–148. Willard, P. (2002). Secrets of Saffron: The Vagabond Life of the World’s Most Seductive Spice. Boston, Massachusetts, United States: Beacon Press. Warren, P. (2005). Flowers for the goddess? New fragments of wall paintings from Knossos. Br Sch Athens Stud 13, 131–148.fh Willard, P. (2002). Secrets of Saffron: The Vagabond Life of the World’s Most Seductive Spice. Boston, Massachusetts, United States: Beacon Press. Frontiers in Plant Science \| www.frontiersin.org REFERENCES doi: 10.1016/ S0305-1978(02)00020-0 van Heerwaarden, J., Doebley, J., Briggs, W. H., Glaubitz, J. C., Goodman, M. M., de Jesus Sanchez Gonzalez, J., et al. (2011). Genetic signals of origin, spread, and introgression in a large sample of maize landraces. Proc. Natl. Acad. Sci. U. S. A. 108, 1088–1092. doi: 10.1073/pnas.1013011108 February 2022 \| Volume 13 \| Article 834416 Frontiers in Plant Science \| www.frontiersin.org 9 Tracing Saffron: Artworks and Genetics Kazemi-Shahandashti et al. Wachsmann, S., and Wachsmann, A. (1987). Aegeans in the Theban Tombs. Leuven, Belgium: Peeters Publishers.h Publisher’s Note: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Copyright © 2022 Kazemi-Shahandashti, Mann, El-nagish, Harpke, Nemati, Usadel and Heitkam. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. February 2022 \| Volume 13 \| Article 834416 Frontiers in Plant Science \| www.frontiersin.org 10
https://openalex.org/W2734503902	https://www.nature.com/articles/oncsis201764.pdf	English	null	Aspartate β-hydroxylase disrupts mitochondrial DNA stability and function in hepatocellular carcinoma	Oncogenesis	2,017	cc-by	9,046	1Department of Hepatic Surgery, The Eastern Hepatobiliary Surgery Hospital, Second Military Medical University, Shanghai, China; 2Department of Hepatobiliary Surgery, The Daping Hospital, Third Military Medical University, Chongqing, China; 3Department of Clinical Database, The Eastern Hepatobiliary Surgery Hospital, Second Military Medical University, Shanghai, China; 4Department of Surgery, The Ohio State University, Wexner Medical Center, Columbus, OH, USA and 5National Scientiﬁc Center for Liver Cancer, Second Military Medical University, Shanghai, China. Correspondence: Dr F Shen, Department of Hepatic Surgery, The Eastern Hepatobiliary Surgery Hospital, Second Military Medical University, 225 Changhai Road, Shanghai 200438, China. E-mail: shenfengehbh@sina.com 6These authors contributed equally to this work. Received 26 January 2017; revised 12 June 2017; accepted 14 June 2017 OPEN OPEN OPEN Citation: Oncogenesis (2017) 6, e362; doi:10.1038/oncsis.2017.64 www.nature.com/oncsis ORIGINAL ARTICLE Aspartate β-hydroxylase disrupts mitochondrial DNA stability and function in hepatocellular carcinoma C Tang1,2,6, Y Hou1,6, H Wang1,6, K Wang1, H Xiang1, X Wan3, Y Xia1, J Li1, W Wei1, S Xu1, Z Lei1, TM Pawlik4, H Wang5, M Wu1 and F Shen1 The mechanism of aberrant mitochondrial genome and function in hepatocellular carcinoma (HCC) remains largely unknown. Our previous study demonstrated an increased expression of aspartate β-hydroxylase (ASPH) in HCC tissues, which was associated with tumor invasiveness and a worse prognosis. Currently, we unexpectedly observed the presence of ASPH in puriﬁed mitochondrial protein fraction. In addition, immunostaining of both exogenously and endogenously expressed ASPH showed a colocalization with mitochondrial biomarkers. This study aimed to investigate whether the mitochondrial ASPH is involved in mitochondrial malfunction in HCC. Our results showed that ASPH overexpression in HCC tissues was correlated with decreased copy numbers of displacement loop (D-loop) and NADH dehydrogenase subunit 1 (ND-1) and enhanced D-loop mutation, suggesting the disrupted mitochondrial DNA (mtDNA) stability. The reduced mtDNA copy numbers were associated with aggressive clinicopathological features of HCC. The loss of mtDNA integrity induced by enforced expression of ASPH was accompanied with mitochondrial dysfunction, which was characterized by the aberrant mitochondrial membrane potential, decreased ATP generation and enhanced reactive oxygen species. In contrast, knocking down ASPH by siRNA in HCC cell lines showed the opposite impact on mtDNA integrity and function. Mass spectrometry and co-immunoprecipitation further identiﬁed that ASPH interacted with histone H2A member X (H2AX). ASPH overexpression diminished the interaction between H2AX and mitochondrial transcription factor A (mtTFA), an important DNA-binding protein for mtDNA replication, which then reduced the binding of mtTFA to D-loop region. Collectively, our results demonstrate that ASPH overexpression disrupts the mtDNA integrity through H2AX–mtTFA signal, thereby affecting mitochondrial functions in HCC. Oncogenesis (2017) 6, e362; doi:10.1038/oncsis.2017.64; published online 17 July 2017 hotspot accounts for 76% of all detected mutations.20 Moreover, mtDNA copy number aberrations were frequently identiﬁed during HCC development.12,13 Previous studies also demonstrated that over 60% of HCC tissues had a decreased mtDNA copy number when compared with those in non-tumor liver tissues.12 The loss of mtDNA copy number was associated with a larger tumor size, presence of cirrhosis and, of critical importance, shorter patient’s survival.21 Despite these important observations, the understanding of tumor-speciﬁc mechanisms regulating mtDNA alteration in HCC remains limited. ASPH overexpression causes mitochondrial malfunction in HCC cell lines ASPH overexpression causes mitochondrial malfunction in HCC cell lines cell lines We next investigated whether disruption of mtDNA integrity caused by ASPH overexpression was associated with mitochon- drial malfunction. We assayed and compared cellular mitochon- drial activity among different HCC cells stained with MitoTracker dye by ﬂow cytometers. Using the area under curve as a quantitative index, ASPH overexpression resulted in 35% and 37% reduced mitochondrial activities in HepG2 and Huh-7 cells compared with the matched control cells, respectively (HepG2-ASPH vs HepG2-vec: 336.1 ± 58.1 vs 515.0 ± 28.2, Po0.01; Huh-7-ASPH vs Huh-7-vec: 138.3 ± 2.8 vs 219.3 ± 4.6, Po0.01; Figure 4a; Supplementary Figure S3a). In contrast, silencing ASPH enhanced mitochondrial activities by 22% and 42% in MHCC-97 L and EHBC-512 cells compared with the control group, respectively (MHCC-siASPH vs MHCC-vec: 714.5 ± 5.4 vs 582.8 ± 16.9, Po0.01; EHBC-siASPH vs EHBC-vec: 869.7 ± 63.5 vs 611.2 ± 86.5, Po0.05; Figure 4b; Supplementary Figure S3b). ASPH overexpression consistently and markedly reduced the area under curve of mitochondrial membrane potential in HepG2 cells (HepG2-ASPH vs HepG2-vec: 1111.3 ± 76.2 vs 1326.7 ± 146.1, Po0.01) and Huh-7 cells (Huh-7-ASPH vs Huh-7-vec: 206.3 ± 5.1 vs 314.5 ± 8.8, Po0.01; Figure 4c; Supplementary Figure S3c). Knock- ing down of ASPH promoted the potential in MHCC-97 L cells (MHCC-siASPH vs MHCC-vec: 440.0 ± 19.5 vs 349.2 ± 8.6, Po0.01) We next investigated whether disruption of mtDNA integrity caused by ASPH overexpression was associated with mitochon- drial malfunction. We assayed and compared cellular mitochon- drial activity among different HCC cells stained with MitoTracker dye by ﬂow cytometers. Using the area under curve as a quantitative index, ASPH overexpression resulted in 35% and 37% reduced mitochondrial activities in HepG2 and Huh-7 cells compared with the matched control cells, respectively (HepG2-ASPH vs HepG2-vec: 336.1 ± 58.1 vs 515.0 ± 28.2, Po0.01; Huh-7-ASPH vs Huh-7-vec: 138.3 ± 2.8 vs 219.3 ± 4.6, Po0.01; Figure 4a; Supplementary Figure S3a). In contrast, silencing ASPH enhanced mitochondrial activities by 22% and 42% in MHCC-97 L and EHBC-512 cells compared with the control group, respectively (MHCC-siASPH vs MHCC-vec: 714.5 ± 5.4 vs 582.8 ± 16.9, Po0.01; EHBC-siASPH vs EHBC-vec: 869.7 ± 63.5 vs 611.2 ± 86.5, Po0.05; Figure 4b; Supplementary Figure S3b). INTRODUCTION Hepatocellular carcinoma (HCC) is the ﬁfth most common malignancy and second leading cause of cancer-related mortality worldwide.1 There are multiple risk factors associated with the development of HCC, including viral hepatitis, aﬂatoxin exposure and excessive alcohol abuse.2–4 Currently, oxidative stress induced via abnormal mitochondrial regulation has also been proposed as an important factor in the pathogenesis of HCC.5,6 The mitochondria is a cytoplasmic organelle that contributes to generate ATP through oxidative phosphorylation,7 produces reactive oxygen species (ROS)8,9 and mediates cell apoptosis signal cascade,10,11 in addition to other cellular functions. Notably, genomic aberrations affecting mitochondrial DNA (mtDNA) have been reported in some human malignancies, including HCC.12–14 The impact of these genomic aberrations on mtDNA has been suggested to enhance tumorigenesis by promoting genomic insults from carcinogens, antagonizing tumor suppressor function, mediating metabolic transition to glycolysis and promoting tumor cell migration.15–18 Aspartate β-hydroxylase (ASPH) is a type II membrane protein, with hydroxylase enzymatic activity that adds a hydroxyl group to aspartate or asparagine residue within epidermal growth factor-like domains.22 Our own group, as well as other authors, has previously demonstrated that ASPH was one of the most differentially expressed genes in HCC and its overexpression was associated with a reduced survival rate and an increased tumor recurrence rate at 5 years after liver resection.23–25 Although the role of ASPH in promoting tumor cell migration and invasion has been identiﬁed, the underlying mechanism remains uncharacterized.26,27 Currently, an unexpected observation of mitochondrial localization of ASPH in immunostaining assay using g Interestingly, sequencing of the mitochondrial genomes has revealed somatic point mutations in 52% of tested HCC tissues,19 in which the displacement loop (D-loop) region mutational ASPH disrupts mtDNA stability in HCC C Tang et al ASPH disrupts mtDNA stability in HCC 2 Also, more patients in the ASPH-high group obtained somatic mutation in D-loop region compared with patients in the ASPH-low group (47.0%, 16/34 vs 18.9%, 7/30, Po0.05; Figure 2f). As such, the results suggested that ASPH expression level was associated with mtDNA contents and mutagenesis in HCC tissues. our prepared antibody speciﬁc for the catalytic domain of ASPH led us to assume that ASPH might have a role in mitochondrial function in HCC. In this study, we characterized the detection of ASPH within the mitochondria of HCC cells, and investigated the impact of expression of ASPH on mtDNA stability and function. RESULTS ASPH localizes on mitochondria in HCC cells and HCC tissue ASPH localizes on mitochondria in HCC cells and HCC tissues Using an antibody speciﬁc against the catalytic domain of ASPH, colocalization between ASPH and the mitochondrial marker MitoTracker was noted in HCC cell lines of EHBC-512 and MHCC-97 L cells, which had a high endogenous expression of ASPH (Figure 1a; Supplementary Figure S1). An overexpression of green ﬂuorescent protein-tagged ASPH was also noted in HepG2 and Huh-7 cells that displayed low endogenous expression of ASPH (Supplementary Figure S1). The green ﬂuorescent protein staining speciﬁcally localized on mitochondrial-like structures, featured by the presence of MitoTracker (Figure 1b). y p g Both endogenous and exogenous ASPH were present in mitochondrial subfraction (Figures 1c and d). These ﬁndings were validated by the enrichment of mitochondrial biomarkers heat shock protein 60 and voltage-dependent anion channel. Mitochondrial subfraction of HCC tissues was also isolated. Both endogenous ASPH and mitochondrial biomarker voltage- dependent anion channel were present in mitochondrial subfrac- tion of HCC tissues (Supplementary Figure S2). Moreover, ASPH also had a calnexin-featured cytoplasmic localization (Figures 1c and d). As such, the results collectively demonstrated a mitochondrial subcellular localization of ASPH in HCC cell lines and HCC tissues. p py In view of the heterogeneity of tumor cells within the cell lines, the D-loop regions of 96–98 monoclonal mtDNA for each cell type were then sequenced. As shown in Figures 3c, 73.5% (72/98), 20.4% (20/98), 4.1% (4/98) and 2.0% (2/98) of HepG2-vec cells had 0, 1, 2 and 42 mutations detected within the D-loop region, respectively. Nevertheless, the corresponding proportion of mutations was 55.1% (54/98), 33.7% (33/98), 10.2% (10/98) and 1.0% (1/98) in the sequenced HepG2-ASPH clones, suggesting an increased rate of mutagenesis in D-loop regions (Po0.05). Similarly, the mutagenesis in D-loop region of Huh-7-ASPH cells was enhanced when compared with Huh-7-vec cells (Figure 3c). In contrast, when ASPH was silenced in MHCC-siASPH and EHBC-siASPH, fewer mutations in D-loop region were observed compared with the controls (Figure 3d). Thus, ASPH expression level also affected the mutagenesis in D-loop region. INTRODUCTION The results highlight a previously unreported role of mitochondrial ASPH in HCC development. ASPH overexpression disrupts mtDNA stability in HCC cell lines To study the effect of ASPH expression level on mtDNA stability in HCC cells, ASPH expression level in HCC cell lines HepG2, Huh-7, MHCC-97 L and EHBC-512 was manipulated. ASPH was either overexpressed (HepG2-ASPH and Huh-7-ASPH) or silenced (MHCC-siASPH and EHBC-siASPH), followed by an evaluation on changes of copy number of D-loop and ND-1. As shown in Figure 3a, the ND-1 contents in HepG2-ASPH and Huh-7-ASPH decreased to 41.0 ± 25.7% (Po0.05) and 16.4 ± 7.3% (Po0.01), respectively, compared with the respective control cells; in contrast, the ND-1 contents in MHCC-siASPH and EHBC-siASPH increased to 231.3 ± 71.8% (Po0.05) and 612.3 ± 142.4% (Po0.01), respectively. D-loop copy numbers were also down- regulated in HepG2-ASPH (13.6 ± 8.6% in relative to HepG2-vec, Po0.01) and Huh-7-ASPH (41.6 ± 18.4% in relative to Huh-7-vec, Po0.01), while upregulated in MHCC-siASPH (861.3 ± 67.3% in relative to MHCC-vec, Po0.01) and EHBC-siASPH (2365.0 ± 496.5% in relative to EHBC-vec, Po0.01; Figure 3b). Therefore, ASPH expression level affected mtDNA copy number. ASPH expression level is negatively correlated with mtDNA stability in HCC tissues The association between ASPH expression and mtDNA stability based on the changes of mtDNA copy number and mutagenesis was analyzed in tumor and matched non-tumor tissues in a cohort of 71 patients with HCC. A decrease in the copy number status of both NADH dehydrogenase subunit 1 (ND-1) and D-loop was noted in the tumor tissues compared with non-tumor tissues (ND-1: 1.077, 95% conﬁdence interval: 0.904–1.249 vs 1.249, 95% conﬁdence interval 1.133–1.366, Po0.05; D-loop: 0.933, 0.727–1.139 vs 1.164, 1.015–1.314, Po0.05; Figures 2a and b). Similarly, more mutations were detected in the tumor tissues vs the non-tumor tissues (Po0.05; Figure 2c). ASPH overexpression causes mitochondrial malfunction in HCC cell lines HCC tissue samples were further divided into the ND-1-high (n = 27) and ND-1-low (n = 44) groups based on the relative ND-1 content in the tumor tissues compared with the matched non- tumor tissues. As shown in Supplementary Table S1, a reduced ND-1 copy number was associated with an elevated serum alpha- fetoprotein level (P = 0.037), larger tumor diameter (P = 0.017), presence of microvascular invasion (P = 0.013) and advanced tumor stage (P = 0.040). Meanwhile, a reduced D-loop copy number was associated with the absence of tumor capsule (P = 0.028). Collectively, the data strongly suggested a correlation between the decrease of mtDNA copy number and aggressive pathological features of HCC. ASPH overexpression consistently and markedly reduced the area under curve of mitochondrial membrane potential in HepG2 cells (HepG2-ASPH vs HepG2-vec: 1111.3 ± 76.2 vs 1326.7 ± 146.1, Po0.01) and Huh-7 cells (Huh-7-ASPH vs Huh-7-vec: 206.3 ± 5.1 vs 314.5 ± 8.8, Po0.01; Figure 4c; Supplementary Figure S3c). Knock- ing down of ASPH promoted the potential in MHCC-97 L cells (MHCC-siASPH vs MHCC-vec: 440.0 ± 19.5 vs 349.2 ± 8.6, Po0.01) Tumors were then further classiﬁed into the ASPH-high (n = 34) and ASPH-low (n = 37) groups, as described in the Materials and Methods section. Both ND-1 and D-loop DNA contents were diminished in the ASPH-high group compared with the ASPH-low group (ND-1: 0.863, 95% conﬁdence interval: 0.667–1.059 vs 1.499, 95% conﬁdence interval: 1.055–1.944, Po0.05; D-loop: 0.643, 0.509–0.775 vs 0.925, 0.747–1.103, Po0.01; Figures 2d and e). Oncogenesis (2017), 1 – 11 ASPH disrupts mtDNA stability in HCC C Tang et al 3 Figure 1. ASPH localizes on mitochondria in HCC cells. (a) Colocalization of endogenous ASPH immunostained with anti-ASPH antibodies and mitochondrial marker MitoTracker in MHCC-97 L and EHBC-512 cells. Green: ASPH protein staining; red: mitochondrial organelle staining. mages were taken by confocal ﬂuorescence microscopy under × 600 magniﬁcation for oil immersion. (b) Colocalization of exogenously expressed green ﬂuorescent protein (GFP)-fusion ASPH immunostained with anti-GFP antibodies and MitoTracker in HepG2 and Huh-7 cells. mages were taken as mentioned above. (c) Presence of endogenous ASPH in mitochondrial and cytoplasmic fraction detected by mmunoblot in MHCC-97 L and EHBC-512 cells. (d) Presence of exogenously expressed GFP-fusion ASPH in mitochondrial and cytoplasmic fraction detected by immunoblot in HepG2 and Huh-7 cells. For b and d, mito: fraction of mitochondrial protein; cyto: fraction of cytoplasmic proteins. ASPH overexpression causes mitochondrial malfunction in HCC cell lines (a, b) Box plot of the relative copy number of ND-1 and D NA extracted from tumor and matched non-tumor tissues (n = 71) that was detected by real-time PCR and normalized to the am tin DNA. (c) Percentages of subgroups of tumor and non-tumor tissues based on the amount of single-nucleotide variation in th on identiﬁed by Sanger sequencing. The number of patients in each subgroup was marked in the corresponding column. (d, e) oral ND-1 and D-loop copy number to matched non-tumor tissues assayed by real-time PCR in the subgroups of HCC patients w 34) or low (n = 37) ASPH expression. (f) The number of patients with or without somatic mutation in the D-loop region of mtDNA d anger sequencing in HCC patients with high (n = 34) or low (n = 37) ASPH expression. For all box plots, the band inside the box n value and the ends of the whiskers represented the minimum and maximum of 2.5–97.5 percentile of all data. *Po0.01, Po ps of non-tumor tissues or high ASPH expression as indicated. ASPH disrupts mtDNA stability in HCC C Tang et al 4 Figure 2. ASPH expression level affects mtDNA stability in HCC tissues. (a, b) Box plot of the relative copy number of ND-1 and D-loop in mtDNA extracted from tumor and matched non-tumor tissues (n = 71) that was detected by real-time PCR and normalized to the amounts of β-actin DNA. (c) Percentages of subgroups of tumor and non-tumor tissues based on the amount of single-nucleotide variation in the D-loop region identiﬁed by Sanger sequencing. The number of patients in each subgroup was marked in the corresponding column. (d, e) Ratios of tumoral ND-1 and D-loop copy number to matched non-tumor tissues assayed by real-time PCR in the subgroups of HCC patients with high (n = 34) or low (n = 37) ASPH expression. (f) The number of patients with or without somatic mutation in the D-loop region of mtDNA detected by Sanger sequencing in HCC patients with high (n = 34) or low (n = 37) ASPH expression. For all box plots, the band inside the box was the mean value and the ends of the whiskers represented the minimum and maximum of 2.5–97.5 percentile of all data. *Po0.01, Po0.05 vs groups of non-tumor tissues or high ASPH expression as indicated. ASPH overexpression causes mitochondrial malfunction in HCC cell lines Heat shock protein 60 (HSP60) and voltage-dependent anion channel (VDAC) are biomarkers of mitochondrial organelles. Calnexin is a biomarker of endoplasmic reticulum. Figure 1. ASPH localizes on mitochondria in HCC cells. (a) Colocalization of endogenous ASPH immunostained with anti-ASPH antibodies and mitochondrial marker MitoTracker in MHCC-97 L and EHBC-512 cells. Green: ASPH protein staining; red: mitochondrial organelle staining. Images were taken by confocal ﬂuorescence microscopy under × 600 magniﬁcation for oil immersion. (b) Colocalization of exogenously expressed green ﬂuorescent protein (GFP)-fusion ASPH immunostained with anti-GFP antibodies and MitoTracker in HepG2 and Huh-7 cells. Images were taken as mentioned above. (c) Presence of endogenous ASPH in mitochondrial and cytoplasmic fraction detected by immunoblot in MHCC-97 L and EHBC-512 cells. (d) Presence of exogenously expressed GFP-fusion ASPH in mitochondrial and cytoplasmic fraction detected by immunoblot in HepG2 and Huh-7 cells. For b and d, mito: fraction of mitochondrial protein; cyto: fraction of cytoplasmic proteins. Heat shock protein 60 (HSP60) and voltage-dependent anion channel (VDAC) are biomarkers of mitochondrial organelles. Calnexin is a biomarker of endoplasmic reticulum. generation in MHCC-97 L cells (MHCC-siASPH vs MHCC-vec: 19290.6 ± 4963.2 vs 9868.8 ± 1773.7, Po0.01) and EHBC-512 cells (EHBC-siASPH vs EHBC-vec: 20388.6 ± 4367.2 vs 10184.2 ± 3087.3, Po0.01; Figure 4f). Thus, ASPH affected intracellular ATP generation likely though regulating mitochondrial membrane potential. and EHBC-512 cells (EHBC-siASPH vs EHBC-vec: 558.5 ± 44.9 vs 355.3 ± 32.2, Po0.01; Figure 4d; Supplementary Figure S3d). Therefore, the mitochondrial membrane potential could be affected by ASPH expression level. Consistent with aberrant mitochondrial activity and membrane potential, the intracellular ATP concentration in HepG2 and Huh-7 cells overexpressed with ASPH was lower compared with control cells, characterized by markedly less relative luminescence unit in the ATP detection assay (HepG2-ASPH vs HepG2-vec: 8438.6 ± 2500.5 vs 23311.2 ± 7268.3, Po0.01; Huh-7-ASPH vs Huh-7-vec: 6310.8 ± 1006.5 vs 10674.6 ± 1531.6, Po0.01; Figure 4e). Conversely, downregulation of ASPH promoted ATP ASPH overexpression increases ROS In addition to energy metabolism, ROS homeostasis is another important biological function for mitochondria. The exogenously enforced expression of ASPH in HepG2 and Huh-7 cells caused an increase of ROS that was characterized by upregulation of Oncogenesis (2017), 1 – 11 ASPH disrupts mtDNA stability in HCC C Tang et al e 2. ASPH expression level affects mtDNA stability in HCC tissues. Oncogenesis (2017), 1 – 11 ASPH overexpression causes mitochondrial malfunction in HCC cell lines genes was screened in relation to ROS homeostasis using a PCR array (Supplementary Table S4) and then validated the results using real-time PCR. These genes included thyroid peroxidase (TPO), myeloperoxidase (MPO), peroxidasin homolog (PXDN), sequestosome 1 (SQSTM1), glutamate–cysteine ligase modiﬁer subunit (GCLM), phosphatidylinositol-3,4,5-trispho- sphate-dependent Rac exchange factor 1 (PREX1), keratin 1 (KRT1) and superoxide dismutase 3 (SOD3). We found that ASPH overexpression promoted biosynthesis of ROS generation enzymes in HepG2 and Huh-7 cells, such as TPO and MPO. Similarly, we observed a reduced expression of enzymes capable of cleaning up ROS, such as GCLM, PREX1 and SOD3 (Figure 5c). intracellular H2O2 concentration as compared with the controls (HepG2-ASPH vs HepG2-vec: 0.365 ± 0.047 vs 0.291 ± 0.014, Po0.01; Huh-7-ASPH vs Huh-7-vec: 0.164 ± 0.016 vs 0.131 ± 0.012, Po0.01; Figure 5a). Conversely, loss of ASPH via gene silencing in MHCC-97 L and EHBC-512 cells resulted in a reduced ROS level in comparison with MHCC-vec cells (0.414 ± 0.014 vs 0.634 ± 0.127, Po0.01) and EHBC-vec cells (0.212 ± 0.092 vs 1.151 ± 0.626, Po0.01; Figure 5b). Thus, the intracellular ROS level could be also inﬂuenced by ASPH expression level in HCC cell lines. To investigate the inﬂuence of ASPH expression on ROS generation and chelation signaling, mRNA expression of multiple Oncogenesis (2017), 1 – 11 PH expression level affects mtDNA stability in HCC cell lines. (a) Alteration of relative ND-1 (left) and D-loop (right) copy nu of HepG2 and Huh-7 cells overexpressed with ASPH in comparison to the control cells. (b) Alteration of relative ND-1 (l ) copy number in the mtDNA of MHCC-97 L and EHBC-512 cells with silenced ASPH in comparison to the control cells. Fo py number was assayed by real-time PCR and normalized to the amounts of β-actin DNA. (c) Percentages of sub-clones of ells that were overexpressed with ASPH or control vectors based on the amount of single-nucleotide variation in the ercentages of sub-clones of MHCC-97 L and EHBC-512 cells that were overexpressed with ASPH siRNA or control vectors b of single-nucleotide variation in the D-loop region. For c and d, the D-loop mutation was identiﬁed by Sanger sequencing ub-clones in each group was marked in the corresponding column. All data are shown as mean ± s.d. from at leas experiments. Po0.05, Po0.01 vs vector group. vec, ASPH and siASPH indicated cell lines overexpressed with SPH siRNA, respectively. ASPH overexpression causes mitochondrial malfunction in HCC cell lines ASPH disrupts mtDNA stability in HCC C Tang et al ASPH disrupts mtDNA stability in HCC C Tang et al 5 Figure 3. ASPH expression level affects mtDNA stability in HCC cell lines. (a) Alteration of relative ND-1 (left) and D-loop (right) copy number in the mtDNA of HepG2 and Huh-7 cells overexpressed with ASPH in comparison to the control cells. (b) Alteration of relative ND-1 (left) and D-loop (right) copy number in the mtDNA of MHCC-97 L and EHBC-512 cells with silenced ASPH in comparison to the control cells. For a and b, mtDNA copy number was assayed by real-time PCR and normalized to the amounts of β-actin DNA. (c) Percentages of sub-clones of HepG2 and Huh-7 cells that were overexpressed with ASPH or control vectors based on the amount of single-nucleotide variation in the D-loop region. (d) Percentages of sub-clones of MHCC-97 L and EHBC-512 cells that were overexpressed with ASPH siRNA or control vectors based on the amount of single-nucleotide variation in the D-loop region. For c and d, the D-loop mutation was identiﬁed by Sanger sequencing and the number of sub-clones in each group was marked in the corresponding column. All data are shown as mean ± s.d. from at least three independent experiments. Po0.05, Po0.01 vs vector group. vec, ASPH and siASPH indicated cell lines overexpressed with vector, ASPH and ASPH siRNA, respectively. ASPH competes with transcription factor A in interacting with histone H2A member X To identify ASPH-mediated downstream signals regulating mtDNA stability, we analyzed the compositions of protein complex immunoprecipitated by ASPH antibodies, in which a ASPH competes with transcription factor A in interacting with histone H2A member X In support of these ﬁndings, downregulation of ASPH in MHCC- -97 L and EHBC-512 displayed an opposite effect (Figure 5d). Therefore, ASPH overexpression induced the alteration of gene expression patterns that promoted ROS generation while inhibited ROS elimination. To identify ASPH-mediated downstream signals regulating mtDNA stability, we analyzed the compositions of protein complex immunoprecipitated by ASPH antibodies, in which a Oncogenesis (2017), 1 – 11 4. ASPH regulates mitochondrial function in HCC cell lines. (a, b) Mitochondrial activity assayed by MitoTracker Red staining and ted by ﬂow cytometers in HepG2 and Huh-7 cells that was overexpressed with ASPH and control vectors, or MHCC-97 L and EHBC-512 hat was overexpressed with ASPH siRNA and control vectors. ASPH overexpression causes mitochondrial malfunction in HCC cell lines The area under ﬂuorescent count curve (AUC) was used for quantiﬁcation. Mitochondrial membrane potential assayed by tetramethylrhodamine ethyl ester (TMRE) staining and detected by ﬂow cytometers in 2 and Huh-7 cells that was overexpressed with ASPH and control vectors, or MHCC-97 L and EHBC-512 cells that was overexpressed with siRNA and control vectors. The AUC was used for quantiﬁcation. (e, f) Intracellular ATP concentration detected by a commercial minescence assay in HepG2 and Huh-7 cells overexpressed with ASPH and control vectors, or MHCC-97 L and EHBC-512 cells xpressed with ASPH siRNA and control vectors. Relative luminescence unit (RLU) was used for quantiﬁcation. All data are shown as ± s.d. from at least three independent experiments. Po0.01, Po0.05 vs vector group. vec, ASPH and siASPH indicated cell lines xpressed with vector, ASPH and ASPH siRNA, respectively. ASPH disrupts mtDNA stability in HCC C Tang et al ASPH disrupts mtDNA stability in HCC C Tang et al 6 Figure 4. ASPH regulates mitochondrial function in HCC cell lines. (a, b) Mitochondrial activity assayed by MitoTracker Red staining and detected by ﬂow cytometers in HepG2 and Huh-7 cells that was overexpressed with ASPH and control vectors, or MHCC-97 L and EHBC-512 cells that was overexpressed with ASPH siRNA and control vectors. The area under ﬂuorescent count curve (AUC) was used for quantiﬁcation. (c, d) Mitochondrial membrane potential assayed by tetramethylrhodamine ethyl ester (TMRE) staining and detected by ﬂow cytometers in HepG2 and Huh-7 cells that was overexpressed with ASPH and control vectors, or MHCC-97 L and EHBC-512 cells that was overexpressed with ASPH siRNA and control vectors. The AUC was used for quantiﬁcation. (e, f) Intracellular ATP concentration detected by a commercial bioluminescence assay in HepG2 and Huh-7 cells overexpressed with ASPH and control vectors, or MHCC-97 L and EHBC-512 cells overexpressed with ASPH siRNA and control vectors. Relative luminescence unit (RLU) was used for quantiﬁcation. All data are shown as mean ± s.d. from at least three independent experiments. Po0.01, Po0.05 vs vector group. vec, ASPH and siASPH indicated cell lines overexpressed with vector, ASPH and ASPH siRNA, respectively. mitochondrial localized protein, histone H2A member X (H2AX), emerged as a candidate. immunoprecipitant by ASPH or H2AX antibodies in total cell lysates (Figure 6b) and mitochondrial fraction (Supplementary Figures S4b and c). Oncogenesis (2017), 1 – 11 ASPH overexpression causes mitochondrial malfunction in HCC cell lines We then examined whether the interaction between H2AX and mitochondrial transcription factor A (mtTFA) or DNA polymerase gamma (POLG) could be affected by ASPH–H2AX interaction. As shown in Figure 6c, ASPH could interact with H2AX in HepG2-ASPH cells, whereas the interaction between mtTFA and H2AX was greatly diminished compared with Co-immunoprecipitation assay consistently showed that hemagglutinin (HA)-tagged ASPH and myc-tagged H2AX were in a common protein complex (Figure 6a). Exogenous ASPH–H2AX interaction was also conﬁrmed in mitochondrial fraction by co-immunoprecipitation assay (Supplementary Figure S4a). In addition, we detected endogenous H2AX or ASPH presence in Oncogenesis (2017), 1 – 11 HepG2-vec cells. Of note, ASPH overexpression did not POLG–H2AX interaction and failed to inﬂuence mtTFA sion level. Thus, we propose that ASPH competes with in interacting with H2AX in HCC cell lines To clarify the outcome of weakened H2AX–mtTFA inte caused by ASPH overexpression, we tested the binding efﬁ of mtTFA to D-loop region. The mtTFA-binding area was sc by amplifying mtTFA binding fragment using tilling p ASPH regulates ROS generation in HCC cell lines. (a, b) Intracellular ROS generation characterized and assayed by intracellula tration and detected by ﬂow cytometers in HepG2 and Huh-7 cells overexpressed with ASPH and control vectors, or MHCC-97 12 cells overexpressed with ASPH siRNA and control vectors. (c, d) Ratios of mRNA expression level of multiple genes associate nerating and chelating detected by real-time PCR in HepG2 and Huh-7 cells overexpressed with ASPH, or MHCC-97 L and EH th silenced ASPH to the control groups. All data are shown as mean ± s.d. from at least three independent experiments. P 05 vs vector group. vec, ASPH and siASPH indicated cell lines overexpressed with vector, ASPH and ASPH siRNA, respectively. ASPH disrupts mtDNA stability in HCC C Tang et al ASPH disrupts mtDNA stability in HCC C Tang et al C Tang et al 7 Figure 5. ASPH regulates ROS generation in HCC cell lines. (a, b) Intracellular ROS generation characterized and assayed by intracellular H2O2 concentration and detected by ﬂow cytometers in HepG2 and Huh-7 cells overexpressed with ASPH and control vectors, or MHCC-97 L and EHBC-512 cells overexpressed with ASPH siRNA and control vectors. ASPH overexpression causes mitochondrial malfunction in HCC cell lines (c, d) Ratios of mRNA expression level of multiple genes associated with ROS generating and chelating detected by real-time PCR in HepG2 and Huh-7 cells overexpressed with ASPH, or MHCC-97 L and EHBC-512 cells with silenced ASPH to the control groups. All data are shown as mean ± s.d. from at least three independent experiments. Po0.01, Po0.05 vs vector group. vec, ASPH and siASPH indicated cell lines overexpressed with vector, ASPH and ASPH siRNA, respectively. that in HepG2-vec cells. Of note, ASPH overexpression did not affect POLG–H2AX interaction and failed to inﬂuence mtTFA expression level. Thus, we propose that ASPH competes with mtTFA in interacting with H2AX in HCC cell lines. To clarify the outcome of weakened H2AX–mtTFA interaction caused by ASPH overexpression, we tested the binding efﬁciency of mtTFA to D-loop region. The mtTFA-binding area was screened by amplifying mtTFA-binding fragment using tilling primers Oncogenesis (2017), 1 – 11 ASPH disrupts mtDNA stability in HCC C Tang et al Figure 6. ASPH competes with mtTFA in interacting with H2AX in HCC cell lines. (a) Interaction between exogenously express ASPH and myc-tagged H2AX in 293 cells assayed by immunoprecipitation using HA antibody or myc antibody. (b) Interaction o ASPH and H2AX in MHCC-97 L cells assayed by immunoprecipitation using ASPH antibody or H2AX antibody. (c) Inﬂue overexpression on mTFA–H2AX interaction. The protein complex that was immunoprecipitated by H2AX antibody in overexpressed with ASPH and control vectors was further analyzed by immunoblot using ASPH, POLG and mTFA antibodies. (d) of mTFA-binding mtDNA motif in the D-loop region through a chromosome immunoprecipitation (ChIP)-PCR assay. The complex was immunoprecipitated by mTFA antibody in MHCC-97 L cells. The putative bound mtDNA motif was further ampliﬁe primer pairs that covered the D-loop region and were designated from P1 to P8. The amplicon of primer set 4 (P4) showed enri protein–DNA motif. Po0.01 vs IgG group. (e) Inﬂuence of ASPH overexpression on mTFA-binding probability to D-loop reg ChIP-PCR assay. The amounts of D-loop region characterized by P4 amplicon in the mTFA–DNA complex were analyzed in overexpressed with ASPH or MHCC-97 L cells with silenced ASPH in comparison to the control group overexpressed with vecto shown as mean ± s.d. from at least three independent experiments. Po0.01 vs vector group. g 8 Figure 6. ASPH competes with mtTFA in interacting with H2AX in HCC cell lines. ASPH overexpression causes mitochondrial malfunction in HCC cell lines (a) Interaction between exogenously expressed HA-tagged ASPH and myc-tagged H2AX in 293 cells assayed by immunoprecipitation using HA antibody or myc antibody. (b) Interaction of endogenous ASPH and H2AX in MHCC-97 L cells assayed by immunoprecipitation using ASPH antibody or H2AX antibody. (c) Inﬂuence of ASPH overexpression on mTFA–H2AX interaction. The protein complex that was immunoprecipitated by H2AX antibody in HepG2 cells overexpressed with ASPH and control vectors was further analyzed by immunoblot using ASPH, POLG and mTFA antibodies. (d) Identiﬁcation of mTFA-binding mtDNA motif in the D-loop region through a chromosome immunoprecipitation (ChIP)-PCR assay. The protein–DNA complex was immunoprecipitated by mTFA antibody in MHCC-97 L cells. The putative bound mtDNA motif was further ampliﬁed using eight primer pairs that covered the D-loop region and were designated from P1 to P8. The amplicon of primer set 4 (P4) showed enrichment in the protein–DNA motif. Po0.01 vs IgG group. (e) Inﬂuence of ASPH overexpression on mTFA-binding probability to D-loop region through a ChIP-PCR assay. The amounts of D-loop region characterized by P4 amplicon in the mTFA–DNA complex were analyzed in HepG2 cells overexpressed with ASPH or MHCC-97 L cells with silenced ASPH in comparison to the control group overexpressed with vectors. All data are shown as mean ± s.d. from at least three independent experiments. *Po0.01 vs vector group. DISCUSSION covering whole D-loop regions. As shown in Figure 6d, the amplicon ampliﬁed by P4 primer set showed a signiﬁcant enrichment in mtDNA fragment pulled down by mtTFA antibodies when compared with IgG isotype. This amplicon was used as reporter of mtTFA-binding efﬁciency to mtDNA. In this study, we describe a previously unidentiﬁed function of ASPH in disrupting mtDNA stability and mitochondrial function. The data demonstrate that ASPH localizes on mitochondria in HCC cells and HCC tissues and interacts with H2AX. Overexpression of ASPH disrupts the mtDNA integrity through H2AX–mtTFA signal, thereby affecting mitochondrial functions in HCC. The reduction of mtDNA has been observed in some human neoplasms, such as renal carcinoma, gastric cancer, breast cancer and HCC.28–31 We here demonstrated that the changes in mtDNA copy number were associated with aggressive pathological features of HCC, including large tumor size, microvascular invasion, absence of tumor capsule and advanced disease stage. These data suggested an important role of mtDNA instability in disease development. We then examined the content of D-loop region ampliﬁed by P4 primer set in HCC cells with varied ASPH expression level and found that ASPH overexpression greatly impaired mtTFA binding to D-loop region in HepG2 cells (HepG2-ASPH vs HepG2-vec: 6.2 ± 2.9 vs 26.1 ± 8.0, Po0.01), while ASPH silencing facilitated mtTFA binding in MHCC-97 L cells (MHCC-siASPH vs MHCC-vec: 13.9 ± 2.2 vs 7.3 ± 1.8, Po0.01; Figure 6e). Collectively, ASPH could inﬂuence mtTFA-binding probability to mtDNA in HCC cell lines. Oncogenesis (2017), 1 – 11 ASPH disrupts mtDNA stability in HCC C Tang et al ASPH disrupts mtDNA stability in HCC C Tang et al 9 The underlying mechanism of mtDNA aberrance in HCC remains uncharacterized, though the involvement of mtDNA instability in carcinogenesis has been well described.12–14 On the other hand, ASPH has been recognized as an important molecular marker associated with tumor invasion and disease outcome in some types of cancers, including HCC,23–25,32–34 but the related mechanism and downstream signaling are largely unknown. In the current study, we noted mitochondrial localization of ASPH in HCC cells using an antibody speciﬁc for the catalytic domain of ASPH. The data further highlighted a role of mitochondrial ASPH in regulating mtDNA copy numbers and somatic mtDNA mutation. DISCUSSION Both of these features are hallmarks of mtDNA instability in HCC.12,13,19,28 ASPH overexpression in HepG2 and Huh-7 cells signiﬁcantly increased mutation frequency in the D-loop region and decreased copy numbers of D-loop and ND-1 region, the most frequently altered mtDNA region during tumorigenesis. These in vitro results were consistent with our correlation study results between ASPH expression level and mtDNA integrity in HCC tissues, which demonstrated that HCC tissues with high ASPH expression showed less mtDNA copy numbers while more mutations compared with tissue that had low ASPH expression. The underlying mechanism of mtDNA aberrance in HCC remains uncharacterized, though the involvement of mtDNA instability in carcinogenesis has been well described.12–14 On the other hand, ASPH has been recognized as an important molecular marker associated with tumor invasion and disease outcome in some types of cancers, including HCC,23–25,32–34 but the related In conclusion, we reported the mitochondrial localization of ASPH and its role in mtDNA instability and mitochondrial dysfunction in HCC. In the proposed model (Supplementary Figure S5), ASPH localizes to the mitochondria where it interacts with H2AX. This interaction interferes with H2AX–mtTFA binding in a competitive manner. The disruption of H2AX–mtTFA may prevent mtTFA from binding to mtDNA. As an overall outcome, overexpression of ASPH causes a reduced mtDNA copy number and an increased mtDNA mutation. Findings from the current study expand the mechanistic understanding of mitochondrial malfunction in HCC and suggest ASPH as an intervention target for maintaining mtDNA stability. Constructs The cDNA encoding the full open reading frame sequence of ASPH and H2AX was purchased from Origene (Rockville, MD, USA). The siRNA sequences for silencing ASPH was 5′-GCGCAGTGTGAGGATGAT-3′. We also sub-cloned expression constructs of HA-tagged ASPH or myc-tagged H2AX by adding a HA or myc tags at the N terminus of ASPH or H2AX, respectively. The lentivirus of wild-type ASPH and the siRNAi against human ASPH were purchased from Sunbio (Beijing, China). MATERIALS AND METHODS Patients and tissue samples The study was approved by the Institutional Ethics Committee of the Eastern Hepatobiliary Surgery Hospital, and informed consent was obtained from each patient to use their data and surgically resected specimens in this study. Surgical specimens, including tumor and paired non-tumor tissues, from 140 patients who underwent liver resection for histologically proven HCC at the Eastern Hepatobiliary Surgery Hospital between March 2011 and June 2011 were collected and used in this study. 2 The mtDNA instability induced by ASPH overexpression might contribute to mitochondrial dysfunction in HCC cells. We found that ASPH overexpression decreased mitochondrial membrane potential and mitochondrial activity, and resulted in a decreased ATP generation. Simultaneously, the intracellular ROS levels within those cells were greatly enhanced, mainly due to irregular activity of ROS generating and chelating pathway. These consequences might be a direct result of the effect of ASPH on mtDNA integrity, as reduction in the copy number of mtDNA and somatic mutations in D-loop region were accompanied with a reduction of mitochondrial respiratory enzymes, thereby repressing the function of mitochondria and inﬂuencing ROS production.13 Of note, it has been reported that increasing ROS could promote ASPH expression,35 which inevitably worsen mtDNA instability. As previously reported,25 patients were grouped according to the level of ASPH expression. Speciﬁcally, the ratio of ASPH mRNA in the tumor tissues as compared with the matched non-tumor control tissues was tested by real-time PCR. Among 140 HCC patients, 34 patients had ASPH high expression (ratio42), while 37 patients had ASPH low expression in tumor tissues (ratioo0.5). High or low expression of ASPH was also conﬁrmed by immunoblotting (Supplementary Figure S6). These 71 patients were further used to study the correlation between mtDNA integrity, ASPH expression level and the clinicopathological features. The clinicopathological data of these patients are listed in Supplementary Table S2. Antibodies A polyclonal antibody against ASPH was raised in our institution using the synthetic peptide antigen of 12 amino-acid residues around the Fe2+- binding domain of ASPH. The antibodies against mitochondrial biomarkers heat shock protein 60 (ab46798) and voltage-dependent anion channel (ab154856), endoplasmic reticulum biomarkers calnexin (ab195198), green ﬂuorescent protein (ab6556), H2AX (ab11175), POLG (ab207558) and mtTFA (ab176558) were purchased from Abcam (Cambridge, UK). The mouse monoclonal anti-myc (9E10) and HA antibodies (12CA5) were purchased from Sigma-Aldrich (St Louis, MO, USA) and Roche (Basel, Switzerland), respectively. Additional information on secondary antibodies is detailed in the Supplementary Information. Cell cultures Human embryonic kidney 293 cell and human liver cancer cell lines included HepG2, Huh-7, MHCC-97 L and EHBC-512 were maintained as previously described.40 Additional information on cell culture is detailed in the Supplementary Information. p , y y We newly identiﬁed the interplay between ASPH and mito- chondrial H2AX by a high-throughput proteomic strategy of mass spectrometry. H2AX has been identiﬁed as a protein that localizes to the mitochondria and serves as a shuttle protein transporter involved in mitochondrial protein transport.36 This interaction might facilitate the mitochondrial localization of ASPH, which needs to be further investigated. Moreover, the increased occupation of H2AX by ASPH binding might affect other cargo proteins loading to this transporter. In this study, we detected a decreased mtTFA, a previously identiﬁed mitochondrial protein shuttled by H2AX,36 in the immunoprecipitated H2AX protein complex of HCC cells overexpressing ASPH. The mtTFA has been widely known as one of the most important members for mtDNA stability involved in mtDNA replication, transcription and repair.37–39 The weakened H2AX–mtTFA interaction caused by ASPH overexpression might result in two outcomes. First, the ability of mtTFA to translocate to the mitochondria, where mtDNA replication mainly occurs, might be greatly abolished via H2AX–ASPH interactions. Another consequence is that mitochon- drial mtTFA might not be capable of forming a proper protein complex to initial mtDNA replication, for which mtTFA–H2AX interaction itself is also necessary.36 The mtDNA immunoprecipita- tion data demonstrated that mtTFA binding to the D-loop region is greatly reduced by ASPH overexpression in HCC cells. Therefore, H2AX and mtTFA might be the central molecules mediating the function of mitochondrial ASPH in mtDNA integrity. Oncogenesis (2017), 1 – 11 PCR array and real-time PCR All statistical analysis was performed using SPSS (version 18.0, Chicago, IL, USA). Continuous variables were expressed as median (range) and compared using the nonparametric Mann–Whitney U-test, Kruskal–Wallis test and two-independent samples t-test or paired t-test. Categorical variables were reported as the number of cases and the prevalence, and the differences between the groups were compared using the χ2-test with Yates correction or Fisher exact test as appropriate. A Po0.05 was considered as statistically signiﬁcant. The Human Oxidative Stress RT2 Proﬁler PCR Array Kit (Qiagen, Hilden, Germany) was used to analyze the ROS metabolism-related gene expression in HepG2 cells overexpressing ASPH, MHCC-97 L cells with ASPH silenced and their respective controls. RNA extraction and cDNA preparation were performed as previously described.40 The PCR array was performed according to the protocol recommended by the manufacturer. Differentially expressed genes were conﬁrmed by real-time PCR. The primers are listed in Supplementary Table S3. β-actin served as an internal control. Additional information on PCR procedures is detailed in Supplementary Information. Statistical analysis Statistical analysis The authors declare no conﬂict of interest. y The amounts of ND-1 and D-loop fragment to reﬂect the changes of copy number of mtDNA were assayed as previously described.12,13 The copy numbers of ND-1 or D-loop in the mtDNA of HCC samples were deﬁned as the ratio between tumor and matched non-tumor control. The copy numbers of ND-1 or D-loop in the mtDNA of HCC cells were quantiﬁed after normalization to internal control. The primers are listed in Supplementary Table S3. β-actin served as an internal control. Additional information on PCR procedures is detailed in the Supplementary Information. ACKNOWLEDGEMENTS Technical assistance on immunoprecipitation was provided by Prof. Long Yu from the State Key Laboratory of Genetic Engineering of Fudan University. This work was supported by the State Key Project for Infectious Diseases (2012ZX10002-016 and 2012ZX10002-011), the Creative Research Groups of National Natural Science Foundation of China (30921006) and the Postdoctoral Foundation of Shanghai (13R21318400). To detect mutations in the D-loop region of mtDNA, we used mtDNA extracted from paired HCC tissues or multiple clones of HCC cell lines cultured from single tumor cells to reﬂect the intrinsic heterogeneity of mtDNA. The D-loop region was ampliﬁed as previously described,41 using primers listed in Supplementary Table S3. Additional information on PCR procedures is detailed in the Supplementary Information. Mitochondrial membrane potential and mitochondrial activity measurement ochondrial membrane potential and mitochondrial a 7 Boekema EJ. A passive function of mitochondrial ATP synthase: target for tumor killer HAMLET. J Mol Biol 2015; 427: 1863–1865. 7 Boekema EJ. A passive function of mitochondrial ATP synthase: target for tumor killer HAMLET. J Mol Biol 2015; 427: 1863–1865. 8 Sabharwal SS, Schumacker PT. Mitochondrial ROS in cancer: initiators, ampliﬁers or an Achilles' heel? Nat Rev Cancer 2014; 14: 709–721. Mitochondrial activity and membrane potential of HCC cell lines were, respectively, measured using MitoTracker Red CMXRos (MTR, Life Invitrogen) or tetramethylrhodamine ethyl ester (Life Invitrogen) following the manufacturer’s instructions. 9 Zou LY, Zheng BY, Fang XF, Li D, Huang YH, Chen ZX et al. HBx co-localizes with COXIII in HL-7702 cells to upregulate mitochondrial function and ROS generation. Oncol Rep 2015; 33: 2461–2467. 10 Tang C, Lin H, Wu Q, Zhang Y, Bie P, Yang J. Recombinant human augmenter of liver regeneration protects hepatocyte mitochondrial DNA in rats with obstructive jaundice. J Surg Res 2015; 196: 90–101. Mass spectrometry and immunoprecipitation The FLAG-ASPH plasmids were transfected into 293 cells. Crude cell lysate was prepared 72 h after transfection. The protein complex interacting with FLAG- ASPH was obtained using the FLAG HA Tandem Afﬁnity Puriﬁcation Kit (Sigma-Aldrich), according to the manufacturer’s instructions. Mass spectro- metry analysis was performed on ﬂag-tagged immunoprecipitant. The data were analyzed by Research Center for Proteome Analysis, Shanghai, China. Co-immunoprecipitation was performed using the cell lysates or the mitochondrial fraction from 293 cells co-transfected with HA-ASPH and myc-H2AX. The analysis of endogenous interaction among ASPH, H2AX and mtTFA was performed in HCC cell lines using their respective antibodies. 11 Radogna F, Albertini MC, De Nicola M, Diederich M, Bejarano I, Ghibelli L. Melatonin promotes Bax sequestration to mitochondria reducing cell suscept- ibility to apoptosis via the lipoxygenase metabolite 5-hydroxyeicosatetraenoic acid. Mitochondrion 2015; 21: 113–121. 12 Lee HC, Li SH, Lin JC, Wu CC, Yeh DC, Wei YH. Somatic mutations in the D-loop and decrease in the copy number of mitochondrial DNA in human hepatocellular carcinoma. Mutat Res 2004; 547: 71–78. 13 Yin PH, Lee HC, Chau GY, Wu YT, Li SH, Lui WY et al. Alteration of the copy number and deletion of mitochondrial DNA in human hepatocellular carcinoma. Br J Cancer 2004; 90: 2390–2396. 14 Wallace DC. Mitochondria and cancer. Nat Rev Cancer 2012; 12: 685–698. 14 Wallace DC. Mitochondria and cancer. Nat Rev C REFERENCES 1 Stewart BW, Wild C. World Cancer Report 2014. International Agency for Research on Cancer: Lyon, France, 2014, pp 403–414. 1 Stewart BW, Wild C. World Cancer Report 2014. International Agency for Research on Cancer: Lyon, France, 2014, pp 403–414. 2 Wang Q, Xu Y, Zhou W, Zhong L, Wen Z, Yu H et al. The viral oncoprotein HBx of Hepatitis B virus promotes the growth of hepatocellular carcinoma through cooperating with the cellular oncoprotein RMP. Int J Biol Sci 2014; 10: 1181–1192. 2 Wang Q, Xu Y, Zhou W, Zhong L, Wen Z, Yu H et al. The viral oncoprotein HBx of Hepatitis B virus promotes the growth of hepatocellular carcinoma through cooperating with the cellular oncoprotein RMP. Int J Biol Sci 2014; 10: 1181–1192. 3 Moudgil V, Redhu D, Dhanda S, Singh J. A review of molecular mechanisms in the development of hepatocellular carcinoma by aﬂatoxin and hepatitis B and C viruses. J Environ Pathol Toxicol Oncol 2013; 32: 165–175. Subcellular fraction preparation The mitochondrial fraction of cultured cells was isolated using Mitochon- dria Isolation Kit (Thermo) (Waltham, MA, USA) according to the speciﬁcations provided by the manufacturer. The cytosol and mitochon- drial fractions were further lysed with 2% 3-[(3-cholamidopropyl)dimethy- lammonio]propanesulfonate (CHAPS) in Tris-buffered saline, boiled with SDS–polyacrylamide gel electrophoresis sample buffer and applied for gel electrophoresis. The current study had several limitations. As a hydroxylase, whether the enzymatic activity of ASPH is required for its role in regulating mtDNA integrity remains to be investigated in the future study. In addition, the essential role of mtTFA in mtDNA replication has been well established, whereas its contribution to mtDNA mutagenesis requires more investigation. Oncogenesis (2017), 1 – 11 Oncogenesis (2017), 1 – 11 ASPH disrupts mtDNA stability in HCC C Tang et al ASPH disrupts mtDNA stability in HCC C Tang et al 10 Immunoblot and immunoﬂuorescence staining Immunostaining and immunoblot were performed as previously described.40 Fluorescent images were taken using a confocal microscope of FV1000 (Olympus, Tokyo, Japan). Hoechst 33258 (Molecular Probes, Waltham, MA, USA) was used for nuclear staining. crosslinked. The bound DNA was extracted using the QIAamp DNA mini kit (Qiagen) according to the manufacturer’s instructions. We designed eight pairs of primers across the whole mtDNA D-loop region for PCR ampliﬁcation. The primers are listed in Supplementary Table S3. ROS and ATP generation measurement The intracellular ROS generation in HCC cell lines was measured through Amplex Red Hydrogen Peroxide/Peroxidase Assay Kit (Life Invitrogen, Waltham, MA, USA, A22188). The intracellular ATP generation in HCC cell lines was measured using ATP Determination Kit (Biovision, Milpitas, CA, USA). All assays were performed according to the manufacturer’s instructions. Additional information on assay procedures is detailed in the Supplementary Information. 4 Sidharthan S, Kottilil S. Mechanisms of alcohol-induced hepatocellular carcinoma. Hepatol Int 2014; 8: 452–457. 4 Sidharthan S, Kottilil S. Mechanisms of alcohol-induced hepatocellular carcinoma. Hepatol Int 2014; 8: 452–457. 5 Marra M, Sordelli IM, Lombardi A, Lamberti M, Tarantino L, Giudice A et al. Molecular targets and oxidative stress biomarkers in hepatocellular carcinoma: an overview. J Transl Med 2011; 9: 171–184. 6 Sasaki Y. Does oxidative stress participate in the development of hepatocellular carcinoma? J Gastroenterol 2006; 41: 1135–1148. 6 Sasaki Y. Does oxidative stress participate in the development of hepatocellular carcinoma? J Gastroenterol 2006; 41: 1135–1148. CONFLICT OF INTEREST The authors declare no conﬂict of interest. The authors declare no conﬂict of interest. mtDNA immunoprecipitation and PCR 15 Pelicano H, Carney D, Huang P. ROS stress in cancer cells and therapeutic implications. Drug Resist Updat 2004; 7: 97–110. The procedures of mtDNA immunoprecipitation were slightly modiﬁed from conventional chromatin immunoprecipitation. A unit of 1 mg of pre- cleared mitochondrial protein was mixed with 20 μg anti-mtTFA antibody and AminoLink Plus Resin (Thermo Fisher Scientiﬁc, Waltham, MA, USA) by end-over-end inversion overnight at 4 °C. Antibodies of IgG isotype served as a control. The immunoprecipitant was centrifuged, washed and 16 Chatterjee A, Mambo E, Sidransky D. Mitochondrial DNA mutations in human cancer. Oncogene 2006; 25: 4663–4674. 17 Ishikawa K, Hayashi J. Trading mtDNA uncovers its role in metastasis. Cell Adh Migr 2009; 3: 11–13. 18 Carew JS, Huang P. Mitochondrial defects in cancer. Mol Cancer 2002; 1: 9. Oncogenesis (2017), 1 – 11 ASPH disrupts mtDNA stability in HCC ASPH disrupts mtDNA stability in HCC C Tang et al 11 expressed in recombinant Escherichia coli. Hybridoma (Larchmt) 2011; 30: 69–79. expressed in recombinant Escherichia coli. Hybridoma (Larchmt) 2011; 30: 69–79. 19 Yin PH, Wu CC, Lin JC, Chi CW, Wei YH, Lee HC. Somatic mutations of mito- chondrial genome in hepatocellular carcinoma. Mitochondrion 2010; 10: 174–182. 33 Luu M, Sabo E, de la Monte SM, Greaves W, Wang J, Tavares R et al. Prognostic value of aspartyl (asparaginyl)-beta-hydroxylase/humbug expression in non-small cell lung carcinoma. Hum Pathol 2009; 40: 639–644. 20 Wong LJ, Tan DJ, Bai RK, Yeh KT, Chang J. Molecular alterations in mitochondrial DNA of hepatocellular carcinomas: is there a correlation with clinicopathological proﬁle? J Med Genet 2004; 41: e65. 21 Yamada S, Nomoto S, Fujii T, Kaneko T, Takeda S, Inoue S et al. Correlation between copy number of mitochondrial DNA and clinicopathologic parameters of hepatocellular carcinoma. Euro J Surg Oncol 2006; 32: 303–307. 34 Palumbo KS, Wands JR, Safran H, King T, Carlson RI, de la Monte SM. Human aspartyl (asparaginyl) beta-hydroxylase monoclonal antibodies: potential biomarkers for pancreatic carcinoma. Pancreas 2002; 25: 39–44. 35 Lawton M, Tong M, Gundogan F, Wands JR, de la Monte SM. Aspartyl- (asparaginyl) beta-hydroxylase, hypoxia-inducible factor-alpha and Notch cross- talk in regulating neuronal motility. Oxid Med Cell Longev 2010; 3: 347–356. 22 Gronke RS, Welsch DJ, VanDusen WJ, Garsky VM, Sardana MK, Stern AM et al. Partial puriﬁcation and characterization of bovine liver aspartyl beta-hydroxylase. J Biol Chem 1990; 265: 8558–8565. mtDNA immunoprecipitation and PCR 23 Xian ZH, Zhang SH, Cong WM, Yan HX, Wang K, Wu MC. Expression of aspartyl beta-hydroxylase and its clinicopathological signiﬁcance in hepatocellular carcinoma. Mod Pathol 2006; 19: 280–286. 36 Choi YS, Hoon Jeong J, Min HK, Jung HJ, Hwang D, Lee SW et al. Shot-gun proteomic analysis of mitochondrial D-loop DNA binding proteins: identiﬁcation of mitochondrial histones. Mol Biosyst 2011; 7: 1523–1536. 24 Cantarini MC, de la Monte SM, Pang M, Tong M, D'Errico A, Trevisani F et al. Aspartyl-asparagyl beta hydroxylase over-expression in human hepatoma is linked to activation of insulin-like growth factor and notch signaling mechanisms. Hepatology 2006; 44: 446–457. 37 Larsson NG, Wang J, Wilhelmsson H, Oldfors A, Rustin P, Lewandoski M et al. Mitochondrial transcription factor A is necessary for mtDNA maintenance and embryogenesis in mice. Nat Genet 1998; 18: 231–236. 38 Kang D, Kim SH, Hamasaki N. Mitochondrial transcription factor A (TFAM): roles in maintenance of mtDNA and cellular functions. Mitochondrion 2007; 7: 39–44. 25 Wang K, Liu J, Yan ZL, Li J, Shi LH, Cong WM et al. Overexpression of aspartyl-(asparaginyl)-beta-hydroxylase in hepatocellular carcinoma is associated with worse surgical outcome. Hepatology 2010; 52: 164–173. 39 Campbell CT, Kolesar JE, Kaufman BA. Mitochondrial transcription factor A regulates mitochondrial transcription initiation, DNA packaging, and genome copy number. Biochim Biophys Acta 2012; 1819: 921–929. 26 Sepe PS, Lahousse SA, Gemelli B, Chang H, Maeda T, Wands JR et al. Role of the aspartyl-asparaginyl-beta-hydroxylase gene in neuroblastoma cell motility. Lab Invest 2002; 82: 881–891. 40 Hou Y, Zou Q, Ge R, Shen F, Wang Y. The critical role of CD133(+)CD44(+/high) tumor cells in hematogenous metastasis of liver cancers. Cell Res 2012; 22: 259–272. 27 de la Monte SM, Tamaki S, Cantarini MC, Ince N, Wiedmann M, Carter JJ et al. Aspartyl-(asparaginyl)-beta-hydroxylase regulates hepatocellular carcinoma invasiveness. J Hepatol 2006; 44: 971–983. 41 Datta S, Chattopadhyay E, Ray JG, Majumder M, Roy PD, Roy B. D-loop somatic mutations and approximately 5 kb ‘common’ deletion in mitochondrial DNA: important molecular markers to distinguish oral precancer and cancer. Tumour Biol 2015; 36: 3025–3033. 28 Hsu CC, Lee HC, Wei YH. Mitochondrial DNA alterations and mitochondrial dys- function in the progression of hepatocellular carcinoma. World J Gastroenterol 2013; 19: 8880–8886. 29 Zhang G, Qu Y, Dang S, Yang Q, Shi B, Hou P. Variable copy number of mito- chondrial DNA (mtDNA) predicts worse prognosis in advanced gastric cancer patients. mtDNA immunoprecipitation and PCR Diagn Pathol 2013; 8: 173. Oncogenesis is an open-access journal published by Nature Publishing Group. This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ 30 Meierhofer D, Mayr JA, Fink K, Schmeller N, Koﬂer B, Sperl W. Mitochondrial DNA mutations in renal cell carcinomas revealed no general impact on energy metabolism. Br J Cancer 2006; 94: 268–274. 31 Tommasi S, Favia P, Weigl S, Bianco A, Pilato B, Russo L et al. Mitochondrial DNA variants and risk of familial breast cancer: an exploratory study. Int J Oncol 2014; 44: 1691–1698. 32 Yang H, Wang H, Xue T, Xue XP, Huyan T, Wang W et al. Single-chain variable fragment antibody against human aspartyl/asparaginyl beta-hydroxylase © The Author(s) 2017 tary Information accompanies this paper on the Oncogenesis website (http://www.nature.com/oncsis) upplementary Information accompanies this paper on the Oncogenesis website (http://www.nature.com Oncogenesis (2017), 1 – 11 Oncogenesis (2017), 1 – 11
https://openalex.org/W4289032879	https://curis.ku.dk/ws/files/228533674/Mitchell_et_al_Nutrients_2019_Vol_11_9_e2207.pdf	English	null	Protein intake at twice the RDA in older men increases circulatory concentrations of the microbiome metabolite trimethylamine-N-oxide (TMAO)	Carolina Digital Repository (University of North Carolina at Chapel Hill)	2,019	cc-by	11,113	Download date: 24. Oct. 2024 university of copenhagen Protein intake at twice the RDA in older men increases circulatory concentrations of the microbiome metabolite Trimethylamine-N-Oxide (TMAO) Mitchell, Sarah M; Milan, Amber M; Mitchell, Cameron J; Gillies, Nicola A; D'Souza, Randall F; Zeng, Nina; Ramzan, Farha; Sharma, Pankaja; Knowles, Scott O; Roy, Nicole C; Sjödin, Anders Mikael; Wagner, Karl-Heinz; Zeisel, Steven H; Cameron-Smith, David university of copenhagen Protein intake at twice the RDA in older men increases circulatory concentrations of the microbiome metabolite Trimethylamine-N-Oxide (TMAO) Mitchell, Sarah M; Milan, Amber M; Mitchell, Cameron J; Gillies, Nicola A; D'Souza, Randall F; Zeng, Nina; Ramzan, Farha; Sharma, Pankaja; Knowles, Scott O; Roy, Nicole C; Sjödin, Anders Mikael; Wagner, Karl-Heinz; Zeisel, Steven H; Cameron-Smith, David university of copenhagen Protein intake at twice the RDA in older men increases circulatory concentrations of the microbiome metabolite Trimethylamine-N-Oxide (TMAO) Mitchell, Sarah M; Milan, Amber M; Mitchell, Cameron J; Gillies, Nicola A; D'Souza, Randall F; Zeng, Nina; Ramzan, Farha; Sharma, Pankaja; Knowles, Scott O; Roy, Nicole C; Sjödin, Anders Mikael; Wagner, Karl-Heinz; Zeisel, Steven H; Cameron-Smith, David Mitchell, Sarah M; Milan, Amber M; Mitchell, Cameron J; Gillies, Nicola A; D'Souza, Randall F; Zeng, Nina; Ramzan, Farha; Sharma, Pankaja; Knowles, Scott O; Roy, Nicole C; Sjödin, Anders Mikael; Wagner, Karl-Heinz; Zeisel, Steven H; Cameron-Smith, David Mitchell, Sarah M; Milan, Amber M; Mitchell, Cameron J; Gillies, Nicola A; D'Souza, Randall F; Zeng, Nina; Ramzan, Farha; Sharma, Pankaja; Knowles, Scott O; Roy, Nicole C; Sjödin, Anders Mikael; Wagner, Karl-Heinz; Zeisel, Steven H; Cameron-Smith, David Publication date: 2019 Document version Publisher's PDF, also known as Version of record Document version Publisher's PDF, also known as Version of record Document version Publisher's PDF, also known as Version of record Protein Intake at Twice the RDA in Older Men Circulatory Concentrations of the Microbiome Metabolite Trimethylamine-N-Oxide (TMAO) Sarah M. Mitchell 1 , Amber M. Milan 1,2 , Cameron J. Mitchell 1,3 , Nicola A. Gillies 1, Randall F. D’Souza 1,4, Nina Zeng 1 , Farha Ramzan 1 , Pankaja Sharma 1 , Scott O. Knowles 2 , Nicole C. Roy 2,5,6 , Anders Sjödin 7, Karl-Heinz Wagner 8, Steven H. Zeisel 9 and David Cameron-Smith 1,5,10,* Sarah M. Mitchell 1 , Amber M. Milan 1,2 , Cameron J. Mitchell 1,3 , Nicola A. Gillies 1, Randall F. D’Souza 1,4, Nina Zeng 1 , Farha Ramzan 1 , Pankaja Sharma 1 , Scott O. Knowles 2 , Nicole C. Roy 2,5,6 , Anders Sjödin 7, Karl-Heinz Wagner 8, Steven H. Zeisel 9 and David Cameron-Smith 1,5,10,* 1 Liggins Institute, The University of Auckland, Auckland 1023, New Zealand; sarah.mitchell@auckland.ac.nz (S.M.M.); a.milan@auckland.ac.nz (A.M.M.); cameron.mitchell@ubc.ca (C.J.M.); n.gillies@auckland.ac.nz (N.A.G.); r.dsouza@auckland.ac.nz (R.F.D.); n.zeng@auckland.ac.nz (N.Z.); f.ramzan@auckland.ac.nz (F.R.); p.sharma@auckland.ac.nz (P.S.) 2 Food Nutrition & Health Team, Food & Bio-based Products Group, AgResearch, Palmerston North New Zealand; scott.knowles@agresearch.co.nz (S.O.K.); nicole.roy@agresearch.co.nz (N.C.R.) New Zealand; scott.knowles@agresearch.co.nz (S.O.K.); nicole.roy@agresearch.co.nz (N.C.R.) 3 School of Kinesiology, University of British Columbia, Vancouver, BC V6T 1Z1, Canada g ( ) y g ( ) 3 School of Kinesiology, University of British Columbia, Vancouver, BC V6T 1Z1, Canada gy y 4 Discipline of Nutrition, School of Medical Sciences, University of Auckland, Auckland 1023, New Zealand 5 Riddet Institute, Massey University, Palmerston North 4442, New Zealand Discipline of Nutrition, School of Medical Sciences, University of Auckland, Auckland 1023, New Zea 5 Riddet Institute, Massey University, Palmerston North 4442, New Zealand y y 6 The High-Value Nutrition National Science Challenge, Auckland 1023, New Zealand 7 Department of Nutrition, Exercise and Sport, Copenhagen University, 1165 Copenhagen, Denmark amsj@nexs.ku.dk 8 Department of Nutritional Sciences and Research Platform Active Ageing, University of Vienna, 1010 Vienna, Austria; karl-heinz.wagner@univie.ac.at 8 Department of Nutritional Sciences and Research Platform Active Ageing, University of Vienna, 1010 Vi A i k l h i @ i i 8 Department of Nutritional Sciences and Research Platform Active Ageing, University of Vienna, 1010 Vienna, Austria; karl-heinz.wagner@univie.ac.at 1010 Vienna, Austria; karl-heinz.wagner@univie.ac.at 9 Nutrition Research Institute, University of North Carolina, Kannapolis, NC 28081, USA; steven_zeisel@unc.edu 10 Clinical Nutrition Research Centre, Singapore Institute for Technology and Research, Singapore 117609, Singapore 10 Clinical Nutrition Research Centre, Singapore Institute for Clinical Sciences, Agency for Science, Technology and Research, Singapore 117609, Singapore * Correspondence: d.cameron-smith@auckland.ac.nz; Tel.: +64-09-373-7599 nutrients nutrients Document license: CC BY Citation for published version (APA): Mitchell, S. M., Milan, A. M., Mitchell, C. J., Gillies, N. A., D'Souza, R. F., Zeng, N., Ramzan, F., Sharma, P., Knowles, S. O., Roy, N. C., Sjödin, A. M., Wagner, K-H., Zeisel, S. H., & Cameron-Smith, D. (2019). Protein intake at twice the RDA in older men increases circulatory concentrations of the microbiome metabolite Trimethylamine-N-Oxide (TMAO). Nutrients, 11(9), [2207]. https://doi.org/10.3390/nu11092207 Citation for published version (APA): Mitchell, S. M., Milan, A. M., Mitchell, C. J., Gillies, N. A., D'Souza, R. F., Zeng, N., Ramzan, F., Sharma, P., Knowles, S. O., Roy, N. C., Sjödin, A. M., Wagner, K-H., Zeisel, S. H., & Cameron-Smith, D. (2019). Protein intake at twice the RDA in older men increases circulatory concentrations of the microbiome metabolite Trimethylamine-N-Oxide (TMAO). Nutrients, 11(9), [2207]. https://doi.org/10.3390/nu11092207 1. Introduction Adequate protein intake is key for the elderly [1]. The current recommendations for protein intake for all adults, according to the World Health Organization (WHO) [2] and the US Department of Agriculture (USDA) [3], is 0.8 g protein/kg body weight/day. However, a growing number of randomised controlled trials (RCTs) [4], mechanistic [5] and epidemiological [6] studies support an increase from the current recommendations to mitigate the risks of age-related loss of skeletal muscle (sarcopenia), falls, and associated mortality risk [7–9]. Yet, there is also the possibility that a high protein diet may increase the risk for cardiovascular disease (CVD). Epidemiological evidence suggests that high protein diets from both plant and animal sources are associated with reduced CVD risk [10,11]. Furthermore, interventional studies have demonstrated that during an energy deﬁcit, a higher protein intake might be beneﬁcial for cardiovascular and metabolic health, particularly in obese and at-risk populations [12–15]. Conversely, further evidence from epidemiological and association studies suggests that high protein intake heightens the risk of CVD [16–19]. High protein diets are also associated with an increased prevalence of diabetes [20,21] and impaired kidney function [22]. A putative biomarker for CVD risk is the gut microbiota-derived metabolite trimethylamine-N- oxide (TMAO) as it has repeatedly been associated with an increased risk of adverse cardiovascular events [23–26]. Dietary intake of the nutrients choline and carnitine are the main contributors to circulatory TMAO via their metabolism by gut microbiota. Choline is an essential nutrient abundant in protein-rich foods, such as eggs, red meat, poultry, and ﬁsh [27]; similarly, carnitine is found at high levels in red meat, poultry, and dairy products [28]. Clinical studies have demonstrated that plasma and urine TMAO concentrations increase transiently after ingestion of choline or carnitine supplements [29], red meat [25], ﬁsh [30], dairy [31], and egg ingestion [32]. There are limited RCTs that have analysed the impact of high protein dietary interventions on circulatory TMAO concentrations, particularly for the elderly. Consequently, it is inconclusive whether high protein diets modify fasting plasma TMAO concentrations in this population. Hence, the purpose of this study was to determine the eﬀect of the consumption of dietary protein at twice the current RDA compared to consumption at the RDA on TMAO concentrations. Further analysis was made of traditional biomarkers of cardiovascular and metabolic disease risk in the elderly. 2.1. Study Design The study was a 10-week single-blind parallel randomised controlled trial. The study was approved by the Southern Health and Disability Ethics Committee (New Zealand; 15/STH/236) and was conducted in accordance with the Declaration of Helsinki. The study protocol was registered with the Australia New Zealand Clinical Trial Registry #ACTRN12616000310460. All the subjects gave written informed consent to participate in the study. Received: 14 August 2019; Accepted: 10 September 2019; Published: 12 September 2019 Received: 14 August 2019; Accepted: 10 September 2019; Published: 12 September 2019 Abstract: Higher dietary protein intake is increasingly recommended for the elderly; however, high protein diets have also been linked to increased cardiovascular disease (CVD) risk. Trimethylamine-N-oxide (TMAO) is a bacterial metabolite derived from choline and carnitine abundant from animal protein-rich foods. TMAO may be a novel biomarker for heightened CVD risk. The purpose of this study was to assess the impact of a high protein diet on TMAO. Healthy men (74.2 ± 3.6 years, n = 29) were randomised to consume the recommended dietary allowance of protein (RDA: 0.8 g protein/kg bodyweight/day) or twice the RDA (2RDA) as part of a supplied diet for 10 weeks. Fasting blood samples were collected pre- and post-intervention for measurement of TMAO, blood lipids, glucose tolerance, insulin sensitivity, and inﬂammatory biomarkers. An oral glucose tolerance test was also performed. In comparison with RDA, the 2RDA diet increased circulatory TMAO (p = 0.002) but unexpectedly decreased renal excretion of TMAO (p = 0.003). LDL cholesterol was increased in 2RDA compared to RDA (p = 0.049), but no diﬀerences in other biomarkers of CVD risk and insulin sensitivity were evident between groups. In conclusion, circulatory TMAO is responsive to changes in dietary protein intake in older healthy males. Keywords: TMAO; high protein diet; CVD; CVD biomarkers; elderly Nutrients 2019, 11, 2207; doi:10.3390/nu11092207 www.mdpi.com/journal/nutrients www.mdpi.com/journal/nutrients Nutrients 2019, 11, 2207 2 of 14 2.2. Participants Thirty-one men were recruited to take part in the study (Table 1) via public advertisements and were screened by questionnaire. Eligible subjects were males aged 70 years or more, non-smokers, had a body mass index (BMI) between 20 and 32 kg/m2, and were independent community dwellers. The exclusion criteria included prior history of cardiovascular, metabolic, renal, or gastrointestinal disease. Participants were recruited from the Auckland (New Zealand) region and data was collected at the Liggins Institute, University of Auckland, between March and October 2016. 3 of 14 Nutrients 2019, 11, 2207 Table 1. Participant characteristics pre-intervention. Table 1. Participant characteristics pre-intervention. p p Parameter RDA Group 2RDA Group Mean ± SD or Number Mean ± SD or Number Count 1 15 14 Age (years) 74.7 ± 3.9 73.7 ± 3.3 Height (cm) 172.8 ± 8.2 171.7 ± 5.5 Weight (kg) 85.7 ± 20.5 83.0 ± 8.3 BMI (kg/m2) 28.4 ± 5.1 28.2 ± 3.3 Medication usage (n) Statin 4 6 ACE inhibitor 4 3 Aspirin 4 2 Calcium channel blocker 1 2 Proton pump inhibitors 0 2 Alpha blocker 1 2 Beta blocker 1 2 Xanthine oxidase inhibitor 1 1 BMI—body mass index, ACE—angiotensin converting enzyme. 1 Only the participants who completed the study are included (n = 29). BMI—body mass index, ACE—angiotensin converting enzyme. 1 Only the participants who completed the study are included (n = 29). 2.4. Dietary Analysis Prior to the study, participants completed a 3-day food diary at baseline to determine habitual intake and completed fortnightly checklists throughout the study to monitor assigned dietary intake. Macronutrient intakes were estimated by a registered dietician using Foodworks software (Version 8; Xyris). Serving sizes of eggs (1 egg), ﬁsh (100 g), meat (100 g), and dairy (250 mL milk, 50 g cheese, 150 mL yoghurt) were calculated from the habitual diet and study diets, as these were collectively the major contributors of choline and carnitine in the diet. 2.3. Study Procedures Participants were randomly assigned to a dietary intervention for 10 weeks according to a parallel design: (1) the current recommended protein diet (RDA group; 0.8 g protein/kg bodyweight/day), or (2) a high protein diet (2RDA group; twice the RDA). This study reports secondary outcomes from a clinical trial for which the primary outcomes were changes in muscle size and strength [4]. The quantity of protein in the 2RDA diet was chosen as it has previously been recommended as optimal for preserving muscle mass in older adults [33]. Sample size calculation, randomisation, dietary and compliance data was described previously [4]. Brieﬂy, both diets contained approximately 30% energy from fat with the balance made up of carbohydrates. Protein was made up from a combination of animal and plant sources, including dairy, eggs, poultry, ﬁsh, red meat, legumes, grains, nuts, and seeds. All food consumed was provided and delivered weekly to the participants’ homes. The energy contents of the diets were individually calculated based on the Harris-Benedict equation [34] and adjusted for physical activity as assessed by a wrist accelerometer (Fitbit Charge HR; Fitbit, Inc., San Francisco, CA, USA). During the intervention, participants attended the Nutrition Clinic at the Liggins Institute at weeks 4, 6 and 8 for weight measurement and compliance check. Participants were asked to make no changes to their exercise routine throughout their participation in the trial. Fasting anthropometry measurements and biological sampling, and an oral glucose tolerance test (OGTT) were taken at pre-intervention (Pre) and after 10 weeks of intervention (Post). 2.6. OGTT Fasting sample collection was followed by a 2-h OGTT whereby a pre-prepared glucose solution containing 75 g of glucose as a 200-mL beverage was consumed within 5 min. During the OGTT, blood samples were collected in 30-min intervals for 2 h (0, 30, 60, 90, and 120 min) for the measurement of plasma glucose and insulin concentrations to establish whole-body insulin sensitivity, calculated by the Matsuda Index [37] and by using the homeostatic model assessment of insulin resistance (HOMA-IR) [38]. 2.8. Calculations Creatinine clearance was calculated from plasma and 24-h urine creatinine excretion as follows: Creatinine clearance = [(urine creatinine concentration in mmol/L) × (urine volume in mL/1140 min)/(plasma creatinine concentration in (µmol/L) · 1000 mL−1 · min−1)] [39]. Creatinine clearance = [(urine creatinine concentration in mmol/L) × (urine volume in mL/1140 min)/(plasma creatinine concentration in (µmol/L) · 1000 mL−1 · min−1)] [39]. Fractional renal excretion of TMAO was calculated as follows: Fractional renal excretion of TMAO was calculated as follows: Fractional renal excretion of TMAO was calculated as follows: Fractional renal excretion of TMAO = (urinary TMAO × plasma creatinine)/(plasma TMAO × urinary creatinine) × 100 [40]. 2.7. Anthropometry Resting blood pressure was measured with an electronic sphygmomanometer (Omron HEM-71111-AU, Omron Healthcare, Kyoto, Japan). Participants were weighed with a calibrated scale (Tanita HD-351, Arlington Heights, IL, USA) without shoes and height was measured with a stadiometer (HoltainLtd., Crymych, UK). BMI was calculated using these data. Waist circumference was taken with a tape measure at the top of the iliac crest. All measurements were taken twice and the average was reported. Dual-energy X-ray absorptiometry (DXA) was performed for total fat mass and central adiposity (android % fat) (Prodigy, GE-Lunar, Madison, WI, USA). 2.5. Sample Collection and Analysis Fasting blood samples were drawn from an antecubital vein by cannula into both EDTA-coated vacutainers and uncoated serum vacutainers. Serum was left to clot for 20 min before all the tubes Nutrients 2019, 11, 2207 4 of 14 were centrifuged for 15 min at 1900× g and 4 ◦C. Plasma and serum samples were then stored at −80 ◦C until analysis. fuged for 15 min at 1900× g and 4 ◦C. Plasma and serum samples were then stored at −80 ◦C sis. TMAO, choline, and betaine were quantiﬁed in plasma and urine using stable isotope dilution coupled to liquid chromatography-multiple reaction monitoring mass-spectrometry with an ACQUITY UPLC (Waters, Milford, MA, USA), according to previously published methods [35]. , , q p g p coupled to liquid chromatography-multiple reaction monitoring mass-spectrometry with an ACQUITY UPLC (Waters, Milford, MA, USA), according to previously published methods [35]. UPLC (Waters, Milford, MA, USA), according to previously published methods [35]. Carnitine in plasma was quantiﬁed using reversed-phase ultra-high-performance liquid chromatography coupled with tandem mass-spectrometry (UHPLC-MS/MS) (ThermoFisher Scientiﬁc, Santa Clara, CA, USA). Plasma was measured for glucose, creatinine, and lipids (including total cholesterol (T-C), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and triacylglycerides (TAG)), and serum was assessed for C-reactive protein (CRP) by a Cobas C311 Analyser (Roche Diagnostics, Basel, Switzerland) by enzymatic colorimetric assays. Insulin concentration in plasma was measured by a Cobas e411 (Roche Diagnostics, Basel, Switzerland) by immunoassays. Whole blood was measured for glycated hemoglobin (HbA1c) using the Cobas C311 Analyser (Roche Diagnostics, Basal, Switzerland). Participants were asked pre- and post-intervention to collect a 24-hour urine sample in a container provided. Samples were weighed and stored in aliquots at −80 ◦C. The concentration of urinary nitrogen (comprising urea, creatinine, uric acid) was measured with a Cobas C311 Analyser (Roche Diagnostics, Basel, Switzerland) to estimate dietary protein intake and nitrogen balance [36]. 3.1. Dietary Intake A comprehensive summary of macronutrient intake has been published [4]. Overall, participants were highly compliant to both diets (compliance for protein intake was 98.9% in the RDA group and 97.5% in the 2RDA group). This was documented in fortnightly compliance food records and according to urinary nitrogen excretion assessed pre- and post-intervention (RDA: from 1.3 ± 0.2 to 0.9 ± 0.2 g/kg bodyweight/day, p = 0.001; 2RDA: from 1.3 ± 0.2 to 1.5 ± 0.1 g/kg bodyweight/day, p = 0.001). Average protein intake was altered by the intervention (time × diet interaction p < 0.001); it reduced in the RDA group (from 104.8 ± 30 to 80.1 ± 23 g/day, p = 0.004) and increased in the 2RDA group (from 95.6 ± 20 to 136.2 ± 18 g/day, p < 0.001). Dietary cholesterol decreased in the RDA group (from 386.8 ± 142.1 to 299.2 ± 83.2 mg/day, p = 0.063) and increased in the 2RDA group (from 351.1 ± 113.4 to 526.8 ± 73.4 mg/day, p < 0.001) as a result of the intervention (time × diet interaction p < 0.001). Dietary ﬁbre increased (time eﬀect, p < 0.001) for RDA (from 34 ± 9.9 to 47 ± 8.4 g/day, p < 0.001), and for 2RDA (from 28.9 ± 8.1 to 50.3 ± 5.3 g/day, p < 0.001) but was not diﬀerent between the diets (p = 0.410) as a consequence of including the local recommendations for fruit and vegetable intake in both diets [41]. Egg, ﬁsh, red meat, white meat, and dairy which were the major contributors to choline and carnitine intake are listed in Table 2. Table 2. Servings per day of egg, ﬁsh, red meat, white meat, dairy intake, and fruit and vegetables for the RDA and 2RDA groups pre- and post-intervention. Table 2. Servings per day of egg, ﬁsh, red meat, white meat, dairy intake, and fruit and vegetables for the RDA and 2RDA groups pre- and post-intervention. Food Group RDA (Mean ± SD) 2RDA (Mean ± SD) Eﬀect (p) a Pre Post Pre Post Time Diet Time × Diet Egg 1 0.7 ± 0.6 0.6 ± 0.4 0.7 ± 0.5 1. 3.1. Dietary Intake ± 0.4 † # 0.047 * 0.054 0.021 * Fish 2 0.4 ± 0.4 0.1 ± 0.1 # 0.4 ± 0.5 0.6 ± 0.3 † 0.489 0.013 * 0.002 * Red meat 2 1.2 ± 0.9 0.2 ± 0.1 # 1.3 ± 0.7 1.1 ± 0.3 † <0.001 * 0.003 * 0.011 * White meat 2 0.4 ± 0.5 0.2 ± 0.1 0.3 ± 0.4 0.9 ± 0.1 † # 0.024 * 0.001 * <0.001 * Dairy 3 1.1 ± 0.8 0.8 ± 0.5 1.1 ± 0.8 2.4 ± 0.7 † # 0.014 * <0.001 * <0.001 * Fruit & vegetables 4 5.3 ± 3 8.5 ± 1.6 # 4.4 ± 2.1 8.6 ± 2.1 # <0.001 * 0.41 0.535 Serving size calculated as 1 one egg; 2 100 g cooked weight; 3 250 mL milk or 150 g yoghurt, or 50 g cheese; 4 one cup. Fish includes fatty ﬁsh, white ﬁsh, and shellﬁsh; red meat includes beef, lamb, venison, and pork; white meat includes poultry; dairy includes milk, yoghurt, cheese, and ice-cream. * Signiﬁcant main eﬀect or interaction, p < 0.05. # Diﬀerent from pre-intervention within the same group, p < 0.05. † Diﬀerent between diets at indicated time point. p values were controlled using the Sidak post hoc procedure. a Main eﬀects and interactions were calculated by two-way repeated measures ANOVA. Serving size calculated as 1 one egg; 2 100 g cooked weight; 3 250 mL milk or 150 g yoghurt, or 50 g cheese; 4 one cup. Fish includes fatty ﬁsh, white ﬁsh, and shellﬁsh; red meat includes beef, lamb, venison, and pork; white meat includes poultry; dairy includes milk, yoghurt, cheese, and ice-cream. * Signiﬁcant main eﬀect or interaction, p < 0.05. # Diﬀerent from pre-intervention within the same group, p < 0.05. † Diﬀerent between diets at indicated time point. p values were controlled using the Sidak post hoc procedure. a Main eﬀects and interactions were calculated by two-way repeated measures ANOVA. 2.9. Statistical Analysis Sample size was calculated for the primary outcomes which were changes in muscle size and strength [4]. Data are presented as mean ± standard deviation (SD). Statistical analyses were performed using the software package SPSS, version 25.0 (SPSS Inc., Chicago, IL, USA). Between-group diﬀerences were analysed using a two-way repeated measures ANOVA with time (from pre to 5 of 14 Nutrients 2019, 11, 2207 post) as a repeated factor and diet (RDA vs 2RDA) as a between-subject factor. Post hoc multiple comparisons were controlled using Sidak corrections. Normality was assessed with the Shapiro–Wilk test and non-normally distributed data were log transformed before further analysis. Correlation analyses between variables were determined using Spearman’s correlation coeﬃcient for non-normally distributed data. Signiﬁcance was accepted at p < 0.05. Unadjusted means ± SDs are shown in the tables and text. 3. Results Thirty-one participants were randomly assigned to the two diets (RDA: n = 16, 2RDA: n = 15). One withdrew before the start of the intervention (RDA) and one was excluded due to non-compliance (2RDA). The pre-intervention characteristics of the remaining 29 participants are summarised in Table 1 and reported in detail elsewhere [4]. 3.2. TMAO, Choline, Betaine, and Carnitine The concentrations of TMAO, choline, betaine, and carnitine in plasma are shown in Figure 1, and the concentrations of TMAO, choline, and betaine in urine (expressed as the ratio to urinary creatinine concentrations) are shown in Figure 2. There was an interaction eﬀect of plasma TMAO (time × diet interaction, p = 0.002; Figure 1A) which did not change in the RDA group (12.8 ± 9.67 µM vs. 8.05 ± 7.52 µM, p = 0.165) and increased in the 2RDA group (from 8.34 ± 4.79 µM to 29.08 ± 31.53 µM, 6 of 14 Nutrients 2019, 11, 2207 p = 0.004). There was an interaction eﬀect of plasma betaine (time × diet interaction p = 0.006; Figure 1C) which increased in the RDA group (from 38.67 ± 5.25 µM to 43.73 ± 9.16 µM, p = 0.018) and was unchanged in the 2RDA group (from 46.71 ± 13.26 µM to 44.07 ± 11.18 µM, p = 0.219). However, there was a signiﬁcant diﬀerence between groups pre- but not post-intervention (p = 0.030, p = 0.930, respectively) which may have driven this eﬀect. There was no eﬀect of the intervention on plasma choline or carnitine (time × diet interaction p = 0.982 and p = 0.492, respectively). There was an interaction eﬀect of TMAO in urine (time × diet interaction, p = 0.007) which did not change in the RDA group (77.4 ± 77 vs. 125.2 ± 134 mmol/mol creatinine, p = 0.345) and decreased in the 2RDA group (from 209.8 ± 198 to 77.7 ± 46.2 mmol/mol creatinine, p = 0.005; Figure 2A). There was no interaction eﬀect in urinary choline or betaine (time × diet interaction p = 0.389 and p = 0.656, respectively) although urinary betaine decreased in both groups (time eﬀect, p = 0.032). Change in plasma TMAO correlated inversely with change in urinary TMAO (r = −0.698, p = 0.01). Nutrients 2019, 11, 2207 6 of 14 Figure 1. Box-whisker plots illustrating concentrations of (A) TMAO, (B) choline, (C) betaine, (D) carnitine in plasma of the RDA and 2RDA groups pre-intervention (uncoloured boxes) and after 10 weeks (coloured boxes). The central lines represent means and the error bars represent minimum and maximum values. TMAO: trimethylamine-N-oxide. 3.2. TMAO, Choline, Betaine, and Carnitine Two-way repeated measures ANOVA was used for comparison and the difference between time points in same group was controlled using the Sidak post hoc procedure RDA 2RDA 0 40 80 120 T M A O ( u M ) p = 0.002 p = 0.004 RDA 2RDA 0 10 20 C h o l i n e ( u M ) RDA 2RDA 0 40 80 B e t a i ne ( uM ) p = 0.006 p = 0.018 RDA 2RDA 0 50 100 C a r n i t i n e ( u M ) A B C D Figure 1. Box-whisker plots illustrating concentrations of (A) TMAO, (B) choline, (C) betaine, (D) carnitine in plasma of the RDA and 2RDA groups pre-intervention (uncoloured boxes) and after 10 weeks (coloured boxes). The central lines represent means and the error bars represent minimum and maximum values. TMAO: trimethylamine-N-oxide. Two-way repeated measures ANOVA was used for comparison and the diﬀerence between time points in same group was controlled using the Sidak post hoc procedure. RDA 2RDA 0 40 80 120 T M A O ( u M ) p = 0.002 p = 0.004 RDA 2RDA 0 10 20 C h o l i n e ( u M ) A B RDA 2RDA 0 10 20 C h o l i n e ( u M ) B B B A RDA 2RDA 0 40 80 B e t a i ne ( uM ) p = 0.006 p = 0.018 RDA 2RDA 0 50 100 C a r n i t i n e ( u M ) C D RDA 2RDA 0 50 100 C a r n i t i n e ( u M ) D D C Figure 1. Box-whisker plots illustrating concentrations of (A) TMAO, (B) choline, (C) betaine, (D) carnitine in plasma of the RDA and 2RDA groups pre-intervention (uncoloured boxes) and after 10 weeks (coloured boxes). The central lines represent means and the error bars represent minimum and maximum values. TMAO: trimethylamine-N-oxide. Two-way repeated measures ANOVA was used for comparison and the difference between time points in same group was controlled using the Sidak Figure 1. Box-whisker plots illustrating concentrations of (A) TMAO, (B) choline, (C) betaine, (D) carnitine in plasma of the RDA and 2RDA groups pre-intervention (uncoloured boxes) and after 10 weeks (coloured boxes). 3.3. Lipids and Lipoproteins The eﬀects on plasma lipids and lipoproteins are shown in Table 3. Small changes in LDL-C concentrations were dependent on diet and decreased in the RDA group and increased in the 2RDA group (time × diet interaction, p = 0.049). There were no signiﬁcant changes in T-C, HDL-C or TAG in either group. Table 3. Fasting measurements of circulatory cholesterol, triglycerides, glucose, insulin, CRP and anthropometric measurements of the RDA and 2RDA groups pre- and post-intervention. Table 3. Fasting measurements of circulatory cholesterol, triglycerides, glucose, insulin, CRP and anthropometric measurements of the RDA and 2RDA groups pre- and post-intervention. anthropometric measurements of the RDA and 2RDA groups pre and post intervention. 3.3. Lipids and Lipoproteins Variable RDA (Mean ± SD) 2RDA (Mean ± SD) Eﬀect (p) a Pre Post Pre Post Time Diet Time × Diet Lipids & lipoproteins T-C (mmol/L) 4.7 ± 0.9 4.5 ± 1.2 4.6 ± 0.9 4.8 ± 1.2 0.82 0.877 0.055 HDL-C (mmol/L) 1.4 ± 0.5 1.4 ± 0.6 1.3 ± 0.3 1.3 ± 0.3 0.89 0.559 0.662 LDL-C (mmol/L) 3.0 ± 1 2.8 ± 1.1 3.0 ± 0.9 3.2 ± 1.3 0.7 0.668 0.049 * TAG (mmol/L) 1.2 ± 0.4 1.2 ± 0.4 1.2 ± 0.8 1.1 ± 0.5 0.55 0.631 0.871 Metabolic & inﬂammatory biomarkers Glucose (mmol/L) 5.7 ± 0.5 5.6 ± 0.5 5.7 ± 0.6 5.5 ± 0.4 0.020 * 0.639 0.358 Insulin (µU/mL) 9.7 ± 3.9 9.6 ± 3.7 12.5 ± 10.7 9.9 ± 5.6 0.391 0.421 0.42 HbA1c (mmol/mol) 36.1 ± 5 36.8 ± 4.4 36.9 ± 3.7 36.9 ± 3.3 0.397 0.761 0.305 HOMA-IR 2.5 ± 1.2 2.4 ± 1.1 3.5 ± 3.5 2.4 ± 1.3 0.305 0.439 0.425 Matsuda Index 3.2 ± 1.7 3.2 ± 1.8 3.2 ± 1.9 3.5 ± 2.1 0.15 0.649 0.216 CRP (mg/L) 1.8 ± 2.1 2.1 ± 3.3 2 ± 3.1 2.3 ± 1.6 0.644 0.862 0.967 Anthropometrics Total body fat (kg) 25.4 ± 11.5 23.9 ± 11.0 # 23.5 ± 6.8 21.8 ± 6.8 # <0.001 * 0.534 0.954 Android fat (%) 38.4 ± 11 36.7 ± 11 # 39.9 ± 8 36.9 ± 8.0 # <0.001 * 0.831 0.049 * WC (cm) 104.8 ± 15 99.6 ± 14.1 # 101.5 ± 7.6 96 ± 7.5 # <0.001 * 0.377 0.765 Systolic BP (mmHg) 143 ± 14 144 ± 11 142 ± 18 141 ± 16 0.839 0.647 0.663 Diastolic BP (mmHg) 75 ± 7 74 ± 7 76 ± 9 75 ± 9 0.126 0.668 0.975 T-C, total cholesterol; HDL-C, high density lipoprotein cholesterol; LDL-C, low density lipoprotein cholesterol; TAG, triacylglycerides; HbA1c, glycated haemoglobin; HOMA-IR, homeostatic model assessment of insulin resistance; CRP, C-reactive protein; WC, waist circumference; BP, blood pressure. Insulin and glucose concentrations during the OGTT were used to calculate the Matsuda Index. * Signiﬁcant main eﬀect or interaction, p < 0.05. # Diﬀerent from pre-intervention within the same group, p < 0.05. † Diﬀerent between diets at indicated time point. p values were controlled using the Sidak post hoc procedure. a The main eﬀects and interactions were calculated by two-way repeated measures ANOVA. 3.2. TMAO, Choline, Betaine, and Carnitine RDA 2RDA 0 400 800 T M A O ( m m o l / m o l c r e a t i n i n e ) p = 0.007 p = 0.005 RDA 2RDA 0 5 10 C h o l i n e ( m m o l / m o l c r e a t i n i n e ) RDA 2RDA 0 30 60 B e t a i n e ( m m o l / m o l c r e a t i n i n e ) P = 0.032 A B C RDA 2RDA 0 400 800 T M A O ( m m o l / m o l c r e a t i n i n e ) p = 0.007 p = 0.005 RDA 2RDA 0 5 10 C h o l i n e ( m m o l / m o l c r e a t i n i n e ) A B C RDA 2RDA 0 30 60 B e t a i n e ( m m o l / m o l c r e a t i n i n e ) P = 0.032 C ure 2. Box whisker plots illustrating concentrations of (A) TMAO, (B) choline, and (C) betaine in ne of the RDA and 2RDA groups pre-intervention (uncoloured boxes) and after 10 weeks (coloured es). The central lines represent means and the error bars represent minimum and maximum ues. TMAO: trimethylamine-N-oxide. Two-way repeated measures ANOVA was used for Figure 2. Box whisker plots illustrating concentrations of (A) TMAO, (B) choline, and (C) betaine in urine of the RDA and 2RDA groups pre-intervention (uncoloured boxes) and after 10 weeks (coloured boxes). The central lines represent means and the error bars represent minimum and maximum values. TMAO: trimethylamine-N-oxide. Two-way repeated measures ANOVA was used for comparison and diﬀerence between time points in same group was controlled using the Sidak post hoc procedure. 3.2. TMAO, Choline, Betaine, and Carnitine The central lines represent means and the error bars represent minimum and maximum values. TMAO: trimethylamine-N-oxide. Two-way repeated measures ANOVA was used for comparison and the diﬀerence between time points in same group was controlled using the Sidak post hoc procedure. 7 of 14 7 of 14 Nutrients 2019, 11, 2207 post hoc procedure. Figure 2. Box whisker plots illustrating concentrations of (A) TMAO, (B) choline, and (C) betaine in urine of the RDA and 2RDA groups pre-intervention (uncoloured boxes) and after 10 weeks (coloured boxes). The central lines represent means and the error bars represent minimum and maximum values. TMAO: trimethylamine-N-oxide. Two-way repeated measures ANOVA was used for RDA 2RDA 0 400 800 T M A O ( m m o l / m o l c r e a t i n i n e ) p = 0.007 p = 0.005 RDA 2RDA 0 5 10 C h o l i n e ( m m o l / m o l c r e a t i n i n e ) RDA 2RDA 0 30 60 B e t a i n e ( m m o l / m o l c r e a t i n i n e ) P = 0.032 A B C Figure 2. Box whisker plots illustrating concentrations of (A) TMAO, (B) choline, and (C) betaine in urine of the RDA and 2RDA groups pre-intervention (uncoloured boxes) and after 10 weeks (coloured boxes). The central lines represent means and the error bars represent minimum and maximum values. TMAO: trimethylamine-N-oxide. Two-way repeated measures ANOVA was used for comparison and diﬀerence between time points in same group was controlled using the Sidak post hoc procedure. 3 6 Creatinine and Creatinine Clearance 3.6. Creatinine and Creatinine Clearance 3 6 Creatinine and Creatinine Clearance 3.6. Creatinine and Creatinine Clearance The effects on creatinine in plasma and urine and creatinine clearance are summarised in Table 4. There was an interaction effect on plasma creatinine concentrations (time × diet interaction, p = 0.020), which decreased in the RDA group (p = 0.014) but was unchanged in the 2RDA group (p = 0.776). This contributed to a significant difference in creatinine clearance (time × diet interaction, p = 0.013) whereby it increased in the RDA group (p = 0.010) with no change in the 2RDA group. There was an interaction effect of the fraction renal excretion of TMAO (time × diet interaction, p = 0.002) which increased insignificantly in the RDA group (p = 0.229) and decreased in the 2RDA group (p = The eﬀects on creatinine in plasma and urine and creatinine clearance are summarised in Table 4. There was an interaction eﬀect on plasma creatinine concentrations (time × diet interaction, p = 0.020), which decreased in the RDA group (p = 0.014) but was unchanged in the 2RDA group (p = 0.776). This contributed to a signiﬁcant diﬀerence in creatinine clearance (time × diet interaction, p = 0.013) whereby it increased in the RDA group (p = 0.010) with no change in the 2RDA group. There was an interaction eﬀect of the fraction renal excretion of TMAO (time × diet interaction, p = 0.002) which increased insigniﬁcantly in the RDA group (p = 0.229) and decreased in the 2RDA group (p = 0.002). 02). T bl 4 M f i i f l d i l l d i i l d Table 4. Measurements of creatinine from plasma and urine, calculated creatinine clearance, and calculated fractional renal excretion of TMAO of the RDA and 2RDA groups pre- and post-intervention. 0.002). Table 4. Measurements of creatinine from plasma and urine, calculated creatinine clearance, and calculated fractional renal excretion of TMAO of the RDA and 2RDA groups pre- and post- intervention. Variable RDA (Mean ± SD) 2RDA (Mean ± SD) Effect (p) a Pre Post Pre Post Time Diet Time Renal biomarkers Plasma Cr (μmol/L) 90.8 ± 18.9 82 ± 16.1 # 98.1 ± 11 97.4 ± 16.7 † 0.010 * 0.057 0.0 Urinary Cr (mmol/L) 6.3 ± 2.9 5.7 ± 3.8 6.8 ± 2.1 6.3 ± 2.3 0.174 0.673 0. 3.4. Metabolic and Inﬂammatory Biomarkers Eﬀects on metabolic and inﬂammatory biomarkers are shown in Table 3. There was a decrease in fasting blood glucose in both groups (time eﬀect, p = 0.020) without interaction. Other biomarkers of 8 of 14 Nutrients 2019, 11, 2207 glucose tolerance and insulin sensitivity, namely insulin, HOMA-IR, and HbA1c, were not aﬀected by the intervention in either group, nor were the results of the OGTT. CRP concentrations were also not aﬀected by the dietary intervention. 3 6 Creatinine and Creatinine Clearance 3.6. Creatinine and Creatinine Clearance Cr clearance (mL/min) 128.1 ± 50.5 101 ± 37.1 # 116.8 ± 25.2 116 ± 24.4 0.010 * 0.882 0.0 ional renal excretion of TMAO (%) 76.7 ± 68.7 173.2 ± 213.7 305.2 ± 288.3 † 54.3 ± 49.4 † # 0.095 0.619 0.0 Cr, creatinine; TMAO, trimethylamine-N-oxide. * Significant main effect or interaction, p < 0.05. # Table 4. Measurements of creatinine from plasma and urine, calculated creatinine clearance, and calculated fractional renal excretion of TMAO of the RDA and 2RDA groups pre- and post-intervention. Variable RDA (Mean ± SD) 2RDA (Mean ± SD) Eﬀect (p) a Pre Post Pre Post Time Diet Time × Diet Renal biomarkers Plasma Cr (µmol/L) 90.8 ± 18.9 82 ± 16.1 # 98.1 ± 11 97.4 ± 16.7 † 0.010 * 0.057 0.027 * Urinary Cr (mmol/L) 6.3 ± 2.9 5.7 ± 3.8 6.8 ± 2.1 6.3 ± 2.3 0.174 0.673 0.730 Cr clearance (mL/min) 128.1 ± 50.5 101 ± 37.1 # 116.8 ± 25.2 116 ± 24.4 0.010 * 0.882 0.013 * Fractional renal excretion of TMAO (%) 76.7 ± 68.7 173.2 ± 213.7 305.2 ± 288.3 † 54.3 ± 49.4 † # 0.095 0.619 0.002 * Cr, creatinine; TMAO, trimethylamine-N-oxide. * Signiﬁcant main eﬀect or interaction, p < 0.05. # Diﬀerent from pre-intervention within the same group, p < 0.05. † Diﬀerent between diets at indicated time point. p values were controlled using the Sidak post hoc procedure. a Main eﬀects and interactions were calculated by two-way repeated measures ANOVA. Urinary Cr (mmol/L) 6.3 ± 2.9 5.7 ± 3.8 6.8 ± 2.1 6.3 ± 2.3 0.174 0.673 0 Cr clearance (mL/min) 128.1 ± 50.5 101 ± 37.1 # 116.8 ± 25.2 116 ± 24.4 0.010 * 0.882 0 onal renal excretion of TMAO (%) 76.7 ± 68.7 173.2 ± 213.7 305.2 ± 288.3 † 54.3 ± 49.4 † # 0.095 0.619 0 Cr, creatinine; TMAO, trimethylamine-N-oxide. * Significant main effect or interaction, p < 0.05. # Cr, creatinine; TMAO, trimethylamine-N-oxide. * Signiﬁcant main eﬀect or interaction, p < 0.05. # Diﬀerent from pre-intervention within the same group, p < 0.05. † Diﬀerent between diets at indicated time point. p values were controlled using the Sidak post hoc procedure. a Main eﬀects and interactions were calculated by two-way repeated measures ANOVA. 3.5. Body Composition and Blood Pressure The eﬀects of selected measures of body composition are summarised in Table 3 and changes in lean mass and total body weight are reported in detail elsewhere [4]. There was an eﬀect of the intervention on android (abdominal fat deposition) % fat (time × diet interaction, p = 0.049; Figure 3). Total fat mass and waist circumference also decreased in both groups (time eﬀect, all p < 0.001). There was no eﬀect of time or diet on blood pressure. Nutrients 2019, 11, 2207 8 of 14 Figure 3. Change in android % fat from before to after 10 weeks of consuming the RDA (circles) and 2RDA (squares) diets. Each point represents a participant. The central lines represent means and the error bars represent SD. Δ, absolute change; Two-way repeated measures ANOVA was used for comparison. RDA 2RDA -8 -6 -4 -2 0 2 Δ a n d r o i d % f a t p = 0.049 Figure 3. Change in android % fat from before to after 10 weeks of consuming the RDA (circles) and 2RDA (squares) diets. Each point represents a participant. The central lines represent means and the error bars represent SD. ∆, absolute change; Two-way repeated measures ANOVA was used for comparison. RDA 2RDA -8 -6 -4 -2 0 2 Δ a n d r o i d % f a t p = 0.049 Figure 3. Change in android % fat from before to after 10 weeks of consuming the RDA (circles) and 2RDA (squares) diets. Each point represents a participant. The central lines represent means and the error bars represent SD. Δ, absolute change; Two-way repeated measures ANOVA was used for comparison. Figure 3. Change in android % fat from before to after 10 weeks of consuming the RDA (circles) and 2RDA (squares) diets. Each point represents a participant. The central lines represent means and the error bars represent SD. ∆, absolute change; Two-way repeated measures ANOVA was used for comparison. Different time poin 4. Discussion p p g p p interactions were calculated by two-way repeated measures ANOVA. 4 Discussion This study demonstrates that in the short term (10 weeks), a diet containing twice the RDA of protein can elevate fasting circulatory TMAO. Furthermore, this dietary interventional also resulted in 9 of 14 Nutrients 2019, 11, 2207 a marginal increase in LDL-C. However, in contrast to the hypothesis, a high protein diet does not aﬀect other established biomarkers of CVD risk and insulin sensitivity. After 10 weeks of the 2RDA diet, an increase in plasma TMAO concentrations was observed in the majority of these participants, where on average, plasma TMAO increased ﬁve-fold. This response follows previously reported studies where the dietary precursors of TMAO were increased. For instance, observational studies demonstrated a positive correlation between fasting TMAO plasma concentrations and self-reported red meat and ﬁsh intake [30] and comparatively higher concentrations of TMAO were found in the plasma and urine of omnivores compared to vegetarians [24,42]. Randomised studies showed that chronic ingestion of red meat for 4 weeks [40] and daily carnitine supplementation for 12 weeks [43] elevated fasting TMAO plasma concentrations by three-fold and ten-fold, respectively. In acute dietary trials, ingestion of meats, ﬁsh, eggs, and dairy resulted in postprandial transient increases in TMAO in the circulation and in urine, peaking at around 6–8 hours [32], or up to 12 hours after eating ﬁsh [44]. g The TMAO concentrations pre-intervention were diverse and the magnitude of response was varied, particularly in the 2RDA group. In the present study, the pre-intervention TMAO values averaged 10 µM, which is higher than the reported reference values of circulatory TMAO in healthy adults, which are in the range of 3.5 µM [45]. However, previous studies have reported that TMAO concentrations increase with age [45,46]. The inter-individual variation of circulatory TMAO is not unique to this study and can be driven by several factors. In addition to food intake, the gut microbiota strongly inﬂuences circulatory concentrations of TMAO [47]. A higher Firmicutes to Bacteroidetes ratio and reduced microbial diversity is related to a greater TMAO production [29]. TMAO is not only produced by microbiota fermentation of dietary choline and carnitine from TMA [24,25] but can also be absorbed directly from eating ﬁsh [29]. Indeed, ﬁsh is the highest source of circulatory TMAO in postprandial studies when compared to other foods enriched in choline and carnitine [29]. Different time poin 4. Discussion The conversion of TMA produced by the gut microbiota to TMAO is dependent on hepatic ﬂavin-containing monooxygenases (FMOs) [48], therefore, variations in the expression of hepatic FMO can aﬀect TMAO concentrations. Additionally, as the majority of TMAO is ﬁltered through the kidneys and excreted through the urine [49], renal function can have an impact on circulatory TMAO. In our study, the concentrations of circulatory TMAO in the 2RDA group post-intervention reached comparable levels described in a previous carnitine supplementation study [43] although in some, the concentrations of TMAO reached levels seen in chronic kidney disease (CKD) patients, which can be 40-fold greater relative to healthy cohorts [50]. Renal function calculated from creatinine clearance was within the reported reference values for healthy males, which are between 97 to 137 mL/min [51]. The 2RDA diet did not have an eﬀect on creatinine clearance, which is in agreement with ﬁndings from a previous study in older pre-diabetic adults [52], although the eﬀect of higher protein intakes on renal function in healthy adults continues to be debated [53]. Kidney function is an independent risk factor for CVD [54], and impaired renal function is correlated with elevated plasma TMAO concentrations [55,56]. Although creatinine clearance was not aﬀected in the 2RDA group, urinary TMAO levels and the fractional rate of TMAO clearance decreased. This is interesting as it was expected that urinary TMAO concentrations would increase correspondingly with circulating TMAO, such as has been observed in previous dietary studies with high dietary TMAO precursors [29,40]. In particular, the change in urinary TMAO concentrations in the current study was inversely correlated with plasma TMAO concentrations. This suggests that reduced eﬃciency in renal clearance for TMAO may be a contributory factor alongside increased protein intake for elevated circulatory TMAO. Despite the observed increases in TMAO, these did not correspond with concurrent alterations in CVD biomarkers within a 10-week period, although there was an increase in LDL-C in the 2RDA group. Circulating lipids are a common proxy for the prediction of CVD risk and elevated circulating levels of LDL-C and in particular, have a pivotal role in the initiation and progression of atherosclerosis [57]. Despite this, HDL-C, T-C, and TAGs were unchanged by the dietary intervention. Different time poin 4. Discussion The increased 10 of 14 10 of 14 Nutrients 2019, 11, 2207 LDL-C is in contrast with previous studies in overweight adults [58] and type 2 diabetics [59] where there was no eﬀect of increased protein intake on LDL-C, and a recent meta-analysis reported no eﬀect of higher protein diets on LDL-C [60]. In the present study, dietary cholesterol intake also increased in the 2RDA group from pre-intervention compared to the RDA group. Although still controversial, a recent meta-analysis indicated that while dietary cholesterol may not be associated with CVD, there is evidence that it does increase both total cholesterol and LDL-C [61], which could be a contributory factor toward the observed increase in LDL-C. There were improvements in some additional indicators of metabolic health observed in both groups that were not unique to the 2RDA diet. Central adiposity (measured by android % fat and waist circumference) decreased in both groups. Numerous short-term high protein diet studies have demonstrated increased weight loss [62] and reduction in other indices, such as waist circumference [63,64], however, these studies are often conducted in populations who are obese or with metabolic disease. Since the fat loss was also observed in the RDA group, it is possible that the increased ﬁbre intake in both groups contributed to the overall observed fat loss [65]. Fasting insulin and insulin sensitivity (calculated by the Matsuda Index [37]) was not changed in either group, and fasting glucose decreased in both groups over time. A strength of this study was that the macronutrient components were entirely composed of wholefoods rather than relying on supplements. However, certain dietary components may also have had an inadvertent inﬂuence on the measured outcomes. For example, circulatory TMAO levels were elevated signiﬁcantly by seafood intake for up to 12 hours or more postprandially [44]. Despite all blood draws taken after a minimum 10-hour fast, recent seafood intake may have contributed to the TMAO concentration pool, and therefore, incorporating a standardised pre-test diet as a control should be a consideration for future studies. Furthermore, the increased ﬁbre intake in this study may have confounded the putative negative eﬀect of protein due to the modest but well-known consequences of dietary ﬁbre on reducing cholesterol and body weight [66]. However, it should be noted that the diet was designed within the guidelines of healthy eating and so within the context of a healthy diet, the results are relevant. Different time poin 4. Discussion Finally, this study was conducted with healthy elderly men, therefore, the ﬁndings may not extend to healthy women in the same age bracket. Given the known gender diﬀerences in CVD risk biomarkers, particularly post menopause [67], further investigation in women is required. Author Contributions: The authors responsibilities were as follows: Conceptualization, D.C.-S. and C.J.M.; Methodology, D.C.-S., C.J.M. and A.M.M.; Formal analysis, S.M.M., A.M.M. and S.H.Z.; Investigation, C.J.M., A.M.M., S.M.M., R.F.D., N.Z., F.R. and P.S.; Data curation, C.J.M., A.M.M., S.M.M. and N.A.G.; Writing—original draft preparation, S.M.M.; Writing—review and editing, A.M.M., C.J.M., R.F.D., N.Z., F.R., P.S., S.O.K., N.C.R., S.H.Z. and D.C.-S.; Supervision, D.C.-S., C.J.M. and A.M.M.; funding acquisition, D.C.-S., N.C.R. and S.O.K. All authors approved the ﬁnal version of the manuscript for submission. Funding: This study was supported by New Zealand Ministry of Business, Innovation and Employment International Relationships and the European Union (IRSES-318962-BIOAGE), AgResearch Limited through the Strategic Science Investment Fund (Contracts A19079 and A21246: Nutritional strategies for an ageing population), and by a grant from the National Institutes of Health (NIH) to the University of North Carolina (UNC) Clinical Nutrition Research Unit (DK056350). Nutrients 2019, 11, 2207 Nutrients 2019, 11, 2207 Acknowledgments: We thank Petra Hinterleitner, Elisabet Boman, Evelina Malmquist, Elina Holmstrand, Linnea Lind, Faith Chege, Fernando Tom, and Elise Penning for their help with diet preparation and data collection, and Stephen Orena from UNC for TMAO and related metabolite analysis. Conﬂicts of Interest: A.M.M., S.O.K. and N.C.R. are current employees of AgResearch Ltd. None of the remaining authors reported a conﬂict of interest related to the study. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results. References 1. Bauer, J.; Biolo, G.; Cederholm, T.; Cesari, M.; Cruz-Jentoft, A.J.; Morley, J.E.; Phillips, S.; Sieber, C.; Stehle, P.; Teta, D. Evidence-based recommendations for optimal dietary protein intake in older people: A position paper from the PROT-AGE Study Group. J. Am. Med Dir. Assoc. 2013, 14, 542–559. [CrossRef] [PubMed] . WHO. Protein and Amino Acid Requirements in Human Nutrition; World Health Organization: Gen Switzerland, 2007. 3. USDA. 2015–2020 Dietary Guidelines for Americans; USDA: Washington, DC, USA, 2015. 4. Mitchell, C.J.; Milan, A.M.; Mitchell, S.M.; Zeng, N.; Ramzan, F.; Sharma, P.; Knowles, S.O.; Roy, N.C.; Sjödin, A.; Wagner, K.-H.; et al. The eﬀects of dietary protein intake on appendicular lean mass and muscle function in elderly men: A 10-wk randomized controlled trial. Am. J. Clin. Nutr. 2017, 106, 1375–1383. [CrossRef] [PubMed] 5. Raﬁi, M.; Chapman, K.; Elango, R.; Campbell, W.W.; Ball, R.O.; Pencharz, P.B.; Courtney-Martin, G. Dietary Protein Requirement of Men >65 Years Old Determined by the Indicator Amino Acid Oxidation Technique is Higher than the Current Estimated Average Requirement. J. Nutr. 2015, 146, 681–687. [CrossRef] [PubMed] 6. Vellas, B.J.; Hunt, W.C.; Romero, L.J.; Koehler, K.M.; Baumgartner, R.N.; Garry, P.J. Changes in nutritional status and patterns of morbidity among free-living elderly persons: A 10-year longitudinal study. Nutrition 1997, 13, 515–519. [CrossRef] 7. Houston, D.K.; Nicklas, B.J.; Ding, J.; Harris, T.B.; Tylavsky, F.A.; Newman, A.B.; Lee, J.S.; Sahyoun, N.R.; Visser, M.; Kritchevsky, S.B.; et al. Dietary protein intake is associated with lean mass change in older, community-dwelling adults: The Health, Aging, and Body Composition (Health ABC) Study. Am. J. Clin. Nutr. 2008, 87, 150–155. [CrossRef] 8. Beasley, J.M.; Wertheim, B.C.; Lacroix, A.Z.; Prentice, R.L.; Neuhouser, M.L.; Tinker, L.F.; Kritchevsky, S.; Shikany, J.M.; Eaton, C.; Chen, Z.; et al. Biomarker-Calibrated Protein Intake and Physical Function in the Women’s Health Initiative. J. Am. Geriatr. Soc. 2013, 61, 1863–1871. [CrossRef] 9. Arango-Lopera, V.E.; Arroyo, P.; Gutiérrez-Robledo, L.M.; Perez-Zepeda, M.U.; Cesari, M. Mortality as an adverse outcome of sarcopenia. J. Nutr. Health Aging 2013, 17, 259–262. [CrossRef] 10. Hu, F.B.; Stampfer, M.J.; Manson, J.E.; Rimm, E.; Colditz, G.A.; Speizer, F.E.; Hennekens, C.H.; Willett, W.C. Dietary protein and risk of ischemic heart disease in women. Am. J. Clin. Nutr. 1999, 70, 221–227. [CrossRef] 11 Mirmiran P; Hajifaraji M ; Bahadoran Z ; Sarvghadi F ; Azizi F Dietary protein intake is associated with 10. 5. Conclusions In this study, a diet containing 1.6 g protein/kg bodyweight/day increased circulating TMAO concentrations. However, there were no changes in other traditional measures of CVD risk, suggesting that TMAO was regulated by protein intake in elderly males. Based on the variable response, it is also likely that TMAO production and clearance is contingent on individualistic and multi-faceted responses drawn from dietary protein load, gut microbiota composition, and kidney function. The elevated microbial product TMAO seen in this study further demonstrates the potential impact of the microbiome in metabolic and cardiovascular health, where further investigation on the composition and activity in the gut microbiota to detect correlations is warranted. Author Contributions: The authors responsibilities were as follows: Conceptualization, D.C.-S. and C.J.M.; Methodology, D.C.-S., C.J.M. and A.M.M.; Formal analysis, S.M.M., A.M.M. and S.H.Z.; Investigation, C.J.M., A.M.M., S.M.M., R.F.D., N.Z., F.R. and P.S.; Data curation, C.J.M., A.M.M., S.M.M. and N.A.G.; Writing—original draft preparation, S.M.M.; Writing—review and editing, A.M.M., C.J.M., R.F.D., N.Z., F.R., P.S., S.O.K., N.C.R., S.H.Z. and D.C.-S.; Supervision, D.C.-S., C.J.M. and A.M.M.; funding acquisition, D.C.-S., N.C.R. and S.O.K. All authors approved the ﬁnal version of the manuscript for submission. Funding: This study was supported by New Zealand Ministry of Business, Innovation and Employment International Relationships and the European Union (IRSES-318962-BIOAGE), AgResearch Limited through the Strategic Science Investment Fund (Contracts A19079 and A21246: Nutritional strategies for an ageing population), and by a grant from the National Institutes of Health (NIH) to the University of North Carolina (UNC) Clinical Nutrition Research Unit (DK056350). 11 of 14 11 of 14 References Hu, F.B.; Stampfer, M.J.; Manson, J.E.; Rimm, E.; Colditz, G.A.; Speizer, F.E.; Hennekens, C.H.; Willett, W.C. Dietary protein and risk of ischemic heart disease in women. Am. J. Clin. Nutr. 1999, 70, 221–227. [CrossRef] 11. Mirmiran, P.; Hajifaraji, M.; Bahadoran, Z.; Sarvghadi, F.; Azizi, F. Dietary protein intake is associated with favorable cardiometabolic risk factors in adults: Tehran Lipid and Glucose Study. Nutr. Res. 2012, 32, 11. Mirmiran, P.; Hajifaraji, M.; Bahadoran, Z.; Sarvghadi, F.; Azizi, F. Dietary protein intake is associated with favorable cardiometabolic risk factors in adults: Tehran Lipid and Glucose Study. Nutr. Res. 2012, 32, 169–176. [CrossRef] 12. Anderson, J.W.; Fuller, J.; Patterson, K.; Blair, R.; Tabor, A. Soy compared to casein meal replacement shakes with energy-restricted diets for obese women: Randomized controlled trial. Metabolism 2007, 56, 280–288. [CrossRef] 13. Clifton, P.; Bastiaans, K.; Keogh, J.; Clifton, P.; Keogh, J. High protein diets decrease total and abdominal fat and improve CVD risk proﬁle in overweight and obese men and women with elevated triacylglycerol. Nutr. Metab. Cardiovasc. Dis. 2009, 19, 548–554. [CrossRef] [PubMed] 14. Noakes, M.; Keogh, J.B.; Foster, P.R.; Clifton, P.M. Eﬀect of an energy-restricted, high-protein, low-fat diet relative to a conventional high-carbohydrate, low-fat diet on weight loss, body composition, nutritional status, and markers of cardiovascular health in obese women. Am. J. Clin. Nutr. 2005, 81, 1298–1306. [CrossRef] [PubMed] 15. Layman, D.K.; Boileau, R.A.; Erickson, D.J.; Painter, J.E.; Shiue, H.; Sather, C.; Christou, D.D. A reduced ratio of dietary carbohydrate to protein improves body composition and blood lipid proﬁles during weight loss in adult women. J. Nutr. 2003, 133, 411–417. [CrossRef] [PubMed] 12 of 14 12 of 14 Nutrients 2019, 11, 2207 16. Micha, R.; Michas, G.; Mozaﬀarian, D. Unprocessed Red and Processed Meats and Risk of Coronary Artery Disease and Type 2 Diabetes—An Updated Review of the Evidence. Curr. Atheroscler. Rep. 2012, 14, 515–524. [CrossRef] [PubMed] Trichopoulou, A.; Psaltopoulou, T.; Orfanos, P.; Hsieh, C.; Trichopoulos, D. Low-carbohydrate–high-protein diet and long-term survival in a general population cohort. Eur. J. Clin. Nutr. 2007, 61, 575. [CrossRef] 18. Lagiou, P.; Sandin, S.; Lof, M.; Trichopoulos, D.; Adami, H.-O.; Weiderpass, E. Low carbohydrate-high protein diet and incidence of cardiovascular diseases in Swedish women: Prospective cohort study. BMJ 2012, 344, e4026. [CrossRef] [PubMed] 19. Sjogren, P.; Becker, W.; Warensjö, E.; Olsson, E.; Byberg, L.; Gustafsson, I.-B.; Karlström, B.; Cederholm, T. References Mediterranean and carbohydrate-restricted diets and mortality among elderly men: A cohort study in Sweden. Am. J. Clin. Nutr. 2010, 92, 967–974. [CrossRef] 20. Song, Y.; Manson, J.E.; Buring, J.E.; Liu, S. A prospective study of red meat consumption and type 2 diabetes in middle-aged and elderly women: The women’s health study. Diabetes Care 2004, 27, 2108–2115. [CrossRef] 21. Pounis, G.; Tyrovolas, S.; Antonopoulou, M.; Zeimbekis, A.; Anastasiou, F.; Bountztiouka, V.; Metallinos, G.; Gotsis, E.; Lioliou, E.; Polychronopoulos, E.; et al. Long-term animal-protein consumption is associated with an increased prevalence of diabetes among the elderly: The Mediterranean islands (MEDIS) study. Diabetes Metab. 2010, 36, 484–490. [CrossRef] 22. Friedman, A.N. High-protein diets: Potential eﬀects on the kidney in renal health and disease. Am. J. Kidney Dis. 2004, 44, 950–962. [CrossRef] 23. Tang, W.W.; Wang, Z.; Levison, B.S.; Koeth, R.A.; Britt, E.B.; Fu, X.; Wu, Y.; Hazen, S.L. Intestinal Microbial Metabolism of Phosphatidylcholine and Cardiovascular Risk. N. Engl. J. Med. 2013, 368, 1575–1584. [CrossRef] [PubMed] 24. Koeth, R.A.; Wang, Z.; Levison, B.S.; Buﬀa, J.A.; Org, E.; Sheehy, B.T.; Britt, E.B.; Fu, X.; Wu, Y.; Li, L.; et al. Intestinal microbiota metabolism of L-carnitine, a nutrient in red meat, promotes atherosclerosis. Nat. Med. 2013, 19, 576–585. [CrossRef] [PubMed] 25. Wang, Z.; Klipfell, E.; Bennett, B.J.; Koeth, R.; Levison, B.S.; Dugar, B.; Feldstein, A.E.; Britt, E.B.; Fu, X.; Chung, Y.-M.; et al. Gut ﬂora metabolism of phosphatidylcholine promotes cardiovascular disease. Nature 2011, 472, 57–63. [CrossRef] [PubMed] 26. Zeisel, S.H.; Warrier, M. Trimethylamine N -Oxide, the Microbiome, and Heart and Kidney Disease. Annu. Rev. Nutr. 2017, 37, 157–181. [CrossRef] [PubMed] 27. Zeisel, S.H.; Da Costa, K.-A. Choline: An Essential Nutrient for Public Health. Nutr. Rev. 2009, 67, 615–623. [CrossRef] [PubMed] 28. Demarquoy, J.; Georges, B.; Rigault, C.; Royer, M.-C.; Clairet, A.; Soty, M.; Lekounoungou, S.; Le Borgne, F. Radioisotopic determination of l-carnitine content in foods commonly eaten in Western countries. Food Chem. 2004, 86, 137–142. [CrossRef] 29. Cho, C.E.; Taesuwan, S.; Malysheva, O.V.; Bender, E.; Tulchinsky, N.F.; Yan, J.; Sutter, J.L.; Caudill, M.A. Trimethylamine-N-oxide (TMAO) response to animal source foods varies among healthy young men and is inﬂuenced by their gut microbiota composition: A randomized controlled trial. Mol. Nutr. Food Res. 2016, 61, 1600324. [CrossRef] [PubMed] 30. Krüger, R.; Merz, B.; Rist, M.J.; Ferrario, P.G.; Bub, A.; Kulling, S.E.; Watzl, B. References Associations of current diet with plasma and urine TMAO in the KarMeN study: Direct and indirect contributions. Mol. Nutr. Food Res. 2017, 61, 1700363. [CrossRef] 31. Rohrmann, S.; Allenspach, M.; Linseisen, J.; Von Eckardstein, A.; Müller, D. Plasma Concentrations of Trimethylamine-N-oxide Are Directly Associated with Dairy Food Consumption and Low-Grade Inﬂammation in a German Adult Population. J. Nutr. 2015, 146, 283–289. [CrossRef] 32. Miller, C.A.; Corbin, K.D.; Da Costa, K.-A.; Zhang, S.; Zhao, X.; Galanko, J.A.; Blevins, T.; Bennett, B.J.; O’Connor, A.; Zeisel, S.H. Eﬀect of egg ingestion on trimethylamine-N-oxide production in humans: A randomized, controlled, dose-response study. Am. J. Clin. Nutr. 2014, 100, 778–786. [CrossRef] 33. Morley, J.E.; Argilés, J.M.; Evans, W.J.; Bhasin, S.; Cella, D.; Deutz, N.E.P.; Doehner, W.; Fearon, K.C.H.; Ferrucci, L.; Hellerstein, M.K.; et al. Nutritional Recommendations for the Management of Sarcopenia. J. Am. Med Dir. Assoc. 2010, 11, 391–396. [CrossRef] [PubMed] 34. Gerrior, S.; Juan, W.; Peter, B. An Easy Approach to Calculating Estimated Energy Requirements. Prev. Chronic Dis. 2006, 3, 3. 13 of 14 13 of 14 Nutrients 2019, 11, 2207 35. Zhao, X.; Zhang, S.; Zeisel, S.H.; Zeisel, S.H.; Zeisel, S. Rapid LC-MRM-MS assay for simultaneous quantiﬁcation of choline, betaine, trimethylamine, trimethylamineN-oxide, and creatinine in human plasma and urine. Electrophoresis 2015, 36, 2207–2214. [CrossRef] [PubMed] 35. Zhao, X.; Zhang, S.; Zeisel, S.H.; Zeisel, S.H.; Zeisel, S. Rapid LC-MRM-MS assay for simultaneous quantiﬁcation of choline, betaine, trimethylamine, trimethylamineN-oxide, and creatinine in human plasma and urine. Electrophoresis 2015, 36, 2207–2214. [CrossRef] [PubMed] 36. Mackenzie, T.; Clark, N.G.; Bistrian, B.R.; Flatt, J.P.; Hallowell, E.M.; Blackburn, G.L. A simple method for estimating nitrogen balance in hospitalized patients: A review and supporting data for a previously proposed technique. J. Am. Coll. Nutr. 1985, 4, 575–581. [CrossRef] [PubMed] 37. Matsuda, M.; DeFronzo, R.A. Insulin sensitivity indices obtained from oral glucose tolerance testing: Comparison with the euglycemic insulin clamp. Diabetes Care 1999, 22, 1462–1470. [CrossRef] 38. Matthews, D.; Hosker, J.; Rudenski, A.; Naylor, B.; Treacher, D.; Turner, R. Homeostasis model assessment: Insulin resistance and β-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985, 28, 412–419. [CrossRef] 39. Luscombe-Marsh, N.D.; Noakes, M.; Wittert, G.A.; Keogh, J.B.; Foster, P.; Clifton, P.M. Carbohydrate-restricted diets high in either monounsaturated fat or protein are equally eﬀective at promoting fat loss and improving blood lipids. Am. J. Clin. Nutr. 2005, 81, 762–772. [CrossRef] 40. References Wang, Z.; Bergeron, N.; Levison, B.S.; Li, X.S.; Chiu, S.; Jia, X.; Koeth, R.A.; Li, L.; Wu, Y.; Tang, W.H.W.; et al. Impact of chronic dietary red meat, white meat, or non-meat protein on trimethylamine N-oxide metabolism and renal excretion in healthy men and women. Eur. Hear. J. 2019, 40, 583–594. [CrossRef] 41. Health, M.O. Food and Nutrition Guidelines for Healthy Older People: A background Paper; Ministry of Health: Wellington, New Zealand, 2013. 42. De Filippis, F.; Pellegrini, N.; Vannini, L.; Jeﬀery, I.B.; la Storia, A.; Laghi, L.; Serrazanetti, D.I.; di Cagno, R.; Ferrocino, I.; Lazzi, C. High-level adherence to a Mediterranean diet beneﬁcially impacts the gut microbiota and associated metabolome. Gut 2016, 65, 1812–1821. [CrossRef] 43. Samulak, J.J.; Sawicka, A.K.; Hartmane, D.; Grinberga, S.; Pugovics, O.; Lysiak-Szydlowska, W.; Olek, R.A. l-Carnitine Supplementation Increases Trimethylamine-N-Oxide but not Markers of Atherosclerosis in Healthy Aged Women. Ann. Nutr. Metab. 2019, 74, 11–17. [CrossRef] 44. Stanstrup, J.; Schou, S.S.; Holmer-Jensen, J.; Hermansen, K.; Dragsted, L.O. Whey Protein Delays Gastric Emptying and Suppresses Plasma Fatty Acids and Their Metabolites Compared to Casein, Gluten, and Fish Protein. J. Proteome Res. 2014, 13, 2396–2408. [CrossRef] [PubMed] 45. Wang, Z.; Levison, B.S.; Hazen, J.E.; Donahue, L.; Li, X.-M.; Hazen, S.L. Measurement of trimethylamine-N-oxide by stable isotope dilution liquid chromatography tandem mass spectrometry. Anal. Biochem. 2014, 455, 35–40. [CrossRef] [PubMed] 46. Li, T.; Chen, Y.; Gua, C.; Li, X. Elevated Circulating Trimethylamine N-Oxide Levels Contribute to Endothelial Dysfunction in Aged Rats through Vascular Inﬂammation and Oxidative Stress. Front. Physiol. 2017, 8, 8. [CrossRef] [PubMed] 47. Romano, K.A.; Vivas, E.I.; Amador-Noguez, D.; Rey, F.E. Intestinal Microbiota Composition Modulates Choline Bioavailability from Diet and Accumulation of the Proatherogenic Metabolite Trimethylamine-N-Oxide. mBio 2015, 6, 02481. [CrossRef] [PubMed] 48. Lang, D.H.; Yeung, C.K.; Peter, R.M.; Ibarra, C.; Gasser, R.; Itagaki, K.; Philpot, R.M.; Rettie, A.E. Isoform speciﬁcity of trimethylamine N-oxygenation by human ﬂavin-containing monooxygenase (FMO) and P450 enzymes: Selective catalysis by FMO3. Biochem. Pharmacol. 1998, 56, 1005–1012. [CrossRef] 49. Zeisel, S.H.; Wishnok, J.S.; Blusztajn, J.K. Formation of methylamines from ingested choline and lec Pharmacol. Exp. Ther. 1983, 225, 320–324. [PubMed] 50. Missailidis, C.; Hallqvist, J.; Qureshi, A.R.; Barany, P.; Heimbürger, O.; Lindholm, B.; Stenvinkel, P.; Bergman, P. Serum Trimethylamine-N-Oxide Is Strongly Related to Renal Function and Predicts Outcome in Chronic Kidney Disease. PLoS ONE 2016, 11, e0141738. [CrossRef] 51. References James, G.D.; Sealey, J.E.; Alderman, M.; Ljungman, S.; Mueller, F.B.; Pecker, M.S.; Laragh, J.H. A Longitudinal Study of Urinary Creatinine and Creatinine Clearance in Normal Subjects: Race, Sex, and Age Diﬀerences. Am. J. Hypertens. 1988, 1, 124–131. [CrossRef] 52. Møller, G.; Andersen, J.R.; Ritz, C.; Silvestre, M.P.; Navas-Carretero, S.; Jalo, E.; Christensen, P.; Simpson, E.; Taylor, M.; Martinez, J.A.; et al. Higher Protein Intake Is Not Associated with Decreased Kidney Function in Pre-Diabetic Older Adults Following a One-Year Intervention—A Preview Sub-Study. Nutrients 2018, 10, 54. [CrossRef] Nutrients 2019, 11, 2207 14 of 14 14 of 14 53. Martin, W.F.; Armstrong, L.E.; Rodriguez, N.R. Dietary protein intake and renal function. Nutr. Metab. 2005, 2, 25. [CrossRef] 53. Martin, W.F.; Armstrong, L.E.; Rodriguez, N.R. Dietary protein intake and renal function. Nutr. Metab. 2005, 2, 25. [CrossRef] 54. Sarnak, M.J.; Levey, A.S.; Schoolwerth, A.C.; Coresh, J.; Culleton, B.; Hamm, L.L.; McCullough, P.A.; Kasiske, B.L.; Kelepouris, E.; Klag, M.J. Kidney disease as a risk factor for development of cardiovascular disease: A statement from the American Heart Association Councils on Kidney in Cardiovascular Disease, High Blood Pressure Research, Clinical Cardiology, and Epidemiology and Prevention. Circulation 2003, 108, 2154–2169. [CrossRef] [PubMed] 55. Kim, R.B.; Morse, B.L.; Djurdjev, O.; Tang, M.; Muirhead, N.; Barrett, B.; Holmes, D.T.; Madore, F.; Clase, C.M.; Rigatto, C. Advanced chronic kidney disease populations have elevated trimethylamine N-oxide levels associated with increased cardiovascular events. Kidney Int. 2016, 89, 1144–1152. [CrossRef] [PubMed] 56. Bain, M.A.; Faull, R.; Fornasini, G.; Milne, R.W.; Evans, A.M. Accumulation of trimethylamine and trimethylamine-N-oxide in end-stage renal disease patients undergoing haemodialysis. Nephrol. Dial. Transplant. 2006, 21, 1300–1304. [CrossRef] [PubMed] 57. Witztum, J.L.; Steinberg, D. Role of oxidized low density lipoprotein in atherogenesis. J. Clin. Investig. 1991, 88, 1785–1792. [CrossRef] [PubMed] 58. Wright, C.S.; Zhou, J.; Sayer, R.D.; Kim, J.E.; Campbell, W.W. Eﬀects of a High-Protein Diet Including Whole Eggs on Muscle Composition and Indices of Cardiometabolic Health and Systemic Inﬂammation in Older Adults with Overweight or Obesity: A Randomized Controlled Trial. Nutrients 2018, 10, 946. [CrossRef] 59. Pearce, K.L.; Clifton, P.M.; Noakes, M. Egg consumption as part of an energy-restricted high-protein diet improves blood lipid and blood glucose proﬁles in individuals with type 2 diabetes. Br. J. Nutr. 2011, 105, 584–592. [CrossRef] [PubMed] 60. Santesso, N.; Akl, E.A.; Bianchi, M.; Mente, A.; Mustafa, R.; Heels-Ansdell, D.; Schünemann, H.J. References Eﬀects of higher- versus lower-protein diets on health outcomes: A systematic review and meta-analysis. Eur. J. Clin. Nutr. 2012, 66, 780–788. [CrossRef] 61. Berger, S.; Raman, G.; Vishwanathan, R.; Jacques, P.F.; Johnson, E.J. Dietary cholesterol and cardiovascular disease: A systematic review and meta-analysis. Am. J. Clin. Nutr. 2015, 102, 276–294. [CrossRef] 62. Weigle, D.S.; Breen, P.A.; Matthys, C.C.; Callahan, H.S.; Meeuws, K.E.; Burden, V.R.; Purnell, J.Q. A high-protein diet induces sustained reductions in appetite, ad libitum caloric intake, and body weight despite compensatory changes in diurnal plasma leptin and ghrelin concentrations. Am. J. Clin. Nutr. 2005, 82, 41–48. [CrossRef] 63. Krieger, J.W.; Sitren, H.S.; Daniels, M.J.; Langkamp-Henken, B. Eﬀects of variation in protein and carbohydrate intake on body mass and composition during energy restriction: A meta-regression. Am. J. Clin. Nutr. 2006, 83, 260–274. [CrossRef] 64. Beavers, K.M.; Nesbit, B.A.; Kiel, J.R.; Sheedy, J.L.; Arterburn, L.M.; Collins, A.E.; Ford, S.A.; Henderson, R.M.; Coleman, C.D.; Beavers, D.P. Eﬀect of an Energy-Restricted, Nutritionally Complete, Higher Protein Meal Plan on Body Composition and Mobility in Older Adults with Obesity: A Randomized Controlled Trial. J. Gerontol. Ser. 2019, 74, 929–935. [CrossRef] [PubMed] 65. Mytton, O.T.; Nnoaham, K.; Eyles, H.; Scarborough, P.; Ni Mhurchu, C. Systematic review and meta-analysis of the eﬀect of increased vegetable and fruit consumption on body weight and energy intake. BMC Public Heal. 2014, 14, 886. [CrossRef] [PubMed] 66. Djoussé, L.; Arnett, D.K.; Coon, H.; Province, M.A.; Moore, L.L.; Ellison, R.C. Fruit and vegetable consumption and LDL cholesterol: The National Heart, Lung, and Blood Institute Family Heart Study. Am. J. Clin. Nutr. 2004, 79, 213–217. [CrossRef] [PubMed] 67. Bello, N.; Mosca, L. Epidemiology of coronary heart disease in women. Prog. Cardiovasc. Dis. 2004, 46, 287–295. [CrossRef] [PubMed] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
https://openalex.org/W2006702661	https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0075984&type=printable	English	null	Lactate Dehydrogenase A Expression Is Necessary to Sustain Rapid Angiogenesis of Pulmonary Microvascular Endothelium	PloS one	2,013	cc-by	8,027	Abstract doi:10.1371/journal.pone.0075984 Editor: Tim Lahm, Indiana University, United States of America Received May 31, 2013; Accepted August 19, 2013; Published September 26, 2013 Copyright: © 2013 Parra-Bonilla et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This work was supported by HL60024 and HL66299 (to TS) and an American Heart Association Predoctoral Fellowship (to GPB). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist. * E-mail: tstevens@jaguar1.usouthal.edu Received May 31, 2013; Accepted August 19, 2013; Published September 26, 2013 require rapid endothelial cell proliferation. In more severe cases, capillary repair involves an angiogenic response to replace lost vessels. Capillary endothelium undergoes robust angio-proliferative responses, both in vivo and in vitro [6,7,8,9]. PMVECs, for example, are rapidly angiogenic [10,11]. These cells proliferate more quickly than do pulmonary artery endothelial cells (PAECs), and they are enriched with a higher number of replication-competent cells. Some of the molecular components that contribute to the angio-proliferative phenotype of PMVECs have been elucidated, including nucleosome assembly protein-1 (NAP-1) [11], cyclin D1, cdk2 and cdk4 [12]. NAP-1 is an epigenetic factor implicated in transcriptional regulation, and cyclin D1 and cdk are involved in cell cycle control. Therefore, all of these proteins constitute key molecular fingerprints of PMVEC hyperproliferation. Glenda Parra-Bonilla1,5, Diego F. Alvarez1,2,5, Mikhail Alexeyev3,5, Audrey Vasauskas4,5, Troy Stevens1,2,5* 1 Department of Pharmacology, University of South Alabama, Mobile, Alabama, United States of America, 2 Department of Medicine, University of South Alabama, Mobile, Alabama, United States of America, 3 Department of Cell Biology and Neuroscience, University of South Alabama, Mobile, Alabama, United States of America, 4 Department of Biochemistry and Molecular Biology, University of South Alabama, Mobile, Alabama, United States of America, 5 Center for Lung Biology, University of South Alabama, Mobile, Alabama, United States of America Abstract Angiogenesis is a fundamental property of endothelium, yet not all endothelial cells display equivalent angiogenic responses; pulmonary microvascular endothelial cells undergo rapid angiogenesis when compared to endothelial cells isolated from conduit vessels. At present it is not clear how pulmonary microvascular endothelial cells fulfill the bioenergetic demands that are necessary to sustain such rapid blood vessel formation. We have previously established that pulmonary microvascular endothelial cells utilize aerobic glycolysis to generate ATP during growth, a process that requires the expression of lactate dehydrogenase A to convert pyruvate to lactate. Here, we test the hypothesis that lactate dehydrogenase A is required for pulmonary microvascular endothelial cells to sustain rapid angiogenesis. To test this hypothesis, Tet-On and Tet-Off conditional expression systems were developed in pulmonary microvascular endothelial cells, where doxycycline is utilized to induce lactate dehydrogenase A shRNA expression. Expression of LDH-A shRNA induced a time-dependent decrease in LDH-A protein, which corresponded with a decrease in glucose consumption from the media, lactate production and cell growth; re-expression of LDH-A rescued each of these parameters. LDH-A silencing greatly reduced network formation on Matrigel in vitro, and decreased blood vessel formation in Matrigel in vivo. These findings demonstrate that LDH-A is critically important for sustaining the rapid angiogenesis of pulmonary microvascular endothelial cells. Citation: Parra-Bonilla G, Alvarez DF, Alexeyev M, Vasauskas A, Stevens T (2013) Lactate Dehydrogenase A Expression Is Necessary to Sustain Rapid Angiogenesis of Pulmonary Microvascular Endothelium. PLoS ONE 8(9): e75984. doi:10.1371/journal.pone.0075984 Editor: Tim Lahm, Indiana University, United States of America Received May 31, 2013; Accepted August 19, 2013; Published September 26, 2013 Copyright: © 2013 Parra-Bonilla et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This work was supported by HL60024 and HL66299 (to TS) and an American Heart Association Predoctoral Fellowship (to GPB). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist. * E-mail: tstevens@jaguar1.usouthal.edu Citation: Parra-Bonilla G, Alvarez DF, Alexeyev M, Vasauskas A, Stevens T (2013) Lactate Dehydrogenase A Expression Is Necessary to Sustain Rapid Angiogenesis of Pulmonary Microvascular Endothelium. PLoS ONE 8(9): e75984. Lactate Dehydrogenase A Expression Is Necessary to Sustain Rapid Angiogenesis of Pulmonary Microvascular Endothelium Glenda Parra-Bonilla1,5, Diego F. Alvarez1,2,5, Mikhail Alexeyev3,5, Audrey Vasauska Citation: Parra-Bonilla G, Alvarez DF, Alexeyev M, Vasauskas A, Stevens T (2013) Lactate Dehydrogenase A Expression Is Necessary to Sustain Rapid Angiogenesis of Pulmonary Microvascular Endothelium. PLoS ONE 8(9): e75984. doi:10.1371/journal.pone.0075984 September 2013 \| Volume 8 \| Issue 9 \| e75984 Citation: Parra-Bonilla G, Alvarez DF, Alexeyev M, Vasauskas A, Stevens T (2013) Lactate Dehydrogenase A Expression Is Necessary to Sustain Rapid Angiogenesis of Pulmonary Microvascular Endothelium. PLoS ONE 8(9): e75984. doi:10.1371/journal.pone.0075984 Editor: Tim Lahm, Indiana University, United States of America Received May 31, 2013; Accepted August 19, 2013; Published September 26, 2013 Copyright: © 2013 Parra-Bonilla et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This work was supported by HL60024 and HL66299 (to TS) and an American Heart Association Predoctoral Fellowship (to GPB). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist. * E-mail: tstevens@jaguar1.usouthal.edu Isolation and culture of pulmonary endothelial cells Cells used for this study were isolated and sub-cultured from male CD rats weighing 200-250 g using methods detailed previously [3,10,13]. Figure 1. Map of the Tet-Off construct and induction of LDH-A shRNA. (A) PMVECs were infected with a retrovirus, enabling reverse tetracycline-controlled transactivator protein (tTA) expression. Cells were selected to homogeneity using blasticidin, reinfected with a lentivirus [shown in (B)], and selected to homogeneity using puromycin. The resulting double-transfection enabled the expression of a LDH-A short hairpin RNA (shRNA) in the absence of doxycycline. Bsr, blasticidin resistance gene; EGFP, enhanced green fluorescent protein; HIV RRE, human immunodeficiency virus Rev response element; IRES EMV, encephalomyocarditis virus internal ribosome entry site; LTR, retro/lentiviral long terminal repeat; PAC, puromycin resistance gene; PSV40, simian virus 40 promoter; PTet, doxycycline-regulated promoter; wPRE, woodchuck hepatitis virus posttranscriptional regulatory element; mir, 5’–3’ flanking sequence derived from the murine mir (micro-RNA gene)-15. (C) Doxycycline retrieval promoted expression of the red fluorescent protein mCherry in infected PMVECs. Figure 1. Map of the Tet-Off construct and induction of LDH-A shRNA. (A) PMVECs were infected with a retrovirus, enabling reverse tetracycline-controlled transactivator protein (tTA) expression. Cells were selected to homogeneity using blasticidin, reinfected with a lentivirus [shown in (B)], and selected to homogeneity using puromycin. The resulting double-transfection enabled the expression of a LDH-A short hairpin RNA (shRNA) in the absence of doxycycline. Bsr, blasticidin resistance gene; EGFP, enhanced green fluorescent protein; HIV RRE, human immunodeficiency virus Rev response element; IRES EMV, encephalomyocarditis virus internal ribosome entry site; LTR, retro/lentiviral long terminal repeat; PAC, puromycin resistance gene; PSV40, simian virus 40 promoter; PTet, doxycycline-regulated promoter; wPRE, woodchuck hepatitis virus posttranscriptional regulatory element; mir, 5’–3’ flanking sequence derived from the murine mir (micro-RNA gene)-15. (C) Doxycycline retrieval promoted expression of the red fluorescent protein mCherry in infected PMVECs. Introduction Lung endothelium is generally thought to be quiescent in the healthy postnatal vasculature [1]. This concept has been disputed, as some investigators believe that there is a program for lung tissue homeostasis, referred as the lung structure maintenance program, in which cells within the alveolus continuously undergo apoptosis and proliferation [2]. The principle of the lung structure maintenance program is that such turnover is necessary to replace injured or senescent cells [2]; disruption of this delicate balance results in vascular disease. Pulmonary microvascular endothelial cells (PMVEC) form the vascular barrier of the alveolar-capillary membrane, and limit fluid filtration into the alveoli and interstitium, protecting the gas-exchange region [3,4,5]. Any disruption of the endothelium in this region must be promptly repaired, a process that may We recently sought to determine how PMVECs sustain their bioenergetic needs during rapid proliferation, and observed that PLOS ONE \| www.plosone.org September 2013 \| Volume 8 \| Issue 9 \| e75984 1 LDH-A in Pulmonary Endothelial Angiogenesis Figure 1. Map of the Tet-Off construct and induction of LDH-A shRNA. (A) PMVECs were infected with a retrovirus, enabling reverse tetracycline-controlled transactivator protein (tTA) expression. Cells were selected to homogeneity using blasticidin, reinfected with a lentivirus [shown in (B)], and selected to homogeneity using puromycin. The resulting double-transfection enabled the expression of a LDH-A short hairpin RNA (shRNA) in the absence of doxycycline. Bsr, blasticidin resistance gene; EGFP, enhanced green fluorescent protein; HIV RRE, human immunodeficiency virus Rev response element; IRES EMV, encephalomyocarditis virus internal ribosome entry site; LTR, retro/lentiviral long terminal repeat; PAC, puromycin resistance gene; PSV40, simian virus 40 promoter; PTet, doxycycline-regulated promoter; wPRE, woodchuck hepatitis virus posttranscriptional regulatory element; mir, 5’–3’ flanking sequence derived from the murine mir (micro-RNA gene)-15. (C) Doxycycline retrieval promoted expression of the red fluorescent protein mCherry in infected PMVECs. / they utilize aerobic glycolysis to maintain ATP concentrations that are roughly 2-fold higher than in pulmonary artery endothelial cells [13]. Lactate dehydrogenase-A (LDH-A) expression was needed to consume glucose and generate lactic acidosis that paralleled rapid proliferation. It is not presently clear whether PMVECs similarly utilize aerobic glycolysis to support neo-angiogenesis. Our present studies therefore tested the hypothesis that LDH-A expression is necessary to sustain rapid neo-angiogenesis in PMVECs. Evaluation of the Tet-Off system to knockdown LDH-A Evaluation of the Tet-Off system to knockdown LDH-A Growth curves with uninduced (with doxycycline) and induced (no doxycycline) cells containing the shRNA sequence that successfully inhibited LDH-A expression were performed [13]. Daily protein samples were collected for western blots as well as medium samples to measure glucose depletion from the media and lactate production. Rescue experiments were performed where cells were grown in the absence of doxycycline (induced cells) for 6 days, and then doxycycline was added to the media. During this experiment the same parameters were measured: cell number, protein levels, glucose depletion and lactate production. Ethics Statement Animal studies were conducted according to guidelines, and following approval, of the University of South Alabama Animal Care and Use Committee. Generation of LDH-A knockdown using a Tet-Off system Using a retro-lentiviral infection method [14], we generated a double transfected PMVEC line that expresses the tetracycline transactivator protein (tTA) and LDH-A shRNA (Figure 1A and 1B). In the first step, cells were infected with a retrovirus encoding rTA and selected to homogeneity with blasticidin, and reinfected with lentivirus containing the shRNA. Cells were then selected with puromycin, and infected cells were grown and expanded in the presence of doxycycline (2 µg/ml). Cells were induced by doxycycline-withdrawal to allow the interaction of tTA with TRE, which promoted shRNA transcription. Transcription efficiency was assessed by the intensity of red fluorescent protein mCherry, which is transcribed along with LDH-A shRNA (Figure 1C). The same shRNA sequence that we previously used to knockdown LDH-A in a Tet-On system was used to develop the Tet-Off system [13]. doi: 10.1371/journal.pone.0075984.g001 containing 1% SDS, 10% DMSO and 200 mM sodium acetate, plus a proteinase inhibitor cocktail. Protein concentration was quantified using a modified Lowry assay from Sigma and albumin was used to make a standard curve for the assay. Samples were boiled for 5 minutes, and equal amounts of total protein were loaded into precast SDS/PAGE 4-12% gradient gels from BioRad. The gels were run at 200 V for 1.5 hours. Proteins were transferred overnight at 4°C using nitrocellulose membranes from BioRad. Primary antibody from Santa Cruz Biotechnology was used for the immunoblotting of LDH-A (sc-27230) (1:200 dilution). After two hours of incubation with the primary antibody, the membrane was incubated with a secondary antibody conjugated with horseradish peroxidase (1:5000). Finally, a pico-western detection kit from Pierce (SuperSignal West Pico) was used to visualize the bands. Network formation on Matrigel in vitro using Tet-On and Tet-Off system Network formation on Matrigel in vitro using Tet-On and Tet-Off system Statistical analyses One- and two-way ANOVA and Student’s t-tests were used for statistical analyses, as appropriate. Bonferroni post hoc was used, as appropriate. Significance was denoted as P < 0.05. Removal of the Matrigel plugs After 10 days, the animals were anesthetized using either ketamine 75 mg/kg of body weight or Nembutal (50 mg/kg of body weight) and the skin from the abdomen was incised to reveal the plugs [10]. Plugs were carefully removed and placed into plastic cassettes and immersion fixed in 10% buffered formalin. Pictures were taken from the un-processed plugs. Paraffin blocks were made from the fixed specimens and cut in 5 µm sections. These sections were used to make un-stained and hematoxylin and eosin (HE) stained slides. Anesthetized animals were euthanized by exsanguination. Angiogenesis assay in vivo: Matrigel plugs in rats CD male rats (Charles River Laboratories) were anesthetized using isoflurane gas according to University of South Alabama Animal Care and Use Committee. Once an anesthetic plane was achieved, cells mixed with Matrigel were injected in the abdomen of the animals (approximately 5 minutes per rat) [10]. Cells engineered using Tet-On (non-induced cells) and Tet Off (induced cells) systems and control cell lines, including wild type PMVECs and cells engineered with Tet-On and Tet-Off constructs lacking LDH-A shRNA, were grown to confluence. Cells were trypsinized and counted. 3.75 x 105 cells were suspended in 250 µl and mixed with 500 µl of un-polymerized Matrigel (BD laboratories) at the moment of the injection [10]. The mixture was injected subcutaneously into the left and right abdomen of the rats using a 23-gauge needle. Tet-On and Tet- Off cells were injected into the same rat to minimize variations between animals. Control cells and Matrigel alone without cells were randomly distributed in different animals. The Matrigel plugs were left in the animal for 10 days. Following the procedure rats were transferred back to their cages where they rapidly recovered. Western blots Samples from the lentivirus-infected cells (PMVECs, Tet-Off cell line) were collected during log phase growth and used to extract total protein. Cells were lysed using a lysis buffer September 2013 \| Volume 8 \| Issue 9 \| e75984 PLOS ONE \| www.plosone.org 2 LDH-A in Pulmonary Endothelial Angiogenesis Network formation on Matrigel in vitro using Tet-On and Tet-Off system Immunohistochemistry Unstained paraffin sections were heated to 60°C for 20 minutes and deparaffinized with three washes of xylene for 5 minutes each. Tissue was rehydrated using a gradual decrease in alcohol (100% and 95%) content for 5 minutes and two times each concentration. Slides were then immersed in distilled water for 5 minutes total. Antigen retrieval was performed by microwaving slides in pre-boiled 0.01 M sodium citrate buffer (pH 5.0) for 1 minute at low power followed by 2 minutes at room temperature. This process was repeated nine additional times. Slides were allowed to cool for 20 minutes at room temperature and washed once in distilled water for 5 minutes followed by three 5-minute PBST washes. Sections were incubated with 3% H2O2 for 10 minutes to suppress endogenous peroxidase activity followed by three 5-minute PBST washes. Slides were blocked for 1 hour in a humidified chamber at room temperature with blocking buffer: 5 mL PBST plus one drop (~50 µL) horse serum from VECTASTAIN. Then, a primary antibody for GFP (Ab290 from Abcam, used at 1:200) was diluted as noted in fresh blocking buffer and incubated overnight at 4°C in a humidified chamber. Slides were rinsed with PBST three times for 5 minutes each and then incubated with biotinylated horse anti-rabbit secondary antibody (VECTOR) for 45 minutes at room temperature. Slides were washed three times with PBST for 5 minutes each followed by incubation with Avidin-Biotin complex for 45 minutes at room temperature (VECTASTAIN UNIVERSAL Elite ABC Kit, PK-6200). Slides were rinsed two times with PBST for 5 minutes each, and then developed using 3,3’- diaminobenzidine (DAB) solution substrate-chromogen (Thermo Ultravision Plus, Thermo Scientific) and immediately immersed in distilled water to stop the reaction. Slides were counterstained with Hematoxylin QS (VECTOR) for 2 minutes and rinsed several times with distilled water. Finally, a coverslip was placed to overlay slides using SHUR/Mount (Electron Microscopy Services). Tet-On (induced), Tet-Off (induced and non-induced) and control cells (wild type cells) were seeded on Matrigel (100 µl) in 48 well plates at a density of 3.5 x 104 cells, and 200 µl of media was added [10]. Cells were incubated for 24 hours, and then observations made and pictures taken to report the presence or absence of networks [10]. Analysis of slides One-way ANOVA was used to assess significance over the 12-day time course, two-way ANOVA was used to compare between groups, and Bonferroni post hoc test was performed as needed. Significant differences over time (^) and due to doxycycline treatment () are shown (P < 0.05). Data represent averages + SEM from 3-5 experiments per group, each performed in triplicate. doi: 10.1371/journal.pone.0075984.g003 Figure 2. LDH-A is required for PMVECs to sustain aerobic glycolysis and rapid growth. Western blot analysis demonstrates that LDH-A protein decreases in the absence of doxycycline over a 7-day time course (A) resulting in decreased lactate production (B), glucose consumption (C) and PMVEC growth (D). Panel A shows a representative western blot taken from three different experiments performed in parallel with the growth curve. Two-way ANOVA was used to compare between groups, and Bonferroni post hoc test was performed as needed. Significant differences over time (^) and between doxycycline treatments () are shown (P < 0.05). Data represent averages + SEM from 3-5 experiments per group, each performed in triplicate. doi: 10.1371/journal.pone.0075984.g002 Figure 3. LDH-A rescue promotes PMVEC growth. Cells were grown in the absence of doxycycline for five days, and then on day six, doxycycline treatment was initiated. Doxycycline treatment rescued LDH-A protein (A), lactate production (B), glucose consumption (C), and PMVEC growth (D). One-way ANOVA was used to assess significance over the 12-day time course, two-way ANOVA was used to compare between groups, and Bonferroni post hoc test was performed as needed. Significant differences over time (^) and due to doxycycline treatment () are shown (P < 0.05). Data represent averages + SEM from 3-5 experiments per group, each performed in triplicate. doi: 10.1371/journal.pone.0075984.g003 grew from 1 x 105 cells to 5 x 106 cells over a seven-day time course, and networks started to form on Matrigel as early as 2 hours after cell seeding; 24 hours later networks were completely formed (Figure 4, LDH-A normal levels). During the growth curve, PMVECs consumed glucose, generated lactate and decreased media pH (data not shown). In contrast, glucose consumption, lactate production and PMVEC growth were all attenuated following suppression of LDH-A expression in both systems. Most of the cells lacking LDH-A were not able to attach on Matrigel, and after 24 hours just irregular networks were formed (Figure 4, low LDH-A levels). Analysis of slides Hematoxylin and eosin stained slides were observed under the light microscope and pictures were taken. Sections were extensively examined to count the number of vessels, defined by lumen-containing red blood cells [10]. Surface area was calculated by taking pictures of the plugs at 2x magnification, in many cases more than two pictures were required to cover all of the surface area. The pictures were then used to draw the area of the plug to be analyzed with the imageJ software (NIH). Measured values were standardized to mm2 based on the value of a known area. The results were then reported as number of formed blood vessels per mm2. We conditionally expressed LDH-A shRNA using a Tet-Off approach to confirm our previous results [13], that LDH-A is essential for PMVEC to sustain their glycolytic phenotype necessary for rapid proliferation, and to exclude an independent effect of doxycycline on cell growth. Tet-Off engineered cells were grown and expanded in the presence of doxycycline (2 µg/ml). Growth curves were performed with both induced (no doxycycline) and uninduced (doxycycline) cells. In induced cells, LDH-A levels progressively decreased throughout the growth curve. However, six days were required for LDH-A levels to dramatically decrease after doxycycline PLOS ONE \| www.plosone.org September 2013 \| Volume 8 \| Issue 9 \| e75984 3 LDH-A in Pulmonary Endothelial Angiogenesis Figure 2. LDH-A is required for PMVECs to sustain aerobic glycolysis and rapid growth. Western blot analysis demonstrates that LDH-A protein decreases in the absence of doxycycline over a 7-day time course (A) resulting in decreased lactate production (B), glucose consumption (C) and PMVEC growth (D). Panel A shows a representative western blot taken from three different experiments performed in parallel with the growth curve. Two-way ANOVA was used to compare between groups, and Bonferroni post hoc test was performed as needed. Significant differences over time (^) and between doxycycline treatments () are shown (P < 0.05). Data represent averages + SEM from 3-5 experiments per group, each performed in triplicate. doi: 10.1371/journal.pone.0075984.g002 Figure 3. LDH-A rescue promotes PMVEC growth. Cells were grown in the absence of doxycycline for five days, and then on day six, doxycycline treatment was initiated. Doxycycline treatment rescued LDH-A protein (A), lactate production (B), glucose consumption (C), and PMVEC growth (D). Analysis of slides These results demonstrate that LDH-A is required for network formation on Matrigel in vitro, an approach that has been used as an indication of angiogenic potential. retrieval (Figure 2A, upper). In contrast, in un-induced cells LDH-A levels remained relatively constant for the entire growth period (Figure 2A, bottom). Lactate levels were decreased to half of that seen in uninduced cells (30 vs. 15 mM) (Figure 2B), as was glucose consumption (Figure 2C). Cell proliferation was inhibited in the induced cells (low LDH-A) (Figure 2D), confirming our previous results using the Tet-On system [13]. One advantage of the conditional expression system is that rescue experiments can be performed over time in the same cell population. Rescue experiments demonstrated that addition of doxycycline restored LDH-A expression (Figure 3A), lactate production (Figure 3B), glucose consumption (Figure 3C) and PMVEC proliferation (Figure 3D). These findings indicate that LDH-A expression is essential to the proliferative phenotype of PMVECs. LDH-A expression is necessary for network formation on Matrigel in vitro One-way ANOVA and Bonferroni post hoc test was performed to assess significance of number of vessels formed by the different cell lines. Significantly different p<0.05. Micrographs were taken at 20x. Arrows indicate vessels containing red blood cells. doi: 10.1371/journal.pone.0075984.g005 Figure 4. LDH-A is essential for network formation on Matrigel in vitro. 3x104 cells from Tet-On and Tet-Off systems were seeded on Matrigel in 48 wells plate and kept at 37°C overnight. Cells expressing high LDH-A levels from both systems started to form networks as early as two hours after seeding, and complete networks were observed after 24 hours. Cells with low LDH-A levels failed to form networks. doi: 10.1371/journal.pone.0075984.g004 Figure 4. LDH-A is essential for network formation on Matrigel in vitro. 3x104 cells from Tet-On and Tet-Off systems were seeded on Matrigel in 48 wells plate and kept at 37°C overnight. Cells expressing high LDH-A levels from both systems started to form networks as early as two hours after seeding, and complete networks were observed after 24 hours. Cells with low LDH-A levels failed to form networks. doi: 10.1371/journal.pone.0075984.g004 Figure 5. LDH-A expression promotes angiogenesis in vivo. Tet-On PMVECs expressing normal LDH-A levels or Tet- Off PMVECs expressing low LDH-A levels were mixed with Matrigel, and then injected in the abdomen of rats. Matrigel plugs were removed after 10 days and paraffin-embebeded slides were made and stained with hematoxylin and eosin for vessel counting. Representative micrographs show high density of vessel formation in Matrigel containing cells expressing normal LDH-A levels (A), and low density of vessel formation in Matrigel containing cells expressing low LDH-A levels (B). Arrows show blood vessels that formed within the Matrigel plug. (C) Vessel density assessment demonstrated that PMVECs expressing normal LDH-A levels (Tet-On, Tet-Off control and Tet-On control) form a significantly greater number of vessels than do PMVECs expressing low LDH-A levels (Tet- Off). One-way ANOVA and Bonferroni post hoc test was performed to assess significance of number of vessels formed by the different cell lines. Significantly different p<0.05. Micrographs were taken at 20x. Arrows indicate vessels containing red blood cells. doi: 10.1371/journal.pone.0075984.g005 Figure 5. LDH-A expression promotes angiogenesis in vivo. Tet-On PMVECs expressing normal LDH-A levels or Tet- Off PMVECs expressing low LDH-A levels were mixed with Matrigel, and then injected in the abdomen of rats. LDH-A expression is necessary for network formation on Matrigel in vitro The development of Tet-On and Tet-Off systems for the inducible expression of shRNA allows us to regulate LDH-A expression in vivo without the need to feed animals with doxycycline. We used Tet-On PMVECs to maintain normal LDH-A levels, and Tet-Off PMVECs to inhibit LDH-A expression; in this manner we were able to control LDH-A shRNA expression. In the first case, Tet-On PMVECs contain the shRNA, but the shRNA is not expressed if cells are not exposed to doxycycline (inducer). In contrast, Tet-Off PMVECs express the shRNA after doxycycline retrieval. We also tested a cell line engineered with the Tet-On system lacking LDH-A shRNA, and a second cell line engineered with the Tet-Off system lacking LDH-A shRNA. These two cell lines served as controls since they each express normal LDH-A levels. Wild type PMVECs were also used as controls. Cells were mixed with September 2013 \| Volume 8 \| Issue 9 \| e75984 To address the angiogenic potential of PMVECs in vitro, with normal and decreased LDH-A levels, a network formation assay was performed. Both the Tet-On and Tet-Off systems were tested. Tet-On cells grown in the absence of doxycycline and Tet-Off cells grown in the presence of doxycycline were considered to be controls, as both of these cell types express normal LDH-A levels. Tet-On cells grown in the presence of doxycycline and Tet-Off cells grown in the absence of doxycycline both express the LDH-A shRNA, and therefore possess low LDH-A levels. In control experiments, PMVECs PLOS ONE \| www.plosone.org 4 LDH-A in Pulmonary Endothelial Angiogenesis Figure 5. LDH-A expression promotes angiogenesis in vivo. Tet-On PMVECs expressing normal LDH-A levels or Tet- Off PMVECs expressing low LDH-A levels were mixed with Matrigel, and then injected in the abdomen of rats. Matrigel plugs were removed after 10 days and paraffin-embebeded slides were made and stained with hematoxylin and eosin for vessel counting. Representative micrographs show high density of vessel formation in Matrigel containing cells expressing normal LDH-A levels (A), and low density of vessel formation in Matrigel containing cells expressing low LDH-A levels (B). Arrows show blood vessels that formed within the Matrigel plug. (C) Vessel density assessment demonstrated that PMVECs expressing normal LDH-A levels (Tet-On, Tet-Off control and Tet-On control) form a significantly greater number of vessels than do PMVECs expressing low LDH-A levels (Tet- Off). LDH-A expression is necessary for network formation on Matrigel in vitro These findings suggest it is the replication competent cells that can be responsive to vascular injury in order to repopulate or repair the endothelium, and further, that the microcirculation is highly adapted to respond to injurious stimuli. Indeed, gene profiling of freshly isolated adult pulmonary capillary endothelium show expression of multiple genes characteristic of embryogenesis and vascular development [17]. Such cell growth may be necessary to generate new cells that line an existing blood vessel, may be necessary to increase the length of an existing vessel, or may be needed to generate de novo vessels within the postnatal circulation. Immunohistochemical analysis was performed using GFP antibody, which revealed that cells lining the formed vessels were indeed the injected PMVECs that contained the GFP sequence (Figure 6). These results confirmed the in vitro observations shown in Figure 4, where Tet-Off PMVECs were not able to form networks on Matrigel. Taken together, these in vitro and in vivo experiments indicate that LDH-A expression is essential to the neo-angiogenic capacity of PMVECs. LDH-A expression is necessary for network formation on Matrigel in vitro Matrigel plugs were removed after 10 days and paraffin-embebeded slides were made and stained with hematoxylin and eosin for vessel counting. Representative micrographs show high density of vessel formation in Matrigel containing cells expressing normal LDH-A levels (A), and low density of vessel formation in Matrigel containing cells expressing low LDH-A levels (B). Arrows show blood vessels that formed within the Matrigel plug. (C) Vessel density assessment demonstrated that PMVECs expressing normal LDH-A levels (Tet-On, Tet-Off control and Tet-On control) form a significantly greater number of vessels than do PMVECs expressing low LDH-A levels (Tet- Off). One-way ANOVA and Bonferroni post hoc test was performed to assess significance of number of vessels formed by the different cell lines. *Significantly different p<0.05. Micrographs were taken at 20x. Arrows indicate vessels containing red blood cells. doi: 10.1371/journal.pone.0075984.g005 shRNA, and a second cell line engineered with the Tet-Off system lacking LDH-A shRNA. These two cell lines served as controls since they each express normal LDH-A levels. Wild type PMVECs were also used as controls. Cells were mixed with Matrigel, and subcutaneously injected into the abdomen of the rats. After 10 days, we quantified the number of blood vessels in excised plugs that were formed by PMVECs expressing normal (Figure 5A) and low LDH-A levels (Figure 5B), as well as in control cells. Every lumen containing red blood cells was considered a vessel. We found that decreasing LDH-A expression in PMVECs decreased their angiogenic capacity by 3.2-fold when compared to PMVECs expressing normal LDH-A levels (Tet-On PMVECs) and control cell lines (Figure 5C). The majority of vessels formed by PMVECs expressing normal LDH-A levels were capillary-like size, although bigger vessels were infrequently observed. We also noted some areas of hemorrhagic-like events in the plugs containing Tet-Off PMVECs (data not shown), which supports the idea of a failed attempt to generate a neo-angiogenic response in cells with low LDH-A protein levels. do pulmonary artery endothelial cells [3,10,11,12]. Single cell cloning of these endothelial cell populations reveal progeny with high, intermediate and low proliferative potentials [10]. PMVEC populations can be comprised of 30-50% highly proliferative cells, whereas pulmonary artery endothelial cell populations are comprised of less than 10% of the highly proliferative cells. LDH-A in Pulmonary Endothelial Angiogenesis The presence of GFP was detected using anti-GFP antibody followed by biotinylated horse anti-rabbit secondary antibody, visualized using the DAB substrate system. Arrows highlight positive endothelial cell staining in intact vessels. (B) Micrographs depicting negative staining in rat sections using an isotype control antibody. Arrows indicate endothelial cells that do not display staining using the control antibody. All micrographs were taken using a Nikon 80i upright microscope fitted with a 60X objective. doi: 10 1371/journal pone 0075984 g006 Figure 6. Cells injected in Matrigel are responsible for formation of new blood vessels. (A) Representative micrographs of GFP-positive endothelial cells. The presence of GFP was detected using anti-GFP antibody followed by biotinylated horse anti-rabbit secondary antibody, visualized using the DAB substrate system. Arrows highlight positive endothelial cell staining in intact vessels. (B) Micrographs depicting negative staining in rat sections using an isotype control antibody. Arrows indicate endothelial cells that do not display staining using the control antibody. All micrographs were taken using a Nikon 80i upright microscope fitted with a 60X objective. doi: 10 1371/journal pone 0075984 g006 doi: 10.1371/journal.pone.0075984.g006 Angiogenesis and vasculogenesis are fundamental features of all endothelia. The ability to generate new blood vessels requires that existing endothelial cells proliferate as they build vascular tubes and establish connections with existing vessels conducting blood. Rapidly growing endothelial cells generate more blood vessels in angiogenesis assays. PMVECs are highly neo-angiogenic, generating 2-3 fold more blood vessels in a shorter time period when compared to pulmonary artery endothelial cells [10]. These findings suggest that the microcirculation is adapted to generate new blood vessels following vascular injury. While our studies demonstrate that PMVECs display the intrinsic ability to rapidly generate new blood vessels, neo- angiogenesis of the pulmonary endothelium in the adult circulation in vivo has not been clearly demonstrated and remains controversial. This controversy may be partly due to the lung’s extensive vascular network, and the unique anatomy of the lung’s capillaries, which enable recruitment and distention and make it difficult to quantify neo-angiogenesis. These technical difficulties are compounded by the need to systematically separate angiogenesis of the lung’s systemic and pulmonary circulations. There is clear evidence that in airway inflammation [21,22,23] and pulmonary hypertension [24,25,26,27] the lung’s systemic blood supply undergoes a reversible angiogenic response. Further, following pulmonary artery ligation and lung ischemia the intercostal circulation invades the lung and forms an anastomosis with the pulmonary circulation [28]. Discussion Proliferation is a fundamental property of living cells, although not all cells are similarly replication competent [15,16]. Endothelial cells isolated from the adult pulmonary circulation display variable growth potentials; PMVECs grow faster than September 2013 \| Volume 8 \| Issue 9 \| e75984 PLOS ONE \| www.plosone.org 5 LDH-A in Pulmonary Endothelial Angiogenesis LDH-A in Pulmonary Endothelial Angiogenesis LDH-A in Pulmonary Endothelial Angiogenesis In the present study, we sought to determine whether LDH-A was necessary for the neo-angiogenic capacity of PMVECs. To test this idea in both in vitro and in vivo angiogenesis assays, we developed a novel experimental system in which LDH-A could be conditionally silenced. PMVECs were generated for conditional expression of LDH-A shRNA using both Tet-On and Tet-Off approaches. Irrespective of whether Tet-On or Tet-Off approaches were used, expression of the LDH-A shRNA reduced LDH-A protein and decreased vascular tube formation in vitro. Using Tet-Off cell lines as a system in which the angiogenic potential of PMVECs lacking LDH-A could be tested in in vivo assays, we observed that LDH-A silencing similarly decreased blood vessel formation. We have interpretated these results to mean that aerobic glycolysis provided the ATP needed for neo-angiogenesis. However, we have not presently ruled out other possible contributing mechanisms. Lactate itself may be an important intracellular signal. LDH is directly involved in the redox status of the cell [20], and in gene regulation responsible for the metabolic switch that is necessary to accomplish demanding processes, such as neo- angiogenesis. LDH may form a protein complex consisting of the transcription factor oct-1 and its co-activator OCA-S, which is required for transcription of histone 2B (H2B). GAPDH and LDH are part of OCA-S, and even the enzymatic activity of LDH is necessary for the regulation of transcription of H2B [20]. Irrespective of the downstream mechanism, our present findings conclusively demonstrate the importance of LDH-A in establishing the bioenergetic demands for rapid endothelial cell neo-angiogenesis. Figure 6. Cells injected in Matrigel are responsible for formation of new blood vessels. (A) Representative micrographs of GFP-positive endothelial cells. The presence of GFP was detected using anti-GFP antibody followed by biotinylated horse anti-rabbit secondary antibody, visualized using the DAB substrate system. Arrows highlight positive endothelial cell staining in intact vessels. (B) Micrographs depicting negative staining in rat sections using an isotype control antibody. Arrows indicate endothelial cells that do not display staining using the control antibody. All micrographs were taken using a Nikon 80i upright microscope fitted with a 60X objective. doi: 10 1371/journal pone 0075984 g006 Figure 6. Cells injected in Matrigel are responsible for formation of new blood vessels. (A) Representative micrographs of GFP-positive endothelial cells. References Reynolds LP, Redmer DA (1998) Expression of the angiogenic factors, basic fibroblast growth factor and vascular endothelial growth factor, in the ovary. J Anim Sci 76: 1671-1681. PubMed: 9655588. 21. Ezaki T, Baluk P, Thurston G, La Barbara A, Woo C et al. (2001) Time course of endothelial cell proliferation and microvascular remodeling in chronic inflammation. Am J Pathol 158: 2043-2055. doi:10.1016/ S0002-9440(10)64676-7. PubMed: 11395382. 9. Sypniewska G, Björntorp P (1987) Increased DNA synthesis in adipocytes and capillary endothelium in rat adipose tissue during overfeeding. Eur J Clin Invest 17: 202-207. doi:10.1111/j. 1365-2362.1987.tb01236.x. PubMed: 3113964. 22. Kwan ML, Gómez AD, Baluk P, Hashizume H, McDonald DM (2001) Airway vasculature after mycoplasma infection: chronic leakiness and selective hypersensitivity to substance P. Am J Physiol Lung Cell Mol Physiol 280: L286-L297. PubMed: 11159008. 10. Alvarez DF, Huang L, King JA, ElZarrad MK, Yoder MC et al. (2008) Lung microvascular endothelium is enriched with progenitor cells that exhibit vasculogenic capacity. Am J Physiol Lung Cell Mol Physiol 294: L419-L430. PubMed: 18065657. y 23. McDonald DM (2001) Angiogenesis and remodeling of airway vasculature in chronic inflammation. Am J Respir Crit Care Med 164: S39-S45. doi:10.1164/ajrccm.164.supplement_2.2106065. PubMed: 11734465. 11. Clark J, Alvarez DF, Alexeyev M, King JA, Huang L et al. (2008) Regulatory role for nucleosome assembly protein-1 in the proliferative and vasculogenic phenotype of pulmonary endothelium. Am J Physiol Lung Cell Mol Physiol 294: L431-L439. PubMed: 17981956. 24. Davie NJ, Crossno JT Jr., Frid MG, Hofmeister SE, Reeves JT et al. (2004) Hypoxia-induced pulmonary artery adventitial remodeling and neovascularization: contribution of progenitor cells. Am J Physiol Lung Cell Mol Physiol 286: L668-L678. PubMed: 12754186. 12. Solodushko V, Fouty B (2007) Proproliferative phenotype of pulmonary microvascular endothelial cells. Am J Physiol Lung Cell Mol Physiol 292: L671-L677. PubMed: 17085521. 25. Davie NJ, Gerasimovskaya EV, Hofmeister SE, Richman AP, Jones PL et al. (2006) Pulmonary artery adventitial fibroblasts cooperate with vasa vasorum endothelial cells to regulate vasa vasorum neovascularization: a process mediated by hypoxia and endothelin-1. Am J Pathol 168: 1793-1807. doi:10.2353/ajpath.2006.050754. PubMed: 16723696. 13. Parra-Bonilla G, Alvarez DF, Al-Mehdi AB, Alexeyev M, Stevens T (2010) Critical role for lactate dehydrogenase A in aerobic glycolysis that sustains pulmonary microvascular endothelial cell proliferation. Am J Physiol Lung Cell Mol Physiol 299: L513-L522. doi:10.1152/ajplung. 00274.2009. PubMed: 20675437. 26. Zhang J, Luo B, Tang L, Wang Y, Stockard CR et al. References 1. Hobson B, Denekamp J (1984) Endothelial proliferation in tumours and normal tissues: continuous labelling studies. Br J Cancer 49: 405-413. doi:10.1038/bjc.1984.66. PubMed: 6201181. 1. Hobson B, Denekamp J (1984) Endothelial proliferation in tumours and normal tissues: continuous labelling studies. Br J Cancer 49: 405-413. doi:10.1038/bjc.1984.66. PubMed: 6201181. contain a complete hierarchy of endothelial progenitor cells. Blood 105: 2783-2786. doi:10.1182/blood-2004-08-3057. PubMed: 15585655. contain a complete hierarchy of endothelial progenitor cells. Blood 105: 2783-2786. doi:10.1182/blood-2004-08-3057. PubMed: 15585655. 16. Ingram DA, Mead LE, Tanaka H, Meade V, Fenoglio A et al. (2004) Identification of a novel hierarchy of endothelial progenitor cells using human peripheral and umbilical cord blood. Blood 104: 2752-2760. doi: 10.1182/blood-2004-04-1396. PubMed: 15226175. j 2. Voelkel NF, Douglas IS, Nicolls M (2007) Angiogenesis in chronic lung disease. Chest 131: 874-879. doi:10.1378/chest.06-2453. PubMed: 17356107. j 2. Voelkel NF, Douglas IS, Nicolls M (2007) Angiogenesis in chronic lung disease. Chest 131: 874-879. doi:10.1378/chest.06-2453. PubMed: 17356107. 17. Favre CJ, Mancuso M, Maas K, McLean JW, Baluk P et al. (2003) Expression of genes involved in vascular development and angiogenesis in endothelial cells of adult lung. Am J Physiol Heart Circ Physiol 285: H1917-H1938. PubMed: 12842817. 3. King J, Hamil T, Creighton J, Wu S, Bhat P et al. (2004) Structural and functional characteristics of lung macro- and microvascular endothelial cell phenotypes. Microvasc Res 67: 139-151. doi:10.1016/j.mvr. 2003.11.006. PubMed: 15020205. 4. Stevens T (2005) Molecular and cellular determinants of lung endothelial cell heterogeneity. Chest 128: 558S-564S. doi:10.1378/ chest.128.6_suppl.558S. PubMed: 16373825. 18. Blouin A, Bolender RP, Weibel ER (1977) Distribution of organelles and membranes between hepatocytes and nonhepatocytes in the rat liver parenchyma. A stereological study. J Cell Biol 72: 441-455. doi: 10.1083/jcb.72.2.441. PubMed: 833203. 5. Ware LB, Matthay MA (2000) The acute respiratory distress syndrome. N Engl J Med 342: 1334-1349. doi:10.1056/NEJM200005043421806. PubMed: 10793167. j 19. Vander Heiden MG, Cantley LC, Thompson CB (2009) Understanding the Warburg effect: the metabolic requirements of cell proliferation. Science 324: 1029-1033. doi:10.1126/science.1160809. PubMed: 19460998. 6. Gargett CE, Rogers PA (2001) Human endometrial angiogenesis. Reproduction 121: 181-186. doi:10.1530/rep.0.1210181. PubMed: 11226042. 20. Dai RP, Yu FX, Goh SR, Chng HW, Tan YL et al. (2008) Histone 2B (H2B) expression is confined to a proper NAD+/NADH redox status. J Biol Chem 283: 26894-26901. doi:10.1074/jbc.M804307200. PubMed: 18682386. 7. Hudlicka O, Brown M, Egginton S (1992) Angiogenesis in skeletal and cardiac muscle. Physiol Rev 72: 369-417. PubMed: 1372998. 8. LDH-A in Pulmonary Endothelial Angiogenesis understanding of the bioenergetic demands of angiogenesis in order to intercede in vascular injury and promote repair. understanding of the bioenergetic demands of angiogenesis in order to intercede in vascular injury and promote repair. Acknowledgements The authors wish to thank the Center for Lung Biology Cell Culture and Small Animal Core and the Gene Delivery Core for assistance in endothelial cell isolation, culture and use and development of the Tet-On and Tet-Off constructs, respectively. LDH-A in Pulmonary Endothelial Angiogenesis In these instances, there is no evidence for angiogenesis of the pulmonary endothelium. However, in response to other, varied stimuli evidence has accumulated to suggest that pulmonary capillary endothelium undergoes angiogenesis, including: hepatopulmonary syndrome [26,29], post-transplant obliterative airway disease [30], pulmonary hypertension [31,32], Pseudomonas aeruginosa airway infection [33], pneumonectomy [34], cavopulmonary anastomosis [35], estrogen supplementation in ovarectomized rats [36], the early luteal phase of the menstrual cycle [36], and following L-arginine supplementation in pulmonary hypertension [37]. Provocatively, intra-tracheal delivery of hepatocyte growth factor [38,39], fibroblast growth factor-2 [40] Rapid proliferation and neo-angiogenesis exert bioenergetic demands not typical for quiescent cells. Endothelial cells possess a relatively low number of mitochondria, and are not thought to possess high aerobic demands [3,18]. In response to growth-provoking stimuli, PMVECs utilize aerobic glycolysis to sustain the bioenergetic needs of rapid proliferation [13]. Aerobic glycolysis is defined by increased glycolytic flux with sufficient oxygen delivery to mitochondria, and is characterized by glucose consumption and production of lactate [19]. Glut-1 surface expression is necessary to transport glucose into the cell, hexokinase phosphorylates glucose in order to retain glucose in the cytoplasm, and LDH converts pyruvate into lactate regenerating NAD+ necessary to sustain glycolytic flux. Expression of each of these three enzymes is higher in PMVECs than in pulmonary artery endothelial cells, enabling PMVECs to generate 2-3 times higher ATP concentrations during growth [13]. LDH is a tetramer comprised of LDH-A and LDH-B proteins, which organize into one of five different combinations. The presence of LDH-A in the tetramer increases enzymatic activity, and is essential for PMVECs to display their rapid neo-angiogenic phenotype. Indeed, in the absence of LDH-A, PMVECs do not rapidly consume glucose, do not generate as much lactate, and do not rapidly proliferate relative to pulmonary artery endothelial cells. September 2013 \| Volume 8 \| Issue 9 \| e75984 PLOS ONE \| www.plosone.org 6 LDH-A in Pulmonary Endothelial Angiogenesis and vascular endothelial cell growth factor [41] may induce angiogenesis and promote alveolar growth in lungs showing emphysematous changes. These reported angiogenic responses are beneficial to the organ, yet there are also examples of disordered angiogenesis in the setting of pulmonary hypertension, both in plexiform lesions of small precapillary vessels [42,43,44] and in pulmonary microvasculopathy of the capillaries [45]. The array of stimuli that appear to induce angiogenesis of pulmonary capillaries suggest this process is critically important in the adult circulation. Author Contributions Angiogenesis is a fundamental property of endothelium, including in the lung, and especially lung microvascular endothelium. Expression of LDH-A is essential for sustaining aerobic glycolysis necessary to provide the bioenergetic demands of proliferation and angiogenesis. With increasing evidence that angiogenesis is an important biological phenomenon of the adult circulation, we must expand our Conceived and designed the experiments: GPB DFA TS. Performed the experiments: GPB MA AV. Analyzed the data: GPB DFA AV TS. Contributed reagents/materials/analysis tools: DFA MA TS. Wrote the manuscript: GPB TS. LDH-A in Pulmonary Endothelial Angiogenesis idiopathic pulmonary arterial hypertension. Am J Respir Crit Care Med 184: 116-123. doi:10.1164/rccm.201006-0905OC. PubMed: 21471108. mechanism ameliorating pulmonary hypertension. Am J Physiol Lung Cell Mol Physiol 296: L1042-L1050. doi:10.1152/ajplung.90327.2008. PubMed: 19346433. 28. Mitzner W, Lee W, Georgakopoulos D, Wagner E (2000) Angiogenesis in the mouse lung. Am J Pathol 157: 93-101. doi:10.1016/ S0002-9440(10)64521-X. PubMed: 10880380. 38. Ono M, Sawa Y, Matsumoto K, Nakamura T, Kaneda Y et al. (2002) In vivo gene transfection with hepatocyte growth factor via the pulmonary artery induces angiogenesis in the rat lung. Circulation 106: I264-I269. PubMed: 12354744. ( ) 29. Schraufnagel DE, Malik R, Goel V, Ohara N, Chang SW (1997) Lung capillary changes in hepatic cirrhosis in rats. Am J Physiol 272: L139- L147. PubMed: 9038913. 39. Shigemura N, Sawa Y, Mizuno S, Ono M, Ohta M et al. (2005) Amelioration of pulmonary emphysema by in vivo gene transfection with hepatocyte growth factor in rats. Circulation 111: 1407-1414. doi: 10.1161/01.CIR.0000158433.89103.85. PubMed: 15781752. 30. Dutly AE, Andrade CF, Verkaik R, Kugathasan L, Trogadis J et al. (2005) A novel model for post-transplant obliterative airway disease reveals angiogenesis from the pulmonary circulation. Am J Transplant 5: 248-254. doi:10.1111/j.1600-6143.2004.00680.x. PubMed: 15643984. y g 10.1161/01.CIR.0000158433.89103.85. PubMed: 15781752. 40. Morino S, Toba T, Tao H, Araki M, Shimizu Y et al. (2007) Fibroblast growth factor-2 promotes recovery of pulmonary function in a canine models of elastase-induced emphysema. Exp Lung Res 33: 15-26. doi: 10.1080/01902140601113070. PubMed: 17364909. 31. Hopkins N, McLoughlin P (2002) The structural basis of pulmonary hypertension in chronic lung disease: remodelling, rarefaction or angiogenesis? J Anat 201: 335-348. doi:10.1046/j. 1469-7580.2002.00096.x. PubMed: 12430958. 41. Thébaud B, Ladha F, Michelakis ED, Sawicka M, Thurston G et al. (2005) Vascular endothelial growth factor gene therapy increases survival, promotes lung angiogenesis, and prevents alveolar damage in hyperoxia-induced lung injury: evidence that angiogenesis participates in alveolarization. Circulation 112: 2477-2486. doi:10.1161/ CIRCULATIONAHA.105.541524. PubMed: 16230500. 32. Howell K, Preston RJ, McLoughlin P (2003) Chronic hypoxia causes angiogenesis in addition to remodelling in the adult rat pulmonary circulation. J Physiol 547: 133-145. doi:10.1113/jphysiol.2002.030676. PubMed: 12562951. 33. Hopkins N, Cadogan E, Giles S, McLoughlin P (2001) Chronic airway infection leads to angiogenesis in the pulmonary circulation. J Appl Physiol 91: 919-928. PubMed: 11457811. 42. Abe K, Toba M, Alzoubi A, Ito M, Fagan KA et al. (2010) Formation of plexiform lesions in experimental severe pulmonary arterial hypertension. Circulation 121: 2747-2754. doi:10.1161/ CIRCULATIONAHA.109.927681. PubMed: 20547927. 34. References (2009) Pulmonary angiogenesis in a rat model of hepatopulmonary syndrome. Gastroenterology 136: 1070-1080. doi:10.1053/j.gastro.2008.12.001. PubMed: 19109954. 14. Alexeyev MF, Fayzulin R, Shokolenko IN, Pastukh V (2010) A retro- lentiviral system for doxycycline-inducible gene expression and gene knockdown in cells with limited proliferative capacity. Mol Biol Rep 37: 1987-1991. doi:10.1007/s11033-009-9647-7. PubMed: 19655272. 27. Montani D, Perros F, Gambaryan N, Girerd B, Dorfmuller P et al. (2011) C-kit-positive cells accumulate in remodeled vessels of 15. Ingram DA, Mead LE, Moore DB, Woodard W, Fenoglio A et al. (2005) Vessel wall-derived endothelial cells rapidly proliferate because they 7 PLOS ONE \| www.plosone.org September 2013 \| Volume 8 \| Issue 9 \| e75984 LDH-A in Pulmonary Endothelial Angiogenesis LDH-A in Pulmonary Endothelial Angiogenesis LDH-A in Pulmonary Endothelial Angiogenesis Foster DJ, Moe OW, Hsia CC (2004) Upregulation of erythropoietin receptor during postnatal and postpneumonectomy lung growth. Am J Physiol Lung Cell Mol Physiol 287: L1107-L1115. doi:10.1152/ajplung. 00119.2004. PubMed: 15286000. 43. Rai PR, Cool CD, King JA, Stevens T, Burns N et al. (2008) The cancer paradigm of severe pulmonary arterial hypertension. Am J Respir Crit Care Med 178: 558-564. doi:10.1164/rccm.200709-1369PP. PubMed: 18556624. 35. Starnes SL, Duncan BW, Kneebone JM, Fraga CH, States S et al. (2000) Pulmonary microvessel density is a marker of angiogenesis in children after cavopulmonary anastomosis. J Thorac Cardiovasc Surg 120: 902-907. doi:10.1067/mtc.2000.110248. PubMed: 11044316. 44. Tuder RM, Groves B, Badesch DB, Voelkel NF (1994) Exuberant endothelial cell growth and elements of inflammation are present in plexiform lesions of pulmonary hypertension. Am J Pathol 144: 275-285. PubMed: 7508683. 36. Farha S, Asosingh K, Laskowski D, Licina L, Sekiguchi H et al. (2007) Pulmonary gas transfer related to markers of angiogenesis during the menstrual cycle. J Appl Physiol 103: 1789-1795. doi:10.1152/ japplphysiol.00614.2007. PubMed: 17717117. 45. Pietra GG, Capron F, Stewart S, Leone O, Humbert M et al. (2004) Pathologic assessment of vasculopathies in pulmonary hypertension. J Am Coll Cardiol 43: 25S-32S. doi:10.1016/S0735-1097(04)90103-9. PubMed: 15194175. j pp p y 37. Howell K, Costello CM, Sands M, Dooley I, McLoughlin P (2009) L- Arginine promotes angiogenesis in the chronically hypoxic lung: a novel PLOS ONE \| www.plosone.org PLOS ONE \| www.plosone.org September 2013 \| Volume 8 \| Issue 9 \| e75984 8
https://openalex.org/W3197617774	https://europepmc.org/articles/pmc8450565?pdf=render	English	null	Biomarkers and Future Perspectives for Hepatocellular Carcinoma Immunotherapy	Frontiers in oncology	2,021	cc-by	13,985	REVIEW published: 06 September 2021 doi: 10.3389/fonc.2021.716844 Biomarkers and Future Perspectives for Hepatocellular Carcinoma Immunotherapy Yuqing He 1, Mengyao Lu 1, Jing Che 2, Qian Chu 1, Peng Zhang 1* and Yuan Chen 1 1 Department of Oncology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China, 2 College of Life Sciences, Wuhan University, Wuhan, China Yuqing He 1, Mengyao Lu 1, Jing Che 2, Qian Chu 1, Peng Zhang 1* and Yuan Chen 1 1 Department of Oncology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China, 2 College of Life Sciences, Wuhan University, Wuhan, China Hepatocellular cancer is the sixth most frequently diagnosed malignant disease worldwide, and was responsible for tens of millions of deaths in 2020; however, treatment options for patients with advanced hepatocellular carcinoma remain limited. Immunotherapy has undergone rapid development over recent years, especially in the ﬁeld of immune checkpoint inhibitors (ICIs). These drugs aim to activate and enhance antitumor immunity and represent a new prospect for the treatment of patients with advanced cancer. Nevertheless, only a small proportion of liver cancer patients currently beneﬁt from ICI-based treatment, highlighting the need to better understand how ICIs and tumors interact, as well as identify predictive biomarkers for immunotherapeutic responses. In this review, we highlight clinical trials and basic research in hepatocellular carcinoma, with a particular focus on predictive biomarkers for the therapeutic efﬁcacy of ICIs. Predictive biomarkers for immune-related adverse events are also discussed. Reviewed by: Keywords: hepatocellular carcinoma, biomarkers, immune checkpoint inhibitors, immune-related adverse events, immunotherapy Reviewed by: Alessandro Rizzo, Sant’Orsola-Malpighi Polyclinic, Italy Fengbiao Mao, University of Michigan, United States Correspondence: Peng Zhang pengzhang@tjh.tjmu.edu.cn Reviewed by: Alessandro Rizzo, Sant’Orsola-Malpighi Polyclinic, Italy Fengbiao Mao, University of Michigan, United States University of Michigan, United States Keywords: hepatocellular carcinoma, biomarkers, immune checkpoint inhibitors, immune-related adverse events, immunotherapy Correspondence: Peng Zhang Correspondence: Peng Zhang Correspondence: Peng Zhang pengzhang@tjh.tjmu.edu.cn Hepatocellular carcinoma (HCC) accounts for over 80% of all primary liver malignancies, while liver cancer ranked as the sixth most frequently diagnosed cancer in 2020, resulting in 83 million deaths (1). Despite these statistics, HCC treatment remains a major healthcare challenge globally. Additionally, because symptoms and physical characteristics of HCC are not easily detected, 80% of patients diagnosed with HCC miss out on curative treatment (2). Specialty section: This article was submitted to Cancer Genetics, a section of the journal Frontiers in Oncology Received: 29 May 2021 Accepted: 18 August 2021 Published: 06 September 2021 Citation: He Y, Lu M, Che J, Chu Q, Zhang P and Chen Y (2021) Biomarkers and Future Perspectives for Hepatocellular Carcinoma Immunotherapy. Front. Oncol. 11:716844. doi: 10.3389/fonc.2021.716844 Specialty section: This article was submitted to Cancer Genetics, a section of the journal Frontiers in Oncology Received: 29 May 2021 Accepted: 18 August 2021 Published: 06 September 2021 Specialty section: This article was submitted to Cancer Genetics, a section of the journal Frontiers in Oncology HCC typically develops in a background of underlying inﬂammatory liver disease, especially that associatedwithhepatitisB(HBV) orhepatitisCvirus(HCV) infection (3),while nonalcoholic fatty liver disease (NAFLD) israpidlybecoming a keyetiological factorfor HCC inmany Western countries (4, 5). Current treatment modalities for patients with nonadvanced HCC include resection, transplantation, ablation, or chemoembolization, while patients with advanced HCC receive systemic treatment (6). However, progress in the development of treatments for advanced HCC has been limited, partly due to the complex and heterogeneous etiology of this disease. Additionally, the most common driver mutations (TERT promoter, CTNNB1, TP53, and ARID1A mutations) have not yet been shown to be suitable therapeutic targets (7).Although ﬁrst-line multikinase inhibitors (sorafeniband lenvatinib) can prolong the survival of patients with advanced HCC (8–10), and multitarget tyrosine inhibitors (e.g., regorafenib and cabozantinib) and vascular endothelial growth factor (VEGF) receptor inhibitors Received: 29 May 2021 Accepted: 18 August 2021 Published: 06 September 2021 Edited by: Edited by: Heng-Hong Li, Georgetown University, United States Citation: Based on encouraging results from the CheckMate 040 and KEYNOTE-224 clinical trials, the United States Federal Drug Administration (FDA) has granted accelerated approval for the PD-1 inhibitors nivolumab and pembrolizumab as second-line treatments for advanced HCC. The CheckMate 040 phase I/II trial obtained objective response rates (ORRs) of 15% (dose- escalation phase) and 20% (dose-expansion phase) in patients treated with nivolumab (14). Meanwhile, the KEYNOTE-224 phase II trial reported an ORR of 17% for pembrolizumab monotherapy for HCC patients previously treated with sorafenib (15). Disappointingly, however, both the CheckMate 459 and KEYNOTE-240 phase III trials, which evaluated nivolumab versus sorafenib and pembrolizumab versus best supportive care, respectively, failed to meet their predetermined primary endpoints of OS (16, 17). Overall, ICI monotherapy has shown limited efﬁcacy in HCC, beneﬁting only a limited subgroup of patients. More results of clinical trials for ICIs in HCC are summarized in Table 1. Citation: The developmentofanti-PD-1, anti- programmed cell death-ligand 1 (PD-L1), and anti-CTLA-4 monoclonal antibodies has advanced the treatment for advanced cancer, resulting in numerous attempts to apply ICIs for the treatment ofmultipleadvancedsolidmalignancies, includingHCC. Based on encouraging results from the CheckMate 040 and KEYNOTE-224 clinical trials, the United States Federal Drug Administration (FDA) has granted accelerated approval for the PD-1 inhibitors nivolumab and pembrolizumab as second-line treatments for advanced HCC. The CheckMate 040 phase I/II trial obtained objective response rates (ORRs) of 15% (dose- escalation phase) and 20% (dose-expansion phase) in patients treated with nivolumab (14). Meanwhile, the KEYNOTE-224 phase II trial reported an ORR of 17% for pembrolizumab monotherapy for HCC patients previously treated with sorafenib (15). Disappointingly, however, both the CheckMate 459 and KEYNOTE-240 phase III trials, which evaluated nivolumab versus sorafenib and pembrolizumab versus best supportive care, respectively, failed to meet their predetermined primary endpoints of OS (16, 17). Overall, ICI monotherapy has shown limited efﬁcacy in HCC, beneﬁting only a limited subgroup of patients. More results of clinical trials for ICIs in HCC are summarized in Table 1. Cancer immunotherapy has undergone rapid development in recent years, especially in the ﬁeld of immune checkpoint inhibitors (ICI). Immune checkpoint-related molecules, such as programmed cell death-1 (PD-1), cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4), T-cell immunoglobulin, mucin domain-3 (TIM-3), and lymphocyteactivating-3(LAG-3),areimportantcomponentsofthe negative feedback regulatory mechanism that serves to suppress excessiveimmune responses. Theyare constitutivelyupregulatedin varioustumors,generatingT-cellexhaustionoranergy,andthereby helping tumors evade immune surveillance (13). The rationale behind utilizing ICIs is to restore and enhance antitumor immunity by relieving the immunosuppressive effects of immune checkpoint-related molecules. The developmentofanti-PD-1, anti- programmed cell death-ligand 1 (PD-L1), and anti-CTLA-4 monoclonal antibodies has advanced the treatment for advanced cancer, resulting in numerous attempts to apply ICIs for the treatment ofmultipleadvancedsolidmalignancies, includingHCC. Even though the prospect for ICIs seems to be excellent, numerous difﬁculties remain to be resolved. Pivotal among these is the low response rate in patients treated with ICI monotherapy, with treatment beneﬁting only 15%–20% of patients with advanced HCC (14, 15). The incidence of immune-related adverse events (irAEs) is another important concern. A better understanding of the dialog between ICIs and tumors is essential, as is the identiﬁcation of predictive biomarkers for treatment response and toxicity. In this review, we focus on clinical trials and basic research in HCC, with particular emphasis on predictive biomarkers for the therapeutic efﬁcacy of ICIs. Predictive biomarkers for irAEs are also discussed. Citation: He Y, Lu M, Che J, Chu Q, Zhang P and Chen Y (2021) Biomarkers and Future Perspectives for Hepatocellular Carcinoma Immunotherapy. Front. Oncol. 11:716844. doi: 10.3389/fonc.2021.716844 September 2021 \| Volume 11 \| Article 716844 Frontiers in Oncology \| www.frontiersin.org He et al. Biomarkers of ICIs in HCC He et al. In contrast, the results from the IMbrave150 phase III trial have been encouraging (22). This trial enrolled 501 patients with advanced HCC who had not previously received systemic treatment, two-thirds of whom received atezolizumab (anti- PD-L1 blockade) plus bevacizumab (VEGF monoclonal antibody), while the others received sorafenib monotherapy. The median progression-free survival (PFS) was 6.8 months in the combination arm and 4.3 months in the sorafenib group. In addition, OS at 12 months was 67.2% with atezolizumab + bevacizumab and 54.6% with sorafenib. The two subgroups displayed similar toxic effects, with an incidence of 56.5% for grade 3 or 4 adverse events in the combination arm and 55.1% in the sorafenib arm. Of note, these data are momentous, as they identify the ﬁrst therapy to improve OS and PFS beyond the standard of care sorafenib in treatment-naïve patients (23). After more than a decade of stagnation in the treatment of advanced HCC, these therapeutic strategies changed the status quo and have entered into clinical practice. The mechanisms underlying the effects of ICI/anti-VEGF-agent combination treatment have also been elucidated, and have been reviewed in detail (24, 25). (e.g.,ramucirumab) can provide beneﬁt for patientswhopreviously tolerated sorafenib (11), most cases of HCC show tolerance or become refractory to these agents during the clinical course of the disease (12). Accordingly, the median overall survival (OS) for patients treated with these agents remains under 15 months. patients treated with these agents remains under 15 months. Cancer immunotherapy has undergone rapid development in recent years, especially in the ﬁeld of immune checkpoint inhibitors (ICI). Immune checkpoint-related molecules, such as programmed cell death-1 (PD-1), cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4), T-cell immunoglobulin, mucin domain-3 (TIM-3), and lymphocyteactivating-3(LAG-3),areimportantcomponentsofthe negative feedback regulatory mechanism that serves to suppress excessiveimmune responses. Theyare constitutivelyupregulatedin varioustumors,generatingT-cellexhaustionoranergy,andthereby helping tumors evade immune surveillance (13). The rationale behind utilizing ICIs is to restore and enhance antitumor immunity by relieving the immunosuppressive effects of immune checkpoint-related molecules. POTENTIAL PREDICTIVE BIOMARKERS FOR ICI-BASED TREATMENT Because immunotherapy for HCC is still in its infancy, studies relating to predictive biomarkers for ICI treatment response are scarce. Nevertheless, several valuable data on potential TABLE 1 \| Clinical trials of immune-checkpoint inhibitors in hepatocellular carcinoma. Study Name PFS OS Estimated (Reference) Agent ORR (%) DCR (%) (median, months) (median, months) Phase Enrolment (n) CheckMate 040 (14) Nivolumab 19.6 64.5 4.0 15.0 I/II 214 CheckMate 040; Asian cohort analysis (18) Nivolumab 15.2 49.4 NA 14.9 I/II 85 NCT01693562 (19) Durvalumab 10.3 33.3 NA 13.2 I/II 39 NCT01008358 (20) Tremelimumab 17.6 76.4 6.5 8.2 II 20 NCT02658019 (21) Pembrolizumab 32.1 46.4 4.5 13.0 II 28 Keynote-224 (15) Pembrolizumab 17.3 61.5 4.9 12.9 II 104 CheckMate 459 (17) Nivolumab vs. sorafenib 15.4 vs. 7.0 NA 3.7 vs. 3.8 16.4 vs. 14.7 III 743 (371 vs. 372) Keynote-240 (16) Pembrolizumab vs. placebo 18.3 vs. 4.0 62.2 vs. 53.3 3.0 vs. 2.8 13.9 vs. 10.6 III 413 (278 vs. 135) ORR, overall response rate; DCR, disease control rate; OS, overall survival; PFS, progression-free survival; NA, not available. TABLE 1 \| Clinical trials of immune-checkpoint inhibitors in hepatocellular carcinoma. September 2021 \| Volume 11 \| Article 716844 Frontiers in Oncology \| www.frontiersin.org 2 Biomarkers of ICIs in HCC He et al. However, the predictive value of PD-L1 remains unclear. Disappointing response rates were consistent across all patients in both the CheckMate 040 and NCT02658019 trials, regardless of PD-L1 expression levels (14, 18, 21). Even so, further analysis of patients in the CheckMate 040 trial showed that tumor PD-L1 expression was associated with improved OS, although objective responses could still be observed in PD-L1-negative patients treated with nivolumab. These observations highlight that PD-L1 expression alone may not serve as an adequate biomarker for responses to ICI treatment (29). Although these trials provide valuable clinical data that allow the evaluation of the predictive effect of PD-L1 expression, they have their limitations. First, these are retrospective analyses with small sample sizes. Secondly, there are limitations associated with the detection of PD-L1 expression (discussed in detail later), as well as inconsistencies in sample sources; although they are all collected at baseline, some samples are fresh and some are archival. biomarkers have emerged in recent years, including genetic and protein markers, immune-related cells, and host-related factors (Figures 1, 2), which are described below. PD-L1 Expression Biomarkers of ICIs in HCC FIGURE 2 \| A schematic representation of biomarkers for predicting treatment responses to immune checkpoint inhibitors (ICIs) in vivo and their interactions with each other. Both the genetic characteristics and the expression of PD-L1 in tumor cells, as well as the density and diversity of tumor-inﬁltrating CD8+ T cells, have the potential to predict the efﬁcacy of ICI treatment in hepatocellular carcinoma (HCC). Other tumor-inﬁltrating lymphocytes in the tumor microenvironment affect T- cell activity by secreting cytokines (such as IFN-g and TGF-b), thereby affecting the efﬁcacy of ICI treatment. The gut microbiota also affects the responses to ICI treatment by inﬂuencing the activity of tumor-inﬁltrating lymphocytes (TILs), while circulating biomarkers in peripheral blood can also be used as predictors of immunotherapeutic efﬁcacy. FIGURE 2 \| A schematic representation of biomarkers for predicting treatment responses to immune checkpoint inhibitors (ICIs) in vivo and their interactions with each other. Both the genetic characteristics and the expression of PD-L1 in tumor cells, as well as the density and diversity of tumor-inﬁltrating CD8+ T cells, have the potential to predict the efﬁcacy of ICI treatment in hepatocellular carcinoma (HCC). Other tumor-inﬁltrating lymphocytes in the tumor microenvironment affect T- cell activity by secreting cytokines (such as IFN-g and TGF-b), thereby affecting the efﬁcacy of ICI treatment. The gut microbiota also affects the responses to ICI treatment by inﬂuencing the activity of tumor-inﬁltrating lymphocytes (TILs), while circulating biomarkers in peripheral blood can also be used as predictors of immunotherapeutic efﬁcacy. PD-L1-positive tumor cells—the so-called tumor proportion score (TPS)—and the ratio of PD-L1-stained tumor and immune cells, the so-called combined positive score (CPS). The KEYNOTE-224 phase II trial evaluated PD-L1 using both scoring methods, with the CPS turning out to be a more applicable biomarker (15). In addition, the predictive value of PD-L1 expression may be underestimated as it is commonly evaluated at a single time point, even though PD-L1 expression is dynamic and inducible (32, 33). PD-L1 is heavily glycosylated and such modiﬁcations signiﬁcantly affect the detection performance and therapeutic efﬁcacy of PD-L1 antibodies (34, 35). A method was developed to resolve this issue that involved removing the glycan moieties from cell-surface antigens via enzymatic digestion, which boosted the positivity rate of PD- L1 detection in tumor samples (35). PD-L1 Expression PD-L1 is dynamically and widely expressed on the surface of tumor cells, antigen-presenting cells, and other immune cells. PD-L1 expression is reported to be generally low in HCC (~10% of tumor cells) (26) and is associated with recurrence and shorter OS (27). In addition, high PD-L1 expression in inﬂammatory cells within the tumor microenvironment (TME) correlates with high serum alpha-fetoprotein (AFP) levels, macrovascular invasion, and poor differentiation, resulting in increased tumor aggressiveness (28). PD-L1 was the ﬁrst proposed predictive biomarker for responses to ICIs, and the relationship between PD-L1 expression and ICI treatment response has been extensively explored in HCC-related clinical trials. Several results of clinical trials in which PD-L1 was evaluated as a predictive biomarker for ICI responses have been published (the data are summarized in detail in Table 2). In the CheckMate 459 phase III trial, although nivolumab did not elicit prominent improvements in OS as a ﬁrst-line treatment for HCC, PD-L1- positive tumors nevertheless showed a better response to nivolumab compared to sorafenib (17). This agreed with the results of the KEYNOTE-224 trial, which showed that PD-L1 expression was associated with the response to pembrolizumab, even though the results were not statistically signiﬁcant (15). The contradictory results of these clinical trials may be partly due to limitations in the detection of PD-L1 levels, including a lack of standard methods for evaluating PD-L1 expression and its spatial and temporal heterogeneity, as well as the absence of standard thresholds that allow the determination of “overexpression” (30, 31). The expression of PD-L1 is measured using immunohistochemistry on formalin-ﬁxed parafﬁn-embedded (FFPA) sections; however, two methods are used for deﬁning PD-L1-positive expression, namely, the ratio of FIGURE 1 \| The classiﬁcation of biomarkers. The biomarkers introduced in this review can be divided into protein markers, gene markers, immune-related cells, and host-related cells. Moreover, according to their localization in vivo, they can be further subdivided into intratumoral and extratumoral biomarkers as they are detected using different methods. FIGURE 1 \| The classiﬁcation of biomarkers. The biomarkers introduced in this review can be divided into protein markers, gene markers, immune-related cells, and host-related cells. Moreover, according to their localization in vivo, they can be further subdivided into intratumoral and extratumoral biomarkers as they are detected using different methods. September 2021 \| Volume 11 \| Article 716844 Frontiers in Oncology \| www.frontiersin.org 3 He et al. PD-L1 Expression To examine this association more broadly, another study analyzed the clinical and genomic data of thousands of patients with advanced cancer and identiﬁed an association between a higher TMB and improved survival linked to ICI treatment in most of the cancer types assessed (40). Several studies have shown that a higher TMB is associated with immune microenvironment diversiﬁcation and worse prognosis in HCC patients (41, 42). However, its putative role as a predictive biomarker has not been reported in HCC patients treated with either nivolumab or pembrolizumab, and whether the TMB has value as a predictive marker for ICI efﬁcacy remains unclear. p y A study evaluated the frequency of genomic biomarkers of ICI response in 755 HCC patients and found that the median TMB was four mutations/Mb, with only six tumors (0.8%) found to be TMB-high (43). Furthermore, in a small case series (N=17), the TMB showed no correlation with ICI response. As in the case of PD-L1 expression, however, TMB determination lacks standardized thresholds and there is variability in quantiﬁcation methods (44). Thus, the clinical value of TMB should be interpreted with caution. Recently, Wong et al. evaluated the TMB in 29 HCC patients through targeted next-generation sequencing (tNGS) on both fresh and archival samples. The authors reported that, while fresh HCC samples were better sources of tumor DNA (45), the low median TMB values observed may limit the usefulness of the TMB as a predictor of response to immunotherapy in HCC. Tumor Mutational Burden/Microsatellite Instability y Neoantigens, arising as a consequence of tumor-speciﬁc mutations, are hypothesized to generate robust immune responses (36). The tumor mutational burden (TMB) refers to the number of nonsynonymous mutations found in the genome of a single tumor, including alterations in DNA damage response genes and those encoding the DNA polymerase epsilon (POLE) and delta (POLD) catalytic subunits, and has been assessed for its potential as a biomarker in multiple tumor types (37, 38). It is thought that tumors with a greater TMB can produce a greater number of neoantigens. One cross-cancer study reported that tumors with a high TMB are positively correlated with responses to anti-PD-1/PD-L1 therapy, and a greater TMB is also associated with higher PD-L1 expression in tumor cells (39). DNA mismatch repair (MMR) represents a key mechanism for the maintenance of genomic integrity and stability. A deﬁciency in MMR activity results in a hypermutator phenotype known as microsatellite instability (MSI) (46). High mutation rates in somatic cells are believed to amplify the September 2021 \| Volume 11 \| Article 716844 Frontiers in Oncology \| www.frontiersin.org 4 Biomarkers of ICIs in HCC He et al. He et al. TABLE 2 \| Predictive biomarkers in clinical trials of immune-checkpoint inhibitors in hepatocellular carcinoma. Biomarker Cut-off Agent Signiﬁcant Association Findings Study Name (reference) Phase PD-L1 expression CPS ≥1 Pembrolizumab Better ORR and PFS PD-L1 expression was correlated with ORR and PFS KEYNOTE-224 (15) II TPS ≥1 Pembrolizumab No No correlation between PD-L1 expression and ORR or PFS KEYNOTE-224 (15) II TPS ≥1 Nivolumab Better OS Median OS: 28.1 vs. 16.6 months CheckMate 040 (29) I/II TPS ≥1 Nivolumab No ORR: 29% vs. 20% (ITT population) ORR: 13% vs. 22% (Asian cohort) CheckMate 040 (Asian cohort) (18) I/II NA Nivolumab vs. sorafenib Better ORR ORR: 28% vs. 12% CheckMate 459 (17) III NA Pembrolizumab No No correlation between PD-L1 expression and ORR NCT02658019 (21) II Four-gene signature / Nivolumab Better OS Four-gene inﬂammatory signature (CD274, CD8A, LAG3, and STAT1) was associated with improved OS CheckMate 040 (29) I/II Plasma TGF-b TGF-b <200 pg/mL Pembrolizumab Better OS and PFS Median OS: >25 months vs. 7 months; median PFS: >25 months vs. 2 months NCT02658019 (21) II NLR or PLR / Nivolumab Better OS OS was longer in patients with NLR or PLR in the lower tertile CheckMate 040 (29) I/II Serum AFP AFP <400 µg/L Nivolumab Better OS Median OS: 16.8 vs. Tumor Mutational Burden/Microsatellite Instability 13 months CheckMate 040 (29) I/II CPS, combined positive score; TPS, tumor proportion score; ORR, overall response rate; PFS, progression-free survival; ITT, intent-to-treat; NA, not available; OS, overall survival; NLR, neutrophil-to-lymphocyte ratio; PLR, platelet-to-lymphocyte ratio. on score; ORR, overall response rate; PFS, progression-free survival; ITT, intent-to-treat; NA, not available; OS, overall survival; NLR, CPS, combined positive score; TPS, tumor proportion score; ORR, overall response rate; PFS, progression-free survival; ITT, intent-to-treat; NA, not available; OS, overall survival; NLR, neutrophil-to-lymphocyte ratio; PLR, platelet-to-lymphocyte ratio. neoantigen load, leading to lymphocyte activation and enhanced cancer susceptibility to immunotherapy (47, 48). A study assessed the efﬁcacy of PD-1 blockade with pembrolizumab monotherapy in 86 patients with advanced MMR-deﬁcient cancers across 12 different cancer types. Based on ORRs and complete response (CR) rates of 53% and 21%, respectively, the results of this study support that MSI may serve as a predictor of the clinical response of solid tumors to PD-1 blockade (49). Although MSI is considered useful as an agnostic histological indicator for the selection of responders to ICI therapy, there is nonetheless a lack of data for HCC patients. Ang et al. assessed 542 HCC specimens for MSI and found that only one (0.2%) was MSI-high and TMB-high (43), while several other studies have reported that the prevalence of MSI in HCC is also generally low (50, 51). These ﬁndings highlight that MSI may not be an ideal biomarker to predict responses to ICIs in HCC patients. increased frequency of CD3+ T cells and the best overall response. A different study reported that an increase in CD8+ T cells in six-week tumor biopsies was connected with clinical beneﬁts for HCC patients who received tremelimumab plus ablation combination therapy (55). Furthermore, Kaseb et al. found that the clinical response of HCC patients who received perioperative immunotherapy (nivolumab plus ipilimumab) followed by surgical resection was correlated with an increase in CD8+ T-cell inﬁltration, and, speciﬁcally, with two effector T- cell clusters (CD3+CD8+CD45RO+Eomes+ and CD3+CD8+ CD45RO+Eomes+CD57+CD38low) (56). The connection between TILs and ICI treatment response has also been investigated in other immune cell types. Ng et al. analyzed 49 HCC samples from patients treated with ICIs and reported that patients with a high intratumoral CD38+CD68+ macrophage density had a better median OS compared with those with low CD38+CD68+ macrophage density (34.43 vs. 9.66 months) (57), likely because CD38hi macrophages produce greater amounts of interferon-gamma (IFN-g) and related cytokines. Tumor Mutational Burden/Microsatellite Instability Notably, some of these 49 patients had received combination immunotherapy. The inﬂuence of immune cell inﬁltration on the effectiveness of ICIs has also been studied in multifocal HCC, with the results showing that small nodules are more sensitive to anti-PD-1 therapy than large nodules, while small tumors exhibit greater immune cell inﬁltration and an upregulated interferon signature compared with large tumors (58). The low frequency of high TMB and high MSI limits their exploration as predictive biomarkers in HCC, while the number of cases included in existing studies is also relatively low (44). However, given that MSI has shown good predictability in immunotherapy for other types of cancers, research efforts should continue to focus on exploring its predictive potential in HCC, including combining data from multiple clinical trials to obtain a sufﬁciently large sample size that would allow the evaluation of its role in HCC-targeted immunotherapy. Speciﬁc Gene Alterations Over recent years, molecular proﬁling associated with the advent of NGS has provided information on actionable targets and identiﬁed speciﬁc gene alterations associated with responses to ICIs (60). Mutations in the CTNNB1 gene, which lead to the activation of the WNT/b-catenin signaling pathway, are characteristic of immune-excluded HCC types (cold tumors) and are associated with signiﬁcantly lower enrichment scores for several immune signatures (61, 62). Harding et al. employed NGS to determine which type of patient with advanced HCC might beneﬁt from systemic treatments (63). For a subgroup of 31 patients treated with diverse ICIs, activating alterations in the WNT/b-catenin pathway were associated with lower disease control rates (DCRs) (0% vs. 53%), shorter median PFS (2.0 vs. 7.4 months), and shorter median OS (9.1 vs. 15.2 months) compared with those in WNT wild-type HCC. Additionally, b-catenin-driven tumors were reported to be resistant to PD-1 therapy in a mouse model of HCC, while the expression of chemokine (C-C motif) ligand 5 (CCL5) could restore the b-catenin-associated loss of immune surveillance (64). In addition, TP53 gene alterations were the most frequently identiﬁed mutations in HCC patients and were mutually exclusive with CTNNB1 mutations (65). Studies have shown that TP53 alterations are strongly related to the immune microenvironment in HCC, with less CD8+ T cell inﬁltration and more FOXP3+ Treg inﬁltration, resulting in the downregulation of the immune response (66–68). TP53 dysfunction was shown to be linked to chromosomal instability (deﬁned as high broad copy-number alteration loads) and immune-excluded traits in HCC (69). Conversely, Yang et al. reported that HCC patients carrying TP53 neoantigens showed higher cytotoxic lymphocyte inﬁltration and longer OS (70). Combined, these observations indicate that the presence of speciﬁc gene alternations, especially those related to CTNNB1, TP53, noncoding RNAs, and methylation, may inﬂuence the response to ICI treatment through interacting with the immune microenvironment. Although these studies are still preliminary, such alterations could represent novel biomarkers for HCC patient selection, and patient exclusion in particular. Recent studies have indicated that immune-related long noncoding RNAs (lncRNAs) may predict ICI treatment responses in HCC. Peng et al. reported that the host gene of lncRNA MIR155 was strongly positively correlated with the expression of CTLA-4 and PD-L1 in HCC tissues, and showed predictive value for the curative effect of ICI therapy (71). Additionally, Zhang et al. Tumor-Inﬁltrating Lymphocytes Because T-cell inﬁltration within the TME is a prerequisite for immune checkpoint blocking (52), baseline intratumoral T-cell density and phenotype have been extensively studied and closely connected with ICI responses in melanoma and other tumors (53, 54). A retrospective biomarker analysis undertaken in the CheckMate 040 trial indicated that, although not statistically signiﬁcant, increased numbers of CD3+ or CD8+ tumor- inﬁltrating T cells were correlated with a trend towards improved OS in nivolumab-treated HCC patients (29). This study further indicated that an association existed between an Deep single-cell RNA sequencing was performed on 5,063 T cells isolated from peripheral blood, tumor tissue, and adjacent normal tissues of six HCC patients (59). The results demonstrated that speciﬁc subsets, such as exhausted CD8+ T cells and regulatory T cells (Tregs), were preferentially enriched in HCC. The authors further identiﬁed that increased expression of layilin (encoded by LAYN) was associated with tumor- inﬁltrating Tregs and activated CD8+ T cells. While these ﬁndings need to be validated in a larger cohort, they will undoubtedly open a new avenue for further research into the September 2021 \| Volume 11 \| Article 716844 Frontiers in Oncology \| www.frontiersin.org 5 Biomarkers of ICIs in HCC He et al. potential of utilizing CD8+ T-cell inﬁltration as a biomarker in HCC. in HCC patients (73–76). However, prospective validation in HCC patients who received ICI treatment is still lacking. MicroRNAs and circular RNAs have also been reported to reshape the immune microenvironment in HCC (77, 78). Mo et al. identiﬁed 5-methylcytosine (5mC)-associated molecular subtypes in HCC and found that they were associated with responses to immunotherapy (79). Although it is important to focus on lymphocytes in the TME, applying them as biomarkers in the clinic remains a distant prospect. How to select lymphocytes with speciﬁc characteristics, how to deﬁne the positive criteria are among the many unanswered questions. Joint evaluation of the number of multiple immune cells and PD-L1 expression to develop TME scores may be one of the answers. Several studies have indicated that cancer cell-intrinsic epigenetic alterations are associated with carcinogenesis and tumor progression (80–83), as well as with changes in the TME, such as inﬁltration of tumor-associated lymphocytes and expression of immune checkpoint molecules (84, 85), indicative of their potential as predictors of immunotherapy. Furthermore, given that DNA methylation can be measured in liquid biopsies, epigenetic biomarkers may provide additional advantages, such as low patient invasiveness (86). Tumor-Inﬁltrating Lymphocytes In patients with melanoma receiving anti-PD-1 immunotherapy, the hypermethylation of the T-cell costimulatory receptor TNFRSF9 was reported to be correlated with poor PFS and treatment response (87). In a multicenter study, Duruisseaux et al. established an epigenomic proﬁle based on a microarray DNA methylation signature (EPIMMUNE) in a discovery set of tumor samples from patients with advanced nonsmall cell lung cancer (NSCLC) who had received anti-PD-1 therapy (88). The authors found that the unmethylated status of the T-cell differentiation factor forkhead box P1 (FOXP1) was associated with improved PFS and OS, and, therefore, possessed good predictive value for the efﬁcacy of anti-PD-1 treatment. Another study identiﬁed a broad DNA methylation signature in peripheral blood mononuclear cells and T cells of HCC patients that differed from that of non- HCC patients (89). Additionally, differences in immune inﬁltrates related to the methylation levels of cell division cycle-associated (CDCA) family genes in HCC were reported to have potential as predictive biomarkers for responses to immunotherapy (90). Notably, Llopiz et al. reported that an epigenetic drug/ICI combination exerted synergistic antitumor effects in a murine model of HCC (91), thereby providing further evidence that DNA methylation signatures may be related to the efﬁcacy of immunotherapy, and also have the potential to serve as biomarkers of combination therapy. Speciﬁc Gene Alterations identiﬁed an immune-related lncRNA signature that correlated with worse survival and was an independent prognostic biomarker for HCC patients (72). That this signature was associated with immune cell inﬁltration and ICI treatment-related molecules (including PD-L1, PD-L2, and IDO1) suggests that it may have the potential to measure the response to ICI immunotherapy. Several studies have reported on the ability of immune-related lncRNAs to predict prognosis Immune-Related Gene Signatures Immune-Related Gene Signatures Comprehensive analyses of tumor transcriptomic proﬁling data have recently been conducted to characterize the responsiveness of the immune microenvironment to ICI treatment. Ayers et al. identiﬁed an 18-gene T-cell-inﬂamed expression proﬁle and a 6- gene IFN-g signature that could predict responses to pembrolizumab therapy across multiple solid tumors (92). Based on gene expression proﬁles, Sia et al. classiﬁed 25% of HCCs as an “immune-speciﬁc class” characterized by high expression levels of September 2021 \| Volume 11 \| Article 716844 Frontiers in Oncology \| www.frontiersin.org 6 Biomarkers of ICIs in HCC He et al. OS and PFS (21). The therapeutic co-administration of TGF-b- blocking and anti-PD-L1 antibodies was reported to facilitate T- cell penetration into the center of tumors and elicit strong antitumor immunity (101). Clinical trials of the TGF-b receptor I inhibitor galunisertib have been conducted, and have reported median OS durations of 16.8 months in patients with advanced HCC with baseline AFP <1.5 × ULN (102). Brieﬂy, TGF-b may serve as a negative predictive biomarker for ICI therapy given that HCCs with strong TGF-b and exhausted immune signatures may be resistant to PD-1 blockade, while HCC patients with an activated TGF-b signature are expected to beneﬁt from a combination of ICI and TGF-b inhibitor therapy (99). Thus, it seems likely that drugs targeting both TGF-b and PD-1/PD-L1, such as bintrafusp alfa, will play a role in the treatment of HCC in the future. PD-L1 and markers of cytolytic activity. This class further comprised two subtypes, namely, an active immune subtype, characterized by signiﬁcant enrichment of T cells and IFN signatures, and an exhausted immune subtype, associated with a T-cell exhaustion signature and immunosuppressive components (TGF-b and M2 macrophages) (62). The previously-mentioned retrospective analysis of the ﬁndings of the CheckMate 040 trial also included the evaluation of the predictive value of a four-gene inﬂammatory signature (CD274 [PD-L1],CD8A,LAG3,andSTAT1).Thissignature wasfoundtobe associated with improved response and OS related to nivolumab therapy, both in the dose-escalation and dose-expansion phases (29), and may be indicative of IFN-g/STAT1-dependent CD8+ T- cell expansion, LAG-3-dependent T-cell exhaustion, and/or an immune-suppressed TME with high PD-L1 expression. Importantly, however, although the differences in the above- mentioned clinical trial data were reported to be statistically signiﬁcant, this gene signature does not adequately distinguish between responders and nonresponders. Immune-Related Gene Signatures Recently, a ﬁve-gene immune-related signature (LDHA, PPAT, BFSP1, NR0B1, and PFKFB4) was identiﬁed and used to establish a prognostic model for responses to HCC treatment that could stratify patients who were sensitive toimmunotherapy (93).Otherstudiesusing different gene combination strategies have also reported their potential to predictICIresponsesbyreﬂectingthecharacteristicsoftheimmune microenvironment (94, 95). Nevertheless, these signatures need further testing for clinical application. Several soluble immune checkpoint-related proteins were recently shown to have promising predictive value in various cancer types. Chen et al. reported that metastatic melanomas released exosomes carrying PD-L1 that suppressed CD8+ T-cell function and supported tumor growth. Additionally, the authors reported that the magnitude of the increase in the levels of circulating exosomal PD-L1 during the early stages of treatment could discriminate clinical responders from nonresponders (103). One preclinical study using mouse models indicated that the removal of exosomal PD-L1 inhibits tumor growth, even in models resistant to PD-L1 blockade (104). Soluble PD-L1 (sPD-L1) has been reported to be correlated with responses to immunotherapy in patients with NSCLC (105). Additionally, several studies have focused on the prognostic value of sPD-L1 and identiﬁed high serum sPD-L1 levels as an independent predictive factor for poor outcomes in HCC patients (106–108). Whether an association exists between sPD-L1 and intratumoral PD-L1 expression levels remains unclear, with contradictory results having been reported (106, 108). Moreover, there is a lack of data on the ability of exosomal PD-L1 or sPD-L1 to predict clinical outcomes in HCC patients following ICI therapy, which warrants further study. Immune-related gene signatures are associated with the same obstacles as TILs, namely, how to select the best gene combinations and positive cut-off values. A feasible strategy involves undertaking a wide-ranging bioinformatics analysis of the existing databases to identify possible gene combinations and then verify them through basic research and clinical trials. Biomarkers in Peripheral Blood The continuous availability of tumor samples from ICI-treated patients is crucial for biomarker research; however, this is difﬁcult to achieve owing to the invasive nature of biopsies. In contrast, circulating biomarkers can be easily collected and repeatedly measured after treatment, rendering them more convenient for use in the clinic. Circulating immune cells in peripheral blood have been extensively evaluated as predictive biomarkers (105, 109). Agdashian et al. tested the combination of anti-CTLA-4 treatment (tremelimumab) with locoregional therapy in HCC patients and found that the frequency of CD4+PD-1+ cells among peripheral blood mononuclear cells at baseline was higher in patients responding to therapy than in nonresponding patients (110). In addition, low baseline peripheral B cell PD-1 positivity and constant posttreatment monocyte PD-L1 positivity were observed to be associated with disease control in 16 HCC patients treated with nivolumab (111). Dharmapuri et al. evaluated the relationship between the neutrophil–lymphocyte (NLR) and platelet– lymphocyte (PLR) ratios and survival outcomes in HCC patients treated with nivolumab and reported that patients who achieved a partial or CR had signiﬁcantly lower posttreatment NLRs and PLRs (112). The predictive value of the NLR and the PLR was also indicated in the CheckMate 040 trial (29). Studies have shown that TGF-b plays a central role in immune suppression within the TME and tumor immune evasion (96). In HCC, the potent immune inhibitory function of Tregs is a major obstacle to generating an effective antitumor response, and Treg activation is modulated by the TGF-b pathway (97). In addition, TGF-b enables tumor evasion from host immune responses in part through enhancing SMAD3- mediated PD-1 expression on TILs (98). The TGF-b signaling pathway is activated at the transcriptional level in most HCCs (99, 100). A strong association was identiﬁed between the TGF-b signature and the exhausted immune signature in HCC (62, 99). In a phase II trial, several representative circulating biomarkers were analyzed in 29 patients with unresectable HCC treated with pembrolizumab, with the results showing that baseline plasma TGF-b levels of <200 pg/mL were an effective predictor of better Necrotic or apoptotic tumor cells release DNA carrying tumor-related genetic and epigenetic alterations into the September 2021 \| Volume 11 \| Article 716844 Frontiers in Oncology \| www.frontiersin.org 7 Biomarkers of ICIs in HCC He et al. He et al. Biomarkers in Peripheral Blood However, a study analyzing the mutational landscape of advanced HCC using ctDNA reported that WNT pathway-related mutations were not associated with clinical outcomes after ICI therapy (65), highlighting that further investigation is needed to determine whether ctDNA can indeed serve as a predictive biomarker in HCC. Meanwhile, although the blood TMB was reported to have good predictive value in some types of cancer (116, 117), evidence is lacking for HCC. Winograd et al. sought to detect PD-L1-expressing circulating tumor cells (CTCs) in HCC patients and found that there was a strong association between the presence of PD-L1+ CTCs and favorable treatment responses to PD-1 blockade (118). High AFP levels are considered to be a prognostic marker for poor clinical outcomes in HCC patients. Recently, a posttreatment decline in serum AFP levels was reported to be associated with higher ICI treatment efﬁcacy in advanced HCC (119, 120). In addition, Spahn et al. reported that baseline levels of AFP of <400 µg/LatthestartofICItreatmentwereassociatedwithhigherratesof CR or partial response (PR) as best responses (121). However, the results of the checkmate040 clinical trial showed that although baseline AFP <400 µg/L was associated with longer OS compared with AFP ≥400 µg/L, the ORR and DCR were similar regardless of baselineAFPlevels(29).GiventhattheAFPleveliscloselyrelatedto the baselinecharacteristics of patients, its predictive effect shouldbe interpreted with caution. Biomarkers in peripheral blood have the great advantage of low invasiveness, while a large number of studies have also shown their potential as predictors of immunotherapeutic efﬁcacy. However, supporting data from large-sample clinical trials are still lacking. In addition, there is currently no evidence that the predictive accuracy of blood samples is better than that of tissue samples for any given biomarker. Nevertheless, these studies provide insights for future research into biomarkers for HCC immunotherapy. However, in the aforementioned studies, there was no overlap in gut microbiota associated with responses, which may be due to differences in etiology, geographic location, nutritional intake, and techniques used to analyze the samples. Moreover, Rosshart et al. found that the gut microbiome of laboratory mice differs signiﬁcantly from that of closely related species in the wild (133), implying that caution is needed when generalizing these research results. The gut microbiota can be inﬂuenced by many environmental, dietary, and lifestyle factors, all of which can potentially affect the immune system and, consequentially, regulate the response to ICIs (134). Biomarkers in Peripheral Blood hepatic exposure to bacterial products, it seems likely that a relationship exists between the gut microbiota and responses to ICItherapy.Indeed,increasingevidencepointstowardsakeyroleof the bacterial microbiome in promoting the development of HCC (125). In the context of chronic inﬂammation, intestinal bacterial translocation is detected by Toll-like receptor (TLR) 4 present on resident liver cells through its ligand lipopolysaccharide, which leads to the upregulation of the expression of the hepatomitogen epiregulin and, consequently, the promotion of hepatocarcinogenesis (126). Microbiota-derived metabolites can also affect the development of HCC. For instance, the gut microbial metabolite deoxycholic acid acts in concert with lipoteichoic acid to enhance the tumor-promoting phenotype of hepatic stellate cells and promote the expression of COX2 through TLR2, resulting in the suppression of antitumor immunity (127). Additionally, gut microbial-dependent bile acid metabolism modulates liver tumor growth by regulating the hepatic expression of CXCL16, a mediator of natural killer T (NKT) cell recruitment (128). Furthermore, Arpaia et al. found that microbe-derived short-chain fatty acids facilitate extrathymic Treg generation (129). Studies have also shown that the gut microbiota may play an important role in regulating ICI treatment responses (130, 131). For instance, fecal microbiota transplantation from cancer patients who responded to ICIs into sterile orantibiotic-treated mice improved the response to anti-PD-1 therapy (122, 123). Moreover, Zheng and colleagues recently reported the dynamic variation in the composition of the gut microbiome during anti-PD-1 immunotherapy in HCC by metagenomic sequencing (132). They observed that fecal samples from patients responding to immunotherapy showed higher taxonomic richness and greater gene counts compared with those ofnonresponders;microbialcompositionremainedrelativelystable in the responder group, whereas in nonresponders, Proteobacteria abundance markedly increased from week 3 and became predominant at week 12. In addition, antibiotic administration at the initiation of ICI treatment was reportedly associated with worse outcomes, indicative of the inﬂuence of the gut microbiota on HCC treatment (121). A clinical trial (NCT03785210) to evaluate the effects of combined antibiotic (vancomycin) and ICI therapies is currently underway. bloodstream, thereby helping to overcome the limitations related to sample availability. Circulating tumor DNA (ctDNA) can reportedly predict tumor responses to ICI therapy in several types of tumors; speciﬁcally, it can be used to distinguish pseudoprogression from true progression (113–115). Gut Microbiota Increasing evidence has indicated that the gut microbiota plays a crucial role in the development and regulation of innate and adaptive immunity, while several studies have described its value in predicting the efﬁcacy of ICIs. For instance, analysis of baseline gut microbiota composition of fecal samples from patients with melanoma or NSCLC before immunotherapy treatment has indicated that commensal microbial composition is associated with an ICI response (122, 123). Ameta-analysisundertakenon2,424samplesthrough16SRNA genesequencingand machine based-learningindicated that among several major dysbiosis-related diseases, liver cirrhosis is the condition where changes in the gut microbiome most accurately predict the presence of disease (124). Given that there is an anatomical connection between the liver and the gut, and that HCC occurs in the context of chronic liver inﬂammation concomitant with a defective intestinal barrier and increased Frontiers in Oncology \| www.frontiersin.org Biomarkers in Peripheral Blood Given these complications, the application of gut microbiota as a biomarker for clinical use remains a distant possibility. Potential Biomarkers for irAEs Between 15% and 25% of ICI-treated HCC patients undergo grade 3/4 treatment for immune-related adverse events, including fatigue, pruritus, rash, diarrhea, and increases in aspartate aminotransferase and alanine aminotransferase levels (14, 15). Although ICI treatment-related adverse events are manageable and less frequent than those seen with chemotherapy, it is still necessary to identify biomarkers that can predict irAEs to alleviate unnecessary suffering in patients. p y g p Most studies have focused on identifying biomarkers for predicting the efﬁcacy of immunotherapy, while relatively few studies have investigated biomarkers relating to irAEs. Moreover, a large proportion of research ﬁndings come from patients with melanoma (148). Nevertheless, these studies can serve as a reference for identifying biomarkers that can predict irAEs in HCC. Baseline serum IL-6 and IL-17 levels were signiﬁcantly associated with an increased risk of severe toxicity in patients treated with ipilimumab (149, 150), with IL-17 being related to severe diarrhea/colitis. A retrospective review involving 167 adult patients with solid tumors indicated that increased baseline lymphocyte counts are associated with a greater risk for irAEs in patients treated with nivolumab or pembrolizumab (151). The detection of autoantibodies has been suggested to predict the development of irAEs related to the autoantibodies, and two studies evaluating antithyroid antibodies and diabetes-related autoantibodies have been reported (152, 153). Baseline gut microbiota enriched in Faecalibacterium spp. and other members of the Firmicutes is associated with a more frequent occurrence of ipilimumab-induced colitis (154). Pﬁster et al. conducted a meta-analysis that incorporated more than 1,600 patients with advanced HCC in three randomized phase III clinical trials and reported that PD-L1 or PD-1 inhibitors did not improve survival in patients with nonviral HCC, particularly NAFLD (142). This was likely due to the progressive accumulation of exhausted, unconventionally activated CD8+PD-1+ T cells, which contributed to inducing NAFLD/HCC rather than carrying out or enhancing immune surveillance. However, in the CheckMate 040 phase I/II trial, which enrolled HCC patients with or without HBV or HCV infection, responses to nivolumab were observed irrespective of HCC etiology (14). Epithelial-to-mesenchymal transition (EMT) has been implicated as a resistance mechanism that helps to promote the immune evasion of cancer cells (143). High expression of PDL1 in HCC patients is reported to be associated with an EMT phenotype and be a predictor of poor survival (41). Potential Biomarkers for irAEs The correlation between PD-L1 expression and EMT presents a theoretical foundation to investigate EMT as a negative biomarker for ICI responses. Rogado et al. reported that ICI treatment was markedly more beneﬁcial for patients with advanced cancer presenting with irAEs than for those without irAEs (ORR: 82.5% vs. 16.6%; PFS: 10 vs. 3 months) (155). Cutaneous or early irAEs are associated with improved survival in melanoma patients treated with nivolumab (156, 157). Future studies should address this association to explore the underlying biological mechanisms related to ICI efﬁcacy, while how to balance the incidence of irAEs and the immunotherapeutic response also merits serious consideration. Matrix metalloproteinase 9 (MMP-9) secreted by tumor- associated macrophages was recently reported to be a potential predictor of immune characteristics and immunotherapeutic responses in HCC (144, 145). One study measured ADAM9 mRNA levels in blood samples derived from patients with advanced HCC. Among four patients treated with nivolumab therapy, two who exhibited a clinical response also showed signiﬁcant decreases in serum ADAM9 mRNA levels, whereas the two who displayed no response to nivolumab did not. Although the sample size was small, the results of this study nevertheless suggested that ADAM9 mRNA might serve as a predictive biomarker for clinical responses (146). Others The T-cell receptor (TCR) is composed of multiple antigen- speciﬁc peptide chains. Recent studies have used high- throughput sequencing for the in-depth elucidation of the September 2021 \| Volume 11 \| Article 716844 8 He et al. Biomarkers of ICIs in HCC composition and distribution of the TCR. There have been several attempts to use the TCR as a predictive biomarker for ICI responses. A more clonal TCR repertoire or oligoclonal TIL expansion is associated with a better response to PD-1 blockade in melanoma patients (135, 136). In addition, baseline TCR diversity in peripheral blood has been associated with clinical outcomes following ipilimumab treatment in metastatic melanoma (137). Despite the lack of data regarding the suitability of employing TCR as a biomarker in HCC, its predictive potential nevertheless warrants further investigation given the novel TCR analysis approaches proposed (138, 139). Lin et al. found that the combination of TCR repertoires and TNM stage could serve as an efﬁcient prognostic indicator in patients with HBV-associated HCC (140). Additionally, Han et al. identiﬁed several speciﬁc TRBV–TRBJ combinations that could distinguish the TCR repertoires of HCC patients from those of healthy adults and thus have the potential to serve as novel biomarkers (141). predictor of immunotherapy outcome in HCC and opens up new avenues for predictor selection. REFERENCES 1. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA: Cancer J Clin (2021) 71(3):209–49. doi: 10.3322/caac.21660 13. Chen DS, Mellman I. Oncology Meets Immunology: The Cancer-Immunity Cycle. Immunity (2013) 39(1):1–10. doi: 10.1016/j.immuni.2013.07.012 2. Lin S, Hoffmann K, Schemmer P. Treatment of Hepatocellular Carcinoma: A Systematic Review. Liver Cancer (2012) 1(3-4):144–58. doi: 10.1159/000343828 14. El-Khoueiry AB, Sangro B, Yau T, Crocenzi TS, Kudo M, Hsu C, et al. Nivolumab in Patients With Advanced Hepatocellular Carcinoma (CheckMate 040): An Open-Label, Non-Comparative, Phase 1/2 Dose Escalation and Expansion Trial. Lancet (London England) (2017) 389 (10088):2492–502. doi: 10.1016/s0140-6736(17)31046-2 3. Park JW, Chen M, Colombo M, Roberts LR, Schwartz M, Chen PJ, et al. Global Patterns of Hepatocellular Carcinoma Management From Diagnosis to Death: The BRIDGE Study. Liver International: Off J Int Assoc Study Liver (2015) 35(9):2155–66. doi: 10.1111/liv.12818 15. Zhu AX, Finn RS, Edeline J, Cattan S, Ogasawara S, Palmer D, et al. Pembrolizumab in Patients With Advanced Hepatocellular Carcinoma Previously Treated With Sorafenib (KEYNOTE-224): A Non-Randomised, Open-Label Phase 2 Trial. Lancet Oncol (2018) 19(7):940–52. doi: 10.1016/ s1470-2045(18)30351-6 4. Personeni N, Rimassa L. Hepatocellular Carcinoma: A Global Disease in Need of Individualized Treatment Strategies. J Oncol Pract (2017) 13 (6):368–9. doi: 10.1200/jop.2017.024604 4. Personeni N, Rimassa L. Hepatocellular Carcinoma: A Global Disease in Need of Individualized Treatment Strategies. J Oncol Pract (2017) 13 (6):368–9. doi: 10.1200/jop.2017.024604 5. Zhang X. NAFLD Related-HCC: The Relationship With Metabolic Disorders. Adv Exp Med Biol (2018) 1061:55–62. doi: 10.1007/978-981- 10-8684-7_5 5. Zhang X. NAFLD Related-HCC: The Relationship With Metabolic Disorders. Adv Exp Med Biol (2018) 1061:55–62. doi: 10.1007/978-981- 10-8684-7_5 16. Finn RS, Ryoo BY, Merle P, Kudo M, Bouattour M, Lim HY, et al. Pembrolizumab As Second-Line Therapy in Patients With Advanced Hepatocellular Carcinoma in KEYNOTE-240: A Randomized, Double- Blind, Phase III Trial. J Clin Oncol: Off J Am Soc Clin Oncol (2020) 38 (3):193–202. doi: 10.1200/jco.19.01307 6. Forner A, Reig M, Bruix J. Hepatocellular Carcinoma. Lancet (London England) (2018) 391(10127):1301–14. doi: 10.1016/s0140-6736(18)30010-2 7. Bouattour M, Mehta N, He AR, Cohen EI, Nault JC. Systemic Treatment for Advanced Hepatocellular Carcinoma. Liver Cancer (2019) 8(5):341–58. doi: 10.1159/000496439 17. Yau T, Park JW, Finn RS, Cheng AL, Mathurin P, Edeline J, et al. ACKNOWLEDGMENTS The authors would like to acknowledge Yuhao Hu for assistance with referencing. 12. Nishida N, Kitano M, Sakurai T, Kudo M. Molecular Mechanism and Prediction of Sorafenib Chemoresistance in Human Hepatocellular Carcinoma. Digest Dis (Basel Switzerland) (2015) 33(6):771–9. doi: 10.1159/000439102 CONCLUSION In HCC, although several studies have been conducted to identify predictive biomarkers that would allow the stratiﬁcation of patients who could beneﬁt from ICI treatment, few have been prospectively validated and none have resulted in the rewriting of the current clinical guidelines or entered into clinical practice. Here, we summarized the progress of immunotherapy for HCC over recent years, with a particular emphasis on predictive biomarkers. However, as HCC-related immunotherapy is still in its infancy, basic research and clinical trials exploring the predictive efﬁcacy of immunotherapy biomarkers are still In addition, Qayyum et al. undertook an interesting prospective study that included 15 patients with advanced HCC treated with pembrolizumab and found that the changes in HCC stiffness as measured by magnetic resonance elastography (MRE) at 6 weeks was signiﬁcantly associated with OS (147). This was the ﬁrst proposed imaging-based September 2021 \| Volume 11 \| Article 716844 Frontiers in Oncology \| www.frontiersin.org 9 He et al. Biomarkers of ICIs in HCC limited, and it is not yet possible to determine which biomarker(s) can effectively predict the efﬁcacy of immunotherapy. It is particularly noteworthy that biomarkers represent continuums and undergo dynamic changes in a population of patients. Accordingly, it is pivotal to obtain samples from patients both before and during treatment to evaluate these dynamic changes and properly determine the predictive value of the assessed biomarkers, while adequate consideration should be given to their application in clinical practice. Furthermore, we only reviewed the biomarkers that predict responses to immune monotherapy, especially anti-PD-L1 and anti-PD-1 agents. However, given the success of the IMbrave150 phase III clinical trial, it is clear that antivascular therapy combined with immunotherapy has great potential in patients with advanced HCC, and combination therapy may be the direction of cancer treatment in the future. But while combination therapy can prolong the OS of HCC patients, it also complicates the patient selection process. The development of research techniques such as NGS, single-cell RNA sequencing, and artiﬁcial intelligence should allow for a more comprehensive understanding of the various components of the TME and their interactions, and potential biomarkers could be widely screened on a genomic scale to identify the predictors of treatment efﬁcacy. FUNDING This work was funded by the National Natural Science Foundation of China (No. 81974483). This work was funded by the National Natural Science Foundation of China (No. 81974483). AUTHOR CONTRIBUTIONS PZ conceived of the topic for this review. All authors contributed to the article and approved the submitted version. REFERENCES Chang L, Chang M, Chang HM, Chang F. Microsatellite Instability: A Predictive Biomarker for Cancer Immunotherapy. Appl Immunohistochem Mol Morphology: AIMM (2018) 26(2):e15–21. doi: 10.1097/pai.0000000000000575 29. Sangro B, Melero I, Wadhawan S, Finn RS, Abou-Alfa GK, Cheng AL, et al. Association of Inﬂammatory Biomarkers With Clinical Outcomes in Nivolumab-Treated Patients With Advanced Hepatocellular Carcinoma. J Hepatol (2020) 73(6):1460–9. doi: 10.1016/j.jhep.2020.07.026 49. LeDT,DurhamJN,SmithKN,WangH,BartlettBR,AulakhLK,etal.Mismatch Repair Deﬁciency Predicts Response of Solid Tumors to PD-1 Blockade. Sci (New York NY) (2017) 357(6349):409–13. doi: 10.1126/science.aan6733 30. Hansen AR, Siu LL. PD-L1 Testing in Cancer: Challenges in Companion Diagnostic Development. JAMA Oncol (2016) 2(1):15–6. doi: 10.1001/ jamaoncol.2015.4685 50. Yamamoto H, Itoh F, Fukushima H, Kaneto H, Sasaki S, Ohmura T, et al. Infrequent Widespread Microsatellite Instability in Hepatocellular Carcinomas. Int J Oncol (2000) 16(3):543–7. doi: 10.3892/ijo.16.3.543 31. Topalian SL, Taube JM, Anders RA, Pardoll DM. Mechanism-Driven Biomarkers to Guide Immune Checkpoint Blockade in Cancer Therapy. Nat Rev Cancer (2016) 16(5):275–87. doi: 10.1038/nrc.2016.36 51. Goumard C, Desbois-Mouthon C, Wendum D, Calmel C, Merabtene F, Scatton O, et al. Low Levels of Microsatellite Instability at Simple Repeated Sequences Commonly Occur in Human Hepatocellular Carcinoma. Cancer Genomics Proteomics (2017) 14(5):329–39. doi: 10.21873/cgp.20043 32. Vilain RE, Menzies AM, Wilmott JS, Kakavand H, Madore J, Guminski A, et al. Dynamic Changes in PD-L1 Expression and Immune Inﬁltrates Early During Treatment Predict Response to PD-1 Blockade in Melanoma. Clin Cancer Research: an Off J Am Assoc Cancer Res (2017) 23(17):5024–33. doi: 10.1158/1078-0432.Ccr-16-0698 52. Teng MW, Ngiow SF, Ribas A, Smyth MJ. Classifying Cancers Based on T- Cell Inﬁltration and PD-L1. Cancer Res (2015) 75(11):2139–45. doi: 10.1158/ 0008-5472.Can-15-0255 33. Abiko K, Matsumura N, Hamanishi J, Horikawa N, Murakami R, Yamaguchi K, et al. IFN-g From Lymphocytes Induces PD-L1 Expression and Promotes Progression of Ovarian Cancer. Br J Cancer (2015) 112 (9):1501–9. doi: 10.1038/bjc.2015.101 53. Thommen DS, Koelzer VH, Herzig P, Roller A, Trefny M, Dimeloe S, et al. A Transcriptionally and Functionally Distinct PD-1(+) CD8(+) T Cell Pool With Predictive Potential in Non-Small-Cell Lung Cancer Treated With PD-1 Blockade. Nat Med (2018) 24(7):994–1004. doi: 10.1038/s41591-018-0057-z 34. Chan LC, Li CW, Xia W, Hsu JM, Lee HH, Cha JH, et al. IL-6/JAK1 Pathway Drives PD-L1 Y112 Phosphorylation to Promote Cancer Immune Evasion. J Clin Invest (2019) 129(8):3324–38. doi: 10.1172/jci126022 54. Chen PL, Roh W, Reuben A, Cooper ZA, Spencer CN, Prieto PA, et al. REFERENCES doi: 10.3389/ fonc.2018.00269 42. Xie F, Bai Y, Yang X, Long J, Mao J, Lin J, et al. Comprehensive Analysis of Tumour Mutation Burden and the Immune Microenvironment in Hepatocellular Carcinoma. Int Immunopharmacol (2020) 89(Pt A):107135. doi: 10.1016/j.intimp.2020.107135 23. Kelley RK. Atezolizumab Plus Bevacizumab - A Landmark in Liver Cancer. New Engl J Med (2020) 382(20):1953–5. doi: 10.1056/NEJMe2004851 24. Rizzo A, Ricci AD, Brandi G. Immune-Based Combinations for Advanced Hepatocellular Carcinoma: Shaping the Direction of First-Line Therapy. Future Oncol (London England) (2021) 17(7):755–7. doi: 10.2217/fon-2020-0986 43. Ang C, Klempner SJ, Ali SM, Madison R, Ross JS, Severson EA, et al. Prevalence of Established and Emerging Biomarkers of Immune Checkpoint Inhibitor Response in Advanced Hepatocellular Carcinoma. Oncotarget (2019) 10(40):4018–25. doi: 10.18632/oncotarget.26998 25. Rizzo A, Ricci AD, Brandi G. Atezolizumab in Advanced Hepatocellular Carcinoma: Good Things Come to Those Who Wait. Immunotherapy (2021) 13(8):637–44. doi: 10.2217/imt-2021-0026 44. Merino DM, McShane LM, Fabrizio D, Funari V, Chen SJ, White JR, et al. Establishing Guidelines to Harmonize Tumor Mutational Burden (TMB): In Silico Assessment of Variation in TMB Quantiﬁcation Across Diagnostic Platforms: Phase I of the Friends of Cancer Research TMB Harmonization Project. J Immunother Cancer (2020) 8(1):e000147. doi: 10.1136/jitc-2019- 000147 26. Pinato DJ, Mauri FA, Spina P, Cain O, Siddique A, Goldin R, et al. Clinical Implications of Heterogeneity in PD-L1 Immunohistochemical Detection in Hepatocellular Carcinoma: The Blueprint-HCC Study. Br J Cancer (2019) 120(11):1033–6. doi: 10.1038/s41416-019-0466-x 27. Gao Q, Wang XY, Qiu SJ, Yamato I, Sho M, Nakajima Y, et al. Overexpression of PD-L1 Signiﬁcantly Associates With Tumor Aggressiveness and Postoperative Recurrence in Human Hepatocellular Carcinoma. Clin Cancer Research: an Off J Am Assoc Cancer Res (2009) 15(3):971–9. doi: 10.1158/1078-0432.Ccr-08-1608 45. Wong CN, Fessas P, Dominy K, Mauri FA, Kaneko T, Parcq PD, et al. Qualiﬁcation of Tumour Mutational Burden by Targeted Next-Generation Sequencing as a Biomarker in Hepatocellular Carcinoma. Liver International: Off J Int Assoc Study Liver (2021) 41(1):192–203. doi: 10.1111/liv.14706 Off J Int Assoc Study Liver (2021) 41(1):192–203. doi: 10.1111/liv.14 28. Calderaro J, Rousseau B, Amaddeo G, Mercey M, Charpy C, Costentin C, et al. Programmed Death Ligand 1 Expression in Hepatocellular Carcinoma: Relationship With Clinical and Pathological Features. Hepatol (Baltimore Md) (2016) 64(6):2038–46. doi: 10.1002/hep.28710 46. Jiricny J. The Multifaceted Mismatch-Repair System. Nat Rev Mol Cell Biol (2006) 7(5):335–46. doi: 10.1038/nrm1907 47. Baretti M, Le DT. DNA Mismatch Repair in Cancer. Pharmacol Ther (2018) 189:45–62. doi: 10.1016/j.pharmthera.2018.04.004 48. REFERENCES Analysis of Immune Signatures in Longitudinal Tumor Samples Yields Insight Into Biomarkers of Response and Mechanisms of Resistance to Immune Checkpoint Blockade. Cancer Discov (2016) 6(8):827–37. doi: 10.1158/2159-8290.Cd-15-1545 35. Lee HH, Wang YN, Xia W, Chen CH, Rau KM, Ye L, et al. Removal of N- Linked Glycosylation Enhances PD-L1 Detection and Predicts Anti-PD-1/ PD-L1 Therapeutic Efﬁcacy. Cancer Cell (2019) 36(2):168–78.e4. doi: 10.1016/j.ccell.2019.06.008 55. Duffy AG, Ulahannan SV, Makorova-Rusher O, Rahma O, Wedemeyer H, Pratt D, et al. Tremelimumab in Combination With Ablation in Patients With Advanced Hepatocellular Carcinoma. J Hepatol (2017) 66(3):545–51. doi: 10.1016/j.jhep.2016.10.029 36. Schumacher TN, Schreiber RD. Neoantigens in Cancer Immunotherapy. Sci (New York NY) (2015) 348(6230):69–74. doi: 10.1126/science.aaa4971 37. Rayner E, van Gool IC, Palles C, Kearsey SE, Bosse T, Tomlinson I, et al. A Panoply of Errors: Polymerase Proofreading Domain Mutations in Cancer. Nat Rev Cancer (2016) 16(2):71–81. doi: 10.1038/nrc.2015.12 56. Kaseb AO, Vence L, Blando J, Yadav SS, Ikoma N, Pestana RC, et al. Immunologic Correlates of Pathologic Complete Response to Preoperative Immunotherapy in Hepatocellular Carcinoma. Cancer Immunol Res (2019) 7(9):1390–5. doi: 10.1158/2326-6066.Cir-18-0605 38. Gong J, Wang C, Lee PP, Chu P, Fakih M. Response to PD-1 Blockade in Microsatellite Stable Metastatic Colorectal Cancer Harboring a POLE Mutation. J Natl Compr Cancer Netw: JNCCN (2017) 15(2):142–7. doi: 10.6004/jnccn.2017.0016 57. Ng HHM, Lee RY, Goh S, Tay ISY, Lim X, Lee B, et al. Immunohistochemical Scoring of CD38 in the Tumor Microenvironment Predicts Responsiveness to Anti-PD-1/PD-L1 Immunotherapy in Hepatocellular Carcinoma. J Immunother Cancer (2020) 8(2):e000987. doi: 10.1136/jitc-2020-000987 39. Yarchoan M, Hopkins A, Jaffee EM. Tumor Mutational Burden and Response Rate to PD-1 Inhibition. New Engl J Med (2017) 377(25):2500– 1. doi: 10.1056/NEJMc1713444 58. Huang M, He M, Guo Y, Li H, Shen S, Xie Y, et al. The Inﬂuence of Immune Heterogeneity on the Effectiveness of Immune Checkpoint Inhibitors in Multifocal Hepatocellular Carcinomas. Clin Cancer Research: an Off J Am Assoc Cancer Res (2020) 26(18):4947–57. doi: 10.1158/1078-0432.Ccr-19-3840 40. Samstein RM, Lee CH, Shoushtari AN, Hellmann MD, Shen R, Janjigian YY, et al. Tumor Mutational Load Predicts Survival After Immunotherapy Across Multiple Cancer Types. Nat Genet (2019) 51(2):202–6. doi: 10.1038/s41588-018-0312-8 59. Zheng C, Zheng L, Yoo JK, Guo H, Zhang Y, Guo X, et al. Landscape of Inﬁltrating T Cells in Liver Cancer Revealed by Single-Cell Sequencing. Cell (2017) 169(7):1342–56.e16. doi: 10.1016/j.cell.2017.05.035 41. REFERENCES LBA38_PR - CheckMate 459: A Randomized, Multi-Center Phase III Study of Nivolumab (NIVO) vs Sorafenib (SOR) as First-Line (1L) Treatment in Patients (Pts) With Advanced Hepatocellular Carcinoma (aHCC). Ann Oncol (2019) 30:v874–v5. doi: 10.1093/annonc/mdz394.029 8. Llovet JM, Ricci S, Mazzaferro V, Hilgard P, Gane E, Blanc JF, et al. Sorafenib in Advanced Hepatocellular Carcinoma. New Engl J Med (2008) 359(4):378–90. doi: 10.1056/NEJMoa0708857 18. Yau T, Hsu C, Kim TY, Choo SP, Kang YK, Hou MM, et al. Nivolumab in Advanced Hepatocellular Carcinoma: Sorafenib-Experienced Asian Cohort Analysis. J Hepatol (2019) 71(3):543–52. doi: 10.1016/j.jhep.2019.05.014 9. Cheng AL, Kang YK, Chen Z, Tsao CJ, Qin S, Kim JS, et al. Efﬁcacy and Safety of Sorafenib in Patients in the Asia-Paciﬁc Region With Advanced Hepatocellular Carcinoma: A Phase III Randomised, Double-Blind, Placebo-Controlled Trial. Lancet Oncol (2009) 10(1):25–34. doi: 10.1016/ s1470-2045(08)70285-7 19. Wainberg ZA, Segal NH, Jaeger D, Lee K-H, Marshall J, Antonia SJ, et al. Safety and Clinical Activity of Durvalumab Monotherapy in Patients With Hepatocellular Carcinoma (HCC). J Clin Oncol (2017) 35(15_suppl):4071–. doi: 10.1200/JCO.2017.35.15_suppl.4071 10. Kudo M, Finn RS, Qin S, Han KH, Ikeda K, Piscaglia F, et al. Lenvatinib Versus Sorafenib in First-Line Treatment of Patients With Unresectable Hepatocellular Carcinoma: A Randomised Phase 3 Non-Inferiority Trial. Lancet (London England) (2018) 391(10126):1163–73. doi: 10.1016/s0140- 6736(18)30207-1 20. Sangro B, Gomez-Martin C, de la Mata M, Iñarrairaegui M, Garralda E, Barrera P, et al. A Clinical Trial of CTLA-4 Blockade With Tremelimumab in Patients With Hepatocellular Carcinoma and Chronic Hepatitis C. J Hepatol (2013) 59(1):81–8. doi: 10.1016/j.jhep.2013.02.022 11. Bruix J, Qin S, Merle P, Granito A, Huang YH, Bodoky G, et al. Regorafenib for Patients With Hepatocellular Carcinoma Who Progressed on Sorafenib Treatment (RESORCE): A Randomised, Double-Blind, Placebo-Controlled, Phase 3 Trial. Lancet (London England) (2017) 389(10064):56–66. doi: 10.1016/s0140-6736(16)32453-9 21. Feun LG, Li YY, Wu C, Wangpaichitr M, Jones PD, Richman SP, et al. Phase 2 Study of Pembrolizumab and Circulating Biomarkers to Predict Anticancer Response in Advanced, Unresectable Hepatocellular Carcinoma. Cancer (2019) 125(20):3603–14. doi: 10.1002/cncr.32339 September 2021 \| Volume 11 \| Article 716844 Frontiers in Oncology \| www.frontiersin.org 10 He et al. Biomarkers of ICIs in HCC 22. Finn RS, Qin S, Ikeda M, Galle PR, Ducreux M, Kim TY, et al. Atezolizumab Plus Bevacizumab in Unresectable Hepatocellular Carcinoma. New Engl J Med (2020) 382(20):1894–905. doi: 10.1056/NEJMoa1915745 in Hepatocellular Carcinoma. Front Oncol (2018) 8:269. doi: 10.3389/ fonc.2018.00269 in Hepatocellular Carcinoma. Front Oncol (2018) 8:269. REFERENCES Shrestha R, Prithviraj P, Anaka M, Bridle KR, Crawford DHG, Dhungel B, et al. Monitoring Immune Checkpoint Regulators as Predictive Biomarkers September 2021 \| Volume 11 \| Article 716844 Frontiers in Oncology \| www.frontiersin.org 11 Biomarkers of ICIs in HCC He et al. Hepatocellular Carcinoma via the miR-30-5p/Snail/DPP4 Axis. Mol Cancer (2020) 19(1):92. doi: 10.1186/s12943-020-01213-6 60. Berger MF, Mardis ER. The Emerging Clinical Relevance of Genomics in Cancer Medicine. Nat Rev Clin Oncol (2018) 15(6):353–65. doi: 10.1038/ s41571-018-0002-6 Hepatocellular Carcinoma via the miR-30-5p/Snail/DPP4 Axis. Mol Cancer (2020) 19(1):92. doi: 10.1186/s12943-020-01213-6 78. Li Y, He X, Zhang X, Xu Y, Wu Y, Xu X. Immune-Related microRNA Signature for Predicting Prognosis and the Immune Microenvironment in Hepatocellular Carcinoma. Life Sci (2021) 265:118799. doi: 10.1016/ j.lfs.2020.118799 61. Pinyol R, Sia D, Llovet JM. Immune Exclusion-Wnt/CTNNB1 Class Predicts Resistance to Immunotherapies in HCC. Clin Cancer Research: an Off J Am Assoc Cancer Res (2019) 25(7):2021–3. doi: 10.1158/1078-0432.Ccr-18-3778 79. Mo Z, Cao Z, Luo S, Chen Y, Zhang S. Novel Molecular Subtypes Associated With 5mc Methylation and Their Role in Hepatocellular Carcinoma Immunotherapy. Front Mol Biosci (2020) 7:562441. doi: 10.3389/ fmolb.2020.562441 62. Sia D, Jiao Y, Martinez-Quetglas I, Kuchuk O, Villacorta-Martin C, Castro de Moura M, et al. Identiﬁcation of an Immune-Speciﬁc Class of Hepatocellular Carcinoma, Based on Molecular Features. Gastroenterology (2017) 153(3):812–26. doi: 10.1053/j.gastro.2017.06.007 63. Harding JJ, Nandakumar S, Armenia J, Khalil DN, Albano M, Ly M, et al. Prospective Genotyping of Hepatocellular Carcinoma: Clinical Implications of Next-Generation Sequencing for Matching Patients to Targeted and Immune Therapies. Clin Cancer Research: an Off J Am Assoc Cancer Res (2019) 25(7):2116–26. doi: 10.1158/1078-0432.Ccr-18-2293 80. De Carvalho DD, Sharma S, You JS, Su SF, Taberlay PC, Kelly TK, et al. DNA Methylation Screening Identiﬁes Driver Epigenetic Events of Cancer Cell Survival. Cancer Cell (2012) 21(5):655–67. doi: 10.1016/ j.ccr.2012.03.045 81. Yoshimi A, Lin KT, Wiseman DH, Rahman MA, Pastore A, Wang B, et al. Coordinated Alterations in RNA Splicing and Epigenetic Regulation Drive Leukaemogenesis. Nature (2019) 574(7777):273–7. doi: 10.1038/s41586- 019-1618-0 64. Ruiz de Galarreta M, Bresnahan E, Molina-Sánchez P, Lindblad KE, Maier B, Sia D, et al. b-Catenin Activation Promotes Immune Escape and Resistance to Anti-PD-1 Therapy in Hepatocellular Carcinoma. Cancer Discov (2019) 9 (8):1124–41. doi: 10.1158/2159-8290.Cd-19-0074 82. Orouji E, Federico A, Larribère L, Novak D, Lipka DB, Assenov Y, et al. Histone Methyltransferase SETDB1 Contributes to Melanoma Tumorigenesis and Serves as a New Potential Therapeutic Target. REFERENCES Int J Cancer (2019) 145(12):3462–77. doi: 10.1002/ijc.32432 65. von Felden J, Craig AJ, Garcia-Lezana T, Labgaa I, Haber PK, D’Avola D, et al. Mutations in Circulating Tumor DNA Predict Primary Resistance to Systemic Therapies in Advanced Hepatocellular Carcinoma. Oncogene (2021) 40(1):140–51. doi: 10.1038/s41388-020-01519-1 83. Giacopelli B, Zhao Q, Ruppert AS, Agyeman A, Weigel C, Wu YZ, et al. Developmental Subtypes Assessed by DNA methylation-iPLEX Forecast the Natural History of Chronic Lymphocytic Leukemia. Blood (2019) 134 (8):688–98. doi: 10.1182/blood.2019000490 66. Long J, Wang A, Bai Y, Lin J, Yang X, Wang D, et al. Development and Validation of a TP53-Associated Immune Prognostic Model for Hepatocellular Carcinoma. EBioMedicine (2019) 42:363–74. doi: 10.1016/ j.ebiom.2019.03.022 84. Brennan K, Shin JH, Tay JK, Prunello M, Gentles AJ, Sunwoo JB, et al. NSD1 Inactivation Deﬁnes an Immune Cold, DNA Hypomethylated Subtype in Squamous Cell Carcinoma. Sci Rep (2017) 7(1):17064. doi: 10.1038/s41598- 017-17298-x 67. Hu ZQ, Xin HY, Luo CB, Li J, Zhou ZJ, Zou JX, et al. Associations Among the Mutational Landscape, Immune Microenvironment, and Prognosis in Chinese Patients With Hepatocellular Carcinoma. Cancer Immunol Immunother: CII (2021) 70(2):377–89. doi: 10.1007/s00262-020-02685-7 85. Lingohr P, Dohmen J, Semaan A, Branchi V, Dietrich J, Bootz F, et al. Clinicopathological, Immune and Molecular Correlates of PD-L2 Methylation in Gastric Adenocarcinomas. Epigenomics (2019) 11(6):639– 53. doi: 10.2217/epi-2018-0149 68. Cai Y, Tian Y, Wang J, Wei W, Tang Q, Lu L, et al. Identiﬁcation of Driver Genes Regulating the T-Cell-Inﬁltrating Levels in Hepatocellular Carcinoma. Front Genet (2020) 11:560546. doi: 10.3389/fgene.2020.560546 69. Bassaganyas L, Pinyol R, Esteban-Fabró R, Torrens L, Torrecilla S, Willoughby CE, et al. Copy-Number Alteration Burden Differentially Impacts Immune Proﬁles and Molecular Features of Hepatocellular Carcinoma. Clin Cancer Research: An Off J Am Assoc Cancer Res (2020) 26(23):6350–61. doi: 10.1158/1078-0432.Ccr-20-1497 86. Liu L, Toung JM, Jassowicz AF, Vijayaraghavan R, Kang H, Zhang R, et al. Targeted Methylation Sequencing of Plasma Cell-Free DNA for Cancer Detection and Classiﬁcation. Ann Oncol: Off J Eur Soc Med Oncol (2018) 29 (6):1445–53. doi: 10.1093/annonc/mdy119 87. Fröhlich A, Loick S, Bawden EG, Fietz S, Dietrich J, Diekmann E, et al. Comprehensive Analysis of Tumor Necrosis Factor Receptor TNFRSF9 (4- 1BB) DNA Methylation With Regard to Molecular and Clinicopathological Features, Immune Inﬁltrates, and Response Prediction to Immunotherapy in Melanoma. EBioMedicine (2020) 52:102647. doi: 10.1016/j.ebiom.2020.102647 70. Yang H, Sun L, Guan A, Yin H, Liu M, Mao X, et al. REFERENCES Unique TP53 Neoantigen and the Immune Microenvironment in Long-Term Survivors of Hepatocellular Carcinoma. Cancer Immunol Immunother: CII (2021) 70 (3):667–77. doi: 10.1007/s00262-020-02711-8 71. Peng L, Chen Z, Chen Y, Wang X, Tang N. MIR155HG is a Prognostic Biomarker and Associated With Immune Inﬁltration and Immune Checkpoint Molecules Expression in Multiple Cancers. Cancer Med (2019) 8(17):7161–73. doi: 10.1002/cam4.2583 88. Duruisseaux M, Martı́nez-Cardús A, Calleja-Cervantes ME, Moran S, Castro de Moura M, Davalos V, et al. Epigenetic Prediction of Response to Anti- PD-1 Treatment in Non-Small-Cell Lung Cancer: A Multicentre, Retrospective Analysis. Lancet Respir Med (2018) 6(10):771–81. doi: 10.1016/s2213-2600(18)30284-4 72. Zhang Y, Zhang L, Xu Y, Wu X, Zhou Y, Mo J. Immune-Related Long Noncoding RNA Signature for Predicting Survival and Immune Checkpoint Blockade in Hepatocellular Carcinoma. J Cell Physiol (2020) 235(12):9304– 16. doi: 10.1002/jcp.29730 89. Zhang Y, Petropoulos S, Liu J, Cheishvili D, Zhou R, Dymov S, et al. The Signature of Liver Cancer in Immune Cells DNA Methylation. Clin Epigenet (2018) 10:8. doi: 10.1186/s13148-017-0436-1 90. Wang Y, Yang Y, Gao H, Ouyang T, Zhang L, Hu J, et al. Comprehensive Analysis of CDCAs Methylation and Immune Inﬁltrates in Hepatocellular Carcinoma. Front Oncol (2020) 10:566183. doi: 10.3389/fonc.2020.566183 73. Yuan M, Wang Y, Sun Q, Liu S, Xian S, Dai F, et al. Identiﬁcation of a Nine Immune-Related lncRNA Signature as a Novel Diagnostic Biomarker for Hepatocellular Carcinoma. BioMed Res Int (2021) 2021:9798231. doi: 10.1155/2021/9798231 91. Llopiz D, Ruiz M, Villanueva L, Iglesias T, Silva L, Egea J, et al. Enhanced Anti-Tumor Efﬁcacy of Checkpoint Inhibitors in Combination With the Histone Deacetylase Inhibitor Belinostat in a Murine Hepatocellular Carcinoma Model. Cancer Immunol Immunother: CII (2019) 68(3):379– 93. doi: 10.1007/s00262-018-2283-0 74. Xu Q, Wang Y, Huang W. Identiﬁcation of Immune-Related lncRNA Signature for Predicting Immune Checkpoint Blockade and Prognosis in Hepatocellular Carcinoma. Int Immunopharmacol (2021) 92:107333. doi: 10.1016/j.intimp.2020.107333 92. Ayers M, Lunceford J, Nebozhyn M, Murphy E, Loboda A, Kaufman DR, et al. IFN-g-Related mRNA Proﬁle Predicts Clinical Response to PD-1 Blockade. J Clin Invest (2017) 127(8):2930–40. doi: 10.1172/jci91190 75. Xiao S, Huang S, Yang J. Overexpression of GIHCG is Associated With a Poor Prognosis and Immune Inﬁltration in Hepatocellular Carcinoma. OncoTargets Ther (2020) 13:11607–19. doi: 10.2147/ott.S271966 76. Ye Y, Guo J, Xiao P, Ning J, Zhang R, Liu P, et al. Macrophages-Induced Long Noncoding RNA H19 Up-Regulation Triggers and Activates the miR-193b/ MAPK1 Axis and Promotes Cell Aggressiveness in Hepatocellular Carcinoma. REFERENCES Analysis of Genomes and Transcriptomes of Hepatocellular Carcinomas Identiﬁes Mutations and Gene Expression Changes in the Transforming Growth Factor-b Pathway. Gastroenterology (2018) 154(1):195–210. doi: 10.1053/j.gastro.2017.09.007 116. Gandara DR, Paul SM, Kowanetz M, Schleifman E, Zou W, Li Y, et al. Blood- Based Tumor Mutational Burden as a Predictor of Clinical Beneﬁt in Non- Small-Cell Lung Cancer Patients Treated With Atezolizumab. Nat Med (2018) 24(9):1441–8. doi: 10.1038/s41591-018-0134-3 100. Chen J, Gingold JA, Su X. Immunomodulatory TGF-b Signaling in Hepatocellular Carcinoma. Trends Mol Med (2019) 25(11):1010–23. doi: 10.1016/j.molmed.2019.06.007 117. Wang Z, Duan J, Cai S, Han M, Dong H, Zhao J, et al. Assessment of Blood Tumor Mutational Burden as a Potential Biomarker for Immunotherapy in Patients With Non-Small Cell Lung Cancer With Use of a Next-Generation Sequencing Cancer Gene Panel. JAMA Oncol (2019) 5(5):696–702. doi: 10.1001/jamaoncol.2018.7098 101. Mariathasan S, Turley SJ, Nickles D, Castiglioni A, Yuen K, Wang Y, et al. TGFb Attenuates Tumour Response to PD-L1 Blockade by Contributing to Exclusion of T Cells. Nature (2018) 554(7693):544–8. doi: 10.1038/ nature25501 102. Kelley RK, Gane E, Assenat E, Siebler J, Galle PR, Merle P, et al. A Phase 2 Study of Galunisertib (TGF-b1 Receptor Type I Inhibitor) and Sorafenib in Patients With Advanced Hepatocellular Carcinoma. Clin Trans Gastroenterol (2019) 10(7):e00056. doi: 10.14309/ctg.0000000000000056 118. Winograd P, Hou S, Court CM, Lee YT, Chen PJ, Zhu Y, et al. Hepatocellular Carcinoma-Circulating Tumor Cells Expressing PD-L1 Are Prognostic and Potentially Associated With Response to Checkpoint Inhibitors. Hepatol Commun (2020) 4(10):1527–40. doi: 10.1002/hep4.1577 103. Chen G, Huang AC, Zhang W, Zhang G, Wu M, Xu W, et al. Exosomal PD- L1 Contributes to Immunosuppression and is Associated With Anti-PD-1 Response. Nature (2018) 560(7718):382–6. doi: 10.1038/s41586-018-0392-8 119. Shao YY, Liu TH, Hsu C, Lu LC, Shen YC, Lin ZZ, et al. Early Alpha- Foetoprotein Response Associated With Treatment Efﬁcacy of Immune Checkpoint Inhibitors for Advanced Hepatocellular Carcinoma. Liver International: Off J Int Assoc Study Liver (2019) 39(11):2184–9. doi: 10.1111/liv.14210 104. Poggio M, Hu T, Pai CC, Chu B, Belair CD, Chang A, et al. Suppression of Exosomal PD-L1 Induces Systemic Anti-Tumor Immunity and Memory. Cell (2019) 177(2):414–27.e13. doi: 10.1016/j.cell.2019.02.016 105. Mazzaschi G, Minari R, Zecca A, Cavazzoni A, Ferri V, Mori C, et al. Soluble PD-L1 and Circulating CD8+PD-1+ and NK Cells Enclose a Prognostic and Predictive Immune Effector Score in Immunotherapy Treated NSCLC Patients. Lung Cancer (Amsterdam Netherlands) (2020) 148:1–11. doi: 10.1016/j.lungcan.2020.07.028 120. REFERENCES Cancer Lett (2020) 469:310–22. doi: 10.1016/j.canlet.2019.11.001 93. Gu X, Guan J, Xu J, Zheng Q, Chen C, Yang Q, et al. Model Based on Five Tumour Immune Microenvironment-Related Genes for Predicting Hepatocellular Carcinoma Immunotherapy Outcomes. J Trans Med (2021) 19(1):26. doi: 10.1186/s12967-020-02691-4 94. Peng Y, Liu C, Li M, Li W, Zhang M, Jiang X, et al. Identiﬁcation of a Prognostic and Therapeutic Immune Signature Associated With 77. Huang XY, Zhang PF, Wei CY, Peng R, Lu JC, Gao C, et al. Circular RNA circMET Drives Immunosuppression and Anti-PD1 Therapy Resistance in September 2021 \| Volume 11 \| Article 716844 Frontiers in Oncology \| www.frontiersin.org 12 Biomarkers of ICIs in HCC He et al. Hepatocellular Carcinoma. Cancer Cell Int (2021) 21(1):98. doi: 10.1186/ s12935-021-01792-4 Treated With Anti-PD-1 Therapy. Cancer Med (2020) 9(14):4962–70. doi: 10.1002/cam4.3135 Treated With Anti-PD-1 Therapy. Cancer Med (2020) 9(14):4962–70. doi: 10.1002/cam4.3135 95. Xu Y, Wang Z, Li F. Survival Prediction and Response to Immune Checkpoint Inhibitors: A Prognostic Immune Signature for Hepatocellular Carcinoma. Trans Oncol (2021) 14(1):100957. doi: 10.1016/j.tranon.2020.100957 113. Lipson EJ, Velculescu VE, Pritchard TS, Sausen M, Pardoll DM, Topalian SL, et al. Circulating Tumor DNA Analysis as a Real-Time Method for Monitoring Tumor Burden in Melanoma Patients Undergoing Treatment With Immune Checkpoint Blockade. J Immunother Cancer (2014) 2(1):42. doi: 10.1186/s40425-014-0042-0 96. Batlle E, Massagué J. Transforming Growth Factor-b Signaling in Immunity and Cancer. Immunity (2019) 50(4):924–40. doi: 10.1016/j.immuni.2019.03.024 97. Ringelhan M, Pﬁster D, O’Connor T, Pikarsky E, Heikenwalder M. The Immunology of Hepatocellular Carcinoma. Nat Immunol (2018) 19(3):222– 32. doi: 10.1038/s41590-018-0044-z 114. Guibert N, Mazieres J, Delaunay M, Casanova A, Farella M, Keller L, et al. Monitoring of KRAS-Mutated ctDNA to Discriminate Pseudo-Progression From True Progression During Anti-PD-1 Treatment of Lung Adenocarcinoma. Oncotarget (2017) 8(23):38056–60. doi: 10.18632/ oncotarget.16935 98. Park BV, Freeman ZT, Ghasemzadeh A, Chattergoon MA, Rutebemberwa A, Steigner J, et al. TGFb1-Mediated SMAD3 Enhances PD-1 Expression on Antigen-Speciﬁc T Cells in Cancer. Cancer Discov (2016) 6(12):1366–81. doi: 10.1158/2159-8290.Cd-15-1347 115. Lee JH, Long GV, Menzies AM, Lo S, Guminski A, Whitbourne K, et al. Association Between Circulating Tumor DNA and Pseudoprogression in Patients With Metastatic Melanoma Treated With Anti-Programmed Cell Death 1 Antibodies. JAMA Oncol (2018) 4(5):717–21. doi: 10.1001/ jamaoncol.2017.5332 99. Chen J, Zaidi S, Rao S, Chen JS, Phan L, Farci P, et al. REFERENCES Lee PC, Chao Y, Chen MH, Lan KH, Lee CJ, Lee IC, et al. Predictors of Response and Survival in Immune Checkpoint Inhibitor-Treated Unresectable Hepatocellular Carcinoma. Cancers (2020) 12(1):182. doi: 10.3390/cancers12010182 121. Spahn S, Roessler D, Pompilia R, Gabernet G, Gladstone BP, Horger M, et al. Clinical and Genetic Tumor Characteristics of Responding and Non- Responding Patients to PD-1 Inhibition in Hepatocellular Carcinoma. Cancers (2020) 12(12):3830. doi: 10.3390/cancers12123830 106. Chang B, Huang T, Wei H, Shen L, Zhu D, He W, et al. The Correlation and Prognostic Value of Serum Levels of Soluble Programmed Death Protein 1 (sPD-1) and Soluble Programmed Death-Ligand 1 (sPD-L1) in Patients With Hepatocellular Carcinoma. Cancer Immunol Immunother: CII (2019) 68(3):353–63. doi: 10.1007/s00262-018-2271-4 122. Matson V, Fessler J, Bao R, Chongsuwat T, Zha Y, Alegre ML, et al. The Commensal Microbiome is Associated With Anti-PD-1 Efﬁcacy in Metastatic Melanoma Patients. Sci (New York NY) (2018) 359(6371):104– 8. doi: 10.1126/science.aao3290 107. Finkelmeier F, Canli Ö, Tal A, Pleli T, Trojan J, Schmidt M, et al. High Levels of the Soluble Programmed Death-Ligand (sPD-L1) Identify Hepatocellular Carcinoma Patients With a Poor Prognosis. Eur J Cancer (Oxford England: 1990) (2016) 59:152–9. doi: 10.1016/j.ejca.2016.03.002 123. Routy B, Le Chatelier E, Derosa L, Duong CPM, Alou MT, Daillère R, et al. Gut Microbiome Inﬂuences Efﬁcacy of PD-1-Based Immunotherapy Against Epithelial Tumors. Sci (New York NY) (2018) 359(6371):91–7. doi: 10.1126/ science.aan3706 108. Han X, Gu YK, Li SL, Chen H, Chen MS, Cai QQ, et al. Pre-Treatment Serum Levels of Soluble Programmed Cell Death-Ligand 1 Predict Prognosis in Patients With Hepatitis B-Related Hepatocellular Carcinoma. J Cancer Res Clin Oncol (2019) 145(2):303–12. doi: 10.1007/s00432-018-2758-6 124. Pasolli E, Truong DT, Malik F, Waldron L, Segata N. Machine Learning Meta-Analysis of Large Metagenomic Datasets: Tools and Biological Insights. PloS Comput Biol (2016) 12(7):e1004977. doi: 10.1371/ journal.pcbi.1004977 109. Huang AC, Postow MA, Orlowski RJ, Mick R, Bengsch B, Manne S, et al. T- Cell Invigoration to Tumour Burden Ratio Associated With Anti-PD-1 Response. Nature (2017) 545(7652):60–5. doi: 10.1038/nature22079 125. Yu LX, Schwabe RF. The Gut Microbiome and Liver Cancer: Mechanisms and Clinical Translation. Nat Rev Gastroenterol Hepatol (2017) 14(9):527– 39. doi: 10.1038/nrgastro.2017.72 110. Agdashian D, ElGindi M, Xie C, Sandhu M, Pratt D, Kleiner DE, et al. The Effect of Anti-CTLA4 Treatment on Peripheral and Intra-Tumoral T Cells in Patients With Hepatocellular Carcinoma. Cancer Immunol Immunother: CII (2019) 68(4):599–608. doi: 10.1007/s00262-019-02299-8 126. REFERENCES Sex and Interleukin-6 are Prognostic Factors for Autoimmune Toxicity Following Treatment With Anti-CTLA4 Blockade. J Trans Med (2018) 16 (1):94. doi: 10.1186/s12967-018-1467-x 132. Zheng Y, Wang T, Tu X, Huang Y, Zhang H, Tan D, et al. Gut Microbiome Affects the Response to Anti-PD-1 Immunotherapy in Patients With Hepatocellular Carcinoma. J Immunother Cancer (2019) 7(1):193. doi: 10.1186/s40425-019-0650-9 150. Tarhini AA, Zahoor H, Lin Y, Malhotra U, Sander C, Butterﬁeld LH, et al. Baseline Circulating IL-17 Predicts Toxicity While TGF-b1 and IL-10 are Prognostic of Relapse in Ipilimumab Neoadjuvant Therapy of Melanoma. J Immunother Cancer (2015) 3:39. doi: 10.1186/s40425-015-0081-1 133. Rosshart SP, Vassallo BG, Angeletti D, Hutchinson DS, Morgan AP, Takeda K, et al. Wild Mouse Gut Microbiota Promotes Host Fitness and Improves Disease Resistance. Cell (2017) 171(5):1015–28.e13. doi: 10.1016/ j.cell.2017.09.016 Immunother Cancer (2015) 3:39. doi: 10.1186/s40425-015-0081-1 151. Diehl A, Yarchoan M, Hopkins A, Jaffee E, Grossman SA. Relationships Between Lymphocyte Counts and Treatment-Related Toxicities and Clinical Responses in Patients With Solid Tumors Treated With PD-1 Checkpoint Inhibitors. Oncotarget (2017) 8(69):114268–80. doi: 10.18632/ oncotarget.23217 134. Gopalakrishnan V, Helmink BA, Spencer CN, Reuben A, Wargo JA. The Inﬂuence of the Gut Microbiome on Cancer, Immunity, and Cancer Immunotherapy. Cancer Cell (2018) 33(4):570–80. doi: 10.1016/ j.ccell.2018.03.015 152. Kimbara S, Fujiwara Y, Iwama S, Ohashi K, Kuchiba A, Arima H, et al. Association of Antithyroglobulin Antibodies With the Development of Thyroid Dysfunction Induced by Nivolumab. Cancer Sci (2018) 109 (11):3583–90. doi: 10.1111/cas.13800 135. Tumeh PC, Harview CL, Yearley JH, Shintaku IP, Taylor EJ, Robert L, et al. PD-1 Blockade Induces Responses by Inhibiting Adaptive Immune Resistance. Nature (2014) 515(7528):568–71. doi: 10.1038/nature13954 153. Stamatouli AM, Quandt Z, Perdigoto AL, Clark PL, Kluger H, Weiss SA, et al. Collateral Damage: Insulin-Dependent Diabetes Induced With Checkpoint Inhibitors. Diabetes (2018) 67(8):1471–80. doi: 10.2337/dbi18- 0002 136. Inoue H, Park JH, Kiyotani K, Zewde M, Miyashita A, Jinnin M, et al. Intratumoral Expression Levels of PD-L1, GZMA, and HLA-A Along With Oligoclonal T Cell Expansion Associate With Response to Nivolumab in Metastatic Melanoma. Oncoimmunology (2016) 5(9):e1204507. doi: 10.1080/ 2162402x.2016.1204507 154. Chaput N, Lepage P, Coutzac C, Soularue E, Le Roux K, Monot C, et al. Baseline Gut Microbiota Predicts Clinical Response and Colitis in Metastatic Melanoma Patients Treated With Ipilimumab. Ann Oncol: Off J Eur Soc Med Oncol (2017) 28(6):1368–79. doi: 10.1093/annonc/mdx108 137. Postow MA, Manuel M, Wong P, Yuan J, Dong Z, Liu C, et al. REFERENCES Dapito DH, Mencin A, Gwak GY, Pradere JP, Jang MK, Mederacke I, et al. Promotion of Hepatocellular Carcinoma by the Intestinal Microbiota and TLR4. Cancer Cell (2012) 21(4):504–16. doi: 10.1016/j.ccr.2012.02.007 111. Hung YP, Shao YY, Lee JM, Hsu C, Hsu CH, Yang MH, et al. Potential of Circulating Immune Cells as Biomarkers of Nivolumab Treatment Efﬁcacy for Advanced Hepatocellular Carcinoma. J Chin Med Association: JCMA (2021) 84(2):144–50. doi: 10.1097/jcma.0000000000000477 127. Loo TM, Kamachi F, Watanabe Y, Yoshimoto S, Kanda H, Arai Y, et al. Gut Microbiota Promotes Obesity-Associated Liver Cancer Through PGE(2)- Mediated Suppression of Antitumor Immunity. Cancer Discov (2017) 7 (5):522–38. doi: 10.1158/2159-8290.Cd-16-0932 112. Dharmapuri S, Özbek U, Lin JY, Sung M, Schwartz M, Branch AD, et al. Predictive Value of Neutrophil to Lymphocyte Ratio and Platelet to Lymphocyte Ratio in Advanced Hepatocellular Carcinoma Patients 128. Ma C, Han M, Heinrich B, Fu Q, Zhang Q, Sandhu M, et al. Gut Microbiome-Mediated Bile Acid Metabolism Regulates Liver Cancer via September 2021 \| Volume 11 \| Article 716844 Frontiers in Oncology \| www.frontiersin.org 13 Biomarkers of ICIs in HCC He et al. NKT Cells. Sci (New York NY) (2018) 360(6391):eaan5931. doi: 10.1126/ science.aan5931 146. Oh S, Park Y, Lee HJ, Lee J, Lee SH, Baek YS, et al. A Disintegrin and Metalloproteinase 9 (ADAM9) in Advanced Hepatocellular Carcinoma and Their Role as a Biomarker During Hepatocellular Carcinoma Immunotherapy. Cancers (2020) 12(3):745. doi: 10.3390/cancers12030745 129. Arpaia N, Campbell C, Fan X, Dikiy S, van der Veeken J, deRoos P, et al. Metabolites Produced by Commensal Bacteria Promote Peripheral Regulatory T-Cell Generation. Nature (2013) 504(7480):451–5. doi: 10.1038/nature12726 147. Qayyum A, Hwang KP, Stafford J, Verma A, Maru DM, Sandesh S, et al. Immunotherapy Response Evaluation With Magnetic Resonance Elastography (MRE) in Advanced HCC. J Immunother Cancer (2019) 7 (1):329. doi: 10.1186/s40425-019-0766-y 130. Viaud S, Saccheri F, Mignot G, Yamazaki T, Daillère R, Hannani D, et al. The Intestinal Microbiota Modulates the Anticancer Immune Effects of Cyclophosphamide. Sci (New York NY) (2013) 342(6161):971–6. doi: 10.1126/science.1240537 148. Nakamura Y. Biomarkers for Immune Checkpoint Inhibitor-Mediated Tumor Response and Adverse Events. Front Med (2019) 6:119. doi: 10.3389/fmed.2019.00119 131. Sivan A, Corrales L, Hubert N, Williams JB, Aquino-Michaels K, Earley ZM, et al. Commensal Biﬁdobacterium Promotes Antitumor Immunity and Facilitates Anti-PD-L1 Efﬁcacy. Sci (New York NY) (2015) 350 (6264):1084–9. doi: 10.1126/science.aac4255 149. Valpione S, Pasquali S, Campana LG, Piccin L, Mocellin S, Pigozzo J, et al. REFERENCES Peripheral T Cell Receptor Diversity is Associated With Clinical Outcomes Following Ipilimumab Treatment in Metastatic Melanoma. J Immunother Cancer (2015) 3:23. doi: 10.1186/s40425-015-0070-4 155. Rogado J, Sánchez-Torres JM, Romero-Laorden N, Ballesteros AI, Pacheco- Barcia V, Ramos-Levı́ A, et al. Immune-Related Adverse Events Predict the Therapeutic Efﬁcacy of Anti-PD-1 Antibodies in Cancer Patients. Eur J Cancer (Oxford England: 1990) (2019) 109:21–7. doi: 10.1016/ j.ejca.2018.10.014 138. Kobayashi E, Mizukoshi E, Kishi H, Ozawa T, Hamana H, Nagai T, et al. A New Cloning and Expression System Yields and Validates TCRs From Blood Lymphocytes of Patients With Cancer Within 10 Days. Nat Med (2013) 19 (11):1542–6. doi: 10.1038/nm.3358 156. Freeman-Keller M, Kim Y, Cronin H, Richards A, Gibney G, Weber JS. Nivolumab in Resected and Unresectable Metastatic Melanoma: Characteristics of Immune-Related Adverse Events and Association With Outcomes. Clin Cancer Research: an Off J Am Assoc Cancer Res (2016) 22 (4):886–94. doi: 10.1158/1078-0432.Ccr-15-1136 139. Jokinen E, Huuhtanen J, Mustjoki S, Heinonen M, Lähdesmäki H. Predicting Recognition Between T Cell Receptors and Epitopes With TCRGP. PloS Comput Biol (2021) 17(3):e1008814. doi: 10.1371/ journal.pcbi.1008814 140. Lin KR, Deng FW, Jin YB, Chen XP, Pan YM, Cui JH, et al. T Cell Receptor Repertoire Proﬁling Predicts the Prognosis of HBV-Associated Hepatocellular Carcinoma. Cancer Med (2018) 7(8):3755–62. doi: 10.1002/ cam4.1610 157. Teraoka S, Fujimoto D, Morimoto T, Kawachi H, Ito M, Sato Y, et al. Early Immune-Related Adverse Events and Association With Outcome in Advanced Non-Small Cell Lung Cancer Patients Treated With Nivolumab: A Prospective Cohort Study. J Thoracic Oncol: Off Publ Int Assoc Study Lung Cancer (2017) 12(12):1798–805. doi: 10.1016/j.jtho.2017.08.022 141. Han Y, Liu X, Wang Y, Wu X, Guan Y, Li H, et al. Identiﬁcation of Characteristic TRB V Usage in HBV-Associated HCC by Using Differential Expression Proﬁling Analysis. Oncoimmunology (2015) 4(8):e1021537. doi: 10.1080/2162402x.2015.1021537 Conﬂict of Interest: The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. 142. Pﬁster D, Núñez NG, Pinyol R, Govaere O, Pinter M, Szydlowska M, et al. NASH Limits Anti-Tumour Surveillance in Immunotherapy-Treated HCC. Nature (2021) 592(7854):450–6. doi: 10.1038/s41586-021-03362-0 Publisher’s Note: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their afﬁliated organizations, or those of the publisher, the editors and the reviewers. REFERENCES Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. 143. Terry S, Savagner P, Ortiz-Cuaran S, Mahjoubi L, Saintigny P, Thiery JP, et al. New Insights Into the Role of EMT in Tumor Immune Escape. Mol Oncol (2017) 11(7):824–46. doi: 10.1002/1878-0261.12093 144. Liu F, Qin L, Liao Z, Song J, Yuan C, Liu Y, et al. Microenvironment Characterization and Multi-Omics Signatures Related to Prognosis and Immunotherapy Response of Hepatocellular Carcinoma. Exp Hematol Oncol (2020) 9:10. doi: 10.1186/s40164-020-00165-3 Copyright © 2021 He, Lu, Che, Chu, Zhang and Chen. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. 145. Ding H, Hu H, Tian F, Liang H. A Dual Immune Signature of CD8+ T Cells and MMP9 Improves the Survival of Patients With Hepatocellular Carcinoma. Bioscience Rep (2021) 41(3):BSR20204219. doi: 10.1042/ bsr20204219 September 2021 \| Volume 11 \| Article 716844 Frontiers in Oncology \| www.frontiersin.org 14
https://openalex.org/W4362472541	https://www.frontiersin.org/articles/10.3389/fpls.2023.1154051/pdf	English	null	Performance valuation of onion (Allium cepa L.) genotypes under different levels of salinity for the development of cultivars suitable for saline regions	Frontiers in plant science	2,023	cc-by	15,100	Md. Ashraful Alam 1, Md. Atikur Rahman 2, Md. Marufur Rahman 3, Md. Mahmudul Hasan 4, Shamsun Naher 2, Abu Hena Faisal Fahim 5, Md. Abdul Mottalib 6, Srabanti Roy 7, Md. Raﬁqul Islam 8, Shailendra Nath Mozumder 4, Amnah Mohammed Alsuhaibani 9, Ahmed Gaber 10 and Akbar Hossain 11 Md. Ashraful Alam 1, Md. Atikur Rahman 2, Md. Marufur Rahman 3, Md. Mahmudul Hasan 4, Shamsun Naher 2, Abu Hena Faisal Fahim 5, Md. Abdul Mottalib 6, Srabanti Roy 7, Md. Raﬁqul Islam 8, Shailendra Nath Mozumder 4, Amnah Mohammed Alsuhaibani 9, Ahmed Gaber 10 and Akbar Hossain 11 Md. Ashraful Alam 1, Md. Atikur Rahman 2, Md. Marufur Rahman 3, Md. Mahmudul Hasan 4, Shamsun Naher 2, Abu Hena Faisal Fahim 5, Md. Abdul Mottalib 6, Srabanti Roy 7, Md. Raﬁqul Islam 8, Shailendra Nath Mozumder 4, Amnah Mohammed Alsuhaibani 9, Ahmed Gaber 10 and Akbar Hossain 11 1Plant Breeding Division, Spices Research Centre, Bangladesh Agricultural Research Institute, Bogura, Bangladesh, 2Division of Soil Science, Spices Research Centre, Bangladesh Agricultural Research Institute, Bogura, Bangladesh, 3Regional Station, Bangladesh Institute of Research and Training on Applied Nutrition, Rangpur, Bangladesh, 4Division of Horticulture, Spices Research Centre, Bangladesh Agricultural Research Institute, Bogura, Bangladesh, 5Division of Agronomy, Spices Research Centre, Bangladesh Agricultural Research Institute, Bogura, Bangladesh, 6Division of Agricultural Engineering, Spices Research Centre, Bangladesh Agricultural Research Institute, Bogura, Bangladesh, 7Department of Agricultural Chemistry, Bangladesh Agricultural University, Mymensingh, Bangladesh, 8Division of Agronomy, Regional Agricultural Research Station, Bangladesh Agricultural Research Institute (BARI), Ishwardi, Pabna, Bangladesh, 9Department of Physical Sport Science, College of Education, Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia, 10Department of Biology, College of Science, Taif University, Taif, Saudi Arabia, 11Division of Soil Science, Bangladesh Wheat and Maize Research Institute, Dinajpur, Bangladesh CITATION Alam MA, Rahman MA, Rahman MM, Hasan MM, Naher S, Fahim AHF, Mottalib MA, Roy S, Islam MR, Mozumder SN, Alsuhaibani AM, Gaber A and Hossain A (2023) Performance valuation of onion (Allium cepa L.) genotypes under different levels of salinity for the development of cultivars suitable for saline regions. Front. Plant Sci. 14:1154051. doi: 10.3389/fpls.2023.1154051 Abiotic stress, especially salt stress, is one of the major barriers to crop production worldwide. Crops like onion that belong to the glycophytic group are more sensitive to salinity stress. A huge study regarding the inﬂuence of salinity stress on the growth and development of crops has already been done and is still ongoing. One of the major targets of the research is to develop genotypes that have enhanced performance under stress environments. The world needs more of these types of genotypes to combat the ever-growing salt- stressed soils. frontiersin.org Frontiers in Plant Science OPEN ACCESS OPEN ACCESS EDITED BY Bhumi Nath Tripathi, Indira Gandhi National Tribal University, India REVIEWED BY Owais Ali Wani, Sher-e-Kashmir University of Agricultural Sciences and Technology of Kashmir, India Zahoor Ahmad, University of Central Punjab, Pakistan CORRESPONDENCE Md. Ashraful Alam a.alam_83@yahoo.com Akbar Hossain akbarhossainwrc@gmail.com SPECIALTY SECTION This article was submitted to Plant Abiotic Stress, a section of the journal Frontiers in Plant Science RECEIVED 30 January 2023 ACCEPTED 13 March 2023 PUBLISHED 31 March 2023 CITATION Alam MA, Rahman MA, Rahman MM, Hasan MM, Naher S, Fahim AHF, Mottalib MA, Roy S, Islam MR, Mozumder SN, Alsuhaibani AM, Gaber A and Hossain A (2023) Performance valuation of onion (Allium cepa L.) genotypes under different levels of salinity for the development of cultivars suitable for saline regions. Front. Plant Sci. 14:1154051. doi: 10.3389/fpls.2023.1154051 COPYRIGHT © 2023 Alam, Rahman, Rahman, Hasan, Naher, Fahim, Mottalib, Roy, Islam, EDITED BY Bhumi Nath Tripathi, Indira Gandhi National Tribal University, India 1 Introduction production as well (Shoaib et al., 2018; Regessa et al., 2022; Sanwal et al., 2022; Venâncio et al., 2022). Germination and emergence become difﬁcult (Khan, 2003; Regessa et al., 2010; Hanci and Cebeci, 2015; Ullah and Bano, 2019) and subsequent yield reduction occurs (Chinnusamy et al., 2005) for glycophytes under saline-affected soils (Hanci et al., 2016), although they have a different threshold level of salinity, such as the onion, which is very sensitive to salinity beyond 1.2 dS m−1 (Maas and Hoffman, 1977). production as well (Shoaib et al., 2018; Regessa et al., 2022; Sanwal et al., 2022; Venâncio et al., 2022). Germination and emergence become difﬁcult (Khan, 2003; Regessa et al., 2010; Hanci and Cebeci, 2015; Ullah and Bano, 2019) and subsequent yield reduction occurs (Chinnusamy et al., 2005) for glycophytes under saline-affected soils (Hanci et al., 2016), although they have a different threshold level of salinity, such as the onion, which is very sensitive to salinity beyond 1.2 dS m−1 (Maas and Hoffman, 1977). Onion (Allium cepa L.) is one of the most important vegetables and spice crops. It is one of the most popular vegetables around the globe, carrying a pungent ﬂavor and often used as a condiment to prepare multiple delicious cuisines in every corner of the world. It has important medicinal properties to combat several diseases especially blood pressure and heart disease. Two-thirds of the total onion production come from Asia, among which India and China hold the major share (FAOSTAT, [[NoYear]]). Bangladesh ranks third in the list in terms of production (Star Business Report, 2022). Bangladesh produced 19.54 lakh M tons of onion bulbs from 1.85 lakh ha of land in 2020, with an average yield of 10.55 t/ha (FAOSTAT, [[NoYear]]), which is very low compared to other countries. Onion ranks the highest among the spice crops in Bangladesh based on production and generally grows all over the country mostly in the winter season. It is an important ingredient in many food preparations and is mostly used as a spice rather than as a vegetable in different daily dishes. Although the country is producing a lot of onion, it is still has a huge shortage and, thus, has to import a large amount from abroad to meet the domestic demand (BBS, 2019). 1 Introduction In Bangladesh, onion is being grown all over the country, but production is hampered in saline-prone areas around the coastal belt as it is a glycophytic crop. A total of 1.06 million ha of land area (32% of the total coastal and offshore land) in the country is affected by different degrees of salinity (Ahsan and Bhuiyan, 2010; SRDI, 2010; Parvin et al., 2017). Soil salinity was classiﬁed by Soil Research Development Institute (SRDI) (2010) as non-saline (2.0–4.0 dS m−1), very slightly saline (4.1–8.0 dS m−1), moderately saline (8.1– 12.0 dS m−1), strongly saline (12.1–16.0 dS m−1), and very strongly saline (>16 dS m−1), which occupied approximately 0.328 (31%), 0.274 (26%), 0.190 (18%), 0.162 (15%), and 0.102 (10%) m ha of land, respectively (Ahsan and Bhuiyan, 2010; SRDI, 2010). During the growing season, salinity level varies between 6 and 12 dS m−1, which reaches up to 20 dS m−1 in extreme cases (SRDI, 2010). Thus, the cropping intensity is low in the coastal areas compared to the national average. A substantial amount of land has always remained fallow in coastal areas during the winter season after Aman rice (wet season rice) cultivation due to salinity problems. Onion has the potential as a cash crop to ﬁt in this area to increase cropping intensity and save foreign currency by reducing the import of the crop, if cultivated after harvesting T. Aman rice. Moreover, onion cultivation in the mainland has almost plateaued, thus requiring a higher production to meet the shortage. Public research institutes in the country developed several varieties of onion suitable for both winter and rainy seasons, mainly on the mainland. However, those also suffer from salinity stress when cultivated in coastal areas. Therefore, the yield of onion in this area is very low compared to other parts of the country. Hence, genotypes suitable for cultivation under low to moderate levels of salinity level are a crying need for this zone. The potential genotype tolerant to salt stress will increase To feed the ever-increasing population, food production by irrigation is common in arid and semiarid regions, resulting in 20% to 50% of the land being affected by salinity called secondary salinization, bringing unprecedented agricultural losses over time (Pitman and Läuchli, 2006). Md. Ashraful Alam 1, Md. Atikur Rahman 2, Md. Marufur Rahman 3, Md. Mahmudul Hasan 4, Shamsun Naher 2, Abu Hena Faisal Fahim 5, Md. Abdul Mottalib 6, Srabanti Roy 7, Md. Raﬁqul Islam 8, Shailendra Nath Mozumder 4, Amnah Mohammed Alsuhaibani 9, Ahmed Gaber 10 and Akbar Hossain 11* Therefore, a number of germplasm were studied during the 2019– 2020 and 2020–2021 seasons under different salt concentrations to identify tolerant genotypes as well as to study the plants’ responses at different growth stages against elevated salinity levels. A 2-year study was conducted where germination potential was evaluated in the ﬁrst year and carried out in petri dish culture of seeds, followed by plastic pot culture for plant establishment and bulb development evaluation during the second year. Four different saline water solutions having different salt concentrations (0, 8, 10, and 12 dS m−1) were applied to the petri dishes and pots as the source of water for plants in both seasons. Results indicated that a signiﬁcant reduction in plants’ performance occurs under higher salinity levels. Salt concentration had an adverse impact on germination, leaf development and growth, the height of plants, bulb size and shape, and the bulb weight of onion. All the growth phases of onion are sensitive COPYRIGHT © 2023 Alam, Rahman, Rahman, Hasan, Naher, Fahim, Mottalib, Roy, Islam, Mozumder, Alsuhaibani, Gaber and Hossain. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Frontiers in Plant Science 01 frontiersin.org Alam et al. 10.3389/fpls.2023.1154051 10.3389/fpls.2023.1154051 10.3389/fpls.2023.1154051 to elevated concentrations. Variable performances were observed in the genotypes under stress conditions, and a few genotypes (Ac Bog 409, Ac Bog 414, Ac Bog 424, Ac Bog 430, Ac Bog 417, Ac Bog 419, Ac Bog 420, Ac Bog 422, and Ac Bog 425) having some sort of tolerance to salt stress were identiﬁed, which might be recommended for mass production. Tolerance indices could successfully be applied in selecting the salt-tolerant genotypes. Thus, the present ﬁndings and the identiﬁed genotypes could be further utilized in salt stress improvement research on onion. Frontiers in Plant Science KEYWORDS salinity, onion, STI, stress, bulb growth, germination frontiersin.org salinity, onion, STI, stress, bulb growth, germination 2.5 Experimental plan In the ﬁrst season, all the genotypes were tested for their ability to germinate and subsequent plumule development under different salinity treatments. Autoclaved petri dishes were used, where blotting papers (Whatman no. 1) were placed. Then, seeds of the studied genotypes were placed in different petri dishes. Four levels of saline water were applied for each genotype to indulge the germination process. Spraying of saline water as per treatments was applied every other day. The percentage of germination and growth repression was observed after 7 days. Frontiers in Plant Science 2.2 Properties of soil used in the study Physicochemical analysis of initial soil under ﬁeld conditions was carried out at the central soil science laboratory of BARI. The soil employed in the study had a sandy loam texture and was slightly neutral in nature in response, with a ﬁeld capacity of 29.6% and a pH of 6.0, and organic matter percentage was low (1.33), having an average EC of 2.06 dS m−1. Total nitrogen (0.07%) was very low, available phosphorus (40.77 μg/g soil) was very high, exchangeable potassium was low (0.15 meq/100 g soil), and available sulfur was low (10.57 μg/g soil). Available iron (85.91 μg/g soil) was very high, available zinc (1.63 μg/g soil) was optimum, available boron (0.2 μg/ g soil) was low, available manganese (18.21 μg/g soil) was very high, available copper was very high (2.31 μg/g soil), exchangeable calcium (4.70 meq/100 g soil) was optimum, and exchangeable magnesium (1.59 meq/100 g soil) was high. 2.6 Observations recorded Various morpho-physiological trait observations were recorded using a standard protocol for onion phenotyping. The germination percentage was recorded after 7 days of seed placement on petri dishes. Subsequent growth repression (further growth and development arrested despite seed germination) was observed after 14 days of seed placement on each petri dish. The number of plants with dried leaves, number of green plants, maximum leaf length (MLL), and number of total leaves were counted on each pot basis. Individual bulb weight (IBW), bulb length (BL), and bulb diameter (BD) were recorded at harvest. Data on soil salinity level were observed eight times during the entire crop cycle from the seedling to the harvesting stage with an EC meter (model: HI 993310) after 15 days of each irrigation as indicated by Slavich and Petterson (1993) (Slavich and Petterson, 1993). 2.3 Genotypes used in the study A set of 25 onion genotypes was included in the current study; details of the studied genotypes are presented in Table S1 (Khan et al., 2022). A local cultivar BARI Piaz-4 released from a public research institute (BARI) was incorporated as a check cultivar. 2.1 Location of the current study The present study was carried out at the Spices Research Centre, BARI, Bogura during two consecutive winter seasons (Rabi): 2019 and 2020. The details of prevailing weather conditions during the pot experiment are given in Table S1. During the second season, the genotypes under study were accommodated in small-sized pots placed in the ﬁeld. At ﬁrst, seeds of the different genotypes were sown on a well-prepared seed bed to grow seedlings. The seedlings were then transplanted into the pot and kept in the ﬁeld at 35 days after sowing. Uniform sandy loam soil was collected from AEZ-4 and then recommended doses of compost and chemical fertilizers were incorporated for this experiment. The experimental pot was fabricated by biodegradable plastic materials having an internal volume of 500 ml. Each pot was ﬁlled with 400 g of soil mixture. A 15 cm × 10 cm spacing was maintained from row to row and hill to hill while arranging pots in the ﬁeld. Starting from transplanting, irrigations were applied with particular saline solutions to reach the ﬁeld capacity of soil as per aforesaid treatments. Other intercultural practices were followed as and when required, in which timely irrigation was provided to ensure moisture availability and plant protection measures were taken to repel pest and disease infestation. 1 Introduction A similar statement was also concluded in a prediction that highlights that approximately 50% of today’s arable land worldwide would be lost from agricultural use due to the worse effect of salinity by 2050 (Wang et al., 2003). It is estimated that there is a loss of more than US$12 billion per year worldwide due to salinity-induced agricultural input losses (Shabala, 2013). Soil salinity is one of the harsh outputs of global climate change and has an immense impact on arable land, especially coastal agricultural land (Qadir et al., 2014; Rahman et al., 2018). Soil degradation by salinization is one of the consequences of climate change caused by natural and anthropogenic activities (Yeo, 1998). An elevated salinity level adversely affects the morphology, physiology, and yield of a crop and is similar to the case of onion 02 frontiersin.org Alam et al. 10.3389/fpls.2023.1154051 10.3389/fpls.2023.1154051 ﬁeld and also laid out in a CRD by repeating all treatments two times. the total onion production in the country and improve the socio- economic condition of farmers. It will also play a role in increasing the onion cultivation area in the coastal belt. By keeping the above view, the present study was hypothesized as an attempt to evaluate and identify suitable genotypes that have a tolerance to moderate salinity (8–12 dS m−1) levels to increase onion production in saline- prone regions. 2.4 Treatments and design Four different salinity levels were applied in the present experiment (0, 8, 10, and 12 dS m−1) for both petri dish and pot experiments. The saline solutions were prepared using normal NaCl salt following the method recommended by Yaron and Mokady (1962). In the ﬁrst year, the observation was done in petri dishes by arranging all treatments in a completely randomized design (CRD) and repeated three times. In the second-year experiment, all genotypes were accommodated in small-sized pots placed in the frontiersin.org 2.7 Statistical analysis All the observed data were subjected to statistical analysis following the standard formulas. The Data Analysis tool of the 03 frontiersin.org 10.3389/fpls.2023.1154051 Alam et al. after subjecting all levels of salinity (0, 8, 10, and 12 dS m−1), germination percentage was found to range from 38% to 97.50%. Microsoft Excel program was used to estimate the statistical parameters. CV% was estimated based on output from the analyzed results. Shoot tolerance index (ShTI), stress tolerance index (STI), and percent yield reduction (PYR) were calculated on the MS Excel program following the formula given below (Fischer and Maurer, 1978; Choukan et al., 2006; Takahashi et al., 2015; Zafar et al., 2015; Guellim et al., 2019). g p g g Under all levels of salinity (8, 10, and 12 dS m−1), the growth of germinated onion bulb was repressed in a signiﬁcant way (Table 1) compared to untreated control as it ranged from 1% to 6% and 1% to 22% in control and 12 dS m−1 treatment, respectively. After imposing 8 dS m−1 salinity, the lowest growth repression % was recorded in Ac Gaz 379 (0) and the highest was recorded in Ac Bog 422 (9). After exerting 10 dS m−1 salinity, the lowermost growth repression % was noted in Ac Gaz 418 (3), Ac Gaz 420 (3), and Ac Gaz 421 (3), and the highest was noted in Ac Bog 429 (20) and BARI Piaz-4 (20). When 12 dS m−1 salinity was applied, the lowest growth repression % was found in Ac Gaz 379 (1) and the highest was found in Ac Bog 416 (22). Considering the mean values of all treatments (0, 8, 10, and 12 dS m−1), the lowest growth repression % was observed in Ac Gaz 379 (1.5) and the highest was observed in BARI Piaz-4 (11.25). The ShTI was estimated according to the following equation: ShTI = MLL at different dS=m MLL at 0 dS=m 100 …………………… ½i The STI was estimated according to the following equation: STI = IBW at 0 dS=m x IBW at different dS=m (Grand mean of IBW)2 STI = IBW at 0 dS=m x IBW at different dS=m (Grand mean of IBW)2 ½ii The PYR was estimated according to the following equation: From Figure 1, it was revealed that the germination percentage gradually decreased with the increase of irrigation water salinity concentration. PYR = IBW at 0 dS=m − IBW at different dS=m IBW at 0 dS=m 100 Regression analysis was performed to determine the extent of the relationship between IBW and salinity levels at different phases. Stepwise regression was also performed to ﬁnd out the critical phase for salinity stress to IBW. Regression analysis and visualization were performed by using the “ggplot2” package (Wickham, 2009) in the “R” platform (R Core Team R, 2021). 3.2.1 Analysis of variance Variance analysis was carried out among the studied traits of the second-year study and substantial variations (p< 0.01) were observed (Table S2). Variance due to genotypes (s2g) and salinity levels (s2s) were signiﬁcant for all the studied traits, while variance due to genotypes:salinity level (s2g×s) was signiﬁcant for all the traits except MLL. 2.7 Statistical analysis In contrast, the subsequent growth repression among the genotypes was minimum in the control treatment (0 dS m−1), and it was increased along the salinity level. Ultimately, the highest percentage of seized growth or death of germinated seed was observed at the 12 dS m−1 treatment. 3.2.2 Progression of soil salinity 10.3389/fpls.2023.1154051 TABLE 1 Scanning the germination ability of onion genotypes at different salinity stress levels in the ﬁrst year. 3.2.2 Progression of soil salinity Genotype Germination percentage (%) Growth repressed (%) C (0 dS m−1) 8 dS m−1 10 dS m−1 12 dS m−1 Mean C (0 dS m−1) 8 dS m−1 10 dS m−1 12 dS m−1 Mean Ac Bog 409 90 66 62 44 65.5 6 14 24 42 21.5 Ac Bog 410 74 64 64 56 64.5 8 8 20 18 13.5 Ac Bog 411 78 62 44 50 58.5 8 6 26 24 16 Ac Bog 412 86 50 66 26 57 8 12 24 22 16.5 Ac Bog 414 86 66 48 56 64 4 16 16 24 15 Ac Bog 415 64 68 48 48 57 6 16 22 18 15.5 Ac Bog 416 84 74 76 36 67.5 6 6 26 44 20.5 Ac Bog 417 60 68 80 66 68.5 8 14 26 36 21 Ac Bog 418 96 90 36 76 74.5 4 4 6 4 4.5 Ac Bog 419 86 58 70 52 66.5 8 16 14 20 14.5 Ac Bog 420 90 74 52 56 68 4 8 6 12 7.5 Ac Bog 421 94 72 68 50 71 4 8 6 24 10.5 Ac Bog 422 90 78 48 54 67.5 4 18 10 22 13.5 Ac Bog 423 92 72 72 72 77 2 8 22 12 11 Ac Bog 424 84 42 52 72 62.5 6 12 28 12 14.5 Ac Bog 425 82 74 64 74 73.5 4 6 10 16 9 Ac Bog 426 82 66 54 52 63.5 12 8 12 26 14.5 Ac Bog 427 74 60 60 46 60 8 18 12 22 15 Ac Bog 428 92 74 46 46 64.5 8 12 30 12 15.5 Ac Bog 429 82 58 72 44 64 6 12 40 14 18 Ac Bog 430 86 72 74 54 71.5 4 16 16 16 13 Ac Bog 431 80 70 44 70 66 6 4 14 16 10 Ac Bog 432 84 82 50 56 68 2 4 10 18 8.5 Ac Gaz 379 100 98 92 100 97.5 2 0 8 2 3 BARI Piaz-4 66 36 36 14 38 8 16 40 26 22.5 Mean 83.28 67.76 59.12 54.8 66.24 5.84 10.48 18.72 20.08 13.78 SD 9.8 13.28 14.34 17.28 2.44 5.1 9.86 9.96 Min 60 36 36 14 2 0 6 2 Max 100 98 92 100 12 18 40 44 SD standard deviation; Min minimum; Max maximum; C control treatment 0 dS m−1 3.2.2 Progression of soil salinity Soil salinity markedly inﬂuenced the germination percentage of onion (Table 1). Germination percentage was recorded the highest in Ac Gaz 379 (98%) followed by Ac Bog 418 (90%), Ac Bog 432 (84%), and Ac Bog 422 (78%) compared to the rest of the germplasm, whereas the lowest germination (36%) was recorded in BARI Piaz-4 (check cultivar) under 8 dS m−1. Under normal water treatment, germination percentage was found to be higher in Ac Gaz 379 (100%), Ac Bog 428 (92%), Ac Bog 423 (92%), Ac Bog 422 (90%), Ac Bog 421 (94%), Ac Bog 420 (90%), Ac Bog 422 (96%), and Ac Bog 409 (90%) than the rest of the germplasm, and the lowest germination (60%) was recorded in Ac Bog 417. Under 10 dS m−1, germination percentage ranged from 36% (Ac Bog 418) to 92% (Ac Gaz 379). On the other hand, the highest germination percentage was recorded in Ac Gaz 379 (100%), which was followed by Ac Bog 418 (76%), Ac Bog 423 (72%), and Ac Bog 424 (72%), and the lowest germination was recorded in BARI Piaz- 4, which was only 14% under the 12 dS m−1 salinity level. Finally, The studied soil was non-saline during the entire growing period of onion as the salinity range of soil under untreated control ranged from 1.67 to 2.58 dS m−1 (Table S4). In contrast, soil salinity was gradually built up with the advancement of different phases of onion plants depending on the concentration of salt in the solution (0, 8, 10, and 12 dS m−1) after eight spells of application (Tables S5-S7). After the application of irrigation water having 8, 10, and 12 dS m−1 of salinity, it was found that minimum levels of ECs (electrical conductivity in pot soil) were recorded in the ﬁrst phase and maximum levels were recorded in the eighth phase. For example, 3.14, 2.57, and 3.33 dS m−1 were recorded from the ﬁrst phase of 8, 10, and 12 dS m−1 treatments, respectively, whereas 6.92, 9.18, and 11.26 dS m−1 were obtained from 8, 10, and 12 dS m−1 treatments at the eighth phase, respectively. In the last phase of crops, soil salinity was increased by 168.22%, 255.81%, and 336.43% in 8, 10, and 12 dS m−1 treatments compared to the control treatment (0 dS m−1), respectively. The salinity proﬁle in pot soil Frontiers in Plant Science 04 frontiersin.org Alam et al. Frontiers in Plant Science 3.2.3 Impact of different levels of salinity on leaf- associated traits Development of soil salinity (ECs) over the periods of the onion life cycle at different salinity concentrations. All the data points are the mean values of ECs recorded in replicated experimental pots representing 25 different onion genotypes (150 pots). Development of soil salinity (ECs) over the periods of the onion life cycle at different salinity concentrations. All the data points are the mean values of ECs recorded in replicated experimental pots representing 25 different onion genotypes (150 pots). leaves were found in Ac Bog 414, Ac Bog 415, Ac Bog 416, Ac Bog 417, Ac Bog 418, Ac Bog 419, Ac Bog 420, Ac Bog 422, Ac Bog 424, and Ac Bog 425 where the highest percentage of plants with dried leaves (PPDL) was recorded in Ac Bog 410 (100%), Ac Bog 432 (100%), and Ac Bog 428 (83.33%). After applying 10 dS m−1 salinity, no plants with dried leaves were found in Ac Bog 418, Ac Bog 420, and Ac Bog 424, but the highest PPDL was found in Ac Bog 432 (100%) and Ac Gaz 379 (83.33%). highest MLL (cm) under 12 dS m−1 salinity was recorded in Ac Bog 414 (23.40) and the lowest was recorded in Ac Bog 432 (0.0001). After subjecting all levels of treatments (0, 8, 10, and 12 dS m−1), the average MLL of onion seedlings was between 9.44 and 22.76 cm. The number of leaves (NL) was markedly inﬂuenced by the application of salinity (0, 8, 10, and 12 dS m−1) treatments (Tables 3 and S3). The number of leaves of onion seedlings ranged from 1.67 to 7.17 under untreated control treatment. Under the 8 dS m−1 salinity level, the highest NL was recorded in Ac Bog 420 (5.83) and Ac Bog 425 (5.83), and the lowest NL was recorded in Ac Bog 428 (2.50). The highest NL was found in Ac Bog 416 (6.67) under 10 dS m−1 salinity, whereas the lowest NL was found in Ac Bog 410 (1.33). Under 12 dS m−1 salinity treatment, the lowest NL was observed in Ac Bog 411 (1.83), whereas the highest NL was found in Ac Bog 424 (5.50). Considering the mean values of NL over the different salinity levels, the leaf number varied between 2.63 and 5.71. 3.2.3 Impact of different levels of salinity on leaf- associated traits At 12 dS m−1 salinity, no plants with dried leaves were found in Ac Bog 412, Ac Bog 422, and BARI Piaz-4, but the highest PPDL was found in Ac Bog 411 (100%), Ac Bog 412 (100%), Ac Bog 420 (100%), and Ac Bog 432 (100%). Mean PPDL ranged from 4.17% (Ac Bog 422) to 75% (Ac Bog 432) after subjecting all levels (0, 8, 10, and 12 dS m−1) of salinity. The percentage of green plants (PGP) was noticeably inﬂuenced by imposing salinity (Table 2 and Table S3). Almost all studied plants remain green under control treatments, but under 8 dS m−1, 100% of plants were dead in Ac Bog 410 and Ac Bog 432 genotypes where, under the same treatment, the highest PPG (100%) was observed in Ac Bog 414, Ac Bog 415, Ac Bog 416, Ac Bog 420, and Ac Bog 424. Under 10 dS m−1 of salinity, the highest percentage (100%) of green plants was found in Ac Bog 418, Ac Bog 420, and Ac Bog 424 where the lowest number of green plants was recorded in Ac Bog 432 (0%). Under 12 dS m−1 of salinity, in the Ac Bog 422 genotype, the highest percentage (100%) of green plants was noted and the lowest (0%) was obtained from Ac Bog 410, Ac Bog 411, Ac Bog 430, and Ac Bog 432. After subjecting all levels (0, 8, 10, and 12 dS m−1) of salinity, PGP was found to range from 1.5 to 5.25. FIGURE 3 Soil salinity development after the application of different treatments at the end of the cropping season. Each point represents the mean of ECs values recorded in replicated experimental pots of 25 genotypes (150 pots). MLL was prominently affected by imposing salinity (Table 3 and Table S3). The MLL (cm) of onion seedlings ranged from 13.67 to 26.00 under untreated control treatment. The highest MLL (cm) under 8 dS m−1 was recorded in Ac Bog 409 (25.67), which was followed by Ac Bog 415 (24.33), Ac Bog 414 (23.80), and Ac Bog 416 (23.60), and the lowest MLL was found in Ac Bog 432 (0.0001). Under 10 dS m−1, MLL (cm) was noted to be the highest in Ac Bog 422 (21.17), which was followed by Ac Bog 416 (20.67) and Ac Bog 423 (20.30), and the lowest was obtained in Ac Bog 410 (7.50). Frontiers in Plant Science 3.2.3 Impact of different levels of salinity on leaf- associated traits was in ascending order from the start to the end of the crop cycle, i.e., ﬁrst to eighth phase (Figure 2). The soil salinity levels (ECs) reached at the eighth phase in all treatments (8, 10, and 12 dS m−1) were lower than those of the respective irrigation water salinity (Figure 3). However, it was expected to be the same at the irrigation treatments in an equilibrium state. was in ascending order from the start to the end of the crop cycle, i.e., ﬁrst to eighth phase (Figure 2). The soil salinity levels (ECs) reached at the eighth phase in all treatments (8, 10, and 12 dS m−1) were lower than those of the respective irrigation water salinity (Figure 3). However, it was expected to be the same at the irrigation treatments in an equilibrium state. Drying out of onion plant leaves was markedly inﬂuenced by the different levels (0, 8, 10, and 12 dS m−1) of salinity (Table 2 and Table S3). Almost 100% of the seedlings of onion plants were green under non-saline treatment, but at 8 dS m−1, no plants with dried Frontiers in Plant Science 05 frontiersin.org Alam et al. 10.3389/fpls.2023.1154051 FIGURE 1 Comparison of the mean germination percentage (GP) and growth repression (GR) of 25 onion genotypes at different concentrations or salinity levels of irrigation water. FIGURE 1 Comparison of the mean germination percentage (GP) and growth repression (GR) of 25 onion genotypes at different concentrations or salinity levels of irrigation water. FIGURE 2 Development of soil salinity (ECs) over the periods of the onion life cycle at different salinity concentrations. All the data points are the mean values of ECs recorded in replicated experimental pots representing 25 different onion genotypes (150 pots). FIGURE 2 Development of soil salinity (ECs) over the periods of the onion life cycle at different salinity concentrations. All the data points are the mean values of ECs recorded in replicated experimental pots representing 25 different onion genotypes (150 pots). FIGURE 2 Development of soil salinity (ECs) over the periods of the onion life cycle at different salinity concentrations. All the data points are the mean values of ECs recorded in replicated experimental pots representing 25 different onion genotypes (150 pots). FIGURE 1 Comparison of the mean germination percentage (GP) and growth repression (GR) of 25 onion genotypes at different concentrations or salinity levels of irrigation water. 3.2.3 Impact of different levels of salinity on leaf- associated traits The FIGURE 3 Soil salinity development after the application of different treatments at the end of the cropping season. Each point represents the mean of ECs values recorded in replicated experimental pots of 25 genotypes (150 pots). 06 frontiersin.org Alam et al. 10.3389/fpls.2023.1154051 TABLE 2 Performance of onion genotypes for greenness-related traits under different salinity stress levels in pot culture in the second year. formance of onion genotypes for greenness-related traits under different salinity stress levels in pot culture in the second year. 3.2.5 Associated indices of soil salinity stress BL (cm) ranged from 12.76 to 38.74 in the control treatment. The highest BL (cm) was found in Ac Bog 417 (50.2), which was followed by Ac Bog 424 (49.58), and the lowest was recorded in Ac Bog 432 (6.84) under 8 dS m−1 salinity. Moreover, under 10 dS m−1 salinity, the highest BL (cm) was found in Ac Bog 417 (47.38), which was followed by Ac Bog 424 (47.2), Ac Bog 420 (44.62), and Ac Bog 414 (44.02), and the lowest was recorded in Ac Bog 432 (8.52). ShTI, STI, and PYR were distinctly inﬂuenced by the application of different levels of salinity in comparison with the untreated control (Tables 5 and 6). The ShTI under 8 dS m−1 ranged from 0.00 to 99.31 with a mean of 88.62 (Table 5). The highest ShTI was recorded in Ac Bog 415, which was followed by Ac Bog 414 (99.17) and Ac Bog 409 (98.36), and the lowest ShTI was recorded in Ac Bog 432 under 8 dS m−1 salinity treatment. Under 10 dS m−1, the ShTI ranged from 31.91 to 96.03 with a mean of 75.56. The highest ShTI was documented in Ac Bog 425, and the lowest ShTI was documented in Ac Bog 410. The mean ShTI under 12 dS m−1 was 63.56 with a range from 0.00 to 97.50, where the highest ShTI was recognized in Ac Bog 414 but the lowest was recognized in Ac Bog 411 and Ac Bog 432. On the other hand, the highest BL (cm) was obtained from Ac Bog 426 (54.46), and the lowest was recorded in Ac Bog 432 (17) under 12 dS m−1 salinity. After assessing the mean of BL (cm) of all treatments, BL ranged from 12.64 to 43.32 cm. The BD (cm) of onion seedlings was remarkably inﬂuenced by the salinity treatments (Tables 4 and S3). BD (cm) ranged from 11.2 to 39.56 in the untreated control. The highest BD (cm) under 8 dS m−1 salinity was observed in Ac Bog 426 (38.36) and the lowest was recorded in BARI Piaz-4 (10.4). Furthermore, under 10 dS m−1 salinity, the highest BD (cm) was found in Ac Bog 429 (31.68), which was followed by Ac Bog 420 (31.46), and the lowest was recorded in Ac Bog 432 (7.02). 3.2.3 Impact of different levels of salinity on leaf- associated traits TABLE 2 Continued Genotype Plants with dried leaves (%) Green plants (%) C (0 dS m−1) 8 dS m−1 10 dS m−1 12 dS m−1 Mean C (0 dS m−1) 8 dS m−1 10 dS m−1 12 dS m−1 Mean Ac Gaz 379 0 (1.65) 50 (45) 83.33 (71.54) 66.67 (54.74) 50 (43.23) 100 (88.35) 50 (45) 16.67 (18.46) 33.33 (35.26) 50 (46.77) BARI Piaz-4 0 (1.65) 66.67 (54.74) 16.67 (18.46) 0 (1.65) 20.83 (19.13) 100 (88.35) 33.33 (35.26) 83.33 (71.54) 16.67 (18.46) 58.33 (53.4) Mean 1.33 (3) 29.33 (27.55) 32.67 (31.2) 45.33 (41.99) 27.17 (25.94) 98.67 (87) 62 (54.84) 64.67 (56.68) 44.67 (39.62) 67.5 (59.53) CV % 50.83 – 21.88 – LSD (0.05) 5.23 – 12.93 – CV, coefﬁcient of variation; LSD (0.05), least signiﬁcant difference; C, control treatment, 0 dS m−1; data in parentheses represent the transformed value of the corresponding data. 3.2.3 Impact of different levels of salinity on leaf- associated traits g yp g y p y Genotype Plants with dried leaves (%) Green plants (%) C (0 dS m−1) 8 dS m−1 10 dS m−1 12 dS m−1 Mean C (0 dS m−1) 8 dS m−1 10 dS m−1 12 dS m−1 Mean Ac Bog 409 0 (1.65) 33.33 (35.26) 33.33 (35.26) 33.33 (35.26) 25 (26.86) 100 (88.35) 66.67 (54.74) 83.33 (71.54) 66.67 (54.74) 79.17 (67.34) Ac Bog 410 0 (1.65) 100 (88.35) 66.67 (54.74) 100 (88.35) 66.67 (58.27) 100 (88.35) 0 (1.65) 50 (45) 0 (1.65) 37.5 (34.16) Ac Bog 411 0 (1.65) 50 (45) 66.67 (54.74) 100 (88.35) 54.17 (47.43) 100 (88.35) 33.33 (35.26) 33.33 (35.26) 0 (1.65) 41.67 (40.13) Ac Bog 412 0 (1.65) 33.33 (35.26) 16.67 (18.46) 0 (1.65) 12.5 (14.26) 100 (88.35) 50 (45) 83.33 (71.54) 66.67 (54.74) 75 (64.91) Ac Bog 414 0 (1.65) 0 (1.65) 16.67 (18.46) 33.33 (35.26) 12.5 (14.26) 100 (88.35) 100 (88.35) 83.33 (71.54) 66.67 (54.74) 87.5 (75.74) Ac Bog 415 0 (1.65) 0 (1.65) 16.67 (18.46) 50 (45) 16.67 (16.69) 100 (88.35) 100 (88.35) 83.33 (71.54) 33.33 (35.26) 79.17 (70.87) Ac Bog 416 0 (1.65) 0 (1.65) 16.67 (18.46) 83.33 (71.54) 25 (23.33) 100 (88.35) 100 (88.35) 66.67 (54.74) 16.67 (18.46) 70.83 (62.47) Ac Bog 417 0 (1.65) 0 (1.65) 16.67 (18.46) 66.67 (54.74) 20.83 (19.13) 100 (88.35) 100 (88.35) 83.33 (71.54) 33.33 (35.26) 79.17 (70.87) Ac Bog 418 0 (1.65) 0 (1.65) 0 (1.65) 33.33 (35.26) 8.33 (10.06) 100 (88.35) 50 (45) 100 (88.35) 66.67 (54.74) 79.17 (69.11) Ac Bog 419 0 (1.65) 0 (1.65) 16.67 (18.46) 16.67 (18.46) 8.33 (10.06) 100 (88.35) 83.33 (71.54) 83.33 (71.54) 83.33 (71.54) 87.5 (75.74) Ac Bog 420 0 (1.65) 0 (1.65) 0 (1.65) 100 (88.35) 25 (23.33) 100 (88.35) 100 (88.35) 100 (88.35) 16.67 (18.46) 79.17 (70.87) Ac Bog 421 0 (1.65) 33.33 (35.26) 16.67 (18.46) 16.67 (18.46) 16.67 (18.46) 100 (88.35) 66.67 (54.74) 83.33 (71.54) 66.67 (54.74) 79.17 (67.34) Ac Bog 422 0 (1.65) 0 (1.65) 16.67 (18.46) 0 (1.65) 4.17 (5.86) 100 (88.35) 83.33 (71.54) 66.67 (54.74) 100 (88.35) 87.5 (75.74) Ac Bog 423 0 (1.65) 16.67 (18.46) 33.33 (35.26) 50 (45) 25 (25.09) 100 (88.35) 83.33 (71.54) 66.67 (54.74) 50 (45) 75 (64.91) Ac Bog 424 0 (1.65) 0 (1.65) 0 (1.65) 33.33 (35.26) 8.33 (10.06) 100 (88.35) 100 (88.35) 100 (88.35) 66.67 (54.74) 91.67 (79.94) Ac Bog 425 0 (1.65) 0 (1.65) 16.67 (18.46) 16.67 (18.46) 8.33 (10.06) 100 (88.35) 83.33 (71.54) 66.67 (54.74) 83.33 (71.54) 83.33 (71.54) Ac Bog 426 0 (1.65) 16.67 (18.46) 33.33 (35.26) 33.33 (35.26) 20.83 (22.66) 100 (88.35) 66.67 (54.74) 66.67 (54.74) 33.33 (35.26) 66.67 (58.27) Ac Bog 427 16.67 (18.46) 50 (45) 66.67 (61.8) 83.33 (71.54) 54.17 (49.2) 83.33 (71.54) 33.33 (28.19) 33.33 (35.26) 16.67 (18.46) 41.67 (38.36) Ac Bog 428 16.67 (18.46) 83.33 (71.54) 66.67 (54.74) 33.33 (35.26) 50 (45) 83.33 (71.54) 16.67 (18.46) 33.33 (35.26) 66.67 (54.74) 50 (45) Ac Bog 429 0 (1.65) 16.67 (18.46) 16.67 (18.46) 16.67 (18.46) 12.5 (14.26) 100 (88.35) 66.67 (54.74) 66.67 (54.74) 66.67 (54.74) 75 (63.14) Ac Bog 430 0 (1.65) 16.67 (18.46) 33.33 (35.26) 50 (45) 25 (25.09) 100 (88.35) 66.67 (61.81) 50 (45) 0 (1.65) 54.17 (49.2) Ac Bog 431 0 (1.65) 66.67 (54.74) 50 (45) 16.67 (18.46) 33.33 (29.96) 100 (88.35) 16.67 (18.46) 33.33 (35.26) 66.67 (54.74) 54.17 (49.2) Ac Bog 432 0 (1.65) 100 (88.35) 100 (88.35) 100 (88.35) 75 (66.67) 100 (88.35) 0 (1.65) 0 (1.65) 0 (1.65) 25 (23.33) (Continued) 07 Frontiers in Plant Science frontiersin.org 10.3389/fpls.2023.1154051 Alam et al. 3.2.4 Impact of different levels of salinity on bulb-associated traits found in Ac Bog 420 (17) and the lowest was documented in Ac Bog 428 (1.4). The mean IBW (g) of all salinity treatments varied between 3.9 and 16.5 g for all the studied genotypes. The BL (cm) of onion seedlings was decisively inﬂuenced by the salinity treatments (Table 4 and S3). 3.2.5 Associated indices of soil salinity stress In comparison, the highest IBW (g) under 12 dS m−1 salinity was 08 Frontiers in Plant Science frontiersin.org Alam et al. 10.3389/fpls.2023.1154051 Performance of onion genotypes for leaf-related traits under different salinity stress levels in pot culture in the second year. 3.2.5 Associated indices of soil salinity stress Maximum leaf length CV, coefﬁcient of variation; LSD (0.05), least signiﬁcant different; C, control treatment; 0 dS m−1. among the MR of different indices, though differential results are also in place. Ranking and mean ranking were also estimated for yield reduction (%) (PYR), and a reﬂection of a similar ranking pattern from the previous result was observed (Table 6). Finally, to select better genotypes having good yield, better tolerance, and minimum yield loss under stress, ranking based on IBW and re- ranking of MR (of ShTI, STI, and PYR) was carried out (Table 7). 34.24 to 94.52 with a mean of 73.87. The highest PYR was found in Ac Bog 428, which was followed by Ac Bog 413 (93.64) and Ac Bog 423 (92.40), and the lowest was found in the Ac Bog 420 genotype. 3.2.5 Associated indices of soil salinity stress In contrast, the highest BD (cm) under 12 dS m−1 salinity was found in Ac Bog 420 (29.4) and the lowest was recorded in Ac Bog 432 (7.52). After evaluating the mean of BD (cm) of all treatments, the average BD ranged from 12.1 to 26.28 cm. Under 8 dS m−1, the STI ranged from 0.22 to 4.19 with a mean of 1.79 (Table 5). Maximum STI was noted in Ac Bog 430, and the lowest was noted in Ac Bog 432 under 8 dS m−1 salinity treatment. The mean STI under 10 dS m−1 was 1.19 with a range of 0.13 to 3.43. The highest STI was documented in Ac Bog 424, which was followed by Ac Bog 420 (3.25), and the lowest was noted in Ac Bog 432. The STI under 12 dS m−1 ranged from 0.12 to 3.41 with a mean of 1.16. The highest STI under 12 dS m−1 salinity treatment was found in Ac Bog 420, which was followed by Ac Bog 424 (3.22), but the lowest was found in the Ac Bog 432 genotype. The mean PYR under 8 dS m−1 was 56.00 with a range of 0.89 to 92.59. The highest PYR under 8 dS m−1 salinity treatment was recorded in Ac Bog 424, which was followed by Ac Bog 426 (89.91) and Ac Bog 379 (86.92), and the lowest was noted in Ac Bog 409. The PYR under 10 dS m−1 ranged from 37.33 to 94.13 with a mean of 73.62. Under 10 dS m−1, the highest PYR was obtained in Ac Bog 428, which was followed by Ac Bog 413 (93.64) and Ac Bog 423 (89.52), and the lowest was obtained in Ac Bog 420. Under 12 dS m−1, the PYR ranged from The IBW (g) of onion seedlings was markedly inﬂuenced by the salinity treatments (Tables 4 and S3). IBW (g) oscillated from 7.65 to 38.15 in the untreated control. The highest IBW (g) under 8 dS m−1 salinity was observed in Ac Bog 409 (22.2) and the lowest was recorded in Ac Bog 424 (2.1). Moreover, under 10 dS m−1 salinity, the highest IBW (g) was found in Ac Bog 420 (16.2), which was followed by Ac Bog 424 (15.6), and the lowest was noted in Ac Bog 428 (1.5). Frontiers in Plant Science 3.2.5 Associated indices of soil salinity stress Genotypes Maximum leaf length Number of leaves C (0 dS m−1) 8 dS m−1 10 dS m−1 12 dS m−1 Mean C (0 dS m−1) 8 dS m−1 10 dS m−1 12 dS m−1 Mean Ac Bog 409 25.67 25.25 16.17 17.6 21.17 6.17 4.00 5.17 4.17 4.88 Ac Bog 410 23.5 11.33 7.5 10 13.08 2.83 4.17 1.33 2.83 2.79 Ac Bog 411 22 19.5 11.5 0.0001 13.25 2.83 4.00 2.83 1.83 2.87 Ac Bog 412 25 17 15.8 17.2 18.75 5.33 5.33 4.00 4.50 4.79 Ac Bog 413 21.5 18.83 15.67 8.4 16.10 6.00 4.17 4.83 4.50 4.29 Ac Bog 414 24 23.8 19.83 23.4 22.76 5.17 5.67 5.83 4.67 5.54 Ac Bog 415 24.5 24.33 19 19.4 21.81 4.67 5.00 6.00 4.67 5.21 Ac Bog 416 25.67 23.6 20.67 12 20.49 6.00 4.50 6.67 3.50 4.83 Ac Bog 417 23.33 22.33 17.5 21.8 21.24 3.67 5.50 5.50 4.83 5.46 Ac Bog 419 24.33 14.5 20 18.17 19.25 6.33 4.83 3.17 5.00 4.83 Ac Bog 420 25.83 17.5 18.17 17.5 19.75 6.17 5.83 3.83 4.67 5.13 Ac Bog 421 18.8 18.2 14.83 14.33 16.54 4.50 5.17 5.50 4.00 4.79 Ac Bog 422 24.33 22 21.17 15.33 20.71 5.83 5.17 4.83 4.50 5.08 Ac Bog 423 22.17 20.6 20.3 20.25 20.83 7.17 5.00 5.83 4.83 5.71 Ac Bog 424 25.83 21.67 18.5 20.75 21.69 6.17 5.67 3.17 5.50 5.13 Ac Bog 425 20.17 19.67 19.37 18.5 19.43 6.33 5.83 3.33 4.33 4.96 Ac Bog 426 24.2 22.8 17.83 12.67 19.38 6.00 5.00 5.00 4.00 5.00 Ac Bog 427 19.67 19 12.33 12 15.75 5.33 3.17 5.67 3.17 4.33 Ac Bog 428 13.67 12.5 10.67 8.25 11.27 1.67 2.50 3.00 3.33 2.63 Ac Bog 429 26 23.33 13.83 16.4 19.89 4.83 5.50 4.67 4.83 4.96 Ac Bog 430 21.17 17 19.8 17.4 18.84 5.67 4.83 2.67 4.83 4.50 Ac Bog 431 21.75 20 18.5 15 18.81 2.50 3.00 4.67 3.83 3.50 Ac Bog 432 22.75 0.0001 15 0.0001 9.44 3.17 2.83 2.00 2.67 2.67 Ac Gaz 379 20.33 19 15.67 10 16.25 3.00 2.83 5.33 2.83 3.50 BARI Piaz-4 21.5 18 10 15.25 16.19 2.50 5.00 2.67 4.83 3.75 Mean 22.69 17.17 20.72 15.51 19.11 4.8 4.3 4.58 4.11 4.45 CV% 20.76 – 20.97 – LSD (0.05) – – 0.92 CV, coefﬁcient of variation; LSD (0.05), least signiﬁcant different; C, control treatment; 0 dS m−1. 3.2.6 Ranking of the genotypes Genotypes were ranked against salinity stress tolerance to select potential ones (Table 5). Based on different tolerance indices (ShTI and STI), genotypes were ranked, where dissimilarities in ranking positions were observed within a particular index for some of the potential genotypes. Thus, mean ranking (MR) was estimated for all the indices. Many of the genotypes are observed in the upper order Genotypes’ position in those ranks was marked and their position within the top 10 was counted. The genotypes that present a maximum of four times in the top 10 were regarded as tolerant genotypes and those that present three times were denoted as 09 frontiersin.org Alam et al. 3.2.6 Ranking of the genotypes 10.3389/fpls.2023.1154051 E 4 Performance of onion genotypes for bulb-related traits under different salinity stress levels in pot culture in the second yea Genotype Bulb length (mm) Bulb diameter (mm) Individual bulb weight (g) C (0 dS m−1) 8 dS m−1 10 dS m−1 12 dS m−1 Mean C (0 dS m−1) 8 dS m−1 10 dS m−1 12 dS m−1 Mean C (0 dS m−1) 8 dS m−1 10 dS m−1 12 dS m−1 Mean Ac Bog 409 18.8 35.92 34.56 48.78 34.52 39.56 15.7 24.26 22.2 25.44 22.4 22.2 8.75 8.85 15.55 Ac Bog 410 19.46 15.54 17.64 30.92 20.88 21 19.02 19.48 18.42 19.48 14.2 5.25 4.75 5.25 7.35 Ac Bog 411 15.2 15.98 26.34 37.18 23.68 21.14 12.14 20.74 13.98 17 14.25 7.45 2.8 2.6 6.8 Ac Bog 412 15.38 29.76 31.48 30.3 26.72 16 11.9 16.56 13.96 14.6 12.75 4.9 5.8 6.35 7.45 Ac Bog 413 20.8 18.16 26.04 44.28 27.32 24.72 18.84 9.56 9.34 15.62 25.95 9.4 1.65 1.65 9.65 Ac Bog 414 27.76 26.92 44.02 43.02 35.44 27.8 23.36 22.82 22.64 24.16 30.5 12 11.4 11.85 16.4 Ac Bog 415 28.36 37.08 37.72 40.74 35.98 23.74 26.1 24.3 19.02 23.28 25.55 11.4 6.7 4.1 11.95 Ac Bog 416 25.66 23.84 37.96 44.76 33.06 24.56 20.52 26.5 14.22 21.44 27.2 16.65 3.4 3.65 12.7 Ac Bog 417 28.04 50.2 47.38 47.66 43.32 30.3 22.3 27.54 24.96 26.28 22.85 19.7 5.8 3.6 13 Ac Bog 419 36.08 36.44 36.76 30.18 34.86 14.38 35.72 17.4 16.86 21.1 39.05 5.35 7.45 7.25 14.75 Ac Bog 420 25.32 45.26 44.62 31.66 36.72 20.12 23.62 31.46 29.4 26.14 25.85 6.9 16.2 17 16.5 Ac Bog 421 17.5 27 29.24 35.32 27.26 17.54 15.4 19.68 20.92 18.38 18.45 5.1 5.75 5.4 8.65 Ac Bog 422 27.14 34.46 37.12 33.02 32.94 24.3 27.84 23.56 24.3 25 30.85 10.7 8 8.55 14.5 Ac Bog 423 38.74 34.08 35.66 33.72 35.56 17.06 34 15.34 14.16 20.14 38.15 5.5 4 2.9 12.65 Ac Bog 424 27.9 49.58 47.2 46.46 42.78 20.64 21.62 27.2 25.7 23.78 28.35 2.1 15.6 14.65 15.15 Ac Bog 425 29.72 38.86 40.22 36.28 36.26 27.7 30.08 23.1 23.98 26.22 33.35 12 7.25 7.05 14.9 Ac Bog 426 36.7 33.16 34.76 54.46 39.78 33.26 38.36 14.46 13.26 24.84 35.2 3.55 3.85 3.85 11.6 Ac Bog 427 22.94 30.62 24 27.66 26.3 26.56 16.9 18.28 11.44 18.3 18.2 17.7 2.75 3.15 10.45 Ac Bog 428 16.14 23.2 22.64 23.64 21.4 11.2 15.02 20.84 8.32 13.84 25.55 9 1.5 1.4 9.35 Ac Bog 429 17.64 27 30.82 43.72 29.8 26.62 14.52 31.68 15.38 22.04 20.3 11.75 4.15 4.7 10.25 Ac Bog 430 24.88 39.74 37.76 34.64 34.26 25.9 24.84 24.84 25.02 25.14 30.5 17.7 8.05 8.7 16.25 Ac Bog 431 19.54 24.56 33.04 39.18 29.08 20.78 17.18 16.02 17.48 17.86 17.65 9.25 3.9 3.35 8.55 Ac Bog 432 18.18 6.84 8.52 17 12.64 18.26 15.62 7.02 7.52 12.1 7.65 3.7 2.2 2.1 3.9 Ac Gaz 379 17.22 27.04 30.26 26.36 25.22 16.34 15.9 29.26 18.44 19.98 23.7 3.1 4.55 4.45 8.95 BARI Piaz-4 12.76 36.04 29.82 28.46 26.78 16.44 10.4 14.28 14.36 13.88 9.05 3.5 3.55 4.15 5.05 Mean 23.51 30.69 33.02 36.38 30.90 22.64 21.08 21.05 17.81 20.64 23.90 9.43 5.99 5.86 11.29 CV% 20.52 35.11 48.23 LSD (0.05) 3.11 3.55 1.07 CV, coefﬁcient of variation; LSD (0.05), least signiﬁcant different; C, control treatment; 0 dS m−1. Frontiers in Plant Science 3.2.6 Ranking of the genotypes corresponding to the salinity levels where IBW was considered as the dependent variable. The result showed that 61% of the IBW variation was accounted for by the irrigation water salinity levels (Figure 4). moderate genotypes. The genotypes present four times in the top 10 were Ac Bog 409, Ac Bog 414, Ac Bog 424, and Ac Bog 430, whereas the genotypes that occurred three times within the top 10 were Ac Bog 417, Ac Bog 419, Ac Bog 420, Ac Bog 422, and Ac Bog 425. Those genotypes could be selected as salinity-tolerant candidates. The soil salinity levels (ECs) had signiﬁcant effects on bulb formation and its subsequent development. The soil salinity levels at different phases were regressed on IBWs of the studied genotypes. The regression analysis helps to examine the strength of the relationship between dependent and independent variables. It helped to identify the relative importance of predictor variables in terms of contributing to the variation in dependent variables. The frontiersin.org 3.3 Regression study A simple linear regression analysis was carried out involving salinity concentration (ECw) in irrigation water and IBW of onion 10 frontiersin.org frontiersin.org Alam et al. 10.3389/fpls.2023.1154051 TABLE 5 Stress indices and their ranks for the different studied onion genotypes under different salinity treatments in pot culture in the second year. TABLE 5 Stress indices and their ranks for the different studied onion genotypes under different salinity treatments in pot culture in the second year. s and their ranks for the different studied onion genotypes under different salinity treatments in pot culture in the second year. Frontiers in Plant Science 3.3 Regression study variability and are the most important phases under salinity stress (Tables 8 and 9). regression analysis showed that salinity levels at different phases negatively contributed to the IBW (Table 8). Coefﬁcient of determination (R2) values for linear regression varied between 0.39 and 0.58. Multiple regression results revealed that combining all the phases accounted for 61% of the contribution towards the total IBW variation (Table 8). Figure 5 shows the contribution of different phases to the IBW variation. It was clear from the graph that, after phase 4, the attainment of soil salinity level is very distinct. 3.3 Regression study E 5 Stress indices and their ranks for the different studied onion genotypes under different salinity treatments in pot culture in t Genotype ShTI_8 dS m−1 R ShTI_10 dS m−1 R ShTI_12 dS m−1 R MR STI_8 dS m−1 R STI_10 dS m−1 R STI_12 dS m−1 R MR Ac Bog 409 98.36 3 62.99 20 68.56 13 12 3.86 2 1.52 8 1.54 8 4.5 Ac Bog 410 48.21 24 31.91 25 42.55 22 23.67 0.58 20 0.52 19 0.58 18 14.25 Ac Bog 411 88.64 16 52.27 23 0 24 21 0.82 18 0.31 22 0.29 22 15.5 Ac Bog 412 68 21 63.2 19 68.8 12 17.33 0.48 22 0.57 17 0.63 17 14 Ac Bog 413 87.58 17 72.88 15 39.07 23 18.33 1.89 10 0.33 21 0.33 21 13 Ac Bog 414 99.17 2 82.63 6 97.5 1 3 2.84 6 2.7 3 2.81 3 3 Ac Bog 415 99.31 1 77.55 11 79.18 7 6.33 2.26 9 1.33 9 0.81 12 7.5 Ac Bog 416 91.94 12 80.52 8 46.75 21 13.67 3.52 3 0.72 15 0.77 14 8 Ac Bog 417 95.71 7 75.01 13 93.44 2 7.33 3.49 4 1.03 12 0.64 16 8 Ac Bog 419 59.6 23 82.2 7 74.68 9 13 1.62 14 2.26 4 2.2 4 5.5 Ac Bog 420 67.75 22 70.34 17 67.75 14 17.67 1.38 15 3.25 2 3.41 1 4.5 Ac Bog 421 96.81 5 78.88 9 76.22 8 7.33 0.73 19 0.82 14 0.77 13 11.5 Ac Bog 422 90.42 14 87.01 4 63.01 16 11.33 2.56 7 1.92 5 2.05 6 4.5 Ac Bog 423 92.92 10 91.57 3 91.34 4 5.67 1.63 13 1.18 10 0.86 10 8.25 Ac Bog 424 83.89 18 71.62 16 80.33 6 13.33 0.46 23 3.43 1 3.22 2 6.5 Ac Bog 425 97.52 4 96.03 1 91.72 3 2.67 3.11 5 1.88 7 1.83 7 4.75 Ac Bog 426 94.21 8 73.68 14 52.36 19 13.67 0.97 17 1.05 11 1.05 9 9.25 Ac Bog 427 96.59 6 62.68 21 61.01 17 14.67 2.5 8 0.39 20 0.45 20 12 Ac Bog 428 91.44 13 78.05 10 60.35 18 13.67 1.79 12 0.3 23 0.28 24 14.75 Ac Bog 429 89.73 15 53.19 22 63.08 15 17.33 1.85 11 0.65 16 0.74 15 10.5 Ac Bog 430 80.3 20 93.53 2 82.19 5 9 4.19 1 1.91 6 2.06 5 3 Ac Bog 431 91.95 11 85.06 5 68.97 11 9 1.27 16 0.53 18 0.46 19 13.25 Ac Bog 432 0 25 65.93 18 0 24 22.33 0.22 25 0.13 25 0.12 25 18.75 Ac Gaz 379 93.46 9 77.08 12 49.19 20 13.67 0.57 21 0.84 13 0.82 11 11.25 BARI Piaz-4 83.72 19 46.51 24 70.93 10 17.67 0.25 24 0.25 24 0.29 23 17.75 Mean 88.62 75.56 63.56 1.83 1.19 1.16 SD 26.48 19.64 24.73 1.25 0.94 0.96 Min 0.00 31.91 0.00 0.22 0.13 0.12 Max 99.31 96.03 97.50 4.19 3.43 3.41 ShTI, shoot tolerance index; STI, stress tolerance index; R, rank; MR, mean rank; SD, standard deviation; Min, minimum; Max, maximum. frontiersin.org 4 Discussion Salinity is a major environmental stressor that reduces agricultural production and sustainability in arid and semiarid settings by delaying the commencement of germination and subsequent seedling establishment (Uçarlı, 2021). Worldwide, salt has a negative impact on agricultural yield. Approximately 30 A further analysis (stepwise regression) was done to ﬁnd the most critical phases for crop when undergoing salinity stress. Results showed that phases 3, 5, and 7 were responsible for 63% 11 frontiersin.org 10.3389/fpls.2023.1154051 Alam et al. E 6 The yield reduction (%) and their ranks for the different studied onion genotypes under different saline water treatments. Genotype PYR_8 dS m−1 R PYR_10 dS m−1 R PYR_12 dS m−1 R MR Ac Bog 409 0.89 1 60.94 5 60.49 5 3.67 Ac Bog 410 63.03 14 66.55 7 63.03 7 9.33 Ac Bog 411 47.72 8 80.35 17 81.75 17 14.00 Ac Bog 412 61.57 13 54.51 3 50.2 3 6.33 Ac Bog 413 63.78 15 93.64 24 93.64 24 21.00 Ac Bog 414 60.66 11 62.62 6 61.15 6 7.67 Ac Bog 415 55.38 10 73.78 11 83.95 19 13.33 Ac Bog 416 38.79 4 87.5 21 86.58 21 15.33 Ac Bog 417 13.79 3 74.62 13 84.25 20 12.00 Ac Bog 419 86.3 22 80.92 19 81.43 16 19.00 Ac Bog 420 73.31 20 37.33 1 34.24 1 7.33 Ac Bog 421 72.36 19 68.83 8 70.73 8 11.67 Ac Bog 422 65.32 18 74.07 12 72.29 10 13.33 Ac Bog 423 85.58 21 89.52 23 92.4 23 22.33 Ac Bog 424 92.59 25 44.97 2 48.32 2 9.67 Ac Bog 425 64.02 16 78.26 15 78.86 13 14.67 Ac Bog 426 89.91 24 89.06 22 89.06 22 22.67 Ac Bog 427 2.75 2 84.89 20 82.69 18 13.33 Ac Bog 428 64.77 17 94.13 25 94.52 25 22.33 Ac Bog 429 42.12 6 79.56 16 76.85 12 11.33 Ac Bog 430 41.97 5 73.61 10 71.48 9 8.00 Ac Bog 431 47.59 7 77.9 14 81.02 14 11.67 Ac Bog 432 51.63 9 71.24 9 72.55 11 9.67 Ac Gaz 379 86.92 23 80.8 18 81.22 15 18.67 BARI Piaz-4 61.33 12 60.77 4 54.14 4 6.67 Mean 56.00 73.62 73.87 SD 28.28 14.27 15.48 Min 0.89 37.33 34.24 Max 92.59 94.13 94.52 PYR, percent yield reduction (over 0 dS m−1 treatment); R, rank; MR, mean rank; SD, standard deviation; Min, minimum; Max, maximum. Frontiers in Plant Science 4 Discussion agricultural plants currently provide 90% of plant-based human food, and the bulk of these crops, known as glycophytes, are neither salt-tolerant nor salt-sensitive. Because of salt sensitivity, glycophytes make up the majority of cultivated plants. Osmotic stress, ion toxicity, and oxidative stress all have an impact on seed germination and seedling establishment. The negative inﬂuence of abiotic stresses, such as salt, heat, and drought, has an undesirable impact on seed germination (Wahid et al., 2007). Inhibition of seed germination, fall in germination percentage, and germination delay are the initial outcomes of salinity (Uçarlı, 2021) and are caused by altering the levels of seed germination stimulants (i.e., Gibberellic Acids (GAs), Abscisic Acid (ABA), membrane permeability, and water behavior) in the seed (Uçarlı, 2021). Results of the year I study stated that germination percentage was hampered by the different salinity treatments. Germination percentage decreased with the increase in salt concentration. Seed germination primarily increases at low concentrations of salt (NaCl), but at rising concentrations, it was signiﬁcantly reduced (Abdel-Fattah et al., 1972; Sudha and Riazunnisa, 2015). A similar scenario was also observed in different types of salt solutions (NaCl, Frontiers in Plant Science 12 frontiersin.org Alam et al. 10.3389/fpls.2023.1154051 TABLE 7 Ranking of onion genotypes based on bulb weight, stress indices, and yield reduction (%) under different salinity treatments in pot culture in the second year. TABLE 7 Ranking of onion genotypes based on bulb weight, stress indices, and yield reduction (%) under different salinity tre the second year. 4 Discussion onion genotypes based on bulb weight, stress indices, and yield reduction (%) under different salinity treatments in pot culture in Genotype Rank based on Occurrence* (in the top 10) IBW MR (ShTI) MR (STI) MR (PYR) Ac Bog 409 4 8 2 1 4 Ac Bog 410 22 18 17 7 1 Ac Bog 411 23 16 19 13 0 Ac Bog 412 21 13 16 2 1 Ac Bog 413 16 15 14 18 0 Ac Bog 414 2 2 1 5 4 Ac Bog 415 12 4 6 12 2 Ac Bog 416 10 11 7 15 2 Ac Bog 417 9 5 7 11 3 Ac Bog 419 7 9 4 17 3 Ac Bog 420 1 14 2 4 3 Ac Bog 421 19 5 12 10 2 Ac Bog 422 8 7 2 12 3 Ac Bog 423 11 3 8 19 2 Ac Bog 424 5 10 5 8 4 Ac Bog 425 6 1 3 14 3 Ac Bog 426 13 11 9 20 1 Ac Bog 427 14 12 13 12 0 Ac Bog 428 17 11 18 19 0 Ac Bog 429 15 13 10 9 2 Ac Bog 430 3 6 1 6 4 Ac Bog 431 20 6 15 10 2 Ac Bog 432 25 17 21 8 1 Ac Gaz 379 18 11 11 16 0 BARI Piaz-4 24 14 20 3 1 IBW, individual bulb weight; MR, mean rank; ShTI, shoot tolerance index; STI, stress tolerance index; and PYR, percent yield reduction. *Total number of times a genotype is present within the 10 diff ki Rank based on FIGURE 4 Regression of salinity concentration ECw (dS m−1) of irrigation water on individual bulb weight (g). CaCl2, and MgCl2) (Malik and Singh, 1977). Therefore, the selection of genotypes based on salinity stress performance would produce better results (Regessa et al., 2010; Regessa et al., 2022). Though the germination percentage was negatively affected by salt treatment, the impact was not so detrimental for this particular trait as quite a few genotypes showed better performance in terms of germinability. In contrast, many of the genotypes were greatly inﬂuenced, showing subsequent growth repression and ultimately dying under different salt treatments. Very high growth repression was observed at elevated salinity treatments for different genotypes that previously showed better germination percentage under an ideal environment. Frontiers in Plant Science 4 Discussion As a whole, the germination process (emergence to ﬁrst leaf development) is vulnerable to salt stress. This harmful effect may be prompted by Na and Cl ions’ direct inﬂuence on embryo viability (Jahromi et al., 2008; Daszkowska-Golec, 2011) or CaCl2, and MgCl2) (Malik and Singh, 1977). Therefore, the selection of genotypes based on salinity stress performance would produce better results (Regessa et al., 2010; Regessa et al., 2022). Though the germination percentage was negatively affected by salt treatment, the impact was not so detrimental for this particular trait as quite a few genotypes showed better performance in terms of germinability. In contrast, many of the genotypes were greatly inﬂuenced, showing subsequent growth repression and ultimately dying under different salt treatments. Very high growth repression was observed at elevated salinity treatments for different genotypes that previously showed better germination percentage under an ideal environment. As a whole, the germination process (emergence to ﬁrst leaf development) is vulnerable to salt stress. This harmful effect may be prompted by Na and Cl ions’ direct inﬂuence on embryo viability (Jahromi et al., 2008; Daszkowska-Golec, 2011) or FIGURE 4 Regression of salinity concentration ECw (dS m−1) of irrigation water on individual bulb weight (g). FIGURE 4 Regression of salinity concentration ECw (dS m−1) of irrigation water on individual bulb weight (g). 13 frontiersin.org 10.3389/fpls.2023.1154051 Alam et al. TABLE 8 Regression analysis for IBW based on soil salinity developed at different phases. Item Phase SLR MLR SWR 1 2 3 4 5 6 7 8 b −1.68 −1.36 −1.07 −0.76 −0.61 −0.54 −0.48 −0.44 – – R2 0.39 0.53 0.58 0.55 0.54 0.55 0.57 0.56 0.61 0.63 p-value <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 SLR = simple linear regression; MLR = multiple linear regression; SWR = stepwise regression; b = regression coefﬁcient; R2 = coefﬁcient of determination. Numbers 1 to 8 represent the different dates/phases when observation on ECs was recorded. Item Phase SLR MLR SWR 1 2 3 4 5 6 7 8 b −1.68 −1.36 −1.07 −0.76 −0.61 −0.54 −0.48 −0.44 – – R2 0.39 0.53 0.58 0.55 0.54 0.55 0.57 0.56 0.61 0.63 p-value <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 SLR = simple linear regression; MLR = multiple linear regression; SWR = stepwise regression; b = regression coefﬁcient; R2 = coefﬁcient of determination. 4 Discussion However, practically robust foliage during the vegetative stage is a must for maximum production (Bosch Serra and Casanova, 2000; Sta-Baba et al., 2005). Similar to this (Bernstein and Ayers, 1953), 50% reduced growth in onion in ﬁeld plots in Riverside, CA, at an EC value of 4.1 dS m−1, was also observed. In the present study, the number of dried plants increased with the increase in salinity levels. Drying out or death of seedlings due to exposure to the shallow root system of onion at high salt concentration (Sta-Baba et al., 2005) was also reported under salt water irrigation in a previous study conducted by De Malach et al. (1989). In contrast, the number of green plants was reduced with the elevated salinity level. Similar to this, the number of total leaves also decreased with the increasing salinity treatments. A reduction in leaf numbers by up to 50% was reported after 45 days in onion in an earlier study (Sta-Baba et al., 2010). Apart from this, it was also revealed that MLL was reduced with the increase in salinity levels. This indicates that salt stress has a negative impact on MLL, i.e., the height of onion plants, and approximately 21% reduction in height was recorded due to this abiotic stress (Sta- Baba et al., 2010). Some of the genotypes showed better leaf length despite the elevated salinity level, which indicates that there was some sort of tolerance mechanism. Those plants had the capability to grow under higher salt stress conditions. The height of onion plants was also studied by different research groups in previous studies, and they reported that plant height reduced as salt levels rose (Janki et al., 2020). Interestingly, many research ﬁndings (Bernstein and Hayward, 1958; Bernstein, 1962; Allison, 1964) reported that on saline soils, plant height can be limited or completely inhibited by the following factors: (a) the osmotic effect on plant roots, (b) the toxic effect of accumulated ions in plant tissues, (c) the speciﬁc effect of constituent ions, or (d) a combination of all three. Reduction in onion’s vegetative growth and development in terms of plant height, number of green leaves, leaf length, etc. was also reported in earlier ﬁndings (Hanci and Cebeci, 2018; Shoaib et al., 2018; Regessa et al., 2022; Sanwal et al., 2022). 4 Discussion Numbers 1 to 8 represent the different dates/phases when observation on ECs was recorded. p-value <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 SLR = simple linear regression; MLR = multiple linear regression; SWR = stepwise regression; b = regression coefﬁcient; R2 = coefﬁcient of determination. Numbers 1 to 8 represent the different dates/phases when observation on ECs was recorded. SLR = simple linear regression; MLR = multiple linear regression; SWR = stepwise regression; b = regression coefﬁcient; R2 = coefﬁcient of determination. Numbers 1 to 8 represent the different dates/phases when observation on ECs was recorded. FIGURE 5 Regression of ECs at different phases on the individual bulb weight of 25 different onion genotypes. IGURE 5 FIGURE 5 Regression of ECs at different phases on the individual bulb weight of 25 different onion genotypes. FIGURE 5 Regression of ECs at different phases on the individual bulb weight of 25 different onion genotypes. TABLE 9 Initial and ﬁnal model after stepwise regression of different phases of soil salinity levels (ECs) on individual bulb weight. Type Model Initial model IBW ~ Phase_1 + Phase_2 + Phase_3 + Phase_4 + Phase_5 + Phase_6 + Phase_7 + Phase_8 Final model IBW ~ Phase_3 + Phase_5 + Phase_7 IBW, individual bulb weight; different phases representing the ECs values of all the experimental pots at different phases of the crop cycle. TABLE 9 Initial and ﬁnal model after stepwise regression of different phases of soil salinity levels (ECs) on individual bulb weight. Type Model Initial model IBW ~ Phase_1 + Phase_2 + Phase_3 + Phase_4 + Phase_5 + Phase_6 + Phase_7 + Phase_8 Final model IBW ~ Phase_3 + Phase_5 + Phase_7 IBW, individual bulb weight; different phases representing the ECs values of all the experimental pots at different phases of the crop cycle. TABLE 9 Initial and ﬁnal model after stepwise regression of different phases of soil salinity levels (ECs) on individual bulb weight. 14 Frontiers in Plant Science frontiersin.org 10.3389/fpls.2023.1154051 Alam et al. 10.3389/fpls.2023.1154051 present study, traits related to bulb size and weight were observed under different salinity levels. Differences in the performances of the studied genotypes were recorded at different salt stresses. An increase in BL along with a decrease in BD as well as bulb weight was observed with the increase in salt concentration of irrigation water. 4 Discussion A similar ﬁnding of hampered onion bulb ﬁrmness and size was reported in an earlier study (Venâncio et al., 2022). It denotes that bulb development or growth is adversely inﬂuenced by the salt stress accompanied by soil water unavailability triggered by salt concentration. Bulb growth is the most vulnerable (Kadayifci et al., 2005) and tends to escape the stress by reducing the duration of different stages, leading to a faster bulb growth stage under salinized soil (Kamran et al., 2019; Paudel et al., 2020), and ultimately ends with small-sized bulbs, which was common in previous studies. In a previous study (Ayers and Westcot, 1985), it was concluded that onion yield potential is 100% for ECw = 0.8 dS m−1, 90% for 1.2 dS m−1, 75% for 1.8 dS m−1, 50% for 2.9 dS m−1, and 0% for 5.0 dS m−1. The reduction in yield might be due to fewer bulbs per unit area, as well as reduced bulb size (Sta-Baba et al., 2005). A considerable decrease in fresh bulb weight and bulb volume was associated with the increasing salt concentration (Janki et al., 2020; Regessa et al., 2022; Sanwal et al., 2022; Venâncio et al., 2022). Fifty percent and 80% bulb loss were reported at 3.7 and 9.51 dS m−1 solution plots, respectively, compared to the control plot in an earlier study (Sta- Baba et al., 2010). indirectly by decreasing the water availability around seeds (Koyro et al., 2008). After each irrigation treatment, plants turn brown and ultimately older leaves tend to dry. Sometimes, all the leaves in plants of a few genotypes become completely dry within a week of saline water application. Thus, onion plants’ growth in terms of the number of dried plants, number of green plants, number of total leaves, and MLL was severely affected by salinity stress. The interplay between saline treatments and plant age would result in signiﬁcant changes in onion leaves, especially in terms of weight (Sta-Baba et al., 2005; Sta-Baba et al., 2010), which might be due to the changes in cells’ osmotic potential following water intake reduction (El-Hendawy et al., 2019). Although many of the dried plants started to produce new leaves after the older leaves became dry, in this process, plant growth was hampered and could not contribute to accumulating the assimilates towards the bulb formation and subsequent development. Frontiers in Plant Science frontiersin.org 6 Conclusions Onion is one of the most sensitive crops to salt stress as compared to other spice crops. Starting from germination to bulb maturity, all the stages are vulnerable to salinity stress. Reduction in germination percentage along with repression in subsequent growth indicated the adverse inﬂuence of higher salt concentration towards early plant formation. Interruption in leaf growth due to stress ensured limited or no food production as well as successive translocation to the bulb, which ultimately ends with reduced production. A higher salt concentration during the bulb formation stage interferes with the bulb development process, resulting in a reduced and irregular bulb shape, volume, and weight of onion bulbs. All these ultimately lead to a reduction in bulb fresh yield. The signiﬁcant reduction in bulb weight compared to the control treatment clearly deﬁnes the detrimental effect of salt stress on onion yield. The variable performance of the studied genotypes under stress conditions speciﬁes the availability of variability among the germplasm for salinity tolerance. The better sustainability and the subsequent bulb formation of several salinity-tolerant genotypes can be used in further research and for cultivation in saline-prone areas. The present ﬁndings will improve the current understanding of the salinity tolerance of onion. The present output allows the scope of the developing gene pool to have certain characteristics associated with tolerance to salt stress. Further advanced research related to genomic- level studies can dissect the underlying molecular mechanism behind the salinity tolerance of onion. The result from the regression analysis indicated the signiﬁcant effects of saline water irrigation towards the bulb weight variation. Corresponding to water salinity (ECw), soil salinity had a similar contribution to the variation in IBW. Elevated salinity levels are directly associated with impaired soil nutrient availability, which hampers not only nutrient absorption but also water uptake. Plants under salt stress were forced to undergo physiological changes to combat unfavorable conditions. All these lead to the reduction in onion bulb yield under such stress conditions. Simple as well as multiple linear regressions yielded similar results of soil salinity impact on the IBW. Soil salinity levels of all the phases individually and also combined signiﬁcantly contributed to the yield variation. A report from a regression analysis on onion bulb yield was found to be affected by stress during the growth stage (Lee et al., 2019). 4 Discussion Both varietal potential and salt concentration had an extensive By delaying the start of germination and subsequent seedling establishment, salinity becomes a signiﬁcant environmental constraint that lowers agricultural production and stability in arid and semiarid situations. When salt (NaCl) content is low, seed germination is predominantly increased, but as the concentration rises, it is greatly decreased. Onion leaves grow and develop differently depending on plant age and saline treatments. In this investigation, the number of dried plants increased as salinity levels rose. When the salt concentration was higher, the growth rate was signiﬁcantly slower. MLL, or the height of onion plants, is negatively impacted by salinity exertion. Onion growth and yield characteristics varied greatly depending on both varietal potential and salt content (Hanci and Cebeci, 2018; Shoaib et al., 2018; Regessa et al., 2022; Sanwal et al., 2022). Onion has a low EC threshold, making it extremely susceptible to salt stress. Bulb growth is most susceptible to salinity stress and tends to avoid it by shortening the time between stages. Soil salinity levels gradually increase with the saline water application as irrigation (Venâncio et al., 2022). The progression increased after consecutive applications up to the fourth irrigation for all the levels and reached the highest point at the end of the crop season. At this point, soil salinity (ECs) was almost nearer to the corresponding levels of irrigation water solutions (ECw). Similar to this, the gradual evolution pattern and the highest levels of ECs nearer to ECw at the end of the onion crop cycle were also concluded in a study (Sta-Baba et al., 2010). Stress-associated indices help to select superior genotypes having tolerance to particular stresses. Ranking depending on the stress index values makes it easier to identify the best-performing genotypes. Three different indices were estimated in the current study consisting of indicators based on shoot growth (ShTI), bulb Both varietal potential and salt concentration had an extensive effect on different growth and yield attributes of onion (Janki et al., 2020). Onion is very much vulnerable to salt stress (Mangal et al., 1989) with a low EC threshold (Maas and Hoffman, 1977). In the 15 frontiersin.org Alam et al. 10.3389/fpls.2023.1154051 10.3389/fpls.2023.1154051 weight (STI), and bulb weight loss under stress (PYR). 6 Conclusions The result of stepwise regression indicated that the middle phases of the growth cycle are the vulnerable stage under salinity stress (phases 3 to 7). This may be due to the fact that at the early vegetative stage, crop plants can recover quickly, while at the last stage, salinity level had very minimal effects, if any, on the crop as, at this point, bulb formation had already been completed. Stepwise regression was employed to account for the variability contributed by predictor variables on onion bulb yield in the previous study and identiﬁed the most responsible traits related to the early growth stage (Sanwal et al., 2022). From the regression graph, it was also evident that salinity development in soil due to the saline water was more 4 Discussion ShTI recognizes the genotypes that show better shoot growth under salinity stress, while STI could differentiate the better genotypes that have the potential of producing sizable onion bulbs under a salinity stress environment. On the other hand, PYR identiﬁes the genotypes that show minimum bulb yield loss under stress conditions. Results of the present study revealed that the ranking of ShTI, STI, and PYR at different salinity levels showed a dissimilar genotype at the upper positions. For example, Ac Bog 415, Ac Bog 425, and Ac Bog 414 were at the top of the position (rank 1) under salinity levels 8, 10, and 12 dS m−1, respectively, for the stress index ShTI. A similar pattern was also observed for the other two indices (STI and PYR). Thus, a combined rank position, including the stress indices and the rank of IBW, was intended to be estimated. In doing so, mean rank (MR) was calculated for all the indices, and subsequently, a fresh ranking was employed based on the mean ranking of indices. To identify the best genotypes, presence in the top 10 ranking in multiple categories (i.e., the rank of IBW and MRs of indices) was counted. The maximum occurring genotypes were believed to have better vegetative growth, yielding ability, and minimum yield loss and ultimately had some sort of tolerance mechanism under salt stress conditions. In the present study, nine genotypes that had those characteristic features were recorded. Genotype performance under stress and sustainable selection would result depending on this (Venâncio et al., 2022). distinct after consecutive applications of few irrigations (fourth in the case of the present study). Data availability statement The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s. Author contributions Conceptualization: MA, MAR, MMR, MH, MN, AF, MM, SR, MI, and SM; methodology: MA, MAR, MMR, MH, MN, AF, MM, SR, MI, and SM; validation: MA, MAR, MMR, MH, MN, AF, MM, SR, MI, and SM; formal analysis: MA, AG, AA, and AH; investigation: MA and MAR; resources: MA and AH; data curation: MA, AG, AA, and AH; writing—original draft preparation: MA, MAR, MMR, MH, MN, AF, MM, SR, MI, and SM; editing: AG, AH, and AA; visualization: MA, MAR, MMR, 16 Frontiers in Plant Science frontiersin.org 10.3389/fpls.2023.1154051 Alam et al. References Abdel-Fattah, M. A., Abdel-Salam, A. S., Elmofty, I. A., and Abdel-Gawwad, M. M. (1972). Salt tolerance of onion during germination and early seedling growth. Desert Inst. Bull. A.R.E 22, 157–165. FAOSTAT Statistical database (Rome, Italy: Food and Agriculture Organization of the United Nations). Available at: http://fenix.fao.org/faostat/internal/en/#data (Accessed 13 September 2020). Fischer, R., and Maurer, R. (1978). Droughtresistance in springwheat cultivars. 1: Grain yieldresponse. Aust. J. Agr Res. 29, 897–912. doi: 10.1071/AR9780897 Ahsan, M., and Bhuiyan, M. R. (2010). Soil and water salinity, their management in relation to the climate changes in coastal areas of Bangladesh. Khulna Univ. Stud. 2010, 31–42. doi: 10.53808/KUS.SI.SESB.2010.31-42-ls Guellim, A., Catterou, M., Chabrerie, O., Tetu, T., Hirel, B., Dubois, F., et al. (2019). Ben; kichey, t. identiﬁcation of phenotypic and physiological markers of salt stress tolerance in durum wheat (Triticum durum desf.) through integrated analyses. Agronomy 9, 844. doi: 10.3390/agronomy9120844 Allison, L. E. (1964). Salinity in relation to irrigation. Adv. Agron. 16, 139–180. doi: 10.1016/S0065-2113(08)60023-1 Ayers, R. S., and Westcot, D. W. (1985). Water quality for agriculture Vol. 29 (Rome, italy: FAO UNITED NATIONS), 97. Hanci, F., and Cebeci, E. (2015). Comparison of salinity and drought stress effects on some morphological and physiological parameters in onion (Allium cepa l.) during early growth phase. Bulg. J. Agric. Sci. 21, 1204–1210. BBS (2019). Household income and expenditure survey (Bangladesh: Bangladesh Bureau of Statistics (BBS), Statistics and Informatics Division (SID), Ministry of Planning). Hanci, F., and Cebeci, E. (2018). Improvement of abiotic stress tolerance in onion: Selection studies under salinity conditions. Int J Eng Sci. 7 (9), 45–58. doi: 10.9790/ 1813-0709015458 Bernstein, L. (1962). “Salt-affected soils and plants,” in Proceedings of the proceedings of the Paris symposium, vol. 18. (Paris: UNESCO), 139–174. Hanci, F., Cebeci, E., and Fidanci, A. A. (2016). Rapid screening method for salt stress tolerance of onion genotypes. J. Appl. Hortic. 18, 39–43. doi: 10.37855/ jah.2016.v18i01.09 Bernstein, L., and Ayers, A. D. (1953). Salt tolerance of ﬁve varieties of carrots. J. Am. Soc Hortic. Sci. 61, 360–366. Bernstein, L., and Hayward, H. E. (1958). Physiology of salt tolerance. Annu. Rev. Plant Physiol. 9, 25–46. doi: 10.1146/annurev.pp.09.060158.000325 Jahromi, F., Aroca, R., Porcel, R., and Ruiz-Lozano, J. M. (2008). Inﬂuence of salinity on the in vitro development of glomus intraradices and on the in vivo physiological and molecular responses of mycorrhizal lettuce plants. Microb. Ecol. 55, 45–53. doi: 10.1007/s00248-007-9249-7 Bosch Serra, A. D., and Casanova, D. (2000). Supplementary material The authors would also like to extend their sincere appreciation to Bangladesh Agricultural Research Institute and Princess Nourah bint Abdulrahman University Researchers Supporting Project number (PNURSP2023R65), Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia, for supporting the current research. The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpls.2023.1154051/ full#supplementary-material Publisher’s note This research was partially funded by the “Strengthening spices crops research in Bangladesh” project and Princess Nourah bint Abdulrahman University Researchers Supporting Project number (PNURSP2023R65), Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia. All claims expressed in this article are solely those of the authors and do not necessarily represent those of their afﬁliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Conﬂict of interest MH, MN, AF, MM, SR, MI, and SM; supervision: AH and MA; funding: AA, AG, SM, and AH. All authors contributed to the article and approved the submitted version. The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Frontiers in Plant Science Khan, M. A. A., Alam, M. A., Yousuf, M. N., Raihan, H. Z., Rahman, M. A., and Brahma, S. (2022). Evaluation of short day local and exotic onion genotypes. Bangladesh J. Agric. 47, 27–43. doi: 10.3329/bjagri.v47i2.63324 References Estimation of onion (Allium cepa, l.) biomass and light interception from reﬂectance measurements at ﬁeld level. Acta Hortic. 519, 53–63. doi: 10.17660/actahortic.2000.519.4 Janki, P., Vekaria, L., Hl, S., Parmar, K. B., and HP, P. (2020). Effect of saline irrigation water on growth and yields of onion (Allium cepa l.) varieties. Int. J. Chem. Stud. 8, 966–969. doi: 10.22271/chemi.2020.v8.i4g.9726 Chinnusamy, V., Jagendorf, A., and Zhu, J. K. (2005). Understanding and improving salt tolerance in plants. Crop Sci. 45, 437–448. doi: 10.2135/ cropsci2005.0437 Kadayifci, A., Tuylu, G. I., Ucar, Y., and Cakmak, B. (2005). Crop water use of onion (Allium cepa l.) in Turkey. Agric. Water Manage. 72, 59–68. doi: 10.1016/ j.agwat.2004.08.002 Choukan, R., Taherkhani, T., Ghannadha, M. R., and Khodarahmi, M. (2006). Evaluation of drought tolerance in grain maize inbred lines using drought tolerance indices. Iran. J. Agric. Sci. 8, 79–89. Kamran, M., Parveen, A., Ahmar, S., Malik, Z., Hussain, S., Chattha, M. S., et al. (2019). An overview of hazardous impacts of soil salinity in crops, tolerance mechanisms, and amelioration through selenium supplementation. Int. J. Mol. Sci. 21, 148. doi: 10.3390/ijms21010148 Daszkowska-Golec, A. (2011). Arabidopsis seed germination under abiotic stress as a concert of action of phytohormones. Omi. A J. Integr. Biol. 15, 763–774. doi: 10.1089/ omi.2011.0082 De Malach, Y., Pasternak, D., Mendlinger, S., and Borovic, I. (1989). Irrigation with brackish water under desert conditions. VIII 16, 201–215. doi: 10.1016/0378-3774(89) 90003-6 Khan, M. A. (2003). “Halophyte seed germination: Success and pitfalls,” in Proceedings of the international symposium on optimum resource utilization in salt affected ecosystems in arid and semi arid regions Cairo, Egypt: Desert research centre, 346–358. El-Hendawy, S., Elshafei, A., Al-Suhaibani, N., Alotabi, M., Hassan, W., Dewir, Y. H., et al. (2019). Assessment of the salt tolerance of wheat genotypes during the germination stage based on germination ability parameters and associated SSR markers. J. Plant Interact. 14, 151–163. doi: 10.1080/17429145.2019.1603406 Khan, M. A. A., Alam, M. A., Yousuf, M. N., Raihan, H. Z., Rahman, M. A., and Brahma, S. (2022). Evaluation of short day local and exotic onion genotypes. Bangladesh J. Agric. 47, 27–43. doi: 10.3329/bjagri.v47i2.63324 Frontiers in Plant Science 17 frontiersin.org Alam et al. 10.3389/fpls.2023.1154051 Alam et al. Koyro, H.-W., Lieth, H., and Eisa, S. S. (2008). References “Salt tolerance of chenopodium quinoa willd., grains of the Andes: Inﬂuence of salinity on biomass production, yield, composition of reserves in the seeds, water and solute relations,” in Mangroves and halophytes: Restoration and utilisation, vol. 43 . Eds. M. Garcia Sucre and B. Herzog (Dordrecht: Springer Netherlands), 133–145. yield attributes in onion. Physiol. Mol. Biol. Plants an Int. J. Funct. Plant Biol. 24, 1093– 1101. doi: 10.1007/s12298-018-0570-z Slavich, P. G., and Petterson, G. H. (1993). Estimating the electrical conductivity of saturated paste extracts from 1:5 soil, water suspensions and texture. Soil Res. 31, 73–81. doi: 10.1071/SR9930073 Lee, J., Min, B., Yoon, S., Lee, M., Kim, H., and Hong, K. A. (2019). Multiple- regression model of bulb onion yield in response to meteorological conditions in gyeongsangnam province, republic of Korea. Acta Hortic. 1251, 81–90. doi: 10.17660/ ActaHortic.2019.1251.10 SRDI (2010). Saline soils of Bangladesh (Ministry ofAgriculture, Dhaka: Soil Resources Development Institute). Sta-Baba, R., Bkheder, M., Chaar, H., and Harrabi, M. (2005). Identiﬁcation des critères de sélection pour l’adaptation variétale d’oignon (Allium cepa l. Rev. Inst. Nat. Agr. Tunisie 20, 141–159. Maas, E. V., and Hoffman, G. J. (1977). Crop salt tolerance - current assessment. ASCE J. Irrig Drain Div 103, 115–134. doi: 10.1061/jrcea4.0001137 Sta-Baba, R., Hachicha, M., Mansour, M., Nahdi, H., and Kheder, B. M. (2010). Response of onion to salinity. Afr. J. Plant Sci. Biotechnol. ©2010 4, 7– 12. Malik, Y. S., and Singh, K. (1977). Pandita ML effect of salinity on germination of onion varieties. Haryana J. Hortic. Sci. 6, 67–72. Mangal, J. L., Lal, S., and Hooda, P. S. (1989). Salt tolerance of the onion seed crop. J. Hortic. Sci. 64, 475–477. doi: 10.1080/14620316.1989.11515980 Star Business Report (2022). Onion production increased by 27%: Razzaque. Dly. Star. Parvin, G. A., Ali, M. H., Fujita, K., Abedin, M. A., Habiba, U., and Shaw, R. (2017). “Land use change in southwestern coastal Bangladesh: Consequence to food and water supply,” in Proceedings of the land use management in disaster risk reduction. disaster risk reduction. Eds. M. Banba and R. Shaw (Tokyo: Springer). Sudha, G., and Riazunnisa, K. (2015). EFFECT OF SALT STRESS (NaCl) ON MORPHOLOGICAL PARAMETERS OF ONION (Allium cepa l.) SEEDLINGS. Int. J. plant Anim. Environ. Sci. 5, 125–128. Takahashi, F., Tilbrook, J., Trittermann, C., Berger, B., Roy, S. J., Seki, M., et al. (2015). Comparison of leaf sheath transcriptome proﬁles with physiological traits of bread wheat cultivars under salinity stress. References PLoS One 10, e0133322. doi: 10.1371/ journal.pone.0133322 Paudel, D., Dhakal, S., Parajuli, S., Adhikari, L., Peng, Z., Qian, Y., et al. (2020). “Chapter 38 - use of quantitative trait loci to develop stress tolerance in plants,” in Plant life under changing environment. Eds. D. K. Tripathi, V. Pratap Singh, D. K. Chauhan, S. Sharma, S. M. Prasad, N. K. Dubey and N. Ramawat (United States: Academic Press), 917–965, ISBN: . Uçarlı, C. (2021). Effects of salinity on seed germination and early seedling stage. Abiotic Stress Plants 211. doi: 10.5772/intechopen.93647 Pitman, M. G., and Läuchli, A. (2006). “Global impact of salinity and agricultural ecosystems,” in Salinity: Environment - plants - molecules. Eds. A. Läuchli and U. Lüttge (Dordrecht: Springer), 3–20. Ullah, A., and Bano, A. (2019). Role of PGPR in the reclamation and revegetation of saline land. Pakistan J. Bot. 51, 27–35. doi: 10.30848/PJB2019-1(43 Qadir, M., Quillérou, E., Nangia, V., Murtaza, G., Singh, M., Thomas, R. J., et al. (2014). Economics of salt-induced land degradation and restoration. Nat. Resour. Forum 38, 282–295. doi: 10.1111/1477-8947.12054 Venâncio, J. B., Dias, N.da S., de Medeiros, J. F., de Morais, P. L. D., do Nascimento, C. W. A., Neto, O. N. de S., et al. (2022). Yield and morphophysiology of onion grown under salinity and fertilization with silicon. Sci. Hortic. (Amsterdam). 301, 111095. doi: 10.1016/j.scienta.2022.111095 Rahman, A. K. M. M., Ahmed, K. M., Butler, A. P., and Hoque, M. A. (2018). Inﬂuence of surface geology and micro-scale land use on the shallow subsurface salinity in deltaic coastal areas: A case from southwest Bangladesh. Environ. Earth Sci. 77. doi: 10.1007/s12665-018-7594-0 Wahid, A., Perveen, M., Gelani, S., and Basra, S. M. A. (2007). Pretreatment of seed with H2O2 improves salt tolerance of wheat seedlings by alleviation of oxidative damage and expression of stress proteins. J. Plant Physiol. 164, 283–294. doi: 10.1016/ j.jplph.2006.01.005 R Core Team R (2021). A language and environment for statistical computing. Vienna: R Foundation for Statistical Computing. URL: https://www.R-project.org. Wang, W., Vinocur, B., and Altman, A. (2003). Plant responses to drought, salinity and extreme temperatures: Towards genetic engineering for stress tolerance. Planta 218, 1–14. doi: 10.1007/s00425-003-1105-5 Regessa, M., Hachicha, M., Mansour, M., Nahdi, H., and Kheder, M. (2010). Response of onion to salinity. Afr. J. Plant Sci. Biotechnol. 4, 7–12. Regessa, M., Gemechis, A., and Chala, E. (2022). Growth, physiology and yield of onion (Allium cepa l.) under salt stress. Greener J. Agricultural Sci. References 12 (2), 154–167. Wickham, H. (2009). Ggplot2: Elegant graphics for data analysis (Verlag New York: Springer), ISBN: . Yaron, B., and Mokady, R. A. (1962). Technique for conducting pot experiments with saline water. Plant Soil 17, 392–398. doi: 10.1007/BF01377676 Sanwal, S. K., Kesh, H., Kumar, A., Dubey, B. K., Khar, A., Rouphael, Y., et al. (2022). Salt tolerance potential in onion: Conﬁrmation through physiological and biochemical traits. Plants 11, 3325. doi: 10.3390/plants11233325 Yeo, A. (1998). Predicting the interaction between the effects of salinity and climate change on crop plants. Sci. Hortic. (Amsterdam). 78, 159–174. doi: 10.1016/S0304-4238 (98)00193-9 Shabala, S. (2013). Learning from halophytes: Physiological basis and strategies to improve abiotic stress tolerance in crops. Ann. Bot. 112, 1209–1221. doi: 10.1093/aob/ mct205 Zafar, S., Ashraf, M. Y., Niaz, M., Kausar, A., and Hussain, J. (2015). Evaluation of wheat genotypes for salinity tolerance using physiological indices as screening tool. Pakistan J. Bot. 47, 397–405. Shoaib, A., Meraj, S., Naﬁsa,, Khan, K. A., and Javaid, M. A. (2018). Inﬂuence of salinity and fusarium oxysporum as the stress factors on morpho-physiological and 18 frontiersin.org Frontiers in Plant Science
https://openalex.org/W4362627826	https://figshare.com/articles/journal_contribution/Supplementary_Data_Figure_S5_from_Somatic_HLA_Class_I_Loss_Is_a_Widespread_Mechanism_of_Immune_Evasion_Which_Refines_the_Use_of_Tumor_Mutational_Burden_as_a_Biomarker_of_Checkpoint_Inhibitor_Response/22535716/1/files/39998992.pdf	English	null	Supplementary Data Figure S1 from Somatic HLA Class I Loss Is a Widespread Mechanism of Immune Evasion Which Refines the Use of Tumor Mutational Burden as a Biomarker of Checkpoint Inhibitor Response	null	2,023	cc-by	859	Supplementary Data Figure S5. Effects of hybridization on HLA baiting efficiency. a) 0.0 Before binding affinity adjustment 0.5 1.0 Allele Frequency (HLA-A31:01) a) 0.0 0.5 1.0 Allele Frequency (HLA-A31:01) 0 25 50 75 100 x0 x0.01 x0.02 x0.03 x0.04 x0.05 x0.06 x0.07 x0.08 x0.09 x0.1 x0.11 x0.12 x0.13 x0.14 x0.15 x0.16 x0.17 x0.18 x0.19 x0.2 x0.21 x0.22 x0.23 x0.24 x0.25 x0.26 x0.27 x0.28 x0.29 x0.3 x0.31 x0.32 x0.33 x0.34 x0.35 x0.36 x0.37 x0.38 x0.39 x0.4 x0.41 x0.42 x0.43 x0.44 x0.45 x0.46 x0.47 x0.48 x0.49 x0.5 x0.51 x0.52 x0.53 x0.54 x0.55 x0.56 x0.57 x0.58 x0.59 x0.6 x0.61 x0.62 x0.63 x0.64 x0.65 x0.66 x0.67 x0.68 x0.69 x0.7 x0.71 x0.72 x0.73 x0.74 x0.75 x0.76 x0.77 x0.78 x0.79 x0.8 x0.81 x0.82 x0.83 x0.84 x0.85 x0.86 x0.87 x0.88 x0.89 x0.9 x0.91 x0.92 x0.93 x0.94 x0.95 x0.96 x0.97 x0.98 x0.99 x1 Sample Count 0.0 After binding affinity adjustment 0.5 1.0 Allele Frequency (HLA-A31:01) b) k ≥ 1 k < 1 Arbitrary Distance Unit HLA-A Haplotypes Supplementary Data Figure S5. Effects of hybridization on HLA baiting efficiency. a) Example of how hybrid-capture can pull down HLA target sequences with different efficiency. Allele frequency (AF) of the HLA-A31:01 allele before (top, median AF = 0.3) and after (bottom, median AF = 0.5) adjusting for hybrid-capture binding affinity bias in samples with HLA-A31:01 and HLA-A11:01 alleles. b) Dendrogram of representative sequences for each known two-digit haplotype of HLA-A. A matrix of all pairwise sequence distances was used to cluster the haplotypes. The k affinity constants for haplotypes on the left were all greater than or equal to 1 while the k constant for sequences on the right were all less than 1. Supplementary Data Figure S5. Effects of hybridization on HLA baiting efficiency. Supplementary Data Figure S5. Effects of hybridization on HLA baiting efficiency. mentary Data Figure S5. Effects of hybridization on HLA baiting efficiency. 0.0 Before binding affinity adjustment 0.5 1.0 Allele Frequency (HLA-A31:01) a) 0 25 50 75 100 x0 x0.01 x0.02 x0.03 x0.04 x0.05 x0.06 x0.07 x0.08 x0.09 x0.1 x0.11 x0.12 x0.13 x0.14 x0.15 x0.16 x0.17 x0.18 x0.19 x0.2 x0.21 x0.22 x0.23 x0.24 x0.25 x0.26 x0.27 x0.28 x0.29 x0.3 x0.31 x0.32 x0.33 x0.34 x0.35 x0.36 x0.37 x0.38 x0.39 x0.4 x0.41 x0.42 x0.43 x0.44 x0.45 x0.46 x0.47 x0.48 x0.49 x0.5 x0.51 x0.52 x0.53 x0.54 x0.55 x0.56 x0.57 x0.58 x0.59 x0.6 x0.61 x0.62 x0.63 x0.64 x0.65 x0.66 x0.67 x0.68 x0.69 x0.7 x0.71 x0.72 x0.73 x0.74 x0.75 x0.76 x0.77 x0.78 x0.79 x0.8 x0.81 x0.82 x0.83 x0.84 x0.85 x0.86 x0.87 x0.88 x0.89 x0.9 x0.91 x0.92 x0.93 x0.94 x0.95 x0.96 x0.97 x0.98 x0.99 x1 Sample Count 0.0 After binding affinity adjustment 0.5 1.0 Allele Frequency (HLA-A31:01) Supplementary Data Figure S5. Effects of hybridization on HLA baiting efficiency. 0.0 Before binding affinity adjustment 0.5 1.0 Allele Frequency (HLA-A31:01) a) 0 25 50 75 100 x0 x0.01 x0.02 x0.03 x0.04 x0.05 x0.06 x0.07 x0.08 x0.09 x0.1 x0.11 x0.12 x0.13 x0.14 x0.15 x0.16 x0.17 x0.18 x0.19 x0.2 x0.21 x0.22 x0.23 x0.24 x0.25 x0.26 x0.27 x0.28 x0.29 x0.3 x0.31 x0.32 x0.33 x0.34 x0.35 x0.36 x0.37 x0.38 x0.39 x0.4 x0.41 x0.42 x0.43 x0.44 x0.45 x0.46 x0.47 x0.48 x0.49 x0.5 x0.51 x0.52 x0.53 x0.54 x0.55 x0.56 x0.57 x0.58 x0.59 x0.6 x0.61 x0.62 x0.63 x0.64 x0.65 x0.66 x0.67 x0.68 x0.69 x0.7 x0.71 x0.72 x0.73 x0.74 x0.75 x0.76 x0.77 x0.78 x0.79 x0.8 x0.81 x0.82 x0.83 x0.84 x0.85 x0.86 x0.87 x0.88 x0.89 x0.9 x0.91 x0.92 x0.93 x0.94 x0.95 x0.96 x0.97 x0.98 x0.99 x1 Sample Count 0.0 After binding affinity adjustment 0.5 1.0 Allele Frequency (HLA-A31:01) b) k ≥ 1 k < 1 Arbitrary Distance Unit HLA-A Haplotypes Supplementary Data Figure S5. Effects of hybridization on HLA baiting efficiency. Supplementary Data Figure S5. Effects of hybridization on HLA baiting efficiency. 0.5 Allele Frequency (HLA-A31:01) 0 25 50 75 100 x0 x0.01 x0.02 x0.03 x0.04 x0.05 x0.06 x0.07 x0.08 x0.09 x0.1 x0.11 x0.12 x0.13 x0.14 x0.15 x0.16 x0.17 x0.18 x0.19 x0.2 x0.21 x0.22 x0.23 x0.24 x0.25 x0.26 x0.27 x0.28 x0.29 x0.3 x0.31 x0.32 x0.33 x0.34 x0.35 x0.36 x0.37 x0.38 x0.39 x0.4 x0.41 x0.42 x0.43 x0.44 x0.45 x0.46 x0.47 x0.48 x0.49 x0.5 x0.51 x0.52 x0.53 x0.54 x0.55 x0.56 x0.57 x0.58 x0.59 x0.6 x0.61 x0.62 x0.63 x0.64 x0.65 x0.66 x0.67 x0.68 x0.69 x0.7 x0.71 x0.72 x0.73 x0.74 x0.75 x0.76 x0.77 x0.78 x0.79 x0.8 x0.81 x0.82 x0.83 x0.84 x0.85 x0.86 x0.87 x0.88 x0.89 x0.9 x0.91 x0.92 x0.93 x0.94 x0.95 x0.96 x0.97 x0.98 x0.99 x1 Sample Count 0.0 After binding affinity adjustment 0.5 1.0 Allele Frequency (HLA-A31:01) k ≥1 k < 1 Arbitrary Distance Unit HLA-A Haplotypes b) Arbitrary Distance Unit Supplementary Data Figure S5. Effects of hybridization on HLA baiting efficiency. a) Example of how hybrid-capture can pull down HLA target sequences with different efficiency. Allele frequency (AF) of the HLA-A31:01 allele before (top, median AF = 0.3) and after (bottom, median AF = 0.5) adjusting for hybrid-capture binding affinity bias in samples with HLA-A31:01 and HLA-A*11:01 alleles. b) Dendrogram of representative sequences for each known two-digit haplotype of HLA-A. A matrix of all pairwise sequence distances was used to cluster the haplotypes. The k affinity constants for haplotypes on the left were all greater than or equal to 1 while the k constant for sequences on the right were all less than 1.
https://openalex.org/W2778857876	https://bmcvetres.biomedcentral.com/track/pdf/10.1186/s12917-017-1320-0	English	null	Data distribution in public veterinary service: health and safety challenges push for context-aware systems	BMC veterinary research	2,017	cc-by	8,535	© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Abstract Background: Today’s globalised and interconnected world is characterized by intertwined and quickly evolving relationships between animals, humans and their environment and by an escalating number of accessible data for public health. The public veterinary services must exploit new modeling and decision strategies to face these changes. The organization and control of data flows have become crucial to effectively evaluate the evolution and safety concerns of a given situation in the territory. This paper discusses what is needed to develop modern strategies to optimize data distribution to the stakeholders. Main text: If traditionally the system manager and knowledge engineer have been concerned with the increase of speed of data flow and the improvement of data quality, nowadays they need to worry about data overflow as well. To avoid this risk an information system should be capable of selecting the data which need to be shown to the human operator. In this perspective, two aspects need to be distinguished: data classification vs data distribution. Data classification is the problem of organizing data depending on what they refer to and on the way they are obtained; data distribution is the problem of selecting which data is accessible to which stakeholder. Data classification can be established and implemented via ontological analysis and formal logic but we claim that a context-based selection of data should be integrated in the data distribution application. Data distribution should provide these new features: (a) the organization of situation types distinguishing at least ordinary vs extraordinary scenarios (contextualization of scenarios); (b) the possibility to focus on the data that are really important in a given scenario (data contextualization by scenarios); and (c) the classification of which data is relevant to which stakeholder (data contextualization by users). Short conclusion: Public veterinary services, to efficaciously and efficiently manage the information needed for today’s health and safety challenges, should contextualize and filter the continuous and growing flow of data by setting suitable frameworks to classify data, users’ roles and possible situations. Keywords: Public veterinary services, Institutional veterinary context, Data user, Data distribution, O Data distribution in public veterinary service: health and safety challenges push for context-aware systems Laura Contalbrigo1† , Stefano Borgo2†, Giandomenico Pozza1 and Stefano Marangon1 Correspondence: lcontalbrigo@izsvenezie.it †Equal contributors 1Istituto Zooprofilattico Sperimentale delle Venezie, Viale Dell’Università 10, 35020 Legnaro, (PD), Italy Full list of author information is available at the end of the article Contalbrigo et al. BMC Veterinary Research (2017) 13:397 DOI 10.1186/s12917-017-1320-0 Contalbrigo et al. BMC Veterinary Research (2017) 13:397 DOI 10.1186/s12917-017-1320-0 Open Access Background paradigm as an effective strategy for the prevention and control of zoonoses [2, 3]. In an increasingly globalised world, this new approach encompasses zoonotic infec- tions, food safety, the environment and the health delivery systems. The integration of the epidemiological and economic frameworks within the new technological turn called the Internet of Things (IoT), led to an escalating number of distributed sensors, which, together with the laboratory data, make available large amounts of informa- tion [4]. Indeed, health management systems are fully part of the so-called big data environment [5]. Nowadays inter- national organizations like FAO and OIE are developing The increasing interdependence between humans, ani- mals and their products as well as the close association with companion animals have encouraged a change in the public health system thinking [1]. The study of public veterinary systems has rapidly grown as a domain in itself supported by the adoption since 1984 of the “One Health” * Correspondence: lcontalbrigo@izsvenezie.it †Equal contributors 1Istituto Zooprofilattico Sperimentale delle Venezie, Viale Dell’Università 10, 35020 Legnaro, (PD), Italy Full list of author information is available at the end of the article Contalbrigo et al. BMC Veterinary Research (2017) 13:397 Page 2 of 9 Page 2 of 9 web based applications to collect, store and analyze data from different sources with the aim to combine epidemio- logical, spatial and genetic data as well as data about farm- ing systems in order to provide appropriate information to public authorities and scientists on the emergence and spread of animal diseases [6]. This huge amount of data requires a methodological approach to effectively assess and manage the health and safety of the ecosystems [7]. analysed by public health managers for decision-making. Otherwise, doubts about the real situation can arise with delay in critical decisions. In the worst-case scenario, data can be distributed in a scattered way, leading to missing in- formation and even misinterpretation [17]. The successful management of a veterinary control system depends on the quality of the data as well as on their correct interpretation. The organization of the data distribution with respect to the targeted data user is clearly a key factor [18–20]. This is a significant challenge, which calls for efficient public veterinary data management systems, for quick and focused data distribution as to better help the stake- holders in their understanding of the situation and decision-making processes [8–10]. Background g y y Therefore, to be efficient, a veterinary data management system should comply with the needs of the different stakeholders that operate in the same territory although with different perspectives: monitoring, surveillance, screening and modeling of infectious diseases, prevalence estimation, risk-factor study, quantitative risk assessment, product evaluation, commercialization and so on. Each stakeholder is interested in the goals related to his/her du- ties and these can be quite apart going from health hazard and commercialization approval to animal health moni- toring. The stakeholder’s purposes indicate his/her interest in the situation and the subset of data relevant to him/her. Furthermore, when two stakeholders need the very same data, they might require them in different formats or granularity. Moreover, the very same data can acquire different connotations: they can be considered positive by one actor and negative by another [21]. Every day, public veterinary managers, as well as other stakeholders like farmers, food producers, veterinarians and distributors deal with huge quantities of data on a var- iety of topics and different organization quality, often with several types and forms. These actors need to use them to understand scenarios in the territory and to make deci- sions to manage both ordinary and emergency situations. Here, we use the term ‘situation’ to indicate the state of an area or environment of interest as it is at some point in time. Situations can be classified as positive, negative or neutral by each stakeholder, depending on its role and in- terests, distinguishing, in particular, between ordinary ones (positive or neutral) and emergency ones (negative). In ordinary situations, surveillance, monitoring and eradication plans are the basis of veterinary public health. The implementation of these entails the need to elaborate the data collected during previous programs and to integrate them with the new epidemiological in- formation. This is crucial to make an effective risk as- sessment and to decide on activity plans finalized to guarantee the achievement of sanitary goals with mini- mum cost and risk for the consumers, the producers and the public veterinary system itself [11–13]. Finally, the meaning of data is often determined in com- parison with other data [22]. For example, during an Avian Influenza outbreak in a poultry farm, the poultry producer focuses on the pass/fail result of the laboratory test because this may lead to the stamping out of the entire flock. Background Instead, the public authority evaluates the same data against a larger period of time, a broader spatial region, considering virus genetic features, comparing this information with data originating from other farms or connected production sites and considering a variety of possible epidemiological scenarios [23–25]. Analogously, the management of emergency situations (mitigation and prevention - e.g., risk mapping and hazard identification; preparedness and planning - e.g., contingency planning, personnel training, warning system; response and recovery - e.g., contingency plan activation, epidemiologic surveillance) [14] requires to look at different information types integrating several data sources to correctly identify the kind of incident, the location, the magnitude and to monitor the evolution of the situation (e.g., geographical coordinates, logistic and environmental data, animal, means of transport and people movements, laboratory data). For instance, in veterinary epidemic emergencies the identifica- tion of outbreaks and their eradication can be managed effectively only when the competent authorities have quick access to reliable, well processed and pre-analysed informa- tion on health-related data [15, 16]. This information should be coherently and rapidly organized and presented in a very intuitive way to be quickly assimilated and These observations show that data interpretation can be challenging and that the traditional approach based on com- prehensive reports collecting large numbers of heteroge- neous data cannot be a solution anymore, especially if the data relevant for the user are spread across the whole docu- ment and reported using granularities not promptly compar- able [26, 27]. Of course, beside data interpretation there are other elements that influence the manager in the decision- making process, e.g., scientific background, regulations, time constraints, experience on the field, economic as well as socio-political considerations [28, 29]. While recognizing the importance of the other elements within the whole veterin- ary system, we claim that there is the need to improve the impact and to correct the interpretation of data taking into account the context in which they are collected and the role of the actors that read them [30]. For these reasons, it is time for the public veterinary data management system to exploit Contalbrigo et al. BMC Veterinary Research (2017) 13:397 Page 3 of 9 Page 3 of 9 new data modeling and data distribution strategies that can match today’s information needs. devices, accelerometer) [45–48]. Background In this framework it is es- sential to merge these data using reliable methodologies based on typology and provenance as well as device capaci- ties, location and trustability. While typology and provenance are successfully analyzed via the application of ontological frameworks, see e.g. [49–51], contextualization techniques for the distinct stakeholders need to be better understood. As an example, Livestock Geo-wiki, a platform that aims to develop a global livestock information system, provides in- novative visualization and analysis tools addressing specific requirements of different groups of users but it does not consider contextualization techniques focused on users’ need classification [52]. y Data misleading due to wrong contextualization, distri- bution of information or representation is a daily problem within the big data environment in public health [31, 32]. Many examples are reported in both human and veterinary medicine. Especially computational epidemiology needs large and flexible datasets for epidemiological models based on records of variable forms and origins (e.g., types of individuals at specific locations and points in time, re- cords of movement of individuals between locations, social media data, digital traces and results of diagnostic test and genomic analysis) [33]. For example, EMPRES-i (FAO’s global animal disease information system) offers mapping and graphic functions combining data coming from different databases that do not use the same references to describe host taxonomy and geographical locations, there- fore, at the beginning, it was not possible to establish an accurate correspondence between the related references [34]. Moreover, in the veterinary domain, geographical in- formation systems (GIS) required spatial data of animal holdings. Often these records lack relevant information, leading to data incompatibility and even inconsistency [35]. Furthermore, various techniques need to be used to en- hance decision-making process in the face of data uncer- tainty as happens when developing information systems based on sparse and incomplete datasets of varying quality as for example the case of GLiPHA (Global Livestock Pro- duction and Health Atlas) [36]. Likewise in human medi- cine, epidemiologists face similar concerns when mapping raw data in epidemiological studies like cancer mortality rate: the generation of spurious spatial features and statis- tical artifacts may lead to spurious spatial pattern [37, 38]. Regarding misleading interpretation of statistical analysis results, in oncologic clinical trials, it is necessary to identify accurate outcomes and endpoints as well as censoring in the study, otherwise statistical analyses can easily fail [39]. Background In biology the rapid increase in data volumes due to new molecular technologies like the next-generation sequencing (NGS), and the rise of omics data generate new data types and it is still unclear how to process, store and integrate them to produce usable knowledge in accessible formats for healthcare professionals [40–42]. d h h l f h d d b h By introducing a systematic approach for data classifi- cation and for the delineation of the stakeholders’ per- spective, we can deepen the accuracy of data distributed to the stakeholders, like sensitivity and specificity [53], increase their data understanding and trust, and at the same time reduce the risk of data overflow. The outcomes of the study and contextualization of data have also a positive impact on the integration of in- formation from various datasets, which can be harvested for new purposes beyond early detection of animal diseases or food safety issues, like epidemiological studies, pharmacovigilance or identification of emerging risks [54, 55]. Finally, data contextualization increases awareness of the potential risks and ethical challenges related to data sensitivity, legitimacy requirements, and the relationship between ethics and methodology [56]. The need of data selection strategies in the public veterinary management system The need of data selection strategies in the public veterinary management system y y The public veterinary data management system is an im- portant actor in the territorial network of stakeholders since it functions as an information hub from which the different users receive updated data. This flow of information from the public veterinary data management system and the stakeholders ensures that each actor has a proper under- standing of the situation in the field. The core activities of this management system, for what concerns us here, are the acquisition, integration, elaboration, classification and distribution of veterinary data in the field of food safety, animal health and welfare. Information systems are essential to run public veterinary services since veterinary monitoring and surveillance are largely data-oriented to the point that the term big data environment has become para- digmatic even in this domain [57, 58]. Public veterinary data management systems have been working with a trad- itional approach: they collect information from different sources (like sample collectors, public officers, or their own laboratories) and release it to the stakeholders via the use of reports. Reports are data sets organized according to for- mats fixed once for all. Although there are a few exceptions, It is evident that the selection of the data to distribute, the granularity of the information and the outline of the report should start from the goals of the targeted stakeholder, in- cluding its needs and education [43]. Moreover, following the trend of the IoT [44] and the large use of Information Technology (IT) devices like “precision livestock farming” (PLF), we cannot ignore the problem of integrating and un- derstanding data generated for specific purposes using differ- ent sensors and methodologies by farmers, producers and dealers (e.g., cow pedometer, environment sensors, data- logger, video-taping/recording, sensors installed in moving Contalbrigo et al. BMC Veterinary Research (2017) 13:397 Contalbrigo et al. BMC Veterinary Research (2017) 13:397 Page 4 of 9 Page 4 of 9 This is especially hard when the user is under pressure, like in emergency scenarios and there is no time to trans- late from the different standards for data management or even to make confident comparisons and integration of data coming from different data sets with changing meas- ure units, density, frequency, etc. The need of data selection strategies in the public veterinary management system In short, the management system has no discretional option: it must distribute to every interested party all data relative to a request using some pre-established format. Today the IT evolution and the new challenges due to the IoT and the so-called big data issue, require to re- assess the public veterinary system organization and strat- egies. With the exponential grow of data, it is easy to be overwhelmed by the distributed amount of information that has the inherent risk of overpassing the cognitive capacities to correctly read, understand, compare and evaluate them [59]. One could hope that the increasing number of data remains a problem at the IT level since ad hoc services for data filtering (e.g., customization at the client’s side or online search facilities) could help to select the needed information for each evaluation and decision step. If that were true, each stakeholder would still want to access all the available information on a given scenario and use its own services to select what to focus on. Unfor- tunately, this view is naïve. Data are complex entities that have different meanings when looked at in isolation or in collections. Furthermore, the way data are organized, im- pacts the narrative they give of the territory. For example OpenFluDB, a large database of influenza virus sequences, is isolate-centric rather than sequence-centric like other influenza virus resources. This choice makes easier the association with a comprehensive amount of clinical and epidemiological data [60]. If we rely on services at the stakeholders’ side, we cannot guarantee a coherent reading of the situation across the stakeholders in the territory. The risk is that incompatible interpretations may be adopted leading to implement mu- tually contradictory action plans. The organization of a coordination level across stakeholders could solve this last issue, but it is not practical for at least two reasons. 1) stakeholders (e.g. producers and public officers) have different goals and each would try to enforce its own reading of the data; 2) a coordination level adds an extra lapse of time which is a luxury we cannot afford in emergency situations. The need of data selection strategies in the public veterinary management system More subtlety, the infor- mation in the reports is presented from the general perspective of the veterinary service and considers neither the role and goals of the stakeholders nor the context and motivations for which data have been collected. For ex- ample, data on samples collected in a monitoring program for dairy farms in a certain area cannot be reliably used to define the health status of the bovine population in the same area including beef livestock. The correct integration of these two types of data requires knowledge that the stakeholder may not have. To reduce the risk of misjudg- ment and of missed information, we need to investigate which data should be presented to a given user, in which order and in which format. To answer this, one has to de- velop methodologies for data selection. Data selection can be done in different ways, e.g., task or function oriented. A selection is task-oriented when only data relevant for a determined task (e.g. decision making) are presented to the operator. The selection is functional when the infor- mation is shown depending on the general role the oper- ator has in the system (e.g. surveillance). However, data selection is not an isolated step. Two further steps need to be considered: data classification and data distribution. Data classification is the problem of organizing data according to what they refer to and to the way they are obtained. Data distribution is the problem of choosing which data to distribute to which stakeholder and in which context. Data distribution can be seen as the specific use of a data selection methodology. mainly due to increased sensibility to privacy concerns, the principle is pretty clear: any data request type is associated with a report type. The same report is released to all the stakeholders no matter the context, their roles or the situ- ation in the field. In short, the management system has no discretional option: it must distribute to every interested party all data relative to a request using some pre-established format. mainly due to increased sensibility to privacy concerns, the principle is pretty clear: any data request type is associated with a report type. The same report is released to all the stakeholders no matter the context, their roles or the situ- ation in the field. Focus on stakeholders’ classification In the context of the public veterinary data management sys- tem, the classification of data and metadata, the data selec- tion methodologies and the data distribution should be guided by the stakeholder’s technical knowledge, role and goals. Note that the role of the user constrains its goals, yet the goals must be explicitly included in the data management since they can change depending on contextual information like available resources. Technical background, role and goal are thus the factors that can be used to balance the tension between full vs limited disclosure in order to optimize data understanding and reduce data overflow. This leads to the issue of user’s classification. If the classification of data is based on what data are about and how they are collected, the classification of users is based on their role in the system. These classifications are not affected by the context and are independent from the particular scenario. This means that one can apply the methodologies developed in the applied ontology area to provide guidelines and criteria [49–51]. As said, reports with large amounts of data are diffi- cult to read since the user has: – to look at different sections; – to identify the subsets of data relevant to its role and knowledge; g – to rewrite them in the granularity of interest; – to compare these data; – to find a narrative that explains the data coherently with the territory type and history. Contalbrigo et al. BMC Veterinary Research (2017) 13:397 Page 5 of 9 Page 5 of 9 Contalbrigo et al. BMC Veterinary Research (2017) 13:397 Contalbrigo et al. BMC Veterinary Research (2017) 13:397 Regarding data selection and distribution, we now look at the management of veterinary laboratory data. Labora- tory data furnish a fairly simple case without exogenous factors that could hinder the understanding of the prob- lem. To justify and ground the selection and distribution of data, let us assume that a suitable framework for the classification of the stakeholders is available. This frame- work should be based on the profiling of the user’s types, use this profiling to collect the user’s requirements and to characterize their perspectives as well as aims [61, 62]. Focus on stakeholders’ classification Given this, the veterinary management system can pre- pare a data selection methodology for laboratory data that, depending on the interests, action capacities, technical knowledge and goals of the stakeholder, reasons on the usefulness and completeness of some data set for that user. Note that the goal is not to hide information to the user. On the contrary, the goal is to organize and highlight the data relevant to that user so to put him/her in the best position for data comparison and understanding as it hap- pens using the support of data warehouse in the decision making process [63–65]. understood as an event where a device, a goal, an agent and the environment are at least partially known. The use context is thus a complex entity analyzed from the user’s perspective to identify what function a device (a natural or artifactual part) performs in a certain environment [72]. A veterinary situation presents important differences with respect to Mizoguchi assumptions. Veterinary public health systems deal with information in reports, not physical objects. Furthermore, the process from data collection (like diagnostic activities) to data distribution (like a report re- lease) is not aimed to change the physical world: it starts with the completion of a laboratory test and ends with the users’ knowledge update. In this view, we can think of la- boratory data and metadata like information devices in vet- erinary public health systems that are used to change not the physical reality but the users’ knowledge of physical reality. The analogy is between acting on the world to achieve a physical change and acting on the mind to achieve a knowledge change. This match between user con- texts in functional studies and user contexts in veterinary health systems suggests that the data provided by the la- boratory (the epidemiologic unit, etc.) have a precise func- tional role: they make possible to realize the needed changes in the user’s mind in terms of situation’s knowledge and awareness. The distribution of a set of data, with a suit- able format and organization for a user, impacts the user’s knowledge about the actual situation and, subsequently, on how the user can act to fulfill its role and duties. Focus on stakeholders’ classification In other words, a veterinary management system looks as an agent whose goal is to provide data users with an understanding of the actual situation suitable and reliable for their goals and perspectives, and whose tools are data and metadata which must be carefully tuned (that is, selected and distributed) to achieve the desired result. Following the terminology in [71], we call this special user context the institutional veterinary context. The laboratory data example shows also one point we introduced earlier: the need for context classification. Sev- eral research communities have been discussing the use of contexts [66, 67] and yet no general approach has been developed since proposals like [68–70] are limited to some application domain or to some types of scenario only. In the case of veterinary services, the context should tell the characteristics of the stakeholders in terms of rights to know, official goals and duties, possibly even information on the procedures it applies and actions it can take. This information should be integrated with the situation in which the stakeholder operates, that is, its environment. The notion of context and its impact on the assessment of data knowledge Note that knowledge and action completeness are distinct: the first focuses on understanding the situation and how it may evolve, the latter takes into consideration the user’s aims and its action capacities. knowledge of the situation, can drastically reduce the class of uncontrolled situations [78, 79]. The flexible integration of data and their contextualization are new conceptual tools that can turn uncontrolled situations into controlled ones while increasing awareness and confidence in the data users. We call contextualized the data of a report filtered to match the interests of a specific user, presented depending on the situation at stake and provided with the needed granularity and organization (Fig. 1). When data are contextualized, the data received by a user are even less prone to be misinter- preted, they can effectively help to properly evaluate the risks and to take the correct decisions. Furthermore, we can say that a data set is knowledge (or action) minimal for a data user if none of its proper subsets is knowledge (action, respectively) complete for that user in the sense defined above. Then, institutional veterinary contexts help to assess data knowledge and action completeness and minimality for a user by setting a framework to reliably verify effects of data awareness on the user’s decisions/actions in different scenarios. In conclusion, modern veterinary data management sys- tems should include services like data classification, selection and distribution that include these further contextualization features: (a) the organization of situation types starting from the distinction between ordinary vs emergence scenarios (contextualization of scenarios); (b) the possibility to focus on the data that are important in a given scenario (data contextualization by scenarios); and (c) the classification of which data is relevant to which stakeholder (data contextualization by users). With this approach, veterinary data providers can become active information hubs enabling the generation of sustainable and optimized data flows suit- able for today’s information age based on IoT and big data. While the advantages of this approach are clear, further dis- cussions are needed to analyze how the veterinary manage- ment system would change in relation to, e.g., ethical challenges like privacy concerns and the balance between public health and the individual rights [80, 81]. Let us call controlled a situation in which the data set pro- vided by the veterinary data management system is know- ledge and action complete. The notion of context and its impact on the assessment of data knowledge Being the points cited in the previous paragraph the core elements that a notion of context should manage, a recent proposal by Mizoguchi et al. [71], developed in the area of function context, could help. Mizoguchi and collaborators aim to introduce a general notion of function that applies to both the natural and the engineering domains. They distinguish between three kinds of contexts: systemic, use and design context. The systemic context looks at functionalities of a component from the viewpoint of the system where it is embedded; the use context looks at functionalities of a device in the scenario where the device is used; the design context looks at functionalities of a device from the perspective of the device’s designer. The notion of context in the framework of veterinary public health systems can build upon Mizoguchi’s notion of use context. The use context is associated with a situation and fixed by an intentional agent, which is a participant in the situation aiming to reach some goal(s). The context is Assuming a data user classification and characterization system is available, covering institutional user(s) and con- sumers, producers and all legal and economic stakeholders, for each agent the data provider has to select the data and metadata to distribute, to decide the granularity of the data, and to organize the report format. This choice of data se- lection and distribution is driven by the classification of the data and of the users and by an evaluation of the data and metadata that they expect for their interests and activities including their optimal organization [73–75]. Given this setting, data set can be qualified as follows. A data set is knowledge complete for a user type in a situation if from this data set and from standard background knowledge for that user type, one can derive how much the area is departing from its standard and acceptable states, the pos- sible risks as well as short and long term consequences. A data set is action complete for a user type in a situation if from this data set and from standard background knowledge Page 6 of 9 Contalbrigo et al. BMC Veterinary Research (2017) 13:397 for that user type, one has all the knowledge needed to de- velop an optimal plan of actions to achieve its goals [76, 77]. The notion of context and its impact on the assessment of data knowledge Recall that this means that the user can understand the situation in the area of interest and has all the information to act optimally for its goals. Note that a controlled situation is just a situation for which we know how to optimize data flow. The situation itself can be ordinary (positive or neutral) or an emergence (negative). Situations that are uncontrolled are the most dangerous: they are either wrongly classified or wrongly estimated to the point that the stakeholder makes wrong decisions and the consumers, animals and environment could be exposed to excessive risks, erratic behaviors of the market operators, and unnecessary economic losses. A flexible data integration and distribution, i.e., the selection of data to distribute con- textualized to user types, data interrelation and previous Data classification Data integration Data selection Users’ classification Scenario contextualization ecological data animal anagraphic data other data epidemiological data data from other databases genetic data geographical data farm system data Report for user type 1 Report for user type 2 Report for user type 3 Report distribution Report distribution Report distribution Fig. 1 Example of a data flow diagram with a flexible data distribution in accordance with users’ classification and scenario contextualization animal anagraphic data geographical data ecological data Data integration Data classification Data selection data from other databases Scenario contextualization ample of a data flow diagram with a flexible data distribution in accordance with users’ classification and scenario contextu Page 7 of 9 Contalbrigo et al. BMC Veterinary Research (2017) 13:397 Page 7 of 9 Contalbrigo et al. BMC Veterinary Research (2017) 13:397 Availability of data and materials Not applicable. 11. Jameson LJ, Medlock JM. Tick surveillance in great Britain. Vector Borne Zoonotic Dis. 2011;11(4):403–12. 11. Jameson LJ, Medlock JM. Tick surveillance in great Britain. Vector Borne Zoonotic Dis. 2011;11(4):403–12. 12. Vrbova L, Patrick DM, Stephen C, Robertson C, Koehoorn M, Parmley EJ, Galanis E. Utility of algorithms for the analysis of integrated salmonella surveillance data. Epidemiol Infect. 2016;144(10):2165–75. Abbreviation bb e at o GIS: Geopraphical Information System; IoT: Internet of Things; IT: Information Technology; NGS: Next-Generation Sequencing; PLF: Precision Livestock Farming p g ; 7. Cocker R, Rushton J, Mounier-Jack S, Karimuriba E, Lutumba P, Kambarage D, et al. Towards a conceptual framework to support one-health research for policy on emerging zoonoses. Lancet Infect Dis. 2011;11:326–31. 8. Marabelli R. The role of official veterinary service in dealing with new social challenges: animal health and protection, food safety, and the environment Rev Sci Tech OIE 2003;22(2):363 71 7. Cocker R, Rushton J, Mounier-Jack S, Karimuriba E, Lutumba P, Kambarage D, et al. Towards a conceptual framework to support one-health research for policy on emerging zoonoses. Lancet Infect Dis. 2011;11:326–31. Consent for publication 16. Drewe AJ, Hoinville LJ, Cook AJC, Floyd T, Gunn G, Stärk KDCSERVAL. A new framework for the evaluation of animal health surveillance. Transbound Emerg Dis. 2015;62:33–45. Not applicable. 17. Dusse F, Jùnior PS, Alves AT, Novais R, Vieira V, Mendoça M. Information visualization for emergency management: a systematic mapping study. Expert Syst Appl. 2016;45:424–37. References 1. Van Olmen J, Marchal B, Van Damme W, Kegels G, Hill PS. Health systems frameworks in their political context: framing divergent agendas. BMC Public Health. 2012;12:774. 2. Bidaisee S, Macpherson NLC. Zoonoses and one health: a review of the literature. J Parasitol Res. 2014; https://doi.org/10.1155/2014/874345. 2. Bidaisee S, Macpherson NLC. Zoonoses and one health: a review of the literature. J Parasitol Res. 2014; https://doi.org/10.1155/2014/874345. 3. Dahal R, Kahn L. Zoonotic diseases and one health approach. Epidemiology. 2014; https://doi.org/10.4172/2161-1165.1000e115. 4. Simonsen L, Gog JL, Olson D, Viboud C. Infectious disease surveillance in the big data era: towards faster and locally relevant systems. JID 2016; doi: 10.1093/infdis/jiw376. 5. Lee EC, Aher JM, Goldlust S, Kraemer JD, Lawson AB, Bansal S. Mind the scales: harnessing spatial big data for infectious disease surveillance and inference. JID. 2016; doi:10.1093/infdis/jiw344. 6. Farnsworth ML, Hamilton-West C, Fitchett S, Newman SH, de La Rocque S, De Simone L, et al. Comparing national and global data collection systems for reporting. Outbreaks of H5N1 HPAI. Prev Vet Med. 2010;95:175–85. Acknowledgements Not applicable. 8. Marabelli R. The role of official veterinary service in dealing with new social challenges: animal health and protection, food safety, and the environment. Rev Sci Tech OIE. 2003;22(2):363–71. Author details 1 1Istituto Zooprofilattico Sperimentale delle Venezie, Viale Dell’Università 10, 35020 Legnaro, (PD), Italy. 2Laboratory for Applied Ontology, ISTC-CNR, Via alla Cascata, 56/C, 38123 Trento, Italy. Received: 12 July 2017 Accepted: 13 December 2017 Received: 12 July 2017 Accepted: 13 December 2017 Publisher’s Note In the last decades, the study of health management systems has rapidly increased embracing the public veterinary ser- vices. We claimed that these services, to efficaciously and ef- ficiently manage the information needed today, must include classification and contextualization techniques to filter the increasing and continuous flow of data. Such a step can im- prove the overall quality of the decision processes since these are highly sensitive to the quantity, quality and organization of the received information. This new approach may become particularly important in emergency situations since today the stakeholders need to quickly realize and assess the risk situation while being engaged in assessing large amounts of incoming data. Therefore a coherent and principled notion of context is essential to help filter and promptly distribute the right information to each user. We introduced the notion of institutional veterinary context, among others, which sets a framework to assess the different techniques for flexible data classification, selection and distribution taking into account contextual perspectives. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Ethics approval and consent to participate Not applicable. 15. Iannitti S, Savini L, Palma D, Calistri P, Natale F, Di Lorenzo A, et al. An integrated web system to support veterinary activities in Italy for the management of information in epidemic emergencies. Prev Vet Med. 2014; 113:407–16. Ethics approval and consent to participate Not applicable. Authors’ contributions LC and SB equally contributed to the conception and set-up of the manuscript. SB drew attention to the development of the ontological aspects whereas LC described the public veterinary health related context. GP contributed to set-up the last version of the manuscript. SM is the mentoring author and gave major input to the philosophy behind the article and contributed to the editing of the manuscript. All authors read and approved the final manuscript. 13. Umhang G, Comte S, Hormaz V, Boucher JM, Raton V, Favier S, Boué F. Retrospective analyses of fox feces by real-time PCR to identify new endemic areas of Echinococcus multilocularis in France. Parasitol Res. 2016:1–5. 14. Cova TJ. GIS in emergency management. In: Longley PA, Goodchild MF, Maguire DJ, Rhind DW, editors. Geographical information systems: principles, techniques, applications and management. New York: Wiley; 1999. p. 845–58. Funding h Funding The manuscript was financed by the Italian Ministry of Health in the framework of several research projects about data and metadata management system development. 9. Capua I, Alexander DJ. The challenge of avian influenza to the veterinary community. Avian Pathol. 2006;35(3):189–205. 9. Capua I, Alexander DJ. The challenge of avian influenza to the veterinary community. Avian Pathol. 2006;35(3):189–205. 10. Wendt A, Kreienbrock L, Campe A. Zoonotic disease surveillance – inventory of systems integrating human and animal disease information. Zoonoses Public Health. 2015;62:61–74. 10. Wendt A, Kreienbrock L, Campe A. Zoonotic disease surveillance – inventory of systems integrating human and animal disease information. Zoonoses Public Health. 2015;62:61–74. Availability of data and materials Availability of data and materials Not applicable. Competing interests Wiesbaden: springer Fachmedien; 2016. p. 20–55. 50. Borgo S, Pozza G. Knowledge objects: a formal construct for material, information and role dependences. KMRP. 2012;10(3):227–36. 22. Clarke R. Big data, big risks. Inf Syst J. 2016;26(1):77–90. 51. Pozza G, Borgo S, Oltramari A, Contalbrigo L, Marangon S. Information and organization in public health institutes: an ontology-based modeling of the entities in the reception-analysis-report phases. J Biomed Semant. 2016;7(1):1–15 23. Martell-Moran NK, Mauer WA, Kaneene JB. Assessment of avian influenza surveillance and reporting needs of stakeholders in Michigan, 2007. J Am Vet Med Assoc. 2011;238(12):1570–83. 52. Robinson TP, Wint GRW, Conchedda G, Van Boeckel TP, Ercoli V, Palamara E, et al. Mapping the global distribution of livestock. PLoS One. 2014;9(5): e96084. 24. Comin A, Klinkenberg D, Marangon S, Toffan A, Stegeman A. Transmission dynamics of low pathogenicity avian influenza infections in Turkey flocks. PLoS One. 2011;6(10):e26935. 53. Greiner M, Gardner IA. Application of diagnostic tests in veterinary epidemiologic studies. Prev Vet Med. 2000;45:43–59. 25. Dalla Pozza M, Ceolin C, Marangon S. Emergency response following suspicion of an avian influenza outbreak. Zoonoses Public Health. 2008;55(1):50–3. 54. White RW, Tatonetti NP, Shah NH, Altman RB, Horvitz E. Web-scale pharmacovigilance: listening to signals from the crowd. J Am Med Inform Assoc. 2013;20:404–8. 26. Ferraro S, Braga F, Panteghini M. Laboratory medicine in the new healthcare environment. Clin Chem Lab Med. 2016;54(4):523–33. environment. Clin Chem Lab Med. 2016;54(4):523–33. 27. Huang L, Fernandes H, Zia H, Tavassoli P, Rennert H, Pisapia D, Elemento O. The Precision Medicine Knowledge Base: an online application for collaborative editing, maintenance and sharing of structured clinical-grade cancer mutations interpretations. BioRxiv. 2016:e059824. doi:10.1101/059824. 55. Altman RB. Web-scale pharmacovigilance in the discovery of pharmacological effects and toxicity. San Jose, CA: AAAS Annual Meeting; 2015. https://aaas.confex.com/aaas/2015/webprogram/Paper13785.html. Accessed 10 Nov 2016. cancer mutations interpretations. BioRxiv. 2016:e059824. doi:10 56. Vayena E, Salathé M, Madoff LC, Brownstein JS. Ethical challenges of big data in public health. PLoS Comput Biol. 2015;11(2):e1003904. 28. Giangaspero M, Sekiguchi S. Risk assessment of animal infectious diseases and decision making process. Clin Microbiol. 2016;5:242. 29. Del Giudice M, Caputo F, Evangelista F. How are decision systems changing? The contribution of social media to the management of decisional liquefaction. J Decis Syst. 2016:1–13. 57. Pfeiffer DU, Stevens KB. Spatial and temporal epidemiological analysis in the big data era. Prev Vet Med. 2015;122:213–20. 58. Hill AA, Crotta M, Wall B, Good L, O’Brien SJ, Guitian J. Competing interests Towards an integrated food safety surveillance system: a simulation study to explore the potential of combining genomic and epidemiological metadata. R Soc open sci. 2017;4:160721. https://doi.org/10.1098/rsos.160721. 30. Dey AK. Understanding and using context. Pers Ubiquit Comput. 2001;5:4–7. 31. Metcalf CJ, Edmunds WJ, Lessler J. Six challenges in modeling for public health policy. Epidemics. 2015;10:93–6. 59. Sivarajah U, Kamal MM, Irani Z, Weerakkody V. Critical analysis of big data challenges and analytical methods. J Bus Res. 2017;70:263–86. 32. Hansen HK, Porter T. What do big data do in global governance? Glob Gov. 2017;23(1):31–42. 60. Liechti R, Gleizes A, Kuznetsov D, Bougueleret L, Le Mercier P, Bairoch A, Xenarios I. OpenFluDB, a database for human and animal influenza virus. Database 2010. Doi:10-1093/database/baq004. 33. Marathe M, Vullikanti AKS. Computational epidemiology. The challenge of developing and using computer models to understand and control the diffusion of disease through populations. Commun ACM. 2013;56(7):88–96. 61. Quintarelli E, Rabosio E, Tanca LA. Principled approach to context schema evolution in a data management perspective. Inform. Syst. 2015;49:65–101. 34. Claes F, Kuznetsov D, Liechti R, Von Dobschuetz S, Truong BD, Gleizes A, et al. The EMPRES-i genetic module: a novel tool linking epidemiological outbreak information and genetic characteristics of influenza viruses. Database. 2014; https://doi.org/10.1093/database/bau008. 62. Zapater JJS, Escriva DML, Garcià FRS, Durà JJM. Semantic web service discovery system for road traffic information services. Exp Syst Appl. 2015;42:3833–42. 63. Crauwels APP, de Konig R, Nielen M, Elbers ARW, Dijkhuinzen AA, Tielen MJMA. Concept for a decision support system based on practical experiences from a national disease emergency: the Dutch experience. Acta Vet Scand. 2001;94:61–9. 35. Ferrè N, Kuhn W, Rumor M, Marangon SA. Conceptual holding model for veterinary applications. Geospat Health. 2014;8(2):583–96. 36. Clements ACA, Pfeiffer DU, Otte MJ, Morteo K, Chen LA. Global livestock production and health atlas (GLiPHA) for interactive presentation, integration and analysis of livestock data. Prev Vet Med. 2002;56:19–32. 64. Nielsen AC. Data warehouse for assessing animal health, welfare, risk management and communication. Acta Vet Scand. 2011;53(1):S3. 37. Gelman A, Price NP. All maps of parameter estimates are misleading. Stat Med. 1999;18:3221–34. 65. Xu Z, Lee J, Park D, Chung Y. Multidimensional analysis model for highly pathogenic avian influenza using data cube and data mining techniques. Biosyst Eng. 2017;157:109–21. 38. Huck JJ, Whyatt JD, Coulton P. Visualising patterns in spatially-ambiguous point data. JOSIS. 2015;10:47–66. 66. Koç H, Henning E, Jastram S, Starke C. Competing interests State of the art in context modeling – a systematic literature review. LNBIP. 2014;178:53–64. p 39. Heller J. Epidemiological and statistical considerations for interpreting and communicating oncology clinical trials. Vet J. 2015;205(2):233–7. 67. Mnatsakanyan ZR, Burkom HS, Hashemian MR, Coletta MA. Distributed information fusion models for regional public health surveillance. Inf Fusion. 2012;13:419–32. 40. Cook CE, Bergman MT, Finn RD, Cochrane G, Birney E, Apweiler R. The European bioinformatics institute in 2016: data growth and integration. Nucleic Acids Res. 2016;44:D20–6. 41. Zhulin IB. Databases and microbiologists. J Bacteriol. 2015;197(15):2458–67. 68. Dourish P. What we talk about when we talk about context. Personal Ubiquitous Comput. 2004;8:19–30. 42. Manzoni C, Kia DA, Vandrovcova J, Hardy J, Wood NW, Lewis PA, Ferrari R. Genome, transcriptome and proteome: the rise of omics data and their integration in biomedical research. Brief Bioinform. 2016:1–17. https://doi. org/10.1093/bib/bbw114. 69. Wang H-H, Boukiamp F, Elghamrawy T. Ontology-based approach to context representation and reasoning for managing context-sensitive construction information. J Comput Civ Eng. 2011;25(5):331–46. 43. Jelineck HF, Yatsko A, Stranieri A, Venkatraman S, Bagirov A. Diagnostic with incomplete nominal/discrete data. Airman. 2015;4(1):22–35. 70. Grill T, Tscheligi M. Towards a multi-perspectival approach of describing context. LNCS. 2011;6967:115–8. 71. Mizoghuci R, Kitamura Y, Borgo S. Towards a unified definition of function. Fr Art Int. 2012;239:103–16. 44. Atzori L, Iera A, Morabito G. The internet of things: a survey. Comput Netw. 2010;54(15):2787–805. 72. Mizoguchi R, Kitamura Y, Borgo SA. Unifying definition for artifact and biological functions. Appl Ontol. 2016;11(2):129–54. 45. Ruiz-Garcia L, Lunadei L, Barreiro P, Robla JIA. Review of wireless sensor technologies and applications in agriculture and food industry: state of the art and current trends. Sensors. 2009;9(6):4728–50. 73. Chen Z-Y, Fan Z-P, Sun M. Behavior-aware user response modeling in social media: learning from diverse heterogeneous data. Eur J Oper Res. 2015;241: 422–34. 46. Halachmi I, Guarino M. Editorial: precision livestock farming: a “per animal” approach using advanced monitoring technologies. Animal. 2016;10(9):1482–3. 74. De Andrés J, Pariente B, Gonzalez-Rodriguez M, Lanvin DF. Towards an automatic user profiling system for online information sites: identifying demographic determining factors. Online Inform Rev. 2015;39(1):61–80. 47. Terrasson G, Llaria A, Marra A, Voaden S. Accelerometer based solution for precision livestock farming: geolocation enhancement and animal activity identification. IOP Conf Ser Mater Sci Eng. 2016;138(1):012004. 75. Haesen R, Snoeck M, Lemahieu W, Poelmans S. On the definition of service granularity and its architectural impact. LNCS. 2008;5074:375–89. 48. Competing interests 17. Dusse F, Jùnior PS, Alves AT, Novais R, Vieira V, Mendoça M. Information visualization for emergency management: a systematic mapping study. Expert Syst Appl. 2016;45:424–37. 18. Dorea CF, Sanchez J, Revie CW. Veterinary syndromic surveillance: current initiatives and potential for development Prev Vet Med 2011 101 1 17 In the past five years, none of the authors’ve received funding, fees, salary or reimbursements from organizations that may in any way gain or lose financially from the publication of this manuscript. None of the authors holds stocks of shares in an organization that may in any way gain or lose financially from the publication of this manuscript, neither now nor in the future. None of the authors holds patents relating to the content of the manuscript and they have not received reimbursement, fees, funding or salary from an organization that holds or has applied for patents relating to the content of the manuscript. In the past five years, none of the authors’ve received funding, fees, salary or reimbursements from organizations that may in any way gain or lose financially from the publication of this manuscript. None of the authors holds stocks of shares in an organization that may in any way gain or lose financially from the publication of this manuscript, neither now nor in the future. 18. Dorea CF, Sanchez J, Revie CW. Veterinary syndromic surveillance: current initiatives and potential for development. Prev Vet Med. 2011;101:1–17. 19. Dupuy C, Bronner A, Watson E, Wuyckhuise-Sjouke L, Reist M, Fouillet A, et al. Inventory of veterinary syndromic surveillance initiatives in Europe (triple-S project): current situation and perspectives. Prev Vet Med. 2013;111:220–9. None of the authors holds patents relating to the content of the manuscript and they have not received reimbursement, fees, funding or salary from an organization that holds or has applied for patents relating to the content of the manuscript. 20. Daniels P, Poermadjaja B, Morrissy C, Ngo TL, Selleck P, Kalpravidh W, et al. Development of veterinary laboratory networks for avian influenza and other emerging infectious disease control: the southeast Asian experience. EcoHealth. 2014;11:44–9. None of the authors has any financial and non-financial competing interest to declare. Page 8 of 9 Page 8 of 9 Page 8 of 9 Contalbrigo et al. BMC Veterinary Research (2017) 13:397 Page 8 of 9 21. Fürber C. Data Quality. In: Fürber C, editor. Data quality management with semantic technologies. Contalbrigo et al. BMC Veterinary Research (2017) 13:397 Contalbrigo et al. BMC Veterinary Research (2017) 13:397 80. Allan PD. Decision granularity: preserving uncertainty information in data consolidations. SPE Annual Technical Conference and Exhibition Society of Petroleum Engineers. 2004; https://doi.org/10.2118/90147-MS. 81. Birkhead GS, Klompas M, Shah NR. Uses of electronic health records for public health surveillance to advance public health. Annu Rev Public Health. 2015;18(36):345–59. 78. Kumar RL, Stylianou ACA. Process model for analyzing and managing flexibility in information systems. Eur. J Inf Syst. 2014;23(2):151–84. 79. Jablonski S, Ruf T, Wedekind H. Optimization of distributed processing by using a flexible data distribution mechanism. In: Rishe N, Navathe S, Tal D, editors. Proceedings PARBASE-90. International Conference on Databases, parallel architectures and their applications. Miami Beach, Florida, USA: IEEE Computer Society Press; 1990. p. 460–462. 80. Allan PD. Decision granularity: preserving uncertainty information in data consolidations. SPE Annual Technical Conference and Exhibition Society of Petroleum Engineers. 2004; https://doi.org/10.2118/90147-MS. 81. Birkhead GS, Klompas M, Shah NR. Uses of electronic health records for public health surveillance to advance public health. Annu Rev Public Health. 2015;18(36):345–59. 79. Jablonski S, Ruf T, Wedekind H. Optimization of distributed processing by using a flexible data distribution mechanism. In: Rishe N, Navathe S, Tal D, editors. Proceedings PARBASE-90. International Conference on Databases, parallel architectures and their applications. Miami Beach, Florida, USA: IEEE Computer Society Press; 1990. p. 460–462. 78. Kumar RL, Stylianou ACA. Process model for analyzing and managing flexibility in information systems. Eur. J Inf Syst. 2014;23(2):151–84. Competing interests Junkan A, Masip-Bruin X, Amla N. Smart Computing and Sensing Technologies for Animal Welfare: A Syst Rev. arXiv. 2016;1609.00627. 76. Tempelman-Kluit N, Pearce A. Invoking the user from data to design. Coll Res Libr. 2014;75(5):616–40. 49. Borgo S, Masolo C. Ontological foundations of DOLCE. In: Poli R, Healy M, Kameas A, editors. Theory and applications of ontology: computer applications. Dordrecht: Springer; 2010. p. 279–95. doi:10.1007/978-90- 481-8847-5_13. 77. Pandey S, Srivastava S. Data driven enterprise UX: a case study of enterprise management systems. LNCS. 2014;8522:205–16. Page 9 of 9 Contalbrigo et al. BMC Veterinary Research (2017) 13:397 • We accept pre-submission inquiries • Our selector tool helps you to ﬁnd the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to ﬁnd the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to ﬁnd the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and we will help you at every step: Submit your next manuscript to BioMed Central and we will help you at every step:
https://openalex.org/W4220967030	https://ora.ox.ac.uk/objects/uuid:37f8d178-f86a-44c0-bed2-d1e64fdb79cb/files/ssb397918r	English	null	Resilient by design: Preventing wildfires and blackouts with microgrids	Applied energy	2,022	cc-by	12,324	Resilient by design: Preventing wildfires and blackouts with microgrids Department of Engineering Science, University of Oxford, Parks Road, OX1 3PJ, United Kingdom School of Engineering, University of Edinburgh, EH8 9YL, United Kingdom H I G H L I G H T S • A novel strategy for managing wildfires and blackouts using microgrids is proposed. • The Fire Weather Index is used to describe the fire risk distribution in a grid. • The Grid and Wildfire Index are linked by line locations and fire risk distribution. • Set Victoria Australia as a case study, 68% of the overall system cost can be saved. A R T I C L E I N F O Keywords: Distributed generation Fire weather index Geospatial model Grid resilience Networked microgrid This paper proposes a strategy for managing wildfire risks and preventing blackouts using microgrids. To demonstrate this approach, not seen in previous literature, we use the power network of Victoria, Australia, in December 2019 as a case study. The Fire Weather Index (FWI) is a crucial indicator of global fire behaviour both spatially and temporally, as proved with its robust analysis within many previous studies. The FWI is applied to a Wildfire-Energy System for the first time, contributing to a higher spatial and temporal resolution to position the wildfire risk in a grid. A novel method is proposed to automatically correlate the wildfire risk index and the power network model using geographical information of the transmission lines. The optimal power flow and grid performances are obtained from a grid model which incorporates wildfire risk distributions. It is shown that a system with installed microgrids can maintain operation under severe fire-related conditions without scheduled or unplanned outages. Finally, a cost-benefit analysis is conducted, which demonstrates that 68% of system costs can be recuperated by implementing networked microgrid solutions. conductor lines due to the high temperatures during the wildfire season and flashover events [6]. i Applied Energy 313 (2022) 118793 Applied Energy 313 (2022) 118793 * Corresponding author. E-mail address: weijia.yang@eng.ox.ac.uk (W. Yang). * Corresponding author. E-mail address: weijia.ya Available online 8 March 2022 0306-2619/© 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). * Corresponding author. E-mail address: weijia.yang@eng.ox.ac.uk (W. Yang). https://doi.org/10.1016/j.apenergy.2022.118793 Received 15 October 2021; Received in revised form 28 January 2022; Accepted 18 February 2022 Available online 8 March 2022 0306-2619/© 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). https://doi.org/10.1016/j.apenergy.2022.118793 Received 15 October 2021; Received in revised form 28 January 2022; Accepted 18 February 2022 1. Introduction This control system enabled more load portions to be served in a high wildfire risk period. However, once the overloaded lines exceeded the tolerable time for the protection relay without microgrid supports, it would lead to load shedding and line outages [25]. This method may cause a powerline fire if the highly loaded lines operate for an extended period [6]. Thus, a strategy considering microgrids is motivated to sustainably serve more load without powerline fires and system failure for longer durations. i [11]. In regions prone to wildfires, managing these low-probability and high-risk events has required frequent short-term disconnections, which affect millions of customers and add indirect social loss on top of physical loss. The conventional power system has a one-way structure, with power delivered from generation to transmission, distribution, and demand [12]. Along with this structure, long-distance, high-voltage power lines traversing heavily forested areas are particularly vulnerable to wildfires [4]. According to a recent article in the San Francisco Chronicle, phys ical methods like undergrounding and better insulation of the overhead lines have been outlined by several states to enhance the grid resilience (defined as the ability of a grid to sustain normal operation under extreme conditions) with a greater wildfire risk [13]. Resilience is one of the key grid indicators, which can be improved by network design and physical reinforcements. Network resilience is defined as the ability of a power grid to sustain normal operation when facing high-impact, low probability events like natural hazards [14]. A resilient system should be endowed with fault tolerance, quick response, restorability, and reli ability when facing disruptive events [15]. Both the load shedding rate and the line loading rate can be utilised to measure system resilience [16]. In our paper, four grid resilience metrics are utilised to assess the system resilience performance in different scenarios. The definitions, units and selection reasons for the metrics will be discussed in detail in Section 2.5. i This paper proposes the novel use of microgrids to manage wildfire risks within power systems without resorting to power outages. As an extensively validated measure of fire potential, the Canadian Forest Service Fire Weather Index Rating System (FWI) is innovatively utilised to describe the wildfire risk distribution over a power grid. The previous study [4] utilised a static wildfire risk distribution model that was spatially partitioned into only three levels based on loosely assembled data from [26]. 1. Introduction In recent years, the frequency of extreme weather events has increased worldwide, in part due to climate change, resulting in sig nificant financial and social losses to both power system operators and consumers [1]. Wildfires are initiated by different causes, either natural (such as lightning strikes) or human (e.g., accidental sparks, or delib erate arsons). The predominant cause of wildfires is different in various regions, e.g., human factors account for about 42.9% of ignitions in Victoria, Australia [2], whereas 95% of ignitions attribute to human activities in California [3]. Fuel flammability, low humidity, and high wind conditions can all increase the probability of extreme wildfire conditions [4]. With increasingly frequent wildfires, heat damage to grid infrastructure can occur and lead to the introduction of further hazards [5]. Specific hazards to the energy infrastructure include sagging Wildfire risks to the grid have become an international phenomenon in recent years. For instance, from December 2019 to February 2020, Australia faced an unprecedentedly intense and devastating wildfire season [7]. Multiple wildfires broke out in all states in Australia, particularly in Victoria, resulting in 34 fatalities and the destruction of more than 10 million hectares of land [8,9]. Approximately 20,000 households’ power supplies were disconnected at the peak of wildfires during the New Year period in New South Wales [10]. Also, in 2018, a transmission line belonging to California’s largest utility, Pacific Gas & Electric (PG&E), started a wildfire that led to the company pleading guilty to 84 counts of manslaughter and destruction of the Town of Paradise [11]. This devastating wildfire caused around $13.5 billion of damage to property, such that PG&E was forced to declare bankruptcy * Corresponding author. E-mail address: weijia.yang@eng.ox.ac.uk (W. Yang). W. Yang et al. Applied Energy 313 (2022) 118793 (API) for Malaysian PV in 2014 [24]. More comprehensive influence factors should be considered in the future. Rhodes et al., 2020 created an optimisation model to mitigate powerline fire risk, based on the Institute of Electrical and Electronics Engineers-Reliability Test System of the Grid Modernization Laboratory Consortium test case in Southern Cali fornia [4]. Power Shut-off methods were used to allow lines that were important for load delivery to continue to run even with high loading and high wildfire risks, whilst other lines with less load were discon nected at a lower wildfire risk. 1. Introduction In comparison, in our study the wildfire risk is measured daily and regionally with exact ‘Danger Ratings’ to achieve a higher accuracy. In addition, a cost-benefit analysis of the proposed microgrid method is conducted to prove its feasibility and sustainability. The method in our paper is adaptable and can be applied to protect power networks in other fire-prone areas. The Wildfire-Energy System (here after the Test System) represents an example case of our model based on a practical grid scenario of Victoria, Australia, in the 2019–2020 wildfire season. The Test System consists of two main components: the Wildfire Index that calculates the wildfire risk distribution for Victoria, and the Grid Model that simulates the power network of Victoria, hereby refer to as the Wildfire Index and the Grid Model. Considering the economic and environmental limitations of the physical reinforcements, network im provements with microgrids are proposed in this paper to enhance the grid resilience during fire seasons. The cost of physical reinforcements is prohibitively expensive: it is estimated that it will take over $100 billion for PG&E to underground all its high voltage lines across their entire network (2/3 of California). Besides the significant expense, there is concern about the environ mental effects of undergrounding in bio-diverse regions [13]. With an increasing pursuit of more robust and clean grid operations, the regulation and enhancement of grid resilience using renewables have attracted more public attention [17]. One such method is the use of microgrids. A microgrid is a subset of the network that can operate independently from other parts of the network, made up of small-scale distributed generations, local demand, and energy storage systems. In microgrid strategies, various Distributed Generators (DGs) are installed in the vicinity of end-users, such as photovoltaic (PV) panels and small wind turbines [18]. The capacity of DGs can range from several kilo watts to 300 MW [19]. i In Section 2, the methodology to construct the Test System with geographical data is discussed in detail. The overview and dataset sources of the three scenarios are described in Section 3, aiming to explore how overall wildfire risks, FWI line disconnection thresholds and microgrids influence grid resilience. In Section 4, the results for the resilience performance analysis are demonstrated and analysed with four parameters, including line loading, load shedding, system operating costs, and carbon emission factors. 1. Introduction The cost-benefit analysis for micro grid solutions is also discussed, and Section 5 summarises the main findings and advises potential future developments to the Test System. In our paper, various capacities of distributed generators are installed at each node of a transmission network to share the load burden during wildfire seasons. According to regional wildfire risks, parts of the network transmission lines are disconnected, and the cor responding local demands are supported by the distributed generators to avoid load shedding as much as possible. The remaining operating lines in “fire-free areas” keep the power channel between nodal demand and the grid generation unobstructed. In this case, the nodal demand is partially supplied by either the main grid or the local distributed gen eration depending on the economic benefits (e.g., the real-time elec tricity price). The two operation modes above can represent the grid connection (i.e., normal mode) and the isolated islanding mode, thus being treated as a potential microgrid method to enhance the grid operation under extreme events. 2. Methodology In this section, we describe the proposed method following the model formulation process. The Wildfire Index calculation and the Grid Model construction are first explained in Sections 2.1 and 2.2. A spatial analysis technique is then described to match the Grid Model with the Wildfire Index using the coordinates of transmission lines (as explained in Sec tion 2.3). In this study, microgrid solutions are utilised to share regional demand when bulk supply is significantly reduced due to wildfires. Designs of microgrids and automatic line control strategies are also discussed in this section (as explained in Section 2.4). Jazebi et al., 2020 provided a comprehensive review of previous wildfire management techniques in engineering-related fields. It was found that existing literature mainly focused on how wildfires may physically damage lines in high wildfire risk areas [20,21]. However, wildfire risks can affect grid performance without direct damage, e.g., soot accumulation may lead to current leakage and line off [5]. Solar photovoltaic power is not a recommended local DG source in fire-prone regions since the wildfire smoke negatively affects the power generation – the overall electricity reduction is 7%, with the peak power shedding of 27% in a fire-burn case study in Canberra [22]. Similar phenomena have been found in California and Malaysia – a 30% reduction in average electricity generation in California during September 2020 [23] and 0.43 W power reduction per increment of 1 point Air Pollution Index Furthermore, the simulation methods used to assess system resilience are described as the critical analysis of this paper in Section 2.5. Direct Current Optimal Power Flow (DCOPF) has been implemented as the main simulation to calculate the optimal grid solution under system safety conditions, producing various performance indices. Fig. 1 shows the main components and the main steps to obtain grid resilience under 2 Applied Energy 313 (2022) 118793 W. Yang et al. Fig. 1. Main components and construction steps flow chart for the Wildfire- Energy System. g conditions [30]. In most cases, FFDI and FWI both offered similar climatological patterns of wildfires and resulted in similar warnings for wildfire disasters [31]. Thus, the wildfire distribution is relatively in dependent of the index used. Oldenborgh et al. have applied the FWI to analyse the attribution of Australian bushfires [32]. While FWI was originally derived from Canada, it has been shown to be applicable in many different climatic regions (e.g., Mediterranean Europe [33,34]). 2. Methodology Therefore, FWI was utilised in this study based on the universality and the availability of the index. The FWI of each microclimate region is available from the Coper nicus Climate Data Store [29] in Network Common Data Format (netCDF) files [35]. A microclimate is defined as a region with specific climate conditions and topographic locations [36]. This paper describes a microclimate as a region with a different wildfire risk feature from neighbouring regions. For the case study area, the whole studied region was divided into 0.25◦×0.25◦grid cells in line with the spatial resolu tion of the wildfire risk distribution data. As the geographical coverage of the data is on a global scale, the FWI distribution can be calculated for other regions using the same method [32]. As an example of wildfire risk distribution, Fig. 2 shows the Australian FWI for the 30th of December 2019, highlighting the wildfire risk danger level. i In previous studies, such as [32,33,34], FWI is used as a measure of fire risk irrespective of ignition sources. Determining the actual risk of powerline ignition at a particular FWI value is beyond the scope of this paper as we are unable to quantify the likely occurrence of all possible ignition sources at all locations. This is especially true given a number of ignition sources that are the direct result of human behaviour. In França et al., the FWI was utilised as one of the three main elements to assess the fire-risk-breakdown of the power lines in Northern Brazil [37]. Thus, the FWI was regarded as a reference index here to help determine the proper operating mode of the grid component within the Grid Model to avoid a larger scale outage. In practical power systems, the system operators should design a fire prevention plan considering both the regional FWI and possible ignition sources nearby. Fig. 1. Main components and construction steps flow chart for the Wildfire- Energy System. fire risks in our model. This section describes general methods of power grid analysis under wildfire risks for increased replicability and model implementation across other regions of interest. 2.1. The wildfire index Three main steps to construct the Grid Model are explained in detail in this section. The power network was set up within the Grid Model in Step 1, the power generation information was added in Step 2, and the energy demand distribution was obtained in Step 3. The detailed con struction procedures and data sources specific to the Victoria Case study are discussed in detail in Section 3. The simulation of the wildfire risk distribution is one of the critical parts for the Test System design. Wildfire risk levels can have substantial simultaneous differences within a large country. Therefore, a better temporal and spatial accuracy of the wildfire risk severity is crucial to model the Test System better. A weather-dependent risk index was utilised in our study to simulate the wildfire risk distribution during the 2019–2020 wildfire season. The FWI is a meteorological index that is used to evaluate the risk of wildfires [27]. The FWI ingests meteorological parameters such as temperature, wind speed, relative humidity and 24-hr precipitation [28]. The FWI Danger Rating (DR) is classified into six levels, and the relationship with FWI is described in Table 1. It is notable that our model focuses on grid formation and static power dispatch rather than dynamic energy dispatch. Our grid resilience Fig. 2. Australia FWI colourmap on 30th Dec 2019. An alternative index is the McArthur Forest Fire Danger Index (FFDI), developed based on Australia’s wildfire data [29]. According to findings from Dowdy et al., 2009, both FWI and FDI were mostly sen sitive to the wind, relative humidity, and temperature. The percentile of the two indices showed a consistent representation under severe fire Table 1 FWI Danger Rating classification adapted from [28] (upper bound excluded). Fire Danger FWI Ranges Very Low (0, 5.2) Low (5.2, 11.2) Moderate (11.2, 21.3) High (21.3, 38.0) Very High (38.0, 50.0) Extreme (50.0, +∞) Fig. 2. Australia FWI colourmap on 30th Dec 2019. 3 W. Yang et al. Applied Energy 313 (2022) 118793 analytical parameters are obtained for the worst-case scenario (when the grid is operating at peak load). The current work is assessed with dy namic wildfire risk simulations and grid at peak demand. We only consider wind generation with a capacity factor. However, future work could consider more detailed modelling of local microgrid load profiles, which would allow us to investigate the value of battery storage. Step 2: Generator Dataset While the power station positions were given in Step 1, technical and economic parameters were assigned to generation information. The technical data of generators were collated, including fuel types and registered capacities. Databases containing the unit technical parame ters and operating costs were imported to supplement the information in this step. After filtering generators that did not satisfy the system con ditions (e.g., standing exemption of small-scale DGs according to the local authority), a comprehensive dataset that contained all the neces sary generation information was obtained. Step 1: Network Assembly A Keyhole Markup Language (KML) file is a file format to display geographic data containing pinpoint locations and image overlays [38]. As input, raw KML data were downloaded to obtain the geographical and electrical information about transmission lines, substations, and power stations [39]. State and boundary data were imported in the format of shapefiles [40]. The Statistical Area Level 2 (SA2) population dataset [41] was obtained to estimate the regional demand distribution by the Voronoi tessellation method [42]. Then, the operational grid components belonging to National Electricity Market (NEM) zones were sorted based on the information from energy market operators [43]. As an example, the operational grid and the demand proportion served by each node within the state of Victoria are shown in Fig. 3. In the microgrid scenario, networked microgrids were installed at each node of the Grid Model to enable the regional grid to transform between grid-connection and islanded mode flexibly. Different pro portions of nodal demands were served with additional DG integrations to investigate how much load shedding of the Test System can be avoided with various capacities of microgrids. To be more specific, nodal demands were calculated from Step 3. Different nodal demand proportions were then supported with DGs to share the local demand, particularly at nodes isolated due to high fire risk. Finally, the whole system performance was assessed by comparing the result of resilience parameters (Table 2). A cost-benefit analysis was conducted to seek a better trade-off between the cost of load shedding and the microgrid installation solution. y g y Networked Microgrids Networked Microgrids 2.1. The wildfire index for the power flow simulation in Pandapower [39]. However, most of the public load dispatch data were regional rather than nodal demands divided into smaller clusters. The nodal demand distribution was ob tained using the aforementioned method of Voronoi tessellation. Based on the output from Step 1, the nodal demand was attained by multi plying the total demand by the nodal proportion. 2.4. Combining wildfire and grid information In combining the Wildfire Index with the Grid Model, methods to assess and determine the operational status of grid components are essential. As in most networks, transmission lines, especially high voltage lines, are geographically extended over a large area [45]; long transmission lines may cross different regions with separate and different wildfire risk conditions due to their microclimates. For instance, Victoria has a major transmission line (from Red Cliffs to Kerang), traversing different microclimates as shown and marked with green dots in Fig. 3. Long-extending transmission lines result in diffi culties in making well-justified operational decisions as various line segments may have different failure probabilities. i Finally, the maximum FWI value was selected along each line and compared with the preset FWI threshold under different fire control conditions. A switch was opened within the Grid Model if the maximum FWI along a line exceeded the limit. Similarly, in practice, lines are disconnected to prevent wildfires, followed by customers being disconnected to ensure the power grid is not overloaded during wildfire seasons [51]. Table 2 Table 2 Metrics comparison for resilient system operation (technical, economic, environmental and trade-off). NO Standard Name Mathematical Expression Terms Comment (a) Technical Line Loading Percentage Currentline Ratingmax × df × parallel Ratingmax : maximum rating df : derating factor parallel : Number of parallel lines Both maximum line loading rate and the system average loading rate are investigated (%) (b) Technical Load Shedding Proportion SD-AC SD SD : System Demand AC : Actual Consumption The reduced load divided by the total system demand (%) (c) Economic Short-Run Marginal Cost VOM + Fuelcost × Heatrate VOM :Operations & Maintenance Cost The short-run marginal cost is computed in the DCOPF process (AU$/MWh) (d) Environmental Emission Factor ∑(Emissionintensity×Gen) Totalgen Gen: Electricity Generation Totalgen: Total Generation System becomes eco-friendly as the emission factor decreases (kg CO2eq/kWh) (e) Comprehensive Trade-off Detailed explanation in Section 4 Load shedding cost VS DG cost Metrics comparison for resilient system operation (technical, economic, environmental and trade-off). To be more specific, the coordinates in the Australian transmission line KML files are with a spatial resolution of 0.001˚[44]. Thus, KML files were chosen here. i switches. When any part of the transmission line crossed an area exceeding the preset risk threshold, the switch on the line was auto matically disconnected. In Scenario 3, part of the network connected to a disconnected line would then operate in an islanded mode with DGs to reflect the operation of microgrids. In terms of spatial information within the Wildfire Index, the original netCDF4 data of FWI were calculated for discontinuous grid cell vertices. To overlay the value over the entire region, each sampling point was first selected as the centre of one grid cell. Then the FWI of this grid region was assumed to have the average FWI of the four surrounding vertices. l In the line disconnection decision process, the sampling accuracy can be changed according to the resolution of the data. For instance, since the grid cell length was 0.25◦in Victoria, the coordinates along each line were extracted in steps of 0.01◦latitude to achieve good spatial accu racy. After obtaining the transmission line coordinates, the corre sponding FWI grid cell covering each line sampling point was identified. As the FWI of each regional cell had been computed, each coordinate along the line was assigned with a value of FWI. 2.3. Geospatial modeling The length and the trajectory of transmission lines are essential for the Test System, primarily because some electrical parameters (e.g., resistance) are closely related to the line length. Another reason is that knowing the exact path of lines (rather than only start points and end points) is essential for accurate wildfire risk detection of lines that cross multiple microclimates. To complement the Grid Model with network line geographical information, KML files containing multiple co ordinates were utilised to describe the line paths with a higher accuracy. Step 3: Historical Load and Dispatch Historical load signals were required to provide the actual demand Fig. 3. High voltage transmission network in Victoria, Australia, where “+” represents substations, the size of red circles represents regional energy demand pro portion, the two green spots indicate a 220 kV transmission line from Red Cliffs to Kerang, which will be discussed as an example in Section 2.4. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) Fig. 3. High voltage transmission network in Victoria, Australia, where “+” represents substations, the size of red circles represents regional energy demand pro portion, the two green spots indicate a 220 kV transmission line from Red Cliffs to Kerang, which will be discussed as an example in Section 2.4. (For interpretation of he references to colour in this figure legend, the reader is referred to the web version of this article.) Fig. 3. High voltage transmission network in Victoria, Australia, where “+” represents substations, the size of red circles represents regional energy demand pro portion, the two green spots indicate a 220 kV transmission line from Red Cliffs to Kerang, which will be discussed as an example in Section 2.4. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) Fig. 3. High voltage transmission network in Victoria, Australia, where “+” represents substations, the size of red circles represents regional energy demand pro portion, the two green spots indicate a 220 kV transmission line from Red Cliffs to Kerang, which will be discussed as an example in Section 2.4. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) Applied Energy 313 (2022) 118793 W. Yang et al. 2.5. Simulation methods Two main methods have been proposed to define the failure proba bility of system components that span multiple microclimate regions. The first utilised the weighted-average method: an incremental multi plier of failure rate (IMFR) can obtain the overall failure rate by adding the point-specific failure rates along each line segmentation according to their respective microclimate regions [46]. In the second category, the failure rate for the entire transmission line was determined by the highest failure rate along with any one single point of a line [47]. Grid performances (metrics from Table 2) were obtained from an optimal power flow model incorporating wildfire risk distributions [52]. The Optimal Power Flow (OPF) is the best operating situation of power stations to meet the grid demand, determined by network variables, constraints, and the specific objective of the OPF model utilised. The most common objective for OPF is to minimise operating costs [52]. OPF can be classified into Alternating Current OPF (ACOPF) and Direct Current OPF (DCOPF). ACOPF problems are typically approximated and solved by DCOPF that focuses exclusively on active power constraints due to its linearity and solvability [53]. According to the Australian government, lines with a voltage greater than 66 kV are defined as transmission lines. Those at lower voltages are classified into the distribution level [48]. Thus, only transmission lines with voltage levels over 66 kV were considered in this paper. And the focus was given to cases in which transmission lines cross various microclimate regions. As shown in Fig. 3, most lines crossed different microclimates with different failure rates during the wildfire season. As minor failures can instantaneously have a great impact on the wider high-voltage transmission lines, this paper assumed that the worst fail ure rate along the line was selected to judge the whole line’s failure and following operational modes. y y [ ] The DCOPF is a built-in function in pandapower. The system completed the calculation procedures if the load and generation ach ieved balance with all constraints satisfied. According to the parameters of each generator, the optimal power flow followed the objective to minimise the system load shedding in the DCOPF. The Test System simulation was initialised using the wildfire, network, generator, his torical load dispatch, and geospatial data. Then, DCOPF was carried out to produce system resilience parameters under different fire conditions. 3.1. The test system overview To start with, the original test system without microgrid integration was implemented based on the five-day data with different levels of wildfire risks. Different wildfire risk conditions can negatively impact line conductor temperatures and flowing current capacities [14]. Thus, the relationship between energy system resilience and the overall wildfire risk severity was studied in Scenario 1. The overall wildfire severity rose throughout December 2019. Therefore, five days were selected as these days can evenly represent the wildfire risk severity level growth from the lowest to the extreme in December (1st, 3rd, 5th, 18th, 30th, December 2019). Fig. 4 displays various wildfire risk levels on the five selected days. The detail and the result analysis of each scenario will be discussed in Section 4. To better apply the proposed methods to real-world network oper ations, a real power grid was selected in this paper. As mentioned in Section 1, Australia was attacked by a devastating wildfire from December 2019 to February 2020. Compared to other states in Australia, Victoria has some factors that make it suitable for a case study, such as large transmission networks, highly varying wildfire risks and good data availability. Thus, ‘Victoria, Australia’ (hereafter ‘Victoria’) in December 2019 has been selected as the research region of our case study. Following the three steps explained in Section 2.2, the grid model for Victoria was built. In terms of the input for Step 1, ‘Network Assembly’, raw KML data from Geoscience Australia were utilised to obtain the electrical and geographical information on transmission lines [44], substations [56], and power stations [57]. State and boundary data in the format of shapefiles were accessed from the Australian Statistical Geography Standard in the Australian Bureau of Statistics [40]. Ac cording to the Regional Population Growth Module in the Australian Bureau of Statistics [41], the SA2 population datasets were obtained to estimate the regional demand distribution. NEM zones information were utilised to filter out operational grid components belonging to the Australian Energy Market Operator (AEMO) [43]. The setting of FWI line disconnection thresholds may affect the system operation safety and effectiveness. To be more specific, each transmission line has a calculated FWI value, and more lines will be disconnected as the FWI disconnection threshold is set at a lower value. A lower FWI line disconnection threshold means achieving a stricter fire risk control circumstance. 2.5. Simulation methods Different metrics can be utilised as standards in grid resilience as sessments, such as line loading percentage, load shedding percentage, increased economic profit, reduced death, ecological impact and so on [16]. The Short-Run Marginal Cost (SRMC) can describe the cost of producing one additional unit of service [54]. In the electricity market, SRMC is computed from the efficiency, operation and maintenance costs (O&M) and the fuel cost [55]. SRMC was utilised to evaluate the eco nomic profit of the system in our study. The DCOPF is a built-in function in pandapower. The system completed the calculation procedures if the load and generation ach ieved balance with all constraints satisfied. According to the parameters of each generator, the optimal power flow followed the objective to minimise the system load shedding in the DCOPF. The Test System simulation was initialised using the wildfire, network, generator, his torical load dispatch, and geospatial data. Then, DCOPF was carried out to produce system resilience parameters under different fire conditions. i Different metrics can be utilised as standards in grid resilience as sessments, such as line loading percentage, load shedding percentage, increased economic profit, reduced death, ecological impact and so on [16]. The Short-Run Marginal Cost (SRMC) can describe the cost of producing one additional unit of service [54]. In the electricity market, SRMC is computed from the efficiency, operation and maintenance costs (O&M) and the fuel cost [55]. SRMC was utilised to evaluate the eco nomic profit of the system in our study. The Test System operated according to simulated information from both the Wildfire Index and the Grid Model. The operation method here is similar to the real-world solution: transmission lines have reclosers installed which can be automatically disconnected if a fault or overload occurs [49]. To isolate the section of the power network under high fire risks and avoid a larger-scale outage, a controllable line-bus switch was set on each transmission line in the Test System [50]. While the worst line failure rate (depending on the worst fire risk exerting on the line) was available, the line operational mode was then controlled by line i In the OPF result, four metrics were chosen for resilience analysis in 5 5 W. Yang et al. Applied Energy 313 (2022) 118793 cell were both 0.25◦. Thus, there were 836 grid cells in the selected research region in our model. 2.5. Simulation methods It is noticeable that one degree represents different distances at different latitudes [61] (e.g., 0.25◦denote 28 km at 0◦, whereas 22 km at 36◦S in Victoria). As the latitude range of Victoria is not too wide, the side length of one grid cell can be regarded consis tently as 22 km. i this paper. The performance parameters are displayed and explained in Table 2, which contains the technical, economic, and environmental standards as well as the measurement units. 3. Case study The initial system was simulated based on five days with varying wildfire risks to learn overall wildfire risk impacts on grid resilience in Scenario 1. The FWI line disconnection threshold and the DG proportion were varied in Scenarios 2 and 3 to explore how they may affect system resilience. In this section, the Test System prototype in Victoria, Australia, is first introduced, and then the details of three scenarios to analyse grid resilience performance are viewed. Of these, the first is to explore how overall wildfire risk levels affect grid resilience. Then two others are presented, i.e., preset FWI disconnection thresholds for lines and various capacities of dispatchable microgrids. 4.1. Scenario 1 overall wildfire level influences The Test System Formulation and Simulation 4. Results and discussion Main Components and Parameters In this paper, the Victoria network was filtered so that only high voltage lines were retained. Thus, only lines with voltage levels over 66 kV were kept to focus on the high voltage system. Power stations with less than 5 MW ratings were also filtered as AEMO allowed standing exemptions for these small plants, which should not be categorised as generators [60]. There were 201 transmission lines within the NEM operational region in Victoria before the filter procedures. The infor mation on the remaining system components is provided in the following paragraph (also displayed in Fig. 3). The results are analysed through technical, economic, and environ mental aspects for each scenario. As mentioned in Table 2, four per formance indices are obtained and assessed in each scenario, including system line loading percentage, load shedding rate, system operating cost and emission factor. i There were different operating cases defined for each scenario. In Scenario 1, five operating cases were selected from different days, rep resenting different system fire risk levels. Three operating cases were set up in Scenario 2 with various FWI disconnection thresholds for power lines. In Scenario 3, four operating cases were defined based on various proportions of microgrid integration to the Test System. There were 72 buses in the Test System with multiple voltage levels of 500 kV, 330 kV, and 220 kV. There were 68 transmission lines in the network, including 16 × 500 kV lines, 1 × 330 kV lines, and 51 × 220 kV lines. 84 power stations were in operation in Victoria at the time of data collection, sourced by brown coal, natural gas, hydro, or wind. 4.1. Scenario 1 overall wildfire level influences 3.1. The test system overview Scenario 2 aimed to explore how the FWI line disconnection threshold affected power grid resilience under different fire control conditions. i As the main objective of this paper, microgrid solutions were applied to mitigate the system burden during the wildfire season. As previously mentioned in Section 2.2, power demand at each node was obtained from the historical grid dispatch database [39]. A previous grid rein forcement project has proved that 75% of grid demand can be supported by microgrids, feasibly maintaining a normal operation [62]. In Sce nario 3, the installation capacity of DGs was based on the node load in the Grid Model. Thus, 0%, 25%, 50% and 75% of the node load were selected as renewable capacities added at each node, respectively. The outage and microgrid costs were analysed and compared in this scenario. Then we moved to Step 2, ‘Generator Dataset’, the technical data of generators were collated from the Market Management System Data Model (MMSDM) [58]. The National Transmission Network Develop ment Plan (NTNDP) database also contained information about the unit technical parameters and the operating costs [59]. The technical and economic generator dataset was constructed in Step 2. As for Step 3, ‘Historical Demand and Dispatch’, the nodal demand proportion was obtained using the Voronoi tessellation method. The nodal demand was calculated by multiplying the nodal proportion by the total demand [58]. The Test System Formulation and Simulation In Scenario 1, the Test System was simulated with different levels of wildfire risks based on five-day data in December 2019. Both the average FWI and the maximum FWI in the Test System gradually rose from the operating case for 1st December to the operating case for 30th A geographical rectangle region that could contain the entire Vic toria territory was selected with the latitude range of [−39.25◦, –33.75◦] and the longitude range of [140.50◦, 150.00◦]. Since the resolution of FWI reanalysis data was 0.25◦×0.25◦, the length and width of each grid 6 Applied Energy 313 (2022) 118793 Applied Energy 313 (2022) 118793 W. Yang et al. Applied Energy 313 (2022) 118793 Fig. 4. Scenario 1: FWI colourmaps for five selected days in the 2019–2020 fire season in Victoria, Australia to explore the relationship between overall fire severity and grid resilience (1st, 3rd, 5th, 18th, 30th Dec 2019). An animated video displaying the dynamic FWI during the 2019–2020 Fire Season in Victoria, Australia, is linked here. W. Yang et al. Fig. 4. Scenario 1: FWI colourmaps for five selected days in the 2019–2020 fire season in Victoria, Australia to explore the relationship between overall fire severity and grid resilience (1st, 3rd, 5th, 18th, 30th Dec 2019). An animated video displaying the dynamic FWI during the 2019–2020 Fire Season in Victoria, Australia, is linked here. December. Since the demand may impact system resilience, grid de mands of the following four operating cases were scaled down to the same for 1st December to control the variables (the total system demand). i 131.76%), 3rd (86.27%, 149.23%) to 5th (86.63%, 163.64%). It is notable that the system loading sharply turned to drop in the last two days. The system loading was closer to zero in the next two operating cases as more lines were disconnected than pre-disconnected, causing a large-scale outage. Regions encounter an extreme wildfire risk if the FWI is greater than or equal to 50.0 [29]. In other words, the transmission line is likely to cause powerline ignition in this situation. Therefore, lines sitting in a grid cell with 50.0+ FWI were pre-disconnected automatically in Sce nario 1. Moving to the second evaluation metric, the load shedding and the remaining load in Scenario 1 are displayed in Fig. 5. More lines were automatically switched off as the wildfire risk level increased, resulting in more load shedding. The Test System Formulation and Simulation Nearly 100% of the load was curtailed in the operating case for 30th December. Various lines were switched off as different numbers of lines excee ded the preset FWI threshold (50.0) in the five operating cases. The number of disconnected lines after the OPF simulation remained the same as before if there was no vast load shedding. However, compared to the initial number of pre-disconnected lines, due to a high FWI value, more lines were forced to become out of service if they violated the system safety conditions during the OPF simulation process. For instance, the initial FWI disconnection condition automatically switched off five lines in the operating case for 5th December. In contrast, the final disconnection rose to 22 out of 68 lines, as shown in Table 3, resulting in a large-scale blackout. Without the support of local DGs, large-scale outages can be more frequent due to higher wildfire risks. i The third evaluation is to assess the total and the unit system oper ational cost. As the wildfire risk developed, the total cost of operation gradually declined since the remaining operational grid was greatly curtailed due to a higher fire risk. Though the total amount decreased, the SRMC rose as the served demand dropped at a higher rate. The Value of Lost Load (VoLL) represents the cost of disconnection to The Value of Lost Load (VoLL) represents the cost of disconnection to Fig. 5. Load shedding comparison for days with elevated fire conditions in Victoria, Australia (Scenario 1). i As for the first evaluation metric, there was a slow growth in both the average and the maximum line loading percentages from 1st (85.88%, Table 3 Number of disconnected lines in Scenario 1. (Pre-OPF are lines switched off due to the exceeding of the preset high fire risk threshold before the OPF operation; post-OPF are regional power outages due to safety boundary violation after the OPF implementation). Day FWI Disconnected (Pre-OPF) Actual Disabled (Post-OPF) 1st Dec 2019 1 1 3rd Dec 2019 2 2 5th Dec 2019 5 22 18th Dec 2019 17 62 30th Dec 2019 68 68 Fig. 5. Load shedding comparison for days with elevated fire conditions in Victoria, Australia (Scenario 1). Fig. 5. Load shedding comparison for days with elevated fire conditions in Victoria, Australia (Scenario 1). 7 W. Yang et al. Applied Energy 313 (2022) 118793 operating costs in the first two operating cases (with the FWI thresholds of 50.0 and 38) remained the same since there was no load shedding in the first two operating cases, and the actual consumptions were the same. However, the total system operating cost dropped, and the unit operating cost rose once there was a load shedding, as shown in the third operating case. Thus, under a stricter fire risk control limit, the lower the FWI line disconnection threshold, the higher the unit cost. customers [63]. It should be noted that VoLL was excluded from the simulated result of SRMC in all scenarios since VoLL can vary largely by season, time of the day, end-user demand and customer type (i.e., the VoLL for public infrastructures like hospitals is higher as the indirect loss for a power outage there is higher). VoLL was not necessarily used in the OPF simulation as the main objective of the OPF was to achieve mini mum load shedding. However, the VoLL was still utilised in the cost- benefit analysis since it generally contributed to a significant cost increment [64]. Regarding the total system emissions, it remained at the same level as the threshold decreased. The unit carbon emission became higher as the actually served demand decreased from the loose fire control oper ating case (with the FWI threshold of 50.0) to the strict one (with the FWI threshold of 21.3). With regarding to the fourth evaluation metric, in Scenario 1, there was no newly integrated renewable source, thus the generation fuel types determined the system’s overall emission factor. The unit emission factor measured in kg CO2eq/kWh was mainly decided by the actual load consumption in this scenario. 4.3. Scenario 3 networked microgrids In Scenario 2, the Test System based on historical data from 1st December 2019 was tested with different FWI disconnection thresholds for power lines. According to the Danger Rating classification in Table 1, the FWI disconnection thresholds were set at extreme (50.0), very high (38.0), and high (21.3) in three operating cases, representing the fire control condition becoming stricter in Scenario 2. i In Scenario 3, dispatchable energy sources like DGs were integrated into the network with different capacity levels to mitigate load shed ding. 0%, 25%, 50% and 75% of the node load capacity were injected with DGs at each node of the Grid Model in the four operating cases, respectively. In Scenario 3, the Test System based on historical data from 12th December 2019 was utilised since the Test System without micro grid support on this day had a load shedding of around 70% due to high wildfire risks. Various proportions of microgrid capacities were expected to alleviate the electrical outage to different extents. Cost-effectiveness was also evaluated and compared between load shedding compensa tion and microgrid connections budgets. i Regarding the first evaluation metrics, both the maximum and the overall line loading percentages went up as the FWI disconnection threshold declined. Among the three operating cases, the average loading percentages retained the same order of magnitude (around 86%). In comparison, the maximum line loading gradually rose from 131.76% (operating case with the FWI threshold of 50.0) to 163.64% (operating case with the FWI threshold of 21.3) as more lines were disconnected due to a stricter limit of fire control. To begin with the first evaluation metric, Fig. 7 shows the maximum and the average line loading for four operating cases. As the DG capacity went up, the maximum and the overall system loading percentages decreased by 29.8% and 53.1%, respectively. The DG installation shared part of the local load, which led to a lower grid burden. i For the second evaluation metric, as the FWI line threshold decreased, the load shedding increased as 7% of lines were pre-disabled in the third operating case, and the rest of the Gird Model could not hold the burden. The system maintained regular operation in the first two operating cases (with FWI thresholds of 50.0 & 38.0) since only one or two lines of 68 were switched off. Table 3 Overall emissions became lower from the start to the end of December due to the operational generation decreasing gradually under higher fire risks. As for the unit emission factor, the unit emission factor increased by 98.76% from 1st December to 30th December, as nearly all load was eventually shed. The grid became more polluting per kWh as the wildfire risk rose. Either an FWI of 38.0 or 50.0 can be selected as the disconnection threshold to prevent powerline wildfires which will not affect the normal operation of the Victoria system. The specific FWI threshold value will be judged by the actual needs in the wildfire control levels. If a greater requirement of wildfire control is expected, demand-side response (DSR) services should be integrated so that 100% of the load can be satisfied. 4.3. Scenario 3 networked microgrids However, the system load was sharply shed (25.21%) as five lines were pre-disconnected in the third operating case (with an FWI line disconnection threshold of 21.3). With the FWI threshold of 50.0 set in this scenario, 14 lines were switched off automatically to prevent large-scale outages due to fire control requirements before the OPF simulation process. Without microgrid supports, there would be substantial load shedding in the original network. In the first operating case without DGs, 36 more lines were indirectly disabled after the OPF simulation due to the 14 pre- disconnected lines. It was notable that no line was indirectly discon nected during OPF once DGs were installed to match 25% of the node The third evaluation metric is displayed in Fig. 6. The total and unit Fig. 6. System operating cost and SRMC with different fire control conditions. FWI thresholds set for transmission line disconnections at 50.0, 38.0, and 21.3, representing the fire control condition becoming stricter in Scenario 2. Fig. 7. Average and maximum system line loading percentages with various capacities of DG installation at each node in the grid (Scenario 3). Fig. 6. System operating cost and SRMC with different fire control conditions. FWI thresholds set for transmission line disconnections at 50.0, 38.0, and 21.3, representing the fire control condition becoming stricter in Scenario 2. Fig. 7. Average and maximum system line loading percentages with various capacities of DG installation at each node in the grid (Scenario 3). Fig. 7. Average and maximum system line loading percentages with various capacities of DG installation at each node in the grid (Scenario 3). 8 W. Yang et al. Applied Energy 313 (2022) 118793 Fig. 9. System emission factors with various capacities of DG installation at each node in the grid (Scenario 3). loads. As the DG integration level increased at the step of 25%, the load shedding gradually decreased (69.33%, 50.47%, 37.80%, 21.67%). The Average Mitigating Rate (AMR) of load shedding between operating cases was obtained to assess how much load was recovered by DG integration. The method to calculate the rate is given in Eq. (1). 4.4. Cost-benefit analysis for microgrid solutions The system costs before and after the DG installation were compared based on Eq. (2). There are various natural resources for renewable energy in Victoria, including wind, hydro, solar, and bioenergy. The existing widely distributed onshore wind farms and ongoing offshore wind farm projects have proved the feasibility of wind power utilisation in Victoria [65]. Victoria is an excellent site for wind power generation with an average wind speed measured at 6.5 ms−1 [66]. SOcost = TScost + VoLL + DGcost (2) (2) SOcost = TScost + VoLL + DGcost where SOcost is the Test System overall cost, TScost is the total system cost which can be obtained from the OPF; VoLL contains the economic impact of load shedding (through multiplying the unit VoLL cost by the load shedding); Levelised Cost of Electricity (LCOE) assesses the cost of generation for a generator over its lifespan, considering the capital cost and the O&M cost together [70]; DGcost is calculated by multiplying the DG capacity by the LCOE of a wind turbine in Victoria. In this paper, we propose hypothetical planning of renewable uti lisation in Victoria. Both wind and solar resources are abundant in Victoria. A demonstration of renewable resource distribution in Victoria is shown in Fig. 10. Referring to the Victorian transmission network in Fig. 8. System cost and SRMC with various capacities of DG installation at each node in the grid (Scenario 3). According to the Australian Energy Regulator [71], the average electricity price for the Victoria network was AU$84/MWh in 2019. Referring to the latest 2021 economic analysis report about power grids in the NEM, the NEM’s Market Price Cap of AU$15,000/MWh is almost the highest in the world [72]. Thus, we assume the VoLL induced by outages in the scenario without DGs amounted to AU$15,000/MWh as an upper limit here. According to statistics from IRENA (International Renewable Energy Agency) [73], the LCOE of a wind turbine was assumed to be AU$64/MWh. The capacity factor of wind turbines was assumed to be 41% [74], which should be considered in the real cost of DGs, as expressed in Eq. (3). DGcost = (DGgen÷CF) × LCOE (3) (3) DGcost = (DGgen÷CF) × LCOE where DGcost is the project cost for DG installation, DGgen is the DG generation, CF is the capacity factor. 4.3. Scenario 3 networked microgrids Furthermore, there is a positive correlation between wind speed and wildfire spread [67], i.e., fire-prone areas are likely to have greater wind power potential. There is a relatively high coincident distribution between the wind abundance and the severe fire risk. In addition, wind power operates more reliably diurnally than solar power. As mentioned in Section 1, wildfire smoke can negatively affect solar generation effi ciency. Therefore, we would select wind as the DG source as an example here [68,69]. The average system operating cost decreased as more local genera tions came onto the network. For one reason, the overall system oper ating cost decreased since the SRMC of renewables was much lower than the centralised fossil generation. In addition, the scale of actual con sumption recovered as the DG developed. As displayed in Fig. 9, the total and unit emissions decreased as more renewable DGs were used, and the actual consumption rose. Both the unit operating cost and the unit carbon emission decreased as more DGs were used. The microgrid strategies can enhance system resilience by reducing load shedding during the period of high wildfire risks. Then, the finan cial losses brought by the load shedding can be eased by microgrid in stallations. However, the installation of the local DGs requires both the capital investment and the O&M cost. The trade-off between the saved cost for load shedding and the lifetime budget for DG installation is assessed in this section. 4.3. Scenario 3 networked microgrids AMR = RL2nd −RL1st TSD × 100% (1) AMR = RL2nd −RL1st TSD × 100% (1) where RL2nd is the remaining load in the latter case (the 2nd, 3rd, and 4th operating cases), RL1st is the remaining load in the former case (the 1st, 2nd, and 3rd operating cases), TSD is the total system demand. The average mitigating rates between the four operating cases were 18.86%, 12.67% and 16.13%, respectively. The overall average miti gating rate was 15.89%, lower than the additional DGs (25%) between cases. Primarily because DGs only undertook the local demand at each node when the node was isolated. In addition, there were transmissional power losses. Thus, the AMR was not just equal to 25% between every two operating cases. Fig. 9. System emission factors with various capacities of DG installation at each node in the grid (Scenario 3). Fig. 9. System emission factors with various capacities of DG installation at each node in the grid (Scenario 3). Regarding to the economic evaluation in Fig. 8, the total system operating cost in the first column was relatively lower since 70% of the load was shed in the first operating case without microgrids. In the subsequent three operating cases, there was a surge of 9% in the total system cost first. Then it gradually decreased as cheaper DGs were connected (the O&M cost comparison of DGs and fossil fuel plants will be discussed in Section 4.4). The increase in the total system cost was caused by the system recovery after DG integrations. Fig. 3 and the FWI maps in Fig. 4, wind resource distribution is highly colocated with both the fire risk and the network distribution than solar power. Furthermore, there is a positive correlation between wind speed and wildfire spread [67], i.e., fire-prone areas are likely to have greater wind power potential. There is a relatively high coincident distribution between the wind abundance and the severe fire risk. In addition, wind power operates more reliably diurnally than solar power. As mentioned in Section 1, wildfire smoke can negatively affect solar generation effi ciency. Therefore, we would select wind as the DG source as an example here [68,69]. Fig. 3 and the FWI maps in Fig. 4, wind resource distribution is highly colocated with both the fire risk and the network distribution than solar power. 5. Conclusion and future work This paper has proposed the use of microgrids as a novel strategy to mitigate powerline fire risks without the need for blackouts. The power system resilience performance was assessed in high wildfire risk regions. Victoria, Australia, in December 2019 was selected as an example case study. This paper was driven by the practical need to protect power systems and reduce economic and social loss during a high wildfire risk period. The results obtained from the simulations have shown how can intentional power shut offs and microgrids enhance power system resilience in a financially feasible way. The validated FWI was first utilised in a weather-affected power system, which improved the spatial and temporal resolution of wildfire risk positioning in power grids. i Methodologies to assess grid resilience and to build the weather- affected energy system were introduced and discussed. The control lable power line switch played the role of the bridge between the Wildfire Index and the Grid Model, which was automatically discon nected as the line FWI exceeded the preset thresholds under different fire control conditions. The methodology does not exclusively apply to the Test System in Victoria but can be adjusted to other fire-prone regions, supporting the use of the applied methods across various spatial do mains and wildfire-prone network grids. Table 4 MW) The calculation was based on Eq (2) The DG went up as more increasing variable and ‘ ′ represents a declining variable) Table 4 Cost analysis between intentioned outages without DGs and different proportions of DG installation plans in Scenario 3. DG Penetration DG Capacity DG Cost Load Shedding VoLL Operating Cost Overall Cost (%) (MW) (AU$) (MW) (AU$) (AU$) (AU$) 0 0 0 3,180 47,700,000 348,000 48,048,000 25 1,147 179,000 2,315 34,725,000 382,000 35,107,000 50 2,294 358,000 1,734 26,010,000 363,000 26,373,000 75 3,440 537,000 994 14,910,000 340,000 15,250,000 Cost analysis between intentioned outages without DGs and different proportions of DG installation plans in Scenario 3. increasing variable and ‘−′ represents a declining variable) MW). The calculation was based on Eq. (2). The DGcost went up as more wind turbines were connected to the system. VoLL represented the economic impact caused by load shedding, which was mitigated as more DGs were integrated. The total system operating cost was obtained directly from the OPF simulation, reflecting the system operating cost in various circumstances. The overall cost considering all above was summed up in the last column. The total economic impact was alleviated as more DGs were installed in the Test System. Compared to the initial system, around 68% of the overall cost was eased when the local DGs supplemented 75% of nodal demand. The considerably high VoLL in NEM took the dominant role here as a leading factor in overall system cost saving. Thus, the local renewable generation method was consid ered a financially viable solution to achieve a more resilient power network in Victoria, Australia. 4.5. System summary Among the three scenarios, factors that affect system resilience, economy and environment were simulated and investigated, i.e., the overall fire risk level, the FWI line disconnection threshold and different proportions of microgrid installations. With regard to power system load shedding, DG produced generation mitigated the curtailed demand in a financially feasible way. The unit operational cost was mainly affected by the consumed system demand and the total system cost. The higher the DG proportion, the lower the unit cost. The unit carbon emission factor mainly depended on the served energy consumption, calculated from the OPF simulation. The summary of effects is given in Table 5. (where ‘+′ represents an i Three scenarios were presented to investigate how the Test System was influenced by three main variables, i.e., the overall wildfire risk level, the FWI disconnection thresholds for power lines, and the microgrid proportions. The results showed that the overall line loading percentage was positively correlated with the wildfire risk. The line loading percentage declined as the FWI threshold or the DG proportion went up. Secondly, the change of load shedding had an opposite tendency with the FWI threshold and DG capacity, and load shedding rose as the wildfire risk became higher. As for the unit operational costs, only the severity of wildfire risk negatively impacted costs. The unit carbon emission factor was eased as the FWI thresholds or the DGs increased, but it was worse as the wildfire risk became more severe. Table 5 The Test System performance comparison (where ‘+′ stands for an increasing trend and ‘−′ represents a declining trend). Performances Scenarios I II III Variable Tendency (+) Fire Risk FWI Threshold DGs Line Loading Percentage + − − Load Shedding + − − Unit Operational Cost + − − Unit Emission Factor + − − 4.4. Cost-benefit analysis for microgrid solutions The data from Scenario 3 were analysed in Table 4, with various proportions of DG connections (the total system demand was 4,587 Fig. 8. System cost and SRMC with various capacities of DG installation at each node in the grid (Scenario 3). Fig. 8. System cost and SRMC with various capacities of DG installation at each node in the grid (Scenario 3). 9 Applied Energy 313 (2022) 118793 Fig. 10. Distribution maps of renewable resources in Victoria, Australia. Left: mean wind speed (ms−1) [68]. Right: mean PV power output (kWh/ kWp) [69]. W. Yang et al. W. Yang et al. Applied Energy 313 (2022) 118793 enewable resources in Victoria, Australia. Left: mean wind speed (ms−1) [68]. Right: mean PV power output (kWh/ kWp) [6 Fig. 10. Distribution maps of renewable resources in Victoria, Australia. Left: mean wind speed (ms−1) [68]. Right: mean PV power output (kWh/ kWp) [69]. Table 5 Table 5 The Test System performance comparison (where ‘+′ stands for an increasing trend and ‘−′ represents a declining trend). The Test System performance comparison (where ‘+′ stands for an increasing trend and ‘−′ represents a declining trend). i A higher value in the FWI disconnection thresholds and DG capac ities eased system operational stress, both economically and 10 W. Yang et al. Applied Energy 313 (2022) 118793 [2] Clarke H, Gibson R, Cirulis B, Bradstock RA, Penman TD. Developing and testing models of the drivers of anthropogenic and lightning-caused wildfire ignitions in south-eastern Australia. Journal of Environmental Management 2019;235:34–41. https://doi.org/10.1016/j.jenvman.2019.01.055. i environmentally. According to the cost-benefit analysis, it was proved that the microgrid solution could save 68% of the overall system cost. It provided both a way to improve system resilience economically and feasibly integrate more renewables within the grid. p g j j [3] Syphard AD, Keeley JE. Location, timing and extent of wildfire vary by cause of ignition. International Journal of Wildland Fire 2015;24:37–47. https://doi.org/ 10.1071/WF14024. i FWI data are currently measured daily, at noon Local Standard Time worldwide. However, power systems are usually assessed once every half hour. A more accurate model can be obtained as the temporal res olution of the weather data becomes higher. An FWI curve with a tem poral resolution at half-hour intervals can be linearly formed by FWI related parameter curves. [4] Rhodes N, Ntaimo L, Roald L. Balancing wildfire risk and power outages through optimized power shut-offs. IEEE Transactions on Power Systems 2021;36(4): 3118–28. https://doi.org/10.1109/TPWRS.2020.3046796. [5] Sathaye J., Dale L., Larsen P., Fitts G., Koy K., Lewis S., et al. Estimating risk to california energy infrastructure from projected climate change, 2011. [6] Fire and high voltage transmission line safety – information sheet. POWERLINK Queensland 2015. Since the pandapower is not directly a time-dependent tool for power simulations, no time domain parameter is set in the default power flow calculations. The current system can only model instantaneous and static operations. More frequent monitoring of the grid operation is therefore preferred. More scenarios can be designed in a continuous- time model when energy flow is created rather than just instantaneous power flow, e.g., scenarios about energy storages. [7] Filkov AI, Ngo T, Matthews S, Telfer S, Penman TD. Impact of Australia’s catastrophic 2019/20 bushfire season on communities and environment. Retrospective analysis and current trends. Journal of Safety Science and Resilience 2020;1(1):44–56. https://doi.org/10.1016/j.jnlssr.2020.06.009. [1] Frame DJ, Rosier SM, Noy I, Harrington LJ, Carey-Smith T, Sparrow SN, et al. Climate change attribution and the economic costs of extreme weather events: a study on damages from extreme rainfall and drought. Climatic Change 2020;162 (2):781–97. https://doi.org/10.1007/s10584-020-02729-y. CRediT authorship contribution statement [17] Dyson M., Li B. Reimagining grid resilience a framework for addressing catastrophic threats to the US electricity grid in an era of transformational change, 2020. Weijia Yang: Methodology, Software, Writing – original draft, Writing – review & editing. Sarah Sparrow: Writing – review & editing, Methodology, Supervision. Masa¯o Ashtine: Writing – review & editing, Supervision. David Wallom: Conceptualization, Writing – review & editing, Supervision. Thomas Morstyn: Conceptualization, Methodol ogy, Writing – review & editing, Supervision. [18] Guo Y, Zhao C. Islanding-aware robust energy management for microgrids. IEEE Transactions on Smart Grid 2018;9(2):1301–9. [19] Funabashi T. Integration of distributed energy resources in power systems: implementation, operation and control. Elsevier Inc; 2016. https://doi.org/10.101 6/C2014-0-03911-1. i [20] Jazebi S, de Leon F, Nelson A. Review of wildfire management techniques-Part I: Causes, prevention, detection, suppression, and data analytics. IEEE Transactions on Power Delivery 2020;35(1):430–9. https://doi.org/10.1109/ TPWRD.2019.2930055. i Declaration of Competing Interest [21] Jazebi S, de Leon F, Nelson A. Review of wildfire management techniques-Part II: Urgent call for investment in research and development of preventative solutions. IEEE Transactions on Power Delivery 2020;35(1):440–50. https://doi.org/ 10.1109/TPWRD.2019.2930095. i The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. [22] Perry M, Troccoli A. Impact of a fire burn on solar irradiance and PV power. Solar Energy 2015;114:167–73. https://doi.org/10.1016/J.SOLENER.2015.01.005. i [23] Stephen York. Smoke from California wildfires decreases solar generation in CAISO – Today in Energy – U.S. Energy Information Administration (EIA). EIA 2020. htt ps://www.eia.gov/todayinenergy/detail.php?id=45336. Acknowledgement [24] Ya’acob M. E., Hizam H., Adam M. B., Rahim A. H .M. A. , Radzi M. A. M. , Hashimoto Y., et al. Decreasing of grid-tied PV power output due to thick haze phenomena in Malaysia. In: 2014 Saudi Arabia smart grid conference, SASG 2014. https://doi.org/10.1109/SASG.2014.7274288. The author would like to acknowledge that this work is supported by the UK Centre for Greening Finance and Investment (CERAF) (NERC grant NE/V017756/1); the Newton Fund through the Met Office Weather and Climate Science for Service Partnership Brazil (WCSSP Brazil) (NF_MO_BRA_491); the Local Energy Oxfordshire project (IUK #104781), and the UK Engineering and Physical Sciences Research Council (EPSRC) funded EnergyReV project (EP/S031901/1). [25] Hoseinzadeh B., Amini M. H., Bak C. L. Centralized load shedding based on thermal limit of transmission lines against cascading events, 2016. i [26] Duginski P. Strong winds and high fire danger expected in Northern California this weekend. Los Angeles Times; 2019. i [27] National Wildfire Coordinating Group. Fire Weather Index (FWI) System. NWCG 2021. https://www.nwcg.gov/publications/pms437/cffdrs/fire-weather-ind ex-system. y [28] Copernicus. Fire danger indices historical data from the Copernicus Emergency Management Service. Copernicus 2019. https://cds.climate.copernicus.eu/cds app#!/dataset/cems-fire-historical?tab=form. Table 5 i [8] State Library Victoria. 2019-20 Bushfires in Victoria - Research Guides. State Library Victoria 2020. https://guides.slv.vic.gov.au/bushfires/2019. i [9] The 2019-20 bushfires: a CSIRO explainer. CSIRO 2020. https://www.csiro.au/e n/research/natural-disasters/bushfires/2019-20-bushfires-explainer. i ii [10] Endeavour Energy. Cost pass through application 2019–20 Bushfire disaster event; 2020. https://www.aer.gov.au/system/files/Endeavour%20Energy%20-%202 019-20%20bushfire%20disaster%20event%20cost%20pass%20through%20-% 20Application%20-%20August%202020.pdf [accessed April 22, 2021]. i l While our paper only assesses the transmission network with voltage levels greater than 66 kV, the model can be applied to a lower voltage level network to explore electrically caused wildfires and wildfire- induced electrical faults in distribution networks in the future. i [11] Penn I, Eavis P. PG&E pleads guilty to 84 counts of manslaughter in camp fire case. The New York Times; 2020. In conclusion, the high-resolution wildfire-affected power grid combining the FWI, intelligent power line disconnection strategies, and microgrid solutions was proposed in this paper. It has been demon strated that microgrids have the potential to improve grid resilience performances economically. The methods developed for weather- dependent power system modelling in this paper can be used to inves tigate how microgrids improve power grid resilience in the face of other extreme weather events, such as floods and hurricanes in the future. [12] EIA. Delivery to consumers – U.S. Energy Information Administration (EIA), 2020. https://www eia gov/energyexplained/electricity/delivery to consumers php [12] EIA. Delivery to consumers – U.S. Energy Information Administration (EIA), 2020. https://www.eia.gov/energyexplained/electricity/delivery-to-consumers.php. k d d l d ldfi h ll [13] Baker DR. Underground power lines don’t cause wildfires. But they’re really expensive. San Francisco Chronicle; 2017. https://www.sfchronicle.com/bayarea/ article/Underground-power-lines-don-t-cause-wildfires-12295031.php. k l d b f h i [14] Trakas DN, Hatziargyriou ND. Optimal distribution system operation for enhancing resilience against wildfires. IEEE Transactions on Power Systems 2018;33(2): 2260–71. [15] Bie Z, Lin Y, Li G, Li F. Battling the extreme: a study on the power system resilience. Proceedings of the IEEE 2017;105(7):1253–66. [16] Willis HH, Loa K. Measuring the resilience of energy distribution systems. RAND Corporation 2015. Appendix A. Supplementary material i [29] Copernicus. Fire index user guide. Copernicus; 2017. http://datastore.coperni cus-climate.eu/c3s/published-forms/c3sprod/cems-fire-historical/Fire_In_CDS.pdf Supplementary data to this article can be found online at https://doi. org/10.1016/j.apenergy.2022.118793. [30] Dowdy AJ, Mills GA, Finkele K, Groot W. Australian fire weather as represented by the McArthur Forest Fire Danger Index and the Canadian Forest Fire Weather Index. Bureau of Meteorology, Australian Government; 2009. References [31] Dowdy AJ, Mills GA, Finkele K, de Groot W. Index sensitivity analysis applied to the Canadian Forest Fire Weather Index and the McArthur Forest Fire Danger Index. Meteorological Applications 2010;17(3):298–312. https://doi.org/ 10.1002/met.170. 11 W. Yang et al. Applied Energy 313 (2022) 118793 [53] Sun J, Tesfatsion L, Goldfarb D, Hogan W, Kirschen D, Liu C-C, et al. DC optimal power flow formulation and solution using QuadProgJ*. Iowa State University 2010. [32] van Oldenborgh G, Krikken F, Lewis S, Leach NJ, Lehner F, Saunders KR, et al. Attribution of the Australian bushfire risk to anthropogenic climate change. Natural Hazards and Earth System Sciences 2020. https://doi.org/10.5194/nhess- 2020-69. i [54] John A. Dutton e-Education Institute. Variable cost concepts for power generation. Pennsylvania State University 2019. https://www.e-education.psu.edu/ebf483/ node/584. [33] Camia A, Amatulli G. Weather factors and fire danger in the mediterranean. Earth Observation of Wildland Fires in Mediterranean Ecosystems. In: Chuvieco E, editor. Springer Berlin Heidelberg; 2009. p. 71–82. [55] Moye R, Meyn S. The use of marginal energy costs in the design of U.S. Capacity Markets; 2018. https://core.ac.uk/download/pdf/143481163.pdf [accessed May 5, 2021]. [34] Dimitrakopoulos AP, Bemmerzouk AM, Mitsopoulos ID. Evaluation of the Canadian fire weather index system in an eastern Mediterranean environment. Meteorological Applications 2011;18(1):83–93. https://doi.org/10.1002/met.214. [56] Geoscience Australia. Geoscience Australia – Electricity Transmission Substations. Geoscience Australia; 2019. https://ecat.ga.gov.au/geonetwork/srv/eng/catalog. search#/metadata/83173. g pp p g [35] netCDF4 API documentation. NetCDF4 API; 2019. https://unidata.github.io/netcdf 4-python/ [accessed April 20, 2021]. [57] Geoscience Australia. Geoscience Australia – power stations. Geoscience Australia 2019. https://ecat.ga.gov.au/geonetwork/srv/eng/catalog.search#/metad ata/82326. [36] Luo Y, Zhou X. Temporal and spatial variations in soil respiration. Soil Respiration and the Environment 2006:107–31. https://doi.org/10.1016/B978-012088782-8/ 50006-1. i [58] AEMO. AEMO Market data. NEMWEB 2020. https://www.aemo.com. au/energy-systems/electricity/national-electricity-market-nem/d ata-nem/market-data-nemweb#mms-data-model. [37] França GB, Oliveira AN, Paiva CM, Peres L, Oliveira LMT. A fire-risk-breakdown system for electrical power lines in the North of Brazil. Journal of Applied Meteorology and Climatology 2013. https://www.jstor.org/stable/26176341?re freqid=excelsior%3A4d44ca2b89e1b745051f267c1ac26cd7. [59] AEMO. NTNDP database; 2020. https://www.aemo.com.au/energy-systems/major -publications/integrated-system-plan-isp/national-transmission-networ k-development-plan-ntndp/ntndp-database [accessed April 28, 2021]. i q [38] Google Developers. Keyhole markup language. Google developers 2021. http s://developers.google.com/kml. [60] AEMO. Guide to generator exemptions and classification of generating units; 2021. https://www.aemo.com.au/-/media/Files/Electricity/NEM/Participant_Informati on/New-Participants/Generator-Exemption-and-Classification-Guide.docx [accessed April 28, 2021]. [39] Xenophon A, Hill D. Open grid model of Australia’s national electricity market allowing backtesting against historic data. Scientific Data 2018;5:1–21. https:// doi.org/10.1038/sdata.2018.203. g [40] Australian Bureau of Statistics. Australian Statistical Geography Standard. Australian Bureau of Statistics; 2016. https://www.abs.gov.au/AUSSTATS/abs@. nsf/DetailsPage/1270.0.55.003July%202016?OpenDocument. [61] USGS. References How much distance does a degree, minute, and second cover on your maps? 2018. https://www.usgs.gov/faqs/how-much-distance-does-a-degree-minute -and-second-cover-your-maps?qt-news_science_products=0#qt-news_science _products [accessed June 10, 2021]. [41] Australian Bureau of Statistics. Regional population growth, Australia. Australian Bureau of Statistics; 2016. https://www.abs.gov.au/AUSSTATS/abs@.nsf/Deta ilsPage/3218.02014-15?OpenDocument. [62] Energy Exemplar. Understanding microgrids; their impact and integration in a complex energy environment; 2020. https://energyexemplar.com/wp-conten t/uploads/EE_eBook_Microgrids.pdf [accessed May 5, 2021]. [42] Ju L, Ringer T, Gunzburger M. Voronoi tessellations and their application to climate and global modelling. U.S. Department of Energy Office of Scientific and Technical Information; 2012. https://www.osti.gov/servlets/purl/1090872. t/uploads/EE_eBook_Microgrids.pdf [accessed May 5, 2021]. [63] OFGEM DECC. The value of lost load (VoLL) for electricity in Great Britain. OFGEM 2013. https://www.ofgem.gov.uk/ofgem-publications/82293/london-economic s-value-lost-load-electricity-gbpdf. i [43] AEMO. The national electricity market FACT SHEET 2020. https://aemo.com.au/- /media/files/electricity/nem/national-electricity-market-fact-sheet.pdf. y g p [64] Schr¨oder T, Kuckshinrichs W. Value of lost load: An efficient economic indicator for power supply security? A literature review. Frontiers in Energy Research 2015; 3. https://doi.org/10.3389/fenrg.2015.00055. i [44] Geoscience Australia. Geoscience Australia – electricity transmission lines. Geoscience Australia 2019. https://ecat.ga.gov.au/geonetwork/srv/eng/catalog. search#/metadata/83105. [45] Arderne C, Zorn C, Nicolas C, Koks EE. Predictive mapping of the global power system using open data. Scientific Data 2020;7:1–12. https://doi.org/10.1038/ s41597-019-0347-4. p g g [65] Victoria State Government. Renewable energy action plan 2017 Statement of Progress. Victoria State Government 2017. https://www.energy.vic.gov.au /__data/assets/pdf_file/0014/113432/DELWP006-REAP-SOP_F.pdf. [46] Liu Y, Singh C. A methodology for evaluation of hurricane impact on composite power system reliability. IEEE Transactions on Power Systems 2011;26(1):145–52. https://doi.org/10.1109/TPWRS.2010.2050219. l i [66] Geoscience Australia. Wind Energy \| Geoscience Australia, 2010. https://www.ga. gov.au/scientific-topics/energy/resources/other-renewable-energy-resources/win d-energy. gy [67] Muhs JW, Parvania M, Nguyen HT, Palmer JA. Characterizing probability of wildfire ignition caused by power distribution lines. IEEE Transactions on Power Delivery 2021;36(6):3681–8. https://doi.org/10.1109/TPWRD.2020.3047101. g [47] Panteli M, Mancarella P. Influence of extreme weather and climate change on the resilience of power systems: Impacts and possible mitigation strategies. Electric Power Systems Research 2015;127:259–70. https://doi.org/10.1016/j. epsr.2015.06.012. y p g [68] Global Wind Atlas. Energy data info, 2021 https://globalwindatlas.info/. [48] Government of South Australia. Identifying powerlines. Government of South Australia 2017. https://www.sa.gov.au/topics/energy-and-environment/using-ele ctricity-and-gas-safely/powerline-safety/identifying-powerlines. [69] Global Solar Atlas. Energy data info, 2021. https://globalsolaratlas.info/map? c=-35.201256,144.148881,6&r=AUS:AUS.10_1. [70] Energy Information Administration U. Levelized Costs of New Generation Resources in the Annual Energy Outlook 2021; 2021. https://www.eia.gov/outl ooks/aeo/pdf/electricity_generation.pdf [accessed May 5, 2021]. [49] Eaton. Recloser technical data; 2017. www.eaton.com/cooperpowerseries. [49] Eaton. Recloser technical data; 2017. www.ea [50] Thurner L, Scheidler A, Schafer F, Menke J-H, Dollichon J, Meier F, et al. References Pandapower – an open-source python tool for convenient modeling, analysis, and optimization of electric power systems. IEEE Transactions on Power Systems 2018; 33(6):6510–21. https://doi.org/10.1109/TPWRS.2018.2829021. i [71] Wholesale statistics. Australian energy regulator 2021. https://www.aer.gov.au /wholesale-markets/wholesale-statistics. [72] Simshauser P. Rooftop solar PV and the peak load problem in the NEM’s Queensland region; 2021. [51] National Grid. Guidance for UK fire and rescue services. National Gird 2018. https ://www.nationalgrid.com/sites/default/files/documents/FINAL_Fire%20%26% 20Rescue.pdf. [73] Renewable power generation costs in 2018. IRENA 2018. https://www.irena. org/-/media/Files/IRENA/Agency/Publication/2019/May/IRENA_Renewable- Power-Generations-Costs-in-2018.pdf. [52] Sheble GB, Mccalley JD. Module E3 economic dispatch calculation. Iowa State University; 2019. [74] Blakers A, Lu B, Stocks M. 100% renewable electricity in Australia. Energy 2017; 133:471–82. https://doi.org/10.1016/j.energy.2017.05.168. 12
https://openalex.org/W2917473276	https://www.mdpi.com/2071-1050/11/4/1211/pdf?version=1551088153	English	null	Inclusions in and Exclusions from the S&P 500 Environmental and Socially Responsible Index: A Fuzzy-Set Qualitative Comparative Analysis	Sustainability	2,019	cc-by	21,690	Received: 28 December 2018; Accepted: 19 February 2019; Published: 25 February 2019 Abstract: Socially responsible investment (SRI) indices provide an interesting opportunity to analyse the links between corporate ﬁnancial performance (CFP) and corporate sustainability performance (CSP). However, few studies focus on the antecedents of inclusions in and exclusions from SRI indices. Speciﬁcally, the implications of corporate sustainability disclosure (CSD) have been largely ignored in this ﬁeld. Furthermore, previous literature on the CSP-CSD-CFP links shows inconclusive results that have been attributed to both methodological and measurement problems, which suggest the existence of asymmetry, equiﬁnality and complexity amongst these links. This study targets two under-researched areas regarding the determinants of changes in the composition of SRI indices, and the effects of CSD on CSP. This study also attempts to overcome the methodological and measurement limitations of previous studies on the CFP-CSD-CSP links. The study presents a fuzzy-set qualitative comparative analysis (fsQCA) to explore how different combinations of CFP and CSD indicators are related to inclusions in an SRI index (assumed as expressions of a good CSP), and exclusions from an SRI index (equivalent to a poor CSP). The empirical results reveal that a combination of different CSD indicators is necessary, but not sufﬁcient, to lead to the inclusion in or exclusion from an SRI index, and that CFP measures have asymmetrical effects on CSP. CSD is a relevant antecedent or precondition of CSP that can motivate changes in corporate behaviours towards an improved CSP. Poor CSP, leading to an exclusion from the index, is associated with poor CSD and a deterioration of CFP. The implications for researchers, business managers, SRI rating agencies and policymakers are derived. Keywords: SRI indices; corporate sustainability performance; corporate sustainability disclosure; corporate ﬁnancial performance; equiﬁnality; asymmetry; complexity; fsQCA Sustainability 2019, 11, 1211; doi:10.3390/su11041211 sustainability sustainability sustainability sustainability Inclusions in and Exclusions from the S&P 500 Environmental and Socially Responsible Index: A Fuzzy-Set Qualitative Comparative Analysis Juan Pineiro-Chousa 1,, Noelia Romero-Castro 1 and Marcos Vizcaíno-González 2 1 Department of Financial Economics and Accounting, Universidade de Santiago de Compostela, Santiago de Compostela 15782, Spain; noe.romero@usc.es 2 Department of Business, Universidade da Coruña, A Coruña 15071, Spain; marcos.vizcaino@udc.es Correspondence: j.pineiro@usc.es; Tel.: +34-982824466 Juan Pineiro-Chousa 1,, Noelia Romero-Castro 1 and Marcos Vizcaíno-González 2 1 Department of Financial Economics and Accounting, Universidade de Santiago de Compostela, Santiago de Compostela 15782, Spain; noe.romero@usc.es 2 Department of Business, Universidade da Coruña, A Coruña 15071, Spain; marcos.vizcaino@udc.es Correspondence: j.pineiro@usc.es; Tel.: +34-982824466 Received: 28 December 2018; Accepted: 19 February 2019; Published: 25 February 2019 www.mdpi.com/journal/sustainability 1. Introduction As most sustainability indices are designed and managed to comprise companies with a superior or outstanding CSP, it can be assumed that entering a sustainability index is equivalent to displaying a good CSP and that the exclusion from a sustainability index means a worsening of CSP. This assertion is far from universally accepted and the role of rating agencies behind SRI indices has been criticised. The processes and methods for the assessment of the sustainability commitment and performance of companies, do not clearly specify the reasons behind the high or low sustainability scores of a company, and the diversity of ratings and methodologies only adds to the confusion and distrust [8,9]. Furthermore, inclusions in SRI indices have been associated with factors other than CSP [10]. Nevertheless, given that the meaning of SRI indices is to help investors assessing CSP, it seems logical to assume that they represent CSP, even though they do not measure it directly [11]. The disclosure of sustainability information and the quality of that information are essential factors in the construction of sustainability stock indices, because third-party rating agencies gather this information both through their own questionnaires and from publicly accessible sources [9] such as sustainability reports. Nevertheless, the inﬂuence of CSD on CSP has been under-explored by previous literature. This study aims at addressing that research gap. Companies with a high commitment to CSD can be assumed to have also a good CSP [12,13] and, consequently, a high probability of being included in a sustainability index. Therefore, we argue that a good CSD is a necessary condition to show a good CSP, which leads to the company’s inclusion in a SRI index, and that a poor CSD can explain a poor CSP, which leads to an exclusion from a SRI index. p p CFP is also an essential condition for the inclusion in or exclusion from any stock index, and hence for sustainability-focused ones. Although previous research has mainly focused on the impacts of CSP on CFP [14], the reverse causality has also been explicitly analysed [15]. We also target this less developed strand of research by focusing on the inﬂuence of CFP on CSP and arguing that ﬁrms with superior CFP can be included in a sustainability index because a superior CFP enables a good CSP, which results in being listed or maintained in the index. 1. Introduction The relationship between corporate ﬁnancial performance (CFP) and corporate sustainability performance (CSP) is probably one of the most researched subjects in the ﬁeld of corporate sustainability [1]. The terms corporate sustainability, corporate social responsibility (CSR) or environmental, social and governance (ESG) are used interchangeably both among practitioners and researchers [1]. Two prominent strands within this vast academic literature are those focused on (1) the analysis of socially responsible investment (SRI) strategies in general, and sustainability stock indices in particular, and (2) on the analysis of the relevance and impact of the disclosure of sustainability information in the form of sustainability reports [2]. These two topics are greatly interrelated because corporate sustainability disclosure (CSD) is the basis for the development of SRI in general and for the construction of sustainability stock indices in particular. www.mdpi.com/journal/sustainability 2 of 31 Sustainability 2019, 11, 1211 Sustainability, or SRI indices have been mainly used under this CSP-CFP ﬁeld of research in three ways: (1) As a proxy for CSP, taking into consideration that being listed into a sustainability index is synonymous of having a good CSP [3]; (2) to conduct event studies analysing whether the announcement of the inclusion and exclusion events has any signiﬁcant impact on stock returns [4–6]; and (3) to analyse the ﬁnancial performance of sustainability indices, funds or portfolios against the performance of conventional or non-sustainability-screened ones (see [7] for a meta-analysis of previous research on this topic). Although their contributions are highly relevant, the implications of entering or dropping out of a sustainability index have been object of analysis in a small number of papers compared to other approaches to exploring the relationship between CFP and CSP [6]. g pp g y j y papers compared to other approaches to exploring the relationship between CFP and CSP [6]. This study approaches the analysis of the CSP-CFP link in a rather particular fashion, by combining the ﬁrst two strands of this literature based on SRI indices: On one hand, being listed or delisted from an SRI index are considered proxies for a good and poor CSP, respectively; and on the other hand, we focus on inclusion and exclusion events, but instead of trying to explain their impact on CFP, we explore whether inclusions and exclusions, as expressions or result of a good or poor CSP, are related to, or explained by, CSD and CFP. 1. Introduction Furthermore, a superior CFP, regardless of being linked to a superior CSP, can be valued by the stock market. A situation like that would result in a higher stock price and an increased market capitalization, which facilitates the company’s inclusion in SRI indices because market capitalization is usually considered an important factor. Therefore, considering both accounting and market-based measures of CFP seems obliged in this context, as accounting measures can indicate the availability of ﬁnancial resources for an improved CSP, while market measures can capture long-term expectations regarding CSP and thus directly inﬂuence inclusions in or exclusions from a SRI index. Previous research has dealt with all possible hypotheses regarding the relationship among CFP, CSD and CSP. The three variables have been used as independent and dependent variables in different studies, but the results are inconclusive [14]. This situation has raised demands for more research 3 of 31 Sustainability 2019, 11, 1211 in order to reduce variability in theory and results and facilitate convergence. This study adopts an exploratory approach and assumes (1) that complex interactions exist among the variables, (2) that there are several alternatives to achieve the same outcome and (3) that the same cause can produce different outcomes. These premises imply the application of a conﬁgurational approach to analyse the sufﬁcient and necessary causes of an outcome [16]. Qualitative Comparative Analysis (QCA), which, to our knowledge, has only been incipiently applied in the analysis of the CFP and CSP relationship [17], provides a suitable methodological framework in this context. Sustainability indices are actively managed by their corresponding providers; therefore, at each index revision, companies listed in the index can be removed, and new companies can be added. Interestingly, some companies experience successive inclusions and exclusions from the same index throughout their life span. These cases seem particularly attractive to explore due to the reasons behind this variability. While previous research related to sustainability indices has mainly focused on analysing the implications of just being listed in a sustainability index, or of being included in or excluded from an index as separate events, this study analyses the interrelations between the events of being included in and excluded from an index. In particular, we are interested in the analysis of the exclusion from a sustainability index following an inclusion taking place after the initial index conﬁguration, and the inclusion following a previous exclusion. 1. Introduction Thus, our research objectives are two-fold and aim at ﬁlling current research gaps. First, we explore two issues that seem to have been largely neglected in previous literature, identifying the factors that explain inclusions in or exclusions from SRI indices. Speciﬁcally, the analysis addresses the inﬂuence of CSD on CSP, with a singular focus on the companies that appear to have a rather volatile CSP, leading to subsequent inclusions in or exclusions from an SRI index. Second, the study contributes to the incipient application of the QCA framework to the analysis of the complex and asymmetrical CSP-CSD-CFP links in order to shed a new light on the conﬂicting perspectives and inconclusive ﬁndings of prior research. This study moves away from traditional statistical techniques, and relies on the fuzzy-set Qualitative Comparative Analysis (fsQCA) technique to assess the causal complexity among the concepts of CSD, CFP and CSP. More concretely, this study analyses the necessity and sufﬁciency of different CSD and CFP causal conditions regarding companies’ CSP measured through the inclusion in or exclusion from an SRI index. The sample comprises 126 exclusions from and inclusions in the S&P 500 Environmental and Socially Responsible Index, experienced by 99 S&P 500 companies over the period 2010–2016. The results obtained show complex interrelations among causal conditions, some of which can contribute to both good and poor CSP (inclusion in or exclusion from an SRI index). Furthermore, the analysis identiﬁes different paths or combinations of causal conditions that are sufﬁcient but not necessary for a good or poor CSP. The results also provide evidence that there is no single necessary causal condition for a speciﬁc outcome, and that combinations of causal conditions associated with a good CSP (index inclusion) are not the mirror opposite of those associated with a poor CSP (index exclusion). In particular, we conclude that companies included in the index after a previous exclusion show a comparatively higher commitment to CSD, whereas companies excluded after a previous inclusion show a lower or no commitment to CSD. Regarding the CFP-CSP link, results reﬂect an ambiguous role of alternative CFP measures in relation to inclusions in or exclusions from the index. The remaining of the study has the following structure. 2.1. The CSP-CFP Link There is extensive research on the integration of sustainability or CSR into corporate strategic management [18], and particularly on the analysis of the links with economic value creation (including comprehensive reviews and meta-analysis [2,14,19–22]); however, the results regarding the relationship between CSR and ﬁnancial performance remain inconclusive [17,23–25]. In particular, research ﬁndings support all the possible hypotheses (non-signiﬁcant or neutral effects, positive effects and negative effects of CSP on CFP). Friede et al. [26] and Busch and Friede [19] develop second-level, or ‘review of reviews’ studies on CSP and CFP respectively, accounting for 2200 and 1214 unique or individual studies on the topic, ﬁnding an undeniable business case for CSR. Brammer and Millington [27] conclude that a curvilinear relationship exists between CSP (charitable giving) and CFP, as both low and high CSP are related to a high CFP. Academic research has also focused on the analysis of the CFP consequences of corporate social irresponsibility (CSI), arguing that the CSI-CFP links can be stronger than the CSR-CFP links and that not distinguishing between CSR and CSI can be one of the reasons behind the contradictory results obtained in the numerous studies that explore this relationship [28,29]. A good number of moderating or mediating factors have also been explored [20] based on the assumption that companies’ ability to gain competitive advantages and create economic value through sustainability is largely based on the possession of certain resources and capabilities, usually of an intangible nature [22,25,30], which are inﬂuenced by contextual factors [21]. Besides the inconclusive results of the vast literature devoted to the analysis of the CSP-CFP link, there is also a lack of consensus regarding the direction of the causality between CSP and CFP. Although previous literature has predominantly focused on the impact of CSP on CFP [14], some studies acknowledge that companies with a good CFP could have extra resources for the improvement of their CSP [15,31–35] and support the existence of a bidirectional relation or virtuous circle between CSP and CFP [14,30,36–38]. However, studies on this issue present inconclusive results. In their meta-analysis of the CSP-CFP literature, Wang et al. [21], conclude that prior CSP is related to CFP but not contrariwise. 1. Introduction Section 2 provides a review of previous research on the CSP-CSD-CFP links, justiﬁes the application of a conﬁgurational approach to try to disentangle past inconclusive and even contradictory results and presents ten propositions to be tested using fsQCA. Section 3 explains the choice of an SRI index and of concrete measures of CSD and CFP, describes the sample, introduces the fsQCA technique and presents the speciﬁcation of causal conditions and outcomes as well as four alternative models to test the propositions. Section 4 shows and discusses the results. Finally, Section 5 identiﬁes the main implications and the limitations of the study. Sustainability 2019, 11, 1211 4 of 31 2. Theoretical Background In this section we review previous literature and formulate the study’s propositions over four concrete areas: The CSP-CFP relationship, the role of CSD in relation to CFP and CSP, the relevance of SRI indices as a research object to explore the CFP-CSD-CFP nexus and the convenience of adopting a conﬁgurational approach to study the CFP-CSD-CSP interrelations. 2.1. The CSP-CFP Link We highlight the contribution by Ruggiero and Cuppertino [15], as they adopt a similar approach to ours for the measurement of CSP in terms of inclusion in or exclusion from the Global 100 Index (not an SRI index, but an annual ranking of the top 100 worldwide corporate leaders in CSR). Through a logistic regression analysis, they ﬁnd a signiﬁcant positive correlation between CFP and CSP, considering the mediating (more than moderating) effect of innovation. The inconclusive results of previous literature on the CSP-CFP link have been attributed to conceptual or practical differences in the deﬁnition and measurement of both CSP and CFP, to the underlying theories to approach the CSP-CFP relationship and to the different methodologies applied [20,24,30,39]. Schaltegger and Synnestvedt [40] and Aguinis and Glavas [18] suggest that empirical studies are not conclusive due to the absence of clear theoretical frameworks. The authors highlight the importance of considering the speciﬁc CSR management deployed at each organisation because different CSR strategies or practices will result in different ﬁnancial results. Differences in CSR management and CSP will also depend on business managers’ perceptions and priorities towards CSR [41]. Regarding CFP measurement, both accounting and market-based measures have been used [36,42], sometimes in combination, but it is generally assumed that market measures are more able to capture the long-term value of CSR [14,30] than ﬁnancial ones. Regarding CSP measurement, the multidimensional and complex nature of CSR [41] has forced the approach to CSP through many alternative constructs, including one-dimensional measures related to different ESG dimensions, Sustainability 2019, 11, 1211 5 of 31 aggregated indicators, adhesion to global CSR codes or initiatives, reputation and SRI indices or rating methodologies [8,14,42,43]; however, the credibility and validity of many of these approaches have been questioned [8]. q By associating CSP with the inclusion in or exclusion from a SRI index, this study avoids the controversial choice of speciﬁc measures of CSP (Kappou and Oikonomou [6] also acknowledge this advantage). This choice could seem partial, incomplete or biased in many ways, either in the form of a composite index combining different ESG factors or of different indicators of the CSP components [14,42]. 2.1. The CSP-CFP Link Furthermore, by focusing on the effects of CFP on CSP and not the opposite, we also avoid problems related to the possible counteracting effects of different ESG issues on CFP [23,44–46] that could also be behind the inconclusive results of previous research on the CSP-CFP link. Complexity both in approaching CSP measurement and in conceptualising its impact on CFP has been largely identiﬁed as inherent to the relationship between CSP and CFP, and as a possible cause of the inconsistent results of previous literature [17,22,39]. Aguinis and Glavas [18] and Grewatsch and Kleindienst [20] claim for new theoretical frameworks and new methodological orientations in order to make meaningful advances in the CSR domain, accounting for the mediating and moderating role of diverse internal and external variables in the relationship between CSR and different types of outcomes (as CSP or CFP) and acknowledging that antecedents of CSP may operate in an interdependent manner rather than in isolation. Past literature on the CSP-CFP link has also signalled towards non-linearity and asymmetrical causation [3] as well as equiﬁnality [27] between CSP and CFP. Thus, all these speciﬁc features have motivated the incipient application of QCA to the CSP-CFP ﬁeld of research [17]. Drawing from previous literature suggesting an inﬂuence of CFP on CSP, considering our research objectives and adopting QCA as the appropriate methodological framework, we deﬁne the ﬁrst proposition to be tested through a fsQCA as the two sides of the same coin: Proposition 1a. Companies with a high CFP will also show a good CSP, which will facilitate their inclusion in an SRI index. Proposition 1b. Companies with a low CFP will also show a poor CSP, which will favour their exclusion from an SRI index. 2.2. The Relevance of CSD The analysis of the CSD-CSP-CFP links has also largely attracted the interest of academic research, which is characterized by inconclusive results associated with different theoretical, methodological and measurement approaches [2]. CSD has in fact been brought up to explain the inconclusive results of studies on the CSP-CFP link, signalling the lack of homogeneity in sustainability reports [47] or their exclusive use by a small fraction of larger companies [48] and emphasizing the confusion faced by reports users despite standardisation efforts such as the GRI [49]. The importance of CSD as a mediator or moderator of the CSP-CFP relationship has also been highlighted [3,50,51]. Previous literature analysing CSD is mainly focused on the adoption of the GRI standard [52], which has evolved through different versions of guidelines for the reporting of sustainability information (the GRI G3.1 and G4 being the latest ones). Until the release of the G4, this standard was accompanied by the possibility of distinguishing between different self-referenced application levels, substituted in the G4 by a materiality assessment provided by the GRI itself upon payment. There is also incipient research related to the Integrated Reporting (IR) framework, although its relative novelty and the slow adoption rates limits the availability of empirical research [53]. CSD has been measured through content analysis of sustainability reports [48,54,55], according to the level of compliance with GRI guidelines [56] or through ranking instruments such as Bloomberg’s ESG disclosure scores [51,57]. Crifo et al. [58] used a different approach that included ﬁctitious ESG disclosures to potential private equity investors, ﬁnding an asymmetric effect: Bad ESG information has a greater impact on ﬁrm value and investment decisions than good ESG information. 6 of 31 Sustainability 2019, 11, 1211 Previous research has explored all the possible causal interactions between CFP and CSD and between CSP and CSD [2,59]. Regarding the CFP-CSD relationship, Fifka [59] ﬁnds that US studies have been more focused on the impact of CFP on CSD, with inconclusive results. Despite the inconclusive results, the positive inﬂuence of CFP on CSD appears to have received greater support [54,57,60,61]. Moving towards the effects of CSD on CFP, literature on the impact of global CSD (not only environmental reporting) on CFP is scarce [2] because most of the literature on the impact of CSP on CFP focuses on global CSP rather than on CSD in particular [62]. 2.2. The Relevance of CSD Studies on the impact of CSD on CFP show again inconclusive and even contradictory results, pointing towards equiﬁnality, causal asymmetry and complexity between CFP and CSD. Several studies ﬁnd some evidence of a positive relationship between CSD and CFP [55,62]. Other studies addressing the CSP-CSD-CFP interrelations ﬁnd a negative effect of CSD on ﬁrm value [51] or no effect [3]. There is also a strand of research dealing with the bidirectional links between the quality of sustainability reporting and ﬁnancial reporting [56,63]. Studies exploring the effects of CFP on CSD are usually based on accounting measures of CFP under the slack resources theory [38,64], assuming that companies with a good CFP will devote more resources towards and improved CSD and CSP. By contrast, studies focused on the impact of CSD on CFP are mainly based on market-based measures of CFP to explore the value relevance of CSD for stock markets and its potential impact on the reduction of information asymmetry and the cost of capital [50]. Regarding the CSP-CSD relationship, the effects of CSP on CSD have received a great part of the attention, but they are not clear from a theoretical standpoint, because good CSR performers have natural incentives to improve their disclosure and reduce information asymmetry [50] and to differentiate from bad CSR performers [65]. Nevertheless, bad CSR performers can also look for legitimacy through CSD [66]. Fifka [59] also reports inconclusive results of US studies focused on the impact of CSP on CSD. Schreck and Raithel [65] report a strong non-linear relationship between CSP and CSD and Hummel and Schlick [67] ﬁnd a signiﬁcant positive (negative) relationship between CSP and high-quality (low-quality) CSD. The impact of CSD on CSP has rarely been explored in previous literature [12,13]. There are different arguments that support the positive effect of CSD on CSP. Topping [13] analyses how engagement with the Carbon Disclosure Project is expected to change corporate behaviour towards an improved CSP due to different motivators or drivers: (1) What gets measured gets managed, (2) changes in strategic thinking brought about by disclosure, (3) external demands to compare CSP across companies or industries, (4) competitor benchmarking and (5) SRI. Christensen [12] ﬁnds that CSD reduces the probability of being involved in CSI and that reporting companies involved in CSI experience lower negative stock price impacts compared to uninvolved companies. 2.2. The Relevance of CSD We position our study in this under-researched area of the impacts of CSD on CSP assuming that CSD and the pressure of an increased scrutiny and demand of non-ﬁnancial information are drivers of CSP. In our study, CSP is derived from inclusions in or exclusions from an SRI index; in addition, we assume that a good CSD facilitates inclusion in an SRI index and that a poor CSD leads to exclusion. Again, the review of previous research on the CSP-CSD-CFP links has evidenced signs of complexity, causal asymmetry and equiﬁnality that justify a conﬁgurational methodological approach. We deﬁne the second proposition to be tested through a fsQCA as: Proposition 2a. Companies highly committed to CSD will also show a good CSP, which facilitates their inclusion in an SRI index. Proposition 2b. Companies with a low or no commitment to CSD will also show a poor CSP, which contributes to their exclusion from an SRI index. 2.3. SRI Indices In any case, these studies represent only a small fraction of a greater set of studies focusing on the performance of SRI strategies against traditional investment strategies, showing heterogeneous results that have been also converging throughout time towards the no cost or beneﬁt conclusion [7]. Of greatest interest in the context of this study is the academic literature based on the analysis of the impact of inclusion/exclusion on CFP, usually instrumented through event studies looking for any signiﬁcant stock price reaction to the announcement of the inclusion in or exclusion from an SRI index. A general interesting conclusion of these studies is that exclusions have a greater impact on CFP compared to inclusions [4,6,71,72]. In fact, Doh et al. [4], Kappou and Oikonomou [6] and Oberndorfer et al. [72] do not ﬁnd any signiﬁcant market reaction to inclusions, but conﬁrm negative reactions to exclusions, whereas Consolandi et al. [71] also ﬁnd a positive reaction to inclusions, but softer than for exclusions. Russo and Mariani [73] only focus on exclusions also conﬁrming negative impacts on CFP (Tobins’ Q and stock return). Other studies ﬁnd limited evidence of signiﬁcant impacts of SRI indices’ inclusions and exclusions on stock returns [5,74,75]. In their event study, Doh et al. [4] also analyse the moderating role of previous operating performance (operating income to total assets ratio) and previous CSR reputation over the impact of inclusions and exclusions on CFP. They ﬁnd that companies included (excluded) in (from) an SRI index had higher (lower) previous operating performance than excluded (included) ones. Regarding the inﬂuence of previous CSR reputation, a stronger reputation mitigates the negative impact of exclusions on market value but also the positive effect of inclusions. Doh et al. [4] explicitly advocate the identiﬁcation of inclusions in and exclusions from an SRI index with good or poor CSP. This review of previous research reveals that inclusions in and exclusions from SRI indices have been mainly used in the context of event studies to verify some market reaction to those announcements. Little research has explored the underlying reasons of inclusions or exclusions. Russo and Mariani [73], in their event study on exclusions from the FTSE4Good Index, analyse the effect of exclusions motivated by failure to meet any of the ﬁve different sustainability criteria reported by FTSE4Good itself, concluding that only countering bribery has a signiﬁcant impact on CFP. 2.3. SRI Indices The growing interest on CSR, ESG or corporate sustainability issues and the development of SRI has triggered the appearance of many sustainability or SRI indices that have been also studied under the CSP-CFP research ﬁeld. As Searcy and Elkhawas [11] acknowledge, the best known and most 7 of 31 Sustainability 2019, 11, 1211 mentioned in academic literature are the Dow Jones Sustainability Index series (DJSI), the FTSE4Good Index series and the MSCI ESG Index (formerly KLD and Domini 400 Social Index). y SRI indices have been widely used in the CSP-CFP literature as proxies for CSP [3], assuming that companies listed in an SRI index have a good CSP and that companies not included have a poor CSP; however, it has been criticised for the rather obscure methodologies behind the indices’ conﬁguration and management, and the potential existence of other factors apart from CSP inﬂuencing inclusions in and exclusions from the indices [8–10]. Once again, studies offer inconclusive results, from no effect on CFP to both positive and negative relations. Eom and Nam [68] analyse the impacts of being included in an SRI index on cost of equity and ﬁrm value (Tobins’ Q) and ﬁnd mixed results: No relation for the entire sample, and differences depending on the time of the inclusion, so that inclusions when the index was created showed a negative relation to cost of equity, whereas those related to revisions of the index showed a positive relation. Finally, we highlight the contribution by Ziegler and Schröder [10], as it is quite close to this study’s aim. They examine the antecedents of inclusions in two indices of the DJSI family and ﬁnd an ambiguous inﬂuence of CFP measures (with some support to the slack resources theory) but a signiﬁcant inﬂuence of other factors different from CSP, related to the sustainability assessment and selection processes for the composition of the indices, and questioning the reliability of inclusions in SRI indices as proxies for CSP. A different approach compares the performance of SRI indices with conventional ones; these studies mostly ﬁnd no signiﬁcant statistical differences in risk-adjusted returns [69,70]. Nevertheless, Consolandi et al. [71] show higher returns for the DJSI compared with a surrogate complementary index. 2.3. SRI Indices Ziegler and Schröder [10] focus only on inclusions to analyse the role of CFP (and other geographical and time factors related to the internal assessment process of the rating agency behind the SRI index) as determinants of inclusions. They also question whether inclusions are really equivalent to a good CSP. Sustainability 2019, 11, 1211 8 of 31 The study of Doh et al. [4] is also relevant, although they do not analyse antecedents of inclusions or exclusions, but moderating factors of their impact on CFP. To our knowledge, no previous research has analysed CSD as an antecedent of inclusions to or exclusions from SRI indices. As we have previously argued, CSD is essential for SRI strategies in general and for the conﬁguration of SRI indices in particular. We also provide a more ample approach than previous research by jointly studying both inclusions and exclusions because Russo and Mariani [73] only analyse exclusions and Ziegler and Schröder [10] only focus on inclusions. Therefore, this study focuses both on inclusion and exclusion events to examine the role of CFP and CSD as antecedents of CSP, trusting in the validity of those inclusions and exclusions as expressions of a good or poor CSP. In comparison to studies using being listed in an SRI index as proxy for a good CSP and not being listed as proxy for poor CSP, our approach (shared by the majority of event studies) can be considered more robust because poor CSP is identiﬁed with being delisted. We consider that exclusion from the index is more indicative of a poor CSP than just not being listed. Another important contribution of our study is the selection of the inclusion and exclusion events of the sample. As in Eom and Nam [68], we distinguish between the time of creation of the SRI index and the subsequent revisions. Furthermore, we analyse only inclusions or exclusions that have been preceded by a former exclusion or inclusion. This implies that inclusions at the time of creation of the SRI index are not considered in our analysis. 2.3. SRI Indices This gives us the opportunity to analyse how changes in CFP can be related to changes in CSP, assuming that an improvement in CSP provoking an inclusion in a SRI index after a previous exclusion can be accompanied by an improvement of CFP, and that a deterioration of CSP causing an exclusion is likely in combination with a worsening of CFP. Thus, in Proposition 1a. and Proposition 1b., we analyse CFP from a static perspective (considering speciﬁc CFP measures at a point of time), whereas the next alternative, although complementary, propositions follow a dynamic perspective (based on changes in CFP indicators). Both options seem justiﬁed: Better ﬁnancial conditions at a particular moment can facilitate both better CSD and CSP and the inclusion in an SRI index, while the inclusion/exclusion event could be also explained by an increase/decrease in CFP indicators between the event year and that of the previous exclusion or inclusion. Drawing from previous literature on the antecedents and effects of SRI indices’ inclusions and exclusions, we conclude one more time that asymmetry, equiﬁnality and complexity are likely to be present in the analysis of the relationships between CFP and CSP (inclusions in or exclusions from an SRI index). Therefore, the study is based on a conﬁgurational framework. The third proposition to be tested through a fsQCA is as follows: Proposition 3a: Companies that experience an improvement in their CFP (increase in CFP cators) will also show a better CSP, which will facilitate their inclusion in an SRI index. Proposition 3b: Companies that experience a worsening of their CFP (decrease in CFP indicators) will also show a poorer CSP, which will contribute to their exclusion from an SRI index. Proposition 3b: Companies that experience a worsening of their CFP (decrease in CFP indicators also show a poorer CSP, which will contribute to their exclusion from an SRI index. 2.4. A Conﬁgurational and Complexity Theory Approach to Analyse the Relationship between CSD, CSP and CFP 2.4. A Conﬁgurational and Complexity Theory Approach to Analyse the Relationship between CSD, CSP and CFP The inconclusive results or previous academic literature exploring the links between CSP, CSD and CFP are probably explained by the existence of non-linear, asymmetrical and complex relationships among these concepts. If we want to investigate the inﬂuence of CSD and CFP on CSP (Proposition 1a., 1b., 2a., 2b., 3a. 2.3. SRI Indices Complexity theory assumes some fundamental principles, mainly the following ([79–82,85–88]): (1) Equiﬁnality implies that different combinations of causal conditions can lead to the same outcome; (2) asymmetry assumes that one causal condition can simultaneously explain high and low levels of the outcome; (3) complexity or conjunctural causation imply that the inﬂuence of causal conditions on outcomes depends on the presence or absence of other causal conditions; and (4) causal asymmetry means that causal conditions that lead to the outcome might be different from those leading to the absence of the outcome. QCA allows dealing with this equiﬁnality, asymmetry, complexity and causal asymmetry principles studying multiple conjunctural causation patterns among causal conditions and outcomes [77,82]. There are three QCA variants [16]: Crisp-set QCA (csQCA), which treats each case or observation from a sample as being in or out of the set of cases for each causal condition or outcome, and allocates membership scores of 0 or 1 (0 indicates non-membership, and 1 represents membership); multi-value QCA (mvQCA), which allows multi-value conditions (each category is represented by a natural number in an ordinal scale); and fuzzy-set QCA (fsQCA), which uses membership scores between 0 and 1 through a calibration process. FsQCA is the most applied technique [16,89]. Furthermore, crisp and fuzzy sets may be combined [84]. Although QCA was originally designed for small-N studies, the development of QCA from crisp sets to fuzzy sets and improvements in the underlying software algorithms have opened the door to its application to large-N studies [81,84], which has contributed to its application to management studies [16,89]. FsQCA can provide important insights into areas of research where causal relationships are expected to be heterogeneous across cases [83]. Kan et al. [89] highlight that an important part of the management literature applying QCA, tries to legitimize its use by comparing its application with traditional empirical methods, or to review commonly accepted links that existing theories do not sufﬁciently explain. Some pioneering empirical research has started to apply fsQCA in the CSR, ESG or CSP domains [83,85,90–93]. Because of their focus on the inﬂuence of ESG-related factors on ﬁnancial performance, or their analysis of SRI indices, ﬁve contributions applying fsQCA are closest to this study’s aim. Wang et al. [94] explore the sufﬁciency conditions of reputation in explaining ﬁrm’s market value (common stock price), using three reputation-measurement instruments together with book value and EPS. Paniagua et al. 2.3. SRI Indices and 3b.), it seems advisable to study how CSD and CFP combine rather than compete to explain CSP (inclusion in or exclusion from an SRI index). While previous literature mainly relies on the application of multivariate regression analysis or other similar techniques based on the assumption of a symmetrical relationship between variables, this study employs a different methodology to analyse how CSP (inclusion in or exclusion from an SRI index) can be explained by different combinations of CSD and CFP factors. Qualitative comparative analysis (QCA), proposed by Charles Ragin [76–79], is a half qualitative - half quantitative method based on the use of Boolean algebra to deal with causal complexity and identify conﬁgurations, 9 of 31 Sustainability 2019, 11, 1211 paths or combinations of causal conditions that are necessary or sufﬁcient to explain a speciﬁc level of an outcome [80,81]. In the context of QCA, the concepts of outcome and causal conditions are used to refer to the dependent and independent variables of traditional techniques, but the idea of causality is rather different. In contrast to regression and correlation methods, where each causal condition has an independent impact on the outcome, QCA searches for a combined and not uniform causation where several conditions combined may be present in the outcome, and the same causal conditions may act in favour for or against the outcome [82]. g QCA is built upon the foundations of conﬁgurational or set analysis and complexity theory [17]. The conﬁgurational or set approach explores the different combinations or sets of causal conditions leading to an outcome in terms of sufﬁciency and necessity [78,79], whereby a condition or set of conditions are necessary if they are present in all instances of the outcome (outcome Y always involves condition X) and they do sufﬁce if a particular outcome appears whenever the condition or set of conditions are present (condition X always implies the presence of outcome Y) [16,83,84]. Complexity theory proposes a higher number of complex relationships than of cause-effect relationships between causal conditions and outcomes [85]. 2.3. SRI Indices [95] analyse the impact on ROE of diverse corporate boards and ﬁrm characteristics, and ﬁnd an interesting exception to the general rule where higher leverage and property dispersion demand an extended board of directors to reach a higher ROE. Primc and ˇCater [96] test the inﬂuence on ﬁrm performance (composite index of perceptual ﬁnancial and non-ﬁnancial measures) of companies’ environmental proactivity, life-cycle stage, and the uncertainty, complexity and muniﬁcence of the business environment. Slager [84] applies fsQCA to the FTSE4Good SRI index to analyse listed companies’ reactivity (improvement on scores of the FTSE4Good criteria in 10 of 31 Sustainability 2019, 11, 1211 order to remain included), considering as causal conditions the degree of engagement with the FTSE Responsible Investment team, the number of years in the index and whether the company has ever been excluded or not from the index, gives publicity to the inclusion and is classiﬁed by the FTSE4Good as having high or low CSR- related risks. Finally, Isaksson and Woodside [17], develop a comprehensive study of the heterogeneous relationships that condition the CSP and CFP link, stating that good or bad management needs to be considered together with CSP in order to explain CFP. They test 27 different propositions through alternative fsQCA model speciﬁcations, concluding that there is more evidence of the inﬂuence of bad management and the negation of CSP on negative CFP than of high overall CSP supporting high CFP outcomes. pp g g This study, therefore, applies the fsQCA technique to explore the necessary and sufﬁcient conditions of CSD and CFP that explain CSP (being included in or excluded from an SRI index) and to analyse whether different CSD and CFP factors are able to jointly or individually explain CSP. This analytical choice allows us to study the diversity of causal paths leading to an inclusion in or exclusion from an SRI index (associated to a good or poor CSP) and to explore how CSD and CFP combine to describe or explain a company’s entry or exit. This is a preliminary and exploratory study aimed at both testing the validity of the fsQCA methodology for the analysis of the corporate sustainability and ﬁnancial performance link, and searching for complex relations that offer avenues for further research. Therefore, we deﬁne some additional propositions that uncover the presence of the main principles of complexity theory according to Woodside [88]. Following Cuadrado-Ballesteros et al. 3. Data and Methodology In this section we ﬁrst present and justify the choice of one of the numerous SRI stock indices available and of speciﬁc measures of CFP and CSD. Next, we explain how we have deﬁned and conﬁgured the sample and provide some descriptive information. Then, we outline the methodological foundations of the fsQCA, while the last subsection reports the fsQCA variables (outcome and conditions) and the alternative model speciﬁcations. 2.3. SRI Indices [85] we suggest four additional propositions to be explored through a fsQCA model relating CSD, CFP and CSP (entering/exiting an SRI index): Proposition 4. Entering (or exiting) the index can be achieved through different conﬁgurations of CSD and CFP (equiﬁnality). Proposition 5. The effect of a condition on the inclusion in or exclusion from the index depends on other conditions (complexity). Proposition 5. The effect of a condition on the inclusion in or exclusion from the index depends on other conditions (complexity). Proposition 6. Conﬁgurations associated with exclusion from the index are not the mirror opposite of those associated with inclusions in the index (causal asymmetry). Proposition 6. Conﬁgurations associated with exclusion from the index are not the mirror opposite of those associated with inclusions in the index (causal asymmetry). Proposition 7. The same CFP and CFP characteristics can contribute both to the inclusion in and exclusion from the index (asymmetry). Proposition 7. The same CFP and CFP characteristics can contribute both to the inclusion in and exclusion from the index (asymmetry). These four additional propositions will provide a deeper and better understanding of this study’s results regarding Propositions 1a., 1b., 2a., 2b., 3a. and 3b. In particular, insofar as fsQCA speciﬁes alternative models to analyse both the presence and the absence of the outcome (good CSP or inclusions in an SRI index and poor CSP or exclusions from a SRI index), proposition 7 is meant to conﬁrm that Propositions 1b., 2b. and 3b. are not just the mirror opposite of Propositions 1a., 2a. and 3a. 3.1. Selection of an SRI Index Both indices use ESG scores provided by the investment and rating agency RobecoSAM, but while the former is a standard capitalisation-weighted index, the latter is a smart-beta index that bases its composition on ESG factor scores proposed by RobecoSAM. RobecoSAM develops an annual process of Corporate Sustainability Assessment (CSA), which is also the basis for the Dow Jones Sustainability Indices (DJSI) family, and is based either on a questionnaire proposed to companies, or on publicly available information [9]. For the S&P 500 E&S, RobecoSAM provides an Environmental and Social (E&S) Score that is the weighted average of the scores for the Environmental and Social dimensions received during the CSA process. For the S&P 500 ESG, RobecoSAM provides total scores (or classic scores, also used in the DJSI) that are the weighted average of the scores for the Environmental, Social and Economic dimensions assessed through the CSA. One main difference between the two indices regarding their eligible universe, is that the S&P 500 E&S excludes companies involved in fossil fuel-heavy industries or in the production and sale of tobacco, cluster bombs, landmines, nuclear and other military armaments. Then, the selection of index constituents in the S&P 500 E&S starts sorting the rest of companies of the S&P 500 by 11 GICS sectors and then by E&S Scores, selecting companies in decreasing order till reaching 75% of the ﬂoat-adjusted market capitalization in each sector. For index calculation, constituents are weighted by free ﬂoat market capitalisation. The S&P 500 ESG transform the classic scores derived from the CSA into ESG factor scores under a smart-beta approach, trying to prioritize ﬁnancial materiality and to neutralize the effect of biases such as GICS industries, sub-industries and countries as well as investment factors as momentum, value and size. As a result of these methodological differences, the S&P 500 ESG index is not well-suited for the purpose of our study, as its constituents are practically the same as those of its underlying index (as of 29 June 2018, the S&P 500 showed 505 constituents and the S&P 500 ESG 494). Given that we want to analyse inclusion and exclusion events, the S&P 500 E&S index (with 292 constituents in the same date) is the obliged choice. 3.1. Selection of an SRI Index There are many sustainability, CSR or ESG-related stock indices, based on different investment universes regarding their geographical scope and their relation to general or speciﬁc ESG topics [9]. The United States equity market is of particular interest when analysing investment strategies and results, given its strong efﬁciency and well-established institutional dynamics, and considering that it is a reference market worldwide and a barometer of the US economy in particular, and of the whole world economy generally [97,98]. Regarding the relevance and importance of ESG matters in the US 11 of 31 Sustainability 2019, 11, 1211 capital markets, although in the US, corporate sustainability reporting remains on a voluntary basis, and reporting rates are under the European and Asian ones [99], its business environment seems more demanding of this type of information [100]. US companies are assumed to develop an explicit or more strategically integrated CSR, compared to an implicit CSR in Europe, oriented towards meeting societal expectations [101]. By focusing on the US market, we leave aside the effect of different institutional, cultural and legal environments [62,85] that could condition the CSP-CFP relationship. The study uses the US stock market S&P 500 index, managed by the S&P Dow Jones Indices company because it provides a wide representation of the US market, capturing approximately 80% coverage of available market capitalization [102]. The Government & Accountability Institute [103] has analysed the S&P 500 constituents’ sustainability reporting activities from 2011 onwards, and has found that sustainability reporting has risen from just 20% of the companies reporting in 2011, to 85% in 2017. Consequently, the S&P 500 index seems an appropriate reference context to fulﬁl the goal of this study. Tamimi and Sebastianelli [104] offer a complete analysis of the CSD of S&P 500 companies focusing on their Bloomberg ESG disclosure scores. The S&P Dow Jones Indices company has launched a good number of sustainability-related stock indices with different foci and eligible universes (118 indices were shown in July 2018 on its web), but only two have the S&P 500 index as their underlying index in relation to global sustainability or ESG issues: The S&P 500 Environmental and Socially Responsible Index (S&P 500 E&S) and the S&P 500 ESG Factor Weighted Index (S&P 500 ESG). 3.2. Measures of CSD and CFP Drawing from the review of previous research and on the study’s objectives formalised in Propositions 1a., 1b., 2a., 2b., 3a. and 3b., we have selected different variables to be considered as causal conditions in an fsQCA. Drawing from the review of previous research and on the study’s objectives formalised in Propositions 1a., 1b., 2a., 2b., 3a. and 3b., we have selected different variables to be considered as causal conditions in an fsQCA. CSD is analysed through three dichotomous variables and a continuous one: Being included or not in the GRI sustainability disclosure (GRI SD) database, having or not having a sustainability report about the year before the inclusion or exclusion event, and whether this report explicitly follows some of the GRI standards as dichotomous variables, and the number of published sustainability reports registered on the GRI SD database as continuous variable. The use of the GRI sustainability reporting guidelines is generalised worldwide; it is the most common standard among reporting companies [52,106,107], and the GRI SD database is a common reference to deﬁne study samples in the analysis of the CSP-CSD-CFP links [55,56,62,108]. Considering whether sustainability reports are or not compliant with some of the alternative versions of the GRI guidelines is indicative of reporting quality [56]. Given the exploratory nature of this study, and the need to limit the number of causal conditions included in the fsQCA, we have decided to consider the adherence to the GRI standards globally (similar to Goettsche et al. [62]) without distinguishing between externally assured or non-assured reports, GRI guidelines’ version or application levels (in contrast to previous studies [3,56,63,108]). Moreover, within the period of 2010–2016, many companies changed the format of their reports (from compliant to non-compliant with GRI guidelines or the opposite, and also across different application levels and assured or non-assured states), so that it would be difﬁcult to account for this in the fsQCA. Consequently, we only verify adherence to the GRI guidelines regarding reports available for the year before the inclusion or exclusion event. The availability of a report referred to the year before the inclusion or exclusion event is considered informative, as it is expected that without such relevant disclosure companies previously included in the index could be excluded. 3.2. Measures of CSD and CFP Finally, the number of published reports registered at the GRI SD database or reporting frequency has also been used in previous studies [108] to evaluate the intensity of a company’s commitment with CSD. We consider the total number of reports reﬂected in the GRI database for each company entering or exiting the index, irrespective of whether these reports are adhered to the GRI guidelines, or of the different application levels. According to Proposition 2a., we expect that companies included in the GRI SD database that have published a report referred to the previous year of that of the entry or exit event, that follow the GRI guidelines and that have a high number of reports registered at the GRI SD database will exhibit a high CSP and thus represent a high proportion of the companies that experience an entry event. On the other hand, according to Proposition 2b., we expect that companies not included in the GRI SD database (so that the rest of CSD variables will be absent) will exhibit a poor CSP and thus represent a high proportion of the companies that experience an exclusion event. CFP can be measured through accounting or market measures. Accounting measures are interesting in the analysis of the sustainability-ﬁnancial performance link, because they can reﬂect ﬁnancial consequences of sustainability practices (amount of investment devoted to sustainability, costs and revenue derived from sustainability practices) and even preconditions to sustainability efforts (more ﬁnancially-sound companies can be more involved with sustainability matters). Market measures are also interesting, but a bit more complex to deal with in order to extract straightforward conclusions, as they are based on expectations and these can be inﬂuenced by past, current and expected sustainability and ﬁnancial performances. In this study, CFP is analysed both through accounting and market measures, choosing the absolute revenue ﬁgure and the earnings per share (EPS) ratio as accounting measures of CFP, and the price-earnings ratio (PER) as market measure. Although these indicators are not the most commonly used in previous literature, different literature reviews on the CSP-CSD-CFP links document the use of these alternative measures [20] and indicate mixed results regarding the sense (positive or negative) and direction of causality. We choose revenue and EPS as we assume that the possibility of developing a good CSD and achieving a good CSP is related to the availability of ﬁnancial resources. 3.1. Selection of an SRI Index Furthermore, we understand that the philosophy behind the S&P 500 ESG does not correspond with the aim of this study, as it is much more investment-oriented through a speciﬁc investment strategy (smart-beta strategy). To our knowledge, this study, together with López-Cabarcos et al. [105], are the ﬁrst academic studies focusing on the S&P 500 E&S index. Sustainability 2019, 11, 1211 12 of 31 12 of 31 3.2. Measures of CSD and CFP The PER ratio allows us to introduce in the analysis 13 of 31 13 of 31 Sustainability 2019, 11, 1211 the possible effect of market expectations regarding companies’ CSP, as well as to control for the possible effect of a higher or lower market capitalisation on the inclusion in or exclusion from the SRI index. We use both the value of these indicators at the event year, as well as the change that they have experienced in the period between an entry event and the previous exclusion from the index or between an exit event and the previous inclusion in the index. Given that we are trying to explain why a company is included again in the S&P 500 E&S index after having been excluded, or why a company is excluded from the index after its inclusion (without considering the initial index conﬁguration), this way of approaching the measurement of CFP could yield interesting results. This is a rarely applied approach for the measurement of CFP in previous literature on the relationships of CFP with CSD or CSP. Its application is derived from our distinct and novel approach in this study focusing on companies that have experienced a decline in their CSP (exit the S&P 500 E&S index after a previous inclusion event), or that show an improvement in their CSP (entry the S&P 500 E&S index after a previous exclusion event). Considering CFP dynamically allows us to deal with the debate on CSP and CFP concurrency in time [14]. In contrast to Ruggiero and Cuppertino [15], we do not consider any time lag as we are neither looking for causality between CFP and CSP, nor explicitly advocating the slack resources theory (although it does inﬂuence our choice of CFP measures), that is, we do not want to test if the availability of extra resources in one year increases CSP the following year. Instead, we want to test whether a high CSP leading to an inclusion in an SRI index is accompanied by a high CFP (Proposition 1a.) or an improved CFP (Proposition 3a.), and whether a poor CSP leading to an exclusion from a SRI index is accompanied by a low CFP (Proposition 1b.) or a decrease in CFP (Proposition 3b.). 3.2. Measures of CSD and CFP Although our contemporaneous measurement of CSP and CFP is justiﬁed because we are looking for associations but not for causality, it has been acknowledged that CSP and CFP can under some circumstances be related at the same point in time [14], as long as concurrent measurement does not imply absolute simultaneity, and using a market measure does not impede the interpretation of causality, as we can assume that the PER ratio reﬂects expectations on future CSP and CFP. Regardless, also considering the change experienced in CFP between two consecutive inclusion and exclusion events, allows us to some extent to account for long-term or deferred effects of CFP on CSP. Finally, as an additional casual condition, we consider the number of years between an entry event and the previous exit, or between an exit event and the previous entry, assuming that a long period is needed to cause a signiﬁcant change in the CSP of a company that leads to an inclusion in the S&P 500 E&S index after a previous exclusion, or to an exclusion after a previous inclusion. Note that this variable measures the period of inclusion for exclusion events (that is, number of years a company had been included in the S&P 500 E&S index before being excluded) and the period of exclusion for inclusion events (number of years a company remained excluded before being included again). 3.3. Sample and Data Collection The S&P 500 E&S index was launched in May 2015 with a ﬁrst value date of 30 September 2010. Table 1 shows the evolution of the number of constituents of the S&P 500 E&S index and the S&P 500 during the period 2010–2016. Using the data of companies listed in the S&P 500 and the S&P 500 E&S during this period, we identiﬁed companies’ entries and exits in both indices and generated a database comprising only those entries subsequent to the index’s ﬁrst conﬁguration of 2010 (that is, entries from 2011 onwards) that do not correspond to new listings in the underlying index (S&P 500), and only those exits not related to a delisting or removal from the S&P 500. We have registered a total of 440 companies that were listed at some point in the S&P 500 E&S index in the period 2010–2016 (checked over end-of-year lists of constituents); 237 of them were present at the initial conﬁguration of the index in 2010. These companies have been involved in 488 index inclusion and exclusion events after its initial conﬁguration. In order to consider only the inclusion in the index after a previous exclusion, or the exclusion from the index after a previous inclusion, the inclusions that were not preceded by an exclusion event and all exclusions that happened after the 14 of 31 Sustainability 2019, 11, 1211 inclusion at the initial conﬁguration of the index were removed from the sample. Furthermore, we did not consider inclusions or exclusions motivated by a corresponding inclusion in, or exclusion from the underlying index (S&P 500). inclusion at the initial conﬁguration of the index were removed from the sample. Furthermore, we did not consider inclusions or exclusions motivated by a corresponding inclusion in, or exclusion from the underlying index (S&P 500). Table 1. Companies listed at end of each year (2010–2016). Table 1. Companies listed at end of each year (2010–2016). Table 1. Companies listed at end of each year (2010–2016). Year S&P 500 S&P E 2010 500 237 2011 500 229 2012 500 243 2013 500 258 2014 502 275 2015 504 284 2016 505 284 After this process, the sample comprises 126 events, 65 of which correspond to inclusions after a previous exclusion, and 61 to exclusions after a previous inclusion. These 126 events correspond to 99 different companies. 3.3. Sample and Data Collection Table 2 shows the distribution of these 99 companies by GICS sector, with the Financials and Information Technology categories being the most represented sectors, with 17 companies each one, and the Telecommunications Services category not being represented at all. Table 2. Distribution of companies by GICS sector. GICS Sector Number of Companies in the Sample Number of Companies Included in the GRI Database Financials 17 11 Information Technology 17 11 Consumer Discretionary 16 10 Consumer Staples 13 13 Industrials 10 8 Health Care 6 4 Materials 6 5 Utilities 5 5 Real Estate 5 5 Energy 4 3 Telecommunication Services 0 0 Total 99 75 Table 2. Distribution of companies by GICS sector. Table 2. Distribution of companies by GICS sector. Of the 99 companies, 75 are included in the GRI SD database (see also Table 2 with distribution by GICS sector), two have 15 sustainability reports registered at the database, and another one has 14. Th j it f i h b t t d ﬁ i t d t i bilit t (49) hil l Of the 99 companies, 75 are included in the GRI SD database (see also Table 2 with distribution by GICS sector), two have 15 sustainability reports registered at the database, and another one has 14. The majority of companies have between two and ﬁve registered sustainability reports (49), while only six companies have only one registered sustainability report at the GRI database. For 66 events, there were sustainability reports referred to the previous year at the GRI database (corresponding to 53 companies) but only 37 (corresponding to 28 companies) were adjusted to GRI standards. The majority of companies have between two and ﬁve registered sustainability reports (49), while only six companies have only one registered sustainability report at the GRI database. For 66 events, there were sustainability reports referred to the previous year at the GRI database (corresponding to 53 companies) but only 37 (corresponding to 28 companies) were adjusted to GRI standards. Table 3 shows the distribution of entry and exit events by year and GICS sector. The year 2016 has the highest number of registered events (44), while 2012 has only 11; the ‘Financials’ and ‘Consumer Staples’ GICS sectors are the most represented (21 and 20 events), while the ‘Energy’ (6), ‘Utilities’ (6) and Heath Care (7) sectors are the least represented ones. 3.3. Sample and Data Collection Table 3 shows the distribution of entry and exit events by year and GICS sector. The year 2016 has the highest number of registered events (44), while 2012 has only 11; the ‘Financials’ and ‘Consumer Staples’ GICS sectors are the most represented (21 and 20 events), while the ‘Energy’ (6), ‘Utilities’ (6) and Heath Care (7) sectors are the least represented ones. 15 of 31 Sustainability 2019, 11, 1211 Table 3. Number of events by year and GICS sector. Table 3. Number of events by year and GICS sector. Table 3. Number of events by year and GICS sector. GICS Sector 2012 2013 2014 2015 2016 TOTAL Financials 2 8 4 2 5 21 Information Technology 1 0 6 2 9 18 Consumer Discretionary 1 0 3 4 8 16 Consumer Staples 5 3 6 2 4 20 Industrials 0 1 2 3 6 12 Health Care 0 0 1 3 3 7 Materials 1 3 2 2 2 10 Utilities 0 3 0 1 2 6 Real Estate 1 0 4 2 3 10 Energy 0 2 1 1 2 6 TOTAL 11 20 29 22 44 126 Table 4 shows the distribution of events both by event type (exit or entry) and event year. Finally, regarding the number of years that have passed between an inclusion event and the previous exclusion (exit-entry event) or and exclusion event and its previous inclusion (entry-exit event), most events involve just one year between inclusion-exclusion or exclusion-inclusion events (60 events), 32 events accumulate two years, 22 events accumulate three years, seven events accumulate four years and ﬁve events accumulate ﬁve years. Table 4. Distribution of events by event type and year. Event Type 2012 2013 2014 2015 2016 Exits 2 11 14 11 23 Entries 9 9 15 11 21 Table 4. Distribution of events by event type and year. CFP data about revenue and EPS were collected through the Edgar Online database, while data about historical stock prices were retrieved from the Nasdaq’s website. Table 5 shows a summary of the statistics for the 126 events. Table 5. Summary statistics of CFP variables. Table 5. Summary statistics of CFP variables. CFP Measure Mean Std. Dev. 3.3. Sample and Data Collection Max Min Revenue (event year) $21,293.10 $33,827.04 $179,770.00 $984.36 EPS (event year) 3.52 4.31 40.70 −4.64 PER (event year) 18.71 86.23 422.25 −843.09 Variation in revenue 10.70% 34.79% 285.26% * (increase) 0.09% * (increase) Variation in EPS 18.67% 179.25% 1,275.00% * (decrease) 0.56% * (increase) Variation in PER 33.92% 406.44% 3,692.32% * (decrease) 0.05% * (increase) * Calculated over absolute values of variations in order to properly compare the magnitude of changes. 3.4. Fuzzy-Set Qualitative Comparative Analysis 3.4. Fuzzy-Set Qualitative Comparative Analysis Following Ragin [79], the ﬁrst step in QCA is a purposeful and well-founded selection of cases, with a clear deﬁnition of the outcome to delineate the sample, while trying to guarantee that cases are comparable, and presenting enough variance in their causal conditions to infer conclusions about patterns and trends [84]. The second step is the calibration of each case’s characteristics (outcome and conditions) into sets. These can be crisp and/or fuzzy sets [92], requiring the identiﬁcation of the critical levels for considering full membership, non-membership and the cross-over or ambiguity point. These critical levels should be based on qualitative and substantive knowledge of each characteristic [90]. After coding data, we enter the analytical phase conducting a necessity and a sufﬁciency analysis [83]. Necessary conditions are those present in all cases that display the outcome, and sufﬁcient 16 of 31 16 of 31 Sustainability 2019, 11, 1211 conditions are those that always lead to a particular outcome [92]. Necessary conditions are not always sufﬁcient to produce the outcome by themselves, and sufﬁcient conditions may not be the only conditions that lead to a particular result, so there may be multiple sufﬁcient causes that combine, creating a speciﬁc conﬁguration or path leading to a speciﬁc result for a speciﬁc set of cases. The sufﬁciency analysis is developed both in relation to the outcomes’ presence and absence [84]. g p g p g p p sufﬁciency analysis is developed both in relation to the outcomes’ presence and absence [84]. Two measures are deﬁned to analyse the necessity or sufﬁciency of causal conditions and their combinations: Consistency and coverage. Consistency refers to the degree to which the cases sharing a given combination of conditions coincide in producing the outcome [83,85]. A low consistency indicates that a conﬁguration is not supported by empirical evidence [16]. The researcher must select a cut-off consistency value to decide which paths to include in the ﬁnal solution [92]. Coverage refers to the number of cases for which a conﬁguration is valid [16], that is, the proportion of cases following a speciﬁc path [83]. Low conﬁguration coverages are common when there are many causal conﬁgurations, so it is assumed that they do not imply less relevance as the path is useful to explain a number of cases in which the outcome takes place [16,84]. Thus, high consistency is more important than high coverage [94]. 3.4. Fuzzy-Set Qualitative Comparative Analysis FsQCA provides three types of solutions which show different sets of paths [83,92]: A complex solution where the software makes no simplifying assumptions, hence the most appropriate; a parsimonious solution, which uses the remainders (i.e., combinations of the antecedent conditions not observed in the dataset) to simplify the solution and should only be used if the assumptions are fully justiﬁed; and an intermediate solution, which distinguishes between “easy” and “strong” assumptions and simpliﬁes the solution only considering the “easy” remainders. Finally, the alternative conﬁgurations or paths leading to the outcome’s presence or absence must be interpreted according to the knowledge of the cases as well as the theoretical or practical foundations. 3.5. Model Speciﬁcation Given that we want to analyse the paths through which CSD and CFP contribute to CSP, CSP is the outcome of the fsQCA models to be tested in this study. This outcome is deﬁned in relation to the inclusion and exclusion events involving the S&P 500 E&S index after a previous exclusion or inclusion, considering that inclusion reﬂects a good CSP and exclusion corresponds to a poor CSP. In this way, the outcome of the fsQCA is operationalised through a binary variable that takes value 1 for inclusions (after a previous exclusion) and value 0 for exclusions (after a previous inclusion), so that the presence of the outcome (good CSP) is equivalent to being included in the index after a previous exclusion, and the absence of the outcome (poor CSP) corresponds to being excluded from the index after a previous inclusion. The study comprises the following conditions (including four conditions of CSR disclosure, six of ﬁnancial performance, and one additional condition that accounts for the time elapsed between inclusion/exclusion and previous exclusion/inclusion): (1) A binary condition that takes value 1 if the company is included in the GRI SD database, and value 0 otherwise (“GRI”); (2) a binary condition that takes value 1 if the company has published a sustainability report in the event year (referred to the previous year), and value 0 otherwise (“ReportEvent”); (3) a binary condition that takes value 1 if the report is in accordance with the GRI Guidelines, and value 0 otherwise (ReportGRI); (4) the total number of sustainability reports available at the GRI SD database for the company (“fs_noreports"); (5) the company’s revenue in the event year (“fs_rev”); (6) the company’s earnings-per-share ratio in the event year (“fs_eps”); (7) the company’s price-earnings ratio in the event year (“fs_per”); (8) the variation in revenue between the event year and the year of the previous inclusion or exclusion (“fs_revenuevar”); (9) the variation in EPS between the event year and the year of the previous inclusion or exclusion (“fs_epsvar”); and (10) the variation in PER between the event year and the year of the previous inclusion or exclusion (“fs_pervar”); (11) the number of years between the outcome of being included in or excluded from the index, and the year of the previous inclusion or exclusion (“fs_noyears”). 3.5. Model Speciﬁcation Sustainability 2019, 11, 1211 17 of 31 17 of 31 These six ﬁnancial conditions are used to test two different models: Model I. includes static ﬁnancial variables in the event year (Conditions 5., 6. and 7.), and Model II. includes dynamic ﬁnancial variables between the event year and the year of the previous inclusion or exclusion (Conditions 8., 9. and 10.). This study uses the direct calibration method [79]. Due to a lack of theoretical speciﬁcations of “high” versus “low” scores, calibration of “noyears" and “noreports” and the six ﬁnancial conditions followed the distribution of the actual data. We use the 95th, 5th and 50th percentiles [92,94] to determine full membership, full non-membership and the crossover point, respectively. After calibrating the conditions, the analysis identiﬁes the necessary conditions for being included in the S&P 500 E&S index (CPS), and the necessary conditions for being excluded from the index (~CPS). Here “~” represents the absence of the condition. The second step consists of a sufﬁciency analysis for the following four models. Equations (1) and (2) correspond to Model I. in relation to the outcome’s presence or absence, and Equations (3) and (4) correspond to Model II. in relation to the outcome’s presence or absence: CSP = f (GRI, ReportEvent, ReportGRI, fs_noyears, fs_noreports, fs_rev, fs_eps, fs_per) (1) ~CSP = f (GRI, ReportEvent, ReportGRI, fs_noyears, fs_noreports, fs_rev, fs_eps, fs_per) (2) CSP = f (GRI, ReportEvent, ReportGRI, fs_noyears, fs_noreports, fs_revvar, fs_epsvar, fs_pervar) (3) ~CSP = f (GRI, ReportEvent, ReportGRI, fs_noyears, fs_noreports, fs_revvar, fs_epsvar, fs_pervar) (4) The estimation software is fsQCA 3.0 [109]. The estimation software is fsQCA 3.0 [109]. 4. Results and Discussion This section presents the results of the fsQCA regarding the necessary and sufﬁcient conditions for the outcome of exclusion from or inclusion in the SRI index. The main ﬁndings and implications are discussed. 4.1. Necessity Analysis Table 6 shows the results of the necessity analysis for the all the conditions included in the two alternative models speciﬁed. A condition is necessary when its consistency value is above 0.9 [79,80,110]; coverage scores are provided for completeness, but no criterion exists for determining the necessity of a condition. The negation rows (shown with a “~” preceding the condition code) present the respective logical negation of the identiﬁed conditions. No necessary conditions exist for either the presence of the outcome (CSP) or the absence of the outcome (~CSP); that is, no condition is present every time the outcome is present. Therefore, no unique condition is necessary to explain neither the exclusion from nor the inclusion in the S&P 500 E&S index and, by extension, a good or poor CSP. 4.2. Sufﬁciency Analysis This section presents the analysis of sufﬁcient conditions. As this is a ﬁrst exploratory study, we set a frequency cut-off of only one case for a combination of causal conditions to be considered in the analysis. The estimation of the four models presented in Section 3.5 (Equations (1)–(4)) yields a solution consistency value above 0.75, which is usually the minimum acceptable value [79]. Tables 7–10 show the results of the fsQCA complex solution for Models I. and II., comprising different paths that lead to a good CSP, which motivates inclusion in the S&P 500 E&S index (Table 7; Table 9) or to a poor CSP, which motivates exclusion from the index (Table 8; Table 10). Table 7; Table 8 show the results for Model I. (considering CFP static measures) and Table 9; Table 10 show the results for Model II. (dynamic CFP measures). Consistency (of the entire solution and of each individual path) indicates the extent to which a causal combination leads to the inclusion in or exclusion from the index, and coverage 18 of 31 Sustainability 2019, 11, 1211 represents how much of the outcome (inclusion or exclusion) is explained by each path and by the solution as a whole [92]. Raw coverage is the percentage of events displaying both the speciﬁc event type (exclusion or inclusion, that is, ~CSP and CSP) and the speciﬁc path; as the same events can display various paths, unique coverage shows the percentage of events covered by that single path and no other [79,84]. Table 6. Necessity analysis. “fs” represents a calibrated condition; “~” represents absence of the condition; “” is logic operator AND; “CSP” represents an inclusion event in the S&P 500 E&S index; “GRI” represents inclusion in the GRI SD database; “ReportEvent” indicates if the company published a report in the event year; “ReportGRI” indicates if the report follows the GRI standards; “noyears” indicates the number of years between the event year and the previous exit or entry; “noreports” indicates the number of reports included in the GRI SD database; “rev” stands for the revenue ﬁgure in the event year; “eps” is the EPS in the event year; “per” is the PER in the event year; frequency cut-off: 1; consistency cut-off: 0.77348. 4.2. Sufﬁciency Analysis CSP (Inclusion) ~CSP (Exclusion) Conditions Consistency Coverage Consistency Coverage GRI 0.876923 0.575758 0.688525 0.424242 ~GRI 0.123077 0.296296 0.311475 0.703704 ReportEvent 0.630769 0.621212 0.409836 0.378788 ~ReportEvent 0.369231 0.4 0.590164 0.6 ReportGRI 0.415385 0.72973 0.163934 0.27027 ~ReportGRI 0.584615 0.426966 0.836066 0.573034 fs_noyears 0.394769 0.52755 0.376721 0.472451 ~fs_noyears 0.605231 0.508532 0.623279 0.491469 fs_noreports 0.564461 0.613956 0.378197 0.386044 ~fs_noreports 0.435538 0.427385 0.621803 0.572615 fs_rev 0.414308 0.52444 0.400328 0.47556 ~fs_rev 0.585692 0.50998 0.599672 0.49002 fs_eps 0.437692 0.49651 0.472951 0.50349 ~fs_eps 0.562308 0.532023 0.527049 0.467977 fs_per 0.418 0.497984 0.449016 0.502016 ~fs_per 0.582 0.529535 0.550984 0.470465 fs_revvar 0.475385 0.493768 0.519344 0.506232 ~fs_revvar 0.524615 0.537685 0.480656 0.462315 fs_epsvar 0.444462 0.494946 0.483279 0.505054 ~fs_epsvar 0.555538 0.533935 0.516721 0.466065 fs_pervar 0.477846 0.536535 0.439836 0.463465 ~fs_pervar 0.522154 0.498312 0.560164 0.501688 “fs” represents a calibrated condition; “~” represents absence of the condition or outcome; “CSP” represents corporate sustainability performance, with presence corresponding to an inclusion event in the S&P 500 E&S index (good CSP), and absence corresponding to an exclusion event (poor CSP); “GRI” represents inclusion in the GRI SD database; “ReportEvent” indicates whether the company published a report in the event year; “ReportGRI” indicates whether the report follows the GRI standards; “noyears” indicates the number of years between the event year and the previous exclusion or inclusion; “noreports” indicates the number of reports included in the GRI SD database; “rev” stands for the revenue ﬁgure in the event year; “eps” is the EPS in the event year; “per” is the PER in the event year; “revvar”, “epsvar” and "pervar" stand respectively for the change in the corresponding ﬁgures between the event year and the year of the previous exclusion or inclusion. Table 6. Necessity analysis. Table 7. Sufﬁciency Analysis. Model I-Presence: CSP (inclusion). Table 7. Sufﬁciency Analysis. Model I Presence: CSP (inclusion). 4.2. Sufﬁciency Analysis Combination Raw Coverage Unique Coverage Consistency “fs” represents a calibrated condition; “~” represents absence of the condition; “” is the logic operator AND; “CSP” represents an inclusion in the S&P 500 E&S index; “GRI” represents inclusion in the GRI SD database; “ReportEvent” indicates whether the company published a report in the event year; “ReportGRI” indicates whether the report follows the GRI standards; “noyears” indicates the number of years between the event year and the previous exclusion or inclusion; “noreports” indicates the number of reports included in the GRI SD database; “rev” stands for the revenue ﬁgure in the event year; “eps” is the EPS in the event year; “per” is the PER in the event year; frequency cut-off: 1; consistency cut-off: 0.760926. Table 9. Sufﬁciency Analysis. Model II-Presence: CSP (inclusion). Path Combination Raw Coverage Unique Coverage Consistency 1 GRIReportEventReportGRI~fs_noyears fs_noreports~fs_epsvar~fs_pervar 0.141077 0.0873846 0.764804 2 GRIReportEventReportGRIfs_noyears fs_noreportsfs_epsvar~fs_pervar 0.0795385 0.0166154 0.848933 3 GRIReportEventReportGRIfs_noyears fs_noreports~fs_epsvarfs_pervar 0.107231 0.0470769 0.882278 solution coverage: 0.216308 solution consistency: 0.790332 “fs” represents a calibrated condition; “~” represents absence of the condition; “” is logic operator AND; “CSP” represents an inclusion in the S&P 500 E&S index; “GRI” represents inclusion in the GRI SD database; “ReportEvent” indicates whether the company published a report in the event year; “ReportGRI” indicates whether the report follows the GRI standards; “noyears” indicates the number of years between the event year and the previous exclusion or inclusion; “noreports” indicates the number of reports included in the GRI SD database; “epsvar” and "pervar" stands respectively for the change in the corresponding ﬁgures between the event year and the year of the previous exclusion or inclusion; frequency cut-off: 1; consistency cut-off: 0.755935. Table 10. Sufﬁciency Analysis. Model II-Absence: ~CSP (exclusion). Table 10. Sufﬁciency Analysis. Model II-Absence: ~CSP (exclusion). 4.2. Sufﬁciency Analysis Path Combination Raw Coverage Unique Coverage Consistency 1 GRIReportEventReportGRI fs_noyearsfs_noreportsfs_rev~fs_per 0.0746154 0.00492308 0.911654 2 GRIReportEventReportGRI fs_noyearsfs_noreportsfs_eps~fs_per 0.084 0.0109231 0.845201 3 GRIReportEventReportGRI fs_noyearsfs_noreportsfs_revfs_eps 0.0721538 0.00492309 0.901923 4 GRIReportEventReportGRI~fs_noyears* fs_noreports~fs_rev~fs_eps~fs_per 0.107692 0.0607692 0.77348 5 GRIReportEventReportGRIfs_noyears* fs_noreports~fs_rev~fs_epsfs_per 0.0783077 0.0295385 0.852596 solution coverage: 0.187846 solution consistency: 0.804348 “fs” represents a calibrated condition; “~” represents absence of the condition; “” is logic operator AND; “CSP” represents an inclusion event in the S&P 500 E&S index; “GRI” represents inclusion in the GRI SD database; “ReportEvent” indicates if the company published a report in the event year; “ReportGRI” indicates if the report follows the GRI standards; “noyears” indicates the number of years between the event year and the previous exit or entry; “noreports” indicates the number of reports included in the GRI SD database; “rev” stands for the revenue ﬁgure in the event year; “eps” is the EPS in the event year; “per” is the PER in the event year; frequency cut-off: 1; consistency cut-off: 0.77348. 19 of 31 Sustainability 2019, 11, 1211 Table 8. Sufﬁciency Analysis. Model I-Absence: ~CSP (exclusion). Path Combination Raw Coverage Unique Coverage Consistency 1 ~GRI~ReportEvent~ReportGRI~fs _noyears~fs_noreports~fs_revfs_epsfs_per 0.0621311 0.0314754 0.768763 2 ~GRI~ReportEvent~ReportGRIfs_noyears* ~fs_noreportsfs_revfs_epsfs_per 0.0485246 0.0178689 0.760926 solution coverage: 0.08 solution consistency: 0.792208 “fs” represents a calibrated condition; “~” represents absence of the condition; “” is the logic operator AND; “CSP” represents an inclusion in the S&P 500 E&S index; “GRI” represents inclusion in the GRI SD database; “ReportEvent” indicates whether the company published a report in the event year; “ReportGRI” indicates whether the report follows the GRI standards; “noyears” indicates the number of years between the event year and the previous exclusion or inclusion; “noreports” indicates the number of reports included in the GRI SD database; “rev” stands for the revenue ﬁgure in the event year; “eps” is the EPS in the event year; “per” is the PER in the event year; frequency cut-off: 1; consistency cut-off: 0.760926. Table 8. Sufﬁciency Analysis. Model I-Absence: ~CSP (exclusion). Path Combination Raw Coverage Unique Coverage Consistency 1 ~GRI~ReportEvent~ReportGRI~fs _noyears~fs_noreports~fs_revfs_epsfs_per 0.0621311 0.0314754 0.768763 2 ~GRI~ReportEvent~ReportGRIfs_noyears* ~fs_noreportsfs_revfs_epsfs_per 0.0485246 0.0178689 0.760926 solution coverage: 0.08 solution consistency: 0.792208 “fs” represents a calibrated condition; “~” represents absence of the condition; “” is the logic operator AND; “CSP” represents an inclusion in the S&P 500 E&S index; “GRI” represents inclusion in the GRI SD database; “ReportEvent” Table 8. Sufﬁciency Analysis. Model I-Absence: ~CSP (exclusion). 4.2. Sufﬁciency Analysis Path Combination Raw Coverage Unique Coverage Consistency 1 ~GRI~ReportEvent~ReportGRIfs_noyears~ fs_noreports~fs_revenuevar~fs_epsvar~fs_pervar 0.0521311 0.0162295 0.75 2 ~GRI~ReportEvent~ReportGRI~fs_noyears~ fs_noreportsfs_revenuevarfs_epsvar~fs_pervar 0.0896721 0.0537705 0.762901 solution coverage: 0.105902 solution consistency: 0.769048 “fs” represents a calibrated condition; “~” represents absence of the condition; “*” is logic operator AND; “CSP” represents an inclusion in the S&P 500 E&S index; “GRI” represents inclusion in the GRI SD database; “ReportEvent” indicates whether the company published a report in the event year; “ReportGRI” indicates whether the report follows the GRI standards; “noyears” indicates the number of years between the event year and the previous exclusion or inclusion; “noreports” indicates the number of reports included in the GRI SD database; "revvar", “epsvar” and "pervar" stands respectively for the change in the corresponding ﬁgures between the event year and the year of the previous exclusion or inclusion; frequency cut-off: 1; consistency cut-off: 0.75. The fsQCA complex solution shows ﬁve different paths related to the inclusion in the S&P 500 E&S and based on CSD and CFP static measures (Table 7). All the paths include high levels of the four CSD causal conditions, and all except Path 4. include a high number of years elapsed between the inclusion year and the year of the previous exclusion. Therefore, in a high number of cases, a long period of exclusion favours inclusion in the index. Regarding the static CFP measures, there is one path (Path 4.) that shows low levels for all the three CFP measures considered in Model I. speciﬁcation and a shorter period of exclusion. This path reveals that companies with a high commitment to CSD are included in the index despite a poor CFP. We could argue that cases included in Path 4. are companies that have rapidly reacted to the previous exclusion event by making an extraordinary ﬁnancial effort to improve 20 of 31 Sustainability 2019, 11, 1211 their CSP and return to the index, in line with Slager [84], and that are trying to properly communicate that effort through a high commitment to CSD. The rest of paths show different combinations of high and low CFP static measures. Paths 1. to 3. reveal some support to the slack resources theory, as they include high levels of at least one accounting-based CFP measure (revenue for Path 1., EPS in Path 2., and revenue and EPS in Path 3.), but they are contradicted by Paths 4. and 5. 4.2. Sufﬁciency Analysis showing that low levels of revenue and EPS can also be present when a company is included in the index. Path 5. corresponds to companies that have been included in the index after a long exclusion period, and that exhibit a good CSD, a poor CFP based on accounting measures and a high PER. This could be indicative of a market capitalization effect leading to a “technical” inclusion in the index not directly derived from an improved CSP. The PER ratio shows low levels in Paths 1. and 2., and is not relevant in Path 3., indicating that companies included in the index and exhibiting a high CSD and a high CFP based on accounting measures, are not rewarded by the stock market with a higher stock price. Exclusions are explained by two alternative paths considering CSD and static CFP measures as causal conditions (Table 8). The two paths show low levels of CSD, accompanied in Path 2. by high levels of CFP regarding the three static measures included in Model I. This second path is the mirror opposite of Path 4., leading to inclusions (Table 7), and shows that companies not committed with CSD are excluded from the index despite having a strong CFP and a long inclusion period. Path 1. is related to companies that had been included in the index for a short period before the exclusion event and show a poor CSD together with low revenue and high EPS and PER. A high PER is present in the two paths, suggesting that the stock market does not penalise companies with a poor CSP. Moving to Model II., based on dynamic measures of CFP (variations in revenue, EPS and PER), three alternative paths lead to inclusions (Table 9). Once again, all the paths include high levels of CSD. The variation in revenue is not relevant in any of these paths. Path 1. shows that companies included in the index after a short exclusion period have experienced a decrease in their CFP, perhaps due to a great ﬁnancial effort to improve CSP and return to the index in line with Slager [84]. Paths 2. and 3. alternatively show a decrease in EPS and increase in PER, and an increase in EPS and decrease in PER with a high CSD and a long time of exclusion previous to the inclusion. 4.2. Sufﬁciency Analysis The three paths reﬂect that companies with a high commitment to CSD are included in the index, despite a deterioration of some measures of CFP. Table 10 shows two paths leading to exclusions combining CSD and dynamic measures of CFP as causal conditions. Variation in revenue appears in the two paths, and both are associated with low levels of CSD. Path 1. shows that companies with a low commitment to CSD and with larger periods of presence in the index before the exclusion event, have experienced a decrease in their CFP. Path 2. shows that companies with a poor CSD and a low inclusion period have been excluded, despite an increase in revenue and EPS. This second path also shows an increase in PER that could be motivated by the decrease in EPS. These results allow us to obtain the ﬁrst straightforward conclusion: Proposition 4. (equiﬁnality) is conﬁrmed in the four model speciﬁcations, as we can see that different conﬁgurations or paths explain the same outcome. Inclusions in the S&P 500 E&S index are explained by ﬁve paths under Model I. speciﬁcation (static CFP) and three paths under Model II. (variations in CFP); exclusions from the S&P 500 E&S index are explained by two different paths both in Model I. and Model II. In a similar vein, the results evidence the existence of complexity (Proposition 5.). Both the necessity and the sufﬁciency analyses reveal that there is no single condition related in a univocal manner to a speciﬁc outcome (inclusion or exclusion) neither in Model I. nor in Model II., that is, the effect of each condition depends on others. Regarding Proposition 6. (causal asymmetry), the results validate this proposition for Model II., where the paths leading to exclusions are not just the mirror opposite to any of the paths leading to inclusions; however, Model I. shows an exception, as Path 4. leading to the inclusion in the S&P 500 E&S is exactly the mirror opposite of Path 2. related to the exclusion. 21 of 31 Sustainability 2019, 11, 1211 Furthermore, the speciﬁc conditions considered to measuring CSD and CFP show additional interesting results. 4.2. Sufﬁciency Analysis First, we can conclude that a combination of CSD measures are a necessary precondition for the outcome to take place (either inclusions or exclusions): All the paths leading to the inclusion in the S&P 500 E&S index include the presence of high levels of the four CSD measures, while all the paths leading to the exclusion from the index show the absence of the four CSD measures. Therefore, for this combination of CSD causal conditions, Proposition 6. is rejected because there is no causal asymmetry. Proposition 7. (asymmetry) is also rejected in relation to CSD measures, because high levels of CSD are always related to inclusions and low levels of CSD are always related to exclusions. These results could be interpreted as a conﬁrmation of Propositions 2a. and 2b. In any case, these four CSD antecedents are not sufﬁcient in the four model speciﬁcations as they do not lead to the outcome by themselves, but need the concurrence of different CFP conditions. A high commitment to CSD would be expected to lead to a high CSP, while a subsequent inclusion in the S&P 500 E&S index would demand high levels of static CFP measures (Model I.). In the same vein, a low or no commitment to CSD leading to an exclusion would be associated with low static CFP measures. However, neither Paths 1. to 5. in Table 7 nor Paths 1. to 3. in Table 8 point clearly in this direction, as different paths indistinctly include both high and low levels of the same CFP measures. More strikingly, Path 4. shows that low levels of the three static CFP measures considered (revenue, EPS and PER) lead to the inclusion in the index, while Path 2. shows that high levels of the three measures lead to the exclusion. These paths openly contradict Propositions 1a. and 1b., and are an exception to Proposition 6. of causal asymmetry. Therefore, when a company shows a poor CSD, it is delisted from the index even if its CFP is strong, and when a company shows a good CSD it is included in the index despite having a low CSP. The results do not support the idea that a good CSD should be accompanied by at least one strong CFP measure to lead to an inclusion after a previous exclusion, because Path 4. 4.2. Sufﬁciency Analysis in Table 7 shows that a good CSD together with low levels of all the three CFP measures can explain inclusions. The results also conﬁrm Proposition 7. (asymmetry) because all the static CFP measures individually contribute to inclusions and exclusions in Model I. Speciﬁcally, the results conﬁrm that high levels of CFP contribute to both inclusions and exclusions. In relation to exclusions, high levels of EPS and PER are present in the two paths leading to exclusions. In particular, the role of PER is interesting because this study suggested that exclusions could be motivated by decreases in companies’ market capitalization, which would cause a “technical” delisting and not a delisting motivated by a negative change in CSP. Finding high PER levels associated with exclusions reinforces the validity of inclusions and exclusions as proxies for CSP. It is also worth mentioning that we could expect a negative relationship between PER and EPS in the different paths, because a high EPS could cause a low PER if the stock price does not capture improvements on future cash ﬂows. Pathways 2. and 5. related to inclusions (Table 7) show this interrelation, but it is neither present in Path 4. leading to inclusions nor in the two paths leading to exclusions. Regarding the different behaviour of accounting-based or market-based CFP measures, Paths 1. to 3. seem to support a positive relation between accounting CFP measures and inclusions in line with the slack resources theory, although this is contradicted by Paths 4. and 5. The results for the PER ratio in many paths show low PER related to inclusions and high PER related to exclusions. These results partially support previous evidence by event studies regarding the lack of a relevant effect of inclusions in SRI indices on stock prices, but contradict previous results of negative stock price reactions to exclusions. These results also contradict the accepted idea that a high capacity of market-based CFP measures can better capture the positive inﬂuence of a good CSP. This idea is in line with Endrikat et al. [14], who found a stronger relationship between accounting-based CFP measures and CSP, than between market-based CFP measures and CSP. Next, we focus on Model II., on the dynamic measures of CFP (variations in revenue, EPS and PER between the event year and that of the previous entry or exit). 4.2. Sufﬁciency Analysis We expected high CSD levels leading to inclusions in the index to be accompanied by an improvement in CFP, and low CSD levels leading to exclusions to combine with a deterioration of CFP. We again ﬁnd mixed evidence of the impact of CFP 22 of 31 Sustainability 2019, 11, 1211 change on inclusions (Table 9). On the other hand, Proposition 6. (causal asymmetry) is conﬁrmed, because no path leading to exclusions is the mirror opposite of paths leading to inclusions. Regarding Proposition 7. (asymmetry in the effects of the conditions on the outcome), it is only conﬁrmed for variations (either increases or decreases) in EPS and for decreases in PER, as they are related both to inclusions and exclusions. PER increases are only related to inclusions. It is noticeable that the variation in revenue does not appear in any of these paths. This is interpreted as a sign of irrelevance of this speciﬁc condition when considering the other ones. The results do not support Proposition 3a., because inclusions are not clearly associated with an improved CFP, and Path 1. in Table 9 openly contradicts this idea, because the companies included display a decrease of both EPS and PER. Proposition 3b. is supported by Path 1. in Table 10, showing that exclusions are related to decreases in the three dynamic CFP measures. Furthermore, a decrease in PER is present in the two paths leading to poor CSP (exclusions). This is consistent with the positive CFP-CSP link extensively supported by previous research. More interestingly, the rejection of Proposition 3a. and partial conﬁrmation of Proposition 3b. support the idea that CSI has stronger effects than CSR [28,29], that is, that ﬁnancial markets are more sensitive to a deterioration of CSP or to irresponsible corporate behaviours than to good CSP. Finally, large exclusion periods are associated with a high number of paths leading to inclusions both in Model I. and Model II. (only Path 4. in Table 7 and Path 1. in Table 9 show a low exclusion period related to an inclusion in the index). This result could indicate that companies that have been excluded from the index need a long period to improve their CSP and return to the index. In summary, the results fully conﬁrm Propositions 2a., 2b., 4. and 5., whereas Propositions 3b., 6., 7. receive partial support and Propositions 1a., 1b. and 3a. are rejected. Regarding Propositions 1a. 4.2. Sufﬁciency Analysis and 1b., results in Tables 7 and 8 reveal only limited evidence of a high CFP unambiguously linked to a good CSP, and a low CFP related to a poor CSP. In fact, two of the alternative conﬁgurations leading to inclusions in or exclusions from the index clearly contradict these ideas (Path 4. in Table 7 and Path 2. in Table 8). On the other hand, Propositions 2a. and 2b. are fully conﬁrmed because the fsQCA results show that the four CSD causal conditions are necessary (but not sufﬁcient) to lead to the outcome: High CSD is always related to good CSP (inclusions), and low CSD is always related to poor CSP (exclusions); however, all the paths also include different combinations of the remaining causal conditions. This positive relation between CSD and CSP is consistent with previous research [12,13]. Proposition 3a. (Table 9) is also clearly contradicted by Path 1., and ambiguously supported by Paths 2. and 3. because different inclusion events show decreases in all, or at least one, of the relevant CFP measures (variation in EPS and PER). By contrast, Proposition 3b. is clearly supported by Path 1. of Table 10, showing that low or no CSD together with a decrease in CFP measures are antecedents of an exclusion from the index. Proposition 4. (equiﬁnality) is conﬁrmed by the alternative pathways to inclusions and exclusions under the four alternative model speciﬁcations, and Proposition 5. (complexity) is conﬁrmed, because no single condition in isolation explains the outcome, and a condition’s impact depends on the presence or absence of other causal conditions. Proposition 6. (causal asymmetry) is conﬁrmed by Model II., but has an exception in Model I., where two pathways leading to inclusions and exclusions are the mirror opposite of one another. Finally, Proposition 7. (asymmetry) is partially conﬁrmed. In general, all the causal conditions are both present and absent in the determination of both the presence and absence of the outcome; however, a closer look shows that, in some cases, a speciﬁc state of the antecedent is only related to an speciﬁc state of the outcome (for example, an increase in PER or a low PER are only related to inclusions). The small sample size of this study has allowed a case-oriented interpretation of the results, based on the direct observation and knowledge of the particularities of each set of cases. 5.1. Theoretical Contributions Several decades of research dealing with the business case of sustainability have explored different causal relationships between CSP and CFP, considered multiple mediating and moderating factors and yielding heterogeneous conclusions. This study has approached the analysis of the relationship between CFP and CSP from an alternative perspective to most previous literature on this issue. Facing the inconclusive results of previous extensive research on the CFP-CSD-CSP links, and accounting for the signals of equiﬁnality, complexity and asymmetry revealed by some studies, we have attended the call for applying new research methodologies into this ﬁeld [18]. QCA is a particularly suitable analytical framework to deal with these features, and it has been used by some incipient literature to analyse inconsistent complex results of previous studies [17], and to contrast its results with traditional empirical methods [89]. Rather than exploring the direct link between CFP and CSP, this study applies fsQCA to clarify the complex and asymmetrical links between them. In contrast to previous literature, CFP is not considered a dependent variable but as an antecedent together with CSD, while CSP (considering inclusions in and exclusions from a SRI index as proxy) is considered an outcome to be explained. Little previous research has been done examining the preconditions or antecedents of additions to and deletions from SRI indices (Ziegler and Schröder [10] are an exception). The inﬂuence of CSD on CSP has also received little attention. Furthermore, CSP, CFP or CSD could be moderating or mediating factors of one another, as CFP and CSP can have an inﬂuence on CSD intensity and quality and CSD can have an inﬂuence on CFP and CSP. Although the fsQCA technique does not allow for the examination of mediating models, nor to include control variables [91], it provides valuable information regarding how causal conditions interact to produce a speciﬁc outcome. An important novelty and contribution of this study to the incipient application of the fsQCA in the CFP-CSP research ﬁeld is that the outcome is deﬁned in a dual manner. This approach enables measuring a high level (presence) and low level (absence) of the outcome (good or poor CSP) while analysing two separate but complementary constructs: Inclusions in a SRI index (proxy for a good CSR) and exclusions from a SRI index (proxy for a poor CSP). 4.2. Sufﬁciency Analysis Robustness tests are generally recommended for larger sample sizes [111], and there is no general agreement on how to perform them. They are frequently conducted through changes in the calibration of the variables. This study has tested two alternative model speciﬁcations looking for any difference between static and dynamic CFP measures in leading to CSP. The low coverages obtained are derived from the small-N dataset, and conﬁrm a highly heterogeneous sample and the difﬁculty of searching 23 of 31 23 of 31 Sustainability 2019, 11, 1211 for any straightforward relation between CFP and CSP. Dynamic measures of Model II. slightly improve the solution coverage, and show a speciﬁc set of cases clearly supporting Proposition 3b. Some of the variables included in both Model I. and Model II. have not needed any calibration at all, while calibrated ones (those related with both static and dynamic ﬁnancial performance, together with the number of sustainability reports and the time elapsed between an entry and exit event) cannot be reasonably recalibrated applying any sound alternative argument. As we acknowledge in the limitations subsection, it would be more interesting to explore alternative model speciﬁcations, considering other ﬁnancial performance measures or even other relevant dimensions of corporate performance, than testing alternative calibrations of continuous variables. This could be addressed in future research. 5. Conclusions, Implications and Limitations This study makes several theoretical and practical contributions with implications for both researchers and private and public decision-makers. In particular, we highlight three outstanding contributions in three underdeveloped research areas: The application of the fsQCA to the analysis of the CFP-CSD-CSP links, the analysis of the preconditions of inclusions in and exclusions from SRI indices, and the focus on the role of CSD as an antecedent of CSP. Limitations are also acknowledged. 5.2. Implications for Researchers This study has demonstrated that fsQCA is a valuable alternative and/or complementary methodological approach. FsQCA is particularly useful for regression analyses that consider being listed in a SRI index as proxy for CSP and for event studies that analyse the impact of inclusions in and exclusions from an SRI on CFP. The conﬁrmation of equiﬁnality, complexity and asymmetry for many of the causal conditions both individually and in combination, particularly for CFP measures, indicates that the raw application of statistical techniques to analyse the CFP-CSP links can be misleading. In particular, the effects of high and low CFP on both high and low CSP are complex and depend on other causal conditions. Therefore, we support a redirection of the CFP-CSP research towards case-oriented research exploring the circumstances under which CSR contributes to competitiveness [18,40]. Research should consider the inﬂuence of good or bad management [17], because this study has shown that the search of a univocal and unidirectional relationship between CFP and CSP is not justiﬁed. We encourage future research using fsQCA both in isolation as well as in conjunction with other methodologies to advance knowledge in the CFP-CSP research ﬁeld. Future research could also incorporate these study’s results to improve our understanding of the CSP-CSD-CFP interrelations. We have evidenced the necessity (though not sufﬁciency) of CSD for a company to be included in an SRI index as a consequence, presumably, of a better CSP. The inﬂuence of CSD on CSP remains under explored; in particular, their relationship in the context of SRI indices requires further analysis. Slager [84], for instance, uses a fsQCA to identify the antecedents of the efforts of companies included in the FTSE4Good index to improve their CSP in order to remain included, but does not consider CSD among those antecedents. New contributions could analyse other relevant antecedents of CSD, such as corporate governance or board structure indicators, which have been extensively related to CSD in prior research. Model I. in this study has included sales as proxy of company size and EPS and PER as measures of proﬁtability, revealing mixed interactions with CSD indicators to produce a better or poorer CSP. Alternative measures could be tested in future research. 5.1. Theoretical Contributions Another particular contribution of this study is considering inclusions as proxies for good CSP, and exclusions as proxies for poor CSP (similar to [4]), in contrast to previous literature, which has mainly focused on being listed or not as proxies for CSP. Furthermore, a different approach is used on the analysis of these exclusions and inclusions events by only focusing on those preceded by exclusions from or inclusions in the SRI index. This approach advances the proposal by Eom Sustainability 2019, 11, 1211 24 of 31 and Nam [68], who simply distinguish between the time of the index introduction and those of its subsequent revisions. The results reveal that no single CFP indicator explains the exclusion from or inclusion in an SRI index, and that high CSD (measured through four complementary indicators) is a necessary but not sufﬁcient antecedent of good CSP (inclusion in an SRI index) and low CSD is a necessary but not sufﬁcient condition for poor CSP (exclusions). We have found complexity in the relationship between CFP and CSP, as the impact of each CFP measure depends on other ﬁnancial or non-ﬁnancial causal conditions. The study also identiﬁes alternative paths or combinations of causal conditions to reach the same outcome. Some of these paths provide some support to the slack resources theory as they show that high CSP and CSD are related to an improvement in CFP and that low CSP and CSD are related to a decrease in CFP (fewer resources available to CSP and CSD). However, the analysis identiﬁes speciﬁc paths directly contradicting our expectations of improved CFP associated with good CSP and of a low CFP related to a poor CSP. Furthermore, the results provide some evidence of a stronger relation between CSP and accounting-based CFP measures than with market-based ones, in contrast to most literature ﬁndings and in line with Endrikat et al. [14]. 5.3. Implications for Private and Public Decision-Makers The empirical results clearly support the importance of CSD, because its absence impedes a high CSP and the subsequent inclusion in an SRI index. This result highlights the need to reinforce both the regulatory frameworks demanding greater transparency on non-ﬁnancial issues and the initiatives working on standardization to improve the quality of CSD. US policymakers should acknowledge the powerful incentives of being subjected to a higher scrutiny on non-ﬁnancial issues in order to improve CSP, as the European Union has already done. CSD should also expand among small 25 of 31 Sustainability 2019, 11, 1211 and medium-sized companies in countries with mandatory non-ﬁnancial reporting already in place, and among larger ones, particularly in the Southeast of the European Union [48,49]. Our results also have direct implications for companies willing to gain recognition for their CSP through the inclusion in an SRI index, suggesting the necessity of a higher commitment to CSD. CSD based on the publication of reports following the GRI or IR guidelines enables companies to be assessed under a globally accepted standard. Even those companies that have experienced a deterioration of their CSP could beneﬁt from a well-oriented and focused CSD to avoid the negative impacts on market value that this study’s results have partially supported. Furthermore, an increase in commitment to CSD could lead to an improvement of CSP due to the higher scrutiny and pressure of stakeholders y p y pp CSD could lead to an improvement of CSP due to the higher scrutiny and pressure of stakeho We have found some evidence of the long period required to solve CSP deﬁciencies in companies that had been excluded from the S&P 500 E&S index before returning to the index, as well as mixed or ambiguous associations with CFP measures. Therefore, business managers should also acknowledge the importance of a strategic integration and planning of CSR in order to achieve meaningful CSP improvements in short periods together with CFP improvements. Finally, CSR rating agencies and SRI indices managers should devote bigger efforts to improving and guaranteeing the quality and accuracy of their CSP rating methodologies. Similarly to Eom and Nam [68], we acknowledge that the absence of a clear and direct relationship between CFP and CSP can be a sign of the inability of SRI indices to properly capture companies’ CSP. 5.3. Implications for Private and Public Decision-Makers We have also pointed out the possibility of investors’ mixed perceptions about the ﬁnancial implications of good or poor CSP (Eom and Nam [68] talk about pessimistic and optimistic investors). This leads again to the need for an increase in CSD, so that the potential ﬁnancial beneﬁts derived from CSR are properly understood by ﬁnancial markets. CSD could also contribute to closing the gap between the supply and demand of SRI products. This also demands a high policy commitment to expand SRI and contribute to the integration of CSR matters into investment decisions. Integrated and joint actions are needed from governmental and non-governmental institutions, as well as from CSR rating agencies and companies themselves, to break the irrelevance of CSD in market valuations found in previous research. This is not an easy task because sustainability information is not easily veriﬁable regardless of the adoption of any of the current available reporting standards. Companies should design effective CSD frameworks, focused not only on reporting but also on permanent dialog and interaction with their stakeholders. 5.4. Research Limitations This study has some limitations. Regarding the fsQCA technique, as Cuadrado-Ballesteros et al. [85] note, each model specification in a fsQCA should include a limited number of factors because the higher the number of antecedents considered, the higher the number of possible paths leading to a same outcome, which reduces valid inference. Logically, there are more factors with a possible influence on CSP and on changes in SRI indices’ composition; similarly, CFP can be approached using alternative measures. Different model specifications could also be tested considering CFP or CSD as outcomes instead of antecedents. All these are open avenues for future research applying fsQCA alone or in combination with other techniques. Future research could replicate and further develop this analysis over other SRI indices and other stock markets different from the US market. An important limitation is the use of the GRI SD database to verify whether the sample companies had published a sustainability report, as Northern America shows lower GRI adoption levels [52,106]. Following Goettsche et al. [62], we also acknowledge that there is a remote possibility that some companies issuing sustainability reports (either GRI-aligned or not) may not appear in the GRI SD database. An alternative and more time-consuming and effort-demanding procedure, followed by Schreck and Raithel [65], would be to explore the companies’ websites or even directly contacting them to conﬁrm the availability of sustainability reports for the year previous to the inclusion or exclusion event. Finally, we focus again on the debate surrounding the validity of inclusions and exclusions from an SRI index as proxies for CSP [10]. This study supports these inclusion and exclusion events as accurate 26 of 31 Sustainability 2019, 11, 1211 26 of 31 measures of CSP. However, the rather obscure rating methodologies supporting SRI indices undermine their credibility and make it difﬁcult to discern whether inclusions in or exclusions from SRI indices are really directly derived from a good or poor CSP. Thus, future research could focus on analysing inclusions and exclusions as independent constructs from CSP in order to gain a deeper understanding on the effects of CSP on them, either using aggregated measures or, preferably, decomposing it into speciﬁc ESG issues. Being included in an SRI index could be considered as an antecedent of CSP. 5.4. Research Limitations In fact, Searcy and Elkhawas [11] explain the beneﬁts derived from being listed in a SRI index (e.g., using the index logo at their websites and in corporate reports) so that companies could ﬁnd incentives to try to improve their CSP in order to be included or remain in the index [84]. In summary, this study derives some relevant implications for researchers, business managers and policymakers. Research on the CFP-CSP link, with its long past trajectory and its present controversial state, still has a long future ahead. Author Contributions: Conceptualization and Data collection J.P.-C. and N.R.-C.; Methodology, Software and Formal Analysis, N.R.-C. and M.V.-G; All authors contributed to Writing—Original Draft Preparation and Writing—Review & Editing. All authors have read and approved the manuscript. Funding: This research received no external funding. Acknowledgments: The authors are grateful to the two anonymous reviewers for their feedback and helpful and valuable comments on a previous version of this article. Acknowledgments: The authors are grateful to the two anonymous reviewers for their feedback and helpful and valuable comments on a previous version of this article. Abbreviations CFP corporate ﬁnancial performance CSD corporate sustainability disclosure CSI corporate social irresponsibility CSP corporate sustainability performance CSR corporate social responsibility ESG environmental, social and governance FsQCA fuzzy-set qualitative comparative analysis GRI SD database GRI sustainability disclosure database GRI Global Reporting Initiative IR integrated reporting QCA qualitative comparative analysis ROA return on assets ROE return on equity S&P 500 ESG S&P 500 ESG Factor Weighted Index S&P 500 E&S S&P 500 Environmental and Socially Responsible Index SRI socially responsible investment 5. Hawn, O.; Chatterji, A.K.; Mitchell, W. Do Investors Actually Value Sustainability? New Evidence from Investor Reactions to the Dow Jones Sustainability Index (DJSI). Strateg. Manag. J. 2018, 39, 949–976. [CrossRef] Sustainability Reporting in the European Stock Markets. Bus. Ethics Eur. Rev. 2016, 25, 1–24. [CrossRef] 4. Doh, J.P.; Howton, S.D.; Howton, S.W.; Siegel, D.S. Does the Market Respond to an Endorsement of Social Responsibility? the Role of Institutions, Information, and Legitimacy. J. Manag. 2010, 36, 1461–1485. [CrossRef] References 1. Rezaee, Z. Business Sustainability Research: A Theoretical and Integrated Perspective. J. Account. Lit. 2016, 36, 48–64. [CrossRef] 2. Brooks, C.; Oikonomou, I. The Effects of Environmental, Social and Governance Disclosures and Performance on Firm Value: A Review of the Literature in Accounting and Finance. Brit. Account. Rev. 2018, 50, 1–15. [CrossRef] 3. Kaspereit, T.; Lopatta, K. The Value Relevance of SAM’s Corporate Sustainability Ranking and GRI Sustainability Reporting in the European Stock Markets. Bus. Ethics Eur. Rev. 2016, 25, 1–24. [CrossRef] 4. Doh, J.P.; Howton, S.D.; Howton, S.W.; Siegel, D.S. Does the Market Respond to an Endorsement of Social Responsibility? the Role of Institutions, Information, and Legitimacy. J. Manag. 2010, 36, 1461–1485. [CrossRef] 5. Hawn, O.; Chatterji, A.K.; Mitchell, W. Do Investors Actually Value Sustainability? New Evidence from Investor Reactions to the Dow Jones Sustainability Index (DJSI). Strateg. Manag. J. 2018, 39, 949–976. [CrossRef] 27 of 31 27 of 31 Sustainability 2019, 11, 1211 6. Kappou, K.; Oikonomou, I. Is there a Gold Social Seal? the Financial Effects of Additions to and Deletions from Social Stock Indices. J. Bus. Ethics 2016, 133, 533–552. [CrossRef] 7. Revelli, C.; Viviani, J. Financial Performance of Socially Responsible Investing (SRI): What have we Learned? A Meta-analysis. Bus. Ethics Eur. Rev. 2015, 24, 158–185. [CrossRef] alysis. Bus. Ethics Eur. Rev. 2015, 24, 158–185. [CrossR 8. Chatterji, A.; Levine, D. Breaking Down the Wall of Codes: Evaluating Non-Financial Performance Measurement. Calif. Manag. Rev. 2006, 48, 29–51. [CrossRef] 9. Pagano, M.S.; Sinclair, G.; Yang, T. Understanding ESG ratings and ESG indexes. In Research Handbook of Finance and Sustainability; Boubaker, S., Cumming, D., Nguyen, D.K., Eds.; Cheltenham, UK; pp. 339–371. 10. Ziegler, A.; Schröder, M. What Determines the Inclusion in a Sustainability Stock Index?: A Panel Data Analysis for European Firms. Ecol. Econ. 2010, 69, 848–856. [CrossRef] 11. Searcy, C.; Elkhawas, D. Corporate Sustainability Ratings: An Investigation into how Corporations use the Dow Jones Sustainability Index. J. Clean. Prod. 2012, 35, 79–92. [CrossRef] 12. Christensen, D.M. Corporate Accountability Reporting and High-Proﬁle Misconduct. Account. Rev. 2016, 91, 377–399. [CrossRef] 3. Topping, N. How does sustainability disclosure drive behavior change? J. Appl. Corp. Finan. 2012, 24, 45 [CrossRef] 14. Endrikat, J.; Guenther, E.; Hoppe, H. Making Sense of Conﬂicting Empirical Findings: A Meta-Analytic Review of the Relationship between Corporate Environmental and Financial Performance. Eur. Manag. J. 2014, 32, 735–751. [CrossRef] 15. Ruggiero, P.; Cupertino, S. References CSR Strategic Approach, Financial Resources and Corporate Social Performance: The Mediating Effect of Innovation. Sustainability 2018, 10, 3611. [CrossRef] 16. Roig-Tierno, N.; Gonzalez-Cruz, T.F.; Llopis-Martinez, J. An Overview of Qualitative Comparative Analysis: A Bibliometric Analysis. J. Innovat. Knowl. 2017, 2, 15–23. [CrossRef] 17. Isaksson, L.E.; Woodside, A.G. Modeling Firm Heterogeneity in Corporate Social Performance and Financial Performance. J. Bus. Res. 2016, 69, 3285–3314. [CrossRef] 18. Aguinis, H.; Glavas, A. What we Know and Don’t Know about Corporate Social Responsibility: A Review and Research Agenda. J. Manag. 2012, 38, 932–968. [CrossRef] 9. Busch, T.; Friede, G. The Robustness of the Corporate Social and Financial Performance Rela A Second-Order Meta-Analysis. Corp. Soc. Respons. Environ. Manag. 2018, 25, 583–608. [CrossRef] 20. Grewatsch, S.; Kleindienst, I. When does it Pay to be Good? Moderators and Mediators in the Corporate Sustainability–Corporate Financial Performance Relationship: A Critical Review. J. Bus. Ethics 2017, 145, 383–416. [CrossRef] 21. Wang, Q.; Dou, J.; Jia, S. A Meta-Analytic Review of Corporate Social Responsibility and Corporate Financial Performance: The Moderating Effect of Contextual Factors. Bus. Soc. 2016, 55, 1083–1121. [CrossRef] 22. Perrini, F.; Russo, A.; Tencati, A.; Vurro, C. Deconstructing the Relationship between Corporate Social and Financial Performance. J. Bus. Ethics 2011, 102, 59–76. [CrossRef] 23. Humphrey, J.E.; Lee, D.D.; Shen, Y. Does it Cost to be Sustainable? J. Corp. Financ. 2012, 18, 626–639. [CrossRef] 24. Hussain, N.; Rigoni, U.; Cavezzali, E. Does it Pay to be Sustainable? Looking Inside the Black Box of the Relationship between Sustainability Performance and Financial Performance. Corp. Soc. Respons. Environ. Manag. 2018, 25, 1198–1211. [CrossRef] g 25. McWilliams, A.; Siegel, D. Corporate Social Responsibility and Financial Performance: Correlation Or Misspeciﬁcation? Strateg. Manag. J. 2000, 21, 603–609. [CrossRef] 26. Friede, G.; Busch, T.; Bassen, A. ESG and Financial Performance: Aggregated Evidence from More than 2000 Empirical Studies. J. Sust. Financ. Invest. 2015, 5, 210–233. [CrossRef] 27. Brammer, S.; Millington, A. Does it Pay to be Different? an Analysis of the Relationship between Corporate Social and Financial Performance. Strateg. Manag. J. 2008, 29, 1325–1343. [CrossRef] 28. Lenz, I.; Wetzel, H.A.; Hammerschmidt, M. Can Doing Good Lead to Doing Poorly? Firm Value Implications of CSR in the Face of CSI. J. Acad. Mark. Sci. 2017, 45, 677–697. [CrossRef] 29. Price, J.M.; Sun, W. Doing Good and Doing Bad: The Impact of Corporate Social Responsibility and Irresponsibility on Firm Performance. J. Bus. Res. 2017, 80, 82–97. [CrossRef] 30. References Surroca, J.; Tribó, J.A.; Waddock, S. Corporate Responsibility and Financial Performance: The Role of Intangible Resources. Strateg. Manag. J. 2010, 31, 463–490. [CrossRef] 28 of 31 Sustainability 2019, 11, 1211 31. Artiach, T.; Lee, D.; Nelson, D.; Walker, J. The Determinants of Corporate Sustainability Performance. Account. Financ. 2010, 50, 31–51. [CrossRef] 32. Braun, B. What Makes Environmental Management Systems Successful? An Empirical Study in the German Manufacturing Sector. In The Integration of Social, Environmental and Economic Performance; Schaltegger, S., Wagner, M., Eds.; Greenleaf Publishing: Shefﬁeld, UK, 2006; pp. 589–614. 33. Elsayed, K.; Paton, D. The Impact of Financial Performance on Environmental Policy: Does Firm Life Cycle Matter? Bus. Strat. Environ. 2009, 18, 397–413. [CrossRef] 34. Scholtens, B. A Note on the Interaction between Corporate Social Responsibility and Financial Performance. Ecol. Econ. 2008, 68, 46–55. [CrossRef] 35. Testa, M.; D’Amato, A. Corporate Environmental Responsibility and Financial Performance: Does Bidirectional Causality Work? Empirical Evidence from the Manufacturing Industry. Soc. Res. J. 2017, 13, 221–234. [CrossRef] 36. Orlitzky, M.; Schmidt, F.L.; Rynes, S.L. Corporate Social and Financial Performance: A Meta-Analysis. Organ. Stud. 2003, 24, 403–441. [CrossRef] 37. Martínez Ferrero, J.; Frías Aceituno, J. Relationship between Sustainable Development and Financial Performance: International Empirical Research. Bus. Strat. Environ. 2015, 24, 20–39. [CrossRef] 38. Waddock, S.A.; Graves, S.B. The Corporate Social Performance–ﬁnancial Performance Link. Strateg. Manag. J. 1997, 18, 303–319. [CrossRef] 39. Salzmann, O.; Ionescu-somers, A.; Steger, U. The Business Case for Corporate Sustainability: Literature Review and Research Options. Eur. Manag. J. 2005, 23, 27–36. [CrossRef] 40. Schaltegger, S.; Synnestvedt, T. The Link between ‘green’ and Economic Success: Environmental Management as the Crucial Trigger between Environmental and Economic Performance. J. Environ. Mana. 2002, 65, 339–346. 41. Skouloudis, A.; Avlonitis, G.J.; Malesios, C.; Evangelinos, K. Priorities and Perceptions of Corporate Social Responsibility: Insights from the Perspective of Greek Business Professionals. Manag. Decis. 2015, 53, 375–401. [CrossRef] 42. Peloza, J. The Challenge of Measuring Financial Impacts from Investments in Corporate Social Performance. J. Manag. 2009, 35, 1518–1541. [CrossRef] 43. Galant, A.; Cadez, S. Corporate Social Responsibility and Financial Performance Relationship: A Re Measurement Approaches. Ec. Res.-Ekonomska Istraživanja 2017, 30, 676–693. [CrossRef] pp j 44. Dobre, E.; Stanila, O.G.; Brad, L. References [CrossRef] 29 of 31 29 of 31 Sustainability 2019, 11, 1211 56. Rezaee, Z.; Tuo, L. Are the Quantity and Quality of Sustainability Disclosures Associated with the Innate and Discretionary Earnings Quality? J. Bus. Ethics 2017, 1–24. [CrossRef] 57. Giannarakis, G. Corporate Governance and Financial Characteristic Effects on the Extent of Corporate Social Responsibility Disclosure. Soc. Respons. J. 2014, 10, 569–590. [CrossRef] 58. Crifo, P.; Forget, V.D.; Teyssier, S. The Price of Environmental, Social and Governance Practice Disclosure: An Experiment with Professional Private Equity Investors. J. Corp. Financ. 2015, 30, 168–194. [CrossRef] 59. Fifka, M. Corporate Responsibility Reporting and its Determinants in Comparative Perspective - a Review of the Empirical Literature and a Meta-Analysis. Bus. Strat. Environ. 2013, 22, 1–35. [CrossRef] 60. Stanwick, S.D.; Stanwick, P.A. The Relationship between Environmental Disclosures and Financial Performance: An Empirical Study of US Firms. Eco-Manag. Aud. 2000, 7, 155–164. [CrossRef] 61. Qiu, Y.; Shaukat, A.; Tharyan, R. Environmental and Social Disclosures: Link with Corporate Financial Performance. Brit. Account. Rev. 2016, 48, 102–116. [CrossRef] 62. Goettsche, M.; Steindl, T.; Gietl, S. Do Customers Affect the Value Relevance of Sustainability Reporting? Empirical Evidence on Stakeholder Interdependence. Bus. Strat. Environ. 2016, 25, 149–164. [CrossRef] 63. Martínez Ferrero, J.; Garcia Sanchez, I.; Cuadrado Ballesteros, B. Effect of Financial Reporting Qu Sustainability Information Disclosure. Corp. Soc. Respons. Environ. Manag. 2015, 22, 45–64. [CrossRe 63. Martínez Ferrero, J.; Garcia Sanchez, I.; Cuadrado Ballesteros, B. Effect of Financial Reporting Quality on Sustainability Information Disclosure. Corp. Soc. Respons. Environ. Manag. 2015, 22, 45–64. [CrossRef] 64. Melo, T. Slack-resources Hypothesis: A Critical Analysis Under a Multidimensional Approach to Corporate S i l P f S R J 8 257 269 [C R f] 64. Melo, T. Slack-resources Hypothesis: A Critical Analysis Under a Multidimensional Approach to Co Social Performance. Soc. Respons. J. 2012, 8, 257–269. [CrossRef] 65. Schreck, P.; Raithel, S. Corporate Social Performance, Firm Size, and Organizational Visibility: Distinct and Joint Effects on Voluntary Sustainability Reporting. Bus. Soc. 2018, 57, 742–778. [CrossRef] 66. Clarkson, P.M.; Li, Y.; Richardson, G.D.; Vasvari, F.P. Revisiting the Relation between Environmental Performance and Environmental Disclosure: An Empirical Analysis. Account. Organ. Soc. 2008, 33, 303–327. [CrossRef] 67. Hummel, K.; Schlick, C. The Relationship between Sustainability Performance and Sustainability Disclosure – Reconciling Voluntary Disclosure Theory and Legitimacy Theory. J. Account. Public Policy 2016, 35, 455–476. [CrossRef] 68. Eom, K.; Nam, G. References The Inﬂuence of Environmental and Social Performance on Financial P f E id f R i ’ Li t d E titi S t i bilit 2015 7 2513 2553 [C R f] bre, E.; Stanila, O.G.; Brad, L. The Inﬂuence of Environmental and Social Performance on Financial 44. Dobre, E.; Stanila, O.G.; Brad, L. The Inﬂuence of Environmental and Social Performance on Fi Performance: Evidence from Romania’s Listed Entities. Sustainability 2015, 7, 2513–2553. [CrossRef Dobre, E.; Stanila, O.G.; Brad, L. The Inﬂuence of Environmental and Social Performance on Financia Performance: Evidence from Romania’s Listed Entities. Sustainability 2015, 7, 2513–2553. [CrossRef] 45. Nollet, J.; Filis, G.; Mitrokostas, E. Corporate Social Responsibility and Financial Performance: A Non-Linear and Disaggregated Approach. Econ. Model. 2016, 52, 400–407. [CrossRef] 46. Statman, M.; Glushkov, D. The Wages of Social Responsibility. Financ. Anal. J. 2009, 65, 33–46. [CrossRef] 47. Fortanier, F.; Kolk, A.; Pinkse, J. Harmonization in CSR Reporting. Manag. Int. Rev. 2011, 51, 665. [CrossRef] 48. Evangelinos, K.; Skouloudis, A. European Perspectives on Corporate Non-Financial Disclosure: Evidence from the Southeast. Int. J. Dis. Gover. 2014, 11, 33–53. [CrossRef] Fortanier, F.; Kolk, A.; Pinkse, J. Harmonization in CSR Reporting. Manag. Int. Rev. 2011, 51, 665. [CrossRef 48. Evangelinos, K.; Skouloudis, A. European Perspectives on Corporate Non-Financial Disclosure: Evidence from the Southeast. Int. J. Dis. Gover. 2014, 11, 33–53. [CrossRef] 49. Skouloudis, A.; Evangelinos, K.; Kourmousis, F. Assessing Non-Financial Reports According to the Global Reporting Initiative Guidelines: Evidence from Greece. J. Clean. Prod. 2010, 18, 426–438. [CrossRef] 50. Benlemlih, M. Corporate Social Responsibility and Firm Financing Decisions: A Literature Review. J. Multinat. Financ. Manag. 2017, 42, 1–10. [CrossRef] 51. Fatemi, A.; Glaum, M.; Kaiser, S. ESG Performance and Firm Value: The Moderating Role of Disclosure. Glob. Financ. J. 2018, 38, 45–64. [CrossRef] 52. Marimon, F.; Alonso-Almeida, M.d.M.; Rodríguez, M.d.P.; Cortez Alejandro, K.A. The Worldwide Diffusion of the Global Reporting Initiative: What is the Point? J. Clean. Prod. 2012, 33, 132–144. [CrossRef] 53. Churet, C.; Eccled, R.G. Integrated Reporting, Quality of Management, and Financial Performance. J. Appl. Corp. Financ. 2014, 26, 56–64. 54. Beck, C.; Frost, G.; Jones, S. CSR Disclosure and Financial Performance Revisited: A Cross-Country Analysis. Aust. J. Manag. 2018, 43, 517–537. [CrossRef] 55. Chen, L.; Feldmann, A.; Tang, O. The Relationship between Disclosures of Corporate Social Performance and Financial Performance: Evidences from GRI Reports in Manufacturing Industry. Int. J. Prod. Econ. 2015, 170, 445–456. References Effect of Entry into Socially Responsible Investment Index on Cost of Equity and Firm Value. Sustainability 2017, 9, 717. [CrossRef] 69. Schröder, M. Is there a Difference? the Performance Characteristics of SRI Equity Indices. J. Bus. Financ. Account. 2007, 34, 331–348. [CrossRef] 70. Statman, M. Socially Responsible Indexes. J. Portfolio Manag. 2006, 32, 100–109. [CrossRef] 71. Consolandi, C.; Jaiswal-Dale, A.; Poggiani, E.; Vercelli, A. Global Standards and Ethical Stock Indexes: The Case of the Dow Jones Sustainability Stoxx Index. J. Bus. Ethics 2009, 87, 185–197. [CrossRef] 72. Oberndorfer, U.; Schmidt, P.; Wagner, M.; Ziegler, A. Does the Stock Market Value the Inclusion in a Sustainability Stock Index? an Event Study Analysis for German Firms. J. Environ. Econ. Manag. 2013, 66, 497–509. [CrossRef] 73. Russo, A.; Mariani, M. Drawbacks of a Delisting from a Sustainability Index: An Empirical Analysis. Int. J. Bus. Admin. 2013, 4, 29–40. [CrossRef] 74. Cheung, A.W.K. Do Stock Investors Value Corporate Sustainability? Evidence from an Event Study. J. Bus. Ethics 2011, 99, 145–165. [CrossRef] 75. Curran, M.M.; Moran, D. Impact of the FTSE4Good Index on Firm Price: An Event Study. J. Environ. 2007, 82, 529–537. [CrossRef] [PubMed] 76. Ragin, C.C. The Comparative Method; Moving Beyond Qualitative and Quantitative Strategies; University of California Press: Berkeley and Los Angeles, CA, USA, 1987. 77. Ragin, C.C. Fuzzy-Set Social Science; University of Chicago Press: Chicago, IL, USA, 2000. 78. Ragin, C.C. Set Relations in Social Research: Evaluating their Consistency and Coverage. Polit. Anal. 2006, 14, 291–310. [CrossRef] 79. Ragin, C.C. Redesigning Social Inquiry: Fuzzy Sets and Beyond; University of Chicago Press: Chicago, IL, USA, 2008. 80. Fiss, P.C. A Set-Theoretic Approach to Organizational Conﬁgurations. Acad. Manag. Rev. 2007, 32, 1180–1198. [CrossRef] 81. Fiss, P.C. Building Better Causal Theories: A Fuzzy Set Approach to Typologies in Organization Research. Acad. Manag. J. 2011, 54, 393–420. [CrossRef] 30 of 31 30 of 31 Sustainability 2019, 11, 1211 82. Rihoux, B.; Ragin, C.C. Conﬁgurational Comparative Methods: Qualitative Comparative Analysis (QCA) and Related Techniques; Sage Publications: Thousand Oaks CA, USA, 2009. 83. Maggetti, M. Promoting Corporate Responsibility in Private Banking: Necessary and Sufﬁcient Conditions for Joining the Wolfsberg Initiative Against Money Laundering. Bus. Soc. 2014, 53, 787–819. [CrossRef] 84. Slager, C.H. SRI indices and responsible corporate behaviour: a study of the FTSE4GOOD index. PhD Thesis, University of Nottingham, Nottingham, UK, 2012. 85. Cuadrado-Ballesteros, B.; Martínez-Ferrero, J.; García-Sánchez, I.M. References Board Structure to Enhance Social Responsibility Development: A Qualitative Comparative Analysis of US Companies. Corp. Soc. Respons. Environ. Manag. 2017, 24, 524–542. [CrossRef] 86. Misangyi, V.F.; Greckhamer, T.; Furnari, S.; Fiss, P.C.; Crilly, D.; Aguilera, R. Embracing Causal Complexity: The Emergence of a Neo-Conﬁgurational Perspective. J. Manag. 2017, 43, 255–282. [CrossRef] 87. Woodside, A.G. Moving Beyond Multiple Regression Analysis to Algorithms: Calling for Adoption of a Paradigm Shift from Symmetric to Asymmetric Thinking in Data Analysis and Crafting Theory. J. Bus. Res. 2013, 66, 463–472. [CrossRef] 88. Woodside, A.G. Embrace•perform•model: Complexity Theory, Contrarian Case Analysis, and Multiple Realities. J. Bus. Res. 2014, 67, 2495–2503. [CrossRef] 89. Kan, A.K.S.; Adegbite, E.; El Omari, S.; Abdellatif, M. On the use of Qualitative Comparative Ana Management. J. Bus. Res. 2016, 69, 1458–1463. 90. Gjølberg, M. The Origin of Corporate Social Responsibility: Global Forces Or National Legacies? Socio-Econ. Rev. 2009, 7, 605–637. [CrossRef] 91. Rothenhoefer, L.M. The Impact of CSR on Corporate Reputation Perceptions of the Public-A Conﬁgurational Multi-Time, Multi-Source Perspective. Bus. Ethics Eur. Rev. 2018. [CrossRef] 92. Skarmeas, D.; Leonidou, C.N.; Saridakis, C. Examining the Role of CSR Skepticism using Fuzzy-Set Qualitative Comparative Analysis. J. Bus. Res. 2014, 67, 1796–1805. [CrossRef] 93. Halme, M.; Rintamäki, J.; Knudsen, J.S.; Lankoski, L.; Kuisma, M. When is there a Sustainability Case for CSR? Pathways to Environmental and Social Performance Improvements. Bus. Soc. 2018. [CrossRef] 94. Wang, D.H.; Yu, T.H.; Chiang, C. Exploring the Value Relevance of Corporate Reputation: A Fuzzy-Set Qualitative Comparative Analysis. J. Bus. Res. 2016, 69, 1329–1332. [CrossRef] 95. Paniagua, J.; Rivelles, R.; Sapena, J. Corporate Governance and Financial Performance: The Role of Ownership and Board Structure. J. Bus. Res. 2018, 89, 229–234. [CrossRef] 96. Primc, K.; ˇCater, T. Environmental Proactivity and Firm Performance: A Fuzzy-Set Analysis. Manag. Decis. 2015, 53, 648–667. [CrossRef] 97. Demirguc-Kunt, A. Stock Markets, Corporate Finance, and Economic Growth. World Bank Econ. Rev. 1996, 10, 223–239. [CrossRef] 98. Lucey, B.M.; Muckley, C. Robust Global Stock Market Interdependencies. Int. Rev. Financ. Anal. 2011, 20, 215–224. [CrossRef] 99. Stolowy, H.; Paugam, L. The Expansion of Non-Financial Reporting: An Exploratory Study. Account. Bus. Res. 2018, 48, 525–548. [CrossRef] 100. Williams, C.A.; Aguilera, R.V. Corporate Social Responsibility in a Comparative Perspective. In The Oxford Handbook of Corporate Social Responsibility; Crane, A., Matten, D., McWilliams, A., Moon, J., Siegel, D.S., Eds.; Oxford University Press: Oxford, UK, 2008; pp. 452–472. 101. Matten, D.; Moon, J. References “Implicit” and “Explicit” CSR: A Conceptual Framework for a Comparative Understanding of Corporate Social Responsibility. Acad. Manag. Rev. 2008, 33, 404–424. [CrossRef] 102. S&P Dow Jones Indices. Equity S&P 500, 2018. Available online: https://us.spindices.com/indices/equity/ sp-500 (accessed on 20 December 2018). 103. Government & Accountability Institute. FLASH REPORT: 85% of S&P 500 Index®Companies Publish Sustainability Reports in 2017. 2018. Available online: https://www.ga-institute.com/press-releases/article/ ﬂash-report-85-of-sp-500-indexR-companies-publish-sustainability-reports-in-2017.html (accessed on 20 December 2018). 104. Tamimi, N.; Sebastianelli, R. Transparency among S&P 500 Companies: An Analysis of ESG Disclosure Scores. Manag. Decis. 2017, 55, 1660–1680. 105. López-Cabarcos, Á.M.; Pérez-Pico, M.A.; López-Pérez, L.M. Does Social Network Sentiment Inﬂuence S&P 500 Environmental & Socially Responsible Index? Sustainability 2019, 11, 320. 31 of 31 Sustainability 2019, 11, 1211 31 of 31 106. González, M.; Alonso-Almeida, M.; Dominguez, D. Mapping Global Sustainability Report Scoring: A Detailed Analysis of Europe and Asia. Qual. Quant. 2018, 52, 1041–1055. [CrossRef] 107. Pineiro-Chousa, J.; Vizcaíno-González, M.; Romero-Castro, N. Recent advances in standardizing the reporting of nonﬁnancial information. In Organizational Change and Global Standardization: Solutions to Standards and Norms Overwhelming Organizations; Boje, D.M., Ed.; Routledge: New York, NY, USA, 2015; pp. 169–185. 108. Fernandez-Feijoo, B.; Romero, S.; Ruiz, S. Commitment to Corporate Social Responsibility Measured through Global Reporting Initiative Reporting: Factors Affecting the Behavior of Companies. J. Clean. Prod. 2014, 81, 244–254. [CrossRef] 109. Fuzzy-Set/Qualitative Comparative Analysis, version 3. WINDOWS/MAC Program for Crisp and Fuzzy Sets; Department of Sociology, University of California: Irivine, CA, USA, 2017. 110. Schneider, C.Q.; Wagemann, C. Set-Theoretic Methods for the Social Sciences: A Guide to Qualitative Comparative Analysis; Cambridge University Press: Cambridge, UK, 2012. 111. Emmenegger, P.; Schraff, D.; Walter, A. QCA, the Truth Table Analysis and Large-N Survey Data: The Beneﬁts of Calibration and the Importance of Robustness Tests. In Proceedings of the 2nd International QCA Expert Workshop, Zurich, Switzerland, 5–7 November 2014. © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
https://openalex.org/W945052942	https://zenodo.org/records/6709950/files/3.%20Retailing%20strategy%20of%20products%20and%20customer%20services%20in%20organized%20retail%20sector.pdf	English	null	Retailing strategy of products and customer services in organized retail sector	Zenodo (CERN European Organization for Nuclear Research)	2,016	cc-by	137,708	Year 2015 i DECLARATION BY THE CANDIDATE I, Atul Kumar, declare that thesis entitled “RETAILING STRATEGY OF PRODUCTS & CUSTOMER SERVICES IN ORGANISED RETAIL SECTOR” is my own work conducted under the supervision of Dr. V. K. Soni and co- supervision of Dr. Pradeep Kumar Sinha of Department of Management, Shri Jagdishprasad Jhabarmal Tibrewala University, Jhunjhunu, Rajasthan, India approved by the research Degree Committee. I have put in more than 200 days of attendance with the supervisor/co-supervisor at the centre. I further declare that to the best of my knowledge the thesis does not contain any part of any work which has been submitted for the award of any degree either in this University or any other University/deemed University without proper citation. ii ii CERTIFICATE OF THE SUPERVISOR This is to certify that work entitled “RETAILING STRATEGY OF PRODUCTS & CUSTOMER SERVICES IN ORGANISED RETAIL SECTOR” This is to certify that work entitled “RETAILING STRATEGY OF PRODUCTS & CUSTOMER SERVICES IN ORGANISED RETAIL SECTOR” is a piece of research work done by Mr. Atul Kumar under my supervision for the degree of Doctor of Philosophy in Management of Shri Jagdishprasad Jhabarmal Tibrewala University, Jhunjhunu, Rajasthan, India. That the candidate has put attendance of more than 200 days with me. is a piece of research work done by Mr. Atul Kumar under my supervision for the degree of Doctor of Philosophy in Management of Shri Jagdishprasad Jhabarmal Tibrewala University, Jhunjhunu, Rajasthan, India. That the candidate has put attendance of more than 200 days with me. To the best of my knowledge and belief the thesis To the best of my knowledge and belief the thesis 1. Embodies the work of the candidate himself. 2. Has duly been completed. 3. Fulfils the requirement of ordinances related to the Ph.D. degree of the University and 3. Fulfils the requirement of ordinances related to the Ph.D. degree of the University and 4. Is up to the standard both in respect of the content and language, for being referred to the examiner. 4. Is up to the standard both in respect of the content and language, for being referred to the examiner. iii ACKNOWLEDGEMENT I acknowledge my gratitude with a sense of reverence to the Almighty God and those who have contributed and spared time for the completion of this thesis. Their valuable guidance and wise direction have enabled me to complete my thesis in a systematic way. At the outset, my ethical accountability is to be extremely indebted to Dr. Vinod Tibrewala, Chancellor, Shri Jagdishprasad Jhabarmal Tibrewala University, Jhunjhunu, Rajasthan, India and Dr. D. K. Kaushik, Vice Chancellor, Shri Jagdishprasad Jhabarmal Tibrewala University, Jhunjhunu, Rajasthan, India for his gracious permission to carry out this work. In an ecstasy of delight, I express my profound sense of gratitude to my esteemed guide and supervisor, Dr. V. K. Soni of Department of Management of Shri Jagdishprasad Jhabarmal Tibrewala University, Jhunjhunu, Rajasthan, India for his scholarly, inspiring supervision, incessant painstaking efforts and stimulating guidance patronage. He persevered beyond my expectations. Without his useful advice and valuable guidance, the completion of this thesis would not be possible. I would like to acknowledge the significant contributions to this work made by my esteemed co-guide and co-supervisor, Dr. Pradeep Kumar Sinha of Department of Management of Shri Jagdishprasad Jhabarmal Tibrewala University, Jhunjhunu, Rajasthan, India. He played equally important and complementary role in shaping the work reported in this thesis. I am thankful to Dr. Vinaydeep Brar, Associate Professor, Balaji Institute of Modern Management, Pune for her useful suggestions and cooperation in giving final shape to this work. I am also thankful to Dr. Lakshaydeep, Assistant Professor, Mahatma Jyoti Rao Phoole College of Agriculture and Research (M.J.R.P. University), Jaipur for her useful suggestions in giving final shape to this work. iv I would like to acknowledge Dr. B. N. Gupta, Ex-Professor, Siddhant Institute of Business Management, Pune, (Maharashtra) for his encouraging support and valuable suggestions on various occasions. I also extend my thanks to Dr. V. O. Varkey, Ex Director, Siddhant Group of Institutions, Pune and Dr. S. S. Khot, Principal, Siddhant College of Engineering, Pune for their help and support. I am also thankful to Dr. D. K. Tripathi, Dr. B. D. Vyas, Dr. Pardeep Kumar and Dr. Vinod Meel of Shri Jagdishprasad Jhabarmal Tibrewala University, Jhunjhunu, Rajasthan, India for their kind cooperation on various occasions. I am also thankful to Dr. D. K. Tripathi, Dr. B. D. Vyas, Dr. Pardeep Kumar and Dr. ACKNOWLEDGEMENT Vinod Meel of Shri Jagdishprasad Jhabarmal Tibrewala University, Jhunjhunu, Rajasthan, India for their kind cooperation on various occasions. I also acknowledge my parents who provided moral support to make it possible for me to pursue this goal. It is also my deep duty to express my gratefulness to each and every person of different retail stores, who extended me their help to make this thesis work a successful venture. The incredible help offered by the rest of the persons whose names could not be mentioned is incomparable and deserve the highest degree of appreciation. (Atul Kumar) Research Scholar v v TABLE OF CONTENTS TABLE OF CONTENTS Sr. No. Particulars Page No. TABLE OF CONTENTS 1 Title Page i 2 Declaration by the candidate ii 3 Certificate of the supervisors iii 4 Acknowledgement iv-v 5 Table of contents vi-viii 6 List of tables ix-xiii 7 List of figures xiv-xviii 8 Abbreviations xix-xx 9 Abstract xxi-xxii Section - I Descriptive analysis 10 CHAPTER 1 INTRODUCTION  An overview of Indian retail industry  An evolution of retail in India  Reforms in Indian retail  Investments in the retail sector  Government initiatives for growth of the sector  Key drivers to the exponential growth of the Indian retail industry  Key challenges in Indian retail industry  SWOT analysis of organised retail industry in India  The classification of retail industry  Retailing formats in India  Major retailers in India 1-41 01 06 08 09 11 14 20 24 26 27 33 11 CHAPTER 2 REVIEW OF LITERATURE  Retailing strategies  Customer behaviour  Customer service  Customer satisfaction 42-64 42 49 54 59 12 CHAPTER 3 CONCEPTUAL FRAMEWORK  Retailing  Retailing strategy  Segmentation strategy  Targeting strategies  Positioning strategy  Retail marketing strategy  Defensive strategy  SWOT analysis  PEST analysis 65-85 65 65 66 66 68 68 69 70 vi  Five force analysis  ETOP analysis  BCG matrix  Generic strategy  Retail marketing mix  Brand  Merchandise planning  Retail assortment strategy  Stock planning  Objectives of pricing  Price-adjustment strategies  Customer service  Customer service strategies  Store design elements  Visual merchandising  Consumer behaviour  Customer satisfaction 71 71 72 73 73 74 75 76 77 78 78 79 79 81 81 83 84 85 13 CHAPTER 4 RESEARCH METHODOLOGY  Objectives of the study  Significance of the study  Delimitation of the study  Research hypotheses  Research design  Locale of the study  Universe of the study  Sampling design and selection of samples  Collection of data  Tools for data collection  Pilot study  Scales used for accumulating the responses  The validity of scales  The reliability of scales & internal consistency of reliability  The method used for collecting the data  Methods of reporting  Statistical analysis of data 86-96 86 87 87 88 88 89 89 90 91 91 92 93 93 94 94 95 95 Section – II Data analysis, results & discussion 14 CHAPTER 5 MERCHANDISES (PRODUCTS) & CUSTOMER SERVICES IN ORGANISED RETAIL INDUSTRY  Merchandises offered by organised retail industry  Customer services introduced by organised retail industry 97-154 98 135 15 CHAPTER 6 RETAILING STRATEGIES INTRODUCED BY ORGANISED RETAIL INDUSTRY 155-213 vii Segmentation strategy Targeting strategy Positioning strategy Retail marketing strategy Defensive strategy Strategic tools Generic (business-level) strategy Retail marketing mix Merchandise planning Retail assortment strategy Stock planning Pricing strategy Price-adjustment strategy Promotional strategy Customer service strategy Customer relationship management Store design 156 157 158 159 160 161 162 164 167 169 177 178 179 180 197 205 210 16 CHAPTER 7 CHANGING BUYING BEHAVIOUR OF CUSTOMERS IN THE ORGANISED RETAIL INDUSTRY  Analysis of profile of customers in the organised retail industry  Analysis of buying behaviour of customers in the organised retail industry  Factors influencing buying behaviour of customers in the organised retail industry 214-269 215 225 257 17 CHAPTER 8 CUSTOMER SATISFACTION WITH MERCHANDISE (PRODUCTS) AND SERVICES OFFERED BY ORGANISED RETAIL INDUSTRY  Testing of hypotheses 270-358 331 18 CHAPTER 9 CONCLUSION AND RECOMMENDATIONS  Introduction  Conclusions  Recommendations  Areas for further research 359-394 359 361 385 394 Section – III Supportive information 19 BIBLIOGRAPHY 395-409 20 APPENDIX  Institutional questionnaire  Individual questionnaire  Map of Maharashtra state  Additional information 410-444 410 431 442 443 Segmentation strategy Targeting strategy Positioning strategy Retail marketing strategy Defensive strategy Strategic tools Generic (business-level) strategy Retail marketing mix Merchandise planning Retail assortment strategy Stock planning Pricing strategy Price-adjustment strategy Promotional strategy Customer service strategy Customer relationship management Store design 156 157 158 159 160 161 162 164 167 169 177 178 179 180 197 205 210 16 CHAPTER 7 CHANGING BUYING BEHAVIOUR OF CUSTOMERS IN THE ORGANISED RETAIL INDUSTRY  Analysis of profile of customers in the organised retail industry  Analysis of buying behaviour of customers in the organised retail industry  Factors influencing buying behaviour of customers in the organised retail industry 214-269 215 225 257 17 CHAPTER 8 CUSTOMER SATISFACTION WITH MERCHANDISE (PRODUCTS) AND SERVICES OFFERED BY ORGANISED RETAIL INDUSTRY  Testing of hypotheses 270-358 331 18 CHAPTER 9 CONCLUSION AND RECOMMENDATIONS  Introduction  Conclusions  Recommendations  Areas for further research 359-394 359 361 385 394 Section – III Supportive information 19 BIBLIOGRAPHY 395-409 20 APPENDIX  Institutional questionnaire  Individual questionnaire  Map of Maharashtra state  Additional information 410-444 410 431 442 443 viii LIST OF TABLES LIST OF TABLES LIST OF TABLES Sr. No. Table No. Description Page No. LIST OF TABLES 1 1.1 India in comparison to the world in retailing 4 2 1.2 Percentage of population below the age of 25 years 15 3 1.3 A comparison of rent as %age of sales 21 4 1.4 A comparison of key characteristics of organised and unorganised retails in India 26 5 4.1 Sample size and selection of samples 90 6 6.1 Retail marketing mix (RMM) on importance scale 164 7 6.2 Important factors while planning merchandise 167 8 6.3 Important factors while planning merchandise quality 168 9 6.4 Important characteristics while selecting a vendor 175 10 6.5 Impact of sales promotion 189 11 6.6 Important factors while selecting a retail store location 195 12 6.7 Components of store design 211 13 7.1.1 Residing area of customers 215 14 7.1.2 Gender of customers 216 15 7.1.3 Age of customers 217 16 7.1.4 Education of customers 218 17 7.1.5 Occupation of customers 219 18 7.1.6 Family income (yearly) of customers 220 19 7.1.7 Marital status of customers 221 20 7.1.8 Nature of family of customers 222 21 7.1.9 Family size of customers 223 22 7.1.10 Religion of customers 224 23 7.2.1 Purchase decision maker in the family of customers 225 24 7.2.1 (a) Collection of information, while purchase decision making by customers 226 25 7.2.2 Sources of information, while purchase decision making by customers 227 26 7.2.3 Sources of information from Advertising, while purchase decision making by customers 228 27 7.2.4 Sources of influence for a customer, while purchase decision-making 229 28 7.2.5 Purchaser (buyer) in the family of the 230 ix customers 29 7.2.6 Duration of visiting the store by customers 231 30 7.2.7 Visits of customers to the store in a month 232 31 7.2.8 Preferred days of customers for visiting the stores 233 32 7.2.9 Preferred time for visiting the stores 234 33 7.2.10 Spending amount on purchasing from the stores 234 34 7.2.11 The categories of the products, from which customers purchase products 235 35 7.2.12 Impulse buyers 242 36 7.2.13 Reasons of impulse purchase by customers 248 37 7.2.14 The first preference of the customers while purchasing 249 38 7.2.15 Reasons for visiting a particular store by customers 252 39 7.2.16 Customers of organised retailers, visiting other stores 253 40 7.2.17 Type of stores visited by Customers of organised retailers other than their stores 254 41 7.2.18 Reasons for visiting other stores by customers 255 42 7.3.1 The influence of Sociological / Environmental / External factors on buying behaviour of customers 258 43 7.3.2 The influence of Psychological / Individual / Internal factors on buying behaviour of customers 260 44 7.3.3 The influence of Store related factors on buying behaviour of customers 262 45 8.1 Comparative statistics of customer satisfaction with merchandise assortment 272 46 8.2 Comparative statistics of customer satisfaction with merchandise categories in the assortment 274 47 8.3 Comparative statistics of customer satisfaction with the variety of merchandise in each category 277 48 8.4 Comparative statistics of customer satisfaction with the availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise 279 49 8.5 Comparative statistics of customer satisfaction with the brands in each merchandise category 282 50 8.6 Comparative statistics of customer satisfaction with quality assurance and quality of merchandise 284 x 51 8.7 Comparative statistics of customer satisfaction with the services 287 52 8.8 Comparative statistics of customer satisfaction with the quality of the services 289 53 8.9 Comparative statistics of customer satisfaction with the pricing of merchandise & services 291 54 8.10 Comparative statistics of customer satisfaction with the overall value Vs price of merchandise & services 294 55 8.11 Comparative statistics of customer satisfaction with the exterior atmospherics of the store 296 56 8.12 Comparative statistics of customer satisfaction with the interior atmospherics of the store 299 57 8.13 Comparative statistics of customer satisfaction with the store layout 301 58 8.14 Comparative statistics of customer satisfaction with the visual merchandising 304 59 8.15 Comparative statistics of customer satisfaction with the methods of promotion and communication 306 60 8.16 Comparative statistics of customer satisfaction with the advertising methods 308 61 8.17 Comparative statistics of customer satisfaction with the sales promotion methods 311 62 8.18 Comparative statistics of customer satisfaction with the direct retailing methods 313 63 8.19 Comparative statistics of customer satisfaction with the personal retailing methods 315 64 8.20 Comparative statistics of customer satisfaction with the public relation/publicity methods 318 65 8.21 Comparative statistics of customer satisfaction with the internet/interactive retailing methods 320 66 8.22 Comparative statistics of customer satisfaction with the sales personnel 322 67 8.23 Comparative statistics of customer satisfaction with the contact personnel 325 68 8.24 Comparative statistics of customer satisfaction with the factors pertained to personnel 327 69 8.25 Comparative statistics of customer satisfaction with the other tangible evidence 330 70 8.26 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with merchandise assortment 333 71 8.27 Statistics of Kolmogorov-Smirnov one-sample 334 xi test and one sample t-test for customer satisfaction with merchandise categories in the assortment 72 8.28 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the variety of merchandise in each category 335 73 8.29 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise 336 74 8.30 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the brands in each merchandise category 337 75 8.31 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with quality assurance and quality of merchandise 338 76 8.32 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the services 339 77 8.33 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the quality of the services 340 78 8.34 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the pricing of merchandise & services 341 79 8.35 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the overall value Vs price of merchandise & services 342 80 8.36 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the exterior atmospherics of the store 343 81 8.37 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the interior atmospherics of the store 344 82 8.38 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the store layout 345 xii xii 83 8.39 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the visual merchandising 346 84 8.40 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the methods of promotion and communication 347 85 8.41 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the advertising methods 348 86 8.42 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the sales promotion methods 349 87 8.43 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the direct retailing methods 350 88 8.44 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the personal retailing methods 351 89 8.45 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the public relation/publicity methods 352 90 8.46 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the internet/interactive retailing methods 353 91 8.47 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the sales personnel 354 92 8.48 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the contact personnel 355 93 8.49 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the factors pertained to personnel 356 94 8.50 Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the other tangible evidence 357 xiii LIST OF FIGURES LIST OF FIGURES Sr. No. Figure No. Description Page No. LIST OF FIGURES 1 1.1 Global Retail Development Index 2014 2 2 1.2 Global Retail Development Index 2014: Market attractiveness 3 3 1.3 Significant scope for expansion in organised retail 4 4 1.4 India’s retail market 5 5 1.5 An evolution of retail in India over the years 6 6 1.6 Evolution of retailing in India 7 7 1.7 Merger & acquisition in the Indian retail sector in 2013 9 8 1.8 Strong growth potential attracting high foreign investment 10 9 1.9 FDI Confidence Index 2014 12 10 1.10 Benefits of FDI and its limit in the different segment 13 11 1.11 Business models for entry in Indian retail market prior to September 20, 2012 14 12 1.12 Rising per capita income in India 17 13 1.13 Mall space breakup in India 17 14 1.14 Favourable FDI policy encouraging investment 18 15 1.15 Organised retail in India 19 16 1.16 Growth value proposition in organised retail industry 20 17 1.17 Porter’s Five Forces analysis of the organised retail sector 20 18 1.18 Comparison of total retail & retail e-commerce sale in India 23 19 1.19 SWOT analysis of organised retail industry 24 20 1.20 Retailing formats in India and their basic facility 27 21 1.21 Brief summary of formats in organised retailing 31 22 1.22 Indian retail groups and area of their business 40 23 3.1 Undifferentiating targeting strategy 67 24 3.2 Differentiating targeting strategy 67 xiv 25 3.3 Concentrated targeting strategy 67 26 3.4 Positioning strategies 68 27 3.5 Retail marketing strategy 69 28 3.6 The SWOT analysis 70 29 3.7 The PEST analysis 71 30 3.8 Five force analysis 72 31 3.9 ETOP analysis 72 32 3.10 BCG matrix 73 33 3.11 Generic strategies 73 34 3.12 Retail marketing mix 75 35 3.13 Merchandise planning 76 36 3.14 Grid layout 81 37 3.15 Curving or loop layout 82 38 3.16 Free flow layout 82 39 3.17 Spine layout 87 40 3.18 Consumer behaviour 84 41 3.19 Consumer choice and decision-making 84 42 4.1 Locale and universe of the study 89 43 5.1 Merchandise portfolio analysis of HyperCITY 102 44 5.2 Merchandise portfolio analysis of Star Bazaar 106 45 5.3 Merchandise portfolio analysis of more.MEGAStore 109 46 5.4 Merchandise portfolio analysis of Big Bazaar 114 47 5.5 Merchandise portfolio analysis of D Mart 118 48 5.6 Merchandise portfolio analysis of Big Megamart 121 49 5.7 Merchandise portfolio analysis of West Side 124 50 5.8 Merchandise portfolio analysis of Max 127 51 5.9 Merchandise portfolio analysis of Reliance Digital 131 52 5.10 Merchandise portfolio analysis of NEXT 134 xv xv 53 6.1 Positioning strategies of organised retailers 158 54 6.2 Retail marketing strategies of organised retailers 160 55 6.3 Retail store position in Boston matrix 162 56 6.4 Generic (business-level) strategies of organised retailers 162 57 7.2.1 Types of products, the customers purchase from HyperCITY 236 58 7.2.2 Types of products, the customers purchase from Star Bazaar 236 59 7.2.3 Types of products, the customers purchase from more.MEGAStore 237 60 7.2.4 Types of products, the customers purchase from Big Bazaar 237 61 7.2.5 Types of products, the customers purchase from D Mart 238 62 7.2.6 Types of products, the customers purchase from Big Megamart 239 63 7.2.7 Types of products, the customers purchase from West Side 239 64 7.2.8 Types of products, the customers purchase from Max 240 65 7.2.9 Types of products, the customers purchase from Reliance Digital 240 66 7.2.10 Types of products, the customers purchase from NEXT 241 67 7.2.11 Types of products, the customers buy from HyperCITY while impulse purchase 242 68 7.2.12 Types of products, the customers buy from Star Bazaar while impulse purchase 243 69 7.2.13 Types of products, the customers buy from more.MEGAStore while impulse purchase 243 70 7.2.14 Types of products, the customers buy from Big Bazaar while impulse purchase 244 71 7.2.15 Types of products, the customers buy from D Mart while impulse purchase 244 72 7.2.16 Types of products, the customers buy from Big Megamart while impulse purchase 245 73 7.2.17 Types of products, the customers buy from West Side while impulse purchase 245 74 7.2.18 Types of products, the customers buy from Max while impulse purchase 246 75 7.2.19 Types of products, the customers buy from Reliance Digital while impulse purchase 246 76 7.2.20 Types of products, the customers buy from NEXT while impulse purchase 247 77 7.3.1 The influence of Sociological / Environmental / External factors on buying behaviour of 257 xvi customers 78 7.3.2 The influence of Psychological / Individual / Internal factors on buying behaviour of customers 259 79 7.3.3 The influence of Store related factors on buying behaviour of customers 261 80 7.3.4 Factors influencing buying behaviour of customers of HyperCITY 263 81 7.3.5 Factors influencing buying behaviour of customers of Star Bazaar 263 82 7.3.6 Factors influencing the buying behaviour of customers of more.MEGAStore 264 83 7.3.7 Factors influencing buying behaviour of customers of Big Bazaar 264 84 7.3.8 Factors influencing buying behaviour of customers of D Mart 265 85 7.3.9 Factors influencing buying behaviour of customers of Big Megamart 266 86 7.3.10 Factors influencing buying behaviour of customers of West Side 266 87 7.3.11 Factors influencing buying behaviour of customers of Max 267 88 7.3.12 Factors influencing buying behaviour of customers of Reliance Digital 268 89 7.3.13 Factors influencing buying behaviour of customers of NEXT 268 90 8.1 Weighted average scores (WAS) for customer satisfaction with merchandise assortment 271 91 8.2 Weighted average scores (WAS) for customer satisfaction with merchandise categories in the assortment 273 92 8.3 Weighted average scores (WAS) for customer satisfaction with the variety of merchandise in each category 276 93 8.4 Weighted average scores (WAS) for customer satisfaction with availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise 278 94 8.5 Weighted average scores (WAS) for customer satisfaction with brands in each merchandise category 281 95 8.6 Weighted average scores (WAS) for customer satisfaction with quality assurance and quality of merchandise 283 96 8.7 Weighted average scores (WAS) for customer satisfaction with the services 285 97 8.8 Weighted average scores (WAS) for customer satisfaction with the quality of the services 288 98 8.9 Weighted average scores (WAS) for customer 290 xvii satisfaction with the pricing of merchandise & services 99 8.10 Weighted average scores (WAS) for customer satisfaction with the overall value V/S price of merchandise & services 293 100 8.11 Weighted average scores (WAS) for customer satisfaction with the exterior atmospherics of the store 295 101 8.12 Weighted average scores (WAS) for customer satisfaction with the interior atmospherics of the store 298 102 8.13 Weighted average scores (WAS) for customer satisfaction with the store layout 300 103 8.14 Weighted average scores (WAS) for customer satisfaction with the visual merchandising 302 104 8.15 Weighted average scores (WAS) for customer satisfaction with the methods of promotion and communication 305 105 8.16 Weighted average scores (WAS) for customer satisfaction with the advertising methods 307 106 8.17 Weighted average scores (WAS) for customer satisfaction with the sales promotion methods 309 107 8.18 Weighted average scores (WAS) for customer satisfaction with the direct retailing methods 312 108 8.19 Weighted average scores (WAS) for customer satisfaction with the personal retailing methods 314 109 8.20 Weighted average scores (WAS) for customer satisfaction with the public relation/publicity methods 317 110 8.21 Weighted average scores (WAS) for customer satisfaction with the internet/interactive retailing methods 319 111 8.22 Weighted average scores (WAS) for customer satisfaction with the sales personnel 321 112 8.23 Weighted average scores (WAS) for customer satisfaction with the sales personnel 323 113 8.24 Weighted average scores (WAS) for customer satisfaction with the factors pertained to personnel 326 114 8.25 Weighted average scores (WAS) for customer satisfaction with the other tangible evidence 328 xviii ABBREVIATIONS ABBREVIATIONS Sr. LIST OF FIGURES No. Abbreviation Full-form 1 GDP Gross Domestic Product 2 FICCI Federation of Indian Chambers of Commerce and Industry 3 GRDI Global Retail Development Index 4 NDA National Democratic Alliance 5 BJP Bharatiya Janata Party 6 FDI Foreign Direct Investment 7 USD United States Dollar 8 IBEF India Brand Equity Foundation 9 M&A Merger & Acquisition 10 IKEA Ingvar Kamprad Elmtaryd Agunnaryd 11 USA United States of America 12 D&B Dun & Bradstreet 13 RBI Reserve Bank of India 14 CRM Customer Relationship Management 15 SWOT Strengths, Weaknesses, Opportunities and Threats 16 MBO Multi Brand Outlets 17 SKU Stock Keeping Units 18 ATM Automated Teller Machine 19 CDIT Consumer Durable IT and Telecom 20 FMCG Fast-Moving Consumer Goods 21 RRL Reliance Retail Limited 22 PDS Public Distribution System 23 DMPL Del Monte Pacific Ltd. 24 B2B Business To Business 25 CD Compact Disc 26 DVD Digital Versatile Disc 27 VCD Video Compact Disc 28 CSR Corporate Social Responsibility 29 NYOP Name Your Own Price 30 UK United Kingdom 31 SERVQUAL Service Quality 32 DIY Do It Yourself 33 PEST Political, Economic, Social and Technological xix 34 ETOP Environmental, Threat and Opportunity Profile 5 BCG Boston Consulting Group 36 USP Unique Selling Proposition 37 SBU Strategic Business Units 38 EDLP Everyday Low Pricing 39 CLV Customer Lifetime Value 40 OTC Over The Counter 41 CFL Compact Fluorescent Lamp 42 LCD Liquid-Crystal-Display 43 USB Universal Serial Bus 44 OTG Oven Toaster and Griller 45 HCL Hindustan Computer Limited 46 DTH Direct To Home 47 HUL Hindustan Unilever Limited 48 UPS Uninterruptible Power Supply 50 DSLR Digital Single-Lens Reflex 51 HDMI High-Definition Multimedia Interface 52 CCTV Closed-Circuit Television 53 EMI Equated Monthly Instalment 54 ROI Return On Investment 55 RMM Retail Marketing Mix 56 PRM Partner Relationship Management 57 MRP Maximum Retail Price 58 RCP ResQ Care Plan 59 SMS Short Message Services 60 NGO Non-Governmental Organisation 61 LPA Lacs Per Annum 62 WAS Weighted Average Score xx ABSTRACT The purpose of this research work is to furnish a conclusive picture of retailing strategies of products and customer services in the organised retail sector. Therefore, the research work is multifarious and comprises the analysis of the various products & customer services and retailing strategies introduced by organised retail industry to comprehend the viewpoint of organised retailers. And to perceive the viewpoint of customers, an analysis of their buying behaviour, factors influencing their buying behaviour and satisfaction level with various products & customer services have also been incorporated in this work. Both descriptive and exploratory research approaches were employed in compiling this study. The study was endeavoured in the Maharashtra state of India and the organised retail stores from various segments were selected from various parts of the state. Efforts were made to cover the entire state geographically and regionally, at least 2 well-known cities were selected from each region. The whole Maharashtra state was divided into 5 regions, i.e. northern, southern, eastern, western and central. From each region, two well-known cities were selected. Further, from each city, one retail store was selected for the study. Out of total 10 retail stores, 5 were hyper stores and remaining 5 were super stores. These hyper and super stores have also been selected by a planning that from each region one hyper and one super store must be selected for the study. A non-probability convenience sampling technique was resorted for institutional (store personnel) and individual (customers) respondents. From each retail store, an institutional questionnaire got filled up by store personnel to receive accurate data about various products, services and retailing strategies introduced by the respective retail store. In total 2000 customers of retail stores (i.e. 10 retail stores * 200 customers) were contacted. From the area of each selected retail stores 200 respondents contacted at a convenient basis for obtaining data about their buying behaviour, factors influencing their buying behaviour, and satisfaction level pertained to products and services offered by retail stores. xxi The research work is based on both secondary and primary data. Two instruments, one for personnel of retail stores (institutional questionnaire) and another for customers of retail stores (individual questionnaire), were constructed to accomplish the objectives of the study. Both questionnaires were tested by conducting a pilot study of the few respondents selected on a random basis. ABSTRACT Taking the insight from this pre- testing of the questionnaires, certain items were included and even excluded to modify the questionnaire for the final study. Survey method and printed questionnaires were exercised in accumulating the primary data. The questionnaires were supervised individually practising the face to face approach. The research work yields vivid research results. Most noteworthy, results depicted that retailing strategies introduced by organised retailers are significant to sustain in the hyper-competitive retail market. Results also flaunted that customers are significantly satisfied with various products & services introduced by organised retailers except a few. The research work would be rationalised for its multiple utilities. The study contributes knowledge & facts about the merchandise (products) & customer services penetrated by organised retailers. The study aspires to reveal the retailing strategies determined by organised retailers. This study also delivers evidence about the dilemmas of customers along with some potential solutions to them. The study ascribes details for the betterment needed in the merchandise (products) & services that clients want. The study also determines diverse components and customer services mandatory to persist the retailing industry. The study additionally functions to deliver the factors that serve towards the superior level of customer satisfaction in the organised retail industry. The study also offers the suggestion to the management of organised retail stores to appraise upon the retailing strategy. The study assists in comprehending the retailing environment, retailing culture and the retail decision process. xxii Introductio n n  An overview of Indian retail industry  An evolution of retail in India  Reforms in Indian retail  Investments in the retail sector  Government initiatives for growth of the sector  Key drivers to the exponential growth of the Indian retail industry  Key challenges in Indian retail industry  SWOT analysis of organised retail industry in India  The classification of retail industry  Retailing formats in India  Major retailers in India  An overview of Indian retail industry  An evolution of retail in India  Reforms in Indian retail  Investments in the retail sector  Government initiatives for growth of the sector  Key drivers to the exponential growth of the Indian retail industry  Key challenges in Indian retail industry  SWOT analysis of organised retail industry in India  The classification of retail industry  Retailing formats in India  Major retailers in India 1.1 An overview of Indian retail industry The Indian retail industry is one among the most contributing sectors to Indian GDP at the rate of 14 to 15% (Dikshit, 2011) and 8% in the employment of the country (FICCI, 2011). The Indian retail industry is considered as one among the most vibrant industries in India and one of the pillars of Indian economy. A bright future of the Indian retail sector can be glanced with the growth estimation of this sector, as India is estimated as the fastest growing retail market and will become one of the top five retail markets in the worlds in terms of economic value (Majumder, 2011). In 2011, from the perspective of investment in the retail sector, India has been ranked in Global Retail Development Index as 4th most attractive country among the 30 flourishing market around the globe (FICCI, 2011). Global Retail Development Index (GRDI) 2014 of the US Based global management consulting firm A T Kearney ranked the Indian retail industry at 20th position among the top 30 developing countries. Figure 1.1 explains the competitive position of India in comparison to other 30 promising countries of the world. Moriarty et al. (2014) explain that in GRDI 2014, India has been dropped in 20th place from its 14th place in 2013 due to slow economic growth (4.7% only), currency fluctuations, high rate of consumer price inflation, government debt, high current account deficits and strict FDI policies. But, Moriarty et al. (2014) mention further that India is still an eye-catching long-term retail destination due to several reasons, which are the population of 1.2 billion, younger population (50% of the total population is under 30 years), one-third of the total population is cities resident, increasing disposable income and pro-business policies attitude of the current National Democratic Alliance (NDA) government headed by the Bharatiya Janata Party (BJP) and prime minister Mr. Narendra Modi. Many industry experts have positive feelings for long-term GDP outlook and industrial growth of India. According to Global Retail Development Index (GRDI) 2014, despite the several slumps, Indian retailing sector has a potential to grow in the future and the market is attractive for investment. Figure 1.2 explains the attractiveness in terms of investment in Indian Retail Sector as to consider. 1 1 Figure 1.1: Global Retail Development Index (GRDI) 2014 Source: Adopted from Moriarty et al. 1.1 An overview of Indian retail industry (2014) Figure 1.1: Global Retail Development Index (GRDI) 2014 Source: Adopted from Moriarty et al. (2014) 2 Figure 1.2: Global Retail Development Index 2014: Market attractiveness Source: Adopted from Moriarty et al. (2014) Figure 1.2: Global Retail Development Index 2014: Market attractiveness Source: Adopted from Moriarty et al. (2014) The retail sector of India, especially organised retailing is passing through the transformation from last one decade and despite the several slump, the organised retail market is growing exponentially. The organised retail segment is shifting towards the adoption of modern retailing concepts. In 2007, the Indian retail market size was approximated at USD 339.7 billion; in which USD 13.9 billion was organised retail market size. In 2010, the Indian retail market size was approximated at USD 435 billion, in which 95% market (USD 414) was of an unorganised traditional retail market and only 5% i.e. USD 21 billion was organised retail market size (FICCI, 2011). In 2011-12, the Indian retail market was taped at USD 500 billion in which the organised retail market has been just 7%. The Indian retail market is anticipated to reach USD 738.71 billion by 2016-17 with a compounded growth rate of 15% per annum. While, the organised retail is anticipated to attain the 10.2% share of the total retail market with a growth rate of 24% (The Hindu, 2014). In 2014, the Indian retail market was estimated at USD 490 billion and was projected to reach USD 865 billion by 2023 with a compounded annual growth rate of 6% (The Hindu Business Line, 2014). The figure 1.3 indicates that in 2013, the organised retail market was 8% and expected to grow 24% by 2020. 3 3 Figure 1.3: Significant scope for expansion in organised retail Source: Adopted from (IBEF, 2015) Figure 1.3: Significant scope for expansion in organised retail Source: Adopted from (IBEF, 2015) Table 1.1 depicts the comparative situation of organised and unorganised retailing sector of India and other countries of the world. Table 1.1: India in comparison to the world in retailing Country Organised Unorganised USA 85 15 Taiwan 81 19 Malaysia 55 45 Indonesia 30 70 China 20 80 India 08 92 Source: Adopted from (Madann, 2012) The growth of the organised retail sector in India can also be glanced in terms of sheer space. 1.1 An overview of Indian retail industry In 2012, the total organised retail space supply was 2.5 million square feet, which was increased by 4.7 million square feet with a growth rate of 78% (Chakraborty, 2013). The Secretary General of Assocham, India D. S. Rawat, also glanced a phenomenal growth of the Indian retail sector and said that “favourable demographics, increasing urbanisation, an increase in the number of nuclear families, increasing affluence amid buyers, increasing attraction and preference for branded products and greater aspirations are the factors that will enhance the Indian retail consumption” (The Economic Times, 2014). 4 4 The consumer market in India is getting a robust growth and the consumption level in India is assumed to be double in the next five years from 750 billion USD to 1.5 trillion USD. According to panel members of 7th Food and Grocery Forum of India, the food and grocery retail market of India has ample growth opportunities and that segment of the retail market constitutes about 60% of the total retail market (Phulia and Sharma, 2014; Moriarty et al., 2013). Figure 1.4 explains the breakup of USD 500 billion retail sector of India. Figure 1.4: India’s retail market Source: Adopted from Moriarty et al. (2013) Figure 1.4: India’s retail market Figure 1.4: India’s retail market Source: Adopted from Moriarty et al. (2013) The growing number of middle-class households is also a contributing factor to the growth of the Indian retail sector. It has been estimated that 91 million households will be in the category of the middle-class from its 21 million by 2030. It has also been estimated that 570 million people will prefer to live in the cities in 2030 (Chakraborty, 2013; FICCI, 2011). The growing middle class is redefining the luxury market with their changing buying habits. The growth of retailing sector can also be glanced from the fact that about 40% of the luxury goods sales in India occur from the non-metros customers (Rai and Gopal, 2014). The huge population, especially the young population, higher domestic consumption and the favourable macro trends are other contributing factors to the exponential growth of the Indian retail sector. 5 5 1.2 An evolution of retail in India 1.2 An evolution of retail in India Figure 1.5: An evolution of retail in India over the years Source: Adopted from (IBEF, 2015) Figure 1.5: An evolution of retail in India over the years Source: Adopted from (IBEF, 2015) Retailing is a revolutionary and one among biggest sector in India. The new entrant in Indian retailing signifies the beginning of the retail revolution. Retail market of India is growing with a great pace and it will have great growth opportunities in the future. The origin of retailing in India can be identified with the emergence of Kirana stores and mom-and-pop stores, such kind of stores used to cater to the people of local communities. By the time, government supported rural retail and, many franchise stores enter into India with the assistance of Khadi & Village Industries Commission of India. The economy opened up in 1980s results in a change of retailing. The first few companies who came up in India as retail chains were in the textile sector, for instance, S Kumar’s, Bombay Dyeing, Raymond’s, etc. Later Titan opened their retail showrooms in the organised retail sector of India. With the time passing new entrants started shifting from manufacturing to pure retailing. Retailing in India was considered in an opening stage in the 90s, when some manufacturers opened their own outlets. Many players realised the potential of this market and entered into the apparel segment in the period of 1990-2005. A great expansion can be seen in the period of 2005-10, when large corporate entered into this market with substantial investment in food and general merchandise segment. From 2010 onwards, the period is considered as the consolidation phase. During this period, 6 players are moving towards smaller cities and even rural areas. Figure 1.5 explains the growth of the Indian Retail Sector over the years since its inception. The players have introduced the international brands for their customers. During this period, the FDI in retail is 100% in the case of a single brand and 51% in multi-brand retailers. Private label brands have also emerged in this period. Moreover, the emergence of e- retailing is also the gift of this period. 1.2 An evolution of retail in India Figure 1.6: Evolution of retailing in India Source: Adopted from (Sikri and Wadhwa, 2012) Figure 1.6: Evolution of retailing in India Source: Adopted from (Sikri and Wadhwa, 2012) Retail stores like as Planet M and Music world in music, Crossword in books, Food world in FMCG entered the market before 1995. Shopping Malls started opening in urban areas, giving the world class experience to the customers. Later supermarkets and hypermarkets emerged. The evolution of retail sector in India includes continuous improvement in distribution channels, supply chain management, technology, back-end operations, etc. this will lead to more and more mergers, acquisitions, consolidation and huge investments. Indian retail sector, which has a huge potential for growth in a modern retailing world is just a beginning. In Indian retail sector, the most of the market is unorganised. The key challenge for the organised retail sector is the closest competition from the unorganised sector of the Industry. Unorganised retailing in India is prevalent for centuries. The advantage in unorganised retailing in India is consumer familiarity which runs from one generation to another. Unorganised retailing is a low-cost structure; mostly run by self-owned 7 peoples with a very low retail space as well as labour costs and ultimately lower taxes to pay. The organised retail sector in India is very small but with a great scope for growth. 1.3 Reforms in Indian retail The central government of India was reluctant to allow the FDI in multi-brand retail business till 2011 that resulted in no trade and FDI in any kind of retail outlets by foreign investors to offer multiple products with diversified brands to consumers in India. However, on 24 November 2011, the central government of India announced that India will allow up to 51 percent FDI in “multi-brand retail”; single brand retailers can own 100 percent of their Indian stores, which has been increased from the previous cap of 51 percent; both single brand and multi-brand stores in India have to source nearly a third of their merchandise from small and medium sized Indian suppliers; moreover, all these single or multi-brand stores are obliged to restrict their business in only 53 cities having a population of over one million in India. It is expected that the stores will have full access to over 200 million urban customers in India; multi-brand retailers are obliged to have a minimum investment of USD 100 million, with at least half of the amount invested in cold chains, transportation, back- end infrastructure, packing, refrigeration, sorting and processing to reduce the post- harvest losses; and bring remunerative prices to farmers; and opening of retail competition will be within the India’s federal structure of government. It means the policy is operational to the legal framework of India. The states of India have right to accept it and implement it, or they can reject. Which means actual implementation of the policy will be within the parameters of state laws and regulations. The opening of the doors of the retail industry will lead to global competition and the retail rush to India. Indian retail is equipped with the capabilities of transforming not only the retailing landscape, but also the nation's existing infrastructure. It is expected to help high inflation, but is likely to be opposed by millions of small-scale retailers, who consider these large foreign chains as the biggest threat. Analyst’s claim- the need to control food price inflation, which doubles in previous years, forced the government to open the sector. India’s food supplies have been controlled by tens of millions of middlemen. Traders offer little technical support to help farmers to boost their packaging technology, productivity, but traders add huge markups to farm prices. 1.3 Reforms in Indian retail 8 8 According to analysts, big foreign retailers will provide a force to set up modern supply chains along with refrigerated vans, cold storage and more efficient logistics. On September 20, 2012, the government of India has notified two significant regulations that would go a long way in the development of the Indian retail sector (Nath and Gupt, 2012). First, permitting FDI in multi-brand retail trading and second is, simplifying the rules for single-brand retail trading to make it more business friendly. On January 11, 2012, India government has approved innovation and increased competition in single-brand retail. The reform attempts to attract investments in marketing & production, encourage increased sourcing of goods from India, and increase the competitiveness of Indian enterprises with the access to technologies, management practices and global designs. According to the announcement, India requires single-brand retailers who have more than 51% foreign ownership to source at least 30% of the value of products from Indian village & cottage industries, artisans & craftsmen and small industries. 1.4 Investments in the retail sector In the recent time, Merger & Acquisition (M&A) in the Indian retail sector are also indicating the growth potential of this sector. In 2013, there were a total number of 26 M&A deals of worth UDS 3.5 billion, refer the figure 1.7 & 1.8, which depict the recent M&A in the sector. Figure 1.7: Merger & acquisition in the Indian retails sector in 2013 Source: Adopted from (IBEF, 2015) Figure 1.7: Merger & acquisition in the Indian retails sector in 2013 Source: Adopted from (IBEF, 2015) 9 9 Analysing the strong growth potential, many corporations are attracting for high foreign investment in the sector. Figure 1.8: Strong growth potential attracting high foreign investment Source: Adopted from (IBEF, 2015) Figure 1.8: Strong growth potential attracting high foreign investment Source: Adopted from (IBEF, 2015) According to the Department of Industrial Policies and Promotion, the retail industry in India has received FDI equity inflows USD 275.38 million in the single brand segment from April 2000- January 2015 (IBEF, 2015). Due to the rising need of Indian consumers in different sectors including home appliances and consumer electronics, many companies have invested in India in the past few months (IBEF, 2015). Some of them are as follows: Amazon and Future Group have made an agreement for selling goods online and both the companies will introduce new lines of products that will be offered only at Future Group retail stores as well as Amazon. Initially, Future Group will introduce for selling more than 45 private labels of apparel, the electronics, home, and food category products, while Amazon will be responsible for order fulfilment as well as customer service for the merchandise on its portal. The Japanese firm Soft Bank has invested approximate US $ 627 million in Snapdeal, which made them largest shareholder in the e-commerce. Snapdeal has planned to utilise this investment for 10 expansion and growth of its chain of fulfilment centres. Myntra.com is planning to expand their business by launching a wedding store by November, 2014. This wedding store will cater the needs of wedding wear and accessories of a bride and groom. Free charge plans to launch a service that will promote online shopping through its mobile platforms and Website by giving cash back offers to shoppers who use the service to buy products on eCommerce sites. 1.4 Investments in the retail sector Burger King a well-known American fast food chain store plans to start 12 outlets in India within next three months. At present, Burger King is looking for a company-operated and company- owned model for its store operations in India and its store will be located in the high street as well as food court locations. A logistic service provider, i.e. Gojavas has introduced a ‘try-before-you-buy’ experience in the eCommerce space. In this model, the delivery person waits at the place of the customers until the customer uses the product before accepting or returning it. Paytm is planning to set up 30,000 to 50,000 retail outlets where the customers can load the cash in their digital wallets. It’s searching retailers mostly Kirana stores to enrol them as merchants for accepting digital payments (IBEF, 2015). Mobikwik a mobile wallet company has made a partnership with Jabong.com to give mobile payment services to Jabong.com customers. FashionAndYou has opened three distribution hubs for faster deliveries. Amazon Inc and Flipkart India will invest approximately 2,300 ₹ crores in the near future in order to acquire more customers in the fast-growing online retail market. Abu Dhabi based Lulu Group is planning to invest 25,00 crore in an ₹ integrated meat processing unit, a fruit and vegetable processing unit and a modern shopping mall in Hyderabad, Telangana. Datawind has made a partnership with Homeshop18 to jointly launch special sales programmes for broadcast, mobile and internet media to provide greater access. 1.5 Government initiatives for growth of the sector The Union Minister of Commerce and Industry, Government of India, Ms Nirmala Sitharaman stated that the ministry is stressing to build a branding and 11 marketing culture of the Indian products to the world. She has stated further that the ministry is willing to start a Free Trade Agreement with the European Union. The Indian Government has taken various initiatives to improve the retail industry in the country. Some of the initiatives are as follows: IKEA has entered into a Memorandum of Understanding with the government of Telangana state to set up its first store in Hyderabad (India). IKEA retail outlets adopt a set of standard design and their each location requires an investment around 500-600 crore (US$ 80.38 - 96.46 million). ₹ The Foreign Investment Promotion Board has approved five different proposals of worth 420 ₹ crores (USD 67.53 million) from Puma SA, Flemingo and Bestseller for retailing in India (IBEF, 2015). Furthermore, the board cleared three 100 percent single-brand retail proposals worth 222.5 ₹ crores (USD 35.77 million). The Indian Government is also in the final phase of to implement Goods & Services Tax Bill. This Bill is seen as a key in improving the business climate of our country and to facilitate industrial growth. The strong efforts of the Indian government lead to the investors’ confidence for FDI in India. Figure 1.9 depicts the competitive position of India in the FDI Confidence Index 2014 in comparison to other countries of the world. Figure 1.9: FDI Confidence Index 2014 Source: Adopted from (IBEF, 2015) Figure 1.9: FDI Confidence Index 2014 Source: Adopted from (IBEF, 2015) Figure 1.9: FDI Confidence Index 2014 Source: Adopted from (IBEF, 2015) 12 Earlier, to protect the 15 million small retailers or Kirana stores government of India has mainly closed the organised retailing for other countries and solely permitted 51 % FDI in single-brand retail with prior Government permission (Corporate Catalyst India, 2012). FDI is also allowed in the wholesale business. Policymakers were trying their best to find out new ways by which FDI is possible without hurting the small retailers or unorganised retailing. The government of India was also planning for retail trading in sports goods, stationery, electronics goods and building equipment. Earlier entry option for foreign players  Most widely used entry route by multinational retailers  Domino’s entered in India through master franchise root and Pizza Hut entered through regional franchise  100% FDI is allowed in wholesale trading which involves building of a large distribution infrastructure to assist local manufacturers  The wholesaler deals only with smaller retailers and not consumers Figure 1.11: Business models for entry in Indian retail markets prior to September 20, 2012 Figure 1.11: Business models for entry in Indian retail markets prior to September 20, 2012 Source: Adopted from (IBEF, 2015) 1.5.1 Business models for entry into Indian markets prior to the recent initiative of the Government p ( ) 1.5.1 Business models for entry into Indian markets prior to the recent initiative of the Government 1.5 Government initiatives for growth of the sector The policy of allowing 51 % FDI in single-brand product retailing had led to the entry of solely a few global brands like as Louis Vuitton (shoes, travel accessories, watches, ties, textile ready-to-wear), Nike (footwear), Fendi (luxury products), Toyota (retail trading of cars), Argenterie Greggio (silverware, cutlery, traditional home accessories and gift items), Lladro (porcelain goods) and Damro (Knock-down furniture), into retail trading. French luxury industry of 12 billion Euro is keeping an eye on India to enter into retail business in the domestic luxury segment. Now, the Indian government has approved the 100% FDI in single brand product retailing and 51% FDI in Multi-brand and front-end retailing. The wholesale cash-&-carry sector has already been allowed 100% FDI. This will provide several benefits to the country as depicted in figure 1.10. Figure 1.10: Benefits of FDI and its limit in the different segment Figure 1.10: Benefits of FDI and its limit in the different segment 13 Source: Adopted from (IBEF, 2015) Source: Adopted from (IBEF, 2015) of the Government There were three different routes as stated in figure 1.11 for foreign players to enter into India’s retail sector prior to the approval by the Indian government for the 100 FDI in single brand product retailing and 51% FDI in Multi-brand and front-end retailing (Corporate Catalyst India, 2012). Earlier entry option for foreign players Franchise agreement  Most widely used entry route by multinational retailers  Domino’s entered in India through master franchise root and Pizza Hut entered through regional franchise Cash and Carry Wholesale Trading  100% FDI is allowed in wholesale trading which involves building of a large distribution infrastructure to assist local manufacturers  The wholesaler deals only with smaller retailers and not consumers  Metro AG of Germany was the first significant global player to enter in India through this route Strategic licensing agreement  Foreign company enters into a licensing agreement with a domestic retailers  Mango, the Spanish Apparel brand has entered in India through this route with an agreement with Piramyd, Mumbai  SPAR entered into a similar agreement with Radhakrishna Foodlands Pvt. Ltd. Figure 1.11: Business models for entry in Indian retail markets prior to September 20, 2012 1.6 Key drivers to the exponential growth of the Indian retail industry  The huge population, especially the younger population: India is considered as one among the largest emerging markets of the world due to its population, which is over 1.2 billion (Rahman, 2012). India is also considered as one of the largest economies in terms of purchasing power of the population. The 50% of the total Indian population is below the age of 30  The huge population, especially the younger population: India is considered as one among the largest emerging markets of the world due to its population, which is over 1.2 billion (Rahman, 2012). India is also considered as one of the largest economies in terms of purchasing power of the population. The 50% of the total Indian population is below the age of 30 14 years and about 1/3 of the Indian population resides in cities (Rai and Gopal, 2014). Table 1.2 presents the percentage of population below the age of 25 years in different countries. Table 1.2: Percentage of population below the age of 25 years Country Percentage of Population India 53 China 42 Indonesia 30 USA 30 Japan 27 Germany 26 Source: Adopted from (Madann, 2012)  Changing lifestyle and consumer behaviour: Due to increasing working population, comfortable life style, more travel and leisure are the preferences of Indian customers. These factors are the drivers for the growth of retailing in India; whether it is Appliances, Cosmetics & Toiletries, Electronics etc. (Handa and Grover, 2012).  Changing lifestyle and consumer behaviour: Due to increasing working population, comfortable life style, more travel and leisure are the preferences of Indian customers. These factors are the drivers for the growth of retailing in India; whether it is Appliances, Cosmetics & Toiletries, Electronics etc. (Handa and Grover, 2012).  Rising affluence amid consumers: It is estimated that 91 million households will be in the category of the middle-class from its 21 million by 2030 (Chakraborty, 2013; FICCI, 2011). This increase in the middle-class households is a contributing factor towards the growth of the retail sector.  Demographic changes and improvements in the standard of living: The growing middle class is redefining the luxury market with their changing buying habits. The growth of retailing sector can also be glanced from the fact that about 40% of the luxury goods sales in India occur from the non-metros customers (Rai and Gopal, 2014). 1.6 Key drivers to the exponential growth of the Indian retail industry  Demographic changes and improvements in the standard of living: The growing middle class is redefining the luxury market with their changing buying habits. The growth of retailing sector can also be glanced from the fact that about 40% of the luxury goods sales in India occur from the non-metros customers (Rai and Gopal, 2014).  Favourable demographics: The Demography Dynamics are also favourable for the Indian organised retail sector as approximately 60% of Indian population is young and below the age of 30 years (Rahman, 2012).  Favourable demographics: The Demography Dynamics are also favourable for the Indian organised retail sector as approximately 60% of Indian population is young and below the age of 30 years (Rahman, 2012).  Changing consumer preferences and shopping habits: Shopping behaviour of consumers has been totally changed due to the change in taste and preferences of consumers. Due to the online retailing consumers are well aware of new brands and stores and they buy according to their needs and occasions. Consumers are driving the growth of organised retail. Therefore, the organised retailing is not only growing its market, but also compelling the retailers to offer a wider range of brands as well as a variety (D&B, 2014).  Changing consumer preferences and shopping habits: Shopping behaviour of consumers has been totally changed due to the change in taste and preferences of consumers. Due to the online retailing consumers are well aware of new brands and stores and they buy according to their needs and occasions. Consumers are driving the growth of organised retail. Therefore, the organised retailing is not only growing its market, but also compelling the retailers to offer a wider range of brands as well as a variety (D&B, 2014). 15  The increase in a working population: India is the second largest country in the world in terms of population after China. And, it is the largest consumer market in the world due to its favourable demographics (D&B, 2014). An increase in the number of working women has increased the growth of discretionary items. There has been a 20% increase in the number of working women population in the last decade.  The increase in a working population: India is the second largest country in the world in terms of population after China. And, it is the largest consumer market in the world due to its favourable demographics (D&B, 2014). 1.6 Key drivers to the exponential growth of the Indian retail industry An increase in the number of working women has increased the growth of discretionary items. There has been a 20% increase in the number of working women population in the last decade.  The emergence of nuclear families: With the passage of time there is a shift from joint families to nuclear families. The income level of nuclear families has been increased because both the members have started to earn; which results in an increase in purchasing power and lack of time. Thereby nuclear families are expecting everything under one roof. This led to the development of organised retailing.  Increasing urbanisation: The preference shift of population towards the urban livings is also a good sign of the growth of the Indian organised retail sector. It is estimated that 570 million people will prefer to live in the cities in 2030 (Chakraborty, 2013; FICCI, 2011).  The increase in disposable income and customer aspiration: Booming Indian economy is a consequential source of opportunity for the retailing sector. India is the third largest country of the world in terms of purchasing power of consumers (Rahman, 2012). Due to large working population and an increase in double-income households, the disposable income, as well as the purchasing power of consumers, especially in urban India, has been growing. The growing preference for branded products and services, higher aspirations, demand and increase in the expenditure for luxury items are the major cause of the growth of this sector in India. The branded products, especially in the categories of Cosmetics, Apparels, Shoes, Watches, Beverages, Food and even Jewellery are being preferred as the lifestyle products by the urban consumers of India.  The increase in disposable income and customer aspiration: Booming Indian economy is a consequential source of opportunity for the retailing sector. India is the third largest country of the world in terms of purchasing power of consumers (Rahman, 2012). Due to large working population and an increase in double-income households, the disposable income, as well as the purchasing power of consumers, especially in urban India, has been growing. The growing preference for branded products and services, higher aspirations, demand and increase in the expenditure for luxury items are the major cause of the growth of this sector in India. 1.6 Key drivers to the exponential growth of the Indian retail industry The branded products, especially in the categories of Cosmetics, Apparels, Shoes, Watches, Beverages, Food and even Jewellery are being preferred as the lifestyle products by the urban consumers of India.  Increase in per capita income: The per capita income of Indian consumers is 44,345 ₹ (Rahman, 2012), which is growing and can be referred from figure 1.12. Due to an increase in per capita income, the household consumption is increasing, which is a good sign for organised retailers. As the earnings are increasing, a proportionate increase in spending can also be glanced. Moreover, the Indian economy is expected to grow by a rate of 8% and 16 salaries are also expected to hike with a rate of 15% (Rahman, 2012). This in turn will increase the household consumption, which is a healthy sign for manufacturers and retailers of consumer goods and services. Figure 1.12: Rising per capita income in India Source: Adopted from (IBEF, 2015) Figure 1.12: Rising per capita income in India Source: Adopted from (IBEF, 2015)  Increasing use of plastic money: Among the Indian consumers, the use of plastic money has been increased and due to this the consumers prefer to buy the merchandise in the category of Apparel, Consumer Durable Goods, Food and Grocery etc.  Increasing use of plastic money: Among the Indian consumers, the use of plastic money has been increased and due to this the consumers prefer to buy the merchandise in the category of Apparel, Consumer Durable Goods, Food and Grocery etc.  Improvements in infrastructure and enhanced availability of retail space: Due to the overall betterment of transportation infrastructure and services in India, now the consumers are able to cover a distance to reach a particular retail outlet. Figure 1.13 depicts the enhanced availability of retail space and mall space breakup in India.  Improvements in infrastructure and enhanced availability of retail space: Due to the overall betterment of transportation infrastructure and services in India, now the consumers are able to cover a distance to reach a particular retail outlet. Figure 1.13 depicts the enhanced availability of retail space and mall space breakup in India. 1.6 Key drivers to the exponential growth of the Indian retail industry Figure 1.13: Mall space breakup in India Source: Adopted from (IBEF, 2015) Figure 1.13: Mall space breakup in India Source: Adopted from (IBEF, 2015) 17 17  Easy credit availability: Ease in getting credit cards and ease of use of credit cards for shopping has also boosted the growth of the organised retail sector. Due to the issuance of credit cards at a larger piece by Foreign and Indian banks has increased the retail growth. According to the RBI, as of the financial year 2009, the total number of outstanding credit and debit cards in India was 24.7 million and 137.4 million respectively (D&B, 2014).  Easy credit availability: Ease in getting credit cards and ease of use of credit cards for shopping has also boosted the growth of the organised retail sector. Due to the issuance of credit cards at a larger piece by Foreign and Indian banks has increased the retail growth. According to the RBI, as of the financial year 2009, the total number of outstanding credit and debit cards in India was 24.7 million and 137.4 million respectively (D&B, 2014).  Growing liberalisation of the FDI policy in the past decade: Figure 1.14 states the Growing liberalisation of the FDI policy in India. The main benefit of FDI is that it is both supplementary and complimentary with regards to local investment. Due to FDI companies has gained access to world class technology and supplementary funds. Companies have also access to new management practices around the world and get the chance to become the part of a global market system (Prasad and Harikumar, 2013). With the help of FDI, companies will get new technologies and skills from other countries; which results in infrastructure development of the nation. It will also lead to help in creating more value for money for the customers.  Growing liberalisation of the FDI policy in the past decade: Figure 1.14 states the Growing liberalisation of the FDI policy in India. The main benefit of FDI is that it is both supplementary and complimentary with regards to local investment. Due to FDI companies has gained access to world class technology and supplementary funds. Companies have also access to new management practices around the world and get the chance to become the part of a global market system (Prasad and Harikumar, 2013). 1.6 Key drivers to the exponential growth of the Indian retail industry With the help of FDI, companies will get new technologies and skills from other countries; which results in infrastructure development of the nation. It will also lead to help in creating more value for money for the customers. Figure 1.14: Favourable FDI policy encouraging investment Source: Adopted from (IBEF, 2015) Figure 1.14: Favourable FDI policy encouraging investment Source: Adopted from (IBEF, 2015)  Organised retail concentration in tier II and III cities: Retail revolution started with the tier I cities in India, which thereby has been saturated (D&B, 2014). Now retailers are focusing on tier II and tier III cities as depicted in  Organised retail concentration in tier II and III cities: Retail revolution started with the tier I cities in India, which thereby has been saturated (D&B, 2014). Now retailers are focusing on tier II and tier III cities as depicted in 18 figure 1.15. Due to the changing landscape of Indian retail segment and the increase in competition has forced the retailers to find out the growth opportunities in tier II and III cities. Figure 1.15: Organised retail in India Source: Adopted from (IBEF, 2015) Figure 1.15: Organised retail in India Source: Adopted from (IBEF, 2015)  The internet drives awareness and online purchases: There has been a huge increase in the number of internet users and online buyers in India (D&B, 2014). Online retailing has been increased due to increase in awareness of the internet and online buying; which has opened a whole new channel of retailing in the Indian retail scenario. Online retailing is prevalent due to ease in buying and ease in getting the goods on the doorstep.  The internet drives awareness and online purchases: There has been a huge increase in the number of internet users and online buyers in India (D&B, 2014). Online retailing has been increased due to increase in awareness of the internet and online buying; which has opened a whole new channel of retailing in the Indian retail scenario. Online retailing is prevalent due to ease in buying and ease in getting the goods on the doorstep.  Experimentation with formats: Due to huge competition in retailing market of India the retailers are experimenting with the new formats (Handa and Grover, 2012).  Store design: Shopping malls and supermarkets are growing day by day, which leads to improvements in infrastructure. 1.6 Key drivers to the exponential growth of the Indian retail industry Large availability of retail space & competition has also encouraged the retailers to adopt attractive store designs and make the ambience eye-catching (Handa and Grover, 2012). This results in more visits to the shopping malls which ultimately lead to more sales & higher growth of organised retail in India.  Brief about growth value propositions: Figure 1.16 summarises the major growth value propositions in this Indian retail sector. Which can be described as higher brand consciousness, growing aspiration levels of consumers and appetite to excrement, growing young population and working women, rising incomes and purchasing power, credit availability, changing consumer preference and growing urbanisation, rapid real estate and infrastructure development, emergence of new categories, development of supply chains and 19 improving their efficiency, easy availability of credit to suppliers, expansion of existing players, R&D, innovation and new product development etc. Figure 1.16: Growth value proposition in organised retail industry Source: Adopted from (IBEF, 2015) Figure 1.16: Growth value proposition in organised retail industry Source: Adopted from (IBEF, 2015) 1.7 Key challenges in Indian retail industry  Porter’s Five Forces analysis of the organised retail sector: Figure 1.17 explains the five forces of competition in the retail industry of India.  Porter’s Five Forces analysis of the organised retail sector: Figure 1.17 explains the five forces of competition in the retail industry of India. Figure 1.17: Porter’s Five Forces analysis of organised retail sector Source: Adopted from (IBEF, 2015) Figure 1.17: Porter’s Five Forces analysis of organised retail sector Source: Adopted from (IBEF, 2015) 20  Shortage of skilled/trained manpower: 75 to 80% of the total manpower is required for store operations in the organised retail industry and front/end line assistant profiles are major positions for employment opportunities in the sector (FICCI, 2011). Unfortunately, in India, very few retail operations specific courses are offered at graduate and post graduate level. Moreover, there is also a shortage of niche retail operations specific courses in the area of store merchandising, supply chain management in retailing, CRM in retailing, stock management, customer service management etc.  Shortage of skilled/trained manpower: 75 to 80% of the total manpower is required for store operations in the organised retail industry and front/end line assistant profiles are major positions for employment opportunities in the sector (FICCI, 2011). Unfortunately, in India, very few retail operations specific courses are offered at graduate and post graduate level. Moreover, there is also a shortage of niche retail operations specific courses in the area of store merchandising, supply chain management in retailing, CRM in retailing, stock management, customer service management etc.  Lack of industry status: The retail industry in India is still fighting for the industry status from the government. In the absence of this industry status, the organised retailing is facing issues in procuring the required finance and fiscal benefits.  Lack of industry status: The retail industry in India is still fighting for the industry status from the government. In the absence of this industry status, the organised retailing is facing issues in procuring the required finance and fiscal benefits.  Policy-induced barriers: There a time’s need that Indian government should establish an apex body for governing all the retail operation In India, as the organised retail sector is administrated by two different ministries of the government of India. 1.7 Key challenges in Indian retail industry The Ministries of Commerce formulates policies pertained to retail operations in India, while the Ministry of Consumer Affairs takes care of licensing and other legislation related to retailing. A single apex body can govern retail operations and address the grievances of retailers more effectively.  Real estate market: The real estate market is offering the retail space with high prices and low availability. A brief comparison of rent as %age of sales to Indian retailers and international retailers has been presented in table 1.3. It is very difficult to find appropriate space in a central location of the cities due to an absence of retail space planning in India. Moreover, space which is available for use can easily be interchanged for commercial utilisation. Other reasons for retail space issues are infrequent auctioning of government owned lands, litigation disputes of land owners and fragmented private holdings.  Real estate market: The real estate market is offering the retail space with high prices and low availability. A brief comparison of rent as %age of sales to Indian retailers and international retailers has been presented in table 1.3. It is very difficult to find appropriate space in a central location of the cities due to an absence of retail space planning in India. Moreover, space which is available for use can easily be interchanged for commercial utilisation. Other reasons for retail space issues are infrequent auctioning of government owned lands, litigation disputes of land owners and fragmented private holdings. Table 1.3: A comparison of rent as %age of sales Format Indian Retailers International Retailers Hyper market 5-10% 1-5% Speciality retailers 15-30% 10-12% Source: Adopted from (Madann, 2012) 21  Government regulations: Government regulations, especially the Urban Land Ceiling Act) for the development of real estate is hindering the growth of the retail sector. The existing labour laws are restricting the retailers for offering a 24/7 shopping experience to their customers. Moreover, the multi- stage taxation system and other taxation laws are hindering the growth of this sector.  Government regulations: Government regulations, especially the Urban Land Ceiling Act) for the development of real estate is hindering the growth of the retail sector. The existing labour laws are restricting the retailers for offering a 24/7 shopping experience to their customers. Moreover, the multi- stage taxation system and other taxation laws are hindering the growth of this sector. 1.7 Key challenges in Indian retail industry  Technology: Technology is a major challenge faced by organised retailers for effective management of retail operations. First of all, the technology available in India in this sector is out-dated. A strong need of latest technology (digital merchandising, online inventory monitoring system, mobile payment, etc.) is there to cope up with the customer aspirations.  Technology: Technology is a major challenge faced by organised retailers for effective management of retail operations. First of all, the technology available in India in this sector is out-dated. A strong need of latest technology (digital merchandising, online inventory monitoring system, mobile payment, etc.) is there to cope up with the customer aspirations.  Lack of efficient supply-chain management and distribution channel: Distribution channels are an integral part of the retail industry. The main role of the distribution channels is to deliver the right kind of goods at the right place at the right time. The improper infrastructure and distribution channels are the reasons for inefficient distribution processes in India. Moreover, supply chains are expensive in India. Therefore, the stakeholders in the retail industry are required to improvise distribution channels as well as supply chain systems, so that they can meet the desired quality and service level by their customers.  Lack of efficient supply-chain management and distribution channel: Distribution channels are an integral part of the retail industry. The main role of the distribution channels is to deliver the right kind of goods at the right place at the right time. The improper infrastructure and distribution channels are the reasons for inefficient distribution processes in India. Moreover, supply chains are expensive in India. Therefore, the stakeholders in the retail industry are required to improvise distribution channels as well as supply chain systems, so that they can meet the desired quality and service level by their customers.  Understanding customers: Understanding customer behaviour is not an easy business. However, converting your customers into loyal customers is a more critical task in the retail industry. Retailers are required to have the implementation of effective CRM as well as a loyalty programme.  Understanding customers: Understanding customer behaviour is not an easy business. However, converting your customers into loyal customers is a more critical task in the retail industry. Retailers are required to have the implementation of effective CRM as well as a loyalty programme.  Understanding customers: Understanding customer behaviour is not an easy business. 1.7 Key challenges in Indian retail industry However, converting your customers into loyal customers is a more critical task in the retail industry. Retailers are required to have the implementation of effective CRM as well as a loyalty programme.  Inefficient power supply: Organised retail outlets utilise the electricity for various store management orations like billing systems, cold storing, lighting, lifts, escalators, air conditioning etc. and need large volumes of electricity for the same. Due to inefficient and insufficient power supply, retailers have to invest for ensuring power backups.  Competition from unorganised sector: Competition from the unorganised sector is also a challenge confronted by organised retailers in India as unorganised retailers caters niche and more personalised service to their customers. 22 22  The slow growth of organised retail market: The people are shifting continuously to modern stores from their preferred traditional stores. But this organised (modern) retail format could account only for 8% of the total retail market (Moriarty et al. (2014), which is a challenge for stakeholders.  The slow growth of organised retail market: The people are shifting continuously to modern stores from their preferred traditional stores. But this organised (modern) retail format could account only for 8% of the total retail market (Moriarty et al. (2014), which is a challenge for stakeholders.  Online retailing: Moriarty et al. (2014) estimated that e-commerce market in India will grow more that 50% in the coming five years. This may be a big challenge to store based organised retailers. Figure 1.18 presents a comparison of total retail sales and e-commerce sales. The applications of internet among the young population of India are increasing day by day and speed of the internet as well. Cash on the delivery option by online retailers is a key factor for the growth. However, the inventory management, resource availability and logistic planning are important challenges for the growth of online retailing in India.  Online retailing: Moriarty et al. (2014) estimated that e-commerce market in India will grow more that 50% in the coming five years. This may be a big challenge to store based organised retailers. Figure 1.18 presents a comparison of total retail sales and e-commerce sales. The applications of internet among the young population of India are increasing day by day and speed of the internet as well. Cash on the delivery option by online retailers is a key factor for the growth. 1.8.1 Strengths  The Indian retail market was estimated at USD 490 billion in 2014 and was projected to reach UDS 865 billion by 2023 with a compounded annual growth rate of 6% (The Hindu Business Line, 2014).  Organised retail was at USD 3.31 billion and growing with a rate of 8% (Rahman, 2012).  The retailing sector is the second largest contributor to the Indian GDP at the rate of 14 to 15% after the agriculture (Dikshit, 2011).  The Indian retail industry is contributing 8% to the employment of the country (FICCI, 2011), which is largest after agriculture.  Entry to various sources of financing in the organised retail industry.  Improvements in infrastructure and enhanced availability of retail space, almost 25 million sq. ft. retail space available (Rahman, 2012). 1.7 Key challenges in Indian retail industry However, the inventory management, resource availability and logistic planning are important challenges for the growth of online retailing in India. Figure 1.18: Comparison of total retail & retail e-commerce sale in India Source: Adopted from (Mahajan, 2015) Figure 1.18: Comparison of total retail & retail e-commerce sale in India Source: Adopted from (Mahajan, 2015)  Other bottlenecks: Few other bottlenecks of Indian Retail are as follows: there is a long way to meet international standards by Indian organised retailers. There is a lack of required retail space for organised retailing and there is no fixed consumption pattern. Other bottlenecks are inadequate utilities, IT infrastructure hurdles, policy-related hurdles, limited consumer insights, taxation challenges.  Other bottlenecks: Few other bottlenecks of Indian Retail are as follows: there is a long way to meet international standards by Indian organised retailers. There is a lack of required retail space for organised retailing and there is no fixed consumption pattern. Other bottlenecks are inadequate utilities, IT infrastructure hurdles, policy-related hurdles, limited consumer insights, taxation challenges. 23 1.8 SWOT analysis of organised retail industry in India SWOT Analysis Positive Factors Negative Factors Internal Factors Strengths (See point 1.8.1) Weaknesses (See point 1.8.2) External Factors Opportunities (See point 1.8.3) Threats (See point 1.8.4) Figure 1.19: SWOT analysis of organised retail industry Source: By author 1.8 SWOT analysis of organised retail industry in India Figure 1.19: SWOT analysis of organised retail industry Source: By author 1.8.1 Strengths 1.8.2 Weaknesses  The retail industry in India is still fighting for the industry status from the government.  There is a lack of preferred retail spaces at affordable rates in India.  Indian retailers have to travel a long distance to meet international standards. 24  There is a lack of effective supply chains in India; moreover, these supply chains are expensive.  The improper infrastructure and distribution channels are the reasons for inefficient distribution processes in India.  In India, the variations in consumption pattern can be seen on a large scale.  Retailers in India also face the shortage of trained manpower.  Need to provide Value for Money is squeezing the profit margins of retailers.  Retailers in India also feel shortage of investments for expansion of their operations 1.8.3 Opportunities 1.8.3 Opportunities  The huge population, especially the young population  The lifestyles of consumers, as well as their behaviour, are changing.  Rising affluence amid consumers  Changing demography and improving the living standards  Favourable demographics  The preferences, as well as the shopping habits of consumers, are changing.  Increase in a working population  Social factors, like dual household income of consumers  Emergence of nuclear families  Increasing urbanisation  The aspirations of customers are also increasing as an increase the disposable income  Increase in per capita income  Increasing use of plastic money  Easy credit availability  The basic infrastructure is improving and availability of retail space is also enhanced  Growing liberalisation for the FDI in the sector by the government  High domestic consumption  Favourable macro trends  Increasing spending grocery 1.8.4 Threats 25  The huge population, especially the young population  The lifestyles of consumers, as well as their behaviour, are changing.  The basic infrastructure is improving and availability of retail space is also enhanced  Growing liberalisation for the FDI in the sector by the government  Increasing spending grocery 1.8.4 Threats 25 25  Government regulations (especially the Urban Land Ceiling Act) for the development of real estate.  Increasing rental and lease costs are affecting project viability.  Existing labour laws are restricting the retailers for offering a 24/7 shopping experience.  Multi-stage taxation system and other taxation laws are hindering the growth of this sector.  Retailers in India face a shortage of trained, qualified and professional personnel for supporting the growth of this sector.  Niche and more personalised service catered by Mom & Pop stores.  E-retailing  E-retailing 1.9 The classification of retail industry The Retail industry can be classified into 2 major categories i.e. organised retailing and unorganised retailing. A comparison of key characteristics of organised and unorganised retails in India has been presented in table 1.4.  Organised retail: Organised retailers or traders are those, who have a license for trading activities and registered themselves to pay taxes are called as organised retailers.  Unorganised retail: Unauthorised small shops like Kirana stores, general stores, corner shops, etc. are called as unorganised retail. Table 1.4: A comparison of key characteristics of organised and unorganised retails in India Basis Organised Retailers Unorganised Retailers Average size (sq. ft.) 4000 300 Average Number of SKUs 5000 1200 Service format Self-service Over the counter Customer Expectations Predominantly experience based Predominantly need based Stocks turn High Low Profit Margin High Low Source: Adopted from (Madaan, 2012) 1.10 Retailing formats in India Table 1.4: A comparison of key characteristics of organised and unorganised retails in India 26 The Indian retailing sector is witnessing the drastic changes not only in the terms of growth, but also in the introduction of the different formats of retailing. In India, starting from Flea Markets to shopping malls can be seen. In between, there are various formats like Wholesale Clubs, Factory Outlets, Value Retailers, Off-Price Retailers, Vending, Mom-and-pop stores, E-trailer, MBO’s or Category Killers, Convenience Store, Discount Stores, Specialty Stores, Department Stores, Supermarkets and Hypermarkets. Figure 1.20 explicates the retailing formats in India and their basic facility. Figure 1.20: Retailing formats in India and their basic facility Source: Adopted from Dhillon et al. (2012) Figure 1.20: Retailing formats in India and their basic facility Source: Adopted from Dhillon et al. (2012) 27 27  Shopping malls: Today, shopping malls are largest form of organised retailing in India, which caters a conducive and ideal shopping experience with a wide range of products and services under the one roof. The malls are usually located in metros or near to urban outskirts. These shopping malls exist in the space ranging from 60,000 sq. ft. to 7,00,000 sq. ft. and above (Rahman, 2012). They cater not only products and services, but also provide entertainment, refreshment, spa, beauty salon, etc. The examples of this category are Pune Central, Shoppers Stop, Pantaloon, Inorbit (Levy et al., 2012) etc. 1.9 The classification of retail industry  Shopping malls: Today, shopping malls are largest form of organised retailing in India, which caters a conducive and ideal shopping experience with a wide range of products and services under the one roof. The malls are usually located in metros or near to urban outskirts. These shopping malls exist in the space ranging from 60,000 sq. ft. to 7,00,000 sq. ft. and above (Rahman, 2012). They cater not only products and services, but also provide entertainment, refreshment, spa, beauty salon, etc. The examples of this category are Pune Central, Shoppers Stop, Pantaloon, Inorbit (Levy et al., 2012) etc.  Hypermarkets: It ranges from 50000 to 100000 sq. ft. (Madaan, 2011). Hyper markets are also known as multi-department stores. It is one kind of superstore that combines supermarket as well as a department store (Hyper One Center, 2011). It is a very large retail facility that carries a numerous range of products at one place, including full lines of general merchandise and food & groceries; it may sell some services. When hypermarkets are planned, constructed, and executed correctly, a consumer can satisfy all of their routine weekly shopping needs at one time. Hypemarkets have an extensive width, medium depth and may offer private labels. They have stand-alone locations, malls (as anchor store also). Their pricing policy is lower than market prices and they provide self-service. Their promotional mix includes mass media and loyalty programmes. (Madaan, 2011). Examples are Big Bazaar, Shoppers Stop, Vishal Megamart, Hyper City, Star India Bazaar, Spencer’s Hyper, Reliance Mart etc. (Pradhan, 2009)  Supermarkets: The supermarkets contribute 30% to the total sale of food and grocery retail in India (Rahman, 2012). There are located near to the residential streets, self-service outlets which focus on food, grocery & household items and personal sales by offering the products and services to the varied needs of the shoppers. These supermarkets are categorised as mini supermarkets which typically exist in the space of 1,000 sq. ft. to 2,000 sq. ft. and large supermarkets, which range from of 3,500 sq. ft. to 5,000 sq. ft. Supermarkets have an extensive width, shallow depth and may offer private labels. They have a competitive pricing policy. Their level of service is average and offer self-service. A promotional mix of supermarkets includes mass media and loyalty programmes. (Madaan, 2011). 1.9 The classification of retail industry The example of this 28 category includes Reliance fresh, D’mart, More, Food bazaar, Trumart, Spencer’s Daily, Spencer’s Super (Pradhan, 2009), Reliance Super (Levy et al., 2012), etc.  Speciality stores: Speciality stores are the retail chains, which deals and focus on specific market segments (category) and recognised in their sector. The speciality stores offer general merchandise, single line and multiple lines. They have a narrow width, deep depth and offer usually high quality. They are located in business districts, strip shopping centres and malls. Their pricing policy is average to above average and also premium pricing. Discount stores provide personalised full services. The promotional mix of these stores is suggestive selling, catalogues and visual displays (Madaan, 2011). The example of this category includes Crossword (a book retailer), Kids Kemp (retailer of kids’ products), RPG's Music World, Planet M music chain of the Times Group, Fashion Station, aLL, Depot, Blue Sky, Home Stop, Mothercare, Desi café, Brio, Books & Beyond, RPG Cellucom, (Pradhan, 2009) Reliance Digital, Toy shop Hamleys (Levy et al., 2012), Reebok, etc.  Discount stores: Discount stores or factory outlets offer the products at discounted rates for selling in bulk to achieve economies of scale and to get rid of overstock. Their product categories comprise the variety of perishable as well as non-perishable goods. They have an extensive width, medium depth and offer medium quality. They are located in business districts, stand-alone and strip shopping centres. Discount stores have average services. Their promotional mix includes mass media and price-oriented promotions (Madaan, 2011). Examples of discount stores are Discount Circuit, Brand Factory, (Pradhan, 2009) Fashion Yatra, (Levy et al., 2012) Walmart, Kmart etc.  Discount stores: Discount stores or factory outlets offer the products at discounted rates for selling in bulk to achieve economies of scale and to get rid of overstock. Their product categories comprise the variety of perishable as well as non-perishable goods. They have an extensive width, medium depth and offer medium quality. They are located in business districts, stand-alone and strip shopping centres. Discount stores have average services. Their promotional mix includes mass media and price-oriented promotions (Madaan, 2011). Examples of discount stores are Discount Circuit, Brand Factory, (Pradhan, 2009) Fashion Yatra, (Levy et al., 2012) Walmart, Kmart etc.  Department stores: Large stores ranging from 20000-50000 sq. ft and caters to a variety of needs of the targeted customers. 1.9 The classification of retail industry Further, they have been classified into localised departments such as clothing, toys, home, groceries, general merchandise, apparel, accessories, gift items, sports, etc. They have an extensive width, deep depth and offer good to high quality (Madaan, 2011). Their location is mainly business districts, stand-alone, strip shopping centres. Pricing policy of department stores is above average. They offer assorted to full services. Their promotional mix includes suggestive selling, catalogues  Department stores: Large stores ranging from 20000-50000 sq. ft and caters to a variety of needs of the targeted customers. Further, they have been classified into localised departments such as clothing, toys, home, groceries, general merchandise, apparel, accessories, gift items, sports, etc. They have an extensive width, deep depth and offer good to high quality (Madaan, 2011). Their location is mainly business districts, stand-alone, strip shopping centres. Pricing policy of department stores is above average. They offer assorted to full services. Their promotional mix includes suggestive selling, catalogues 29 and mass media. Examples of department stores are Pantaloons, Westside, Piramyd, Shopper's Stop, (Pradhan, 2009) Sears, Bloomingdale's etc. and mass media. Examples of department stores are Pantaloons, Westside, Piramyd, Shopper's Stop, (Pradhan, 2009) Sears, Bloomingdale's etc.  Convenience store: Convenience stores are located near residential areas and considered as smaller stores of 400-2,000 sq. feet. These stores offer a limited range of convenience products with high-turnover and open 7 days a week as well as extend hours during the day. Slightly higher prices are charged due to the convenience premium. They offer food, grocery and household items (Madaan, 2011). These stores have a medium width of merchandise, shallow depth and average quality. Convenience stores have average and above average pricing policy. They offer highly personalised, average services and they also provide home delivery. A promotional mix of convenience stores includes personal rapport and average promotion. Examples are Spencer’s Express, (Pradhan, 2009) 7-Eleven, Circle K.  MBO’s or category killers: Multi Brand outlets, also known as Category Killers. They offer Specialty products. These stores have a narrow width, deep depth and offer average to high quality. They are located in stand-alone and strip shopping centres. These usually do well in busy market places and Metros. Category killers offer discounts and competitive pricing. Their level of service is assorted and promotional mix includes mass media. Examples of category killers are Toys R Us, PetSmart, etc. Figure 1.21: Brief summary of the formats in organised retailing y g Source: Adopted from D&B (2014) 1.9 The classification of retail industry  MBO’s or category killers: Multi Brand outlets, also known as Category Killers. They offer Specialty products. These stores have a narrow width, deep depth and offer average to high quality. They are located in stand-alone and strip shopping centres. These usually do well in busy market places and Metros. Category killers offer discounts and competitive pricing. Their level of service is assorted and promotional mix includes mass media. Examples of category killers are Toys R Us, PetSmart, etc.  E-trailer (E-tailers): Retailers are offering online products and services for buying and selling (Sikri and Wadhwa, 2012). E-Tailing is popular as the retail E-Commerce that comprises sales as well as purchases of products done through electronic mode between retailers and their customers. The spread of the internet and falling prices of desktops, had enabled many people to turn to E-Tailing. Advances in telecommunications like reliable broadband and the advent of smartphones have made mobile commerce called by the acronym M-Commerce, a fast growing segment. It offers several advantages to the customers in the form of product choice, 24X7 shopping experience, saving time and cost of travelling. Customers feel as a savvy consumer for making a more intelligent purchase by accessing their online catalogues. Customers may purchase goods with a single click of a mouse without travelling. Customers also benefited enormously by getting the products at a lower price in comparison to the conventional stores. (Prasad, 2013). Examples of E-tailers  E-trailer (E-tailers): Retailers are offering online products and services for buying and selling (Sikri and Wadhwa, 2012). E-Tailing is popular as the retail E-Commerce that comprises sales as well as purchases of products done through electronic mode between retailers and their customers. The spread of the internet and falling prices of desktops, had enabled many people to turn to E-Tailing. Advances in telecommunications like reliable broadband and the advent of smartphones have made mobile commerce called by the acronym M-Commerce, a fast growing segment. It offers several advantages to the customers in the form of product choice, 24X7 shopping experience, saving time and cost of travelling. Customers feel as a savvy consumer for making a more intelligent purchase by accessing their online catalogues. Customers may purchase goods with a single click of a mouse without travelling. Customers also benefited enormously by getting the products at a lower price in comparison to the conventional stores. (Prasad, 2013). 1.9 The classification of retail industry Examples of E-tailers 30 Snapdeal.com, Amazon.com etc. Snapdeal.com, Amazon.com etc. Figure 1.21: Brief summary of the formats in organised retailing Source: Adopted from D&B (2014)  Mom-and-pop stores: These are family owned businesses which cater to small sections and are individually handled retail stores which provide a p ( )  Mom-and-pop stores: These are family owned businesses which cater to small sections and are individually handled retail stores which provide a personal touch (Sikri and Wadhwa, 2012). Mom and pop stores are businesses  Mom-and-pop stores: These are family owned businesses which cater to small sections and are individually handled retail stores which provide a personal touch (Sikri and Wadhwa, 2012). Mom and pop stores are businesses 31 that are owned and operated in a single location. Rather than being part of a national chain, these stores offer a distinct shopping experience to the consumers, who want to deal with businesses that are local to a particular city or town and at those places where business owners have created the community. While the proliferations of huge retail chains have considerably decreased the market for independent retail stores. Still, numerous locally owned businesses are continuing and thriving in the competitive economy. An example of this type of store is local general stores. A small, usually family- owned, independently controlled, and operated the store, which has fewer numbers of employees as well as smaller sales volume turnover, and is typically not franchised and open in a single location for business operations.  Vending: Vending is a new entry, in the retail sector, where snacks, beverages and other small items can be purchased via vending machine (Sikri and Wadhwa, 2012). A vending machine is a machine that dispenses items after the customer inserts currency or credit into the machine. E.g. the ATM.  Vending: Vending is a new entry, in the retail sector, where snacks, beverages and other small items can be purchased via vending machine (Sikri and Wadhwa, 2012). A vending machine is a machine that dispenses items after the customer inserts currency or credit into the machine. E.g. the ATM.  Off-price retailers: The off-price retailer provides general merchandise, inconsistent, out of fashion, odd sizes, etc. These stores have a medium width and shallow depth. They are located in stand-alone, shopping strip and at any low rent location. Their pricing policy is very low and offers minimal services & self-service. 1.9 The classification of retail industry They focus on mass media as a promotional mix (Madaan, 2011) Examples of off-price retailers are TJ Maxx, Marshalls, Ross stores, Big Lots, Stein mart etc.  Off-price retailers: The off-price retailer provides general merchandise, inconsistent, out of fashion, odd sizes, etc. These stores have a medium width and shallow depth. They are located in stand-alone, shopping strip and at any low rent location. Their pricing policy is very low and offers minimal services & self-service. They focus on mass media as a promotional mix (Madaan, 2011) Examples of off-price retailers are TJ Maxx, Marshalls, Ross stores, Big Lots, Stein mart etc.  Value retailers: Value retailers provide general merchandise and Multi- brands. And these stores have a medium width and shallow depth. They are located in stand-alone, shopping strips and at any low rent location. Their pricing policy is highly discounted prices. The level of services of such stores is average and the promotional mix is mass media (Madaan, 2011). Indian example of value retailers is a Pantaloon Brand Factory and U.S.A. examples of value retailers are Dollar Tree and Dollar General.  Value retailers: Value retailers provide general merchandise and Multi- brands. And these stores have a medium width and shallow depth. They are located in stand-alone, shopping strips and at any low rent location. Their pricing policy is highly discounted prices. The level of services of such stores is average and the promotional mix is mass media (Madaan, 2011). Indian example of value retailers is a Pantaloon Brand Factory and U.S.A. examples of value retailers are Dollar Tree and Dollar General.  Factory outlets: Factory outlets offer general merchandise, Self- Manufactured items, and are inconsistent. These stores have a moderate width and shallow depth. Such stores are located in a stand-alone location, Factory premises and any low rent location. They provide highly discounted prices and offer self-service. A promotional mix of such stores is low promotions  Factory outlets: Factory outlets offer general merchandise, Self- Manufactured items, and are inconsistent. These stores have a moderate width and shallow depth. Such stores are located in a stand-alone location, Factory premises and any low rent location. They provide highly discounted prices and offer self-service. A promotional mix of such stores is low promotions 32 (Madaan, 2011). Examples of these include Pantaloon Factory Outlets, Lei’s factory Outlets, etc. (Madaan, 2011). 1.11 Major retailers in India The Indian retail industry witnessed a great expansion in the period of 2005- 10. And a large number of corporate houses entered into this market with substantial investment in food and general merchandise segment. Figure 1.22 presents the brief summary of Indian Retail Groups and Area of their Business. The industry is being flourished by the presence of the following players. 1.9 The classification of retail industry Examples of these include Pantaloon Factory Outlets, Lei’s factory Outlets, etc.  Wholesale clubs: Nature of merchandise of wholesale clubs is general merchandise, food items and is inconsistent. These stores have a moderate width and shallow depth. Such stores are located in stand-alone, out of town and industrial areas. They offer very low prices and level of services is low. The promotional mix of these stores is low promotions and direct mail (Madaan, 2011). Examples of wholesale clubs are Costco Wholesale Club and Sam Club.  Flea markets: Flea market offers some general merchandise items and has extensive width but very shallow depth. These are located in open surroundings, low rent, and semi-urban locations. Pricing policy includes bargaining and low prices. Their level of service is low. A promotional mix of flea markets is limited and point of sales promotion (Madaan, 2011). Examples of flea markets are Janpath in Delhi, Crawford and Zaveri Bazaar in Mumbai etc. 1.11.1 Aditya Birla Retail Limited Aditya Birla Retail Limited is US$ 28 billion Corporation. In 2007, the company entered into food and grocery retail sector by acquiring a south India-based supermarket chain ‘Trinethra’. By the time Company expanded its business with the brand named ‘More’ with 2 formats Supermarket and Hypermarket. Supermarket ‘More’ conveniently located in neighbourhoods and it’s catering to the daily, weekly and monthly needs of consumers. Their product includes a wide range of fresh fruits & vegetables, home care, personal care, general merchandise & a basic range of apparels. Presently Aditya Birla Retail Ltd have 640 ‘More’ supermarkets across the country. The Hypermarket ‘More’ Megastore is the complete shopping experience 33 for its customers. Hypermarket ‘More’ Megastore not only offers fruits & vegetables, groceries and FMCG products, but also it offers general merchandise, apparels and CDIT. At present 8 Hypermarket ‘More’ Megastores are operating in New Delhi, Mumbai, Bengaluru, Hyderabad, Indore, Aurangabad & Vadodara. Aditya Birla Retail Ltd has more than 11,000 employees. ‘More’ stores have over 1.6 million members of its loyalty programmes. Aditya Birla retail Ltd has private label brands such as Kitchen’s Promise, Germex, Feasters, Enriche and Best of India. The objective of private label brands is to offer quality products with attractive prices to its customers. Aditya Birla Retail was awarded with ‘Retail Best Employer of the Year’ at Reid & Taylor awards for Retail Excellence by the global jury of the Asia retail Congress 2009. The company was also recognised for impactful Retail & Visual Merchandising at the same forum. The Company was awarded with the ‘Most Admired Retailer of the Year’ in the Smart Strategy category at the Images Retail Awards 2009 at the India Retail Forum, Mumbai. 1.11.2 Bharti Retail Ltd. Bharti Retail Ltd. is a wholly owned subsidiary of Bharti Enterprises. It operates a chain of multiple format stores in India. The neighbourhood formats of the company’s stores are named as ‘Easyday’ and the compact Hypermarket named as ‘Easyday Market’. At present Company is more focusing on food economic sectors with a joint partnership with agriculture company Field Fresh. Field Fresh Foods Private Ltd is a joint venture of DMPL India Ltd (a subsidiary of Del Monte Pacific Ltd.) and Bharti Enterprises. The Company offers processed foods & beverages and fruits & vegetables in the domestic as well as international markets. Presently neighbourhood ‘Easyday’ has 70 stores and hypermarket ‘Easyday Market’ has 6 mid-size compact stores. In order to serve farmers, small scale manufacturers and retailers, Bharti Enterprises and Walmart Stores Inc signed a deal for wholesale, B2B, cash & carry business and back-end supply chain management. They ventured their first B2B Best Price Modern Wholesale cash-and-carry store in May 2009 in Amritsar, Punjab, as of now they have their stores in Jalandhar, Kota, Zirakpur. Their cash & carry stores are about 50,000 and 10,000 square feet and offer a wider variety of products, i.e. fruits & 34 vegetables, fresh, frozen and chilled foods, dry groceries, hotel & restaurant supplies, personal & home care, office supplies, clothing and other general merchandise items. These stores offer a competitive wholesale price and allow retailers to lower the cost of business. This venture of Bharti Enterprises and Walmart Stores Inc is planning to launch their 10 to 15 wholesale cash-and-carry points and will recruit 6,000 to 7,000 people in next 3 coming years. 1.11.3 Reliance Retail Limited Reliance Retail Limited (RRL) is a subsidiary of RIL. Since 2006, Reliance Retail Limited is catering to millions of customers and thousands of farmers & vendors. Its core strategy is backwards integration; it has made huge progress towards building entire value chain starting from farmers to the end customers. Over 3 years of operation, the company has now expanded its presence in more than 85 cities across 14 states in India. It has more than 1,000 stores in the country. The company operates with several ‘Value’ and ‘Speciality’ formats. The ‘value’ formats that Reliance Retail Ltd operates are- Reliance Fresh - a neighbourhood concept, Reliance Mart - supermarket concept, Reliance Super - a mini-mart concept. The ‘value’ formats offer a wide range and assortment of products required for daily household needs. Whereas, the ‘specialty’ formats are: Reliance Trends - an apparel & accessories concept, Reliance Digital - a consumer durables & information technology concept, iStore by Reliance Digital - an exclusive Apple products concept, Reliance Wellness - a health, wellness & beauty concept, Reliance Footprint - a footwear concept, Reliance TimeOut - a books, music & entertainment concept, Reliance Jewels - a jewellery concept, Reliance Living - a furniture, homeware, furnishings, and modular kitchens concept and Reliance AutoZone - an automotive products & services concept. Reliance Retail Ltd expanded the stores network and operating through strategic partnerships with World-Class companies like as Pearl Europe and Marks & Spencer. The company has signed a deal with the Asics Corporation of Japan for distribution arrangement to market Asics brands of shoes & accessories in India. Reliance Retail Ltd. has also launched its flagship store with a franchise arrangement with Hamleys and also planning to enhance their network in a succeeding year. The 35 company has also expanded its presence in B2B office supplies through its joint venture with Office Depot. 1.11.4 Pantaloon Retail Pantaloon Retail is the company of Future Group. Pantaloon Retail (India) Ltd is the India’s leading retailer, which operates multiple retail formats both in the value & lifestyle segment. It was established in the year 1993 and Company headquarter is in Mumbai. It is operating over 16 million square feet of retail space and has over 1,000 stores across 73 cities and 65 rural locations in India, and employs over 30,000 people. The leading formats of the Company are as follows:  Entertainment: Bowling Co.  Entertainment: Bowling Co.  Entertainment: Bowling Co.  Shoes: Shoe Factory  Health & Beauty Care: Star, Sitara  Food & Grocery: Big Bazaar, Food Bazaar  E-tailing: Futurebazaar.com  E-tailing: Futurebazaar.com  Home Solutions: Furniture Bazaar, Hometown, Collection-i  Books, Music & Gifts: Depot  Rural Retail chain: Aadhaar  Consumer Electronics: e-zone Future Value Retail Limited is a fully owned subsidiary of Pantaloon Retail (India) Limited. This company was established by considering the size and growth of the value of the company, which further led by Big Bazaar and Food Bazaar. The company is running its business with more than 120 Big Bazaar stores and 120 Food Bazaar stores in 70 different cities of the country and occupies over 6 million square foot retail space. A subsidiary company, Home Solutions Retail (India) Limited is operating Home Town, a large-format home solutions, store, collection, selling home furniture products and eZone focused on catering to the consumer electronics segment. 1.11.5 Vishal Retail Ltd. Vishal retail Ltd is the company of Vishal Group, which operates its Hypermarkets with an average area of 25,000 to 30,000 square feet with a chain of 36 172 fully integrated stores which spreads over the area of more than 24,00,000 square feet in around 110 cities across India in 24 states. It is the fastest growing retail group in India. Its stores cater to almost all price ranges. The stores have more than 7000 product range which caters to all household needs and offered at one place. The group is maintaining the highest standards in quality and design, these stores have introduced to offer the finest fashion garments at down-to-earth price structure. The store is named as Vishal Megamart, the stores are the best place for bargainers and fashion conscious people. The turnover of the company for the year 2009-10 was 1105 crore. 1.11.6 Tata Group The Tata group is also a leading player of Indian retail industry and business with its subsidiary Trent that has Westside and Star Bazaar for the business. It was launched in 1998. In 2005, it has acquired ‘Landmark’ which is the largest book and music retailer in the country. Trent owns over 4 lakh square foot retail space across the country. It has main categories like apparel, speciality books and music. Its formats include supermarkets and hypermarkets. 1.11.7 Spencer's Retail Limited Spencer’s Retail Limited is the company of RPG Enterprises. It is the largest & growing fastest multi-format retailer in India and it was established in 1988. Its footage is 1 million square feet approximately and has 220 stores, including 30 large format stores across 35 cities in India. It offers the product range in the following categories: fruit and vegetables, food and grocery, personal care, home and office essentials, garments, fashion accessories and electrical & electronics. It was established in 1996 and now become a popular destination for shoppers in India. It is operating with two formats, i.e. hypermarkets and a convenient store for catering the shopping needs consumers. Spencer hyper stores are of more than 15,000 square feet in size and offer everything under the same roof. Their product range comprises the categories like fruits & vegetables, meat, chicken, fish, groceries, bakery, processed foods, chilled and frozen foods, garments and fashion accessories, office stationery, home decor and 37 needs, soft toys and electronics & electrical products. On an average a Spencer’s hyper store stocks 70,000 Stock Keeping Units (SKU’s) across the 35,000 items. Spencer's stores are neighbourhood stores ranging from 1500 square feet to 15000 square feet. The stores offer the widest range and assortment in fruit and vegetables, non-food, staples, FMCG food and frozen foods and cater to the daily and weekly top-up shopping needs of the consumer. Music World is India's largest chain of music store retailing and it offers the widest range of music & home video products (International and Indian). The Music World's product portfolio includes audio CDs, DVDs and VCDs, CD-ROMs, gaming consoles & software of all the leading brands and other music accessories. The company is a key player in the home video market. Music World has successfully formed into high end 'personal audio' gadgets of several top brands. The company also offers home theatre systems, speakers, headphones and DVD players. 1.11.8 Piramal Group Piramyd Retail belongs to Piramal Group. In 1999, it launched its apparel business under the brand name, i.e. Piramyd Retail and Merchandising Pvt. Ltd. (PRMPL). While its food; home and personal care businesses were grouped under Crossroads Shoppertainment Pvt. Ltd. (CSPL). When the apparel and food businesses reached a critical mass, the management merged these 2 into a Piramyd Retail Ltd. Pyramid also offers merchandise with its smaller format of stores called TruMart. It caters to food & personal care products. Piramyd Retail at present is operating with 5 Megastores as well as 8 TruMart stores in the Maharashtra state of India. 1.11.9 Madura Garments- part of AB Nuvo It was established in 1988 and deals with apparels. It offers Louis Philippe, Van Heusen, Allen Solly, SF Jeans and Peter England. It operates with Specialty and Lifestyle formats. Planet Fashion has more than 50 stores and Trouser Town has more than 9 stores. 1.11.10 K Raheja Group 38 38 Shoppers’ Stop, promoted by the real estate group K Raheja, was one of the first retailers to have set up a large retail outlet in New Delhi with international ambience and it was established in the year 1991. Shopper’s Stop Ltd has made its a presence in the country and occupies retail space over 7 lakh square feet and stocks over 200 brands of garments and accessories. The stores are located all over India in the cities like Delhi, Bangalore, Jaipur, Pune, Mumbai, Chennai, Kolkata, Gurgaon, Ghaziabad & Hyderabad. Shopper’s Stop Ltd. has Home Stop, Mothercare, Crosswords, Desicaffe, Brio etc. and its formats include Lifestyle, Department stores and Specialty. 1.11.11 Vivek Group It was established in 1965 and has main categories like Food, Beauty, Specialty-Electronics. It has stores like Vivek, Jaisons and Premier. Vivek has more than 23 stores, Jaisons has more than 26 stores and Premier has more than 3 stores. 1.11.12 Globus It was established in 1998 and deals in apparel. It has stand-alone formats named as Globus and F21. 1.11.13 Nilgiri’s Ltd. It was established in 1904 which has main categories like Food, Specialty- Bakery Products. It has a supermarket format which has more than 21 stores. 1.11.16 Bata India Ltd. It was established in 1931 and has categories named as Footwear and Accessories. It has stand-alone stores named as Bata. It has more than 1,100 stores. 1.11.15 Provogue Ltd. It was established in 1997 and deals with Apparel and Footwear. It has standalone stores named as Provogue and Prozone. It has more than 139 stores. 1.11.14 Trinthera Super Retail Ltd. It was established in 1986 and has categories like Food and Beauty products. It has stores named as Trinthera Quick Shop and Trinthera Super Retail Ltd. Its formats include Convenience, Super & Hypermarkets. It has more than 150 stores. 39 39 Indian Retail Groups Formats / Area of their Business Future Group Big Bazaar, Food Bazaar, Pantaloons, Central, Fashion Station, Brand Factory, Depot, aLL, E-Zone Reliance ADAG Retail Reliance World Lifestyle International Lifestyle, Home Centre, Max, Fun City and International Franchise Brand Stores. Aditya Birla Group "More" Outlets, More Hyper Vishal Retail Group Vishal Mega Mart Fabindia Textiles, Home Furnishings, Handloom Apparel, Jewelry The Tata Group Westside, Star Bazaar, Steel Junction, Landmark, Titan Industries with World of Titans Showrooms, Tanishq Outlets, Croma Next Retail India Ltd. Consumer Electronics Trinethra Fabmall Super Market Chain and Fabcity Hyper Market Chain Rei Agro Ltd. Retail 6TEN and 6TEN Kirana Stores Reliance Retail Reliance MART, Reliance SUPER, Reliance FRESH, Reliance Footprint, Reliance Living, Reliance Digital, Reliance Jewelry, Reliance Trends, Reliance AutoZone, iStore K Raheja Corp Group Shoppers Stop, Crossword, Hyper City, Inorbit Mall Pyramid Retail Pyramid Megastore, TruMart BPCL In & Out Raymond Ltd. Textiles, The Raymond Shop, Park Avenue, Park Avenue Woman, Parx, Colourplus, Neck Ties & More, Shirts & More German Metro Cash & Carry Metro Cash & Carry PGC Retail T-Mart India, Switcher, Respect India, Grand India Bazaar Nilgiri’s Nilgiris’ Supermarket Chain Marks & Spencer Clothing, Lifestyle Products Shoprite Holdings Shoprite Hyper Kapas Cotton Garment Outlets RP - Sanjiv Goenka Group Retail Spencer’s Hyper, Spencer's Daily, Music World, Au Bon Pain (International Bakery Cafeteria), Beverly Hills Polo Club Vivek Limited Retail Formats Viveks, Jainsons, Viveks Service Centre, Viveks Safe Deposit Lockers Paritala Stores Bazar Honey Shine Stores Figure 1.22: Indian retail groups and area of their business 40 1.11.17 Archies Ltd. It was established in 1979 and it deals in Specialty- Cards and Gifts. It has stand-alone stores named as Archies and Stupid Cupid. It has more than 73 stores. 41 41 Review of Literature  Introduction  Retailing strategies  Customer behaviour  Customer service  Customer satisfaction Introduction This chapter comprises the relevant literature review pertained to the research topic. This chapter consists the review of literature pertained to retailing strategies, customer behaviour, customer services and customer satisfaction. The review of relevant literature available assisted to understand the studies, which have been completed in the field. Further, review of relevant literature played a vital role to understand the problems, research gaps, forming the hypothesis, and preparation of the questionnaire. 2.1 Retailing strategies Muzondo and Mutandwa (2011) assessed the significance of seven Ps of marketing in influencing consumer choice of store for main grocery shopping in a hyperinflationary economy. The results of factor analysis confirmed the significance of six viz. product, process, price, place, physical evidence, and promotion of the 7Ps of marketing. Tyagi (2008) expressed that retailers looking to adopt a customer-centric strategy to succeed in the hyper-competitive environment, it is important to understand the impact of such a strategy on retailer’s business functions, on merchandising in terms of sharper assortment and improved presentation and on supply chain needs. This would ensure the alignment of customer-centricity with the organisation’s overall direction and growth of revenues and profitability while increasing customer satisfaction levels. Mai et al. (2011) vocalised that product managers need to better understand the relationships between primary product network size and the complementary product sales. Further, retail marketing strategies should be adjusted accordingly with the different purchase frequency segments of customers. 42 42 Asikhia (2010) found that customer orientation moderated by the marketing information system and managerial attitude is an efficient strategy of providing and sustaining customer advantage for small and medium scale businesses in an ever changing business environment and suggest that managers of small and medium scale businesses should allow their businesses to emerge out of a customers’ need. Omholt (2002) argued that the right horizontal strategy is the one of the critical ways to achieve the superior competitive position in retail organisations. With increased competition from neighbouring and out-of-town centers, increased consumer mobility, and a more diverse population with regard to household structures and lifestyle, sharing a specified location may not be enough for a group of outlets or centers to sustain a competitive advantage. Standop and Grunwald (2009) in an empirical research stated that the compensation, communication and logistics are the elements of product crises response strategies in the retail industry. These elements can influence the evaluations of customers positively; either directly or indirectly which are affected by product problems thus enhances the brand equity. Evans et al. (2008) in the light of recent changes in the international environment, consider whether the drivers of, and impediments to, retail internationalisation and the business strategy adopted has changed. Results suggested that meanwhile a variety of factors drive retail internationalisation, however, the profit growth is the most dominant factor. 2.1 Retailing strategies Sun and Ma (2009) unpacked that the foreign capital Retail business constructs the trade integration management system in the Chinese retail industry, implements the global trade strategy. In China, the majorities of successful foreign retail businessmen choose the latest information technology, implements customer loyalty strategy, retail sales business strategy, implements a localisation strategy in the investment management processes, enhances the competitive advantage and expands the market share. 43 43 43 Medina and Rufín (2009) focused on the effects of three different strategic orientations have on firm performance in the specific domain of retailing. Their results show that market driving proved to be a strong predictor of performance in addition to innovation acting as a mediator between strategic orientations in retailers and business performance. Among the strategic orientations developed by retail firms, market driving proved to be better both in its direct influence on performance. Rani (2012) examined the present retail strategies of store services and store choice and the effect of store services and store choice on Customer satisfaction at MORE and FOODWORLD. Results of study shown that the MORE retail store should concentrate more on improving the store services and store choice Retail strategies for better customer satisfaction and FOOD WORLD should concentrate on some more retail strategies on store choice. Padilla and Eastlick (2009) conducted the exploratory examination of management strategies and urban retail marketing in 6 US cities having the reputation of central business districts that are either flourishing or developing. They classified strategies into three urban retailing and five economic revival themes. The specified strategies vary and are dependent on flourished cities or developing central business districts. Vetrivel (2011) advised that retailers should create a good rapport with their customers for increasing further strength of CRM system among Retailers and the Customers. To enhance profitability and customers’ satisfaction in Marketing sector, the retailers must focus on implementing Customer Relationship Management strategies like a customer database, customer retention, data mining and tracking system that aim to seek, gather and store right information. Coca-Stefaniak et al. (2010) found the key pillars of the “localization” strategic marketing approaches chase by small retailers. They are- meaningful interpersonal relationships between customers and shopkeepers, place attractiveness, community embeddedness, word-of-mouth customer-to-customer marketing and ultimately service beyond simple advice. 2.1 Retailing strategies 44 Cacho-Elizondo and Loussaïef (2010) argued that if retailers want to be identified as socially responsible, they must create the “right” associations to reshape their new brand image. By keeping this in mind, creating an awareness of CSR efforts by the firm is an essential attribute for the assessment of their brand image. Developing a CSR facet in their image-building programs, retailers will meet the expectations of an increasing number of socially-conscious consumers in a better way. Altintas et al. (2010) found that the strategic objective factors have a great effect on the competitive advantage of private label manufacturers in Turkey. They also found that three strategic objective factors have an effect on competitive advantage. These factors are product selling control, production efficiency, and market embeddedness. Datta et al. (2006) examined the shopper behaviour regarding the impact of Iceland Foods Plc store design refit strategy. The findings pointed to the relative importance of the stages of the consumer decision-making process, and also to the relative importance of the factors which influences the consumer decision-making process. Overall, service quality was found to be important to all and the refit strategy, according to the customers was indeed a success. Ilonen et al. (2011) explained the different roles and implications of manufacturer’s branded retail operations of its international strategy. They have examined the roles through two dimensions, the role of branded retail as a sales channel and brand strength. They have also examined the two approaches to implementing branded retail strategy internationally, which can be and frequently overlapping. Their emphasis on branding appears to be very important in both the approaches. Ranganath et al. (2011) said that the technology enables all the managers of the retail company to access important information almost up to the minute and consequently to take timely and quick decisions. Modern technologies are transforming the retail environment and it has also helped a lot many companies to expand their operations. Retail organisations are introducing online retailing at a rapid 45 rate by realising its growing importance of the sales of merchandise through these new distribution channels. Chen and Huddleston (2009) studied the effect of 4 promotional strategies on the purchase intentions of the students. A significant and positive relationship was found between attractiveness and trustworthiness of the sports celebrities and customers' purchase intention in a campus convenience store. 2.1 Retailing strategies To gain purchase intention of customers less expensive promotional strategies are equally effective than celebrity endorsements. Joo et al. (2012) found that how bidding strategies of successful bidders have influenced the savings which they derive from ‘Name Your Own Price’ (NYOP) retailer in relation to buying the same merchandise from a retailer who post prices. They explained in the case of utilising bidding data for hotel room purchases the consumer savings rate depends positively on consumer decision to bargain and on the shape of bid function. Bargainers who employ a constant bid increment strategy and a decreasing bid increment strategy save more in relation to non-bargainers, while those who employ an increasing bid increment strategy fare are desirable. Diona and Arnould (2011) expressed that luxury retail strategy differs from other retail strategies not merely in distinctive formulations of the 4 Ps to customer distinction. Alternatively, it increasingly stands or falls on the justification of a charismatic, creative director, the director offers aesthetic brand ideology. Luxury retail relies on the principles of magic to assemble the charismatic persona of the creative director, it is an art and to diffuse his aesthetic ideology to the brand. Further, they explain that this strategy appears to be to some extent a response to justification crises stimulate by recent strategic extensions of luxury brands into mass marketing. Shankar et al. (2011) written that shopper marketing is the key to managerial practice among manufacturers and retailers, which are desperately accepting innovations in the different aspects of shopper market. Shopper marketing mentions as the planning & execution of all marketing activities which influence a shopper from purchase to post purchase, from the time when a shopper gets the first 46 motivation to purchase, consumption, repurchase and suggestions. The aim of shopper marketing is to enable a win-win-win solution for the manufacturer-retailer-shopper. Barone et al. (2007) examined how retailer-cause fit affects consumer evaluations of retailer’s cause-related marketing strategies. The results suggested that the effects of retailer-cause fit are monitored by consumer perceptions of the retailer’s motive for engaging in cause-related marketing, by the influencing effects associated with the two moderators, as well as by the affinity that consumers hold for the social cause component of the campaign. Gauri et al. (2008) said that two powerful and highly strategic tools that retailers possess are store format decisions and pricing. 2.1 Retailing strategies By using a unique data set which covers all grocery retailers in three states and applying a multinomial logit model to study the determinants of format, price and combination strategies of retailers; it was found out that a consideration of the format or pricing strategy in isolation fails to depict a complete picture, the strategic implications change in important ways when they study format and price strategies in combination; although, some combinations are more similar than others. Dharni and Chawla (2012) made an attempt to explore the impact of IT-related benefits as well as problems in the performance of IT in the retail sector. Results indicated that IT related problems were having a significant impact on the performance of IT in retail. Results also indicate that IT in retail is moving towards the commoditisation phase. Baskaran (2011) identified that customer experience management is an important strategy in retailing and suggests the several ways (e.g., brand, price, promotion, location, advertising, supply chain management, packaging & labeling, atmosphere, and service mix) to provide superior customer experience which must result in higher customer satisfaction, larger wallet shares, more frequent shopping visits and higher profits. Burt et al. (2011) assessed the degree of standardisation within three countries (The UK, China and Sweden) for distinguishable marketing mix activities: location & 47 merchandise, service environment and selling, market communication and store format. These three countries (Sweden, the UK and China) represent different cultural settings and are markets in which IKEA has been operating for a long time. They recommend that while IKEA operates a standardised concept, therefore, degrees of adaptation can be observed in customer facing elements, and in the supporting ‘back office’ processes which support these elements. Etgarand Rachman-Moore (2011) examined the links between the national cultural values that characterise an origin country of retailers and the adoption of their strategy between generalist versus specialist format. Their results indicate that the probability of being a generalist retailer is higher for retailers originating from countries characterised by collectivism, a low power distance, a present orientation and high uncertainty avoidance. Whereas, the chance of occurrence of being a specialist is higher for retailers having a future orientation, low uncertainty avoidance, a greater power distance and originating from countries characterised by individualism. Sheng and Pan (2009) examined how a new brand can take an advantage for bundling with a strong brand. 2.1 Retailing strategies The results of their two studies indicate that consumers’ quality perception of a new brand will be affected by the bundle partner brand image, and the effect is retarded by the bundle forms and the complementarily of bundle components. Therefore, retailers can bundle an unknown brand with a strong brand as a new product introduction strategy. Yan and Pei (2009) developed a model which focuses on the strategic role played by the retail services in a dual-channel competitive market. In their research, results suggest that the improved retail services effectively alleviate the channel competition and conflict and improve the supply chain performance in a competitive market. Campo and Gijsbrechts (2004) examined the conditions for such micro- marketing to be beneficial, and in particular - how these benefits depend on store format. Their results suggest that micro-marketing is beneficial in both types of formats. Still, the appropriate way of localising space allocation patterns is dependent 48 on specific format. Whereas supermarkets should mainly adjust the space shares of food (core) categories and hypermarkets should primarily adapt the space shares of non-food (non-core) categories of local market conditions. Tanabe et al. (2004) considered the relative impact of management’s strategic planning practices and economic changes in company performance. The results derived suggested that variables representing changes in economic conditions and strategic planning were both statistically significant when they were used to explain performance differences. Differences in company management practices were found to play a more important role than changes in the environment. 2.2 Customer behaviour Smith (2006) talked about the large and growing role of vertically integrated supermarket chains, which has raised a buyer power concern. The reason behind it is the potential to harm other suppliers or consumers and retailers. The exercise of buyer power by the chains in relation to their suppliers may indirectly impact adversely on retail competition, to a great extent as barriers to entry into grocery retailing are relatively high. Watchravesringkan and Punyapiroje (2011) contributed towards the understanding of consumer attitude and marketing efforts of hypermarket retailers regarding marketing practices of retailers: Big C, Tesco-Lotus, and Carrefour in Thailand. The findings have depicted that consumer attitude is different towards fair price, retail services and positive advertising; Thai consumers have a similar attitude towards product quality across samples and business provisions. Ruiz and Descals (2008) assessed the impact of sales on temporary retail price discounts on all brands within a merchandise subcategory and other subcategories that are slightly different in terms of tests as well as the composition. The results indicate that the temporary reduction in prices increases the sales of promoting brands, especially in weekends. Omar and Abisoye (2008) examined the role price as a determining factor in consumer patronage of grocery retail stores in the United Kingdom. They suggest that 49 price cuts affect consumer store choice. Price awareness has a positive impact on retail stores which implement low-cost strategies. Whereas, price/quality plan and status sensitivity plans tend to have a positive impact on retail stores that implement high price strategies. They recommend it is good to communicate store pricing policy to the target consumers. Thiruvenkadam and Panchanatham (2011) explored the demographic factors that influence the decisions of the consumers on retail store selection in Chennai. In their opinion the demographic factors like gender, education, income and age have a major role in retail store selection. Shopper groups differ in the selection of a store based on their importance for factors of a store. Lonarkar and Gore (2012) concluded that there is a significant change in the shopping behaviour of the people of the II tier city. This change is more in income group, specific age group, and service group some variations are found when we compare the small and mega retail outlets in case of a distance of outlets and residence of customers. The reason behind the change is the time-saving shopping with satisfactory services. 2.2 Customer behaviour Rajagopal (2011) examined the impact of radio ads on urban customers in relation to buying behaviour in retail. His result revealed that shopping behaviour of the urban consumer is highly influenced by the physical, cognitive and economic variables at retail stores in reaction towards radio advertisements. Radio advertisement listeners get attracted towards promotional messages regarding sales of products and entertaining advertisements which have a quick reaction to the department stores and supermarkets. Shaikh et al. (2012) noticed that brand has a specific impact on consumer psychology. Most of the consumers purchase from supermarket due to its brand name. The rating of a brand name of the supermarket is a compelling force in buying decision and it has co-occurrence with marital status, qualification, gender and occupation of respondents. Quality is a fundamental point to justify the purchase of a particular product. It has also been noticed that the respondents had a price sensitive approach while purchasing products from supermarkets. 50 Prasad and Aryasri (2011) made a detailed study and unpack that shoppers' age, gender, monthly household income, distance travelled to store, occupation, family size and education have a significant relation with retail format choice decisions. The findings from shoppers' psychographic dimensions such as shopping orientations, lifestyle factors and values results in the segmentation of grocery and food retail consumers into autonomous, hedonic, conventional, socialisation and utilitarian type. Rao and Manikyam (2012) gave the customers’ perceptions of the experience related to four areas which are: shopping experience, buying experience, service experience and relationship experience and established the fact that the small scale retailers are providing valued experiences to the customers. The customer experience management requires the efforts of the management for improvements qualitatively and organised efforts of retailers. Khraim et al. (2011) pointed out that retail store attributes are influenced by consumer's religiosity. Consumer religiosity is an important indicator of retail store attribute selection. Results indicate that there was a difference between low, moderate and high consumers' religiosity in evaluating the importance for all retail store factors except for the service factor, while there was no difference between moderate and high levels; however there was a difference between them and low level. Richards et al. (2012) expressed that price promotions are becoming an increasingly important method of managing consumer demand for fresh produce items. 2.2 Customer behaviour They applied a multiple discrete/continuous model of fresh produce demand to study the impact of price promotion on retail apple sales. Their findings showed that the brand switching/category incidence effect of promotion is closer to 65/35 than the more usual 80/20 rule when the nature of the decision is appropriately taken into consideration. González-Benito and Martos-Partal (2012) analysed that recent information acquired by experimentation towards the relationship between store loyalty and store brand purchase recommends a non-monotonic relationship: positive up to a certain 51 store brand level of consumption, later it becomes negative. To investigate this idea further, they analyse the role of the retailer's competitive positioning, and specifically it’s the product category, and price positioning. The more prices oriented retailer's positioning leads to the more favourable relationship between store brand consumption and store loyalty. The relationship between store loyalty and store brand consumption seems to differ across product categories as a result of several factors including perceived risk. White et al. (2012) examined the effects of fairness perceptions of these self- service technology “push” policies on relationships between established antecedents of self-service technology adoption and customer behavioural intentions toward the provider in a retail context. Their results suggested fairness perceptions exert a significant influence on the relationships between these antecedents and customer patronage, negative word of mouth intentions and future spending. Murthi and Rao (2012) discovered that between 40 to 50% of the purchases are made by consumers having expectations of prices rather than posted prices. Consumers who have price expectations may be thought of as being not aware of the prices. Additionally, they found promotions cause few consumers to focus only on promoted brands, and this effect is higher on the price-aware consumers than on the price unaware consumers. Kukar-Kinney et al. (2012) examined the relationship between consumers’ tendencies to buy compulsively and their response to price based on a survey of customers of an internet clothing retailer and found that compulsive buyers are more brand conscious, are prestige sensitive and have greater knowledge of store prices in comparison with non-compulsive buyers. Moreover, compulsive buyers obtain greater transaction value from price promotions; are sale prone and are more price conscious than non-compulsive buyers. Suwelack et al. 2.2 Customer behaviour (2011) resulted from two experimental studies in addition to cognitive effects are: Money-back guarantees evoke a positive response which results in an increase in readiness to pay price premium and consumer purchase intention. Moreover, money-back guarantees have a positive effect on consumer response to 52 experience and search goods. While Money-back guarantees for experience goods should be designed with stricter return conditions in comparison to money-back guarantees for search goods. Demoulin and Zidda (2009) studied consumers’ process of adoption of a new loyalty card in a grocery retail context. Simultaneously, they investigated the impact of behavioural, socio-demographic and attitudinal variables on the likelihood of adoption and the time of adoption. The results of the study have shown that these variables differently affect the adoption likelihood and timing and demonstrate the importance of attitudinal measures of customer loyalty such as commitment to the retail store. The study extended to the attitudinal dimension of loyalty and confirms the so-called self-selection bias. In a series of experiments, Manning and Sprott (2007) found positive effects of multiple unit price promotions on quantity purchase intentions is dependent on the rate of product consumption and the extent of the quantity specified in the offer. Contradictory, offer effectiveness is not influenced by drawing special attention to single unit prices, and it is not limited to the nature of these promotions or aggregate savings. Kukar-Kinney et al. (2007) examined the antecedents and consequences of perceived price-matching policy fairness. The three experiments of them demonstrated that consumer perceptions about the fairness of store pricing policy have an influence on price fairness perception as a result, influence the retail shopping intentions. Meyer-Waarden (2007) inspected the influence of loyalty programs on customer lifetime duration in grocery stores by using behaviour to scan single source panel data. He suggested that loyalty schemes have a positive effect on the share of consumer expenditures and customer lifetimes. Multiple loyalty card holders of geographically close retailers reduce lifetime duration. Mohan et al. (2012) explored the effect of store environment on variety seeking behaviour, with a model which incorporates various components of store 53 environment (music, light, assortment, employee, and layout) and personality variables, optimum stimulation level and deal proneness. They suggested that retailers need to invest in the components of store environment to enhance the variety seeking. 2.2 Customer behaviour Diallo (2012) investigated jointly the effect of store image perceptions, store brand price-image and perceived risk toward store brands on store brand purchase intention in the context of an emerging market (Brazil). The results of the investigation have shown that store image perceptions and store brand price-image influence significantly store brand purchase intention directly or indirectly via the effect of perceived risk toward store brands. Ngobo (2011) analysed that when retailers increase the share of private labels in their product assortments then how does it affect customer store loyalty? The results indicated that the effects of the private label share on the store loyalty depend on - the store private label branding strategy, how the private label share is measured (e.g. private label share vs. relative private label share) and the household's private label usage. Koistinen and Järvinen (2009) studied consumer observations on their choices between various grocery retailing channels. The results indicated that the consumers have one primary store, which is mostly a supermarket or hypermarket. In addition, consumers like to buy in several supplementary stores that are located close to their residence. 2.3 Customer service Dhume (2012) found two dimensions: problem-solving and personal attention had high gap scores which indicate a disparity between what specialty store consumers expected and their perceived service quality and identified the sub- components of ‘Personal Attention’ Dimension which are: Enthusiasm, Polite and courteous salesperson, Sales person behaviour instills confidence, Service delivered when promised, Prompt service, Individual attention, Customised service, interest at heart, and never too busy to respond. For Problem Solving Dimension: Interest in solving problems, Willingness to help customers, knowledgeable salesperson and expected to deal with customer queries, 54 Jayawardhena and Farrell (2011) found that service and customer orientation behaviours are positively related to service encounter quality and service quality; service encounter quality is positively related to service quality and customer satisfaction; service quality is also positively related to customer satisfaction as well as value perception; and customer satisfaction is positively related to retail customers' buying intention. However, the value is not related to customer satisfaction. Lu and Seock (2008) examined the relationships between perceived service qualities their satisfaction to the stores. The findings showed that all three service quality dimensions in the research were positively and significantly in relation to their satisfaction at preferred department stores and overall loyalty behaviour to the stores. Walker (1995) gave the model which has a better understanding of the service satisfaction process. By identifying and separating the core and peripheral dimensions of services, by implementing the concept of passive and active expectations within service encounter, by considering the evaluation process over time, and by including the consumer’s zone of difference, a much more realistic decision process for consumer evaluations of services comes onwards. Grewal et al. (2008) identified the effect of compensation on repurchase intentions of consumers after a service failure. They utilised an experimental process and analysed the impact of compensation instability and locus of responsibility conditions. Findings from their three studies using scenarios from different service industries indicated that compensation is necessary only when the company is responsible for the failure and the failure occurs frequently. If the failure occurs rarely or the company is not responsible for it then, compensation does not affect repurchase intentions. Wiles (2007) studied the announcements of 48 retailers regarding customer service strategies. He found that customer service increases the retailer market values. 2.3 Customer service Further, study suggests that the shareholder value created by the retailer's customer service is affected by the heuristics and cues used to judge the likelihood of service delivery. The result of the study shows that the retailer’s reputation can also prevent the customer services shareholder wealth creation. 55 Kleijnen et al. (2007) focused on the perceived utilitarian value of a new service delivery model, the mobile channel and develop a framework that includes three mode specific benefits: service compatibility, time convenience and user control. And, two costs: cognitive effort and perceived risk as antecedents of perceived value. Findings revealed that the identified antecedents exclude service compatibility which has an effect on mobile channel value perceptions which ultimately affects behavioural intentions. Schau et al. (2007) analysed more than 2,000 service encounters of service retailers and find evidence of two types of code switching suggested by sociolinguists: language and dialect. Further, the idiosyncratic terminologies, the concept of brand codes are introduced in the service provider’s script. Language code switching has a positive outcome at little cost to organisational efficiency, whereas on-script encounters are preferred. Thakor et al. (2008) found the response of young adults towards fellow consumers who are old or middle-aged. The research suggested that stereotypes of older adults vary depending on the person attributes considered and stereotypes are context dependent according to testing predictions based on recent research. Varying the age of the other consumers and service settings they found that the presence of older consumers affected the young adult’s attitude to their patronage intentions as well as the service. Netemeyer and Maxham III (2007) suggested that the ratings assigned by the supervisor for customer service employee performances can be the preferred base for predicting customer outcomes. Maximising in-role performance inputs may have decreasing returns for customer evaluations in the service recovery context; but maximising extra-role performance inputs may actually “delights” customers, which means higher returns for customer evaluations. Singh et al. (2010) expressed that to measure the service quality of retail stores requires an instrument due to the unique attributes of the retail business. Retail Service Quality Scale has been preferred in their research over SERVQUAL. In this study Service quality is measured for a hypermarket in Delhi. 56 Torres-Moraga et al. (2008) expressed that the success of supermarkets and the retail industry is ascertained by the services they offer. 2.3 Customer service The results showed that Supermarket service quality is a multidimensional construct made up of personal attention, hygiene, accessibility, assurance, reliability and tangibility which are different from the standard constructs. Watson (2012) examined the effects of service type, previous service experience, service criticality and service recovery on the loyalty of customer, customer satisfaction and complaint behaviour. The results of a 2 (service criticality: high or low) x2 (service type: personal or non-personal) x2 (previous service experience: failure or non-failure) x3 (service recovery: assistance only, compensation only, or assistance with compensation) between-subjects experimental design indicated significant main effects for each independent variable on the combined dependent measures. Goodrich and Ramsey (2012) with attribution theory they studied service quality theory as well as a disability orientation theory for focusing on perceived quality of the service for people with disabilities, which provides a connecting framework. The research revealed that there is much space for improvement. In a survey with disabilities, people reveal that retailers are rated lower on accessibility than on traditional service quality dimensions. In addition, feelings of disability pride and social activism significantly affect ratings of accessibility. Auh et al. (2011) drawn a self-determination theory and the job characteristics model to develop a multi-level model that examined the relationship between perceived autonomy and perceived service climate and how these relationships alters at varying levels of store-level tenure diversity and suggested that perceived autonomy is positively associated with the employee's perception of service climate. Perceived autonomy had a greater impact on perceived service climate at low and high levels compared to moderate levels of tenure diversity. Kohijoki, Anna-Maija (2011) by analysing the finish household panel data in their longitudinal study look attentively at the accessibility of grocery retail services among the elderly. The results indicated that most of the elderly have not experienced 57 significant difficulties in accessibility. With the increase in consumer’s age, they demand high-quality services and products and they are willing to make an effort to satisfy their needs. Söderlund and Julander (2009) involved two different service settings which were used to manipulate the level of the service worker's physical attractiveness in an experimental approach. The researchers, for both experiments, have demonstrated that a high level of physical attractiveness in comparison to a low level, the service employee generates a greater level of customer satisfaction. 2.3 Customer service Results also indicated that exposure to an attractive service employee set in motion a process in which an attractiveness appraisal affected the attitude toward the service employee, which in turn had a significant as well as positive impact on the satisfaction of customers. Zolfagharian and Paswan (2009) examined consumer perception of service innovativeness and its effect on patronage intention. Using survey data, the empirical analysis suggests: (a) consumer perception of service innovativeness is associated with patronage intention; (b) different consumer perception of service innovativeness dimensions emerges as significant in different service industries, and (c) consumer trait innovativeness modifies the effect of consumer perception of service innovativeness on patronage intention. Cassab (2009) examined the dynamics of multi-channel service attributes and their effect on customer’s intention to renew a service agreement, by using the response surface model based on conjoint analysis of service profiles. The results provided insights for the design of a service that may not be what customers have experienced in the past and demonstrate the qualities of a set of multi-channel service attributes that have not been included in previous studies. Fullerton (2005) found that affective commitment and continuance commitment are mainly partial mediators of the loyalty relationship and service quality. Results revealed that affective commitment to the retailer had a positive impact on customer loyalty while continuance commitment in marketing relationship had a deleterious effect on customer loyalty. 58 Avlonitis and Indounas (2007) analysed the data from 170 companies operating in six different sectors to explore the pricing information that service companies collect in order to price their services and the pricing objectives they follow. The research concluded that the companies frequently follow a hierarchy of pricing objectives with the special importance being placed on the company’s customers. Barber and Tietje C (2004) compared the distribution services offered by the smaller retailers versus Home Depot, using both in-store measures & consumer perception data and look attentively in relation to the importance of distribution services as determinants of store choice. They said the primary economic function of retailers is to deliver products with distribution services. The results have shown that the Home Depot's superiority in pricing and assortment attracts a noteworthy market, but small retailers can secure niche markets by delivering higher levels of ambiance and information. 2.3 Customer service Swinyard (2003) proposed that store type, salesperson mood and shopper behaviour have an effect on the level of customer service provided by the stores. Salespeople in a good mood are more uniformed in their delivery of customer service, whereas salespeople in a bad mood are more probable to make available the poor service to pleasant than to unhappy customers. The research also revealed that department-store salespeople provide a more uniform level of customer service than a discount-store salespeople. 2.4 Customer satisfaction Martínez-Ruiz et al. (2011) proposed a classification of factors and attributes as stated with their importance for customers’ evaluations in different countries for higher levels of customer satisfaction includes three main factors. The first-order factor comprises most valued attributes by all customers which are not depending on the country of residence. The second-order factors include attributes with lesser importance; the importance relies on country of residence. The third-order factor attributes are valued less in comparison to others. 59 59 Haemoon (1999) suggested a model which possesses the explanatory ability and possess practical validity for customer satisfaction. He used this model in the hotel industry, and results are supporting a holistic approach to the hospitality of customers post-purchase decision-making process. Torben et al. (2011) argued that consumer satisfaction is considered as the degree to which consumer preferences or/and expectations are met. Further, they found that when patronising discount stores and upscale stores – consumers who attach high weight to quality and price are likely to become more satisfied than consumers who attach only medium weight to both the parameters. For long- established supermarkets, satisfaction occurs equally in both groups of consumers. It is resulted out in the study of Fornell (1992) that customer satisfaction increases customer loyalty and prevents customer churn, reduces the costs of failed marketing, lowers the customer’s price sensitivity, and new customer creation, improve the effectiveness of the advertising, reduces operating costs due to the increase in the number of customers, and further improve business reputation. Hansen et al. (2011) found whether consumer supermarket satisfaction has an influence by the mere composition of consumer’s preference structure, in contrast to widespread approaches where consumer satisfaction is concerned with the degree to which consumer expectations and/or preferences are met. And expressed their opinion that Consumers' level of satisfaction with various retailers may not solely be determined by matching preferences with retail products & services, but can also be based on considerations of possibilities for mental justification within a certain preference structure. Hamburg and Koschate (2004) explored the role of perceived fairness and customer satisfaction on the repurchase intention after a price increase. Their findings of two experimental studies reveal that perceived fairness has a positive impact on the repurchase intention and that satisfaction moderates this relationship. Yi (1991) in a critical review of customer satisfaction argued that customer satisfaction operates in two different ways: transaction-specific and generally overall. 2.4 Customer satisfaction 60 Where, transaction-specific concept relates to customer satisfaction as the assessment done after a specific purchase. Martenson (2007) analysed the impact of the corporate store image on customer satisfaction and store loyalty in grocery retailing, found that store as a brand is the most important for customer satisfaction. Retailers need to be good at retailing. Consumers become a satisfied consumer if they think that their needs are understood and addressed by the store as well as the store environment is pleasant and clean. Brown and Lam (2008) reported a meta-analysis of relationships linking customer satisfaction to employee job satisfaction and perceived service quality makes an effort to achieve correlate employee data with customer data. Overall, both relationships are found positive and statistically and substantively significant. Utilising the aggregated data, estimation of a path analytic model demonstrated that customer-perceived service quality completely mediates the relationship between employee job satisfaction and customer satisfaction. Esbjerg et al. (2012) developed a conceptual framework for analysing customer satisfaction with individual grocery purchasing experiences within an overall ‘disconfirmation of expectations model’ of customer satisfaction. The contribution of the conceptual framework is twofold. One, framework synthesise and integrates the multiple central concepts from different research groups into a common framework to analyse shopping trip satisfaction and two, by having a focus on satisfaction with individual grocery shopping trips. Kantabutra (2011) in retail stores of Thailand examined the relationships between perceived vision-based leadership, and consumer and staff satisfaction. Store visions characterised by clarity, challenge, stability, future orientation, brevity, ability to inspire, abstractness and containing a reference to employee, sales, consumer & store leadership directly predict and further improve the store manager leadership as perceived by staff. Such visions also indirectly predict improves staff satisfaction, whereas, staff’s perceived leadership enhances staff satisfaction directly. There is no direct impact of staff satisfaction on consumer satisfaction as normally expected elsewhere. 61 Mazaheri et al. (2011) with the help of a lens of a conflict management framework studied service recovery with customer satisfaction; by specifically evaluate the role of pre-existing attitude in determining customer response to service failures. The research of two scenarios based experiments recommended that conflict management style affects service recovery efforts with customer satisfaction. A cooperative recovery style is necessary to satisfy customers and to exceed expectations. Söderlund and Rosengren (2010) assessed the impact of the service worker's display of emotions (unhappiness vs. 2.4 Customer satisfaction happiness) on customer satisfaction under the conditions of different levels of technical service quality (poor vs. good). The research has shown that the impact of the service worker’s emotional display behaviour on customer satisfaction is dependent on the level of technical service quality, which means that customer satisfaction is affected only when technical service quality is better rather than poor. Sánchez-Fernández and Iniesta-Bonillo (2009) analysed the relationship between consumer satisfaction and economic value. Their results provided an operational tool to measure economic value which will allow retailers to design suitable strategies for creating and delivering value to consumers and the usefulness of economic value in the consumer satisfaction. Veseland Zabkar (2009) suggested that the quality of personal interactions bears no direct influences on members’ loyalty; however, it influences the satisfaction of the members strongly than the loyalty program’s perceived quality. However, influence on members’ loyalty is stronger from the mediating variable of customer satisfaction than from perceived quality of the loyalty program. The results have shown that the role of customer satisfaction is the most important determinant of customer loyalty in the DIY setting. Fonseca (2009) applied a new technique and a new conceptual model of customer’s satisfaction as an approach to modelling and to develop an overall satisfaction index and argues that in order to estimate the global customer satisfaction 62 measure we must appeal to methodologies recognising that satisfaction must be understood as a latent variable, quantified through multiple indicators. Orth and Green (2009) indicated in their study that there is higher consumer trust in family business management practices and policies, satisfaction and frontline employee trust but no variations in loyalty. Their research demonstrates differential effects of how image elements affect customer loyalty directly as well as indirectly through trust and satisfaction by examining an integrative loyalty framework. Darley et al. (2008) examined the relationship of perceived salesperson attributes and customer satisfaction behaviour in the purchase experience leading to the patronage of a service department. Their results indicated that gender is a strong moderator of the relationship among the customer satisfaction, future intentions towards service department patronage and perceived salesperson attributes. Demoulin and Zidda (2008) found up to what extent the satisfaction of loyalty card rewards affects the effectiveness of loyalty card programs in the food retail industry. 2.4 Customer satisfaction With the analysis of survey data within the framework of the store, choice models show that the loyalty card owners are more stores most loyal. More precisely, they have indicated that if the reward schemes of the loyalty card program are satisfying the card owners, they are lesser price sensitive and more loyal in comparison to unsatisfied card holders. Zielke (2008) discussed retail stores asymmetric effects in the formation of satisfaction related to price. The research indicated that value for money, price level, and special offers are both satisfiers and dissatisfiers (one-dimensional); price perceptibility, price processability and price fairness tend to be dissatisfiers only (must be requirements); and price advertising and products in the upper price range are indifferent requirements. Bodet (2008) according to a double view of concepts found the satisfaction- loyalty relationships. The empirical analysis highlighted the role of overall satisfaction on attitudinal loyalty and reduces the role of transaction-specific 63 satisfaction. And therefore, found that neither attitudinal loyalty nor customer satisfaction predicts customer repurchase behaviour. Söderlund and Rosengren (2007) inspected how receiving positive and negative word-of-mouth from dissatisfied and satisfied customers influence the potential customer. The results are consonant with not directly expressed assumptions in the existing literature that word-of-mouth of existing customers has a possibility of a significant effect on potential customers, and this examination shows that in the influence process emotional variables play an important role. Vázquez-Carrasco and Foxall (2006) investigated the relationship between aspects of consumers’ personalities and their perception of active or passive loyalty, satisfaction with, and relational benefits toward the provision of a service. The results have shown that the perception of relational benefits leads to passive loyalty and a higher satisfaction. The results are derived by using partial least square analysis. Huddleston et al. (2009) demonstrated that price, employee service, product assortment, and quality, effect store satisfaction despite store type i.e. conventional or specialty stores. The degree of influence of these attributes varied according to store type. The results indicated that specialty store shopper satisfaction characteristics are clearly described whereas conventional store shopper characteristics are more difficult to pinpoint. 64 64 Conceptual Framework  Retailing  Retailing strategy  Segmentation strategy  Targeting strategies  Positioning strategy  Retail marketing strategy  Defensive strategy  SWOT analysis  PEST analysis  Five force analysis  ETOP analysis  BCG matrix  Generic strategies  Retailing  Retailing strategy  Segmentation strategy  Targeting strategies  Positioning strategy  Retail marketing strategy  Defensive strategy  SWOT analysis  PEST analysis  Five force analysis  ETOP analysis  BCG matrix  Generic strategies  Retail marketing mix  Brand  Merchandise planning  Retail assortment strategy  Stock planning  Objectives of pricing  Price-adjustment strategies  Customer service  Customer service strategies  Store design elements  Visual merchandising  Consumer behaviour  Customer satisfaction Introduction This chapter contributes the brief description of diverse conceptions pertained to the present study. The following important conceptions have been incorporated in this chapter: retailing, retail marketing strategies, retail marketing mix, segmentation strategies, targeting strategies, positioning strategies, defensive strategies, SWOT analysis, PEST analysis, Five forces analysis, ETOP analysis, BCG matrix, Generic strategies, brand, types of brands, merchandise planning, retail assortment strategies, stock (inventory) planning, objective of pricing, price-adjustment strategies, customer service, customer service strategies, store design and its elements planning, customer behaviour and customer satisfaction, etc. 3.1 Retailing The word ‘Retail’ originated from an Old French word i.e. ‘tailler’ that intends “to divide” in terms of tailoring (1365). In 1443, it was first time envisioned with the connotation of a “sale in small quantities” (Harper, 2015). Retailing is comprehended as an approach of trading / selling merchandise (products) and/or services directly to the customers at a profit. Waters (2015) defines the term retailing as “the sale of goods or commodities in small quantities directly to consumers”. 3.2 Retailing strategy Retailing strategy is a comprehensive marketing plan, which summarises how a merchant (retailer) has the intention of offering the merchandise and / or services to customers and inciting their buying decisions, which comprises the target market of merchant, the retail format planned to exercise to serve the target market and bases for having sustainable competitive advantages. The target market is the segment of the market, which will be the focal point of the merchant’s resources and offering a mix. The format explicates the elements of its offering mix [i.e. merchandise assortment, 65 service assortment, pricing, promotion and communication, premises (i.e. location, store design, etc.), presentation (i.e. layout and visual merchandise), personnel and other physical evidence], decided to serve the target market. A sustainable competitive advantage is an edge to the merchants over its competitors (Levy et al., 2012). 3.3 Segmentation strategy Market segmentation is a strategy, which segregates broad target market into subsets of consumers, businesses, or countries based on common needs, interests and priorities. Segmentation strategies are exercised to identify and later define the target customers and contribute supporting data for marketing plan elements to triumph specific objectives of the marketing plan. The following are the types of the segmentation strategies:  Single segment concentration strategy  Selective specialisation strategy  Product specialisation strategy  Market specialisation strategy  Full market coverage strategy  Full market coverage strategy The below mentioned are the most common variables for segmentation: The below mentioned are the most common variables for segmentation:  Geographic variables: Population density, size and climate factors, location or region, etc.  Demographic variables: Gender, marital status, age, race, education, income, occupation, etc.  Psychographic variables: Opinions, shared interests, values of customers and attitudes, etc.  Behavioural variables: Usage, benefits, customer status or prospect, loyalty, etc. 3.4 Targeting strategies 3.4 Targeting strategies  Undifferentiating targeting strategy or mass retailing (no specific mix): The firms prefer this strategy do not target any particular segment and select a 66 single marketing strategy with the conviction that their product & appeal will entice numerous individuals. Figure 3.1: Undifferentiating targeting strategy Figure 3.1: Undifferentiating targeting strategy  Differentiated targeting strategy or selective retailing (specific mix): The firms determine this strategy, decide to target many segments of the market and develop different offering mix strategies for each of the segments. Figure 3.2: Differentiating targeting strategy Figure 3.2: Differentiating targeting strategy Figure 3.2: Differentiating targeting strategy ated targeting strategy or niche retailing (small but profitable):  Concentrated targeting strategy or niche retailing (small but profitable): The firms adopt this strategy concentrate all marketing efforts towards a single segment of the market and develop an offering mix, which caters the needs of the segment. g Figure 3.3: Concentrated targeting strategy  Micro (local or individual) targeting strategy or customised retailing: The firms elect this strategy to design their offering mix to suit the tastes of certain locations and individuals. Figure 3.3: Concentrated targeting strategy Figure 3.3: Concentrated targeting strategy  Micro (local or individual) targeting strategy or customised retailing: The firms elect this strategy to design their offering mix to suit the tastes of certain locations and individuals.  Micro (local or individual) targeting strategy or customised retailing: The firms elect this strategy to design their offering mix to suit the tastes of certain locations and individuals. 67 3.5 Positioning strategy Positioning is a systematic process of constructing a unique position in the customer’s mind compared to competitors. Firms construct a unique position either by intensifying the USP or by blueprinting the different brand identity through advertising. The following are the types of positioning strategies: Figure 3.4: Positioning strategies Source: Adopted from (Ripley, 2015) Figure 3.4: Positioning strategies Source: Adopted from (Ripley, 2015) Source: Adopted from (Ripley, 2015) 3.6 Retail marketing strategy  Market penetration: This opportunity comprises achievement of growth by directing efforts towards existing customers with the present retailing format of he merchant. Retailers using this strategy, open more stores in the target market, allow existing stores for longer hours, training salesperson for cross- selling and displaying merchandise to increase impulse purchases.  Market penetration: This opportunity comprises achievement of growth by directing efforts towards existing customers with the present retailing format of he merchant. Retailers using this strategy, open more stores in the target market, allow existing stores for longer hours, training salesperson for cross- selling and displaying merchandise to increase impulse purchases.  Market expansion: Retailers adopting this strategy, use the existing retail format in new market segments to serve them.  Retail format development: Retailers adopting this strategy, develop a new format of retailing with a different retail mix for the same target market.  Diversification: Retailers adopting this strategy come with a new retail format to serve a new market segment. The diversification may be related  Diversification: Retailers adopting this strategy come with a new retail format to serve a new market segment. The diversification may be related 68 diversification or unrelated. In a related one, present target market or retail format of the retailer has something similar with its new opportunity; however, unrelated diversification has nothing common between the present and the venture. Growth Opportunities Target Market Existing New Retail Format Existing Market Penetration Market Expansion New Format Development Diversification Figure 3.5: Retail marketing strategy Source: Adopted from (Levy et al., 2012) Growth Opportunities Figure 3.5: Retail marketing strategy Source: Adopted from (Levy et al., 2012) 3.7 Defensive strategy Mostly, the market leaders employ defensive strategies in strategic decisions (Arbuckle, 2015). However, a small-scale firm may use this strategy if achieved a market leader position. The prime objective of such strategies is to protect position, market share, profitability, remain a market leader and fighting off with the present as well as potential competitors, discouraging competitors, making attacks unattractive. It is not any easy task, but, fortunately, firms have multiple choices of strategies. These strategies assist in maintaining a high level of confidence among customers.  Preemptive defense (more aggressive launch an attack on the enemy): Firms adopting this strategy attack a competitor before they are attacked.  Counteroffensive defense (provides an opportunity to an attacker to attack frontally): Firms adopting this strategy attack a competitor when a competitor attacks the firm.  Flanking defense (serve as a defensive corner for weak front): Firms adopt this strategy to protect their market share by diversifying their operations into new segments or niche segments. 69  Position defense (strategy to retain position): Firms adopting this strategy, try to hold the current market position of them.  Mobile defense (stretches its domain to new territories): Firms adopting this strategy make regular changes in their business operations to discourage the competitors to compete.  Contraction defense (giving up weaker territories or activities): It is the least desirable defense strategy and firms adopting this strategy retreat from the markets. 3.8 SWOT analysis The SWOT is an acronym for strengths, weaknesses, opportunities and threats. A SWOT analysis, also known as SWOT matrix is a comprehensive and systematic planning tool utilised to evaluate strengths, weaknesses, opportunities and threats in any project or any business venture to take prior actions. It is a very common tool and analysis with this tool could be done to any person, place, product, company or industry. It assists in specifying the goals and objectives of the firms or project based on the internal and external factors, which may be favourable or unfavourable to accomplish those goals and objectives. Figure 3.6: The SWOT Analysis Source: Adopted from (Humphrey, 2005) 3.9 PEST analysis The PEST is an acronym for Political, Economic, Social and Technological analysis, which includes the macro-environmental factors important to scanning of the environment for better strategic decisions. Some analysts include the Legal environment factors and rearrange the mnemonic to SLEPT and after adding environmental factors the same becomes PESTEL or PESTLE. Figure 3.7: The PEST analysis Source: Adopted from (Admin, 2011) Figure 3.7: The PEST analysis Source: Adopted from (Admin, 2011) Source: Adopted from (Admin, 2011) Figure 3.6: The SWOT Analysis Source: Adopted from (Humphrey, 2005) 70 70 70 3.10 Five force analysis The initiator of five-force analysis is Prof. Michael E. Porter of Harvard University. Therefore, it is also known as the Porter five forces analysis, which basically offers a framework to understand the level of competition in the industry for developing business strategies. It describes that industrial organisation economics should consider the five forces to understand the intensity of competition and attractiveness of an Industry. Here, attractiveness means the overall profitability of the industry. 71 Figure 3.8: Five force analysis Source: Adopted from (Kazmi, 2007) Figure 3.8: Five force analysis Source: Adopted from (Kazmi, 2007) Figure 3.8: Five force analysis Source: Adopted from (Kazmi, 2007) 3.12 BCG matrix BCG Matrix is a portfolio planning mechanism, which analyse the different SBUs / Products / Businesses of the firm, according to the market growth rate and relative market share. SBUs / Products / Businesses are categorised as stars, cash cows, question marks or dogs based on their market growth rate and relative market share. Figure 3.10: BCG Matrix Source: Adopted from (Kazmi, 2007) Figure 3.10: BCG Matrix Source: Adopted from (Kazmi, 2007) 3.11 ETOP analysis ETOP is an acronym for Environmental, Threat and Opportunity Profile and utilised as a tool to structure environmental concerns. This tool divides the environment into different sectors and the individual sector is further divided into different sub-sectors. Hence, this tool is also used to analyse the impact of each individual sector and sub-sector on the firm. This tool also offers an explanation about the impact in the form of a statement. Figure 3.9: ETOP analysis Source: Adopted from (Glueck and Lauch, 1984) Figure 3.9: ETOP analysis Source: Adopted from (Glueck and Lauch, 1984) 72 72 3.13 Generic strategies The concept of generic strategies was initiated by Michael E. Porter in 1980. Therefore, also known as Porter's generic strategies and explicate that how a firm can pursue competitive advantage in the selected market. There are 4 generic strategies: Figure 3.11: Generic strategies Source: Adopted from (Porter, 1985) Figure 3.11: Generic strategies Source: Adopted from (Porter, 1985) Source: Adopted from (Porter, 1985) 73 73  Cost Leadership (i.e. broadly target with lower cost): Firms adopting this strategy, target the customers in most of the segments of an industry with the lowest price.  Cost Leadership (i.e. broadly target with lower cost): Firms adopting this strategy, target the customers in most of the segments of an industry with the lowest price.  Differentiation (i.e. broadly target with differentiation): Firms adopting this strategy, target the customers in most or all segments on the basis of attributes rather than price. Firms adopting this strategy attempt to differentiate themselves from their competitors. The firms look to minimise costs in all the areas, which do not differentiate.  Cost Focus (i.e. narrowly target with lower cost): Firms adopting this strategy, focus on one or few segments. The firm adopts to offer a lower cost in that scope.  Differentiation Focus (i.e. narrowly target with differentiation): Firms adopting this strategy focus on one or a few segments and differentiate them in that scope. 3.14 Retail marketing mix The retail marketing mix is a set of tools, which work together for catering the customer needs and building a long-term relationship with them. The set of tools comprises merchandise (product), price, promotion, place or premises, presentation, personnel and other physical evidence.  Product: It is the most important element of the retail marketing mix that comprises the classification and brands. With this element, the retailers may satisfy the needs and desires as per the expectations of the customers.  Price: It is considered as an important element of the mix, as brings profits to the firm. Methods of pricing usually based on the target market, competition, margins and brands.  Promotion: This element is designed for promotion and communication pertained to various products and services that include the advertising (in-store or out-store), sales promotion, public relations, personal retailing, direct retailing, internet retailing.  Premises / Place: This element is designed to choose the right channels for reaching to the customers and comprises the locations, sites, space, etc. 74 74 74  Presentation: This element portrays the image of the retailers and comprises the presentation of the merchandise, stores, premises, offices, delivery vans, warehouses, websites, etc.  Personnel: It is an important element of the mix and directly associated with the satisfaction of the customers pertained to services. It comprises the sales personnel, contact personnel and factors pertained to personnel. Figure 3.12: Retail marketing mix Source: Adopted from (Jhureley, 2011) Personnel Presentation Premises or place Promotion Price Product Retail Marketing Mix Product Personnel Retail Marketing Mix Presentation Promotion Premises or place Figure 3.12: Retail marketing mix Figure 3.12: Retail marketing mix Source: Adopted from (Jhureley, 2011) Source: Adopted from (Jhureley, 2011) Source: Adopted from (Jhureley, 2011) 3.15 Brand The brand is a registered name, term, design, symbol, or any other features that differentiate merchandise of one producer or merchant from the others. The legal term of the brand is a trademark (Cohen, 2011). There are different types of brands as follows:  Generic brands: Generic brands are those, which do not have a brand name, logo and promotion.  Local brands: The merchandise distributed and sold in a few specific areas of the country. 75  Manufacturer brands: Manufacturer brands are developed by the manufacturer himself and carry the manufacturer’s name or any other name of the manufacturer’s choice.  National brands: The brands that are distributed nationally, but owned by the manufacturer or distributor.  International brands: The brands that are distributed across the borders (world) but owned by the manufacturer or distributor  Private (store) brands: Products which are distributed in the brand name of the retailer, not the manufacturer. Following are the important factors in merchandise planning: Following are the important factors in merchandise planning:  Target market  Goods & service growth potential  Fashion trends  Image of retailer  Competition  Customer segments  Responsiveness to customers  Amount of investment  Profitability  Risk Following are the important factors in merchandise p  Target market  Goods & service growth potential  Fashion trends  Image of retailer  Competition  Customer segments  Responsiveness to customers  Amount of investment  Profitability  Risk  Target market  Target market  Goods & service growth potential Following are the important factors for planning quality of merchandise: Following are the important factors for planning quality of merchandise:  Target market  Competition  Perceived goods/service benefits  Manufactures vs. private brands 3.16 Merchandise planning It is a strategic choice pertained to planning sales and inventory for maximising ROI and profitability. This is accomplished through increasing sales potential and decreasing the losses from overstock or out of stock conditions (The planning Factory Limited, 2014). Figure 3.13: Merchandise planning Source: Adopted from (Sayed, 2015) Figure 3.13: Merchandise planning Source: Adopted from (Sayed, 2015) 76 76 Following are the important factors in merchandise planning: Following are the important factors in merchandise planning:  A wide variety & a deep assortment: Retailers adopting his strategy offer many goods/service categories and a large assortment in each category.  A wide variety & a deep assortment: Retailers adopting his strategy offer many goods/service categories and a large assortment in each category.  A wide variety & a deep assortment: Retailers adopting his strategy offer many goods/service categories and a large assortment in each category. 3.19.1 Pricing objectives  Survival.  Maximising current profit.  Maximising market share.  Maximising market skimming.  Product quality leadership. 3.18 Stock (inventory) planning Stock (inventory) planning has a consequential impact on the cash flows as well as the profit of the firm. It assists in understanding the optimal time and quantity of inventory. It is the decision pertained to the alignment of sales and production capacity. Retailers inventory planning includes the:  Weekly planning.  Monthly planning.  Yearly planning.  Long term planning. 3.19 Objectives of pricing Pricing objectives are the guiding principles in the overall pricing process. When, firms decide their pricing objectives need to consider the overall marketing, strategic, and financial objectives of the firm; objectives of brand or product offered, consumer price and price elasticity in the market and the resources. The firms have a different pricing objective, approaches, and strategies to choose: 3.17 Retail assortment strategy (product assortment): Retail assortment strategy is a strategic decision pertained to the number and type of merchandise, the merchant (retailer) decides to display in the store for customers. The retailers have few choices to decide about assortment:  A narrow variety & a shallow assortment: Retailers adopting his strategy offer few goods/service categories and a limited assortment in each category.  A wide variety & a shallow assortment: Retailers adopting his strategy offer many goods/service categories and a limited assortment in each category.  A narrow variety & a deep assortment: Retailers adopting his strategy offer few goods/service categories and a large assortment in each category. 77 3.19.4 Price-adjustment strategies  Discount & allowance pricing: Reduction in prices to reward customer responses like as promoting the product or paying early.  Segmented pricing: Adjusting the price to allow for differences in customers, locations or merchandise.  Psychological pricing: Adjustment in prices for psychological effect.  Promotional pricing: temporarily reducing prices to increase short-run sales.  Geographical pricing: Adjusting prices to account for the geographic location of the customer.  Dynamic pricing: Adjustment in prices continues to meet the features and needs of individual customers and various situations.  International pricing: Adjustment in prices for the international market. 3.19.3 Pricing strategies 3.19.3 Pricing strategies 3.19.3 Pricing strategies  Market-skimming pricing.  Market-penetration pricing.  Variable (zone) pricing.  Markup pricing.  Target return pricing.  Perceived value pricing.  Going rate pricing.  Value pricing. o Everyday low pricing (E o High-low pricing. 3.19.3 Pricing strategies  Market-skimming pricing.  Market-penetration pricing.  Variable (zone) pricing.  Markup pricing.  Target return pricing.  Perceived value pricing.  Going rate pricing.  Value pricing. o Everyday low pricing (EDLP). o High-low pricing.  Market-skimming pricing.  Market-penetration pricing.  Variable (zone) pricing.  Markup pricing.  Target return pricing.  Perceived value pricing.  Going rate pricing. o Everyday low pricing (EDLP). o High-low pricing. o High-low pricing. 3.19.2 Pricing approaches  Demand-oriented pricing.  Cost-oriented pricing.  Competition-oriented pricing.  Leadership-oriented pricing. 78 78 3.20.1 The Customer Service skills that matter Ciotti, (2013) identified some skills that every support employee should develop to delight the customers are:  Patience.  Attentiveness.  Clear communication skills.  Knowledge of the product.  Ability to use "positive language".  Acting skills.  Time management skills.  Time management skills.  Ability to "read" customers.  Ability to "read" customers.  A calming presence.  A calming presence.  Goal oriented focus.  Goal oriented focus.  Ability to handle surprises.  Ability to handle surprises.  Persuasion skills.  Tenacity.  Tenacity.  Closing ability.  Closing ability.  Willingness to learn. 3.20 Customer service Customer service is catering the needs of customers with the delivery of high quality and professional assistance before, during and after the customers’ needs are satisfied. The customer services assist in maintaining the long-term relationship with customers and continuing the revenue to the company. Customer service is one among the many key factors for building rapport and brand name of the company. The basic characteristics of good customer service are:  Promptness.  Promptness. 79 79 79  Politeness.  Politeness.  Professionalism.  Personalisation.  Personalisation. 3.20.2 Basic rules for good customer service  Be ready to serve your customers at any point of time at any place.  Do not keep your customers waiting.  Do not keep your customers waiting.  Have answer of all the questions and answer those questions, which have not been asked.  Decide what to promise or not to promise to your customers.  Always cater more than the promise of firm and expectations of customers. 3.20.3 Customer service strategies The service providers may have one approach among the mentioned below:  Patronage builders: Firms adopting his strategy involve in high-cost activities that which are basic reasons behind the customer loyalty). 80  Patronage solidifiers: Firms adopting his strategy involve in the low-cost little things that increase loyalty.  Disappointers: Firms adopting his strategy involve in expansive activities that do no real good.  Basics: Firms adopting his strategy involve in low-cost activities, which are naturally expected. 3.21 Store design elements  Layouts: It is an important element of store design. The retailers use following layouts: o Grid (straight) layout: In this layout, the parallel aisles are used and the products (merchandise) are kept on the shelves on both sides of the aisles. The Cash counter can be located at the exits or entrances of the stores. 81 81 o Free flow layout: Popularly known as a boutique layout, in which the fixtures and aisles are arranged in a symmetric pattern. Figure 3.14: Grid layout Source: Adopted from (Stanton, 2015) o Curving or loop (race track) layout: In this layout, the major aisle is used to loop around the store for guiding the customer traffic in the several departments of the store. o Curving or loop (race track) layout: In this layout, the major aisle is used to loop around the store for guiding the customer traffic in the several departments of the store. o Curving or loop (race track) layout: In this layout, the major aisle is used to loop around the store for guiding the customer traffic in the several departments of the store. Figure 3.15: Curving or loop layout (Source: Stanton, 2015) Figure 3.15: Curving or loop layout (Source: Stanton, 2015) Figure 3.15: Curving or loop layout (Source: Stanton, 2015) o Free flow layout: Popularly known as a boutique layout, in which the fixtures and aisles are arranged in a symmetric pattern. 82 82 Figure 3.16: Free flow layout Source: Adopted from (Stanton, 2015) Source: Adopted from (Stanton, 2015) Source: Adopted from (Stanton, 2015) o Spine layout: In this layout, the main aisle is used as the movements between the several cells established on both sides of the aisles. 3.22 Visual merchandising It is the decision pertained to the presentation of the merchandise and stores to entice the customer visiting the store. The retailers utilises the straight sacks, rounders, four-way fixtures, gondolas to present the merchandise in the stores and also create an appealing store atmosphere (exterior and interior).  Exterior atmospherics: Exterior atmospherics comprises the elements like a storefront, entrances, display window, visibility, adjoining store, etc. These elements assist in creating eye-catching and attractive store exterior  Interior atmospherics- Interior atmospherics comprises the elements like flooring, lighting, colour, personnel, cleanliness, walls, aisles, temperature, sitting areas, music, etc. These elements assist building up the physical shopping environment of a store. o Spine layout: In this layout, the main aisle is used as the movements between the several cells established on both sides of the aisles. o Spine layout: In this layout, the main aisle is used as the movements between the several cells established on both sides of the aisles. Figure 3.17: Spine layout Source: Adopted from (Stanton, 2015) Figure 3.17: Spine layout Figure 3.17: Spine layout Source: Adopted from (Stanton, 2015) Source: Adopted from (Stanton, 2015) 83 83  Signage: Signage is used to assist the customers in locating the merchandise, categories, different departments and promotional offers. Three kinds of signage are used: o Category signage: Used to assist in locating the information within a particular department of the store. o Promotional signage: Used to assist in locating the information pertained to special offers. o Point-of-sale signage: Used to assist in locating the information pertained to price and other details of products. 3.23 Consumer behaviour It is all about the psychological processes, through which a consumer travel for need identification, alternatives to satisfy the needs, making purchase decisions, collecting & interpreting information, taking the decision of purchase, consumption and disposal. 84 84 Figure 3.18: Consumer behaviour 3.23.1 Consumer choice and decision-making Figure 3.19: Consumer choice and decision-making Source: Adopted from (Perner, 2008) Figure 3.18: Consumer behaviour Figure 3.18: Consumer behaviour Figure 3.18: Consumer behaviour Figure 3.18: Consumer behaviour 3.23.1 Consumer choice and decision-making Figure 3.19: Consumer choice and decision-making Source: Adopted from (Perner, 2008) 3.23.1 Consumer choice and decision-making Figure 3.19: Consumer choice and decision-making Source: Adopted from (Perner, 2008) 3.23.1 Consumer choice and decision-making Figure 3.19: Consumer choice and decision-making Source: Adopted from (Perner, 2008) 3.24 Customer satisfaction Customer satisfaction measures the degree how well the expectations of customers pertained to products or services have been met. According to Beard (2014), top reasons, why customer satisfaction is important are:  A tool for evaluating the consumer repurchase intentions as well as loyalty  A point of differentiation.  Helps in decreasing customer churn.  Assist in increasing Customer Lifetime Value.  Assist in creating positive word of mouth.  Assist in retaining customers. 85 85 Chapter 4  Objectives of the study  Significance of the study  Delimitation of the study  Research hypotheses  Research design  Locale of the study  Universe of the study  Sampling design and selection of samples  Collection of data  Tools for data collection  Pilot study  Scales used for accumulating the responses  The validity of scales  The reliability of scales & internal consistency of reliability  The method used for collecting the data  Methods of reporting  Statistical analysis of data  Objectives of the study  Significance of the study  Delimitation of the study  Research hypotheses  Research design  Locale of the study  Universe of the study  Sampling design and selection of samples  Collection of data  Tools for data collection  Pilot study  Scales used for accumulating the responses  The validity of scales  The reliability of scales & internal consistency of reliability  The method used for collecting the data  Methods of reporting  Statistical analysis of data  Objectives of the study  Significance of the study  Delimitation of the study  Research hypotheses  Research design  Locale of the study  Universe of the study  Sampling design and selection of samples  Collection of data  Objectives of the study  Significance of the study  Delimitation of the study  Research hypotheses  Research design  Locale of the study  Universe of the study  Sampling design and selection of samples  Collection of data  Statistical analysis of data Introduction This chapter on methodology provides the description of the objectives, scope, limitations and research procedure followed to carry out the present study. The individual response and institutional responses have been studied independently in this study and the relationship between the two has not been studied so far. The detailed research methodology of the present study is being presented under the following heads: 4.2 Significance of the study The study could be rationalised for its multiple utilities and implements in the consequent significancies: the study contributes knowledge & facts about the variegated merchandise (products) & customer services penetrated by organised retailers in the industry. The study aspires to discover the retailing strategies determined by organised retailers together with finding the dimensional gap to comprehend in the domain of retailing strategies in the line of the hyper-competitive environment. The study observes and reveals appropriate indices for employing modern / high-tech retailing approaches to terminate the gaps. This study also delivers evidence about the dilemmas of customers along with some potential revelations to them. The study ascribes details for the betterment essential in the merchandise (products) & services that clients want. The study also determines diverse components and customer services mandatory to persist the retailing industry in a highly competitive retail environment. The study additionally functions to deliver the factors that serve towards the superior level of customer satisfaction in the organised retail industry. The study assists the management of organised retail stores to appraise upon the retailing strategy. The study benefits in comprehending the retailing environment, retailing culture and the retail decision process. Through this study, the researcher appropriated the opportunity to exercise the theoretical knowledge in the appropriate practical arena. The researcher not only fulfils the requisite of Ph.D. curriculum, additionally comprehended lots about the industry. The study will additionally ascertain to be of comprehensive assistance to the researchers, who yearn to do uniform or accompanied studies in the forthcoming time. 4.1 Objectives of the study After a thorough review of existing literature, it is understood that around the globe, researchers carried out very less research work pertained to retailing strategies of products and customer services in the organised retail industry. The maximum work carried out is presenting either the view of retailers or the customers and the existing work is not able to provide a comprehensive picture of retailing strategies of products and customer services in the organised retail industry. Therefore, to full-fill the gap, the current study has been taken to provide a comprehensive picture of retailing strategies of products and customer services in the organised retail industry from the retailers’ point of view as well as from customers’ point of view. The current research has been carried with the following objectives:  To study the various products and customer services introduced by organised retail industry.  To appraise the various retailing strategies introduced by organised retail industry.  To unpack the changing buying behaviour of customers in the organised retail industry.  To get at the factors that influence the buying behaviour of customers in the organised retail industry.  To assess the customer satisfaction level with regards to various merchandise (products) and services introduced by organised retail industry. 86 4.4. Research hypotheses Based on gaps in the literature review and objectives of the study, the following two null hypotheses have been formulated: H01: Retailing strategies introduced by organised retailers are not significant to sustain in the hyper-competitive retail market. H01: Retailing strategies introduced by organised retailers are not significant to sustain in the hyper-competitive retail market. H02: Customers are not significantly satisfied with various products & services introduced by organised retailers. 4.3 Delimitation of the study  The study focuses only on the organised retail sector.  The study is limited to 3-5 years data.  The study is limited to 3-5 years data.  The study was delimited to the selected cities of Maharashtra state, namely Nagpur, Amravati, Navi Mumbai, Thane, Nashik, Jalgaon, Kolhapur, Sangli, Pune and Aurangabad. 87 87  The study was delimited with 10 retail stores, namely Big Bazaar, NEXT, HyperCITY, Max, more.MEGAStore, D Mart, Star Bazaar, Reliance Digital, Big Megamart and West Side.  The study is limited to 3 formats, i.e. multi-format retail stores, specialised fashion retail stores and specialised electronics retail store. It cannot be implemented in other formats of retailing.  The study is also delimited to the 2000 respondents from selected cities. 4.5 Research Design Philips (1966) defined a good research design as “….The blueprint of the collection, measurement and analysis of data. It aids the scientists in the allocation of his limited resources by posing a crucial choice- is the blueprint to include experiments, interviews, observation, the analysis of records, simulation, or some combination of these? Are the methods of data collection and the research situation to be highly structured? Is an intensive study of a small sample more effective than a less intensive study of a larger sample? Should the analysis be primarily quantitative or qualitative?” Taking the inspirations from this definition, the research design for the present study has been defined as follows: Both descriptive and exploratory research methods were used for compiling this study. An exploratory research focus to develop initial hunches or insights and to provide direction for any further research needed and descriptive research aims to describe something (Parasuraman et al., 2007). In the present study, the exploratory research helped in developing the hypotheses and framing the objectives of the study through the analysis of secondary data, while descriptive research was used to pursue the objectives and the hypothesis of this study. 88 4.6 Locale of the study The study was conducted in the Maharashtra state of India and the organised retail stores from various segments were selected from various parts of the state while choosing the sample unit. Efforts were made to cover the entire state geographically and regionally, at least 2 well-known cities were selected from each region. Figure 4.1: Locale and universe of the study Figure 4.1: Locale and universe of the study 4.7 Universe of the Study For conducting the present study, the whole Maharashtra state was classified into 5 regions, i.e. Northern Maharashtra, Southern Maharashtra, Eastern Maharashtra, Western Maharashtra and Central Maharashtra. From each region, two well-known cities were selected. Further, from each city, one retail store was selected for the study (i.e. five regions * two retail stores per region = ten retail stores). Those retail stores are Big Bazaar, NEXT, HyperCITY, Max, more.MEGAStore, D Mart, Star Bazaar, Reliance Digital, Big Megamart and West Side. Out of total 10 retail stores, 5 were hyper stores and remaining 5 were super stores. These hyper and super 89 stores have also been selected by a planning that from each region one hyper and one super store must be selected for the study. The 5 hyper stores, namely Big Bazaar, HyperCITY, More.MEGAstore, Star Bazaar, and Big MegaMart have been selected while NEXT, Max, D Mart, Reliance Digital and Westside have been selected as the representative of super stores (refer table 4.1). 4.9 Collection of data Neelankavil (2007) discussed that one among the cardinal rules in data accumulation is to deplete all secondary data stimuli before endeavouring a primary study. The present study is based on both secondary and primary data. To achieve the objectives of the present study both primary and secondary data were collected from various sources. The secondary data played a vital role in a review of the literature, formulate hypothesis and questionnaire preparation. It was accumulated from various libraries, books, journals, magazines, Retail Association of India, FICCI, websites and other published sources available, references of the same have been cited at the end of conclusion. Utilising the information from the secondary data, two sets of structured questionnaire were prepared for respondents to accumulate the primary data, comprising open & close ended questions. Self-structured questionnaires were prepared to collect the data from institutions (i.e. personnel of retail stores) and individual customers of stores. Two retail stores from five regions of Maharashtra namely eastern, western, northern, southern and central were selected to collect the primary data. 4.8 Sampling design and selection of samples For the present study, a non-probability convenience sampling technique was resorted for institutional (i.e. store personnel) and individual (i.e. customers) respondents. The non-probability sampling is a subjective approach in which the probability of determination of the population units cannot approximate. The suitability or convenience of the researcher constructs the conjuncture for determining a sample of units in convenience sampling (Parasuraman et al., 2007). Table 4.1: Sample size and selection of samples S.N. Region City Retail Store Type Sample Size (Customers) 1 Eastern Maharashtra Nagpur Big Bazaar Hyper Store 200 Amravati NEXT Super Store 200 2 Western Maharashtra Navi Mumbai HyperCITY Hyper Store 200 Thane Max Super Store 200 3 Northern Maharashtra Nashik More.MEGAstore Hyper Store 200 Jalgaon D Mart Super Store 200 4 Southern Maharashtra Kolhapur Star Bazaar Hyper Store 200 Sangli Reliance Digital Super Store 200 5 Central Maharashtra Pune Big Megamart Hyper Store 200 Aurangabad West Side Super Store 200 Total 5 10 10 5 + 5 2000 Table 4.1: Sample size and selection of samples Total 10 retail stores, namely Big Bazaar, NEXT, HyperCITY, Max, more.MEGAStore, D Mart, Star Bazaar, Reliance Digital, Big Megamart and West Side were selected for the study. From each retail store, an institutional questionnaire got filled up by store personnel (i.e. store manager or assistant store manager or 90 representative of the store) to receive accurate data about various products, services and retailing strategies introduced by the respective retail store. In total 2000 customers of retail stores (i.e. 10 retail stores * 200 customers) were contacted. Only those respondents were selected, who were willingly agreed to respond and had a positive outlook toward the problem. From the area of each selected retail stores 200 respondents contacted at a convenient basis for obtaining data about their buying behaviour, factors influencing their buying behaviour, and satisfaction level pertained to products and services offered by retail stores. Detail plan for selection of respondents has been mentioned in table 4.1. 4.10.2 Individual questionnaire The individual questionnaire was used to know the views of the respondents (i.e. customers of retail stores) on product and service, buying behaviour influencing factors and satisfaction level. The basic purpose of the questionnaire was to analyse the customer profile, buying behaviour of customers, factors influencing buying behaviour, the customer satisfaction with the product and service offered by retail stores. This questionnaire consists of 33 items (questions) in all, however, mostly questions also comprise the succeeding and sub-questions and pattern of questions is different from another. The responses of the respondents have been received on the nominal scales, ordinal scales and Likert’s (1970) five point scale (basically an ordinal scale) in this instrument. The validity, reliability of the scales and internal consistency of reliability were also established. 4.10.1 Institutional questionnaire The institutional questionnaire was used to know the views of the respondents (i.e. personnel of retail stores) on retailing strategy. This questionnaire consists of 69 items (questions) in all, however, mostly questions also comprise the succeeding and sub-questions. The responses of the respondents have been received on the nominal and ordinal scales in this instrument. The validity, reliability of the scales and internal consistency of reliability were also established. 4.10 Tools for data collection Two instruments, one for personnel of retail stores (i.e. institutional questionnaire) and another for customers of retail stores (i.e. individual questionnaire), were constructed to accomplish the objectives of the study. 91 4.12 Scales used for accumulating the responses In the institutional questionnaire, the responses of the respondents have been received on the nominal and ordinal scales. In few questions, the respondents have been asked to rate the variables of 4-point importance scale (basically an ordinal scale), which is ranging like extremely important (4), important (3), partially important (2) and not important (1). While, in few other questions, the respondents have been asked to rate the variables of 4-point agreement scale (basically an ordinal scale), which is ranging like extremely agree (4), agree (3), partially agree (2) and not agree (1). In the individual questionnaire, the responses of the respondents have been received on the nominal scales and ordinal scales. However, the responses pertained to the factors influencing the buying behaviour have been taken on Likert’s (1970) five point scale (basically an ordinal scale) ranging from extremely influenced (2) to not at all influenced (-2) with the middle of the scale recognised by the response alternative neither influenced nor uninfluenced (0). And the responses pertained to the customer satisfaction have been collected on five point expectation cum satisfaction scale (basically an ordinal scale) ranging from greatly exceeding expectations and delighted (5), to much less than expectations and disappointed (1) with the middle of the scale identified by the response alternative matching expectations and satisfied (3). 4.11 Pilot study It is necessary to design a suitable questionnaire, conducting a pilot study and undertake a pre-testing of the questionnaire (Beri, 2000). Both questionnaires were tested by conducting a pilot study of the few respondents selected on a random basis. The institutional questionnaire was pre-tested with personnel of Big Megamart, Star Bazaar and NEXT. Taking the insight from this pre-testing of the institutional questionnaire, certain items were included and even excluded to modify the questionnaire for the final study. The individual questionnaire was also pre-tested with customers 200 customers of organised retail stores selected on a random basis. Utilising the insight from pre-testing of the individual questionnaire, certain items were included and even excluded to modify the questionnaire for the final study. 92 4.13 The validity of scales The face & content validity method was employed to measure the validity of the scales used both individual and institutional questionnaires. “Face validity is the dimension to which a measurement scale seems to measure what is hypothesised to measure” (McDaniel and Gates, 2001). Face validity is determined by the assessment of the researcher, who constructed the questionnaire with divergent scales, which rationally exhibited to correctly reflect what the researcher was hypothesised to evaluate and to evaluate the validity of the content, first researcher signified what precisely required being evaluated. For the topical study, essential variables were determined through the objectives along with hypothesis formulated, which assisted in discovering what 93 required being evaluated. Secondly, by a comprehensive review of the literature to determine all potential items were ascertained. Third, opinions from subject experts, retail store personnel and customers were taken into the consideration on whether conclusive items should be included or even excluded. 4.14 The reliability of scales and internal consistency of reliability Reliability is synonymous with repeatability. It is an evaluation that confirms coherent outcomes over time is remarked to be reliable. When an evaluation is plausible to random error, it shortfalls the reliability. The reliability of an instrument attributes an upper limit on its validity. An evaluation that shortfalls the reliability will also shortfalls the validity. If the scale of reliability is adjacent to 1, it ensures an evidence of good reliability. Reliability analysis is a distinguished and commonly facilitated mechanism for estimating the internal consistency of the variables. The reliability of the scale was computed through the adoption of the research of Copper and Schindler (2006). Cronbach (1951) Alpha (α) is designed as a measure of internal consistency. The alpha values closer to 1, the greater the internal consistency of items in the instrument being considered. The internal consistency of reliability of the scales used in both questionnaires was measured through Cronbach's Alfa. The Alfa values for all importance scales and agreement scales used in the institutional questionnaire are ranging from 0.601 to 0.805, which is an indication of the good internal consistency of reliability of the scales used in the questionnaire. In the case of scales used in the individual questionnaire, the Alfa value of influence scale was 0.791 and for expectation cum satisfaction scale was 0.823, which is an indication of the good internal consistency of reliability of the scales used in the questionnaire. 4.15 The method used for collecting the data Survey method and printed questionnaires were employed in accumulating the primary data. “The personally administrated questionnaires ascertain rapport as well as encourage respondents, moreover hereto, elucidate the questions expeditiously” (Sekaran, 2003). Therefore, the questionnaires were supervised individually practicing the face to face approach to enhance response rate, encourage respondents and elucidate the doubts instantly. The data was collected over a span of 12 months. A 94 survey was done in all seven days, but especially survey has also been done on Saturday and Sunday to get more positive responses. 4.16 Methods of reporting / presenting data The research report text consists of tables and charts for better presentation and understanding of context. The data has been presented with the help of univariate as well as bivariate or multivariate tabulation. The univariate tabulation is also known as one-way tabulation and bivariate or multivariate tabulation is known as cross tabulation (Beri, 2000). Moreover, the data has also been presented with the help of 3D pie charts, 3D clustered column charts, radar with markers charts and line charts. 4.17 Statistical analysis of data The data obtained in the present study were analysed using suitable statistical tools to appraise the various retailing strategies introduced by organised retail industry and assess the customer satisfaction level with regards to various merchandise (products) and services introduced by organised retail industry. The following statistical tools were used for interpretation of data.  Parametric test: One sample t-test  Nonparametric test: Kolmogorov-Smirnov one-sample test  Other statistical tools: Simple percentage method, mean, standard deviation, and weighted average method 4.17.2 Mean and standard deviation It has been used to analyse the data pertained to the customer profile, buying behaviour, factors affecting buying behaviour and customer satisfaction. 4.17.1 Simple percentage method It has been used to analyse the data pertained to the customer profile, buying behaviour, factors affecting buying behaviour and customer satisfaction. 4.17.4 Kolmogorov-Smirnov one-sample test It is similar to the Chi-Square test of goodness of fit. It has been used to analyse the data collected on importance and agreement scales in the institutional questionnaire and also utilised in the individual questionnaire to analyse the data pertained to customer satisfaction. 4.17.3 Weighted average method It has been used to analyse the data pertained to factors affecting the buying behaviour of customers of retail stores and satisfaction of customers. 95 4.17.4 One sample t-test It has been used to analyse the data collected on five point expectation cum satisfaction scale pertained to customer satisfaction. 96 Data Analysis, Results & Discussion CHAPTER 5: Merchandises (products) & customer services in organised retail industry CHAPTER 6: Retailing strategies introduced by organised retail industry CHAPTER 7: Changing buying behaviour of customers in the organised retail industry CHAPTER 8: Customer satisfaction with the merchandise (products) and services offered by organised retail industry and testing of hypotheses CHAPTER 9: Conclusion and recommendations  Merchandise (products) offered by organised retail industry (p ) y g y o Merchandise offered by multi-format organised retailers  Customer services introduced by organised retail industry o Services introduced by multi-format organised retailers o Services introduced by fashion-organised retailers o Services introduced by electronics-organised retailers o Merchandise offered by fashion-organised retailers  Merchandise (products) offered by organised retail industry o Merchandise offered by multi-format organised retailers o Merchandise offered by fashion-organised retailers o Merchandise offered by electronics-organised retailers  Merchandise (products) offered by organised retail industry o Merchandise offered by multi-format organised Chapte r 5  Merchandise (products) offered by organised retail industry (p ) y g y o Merchandise offered by multi-format organised retailers o Merchandise offered by fashion-organised retailers o Merchandise offered by electronics-organised retailers o Services introduced by fashion-organised retailers o Services introduced by electronics-organised retailers Introduction This chapter incorporates the analysis of assorted data pertained to merchandise and services introduced by organised retail industry, which have been accumulated through secondary data sources and personal interaction with responsible personnel of the respective organised retail store. Merchandise portfolio analysis of merchandise offered by organised retail industry has been presented through BCG Matrix. Further, results have been interpreted and discussed to reach on some inferences at the end of the different sections. This chapter has been segregated into two sub-sections as follows: 5.2 Customer services introduced by organised retail industry 97 97 5.1.1 Merchandise offered by HyperCITY HyperCITY offers merchandise under the merchandise lines like food & grocery, health & beauty, home & furniture, fashion, electronics, toys, sports and private (store-level) brands. MERCHANDISE OFFERED BY ORGANISED RETAIL INDUSTRY This sub-section encompasses the data associated with various merchandise offered by organised retailers to their customers. It is not possible to incorporate all the merchandise offered by organised retailers because multi-format organised retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore and Big Bazaar) offers more than 30000 merchandise to their customers. Therefore, merchandise lines offered by organised retailers are determined in this chapter for the study of merchandise offered by organised retailers. Few types of merchandise have also been presented under these merchandise lines as indicators (representative) of respective merchandise lines. The merchandise offered by organised retailers (included in the study) has been exhibited as follows: Merchandise offered by multi-format (item) organised retailers 5.1.1 Merchandise offered by HyperCITY 5.1.2 Merchandise offered by Star Bazaar 5.1.3 Merchandise offered by more.MEGAStore 5.1.4 Merchandise offered by Big Bazaar 5.1.5 Merchandise offered by D Mart Merchandise offered by fashion item organised retailers 5.1.6 Merchandise offered by Big Megamart 5.1.7 Merchandise offered by West Side 5.1.8 Merchandise offered by Max Merchandise offered by the electronics item organised retailers 5.1.9 Merchandise offered by Reliance Digital 5.1.10 Merchandise offered by NEXT Merchandise offered by multi-format (item) organised retailers 5.1.1 Merchandise offered by HyperCITY 5.1.2 Merchandise offered by Star Bazaar 5.1.3 Merchandise offered by more.MEGAStore 5.1.4 Merchandise offered by Big Bazaar 5.1.5 Merchandise offered by D Mart Moreover, the merchandise portfolio analysis has also been done through the BCG Matrix to comprehend the merchandise portfolios of different organised retailers included in this study. 98 Merchandise lines of HyperCITY  Food & Grocery – Fruits & vegetables (onion, cabbage, watermelon, potato, blueberry, royal gala apple, ladyfinger, broccoli, yellow pears and so on), meat & fish (specialty eggs range, chicken salami, chicken cut portions, seer fish, sausages, mutton chops, pomfrets, marinated chicken and so on), ready & instant food (cornflakes, jams, noodles, chocolate spread, muesli, choco pie, pastas, snacks & chips, soups, baked potato chips, chocolates, tomato ketchup, choco fills, cakes & deserts, health foods, pickles & papads, honey, instant & gravy mixes, breakfast and cereals, canned foods, biscuits, popcorns and so on), staples (ghee & vanaspati, olive oils, edible oils, rice, pulses, soya chunks, wheat flour, maida, peanuts, besan, suji, ready spices, whole spices, mukhwas, dry fruits, pohas, Sabu dana, sugar, salt and so on), beverages (cold drink, tea bag, milk, coffee, bournvita, tade, mineral water, squashes, syrups, juices & health drinks, floured milk, etc.), International foods (olive oils, tomato soups, beans, walnuts, different type of olives, tea, bread pieces, sweet corns and so on)  Food & Grocery – Fruits & vegetables (onion, cabbage, watermelon, potato, blueberry, royal gala apple, ladyfinger, broccoli, yellow pears and so on), meat & fish (specialty eggs range, chicken salami, chicken cut portions, seer fish, sausages, mutton chops, pomfrets, marinated chicken and so on), ready & instant food (cornflakes, jams, noodles, chocolate spread, muesli, choco pie, pastas, snacks & chips, soups, baked potato chips, chocolates, tomato ketchup, choco fills, cakes & deserts, health foods, pickles & papads, honey, instant & gravy mixes, breakfast and cereals, canned foods, biscuits, popcorns and so on), staples (ghee & vanaspati, olive oils, edible oils, rice, pulses, soya chunks, wheat flour, maida, peanuts, besan, suji, ready spices, whole spices, mukhwas, dry fruits, pohas, Sabu dana, sugar, salt and so on), beverages (cold drink, tea bag, milk, coffee, bournvita, tade, mineral water, squashes, syrups, juices & health drinks, floured milk, etc.), International foods (olive oils, tomato soups, beans, walnuts, different type of olives, tea, bread pieces, sweet corns and so on)  Health & Beauty - Fragrances (deos, perfumes, premium perfumes and so on), cosmetics (eye range, lip range, make up remover, eyeliner kajal, nail colour remover and so on), dental & oral care (toothpaste, mouthwash, toothbrush and so on), hair care (hair gel range, silky potion, coconut hair oil, shampoos, conditioner, hair colour, OTC and so on), skin care (talcum, hair removal strips, nose strip, disposable razor, fairness cream & lotions, body lotion, lip care, day cream, herbal cosmetics, colour cosmetics, sanitary needs, grooming and so on) personal hygiene (hand sanitiser, body wash range, soap regular, moist soap, germ shield soap, hand washes, neem face pack, whisper ultra and so on), baby care (Lal tail, baby soap, Mamy Poko pants, baby top2toe wash, Woodward gripe water, baby buds, baby wipes / diapers, baby foods, baby / kids grooming and so on) 99 Home & Furniture – Homeware (travel goods, bag packs, outdoor bags, link bags, glassware range, utensils range, bath accessories, non-stick cookware, crockery range, kitchen implements, home decor range, bake wares range, storage / plastics, cutlery range, tumblers, plates, bowls, storage & nonstick containers, lunch boxes, hard & anodised cookware, cookers & cooking vassals, dinner sets, loose crockery, mugs, cups, soccer and so on), home needs (gardening range, tools range, bags and covers range, pet care range, electrical items and lighting, CFL, extension board, hangers, dryers and cleaning solutions, home care range, detergents, bath accessory, cabinet & mirrors, laundry aids, agarbatties, pooja needs, room fresheners, pesticides, scrubs, brooms, brushes, disposable wares, parties napkins, facial & car tissues, toilets & kitchen tools, foils, garbage bags, ladders, slopes, bath ware, artificial flowers, wall frames and so on), home linen (bath towels, cushion fillers, bedding set, bath mats, table covers, pillow covers, curtains, hand towels, pillow fillers, cushion filler, door mats, table mats, cushion covers, face towel, quilts, bed sheets, rugs & carpets, table cloths & napkins, apron, debaan set, lying and so on), furniture (stools, accessories, corner sofa w/d cushion, night stand, wardrobes, sofa beds, dining sets, TV units, sofa sets, bean bags, dresser, recliner, kids bunk beds, beds, corner sofas and so on), stationery (gifting needs, stationery set, school needs, pencil case metal, party items, office supplies, school bags, etc.), household and cleaning needs (mini pen, aluminium foil, dish wash liquid, toilet cleaner, detergent, all out refill, all out ultra-smart chip, fabric conditioner, household cleaning and so on)  Home & Furniture – Homeware (travel goods, bag packs, outdoor bags, link bags, glassware range, utensils range, bath accessories, non-stick cookware, crockery range, kitchen implements, home decor range, bake wares range, storage / plastics, cutlery range, tumblers, plates, bowls, storage & nonstick containers, lunch boxes, hard & anodised cookware, cookers & cooking vassals, dinner sets, loose crockery, mugs, cups, soccer and so on), home needs (gardening range, tools range, bags and covers range, pet care range, electrical items and lighting, CFL, extension board, hangers, dryers and cleaning solutions, home care range, detergents, bath accessory, cabinet & mirrors, laundry aids, agarbatties, pooja needs, room fresheners, pesticides, scrubs, brooms, brushes, disposable wares, parties napkins, facial & car tissues, toilets & kitchen tools, foils, garbage bags, ladders, slopes, bath ware, artificial flowers, wall frames and so on), home linen (bath towels, cushion fillers, bedding set, bath mats, table covers, pillow covers, curtains, hand towels, pillow fillers, cushion filler, door mats, table mats, cushion covers, face towel, quilts, bed sheets, rugs & carpets, table cloths & napkins, apron, debaan set, lying and so on), furniture (stools, accessories, corner sofa w/d cushion, night stand, wardrobes, sofa beds, dining sets, TV units, sofa sets, bean bags, dresser, recliner, kids bunk beds, beds, corner sofas and so on), stationery (gifting needs, stationery set, school needs, pencil case metal, party items, office supplies, school bags, etc.), household and cleaning needs (mini pen, aluminium foil, dish wash liquid, toilet cleaner, detergent, all out refill, all out ultra-smart chip, fabric conditioner, household cleaning and so on)  Fashion- Men’s (formal shirt, corduroy trouser, formal trousers, casual shirts, striped T-shirts, jacket, jeans, denim, ties, belts, shorts, boxer shorts, hanky, cargos, undergarments, sweaters, wind cheaters, track-pants, caps, hand globs, accessories and so on), women’s (printed kurta, dupatta, corduroy trouser, knit Churidar, house coat, nightys, leggings, salwar, denim, hanky, frocks, striped T-shirts, tops, woven pyjama jeans, skirts, inner wears, jackets, sweaters, wind cheaters, stole, ribbons, shorts, checked capris, bags, top with pants, bottom jegging, accessories and so on), boys (shirts, T-shirts, track pants, jeans, cargos, denim, jacket, caps, sweaters, accessories and so on), girls (tops, T- shirts, frocks, jackets, sweaters, ribbons, shorts, jeans, denims, leggings, skirts, inner wears, accessories and so on), newborn baby garments (infant 100 wears, diapers, accessories and so on), men’s footwear (formal shoes, sport shoes, sandals, slippers, socks, casual footwear, accessories and so on), women’s footwear (fashion footwear, shoes, flip flop, slippers, sandals, socks, accessories and so on), kids footwear (sport shoes, formal, school, sandals, socks, accessories and so on)  Electronics – Large appliances (microwaves, air conditioners, refrigerators, washing machines and so on), small appliances (mixer grinder & blender, juicers, toaster, vacuum cleaner, rice cooker, irons, water purifier, personal care, kettle & coffee maker, telephones, CD players, DVD players, printers, mobile & accessory, gas stove, induction earphone and so on), electronics (cameras, camcorders, laptops, accessories and so on), home entertainment (home theatres, LCD TV, flat panel TVs, LCD TVs, 3D TVs, speakers and so on)  Toys - Cars & jeeps, ball toys, board game, puzzles & board games, teddy bears, remote control toys, blocks & sorting, other sports, play sets, action figures, inflatable, soft cushions, do it yourself, other soft toys, track sets, infant utility, clay sets, guns & accessories, beach toys, helicopter auto frequency, cars & bikes, dolls and so on. Merchandise lines of HyperCITY  Sports - Health & fitness (fitness accessories, home gyms, weights, treadmills, health formulas, exercise bike and so on), sports (darts, tennis, badminton, boxing, volleyball, cycles, camping, carom, football, hockey, baseball, swimming, table tennis, cricket, squash, basketball, kids cycles, soccer table and so on)  Sports - Health & fitness (fitness accessories, home gyms, weights, treadmills, health formulas, exercise bike and so on), sports (darts, tennis, badminton, boxing, volleyball, cycles, camping, carom, football, hockey, baseball, swimming, table tennis, cricket, squash, basketball, kids cycles, soccer table and so on)  Private (store-level) Brands - Fresh Basket, Everyday (Everyday tissue hamper, coffee mug, range of Everyday fryums, Everyday range of jams, range of Everyday soups, Everyday range of dry fruits, Everyday range of pulses, ghee, rice and so on), Terzo (scrub pad, wipe and so on), Avorio (bath towel, single & double bed sheet and so on), Ebano (kasroll, porcelian dinner set, duffel, trolley and so on), Technix (Hb, Vc compact, kettel), Maxit (enduro cycle, exercise bike, treadmill kl), Waitrose (cornflakes, baked beans in tomato, porridge oats, olive oil, fusilli penne spaghetti, mince pies, puddings, cakes, biscuits & cookies, confectionery and so on) Merchandise portfolio analysis of HyperCITY through BCG Matrix Merchandise portfolio analysis of HyperCITY through BCG Matrix 101 Market Share High Low Market Growth High Food & Grocery Health & Beauty Home & Furniture Fashion Electronics Toys Stars Question Marks Low Private Brands Sports Cash Caws Dogs Figure 5.1: Merchandise portfolio analysis of HyperCITY Market Share Figure 5.1: Merchandise portfolio analysis of HyperCITY Food & grocery, health & beauty, home & furniture and fashion merchandise lines of HyperCITY are in star position in BCG Matrix, which reflects the high market share of merchandise under these lines and high market growth denotes an increase in the sales of these merchandise lines from past years. Electronics and toy merchandise lines of HyperCITY are in question mark position in BCG Matrix that shows the low market share of merchandise under these lines and high market growth represents an increase in the sales of these merchandise lines from past years. Private brands merchandise line of HyperCITY falls under the category of a cash cow in BCG Matrix that reveals the high market share of merchandise under this line and low market growth denotes that there is no increase in the sales of this merchandise line from past years. Sports merchandise line of HyperCITY is part of dog position in BCG Matrix, which demonstrates low market share and low market growth describes that there is a decrease in the sales of these merchandise lines from past years. 5.1.2 Merchandise offered by Star Bazaar 102 Star Bazaar offers merchandise under the merchandise lines like food & grocery, personal & home care, home & kitchen needs, fashion & accessories, international zone (merchandise) and star bazaar (store level) brands. Merchandise lines of Star Bazaar  Food & Grocery – Staples (pulses, cereals & flours, rice, basmati rice, spices, sugar, edible oil (sunflower oil, soybean oil, olive oil), ghee, dry fruits, flour, grains, sugar, Toor dal, Moong dal, Kabuli chana, green whole moong, chana dal, urad dal, red masoor, raw peanuts, floor lokwan, boiled rice, pohe, many more...), packaged food (biscuits & confectionery, cereals, sauces, jams, breakfast, noodles & vermicelli, soups, chips & savouries, papad & pickles, ready-to-eat preparations, pickles, ketchups, noodles, cornflakes, Indian sweet dessert, ready to mix, breakfast needs, many more...), beverages (tea & coffee, juices, aerated drinks, health drinks, cold drinks, flavoured milk, soya milk, squashes, cold coffee, coconut, many more...), dairy (milk, butter, cheese, dahi & yoghurt, shrikhand, paneer, lassi & chaas, ice cream and many more...), bakery & deli (pastry, cake, bread, toast, buns & breads, mousse, cakes, rolls, veg-mini meals, non-veg mini meals, sweets, veg pizza, pizza paneer, no-veg pizza, veg & chicken burger, chicken biryani, tandoori chicken, many more...), non-veg (meat, chicken, seafood, many more...), fruits & vegetables (fresh fruits and vegetables, imported fruits and vegetables, chopped and packed fruits and vegetables, many more...), frozen (veg snacks, non-veg snacks, green peas, paneer, srikhand, dahi, ghee, milk, cheese, creams & pet needs many more..).  Personal & Home Care – Cleaning aids (detergents - powder & bars, floor cleaners, toilet cleaners, bathroom cleaners, insect repellents, room fresheners, disposable rags, mops, brushes, scrubs, pooja needs, many more...), health & beauty (bathing soap, face pack, shower gel, shampoo & conditioners, creams, lotions & scrub, hand wash, hair care products, men’s grooming, sanitary needs, body wash, hair oil, razor, shaving gel & cream, moisturiser, baby care, diapers, body sprays & deodorants, many more...).  Personal & Home Care – Cleaning aids (detergents - powder & bars, floor cleaners, toilet cleaners, bathroom cleaners, insect repellents, room fresheners, disposable rags, mops, brushes, scrubs, pooja needs, many more...), health & beauty (bathing soap, face pack, shower gel, shampoo & conditioners, creams, lotions & scrub, hand wash, hair care products, men’s grooming, sanitary needs, body wash, hair oil, razor, shaving gel & cream, moisturiser, baby care, diapers, body sprays & deodorants, many more...). Merchandise lines of Star Bazaar  Home & Kitchen Needs – Kitchenware (cookers, container sets, non-stick sets, utensils, kitchen, appliances, dinner sets, crockery, glassware, home plastics, hangers, buckets, bath mug, clip hangers, fridge bottles, lunch boxes,  Home & Kitchen Needs – Kitchenware (cookers, container sets, non-stick sets, utensils, kitchen, appliances, dinner sets, crockery, glassware, home plastics, hangers, buckets, bath mug, clip hangers, fridge bottles, lunch boxes, 103 microwave safe lunch boxes, knife, tea & coffee wares nonstick wares, cook & serve wares, drink wares, tray, glasses, cattle, cups, mugs, many more...), electronics (fans, bells, switches, extension board, CFL, batteries, dry & steam irons, mixer-grinders, gas burners, microwaves, vacuum cleaners, power strips, water purifiers, rice cookers, toasters, fans, small fans, table lamp, water heater, cooler, pedestal fans, table & tower fans, gas stoves, food processor, sewing machine, sandwich maker, speakers, DVD players, telephones, music system, OTG, watches, many more...), home décor (bed sheets, blankets, mats, pillow, pillow covers, bath towels, hand towels, cushion covers, curtains, dining linen, decorative plants, floating stool, floor carpet, plastic table chair, almirah, door mats, floor mats, window curtains, door curtains, artificial flowers, many more...), luggage (baby buggies & with umbrella, strollers, trolley bags, laptop bags, duffel bags, office bags, backpacks, travel bags, helmets, many more...), toys (toys for different age groups below 1 to 10+, trolley, bag trolley, remote control games, friction car, soft toys, board games, dolls, building blocks, ball, educational toys, infant toys, plastic toys, computer games, bat balls, rackets, table tennis, pads, cycle of children, football, volleyball, many more...), stationery (school needs, office needs, writing implements, creative stationery, books, magazine, computer stationery, letter pad, papers, gum, pen, pencil, sketches, pencil box, notebook, diary, stickers, boxes, calculators, box files, story books, books for small children & children, novels, bottles for students, many more...)  Fashion & Accessories – Men's wear (western wear, formal wear, ethnic wear, shirts, T-shirts, fashion shirts, denim jeans, duke T-shirts, denim T- shirts, casual shirts, plain shirts, party wear, trousers, casual pants, shirts, undergarments, shorts, briefs, bath towel, track pants, checked printed shirts, short track pants (of John Player, Belmonte, Perter England brand) & many more…..), women's wear (western wear, formal wear, ethnic wear, T-shirts, printed T-shirts, kurta, stretch leggings, shorts, long sleeveless nighty, plan bra, printed woven pyjama, full pyjamas, leggings, short pyjamas, gagharas, branded kurtas, churidars, dress materials, silk printed sarees, sarees, & many more…), kid's wear (western wear, formal wear, ethnic wear, fashion T-shirts, tops, printed T-shirts, junior boys shirts, track pants, knit Capri, knit shorts, infant wears & utilities, katmos brand for kids and many more…..), 104 accessories (fashionable belts, bags, earrings, bracelets, wallets, handbags, lockets, hair bands, goggles, hanky, many more...), footwear (formal footwear, casual footwear, ethnic footwear, fashion footwear, women sandals, home footwear, kids footwear, fashion sandals, formal shoes, casual buddy shoes, sport shoes, casual shoes, men’s sleeper, chappals, flotter for men, men’s sport shocks, many more...)  International Zone (Merchandise) – Brands of Tesco, blueberry, strawberry, peaches, tomato soups, pears, salsa, cookies, vegetable soups, mushrooms, tomato purees, almonds, vegetables, cashews & many more…  International Zone (Merchandise) – Brands of Tesco, blueberry, strawberry, peaches, tomato soups, pears, salsa, cookies, vegetable soups, mushrooms, tomato purees, almonds, vegetables, cashews & many more…  Star Bazaar (Store Level) Brands – In the categories like food & grocery, cleaning aids, packaged food and beverages, personal care, men’s wear, women wear, kid’s wear, pooja needs, tea, jams, corn flakes, snacks, sugar, pulses, flour, suji, rice, pohe & many more... Merchandise lines of Star Bazaar  Star Bazaar (Store Level) Brands – In the categories like food & grocery, cleaning aids, packaged food and beverages, personal care, men’s wear, women wear, kid’s wear, pooja needs, tea, jams, corn flakes, snacks, sugar, pulses, flour, suji, rice, pohe & many more... Merchandise portfolio analysis of Star Bazaar through BCG Matrix Market Share Merchandise portfolio analysis of Star Bazaar through BCG Matrix Market Share Merchandise portfolio analysis of Star Bazaar through BCG Matrix Market Share Merchandise portfolio analysis of Star Bazaar through BCG Matrix 105 High Low Market Growth High Food & Grocery Fashion & Accessories Personal & Home Care Stars Question Marks Low Home & Kitchen Needs Star Bazaar Brands International Zone (Merchandise) Cash Caws Dogs Figure 5.2: Merchandise portfolio analysis of Star Bazaar Figure 5.2: Merchandise portfolio analysis of Star Bazaar Food & grocery, fashion & accessories and personal & home care merchandise lines of Star Bazaar are in star position in BCG Matrix, which reflects the high market share of merchandise under these lines and high market growth denotes an increase in the sales of these merchandise lines from past years. No merchandise line of Star Bazaar is in question mark position in BCG Matrix. Home & kitchen needs and Star Bazaar brands (store level brands) merchandise lines of Star Bazaar fall under the category of a cash cow in BCG Matrix that shows the high market share of merchandise under this line and low market growth denotes that there is no increase in the sales of this merchandise line from past years. International zone (International merchandise) merchandise line of Star Bazaar is part of dog position in BCG Matrix, which demonstrates low market share and low market growth describes that there is a decrease in the sales of these merchandise lines from past years. Merchandise lines of more.MEGAStore  Grocery: Frozen & dairy products, beverages, imported food, FMCG products, processed food, prepared food, ready to cook / prepared food, packaged & loose staples, milk, butter, curd, ghee, paneer, cheese slices, cornflakes, choco marvel, pickles, fruit drinks, jams, instant soups, honey, papads, oats, pasta, cola, premium tea, green tea, CTC tea, coffee, namkeens, ketchups, noodles, cheese balls, vermicelli, regional specialties, instant mixes, soan papdi, rice, pulses, sugar, salt, flour, refined oil, wheat, bread, spices and so on…  Home & Personal Care: Home needs & home upkeep, cosmetics, toilet cleaners, phenyl, surface cleaners, scouring pads, detergents, detergent bars, dish wash bars, tissues, steel scrubber, room freshener, insect repellants, naphthalene balls, dish wash liquid, soap, facial tissue, toothbrushes, liquid hand wash, deodorants, shaving razor, baby diapers and so on…  Home & Personal Care: Home needs & home upkeep, cosmetics, toilet cleaners, phenyl, surface cleaners, scouring pads, detergents, detergent bars, dish wash bars, tissues, steel scrubber, room freshener, insect repellants, naphthalene balls, dish wash liquid, soap, facial tissue, toothbrushes, liquid hand wash, deodorants, shaving razor, baby diapers and so on…  Fruits & Vegetables: Onion, cabbage, watermelon, potato, blueberry, royal gala apple, ladyfinger, broccoli, yellow pears, fresh non-vegetarian and so on  General Merchandise: books, crockery, cookware, do it yourself & auto accessories, furniture, home decor products, luggage, stationery, toys and so on.  Electronic Merchandise: Audio & video merchandise, information technology merchandises, large white appliances, small white appliances and so on  Electronic Merchandise: Audio & video merchandise, information technology merchandises, large white appliances, small white appliances and so on  Computers & Accessories: Laptops, Printers & Scanners, Software, Memory & Storage devices, Cables & Connectors, Chargers & Adaptors, Headphones & Headsets, Keyboards, Mouse & Mouse Pads, Webcams, CD/DVD Drives & Writers, Routers and so on  Computers & Accessories: Laptops, Printers & Scanners, Software, Memory & Storage devices, Cables & Connectors, Chargers & Adaptors, Headphones & Headsets, Keyboards, Mouse & Mouse Pads, Webcams, CD/DVD Drives & Writers, Routers and so on  Mobile Phones & Accessories: Smartphones, tablets, memory card, car charger, Bluetooth headset, corded phone, cordless phone and so on…  Mobile Phones & Accessories: Smartphones, tablets, memory card, car charger, Bluetooth headset, corded phone, cordless phone and so on…  Apparels: Men, women, infant & children's apparels i.e. 5.1.3 Merchandise offered by more.MEGAstore more.MEGAStore offers merchandise under the merchandise lines like grocery, home & personal care, fruits & vegetables, general merchandise, electronic 106 merchandise, computers & accessories, mobile phones & accessories, apparels, footwear and sports. Merchandise lines of more.MEGAStore western wear, formal wear, ethnic wear, shirts, T-shirts, fashion shirts, denim jeans, duke T-shirts, denim T-shirts, casual shirts, plain shirts, party wear, trousers, casual pants, 107 shirts, undergarments, shorts, briefs, bath towel, track pants, checked printed shirts, short track pants… shirts, undergarments, shorts, briefs, bath towel, track pants, checked printed shirts, short track pants…  Footwear: Footwear & accessories for men, women, kids and infants i.e. casual shoes, formal shoes, sandals, slippers & flip-flops, sports shoes, flats, low heels, mid heels, flip-flops, chappals, infant shoes and so on..  Sports: Fitness (home gyms, weights, fitness accessories, treadmills, exercise bike and so on), sporting goods (squash, basketball, darts, boxing, volleyball, cycles, camping, carom, baseball, swimming, tennis, badminton, table tennis, cricket, kids cycles, soccer table, football, hockey and so on). Merchandise portfolio analysis of more.MEGAstore through BCG Matrix Market Share High Low Merchandise portfolio analysis of more.MEGAstore through BCG Matrix Market Share High Low Merchandise portfolio analysis of more.MEGAstore through BCG Matrix 108 Market Growth High Grocery Home & Personal Care Apparels General Merchandise Mobile Phones & Accessories Computers & Accessories Stars Question Marks Low Footwear Electronics Merchandise Fruits & Vegetables Sports Cash Caws Dogs Figure 5.3: Merchandise portfolio analysis of more.MEGAStore Figure 5.3: Merchandise portfolio analysis of more.MEGAStore Grocery, home & personal care and apparels merchandise lines of more.MEGAStore are in star position in BCG Matrix, which exhibits the high market share of merchandise under these lines and high market growth denotes an increase in the sales of these merchandise lines from past years. General merchandise, mobile phones & accessories and computers & accessories merchandise lines of more.MEGAStore are in question mark position in BCG Matrix that presents the low market share of merchandise under these lines and high market growth represents an increase in the sales of these merchandise lines from past years. Footwear and electronics merchandise lines of more.MEGAStore fall under the category of a cash cow in BCG Matrix that tenders the high market share of merchandise under this line and low market growth denotes that there is no increase in the sales of this merchandise line from past years. Fruits & vegetables and sports merchandise lines of more.MEGAStore are part of dog position in BCG Matrix, which reflects low market share and low market growth describes that there is a decrease in the sales of these merchandise lines from past years. 5.1.4 Merchandise offered by Big Bazaar Big Bazaar offers merchandise under the merchandise lines like grocery & staples bazaar, fashion bazaar, electronics bazaar, personal bazaar, kitchen & dining bazaar, bed & bath bazaar, home bazaar, mobile bazaar, furniture bazaar, toys, sports 109 & outdoors, gifts, office bazaar and luggage bazaar, farm fresh and books, magazines & stationary bazaar. Merchandise lines of Big Bazaar  Grocery & Staples Bazaar: Golden harvest (rice, flour, staples, carrels, cooking mediums, spices, loose grocery etc.), dry fruits, refrigerated products (milk, fruit juices, tea, coffee, cold drinks etc.), biscuits, breakfast products, carrels, canned foods, namkeen, chips & wafers, ready meals, chef’s needs (chocolate, jams, salad dressing, pickles, breads, desserts, spices mixes, pastes, Chinese noodles etc.), International foods (Thai, Italian, Chinese, Mexican, Continental etc.), gourmet food (beverages, daily needs, edible oil, grains, health products, herbs & powders, instant food, spices, pickles & sauces, pulses, refreshments, snacks, etc.) and so on…  Grocery & Staples Bazaar: Golden harvest (rice, flour, staples, carrels, cooking mediums, spices, loose grocery etc.), dry fruits, refrigerated products (milk, fruit juices, tea, coffee, cold drinks etc.), biscuits, breakfast products, carrels, canned foods, namkeen, chips & wafers, ready meals, chef’s needs (chocolate, jams, salad dressing, pickles, breads, desserts, spices mixes, pastes, Chinese noodles etc.), International foods (Thai, Italian, Chinese, Mexican, Continental etc.), gourmet food (beverages, daily needs, edible oil, grains, health products, herbs & powders, instant food, spices, pickles & sauces, pulses, refreshments, snacks, etc.) and so on…  Farm Fresh: Fruits & vegetables  Kitchen & Dining Bazaar: Kitchen linen (aprons, coasters, napkins, sets, table, clothes etc.), cookware (bake ware, kadai, pans, pots, pressure cookers, sets, tawa, tops etc.), serve ware (beverage dispensers, casseroles, jugs, serving bowls, serving dishes, sets, tea pots, trays & platters, etc.), crockery (bowls, cups & mugs, drinkware, plates, sets, etc.), cutlery (children's cutlery, chopsticks, forks, knives, ladles, servers, sets, spoons, tongs, etc.), storage & containers (canisters, food storage, garbage bins, microwave sets, racks & holders, sets, storage bags, tiffin, vacuum flasks, water bottles & jugs, etc.), kitchen utilities (plates, bowls, glasses, baking tools, bar accessories, choppers, graters & slicers, cutters, cutting boards, more utensils & tools, openers, peelers, sieves & spinners, whisks & spatula, etc.) and so on…  Bed & Bath Bazaar: Bed sheets (single bed sheets, double bed sheets), bed covers, pillows & inserts, blankets & quilts, bedding collections (bedding sets, divan sets etc.), towels (bath towels, hand towels, face towels etc.), bath accessories (holders & dispensers, shower accessories, toilet accessories, bathroom sets etc.) and so on…  Bed & Bath Bazaar: Bed sheets (single bed sheets, double bed sheets), bed covers, pillows & inserts, blankets & quilts, bedding collections (bedding sets, divan sets etc.), towels (bath towels, hand towels, face towels etc.), bath accessories (holders & dispensers, shower accessories, toilet accessories, bathroom sets etc.) and so on…  Books, Magazines & Stationary Bazaar: Books (children, comics, educational, fiction, non-fiction etc.), magazines (auto, b2b, business, children, current affairs, fashion, home, lifestyle, photography, technology, travel etc.),  Books, Magazines & Stationary Bazaar: Books (children, comics, educational, fiction, non-fiction etc.), magazines (auto, b2b, business, children, current affairs, fashion, home, lifestyle, photography, technology, travel etc.), 110 stationary (boards & pads, key chains, notebooks & diaries, office stationery, papers, pens, tapes & adhesives, etc.) & so on… stationary (boards & pads, key chains, notebooks & diaries, office stationery, papers, pens, tapes & adhesives, etc.) & so on… stationary (boards & pads, key chains, notebooks & diaries, office stationery, papers, pens, tapes & adhesives, etc.) & so on…  Luggage Bazaar: Airbags & duffel bags, backpacks, briefcases, cabin trolley, hiking bags, luggage sets, messenger bags, suitcases, travel accessories, trolleys & so on…  Luggage Bazaar: Airbags & duffel bags, backpacks, briefcases, cabin trolley, hiking bags, luggage sets, messenger bags, suitcases, travel accessories, trolleys & so on…  Home Bazaar: Home décor (albums & photo frames, art & wall décor, artificial flowers & potpourri, candles & candle stands, clocks, collectables, decorative storage, devotional products, handicrafts, home fragrances, hukkas, lamps & shades, metal ware & antiques, sculptures & figurines, vases, wind chimes & bells, etc.), home furnishing (carpets, rugs & durries, curtains, cushion covers, cushions, doors and bath mats, etc.), auto care (bike, car audio, car exterior, car interior, cleaning, navigation etc.), tools & hardware (bath ware, cleaning, gardening, hand tools etc.), storage & organisers (baskets & containers, cloth hangers & racks, grab n go furniture, laundry organisers, racks & holders, safes & lockers, trash cans, etc.), wedding collection & wedding accessories, utilities, cleaning aids (whitening agents, fabric softness, starch, detergents bars, liquid, powders etc.), pet care & so on…  Home Bazaar: Home décor (albums & photo frames, art & wall décor, artificial flowers & potpourri, candles & candle stands, clocks, collectables, decorative storage, devotional products, handicrafts, home fragrances, hukkas, lamps & shades, metal ware & antiques, sculptures & figurines, vases, wind chimes & bells, etc.), home furnishing (carpets, rugs & durries, curtains, cushion covers, cushions, doors and bath mats, etc.), auto care (bike, car audio, car exterior, car interior, cleaning, navigation etc.), tools & hardware (bath ware, cleaning, gardening, hand tools etc.), storage & organisers (baskets & containers, cloth hangers & racks, grab n go furniture, laundry organisers, racks & holders, safes & lockers, trash cans, etc.), wedding collection & wedding accessories, utilities, cleaning aids (whitening agents, fabric softness, starch, detergents bars, liquid, powders etc.), pet care & so on…  Personal Bazaar: Haircare (hair colours, hair conditioners, hair gels & creams, hair oil, shampoos and so on), skin care (body lotions, creams, face wash, shower gel, skin care accessories, soaps, sunscreens etc.), oral care (toothbrush, toothpaste, tung cleaner, mouth fresheners etc.), medicare products, saving needs, deodorants, perfumes, beauty and wellness for women- eye care, hair care, lip care, makeup, nail care, perfumes, skin care, spa products & so on…  Personal Bazaar: Haircare (hair colours, hair conditioners, hair gels & creams, hair oil, shampoos and so on), skin care (body lotions, creams, face wash, shower gel, skin care accessories, soaps, sunscreens etc.), oral care (toothbrush, toothpaste, tung cleaner, mouth fresheners etc.), medicare products, saving needs, deodorants, perfumes, beauty and wellness for women- eye care, hair care, lip care, makeup, nail care, perfumes, skin care, spa products & so on…  Furniture Bazaar: Stools, accessories, corner sofa w/d cushion, night stand, wardrobes, sofa beds, dining sets, TV units, sofa sets, bean bags, dresser, recliner, kids bunk beds, beds, corner sofas etc. Merchandise lines of Big Bazaar  Furniture Bazaar: Stools, accessories, corner sofa w/d cushion, night stand, wardrobes, sofa beds, dining sets, TV units, sofa sets, bean bags, dresser, recliner, kids bunk beds, beds, corner sofas etc. Merchandise lines of Big Bazaar  Electronics Bazaar: Appliances - geysers & heaters, air conditioners (split ACs, window ACs), fans & air coolers, irons (dry iron, steam irons), vacuum & floor care (steam & carpet cleaners, vacuum cleaners), washing & drying (dryers, washing machine), lighting & power, personal grooming (epilators, hair curlers, hair dryers, hair straighteners, hair stylers, massagers, shavers, skin care, trimmers, etc.), kitchen appliances (blenders, coffee makers, electric cookers, electric kettle, food processors, juicer, mixer, grinder, microwave & 111 oven, pop-up toasters, refrigerators, roti makers, sandwich makers, water purifiers etc.), stoves (chimneys, cooktops, gas stoves), sewing machines, insect killers; entertainment - TV & entertainment (audio video accessories, home audio, home theatre, LCD & LED, media players, sound machine, DJ consoles, DJ control, microphone, 3D glasses, etc.), cameras (camcorders, camera accessories, camera bags & cases, digital cameras, DSLR cameras, flashes, lens accessories, lenses, tripods & monopods, etc.), i-pods & music players, MP3/MP4 players, gaming (games, gaming accessories, gaming consoles, etc.) & so on…  Office Bazaar: Laptops & tablets (computer accessories, desktops, laptop accessories, laptops, tablet accessories, tablets), memory & storage (hard drives, memory cards, and USB flash drives etc.), IT hardware (routers), printers & cartridges, scanners & copiers, software and so on…  Office Bazaar: Laptops & tablets (computer accessories, desktops, laptop accessories, laptops, tablet accessories, tablets), memory & storage (hard drives, memory cards, and USB flash drives etc.), IT hardware (routers), printers & cartridges, scanners & copiers, software and so on…  Mobile Bazaar: Mobiles, telephones, batteries & chargers, Bluetooth, covers & cases, headsets, mobile accessories and so on…  Mobile Bazaar: Mobiles, telephones, batteries & chargers, Bluetooth, covers & cases, headsets, mobile accessories and so on…  Mobile Bazaar: Mobiles, telephones, batteries & chargers, Bluetooth, covers & cases, headsets, mobile accessories and so on…  Toys, Sports & Outdoors: Battery operated toys, blocks, puzzles & learning, board games, CDs & DVDs, dolls & accessories, fun games & toys, guns & action figures, musical, play sets, ride-on, push & pulls, soft toys, sports, vehicles, gear, gym, health care, sports, supplements, digital blood pressure monitor, ladies umbrella, soccer, yoga mat, cricket set, kids umbrella, auto glucose monitor, glucose monitoring system, tennis racket, titanium ball, mega gains, skateboard, rubber roller skates, inflatable ball and so on…  Toys, Sports & Outdoors: Battery operated toys, blocks, puzzles & learning, board games, CDs & DVDs, dolls & accessories, fun games & toys, guns & action figures, musical, play sets, ride-on, push & pulls, soft toys, sports, vehicles, gear, gym, health care, sports, supplements, digital blood pressure monitor, ladies umbrella, soccer, yoga mat, cricket set, kids umbrella, auto glucose monitor, glucose monitoring system, tennis racket, titanium ball, mega gains, skateboard, rubber roller skates, inflatable ball and so on…  Fashion Bazaar: Apparel & Accessories Men’s - Ethnic wear, innerwear, jeans, shirts, shorts, swimwear, sportswear, sweatshirts, T-shirts, trousers, denim, casuals, formals, accessories (bags, belts, caps, cufflinks, gift sets, handkerchief's, socks, sunglasses, ties, wallets, men’s fashion watches & sunglasses, men’s luxury watches, men’s sports watches & sunglasses), Footwear: casual shoes, formal shoes, sandals, slippers & flip-flops, sports shoes, Apparel & Accessories Women’s - Dresses, ethnic wear, jeans, lingerie, salwar kameez, sarees, shorts, sleepwear, sportswear, swimwear, T- shirts & tops, trousers, night wears, fit & healthy wear, casual wear, fashion wear, accessories (scarf & stoles, socks & hosiery, women’s fashion watches & sunglasses, women’s luxury watches,), handbags (clutches, handbags, hobos, sling bags, totes, wallets & pouches), jewelry (anklets, bangles,  Fashion Bazaar: Apparel & Accessories Men’s - Ethnic wear, innerwear, jeans, shirts, shorts, swimwear, sportswear, sweatshirts, T-shirts, trousers, denim, casuals, formals, accessories (bags, belts, caps, cufflinks, gift sets, handkerchief's, socks, sunglasses, ties, wallets, men’s fashion watches & sunglasses, men’s luxury watches, men’s sports watches & sunglasses), Footwear: casual shoes, formal shoes, sandals, slippers & flip-flops, sports shoes, Apparel & Accessories Women’s - Dresses, ethnic wear, jeans, lingerie, salwar kameez, sarees, shorts, sleepwear, sportswear, swimwear, T- shirts & tops, trousers, night wears, fit & healthy wear, casual wear, fashion wear, accessories (scarf & stoles, socks & hosiery, women’s fashion watches & sunglasses, women’s luxury watches,), handbags (clutches, handbags, hobos, sling bags, totes, wallets & pouches), jewelry (anklets, bangles, 112 bracelets, combo sets, earrings, necklaces, nose pins, pendants, rings, sets, jewelry cleaners), Footwear: Women’s - flats, low heels, mid-heels, sandals, slippers & flip-flops, sports shoes, chappals, Apparel & Accessories Kids – girls & boys jeans, shorts, skirts, T-shirts, tops, trousers, accessories (sunglass, women’s sunglasses, kids watch), Footwear: sandal, slipper, shoes, chappals, bracelets, combo sets, earrings, necklaces, nose pins, pendants, rings, sets, jewelry cleaners), Footwear: Women’s - flats, low heels, mid-heels, sandals, slippers & flip-flops, sports shoes, chappals, Apparel & Accessories Kids – girls & boys jeans, shorts, skirts, T-shirts, tops, trousers, accessories (sunglass, women’s sunglasses, kids watch), Footwear: sandal, slipper, shoes, chappals, Apparel Baby – infant clothing, Footwear Baby – infant shoes and so on…  Gifts: Cakes & chocolates, gift vouchers, gift sets, valentine's day, Diwali, charity, flowers, greeting cards, seasonal gifts, sweets & dry fruits  Gifts: Cakes & chocolates, gift vouchers, gift sets, valentine's day, Diwali, charity, flowers, greeting cards, seasonal gifts, sweets & dry fruits Merchandise portfolio analysis of Big Bazaar through BCG Matrix Market Share High Low Merchandise portfolio analysis of Big Bazaar through BCG Matrix Market Share High Low Merchandise portfolio analysis of Big Bazaar through BCG Matrix Market Share High Low Merchandise portfolio analysis of Big Bazaar through BCG Matrix 113 Market Growth High Grocery & Staples Bazaar Fashion Bazaar Electronics Bazaar Personal Bazaar Kitchen & Dining Bazaar Bed & Bath Bazaar Home Bazaar Mobile Bazaar Furniture Bazaar Stars Question Marks Low Toys, Sports & Outdoors Gifts Office Bazaar Luggage Bazaar Farm Fresh Books, Magazines & Stationery Bazaar Cash Caws Dogs Figure 5.4: Merchandise portfolio analysis of Big Bazaar Market Growth Figure 5.4: Merchandise portfolio analysis of Big Bazaar Grocery & staples bazaar, fashion bazaar, electronics bazaar, personal bazaar and kitchen & dining bazaar merchandise lines of Big Bazaar are in star position in BCG Matrix, which reflects the high market share of merchandise under these lines and high market growth denotes an increase in the sales of these merchandise lines from past years. Bed & bath bazaar, home bazaar, mobile bazaar and furniture bazaar merchandise lines of Big Bazaar are in question mark position in BCG Matrix that presents the low market share of merchandise under these lines and high market growth represents an increase in the sales of these merchandise lines from past years. Toys, sports & outdoors, gifts, office bazaar and luggage bazaar merchandise lines of Big Bazaar fall under the category of a cash cow in BCG Matrix that shows 114 the high market share of merchandise under this line and low market growth denotes that there is no increase in the sales of this merchandise line from past years. Farm fresh and books, magazines & stationery bazaar merchandise lines of Big Bazaar are part of dog position in BCG Matrix, which reflects low market share and low market growth describes that there is a decrease in the sales of these merchandise lines from past years. 5.1.5 Merchandise offered by D Mart D Mart offers merchandise under the merchandise lines like foods, garments, toiletries & beauty merchandise, kitchenware, home appliances, bags & trolleys, footwear, bed & bath linen, frozen, toys & games and stationery. 115 Merchandise lines of D Mart Merchandise lines of D Mart  Foods- Ready to cook & eat, noodles, pickles, jams, ketchups, papad, fryams, ghee, sweets, farshan, health drinks, baby foods, grocery items, masala, breakfast items, dry fruits, chocolates, loose grocery, tea, coffee, biscuits, fruits, bakery, vegetables, insecticides and many more  Frozen- Milk, curd, whey, lassi, paneer, cream, cheese, butter, amrakhand, srikahnd, ice creams, pee, ready to eat and much more  Toiletries & Beauty Merchandises- Sanitary, talc, deo, cosmetics, dental care, tissues, hair care, fem care, detergents, hand washers, bath soaps, OTC, household cleanings, accessories, jewellery, bangles and many more  Garments- o Men’s- Shirt, casual shirts, T-shirts, kurta, jacket, trousers, jeans, ties, belts, shorts, boxer shorts, hanky, cargos, pajamas, undergarments, dress materials, sweaters, windcheaters, track-pants, caps, hand gloves, accessories and many more o Ladies- Kurta, kaftan, house coat, nighty, leggings, salwar, hanky, dress materials, frocks, tops, jeans, skirts, inner wears, jackets, sweaters, sarees, windcheaters, stole, ribbons, shorts, dupattas, accessories and many more o Boys- Shirts, T-shirts, jeans, cargos, denim, jacket, dhoti-kurta, caps, sweaters, accessories and many more o Girls- Tops, frocks, jackets, sweaters, ribbons, shorts, jeans, denim, salwar-kurtas, ghagra-choli, leggings, skirts, inner wears, accessories and many more o New Born Baby Garments- Infant wears, diapers, accessories and many more o New Born Baby Garments- Infant wears, diapers, accessories and many more  Bed & Bath Linen- Hand towels, bath towels, kitchen towels, bedsheets, pillows, cushions & covers, curtains, blankets, matrices, door mats, dusters, mops & cleaners, brushes & scrubbers, dustbins, tubs & buckets, accessories and many more  Footwear-  Footwear- 116 o Men’s- Formal shoes, sports shoes, sandals, slippers, socks, casual footwear, accessories and many more o Ladies- Shoes, slippers, sandals, socks, accessories and many more o Kids- Sports shoes, slippers, sandals, socks, accessories  Kitchenware- Induction cooktop, cookware, steel, non-sticks, pans & tawas, kadais, pressure cookers, gas stoves & lighters, kitchen utilities, containers, microwave safeware, Tiffin, crockery, plates & bowls, cutlery, dinner sets, accessories  Home Appliances- Iron, fans (ceiling and table), food processors, music system, rice cookers, torch, bells, water purifiers, toasters, juicer mixer & grinders, sandwich makers, plastic chairs, accessories  Bags & Trolleys- Shopping bags, college bags, laptop bags, office bags, air bags, trolleys, suitcases accessories  Stationery- Books, novels, notebooks, pen, pencils, papers, pen stands, punch machine, staplers, pen drives, CD & DVDs, accessories  Toys & Games- Soft toys, plastic toys, wood toys, mind games Merchandise portfolio analysis of D Mart through BCG Matrix Market Share  Toys & Games- Soft toys, plastic toys, wood toys, mind games Merchandise portfolio analysis of D Mart through BCG Matrix 117 High Low Market Growth High Foods Garments Toiletries & Beauty Merchandise Kitchenware Home Appliances Bags & Trolleys Stars Question Marks Low Footwear Bed & Bath Linen Frozen Toys & Games Stationery Cash Caws Dogs Figure 5.5: Merchandise portfolio analysis of D Mart High Low Market Growth High Foods Garments Toiletries & Beauty Merchandise Kitchenware Home Appliances Bags & Trolleys Stars Question Marks Low Footwear Bed & Bath Linen Frozen Toys & Games Stationery Cash Caws Dogs Figure 5.5: Merchandise portfolio analysis of D Mart Figure 5.5: Merchandise portfolio analysis of D Mart Foods, garments, toiletries & beauty merchandise and kitchenware merchandise lines of D Mart are in star position in BCG Matrix, which reflects the high market share of merchandise under these lines and high market growth denotes an increase in the sales of these merchandise lines from past years. Home appliances and bags & trolleys merchandise lines of D Mart are in question mark position in BCG Matrix that shows the low market share of merchandise under these lines and high market growth represents an increase in the sales of these merchandise lines from past years. Footwear and bed & bath linen merchandise lines of D Mart fall under the category of a cash cow in BCG Matrix that presents the high market share of merchandise under this line and low market growth denotes that there is no increase in the sales of this merchandise line from past years. Frozen, toys & games and stationery merchandise lines of D Mart are part of dog position in BCG Matrix, which reflects low market share and low market growth describes that there is a decrease in the sales of these merchandise lines from past years. 5.1.6 Merchandise offered by Big Megamart 5.1.6 Merchandise offered by Big Megamart 118 Big Megamart offers merchandise under the merchandise lines like youth wear, women's wear, men's wear, kid's wear, home mart, travel mart, footwear, fashion mart and toys. Merchandise lines of Big Megamart  Women's Wear – Ethnic wear (brands- Karigari, W, Nayans, Sushilas, Aurelia, Huur, Neerus, Zariwala, Estyle, Charms by Prafful, Leisha), western wear (brands- Cherokee, Allen Solly, Scullers, Auburn Hill, Expozay, Tantra, Miss Players, Global Trendz), lingerie (brands- Amante, Lovable, City Girl, Juliet)  Men's Wear - Formal wear (brands- Louis Philippe, Ruggers, Van Heusen, Peter England, Excalibur Elitus, Auburn Hill, Pan America, Park Avenue, Celebrity, Options, Wills Lifestyle), casual wear (brands- Cherokee, US Polo, Scullers, Farenheit, Indigo Nation, Allen Solly, Parx, Enryca, V., Colour Plus, Classic Polo, Lerros, Giordano, Dockers, Kmark, Tantra, Prostyle, Allen Solly Youth), ethnic wear (brands- Manu, Karigari, Yash), suits & blazers (brands- Parx, Auburn Hill, Excalibur, Elitus, Belmonte, Shapes, Thomas Scott, Wills Lifestyle, John Players, Notting Hill), innerwear (brands- VIP, Hanes, Skinn, Euro)  Youth Wear – Denims (brands- Colt, Pepe, Newport, Wrangler, Lee, Levis, Lemax, Cherokee, Signature, Spykar, Jealous 21, Flying Machine), casual wear (brands- Ruggers, Dockers, Kappa, Indian Terrain, Colt, Spykar, Flying Machine, Lee, Wrangler, Levis, V., Stori, Bossini, MTV, Bay Island, Cherokee, UCB, Basics, Tantra, Crocodile, Classic Polo, Greg Norman), sportswear (brands- Edge, Reebok, Nike),  Youth Wear – Denims (brands- Colt, Pepe, Newport, Wrangler, Lee, Levis, Lemax, Cherokee, Signature, Spykar, Jealous 21, Flying Machine), casual wear (brands- Ruggers, Dockers, Kappa, Indian Terrain, Colt, Spykar, Flying Machine, Lee, Wrangler, Levis, V., Stori, Bossini, MTV, Bay Island, Cherokee, UCB, Basics, Tantra, Crocodile, Classic Polo, Greg Norman), sportswear (brands- Edge, Reebok, Nike),  Kid's Wear - Apparel wears (brands- Gini & Jony, Drop, Seals, Globe, Disney Kids, Isabelle, Karigari, Myfaa, UCB Kids),  Toys - (Brands - Mera Toys)  Footwear – (brands- Bata, Adidas, Fila, Action and exclusive international brands) o Men’s- Formal shoes, sports shoes, Sandals, Slippers, socks, Casual Footwear, accessories and so on o Ladies- Shoes, Slippers, Sandals, shocks, accessories and so on 119 o Kids- Sports shoes, formal, school, Sandals, socks, accessories and so on  Fashion Mart – Fashion mart offers a complete range of fashion accessories like jewellery, stoles and scarves, handbags & clutches, belts, fragrances, wallets, sunglasses, watches, cosmetics and beauty care products. Merchandise lines of Big Megamart watches (brands - Fastrack, Sonata, Titan, Casio, Citizen, Tommy Hilfiger), sunglasses (brands - Fastrack I Gear, Sorrento, Spykar, Sprint, Scott), jewellery (brands - Zaveri Pearls, Mystic, Adrika), fragrances & deos (brands - Reebok Deos, Udv, Deo Mania Deos, Ferrari, Mango, Antonio Banderas, Bogart, Nautica, Chevignon, Formula 1, Stone Deos, Tommy Hilfiger, Fcuk, Evaflor, Slazenger, Yacht Man, So, Jaques M, Aqua Brava, David Beckham, Xaver Laurent), cosmetics (brands - Lakme, Revlon, Chambor, Diana of London, Ponds)  Home Mart - Living & dining (complete range of kitchenware like crockery, mugs, non-stick cookware, storage containers, curtains and a complete range in kitchen linen (brands - Nirlep, Prestige), bed & bath (bed sheets, comforters, pillow fillers, cushion fillers, towels etc.) (brands - Welhome, Portico, Cherokee, Home Mart, Impulse, Diviniti, Trance, Portico, All Time, Laopala, Crystal)  Home Mart - Living & dining (complete range of kitchenware like crockery, mugs, non-stick cookware, storage containers, curtains and a complete range in kitchen linen (brands - Nirlep, Prestige), bed & bath (bed sheets, comforters, pillow fillers, cushion fillers, towels etc.) (brands - Welhome, Portico, Cherokee, Home Mart, Impulse, Diviniti, Trance, Portico, All Time, Laopala, Crystal)  Travel Mart – Wide range of travel gear like suitcases, backpacks, duffle bags, strollers and attractive range of travel accessories like locks, key chains, laptop bags etc., luggage (brands - American Tourister, F Gear, Cherokee, Ruggers Gear)  Travel Mart – Wide range of travel gear like suitcases, backpacks, duffle bags, strollers and attractive range of travel accessories like locks, key chains, laptop bags etc., luggage (brands - American Tourister, F Gear, Cherokee, Ruggers Gear) Merchandise portfolio analysis of Big Megamart through BCG Matrix 120 Market Share High Low Market Growth High Youth Wear Women's Wear Men's Wear Kid's Wear Home Mart Travel Mart Stars Question Marks Low Footwear Fashion Mart Toys Cash Caws Dogs Figure 5.6: Merchandise portfolio analysis of Big Megamart Market Share Figure 5.6: Merchandise portfolio analysis of Big Megamart Youth wear, women's wear, men's wear and kid's wear merchandise lines of Big Megamart are in star position in BCG Matrix, which reflects the high market share of merchandise under these lines and high market growth denotes an increase in the sales of these merchandise lines from past years. Home mart and travel mart merchandise lines of Big Megamart are in question mark position in BCG Matrix that shows the low market share of merchandise under these lines and high market growth represents an increase in the sales of these merchandise lines from past years. Footwear and fashion mart merchandise lines of Big Megamart fall under the category of a cash cow in BCG Matrix that presents the high market share of merchandise under this line and low market growth denotes that there is no increase in the sales of this merchandise line from past years. Toys merchandise line of Big Megamart is part of dog position in BCG Matrix, which proposes low market share and low market growth describes that there is a decrease in the sales of these merchandise lines from past years. 5.1.7 Merchandise offered by West Side 5.1.7 Merchandise offered by West Side 121 West Side offers merchandise under the merchandise lines like women’s wear, youth wear, men’s wear, cosmetics & fragrances, kid’s wear, home & gift, accessories and footwear and luggage. Merchandise lines of West Side  Men’s Wear- Casual, formal, sports like colour jeans, sweaters, jackets, checks shirts, stripes shirts, dots shirts, textures or plain shirts, Chinese collar shirts, crafted shirts and trousers, checked shirts, woollens, mufflers, ascot sport, graphic printed tees, sports casual, chest printed T-shirts, classic jeans, Nuon denim, undergarments & briefs, suits, T-shirts, kurta, jacket, trousers, jeans, shorts, boxer shorts, cargos, pajamas, wind cheaters, track-pants, & many more  Women’s Wear- Indian, western, casual like salwar kamej, blouges, formal shirts, coloured jeans, regular jeans, slim fit jeans, dhoti pants, chudidar, hand- woven fabrics, dupattas, coloured chinos, snazzy jackets, gowns, scarf, jeggings, leggings or tregging, sweaters, printed tops, tunics, jumpsuits & bottoms, denim, plus size dress, vest, lingerie, kaftan, house coat, nightys, dress materials, frocks, tops, wind cheaters, stole & many more  Youth Wear - Men's & women's like shirts, T-shirts, jeans, cargos, denim, jacket, dhoti-kurta, caps, sweaters, tops, frocks, jackets, sweaters, ribbons, shorts, jeans, denim, salwar-kurtas, ghagra-choli, leggings, skirts, inner wears, & many more  Youth Wear - Men's & women's like shirts, T-shirts, jeans, cargos, denim, jacket, dhoti-kurta, caps, sweaters, tops, frocks, jackets, sweaters, ribbons, shorts, jeans, denim, salwar-kurtas, ghagra-choli, leggings, skirts, inner wears, & many more  Kid’s Wear- For infant (0-2 yrs), junior (2-6 yrs), senior (7-14 yrs) like coloured denim, kids ethnic wear collection, salwar, kurta and leggings, kurta pyjama, & many more  Accessories- Ties, napkins, belts, handbags, jewellery, sunglasses, bracelet, wallet, chic and fun bags, caps, hand globs, ribbons, stole, shocks & many more  Footwear- Men’s footwear, shoes, boots, flip-flop, Nuon footwear, Lilliput brand footwear for kids, kids footwear (India Size: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13; International Size: 27, 28, 29, 30, 31, 32, 33, 34, 35, 36), women’s footwear (India Size: 5, 6, 7, 8, 9, 10; International Size: 36, 37, 38, 39, 40, 41) & formal shoes, sport shoes, sandals, slippers, casual footwear, slippers, many more 122  Cosmetics & Fragrances- Deo, perfumes, international brands in cosmetics, bath and body products, lotions, creams body washes and balms, & many more  Cosmetics & Fragrances- Deo, perfumes, international brands in cosmetics, bath and body products, lotions, creams body washes and balms, & many more  Home & Gift- Kitchenware, household, linen, towels (hand, face, bath), bed covers, bed sheets, bottles & bowls, play shop (toys, dolls, soft toys & so on), mugs, chic & sophisticated home décor), Turkish range of home (collection lamps, candle stands, bed sheets and cushion), table linens, artefacts, home accessories and furnishings, many more  Luggage- Trolley, school bags, office bags, travel bags, briefcases Luggage- Trolley, school bags, office bags, travel bags, briefcases Merchandise portfolio analysis of West Side through BCG Matrix Merchandise portfolio analysis of West Side through BCG Matrix Merchandise portfolio analysis of West Side through BCG Matrix 123 Market Share High Low Market Growth High Women’s Wear Youth Wear Men’s Wear Cosmetics & Fragrances Kid’s Wear Home & Gift Stars Question Marks Low Accessories Footwear Luggage Cash Caws Dogs Figure 5.7: Merchandise portfolio analysis of West Side Figure 5.7: Merchandise portfolio analysis of West Side Women’s wear, youth wear, men’s wear, cosmetics & fragrances and kid’s wear merchandise lines of West Side are in star position in BCG Matrix, which reflects the high market share of merchandise under these lines and high market growth denotes an increase in the sales of these merchandise lines from past years. Home & gift merchandise line of West Side is in question mark position in BCG Matrix that shows the low market share of merchandise under these lines and high market growth represents an increase in the sales of these merchandise lines from past years. Accessories and footwear merchandise lines of West Side fall under the category of a cash cow in BCG Matrix that proposes the high market share of merchandise under this line and low market growth denotes that there is no increase in the sales of this merchandise line from past years. Luggage merchandise line of West Side is part of dog position in BCG Matrix, which tables low market share and low market growth describes that there is a decrease in the sales of these merchandise lines from past years. 5.1.8 Merchandise offered by Max 124 Max offers merchandise under the merchandise lines like ladies’ wear, men’s wear, children’s wear, footwear and accessories. Merchandise lines of Max  Children’s Wear: Infant wears, 2-8 years old girls & boys wear, 8-14 years old girls & boys wear like, prints jersey, badges jersey, embroidery jersey, mock jersey, T-shirts, hoodies and jackets, knit gloves, knit top, denim bottom, ballerina, denim, girl’s top, boys T-shirts, infant tops, shirts, capris, woven top, woven bottom, polo tee, fashion denim & so on…  Ladies’ Wear: Ethnic & fusion wear, western wear, young wear, active wear, sleepwear, lingerie like ethnic fusion outfits, jumpsuits, checked shirt, animal prints dress, pyjamas, denims & coloured bottoms, bird prints dresses, woolen jacket, skirt & shirt, shorts, tees, sheer shirt, skinny jeans, dupattas, kurta and Patiala salwar, blazer, graphic tees, semi-formal, rainbow-hued pants, shorts and capris, girl skirts, tennis dresses and jumpers, sheer blouse, lace skirt, chiffon dress, leggings, classic churidars, salwar kameez dupatta, sweater, knit pullovers, v-neck sweater, woolen trousers, embellished blouse, kalidar kurta, tops, cocktail dresses, prom dresses, formal, casual and fusion party wear, candy striped tee, tropical striped jumpsuits, top with lace, dress with belt, loose top with cutwork, knit top, ballerina, woven top, knit bottom, denim bottom, fashion jeans, sweaters, knit pyjama, printed bottom, woven bottom, pencil fit jeans, knit capris, jackets, medium kurta, fashion top, knit chudidar, printed tee, pite polo, chest print tee, rugby polo tee, chudidar kurta, woven jacket, short kurta, & so on…  Ladies’ Wear: Ethnic & fusion wear, western wear, young wear, active wear, sleepwear, lingerie like ethnic fusion outfits, jumpsuits, checked shirt, animal prints dress, pyjamas, denims & coloured bottoms, bird prints dresses, woolen jacket, skirt & shirt, shorts, tees, sheer shirt, skinny jeans, dupattas, kurta and Patiala salwar, blazer, graphic tees, semi-formal, rainbow-hued pants, shorts and capris, girl skirts, tennis dresses and jumpers, sheer blouse, lace skirt, chiffon dress, leggings, classic churidars, salwar kameez dupatta, sweater, knit pullovers, v-neck sweater, woolen trousers, embellished blouse, kalidar kurta, tops, cocktail dresses, prom dresses, formal, casual and fusion party wear, candy striped tee, tropical striped jumpsuits, top with lace, dress with belt, loose top with cutwork, knit top, ballerina, woven top, knit bottom, denim bottom, fashion jeans, sweaters, knit pyjama, printed bottom, woven bottom, pencil fit jeans, knit capris, jackets, medium kurta, fashion top, knit chudidar, printed tee, pite polo, chest print tee, rugby polo tee, chudidar kurta, woven jacket, short kurta, & so on…  Men’s Wear: Formalwear, semi-formal wear, casual wear, denim, active & sportswear, innerwear, ethnic like shorts and capris, solids and checks shirts, casual shirts range, stripes and patterns shirts, cotton and linens shirts, short sleeve check shirt and solid colour T-shirt, jeans, casual cotton shirts, coats, T- shirt, kappa track pants, T-shirts, denim, shorts, semi-formal trousers, woven bottom, formal trousers, woven top, jackets, track pants, boxer shorts, top knitted track, bottom-shorts, men’s woven shorts, kppa track pants, short track pants, woven shorts, knit top, Capri for home, casual knit top, casual bottom, chest print tee, ringer tee, short tee & so on… 125  Footwear: Men, women, children like infant shoes, kids shoes (in different size 1 to 13 no.), ankle boot, casual shoe, leather slipper, formal slip-on, sports shoes, women - killer heels, shoes, wedges with bow, stilettos, wedges, pumps, ballerina, moccasins, jello shoe, haute wedge heels, catwalk leather wedges, children - casual shoe, girl sleeper, kids ballerina shoes, and so on…  Accessories: Eye-wear girls, neckpiece, bangles, jewelry, bracelets, brooches, nail paints, elephant bangle, handbags (tote bag, tote bag, shoulder bag, laser cut bag, 2 in 1 bag, satchel bag, studded pink bag, leather bag, sling bag, floral printed bag, floral sling bag, tribal printed tote, clutch bags), sunglasses, watches, pens, ring, earring, flowers, umbrellas/boots/raincoats, nail stones, fragrances, colour cosmetics, skin care, scarf, mufflers and scarves, school & stationary, toys of boys & girls, travel items, wallets, belts and accessories, crafted in exclusive, luxurious synthetic leather, ties, metal earrings, metal & stone necklace, roselle, cleo, pencil box, lunch box, water bottle, trolley bags, soft toys, school bags, belt for kids, ladies & gents socks, hair bands & clips, handkerchiefs and so on… Merchandise portfolio analysis of Max through BCG Matrix Merchandise portfolio analysis of Max through BCG Matrix 126 Market Share High Low Market Growth High Ladies’ Wear Men’s Wear Children’s Wear Footwear Stars Question Marks Low Accessories Cash Caws Dogs Figure 5.8: Merchandise portfolio analysis of Max Market Share High Low Market Growth High Ladies’ Wear Men’s Wear Children’s Wear Footwear Stars Question Marks Low Accessories Cash Caws Dogs Figure 5.8: Merchandise portfolio analysis of Max Figure 5.8: Merchandise portfolio analysis of Max Ladies’ wear and men’s wear merchandise lines of Max are in star position in BCG Matrix, which reflects the high market share of merchandise under these lines and high market growth denotes an increase in the sales of these merchandise lines from past years. Children’s wear and footwear merchandise lines of Max are in question mark position in BCG Matrix that tables the low market share of merchandise under these lines and high market growth represents an increase in the sales of these merchandise lines from past years. Accessories merchandise line of Max falls under the category of a cash cow in BCG Matrix that shows the high market share of merchandise under this line and low market growth denotes that there is no increase in the sales of this merchandise line from past years. No merchandise line of Max is part of dog position in BCG Matrix. 5.1.9 Merchandise offered by Reliance Digital 5.1.9 Merchandise offered by Reliance Digital 127 127 Reliance Digital offers merchandise under the merchandise lines like TVs & audio, kitchen appliances, home appliances, mobile phones, computers, accessories, cameras and gaming. Merchandise lines of Reliance Digital  Mobile Phones: Smartphones [life style & value (brands- Apple, BlackBerry, HTC, Huawei, LG, Nokia, Reconnect, Samsung, Sony, ViewSonic, Karbonn)], feature phones (brands- Nokia, Reconnect, Samsung, Sony), mobile accessories [chargers & adapters, screen guards & protectors, cases & pouches, memory cards (Brands- Targus, Apple, Belkin, Samsung, SanDisk, Sony, Transcend, Capdase, Molife)].  Computers: Tablets (brand- Milagrow, Acer, Apple, I-ball, Reconnect, Reliance, Samsung, Sony, Micromax, Amazon, Karbonn, HCL), laptops (brands- Acer, HP, Samsung, Sony, Dell, Lenovo), desktops (brand- Reconnect), printers & scanners (brands- Epson, Canon, HP), Monitors (Brands- BenQ, Others), software (others, Trend micro, Apple, Microsoft, Norton, McAfee), memory & storage (brands- SanDisk, Transcend), hard drives (brands- Others, Seagate, Western Digital), computer accessories [cables & connectors, chargers & adapters, headphones & headsets, keyboards, mouse & mouse pads, webcams, CD/DVD drives & writers, routers (brands- D-Link, Targus, Apple, Belkin, B-Speech, Bose, HP, I-ball, Jabra, JBL, Logitech, Nokia, Panasonic, Philips, Plantronics, Reconnect, Samsung, Sennheiser, Sony, Transcend, Microsoft, Bandridge, Mitashi, ENON, Capdase, Beats by Dr. Dre).  TVs & Audio: Televisions [LCD, LED, plasma, smart, flat (Brands- Onida, Sansui, LG, Panasonic, Philips, Reconnect, Samsung, Sony, Sharp)], Blu-ray & DVD players (brand- Pioneer, LG, Panasonic, Philips, Reconnect, Samsung, Sony, Mitashi), home theatres (brand- Pioneer, LG, Onkyo, Panasonic, Philips, Reconnect, Samsung, Sony, Mitashi, DENON), HiFi systems (brand- Philips, Sony), av receivers (brands- YAMAHA, DENON, HARMAN KARDON), specialty speakers (brand- JAMO), portable audio- video players (brand- Philips, Sony), MP3 players (brands- Apple, Philips, Sony, Transcend), specialty audio. 128  Cameras: Digital cameras [point & shoot, prosumer, mirrorless (brands- Canon, Reconnect, Samsung, Sony, Nikon, Olympus, FUJIFILM), DSLR cameras (brand- Canon, Sony, Nikon), camcorders (brand- Sony), digital camera accessories [batteries & chargers, bags, cases & pouches, cables & cords, digital photo frames, memory cards, tripods & monopods (brands- Belkin, Reconnect, SanDisk, Sony, Transcend, Benro, Slik, Bandridge, Lowepro, Capdase)]  Home Appliances: Air conditioners (split, window, wall mounted) & coolers (brands- Hitachi, LG, Panasonic, Reconnect, Voltas), washing machines (brands- Godrej, LG, Panasonic, Reconnect, Samsung, BOSCH, IFB, Whirlpool, Siemens), fans (brand- Crompton Greaves, Usha, Havells), vacuum cleaners (brands- Eureka Forbes, Panasonic), irons (brand- Panasonic, Philips, Reconnect, Usha, Morphy Richards, Havells), geysers & heaters (brand- A. O. Merchandise lines of Reliance Digital Smith, Crompton Greaves, Reconnect), personal care, satellite receivers, DTH  Kitchen Appliances: Refrigerators (brand- Godrej, Hitachi, LG, Panasonic, Reconnect, Samsung, Whirlpool), microwave ovens (brand- Onida, LG, Panasonic, Reconnect, Samsung, IFB), juicers (brand- Philips, Reconnect, Usha, Morphy Richards, Havells), juicer mixer grinder (brand- Panasonic, Philips, Reconnect, Bajaj, Usha, Maharaja Whiteline, Morphy Richards, Preethi, Wonderchef, Havells, Inalsa), food processors (brand- Philips, Bajaj, Inalsa, Kenstar, Kenwood, Usha, Inalsa), blenders (brand- Philips, Reconnect, Kenstar, Kenwood, Morphy Richards, Braun, Havells, Inalsa), OTGs (Brands- Reconnect, Usha, Morphy Richards), rice cookers (brand- Panasonic, Reconnect), induction cookers (brand- Philips, Reconnect), sandwich makers (brand- Reconnect, Havells, Morphy Richards), toasters (brand- Reconnect, Usha, Havells), electric kettles (brand- Havells, Morphy Richards), water purifiers / dispensers (brand- Eureka Forbes, HUL, Kent), dishwashers.  Kitchen Appliances: Refrigerators (brand- Godrej, Hitachi, LG, Panasonic, Reconnect, Samsung, Whirlpool), microwave ovens (brand- Onida, LG, Panasonic, Reconnect, Samsung, IFB), juicers (brand- Philips, Reconnect, Usha, Morphy Richards, Havells), juicer mixer grinder (brand- Panasonic, Philips, Reconnect, Bajaj, Usha, Maharaja Whiteline, Morphy Richards, Preethi, Wonderchef, Havells, Inalsa), food processors (brand- Philips, Bajaj, Inalsa, Kenstar, Kenwood, Usha, Inalsa), blenders (brand- Philips, Reconnect, Kenstar, Kenwood, Morphy Richards, Braun, Havells, Inalsa), OTGs (Brands- Reconnect, Usha, Morphy Richards), rice cookers (brand- Panasonic, Reconnect), induction cookers (brand- Philips, Reconnect), sandwich makers (brand- Reconnect, Havells, Morphy Richards), toasters (brand- Reconnect, Usha, Havells), electric kettles (brand- Havells, Morphy Richards), water purifiers / dispensers (brand- Eureka Forbes, HUL, Kent), dishwashers. Merchandise lines of Reliance Digital  Gaming: Gaming consoles (brand- Sony, Microsoft), gaming titles (game franchise- Angry Birds, Assassins Creed, Call of Duty, HALO, Hitman, Medal of Honor, Need for Speed, WWE; Gaming Platform- PC, PS3, XBOX 360), gaming mouse & mouse pads (brand- Reconnect, mouse type - gaming optical tracking) 129  Accessories: Memory & storage (memory cards, micro SD, SD cards, pen drives, hard drives, memory card readers), headsets, multimedia speakers, screen guards & protectors, cables & connectors, computer accessories (mouse & mouse pads, webcams, keyboards, CD /DVD drives & writers, printers, memory cards, peripherals, networking devices), chargers & adapters, batteries, software bags, cases & pouches hubs & routers, power & protection, power units (UPS, surge protectors, stabilisers, camera accessories (digital photo frames, tripods & monopods), gaming accessories (gaming mouse & mouse pad), mobile accessories (mobile chargers & adapters, mobile cases & pouches, memory cards)  Accessories: Memory & storage (memory cards, micro SD, SD cards, pen drives, hard drives, memory card readers), headsets, multimedia speakers, screen guards & protectors, cables & connectors, computer accessories (mouse & mouse pads, webcams, keyboards, CD /DVD drives & writers, printers, memory cards, peripherals, networking devices), chargers & adapters, batteries, software bags, cases & pouches hubs & routers, power & protection, power units (UPS, surge protectors, stabilisers, camera accessories (digital photo frames, tripods & monopods), gaming accessories (gaming mouse & mouse pad), mobile accessories (mobile chargers & adapters, mobile cases & pouches, memory cards) Merchandise portfolio analysis of Reliance Digital through BCG Matrix Merchandise portfolio analysis of Reliance Digital through BCG Matrix Merchandise portfolio analysis of Reliance Digital through BCG Matrix 130 Market Share High Low Market Growth High TVs & Audio Kitchen Appliances Home Appliances Mobile Phones Computers Stars Question Marks Low Accessories Cameras Gaming Cash Caws Dogs Figure 5.9: Merchandise portfolio analysis of Reliance Digital Market Share Figure 5.9: Merchandise portfolio analysis of Reliance Digital TVs & audio, kitchen appliances and home appliance merchandise lines of Reliance Digital are in star position in BCG Matrix, which reflects the high market share of merchandise under these lines and high market growth denotes an increase in the sales of these merchandise lines from past years. Mobile phones and computer merchandise lines of Reliance Digital are in question mark position in BCG Matrix that shows the low market share of merchandise under these lines and high market growth represents an increase in the sales of these merchandise lines from past years. Accessories and cameras merchandise lines of Reliance Digital fall under the category of a cash cow in BCG Matrix that puts the high market share of merchandise under this line and low market growth denotes that there is no increase in the sales of this merchandise line from past years. Gaming merchandise line of Reliance Digital is part of dog position in BCG Matrix, which sets low market share and low market growth describes that there is a decrease in the sales of these merchandise lines from past years. 5.1.10 merchandise offered by NEXT 131 NEXT offers merchandise under the merchandise lines like home entertainment, home appliances, phones & accessories, computers & tablets, personal care, cameras & camcorders, games & consoles, fragrances and writing instruments . Merchandise lines of NEXT  Home Entertainment: Televisions [LCD, LED, plasma, DDB, (brands- Akai, Hyundai, LG, Onida, Panasonic, Philips, Samsung, Sansui, Sony, Videocon)], home cinema systems [home theatre systems, speakers (brands- Akai, Genius, LG, Philips, Sony)], video players [DVD players, Blu-ray players, (brands- LG, Philips, Sony), MP3/MP4 players & ipods (brands- Apple, Creative, Philips, Sony, Transcend), D2H (DishTV, Videocon)  Home Appliances: Air conditioners [split, window, (brand- Electrolux, Godrej, Onida, Samsung, Videocon, Voltas, Whirlpool)], refrigerator [direct cool, frost free, side by side (brands- Electrolux, Godrej, Kelvinator, LG, Samsung, Sansui, Videocon, Whirlpool)], washing machines [fully automatic, semi-automatic (brands- Electrolux, Godrej, IFB, Kelvinator, LG, Onida, Samsung, Videocon, Whirlpool)], microwaves & OTG (brands- Electrolux, Godrej, IFB, Kenstar, LG, Morphy Richards, RussellHobbs, Samsung, Videocon, Whirlpool), small home appliances [iron, toaster, rice cooker, coffee maker, tea maker, induction cooker, sandwich maker, water heaters, electric kettles, juicer mixer grinders, water purifiers, vacuum cleaners, hand blenders, food processor, room heater (brands- AOSmith, Bajaj, Black & Decker, Eureka, Eurolex - Usha Shriram, HUL Pureit, Inalsa, Kenstar, Kent, Maharaja, Morphy Richards, Nasaka, Panasonic, Philips, Prestige, Racold, RussellHobbs, Whirlpool)],  Home Appliances: Air conditioners [split, window, (brand- Electrolux, Godrej, Onida, Samsung, Videocon, Voltas, Whirlpool)], refrigerator [direct cool, frost free, side by side (brands- Electrolux, Godrej, Kelvinator, LG, Samsung, Sansui, Videocon, Whirlpool)], washing machines [fully automatic, semi-automatic (brands- Electrolux, Godrej, IFB, Kelvinator, LG, Onida, Samsung, Videocon, Whirlpool)], microwaves & OTG (brands- Electrolux, Godrej, IFB, Kenstar, LG, Morphy Richards, RussellHobbs, Samsung, Videocon, Whirlpool), small home appliances [iron, toaster, rice cooker, coffee maker, tea maker, induction cooker, sandwich maker, water heaters, electric kettles, juicer mixer grinders, water purifiers, vacuum cleaners, hand blenders, food processor, room heater (brands- AOSmith, Bajaj, Black & Decker, Eureka, Eurolex - Usha Shriram, HUL Pureit, Inalsa, Kenstar, Kent, Maharaja, Morphy Richards, Nasaka, Panasonic, Philips, Prestige, Racold, RussellHobbs, Whirlpool)],  Phones & Accessories: Smartphones [android phones, Blackberry, I-phones, Symbian phones, windows phones (brands- Apple, Blackberry, HTC, Huawei, iBall, Karbonn, Lava, Lenovo, LG, Micromax, Nokia, Samsung, Sansui, Sony, Videocon)], value phones (brands- Huawei, iBall, Karbonn, Lava, Micromax, Nokia, Salora, Samsung, Sansui, Swissvoice, Videocon), tablets (brands- Apple, HCL, iBall, Karbonn, Micromax, Milagrow, Samsung, Videocon), mobile accessories [memory card, car charger, Bluetooth headset, (brands- Callmate, Jabra, Nokia, Sony, Transcend)], landline phones [corded phone, cordless phone, (brands- Panasonic, SPCtelecom, Swissvoice)]. Merchandise lines of NEXT  Phones & Accessories: Smartphones [android phones, Blackberry, I-phones, Symbian phones, windows phones (brands- Apple, Blackberry, HTC, Huawei, iBall, Karbonn, Lava, Lenovo, LG, Micromax, Nokia, Samsung, Sansui, Sony, Videocon)], value phones (brands- Huawei, iBall, Karbonn, Lava, Micromax, Nokia, Salora, Samsung, Sansui, Swissvoice, Videocon), tablets (brands- Apple, HCL, iBall, Karbonn, Micromax, Milagrow, Samsung, Videocon), mobile accessories [memory card, car charger, Bluetooth headset, (brands- Callmate, Jabra, Nokia, Sony, Transcend)], landline phones [corded phone, cordless phone, (brands- Panasonic, SPCtelecom, Swissvoice)]. 132  Cameras & Camcorders: Point & shoot (brands- Canon, Fujifilm, Nikon, Olympus, Panasonic, Samsung, Sony), DSLR (brands- Canon, Nikon, Sony), mirrorless cameras (brands- Fujifilm, Nikon, Olympus), camcorder (brands- Panasonic, Sony).  Cameras & Camcorders: Point & shoot (brands- Canon, Fujifilm, Nikon, Olympus, Panasonic, Samsung, Sony), DSLR (brands- Canon, Nikon, Sony), mirrorless cameras (brands- Fujifilm, Nikon, Olympus), camcorder (brands- Panasonic, Sony).  Games & Consoles: Games [platform- DS, PC, PS Vita, PS2, PS3, PSP, Wii, Xbox (genre- action, fighting, others, puzzle, racing, shooter, sports)], gaming consoles (brands- HCL, Microsoft, Nintendo, Sony, Zapak)  Games & Consoles: Games [platform- DS, PC, PS Vita, PS2, PS3, PSP, Wii, Xbox (genre- action, fighting, others, puzzle, racing, shooter, sports)], gaming consoles (brands- HCL, Microsoft, Nintendo, Sony, Zapak)  Computers & Tablets: Laptops (brand- Sony), tablets (brands- Apple, HCL, iBall, Karbonn, Micromax, Milagrow, Samsung, Videocon), computer accessories [handheld scanner, pen drive, surge protector, HDMI cable, mouse, (brands- Avision, Belkin, Moserbaer, Quantum, Transcend)]  Computers & Tablets: Laptops (brand- Sony), tablets (brands- Apple, HCL, iBall, Karbonn, Micromax, Milagrow, Samsung, Videocon), computer accessories [handheld scanner, pen drive, surge protector, HDMI cable, mouse, (brands- Avision, Belkin, Moserbaer, Quantum, Transcend)]  Personal Care: Hair dryer, hair styler, shaver, hair straightener, epilator, trimmer, grooming kit, hair dryer, hair styler (brands- Andis, Braun, Corioliss, Inalsa, Panasonic, Philips, Remington, Trisa, Wahl)  Personal Care: Hair dryer, hair styler, shaver, hair straightener, epilator, trimmer, grooming kit, hair dryer, hair styler (brands- Andis, Braun, Corioliss, Inalsa, Panasonic, Philips, Remington, Trisa, Wahl)  Writing Instruments: Fountain pens (brands- Cross, Lamy, Parker, Sheaffer, Waterman), Ball Pens (Brands- Cross, Lamy, Parker, Sheaffer, Waterman), roller ball pens (brands- Cross, Lamy, Parker, Sheaffer, Waterman), multi- system pens (brand- Lamy), mechanical pencils (brand- Lamy), gift sets (brands- Parker, Sheaffer)  Fragrances: Fragrances, valentine special Merchandise portfolio analysis of NEXT through BCG Matrix Merchandise portfolio analysis of NEXT through BCG Matrix 133 Market Share High Low Market Growth High Home Entertainment Home Appliances Phones & Accessories Computers & Tablets Personal Care Stars Question Marks Low Cameras & Camcorders Games & Consoles Fragrances Writing Instruments Cash Caws Dogs Figure 5.10: Merchandise portfolio analysis of NEXT Figure 5.10: Merchandise portfolio analysis of NEXT Home entertainment, home appliances and phones & accessory merchandise lines of NEXT are in star position in BCG Matrix, which reflects the high market share of merchandise under these lines and high market growth denotes an increase in the sales of these merchandise lines from past years. Computers & tablets and personal care merchandise lines of NEXT are in question mark position in BCG Matrix that shows the low market share of merchandise under these lines and high market growth represents an increase in the sales of these merchandise lines from past years. Cameras & camcorders and games & consoles merchandise lines of NEXT fall under the category of a cash cow in BCG Matrix that sets the high market share of merchandise under this line and low market growth denotes that there is no increase in the sales of this merchandise line from past years. Fragrances and writing instrument merchandise lines of NEXT are part of dog position in BCG Matrix, which reflects low market share and low market growth describes that there is a decrease in the sales of these merchandise lines from past years. 134 Services introduced by multi-format (item) organised retailers 5.2.1 Services introduced by HyperCITY 5.2.2 Services introduced by Star Bazaar 5.2.3 Services introduced by more.MEGAStore 5.2.4 Services introduced by Big Bazaar 5.2.5 Services introduced by D Mart Services introduced by fashion item organised retailers 5.2.6 Services introduced by Big Megamart 5.2.7 Services introduced by West Side 5.2.8 Services introduced by Max Services introduced by electronics item organised retailers 5.2.9 Services introduced by Reliance Digital 5.2.10 Services introduced by NEXT Services introduced by multi-format (item) organised retailers 5.2.1 Services introduced by HyperCITY 5.2.2 Services introduced by Star Bazaar 5.2.3 Services introduced by more.MEGAStore 5.2.4 Services introduced by Big Bazaar 5.2.5 Services introduced by D Mart Services introduced by fashion item organised retailers 5.2.6 Services introduced by Big Megamart 5.2.7 Services introduced by West Side 5.2.8 Services introduced by Max Services introduced by electronics item organised retailers 5.2.9 Services introduced by Reliance Digital 5.2.10 Services introduced by NEXT CUSTOMER SERVICES INTRODUCED BY ORGANISED RETAIL INDUSTRY This sub-section encompasses the data pertained to the various customer services introduced by organised retail industry, which have been collected through secondary data sources and personal interaction with responsible personnel of the respective organised retail store. Services introduced by organised retailers (included in the study) have been flaunted as follows: Services introduced by multi-format (item) organised retailers 135 5.2.1 Services introduced by HyperCITY HyperCITY provides following services to their customers within the store premises.  Credit (finance on 0 % interest, easy monthly installments on electronics and furniture items),  Home delivery,  Alteration facilities,  Installations of electronics and consumer durables,  Packaging / gift wrapping (a separate cell within a store),  Complaints & exchange handling (as per the return policy),  Personal shoppers,  Parking,  Baby strollers,  Babysitting,  Fitting / trial rooms,  Information,  Safety of personal things,  After sale service for electronics and durables,  CCTV for security,  Fire safety arrangement,  Back office for support, 136  ATM terminal of ING Vaisya Bank within the store,  Customers care service with toll free number 18002097172 & 18001020468,  Repairing services,  Gift certificate (of ₹ 50, 100, 250, 500, 1000, 2000),  Trial purchase,  Special sales for regular customers,  Bridal registry,  Restaurants (food planet within the store),  Beauty salon (Kapil’s salon),  Shopping bags,  Refreshment facilities,  Extended warranties on electronics merchandise.  ATM terminal of ING Vaisya Bank within the store,  Customers care service with toll free number 18002097172 & 18001020468,  Repairing services,  Customers care service with toll free number 18002097172 & 18001020468,  Customers care service with toll free number 18002097172 & 18001020468,  Repairing services,  Gift certificate (of ₹ 50, 100, 250, 500, 1000, 2000),  Gift certificate (of ₹ 50, 100, 250, 500, 1000, 2000),  Trial purchase,  Special sales for regular customers,  Restaurants (food planet within the store),  Beauty salon (Kapil’s salon),  Shopping bags,  Refreshment facilities,  Extended warranties on electronics merchandise. 5.2.2 Services introduced by Star Bazaar Star Bazaar provides following services to their customers within the store premises. 5.2.2 Services introduced by Star Bazaar 137  Home delivery,  Alteration facilities,  Installations,  Complaints & exchange handling (as per the return policy),  Extended store hours (8.30 am to 10.30 pm),  Personal shoppers,  Parking,  Fitting / trial room,  Information,  Safety of personal things on baggage counter,  Escalator & staircase facility,  After sale service on electronics items,  Demonstrations,  Entertainment facilities,  Acceptance of all international credit & debit cards,  Shopping carts,  Warranties by manufacturer,  Washroom & drinking water facility,  Air-condition environment,  Lighting arrangements,  Music,  More than 15 billing-counters with 2 express counters for less than 10 item  CCTV for security,  Fire safety arrangement,  Back office support,  More than 5 weight counters for loose grocery items,  ATM terminal of Deutsche Bank,  Customer service desk,  Customer loyalty desk,  Eye testing service through Vision Eyecare centre,  Guarantee on Star branded merchandise,  Special sales for regular customers,  Restaurants (Live Bakery, Dali),  Home delivery,  Alteration facilities,  Installations,  Complaints & exchange handling (as per the return policy),  Extended store hours (8.30 am to 10.30 pm),  Personal shoppers,  Parking,  Fitting / trial room,  Information,  Safety of personal things on baggage counter,  Escalator & staircase facility,  After sale service on electronics items,  Demonstrations,  Entertainment facilities,  Acceptance of all international credit & debit cards,  Shopping carts,  Warranties by manufacturer,  Washroom & drinking water facility,  Air-condition environment,  Lighting arrangements,  Music,  More than 15 billing-counters with 2 express counters for less than  CCTV for security,  Fire safety arrangement,  Back office support,  More than 5 weight counters for loose grocery items,  ATM terminal of Deutsche Bank,  Customer service desk,  Customer loyalty desk,  Eye testing service through Vision Eyecare centre,  Guarantee on Star branded merchandise,  Special sales for regular customers, ( i k li)  Home delivery,  Alteration facilities,  Installations,  Acceptance of all international credit & debit cards,  Shopping carts,  Air-condition environment,  More than 5 weight counters for loose grocery items,  ATM terminal of Deutsche Bank,  Customer service desk,  Eye testing service through Vision Eyecare centre,  Guarantee on Star branded merchandise,  Special sales for regular customers,  Restaurants (Live Bakery, Dali), 138  Fresh grinding service (Chakki),  Sodexo,  Shopping bags,  Refreshment facilities (Coffee Café Day, Confectionary),  Tata credit card desk  Tata credit card desk 5.2.3 Services introduced by more.MEGAstore more.MEGAStore provides following services to their customers within the store premises.  Credit (easy finance),  Home delivery, 139  Cash on delivery,  Alteration,  Installation,  Packaging / gift wrapping,  Complaints & return handling (as per return policy),  Trial purchase,  Phone orders,  Personal shoppers,  Parking,  Baby strollers,  Fitting rooms,  Fur storage,  Information,  Safety of personal things on baggage counter,  Lift, escalator & staircase facility,  After sales service & support,  Demonstrations,  Entertainment facilities,  Acceptance of all international credit & debit cards,  Special facilities for physically handicapped,  Toilets & drinking water facility,  Air-condition environment,  Lighting & music,  More than 5 billing-counters with 2 express checkout counters,  CCTV for security,  Fire safety arrangement with fire alarm,  Customers support service centre / desk,  Customer helpline number service centre (08652906676),  Back office,  Easy 0 % interest, EMI & spot finance facility,  Shopping carts,  Free testing of the products,  Free recipes on store website,  Cash on delivery,  Alteration,  Installation,  Packaging / gift wrapping,  Complaints & return handling (as per return policy),  Trial purchase,  Phone orders,  Personal shoppers,  Parking,  Baby strollers,  Fitting rooms,  Fur storage,  Information,  Easy 0 % interest, EMI & spot finance facility,  Free recipes on store website, 140  SMS facility to find out the nearest store via SMS, type more (space) <yourpincode> and sent to 56070,  To give feedback SMS more<fdbk><yourcomment> to 56070,  Gift certificates,  Gift certificates,  Special sales for regular customers,  Restaurant & refreshment facility - live bakery,  Shopping bags,  Repairing.  Repairing. 5.2.4 Services introduced by Big Bazaar  Credit (through Capital First),  Home delivery,  Cash on delivery,  Alteration, 141  Installation,  Packaging / gift wrapping,  Complaints & return handling (as per exchange policy),  Trial purchase,  Phone orders,  Personal shoppers,  Parking,  Baby strollers,  Babysitting,  Fitting rooms,  Fur storage,  Information,  Safety of personal things on baggage counter,  Lift, escalator & staircase facility,  After sales service & support,  Demonstrations,  Entertainment facilities,  Medical facilities,  Acceptance of all international credit & debit cards,  Special facilities for physically handicapped,  Exchange counter during 10.30 am to 8.30 pm,  Ladies & gents toilets,  Drinking water facility,  Air-condition environment,  Lighting & music,  TV sets at different places in the store,  More than 10 billing-counters with 1 priority billing counter for senio pregnant ladies and physically challenged people, 1 billing co electronics bazaar items, 1 billing counter for dry fruits,  CCTV for security,  Fire safety arrangements with fire exit,  Back office with warehouse,  Installation,  Packaging / gift wrapping,  Complaints & return handling (as per exchange policy),  Phone orders,  Personal shoppers,  Parking,  Baby strollers,  Babysitting,  Fitting rooms,  Fur storage,  Information,  Lift, escalator & staircase facility,  Drinking water facility,  Air-condition environment,  Lighting & music,  TV sets at different places in the store,  TV sets at different places in the store,  More than 10 billing-counters with 1 priority billing counter for senior citizen, pregnant ladies and physically challenged people, 1 billing counter of electronics bazaar items, 1 billing counter for dry fruits,  CCTV for security,  CCTV for security,  Fire safety arrangements with fire exit,  Back office with warehouse,  Weight counter for loose grocery,  Idli & dosa paste grinding, 142  Loudspeaker announcement,  Mirrors in fashion bazaar section,  Loan facility through Capital First,  Customer service desk,  Loyalty card desk,  Foreign currency exchange counter,  ATM terminal of YES BANK,  Customer helpline number service centre (18002002255),  Easy 0 % interest, EMI & spot finance facility,  Shopping carts,  Free testing of the products,  Free talk time for TATA DOCOMO mobile number (shopping of worth ₹ 400 = ₹ 10 talk time free),  Acceptance of meal vouchers of Sadox & Accor,  Gift certificates,  Special sales for regular customers,  Restaurants & refreshment facility (Foodi Counter),  Beauty salon (Javed Habib Hair Expresso),  Shopping bags,  Mehndi corner,  Weight-checking machines,  NAVRAS – Jewelery Bazaar, Fi d d i i h h d b i  Weight-checking machines,  NAVRAS – Jewelery Bazaar,  Fixed deposit with the guaranteed rebate service. 5.2.4 Services introduced by Big Bazaar Big Bazaar provides following services to their customers within the store premises. 5.2.5 Services introduced by D Mart D Mart provides following services to their customers within the store premises.  Complaints & return handling (as per the return policy),  Packaging / gift wrapping,  Parking,  Baby strollers,  Fitting / trial room,  Information,  Safety of personal things, 143  Acceptance of credit & debit cards,  Personal assistance in selecting merchandise,  Shopping carts,  Warranties,  Toilet room,  Drinking water,  Air-condition environment,  Music,  Loudspeaker announcements,  More than 20 billing-counters,  CCTV for security,  Fire safety arrangement,  Alteration,  Shopping bags,  Refreshment facilities.  Drinking water,  Air-condition environment,  Music,  Loudspeaker announcements,  More than 20 billing-counters,  CCTV for security,  Fire safety arrangement,  Alteration,  Shopping bags,  Refreshment facilities. 5.2.6 Services introduced by Big Megamart Big Megamart provides following services to their customers within the store premises.  Alteration facilities (separate alteration cell),  Packaging / gift wrapping,  Packaging / gift wrapping,  Complaints & return handling (as per the return policy),  Personal shoppers,  Parking,  Baby strollers,  Babysitting,  Fitting / trial room (more than 20), 144  Fire safety arrangement,  Gift certificate,  Trial purchase,  Bridal registry,  Restaurants,  Beauty salon,  Shopping bags,  Refreshment facilities. 145 5.2.7 Services introduced by West Side West Side provides following services to their customers within the store premises.  Home delivery of altered garments,  Alteration facilities,  Installations,  Packaging / gift wrapping (a separate cell),  Complaints & exchange / return handling (as per the customer service policy) Complaints & exchange / return handling (as per the customer service policy),  Staff on late hours for members,  Staff on late hours for members,  Extended store hours for members,  Personal shoppers,  Babysitting,  Fitting / trial room (15),  Information, 146  Safety of personal things on baggage counter,  Lift, escalator & staircase facility,  Demonstrations,  Entertainment facilities,  Acceptance of all international credit & debit cards,  Pulling shopping carts,  Quality assurance,  Drinking water facilities,  Well maintained washrooms within the store,  Air-condition environment,  Music - Radio West side,  More than 5 billing counters and one express billing counter for members,  CCTV for security,  Fire safety arrangements with exit point,  Mirror on every floor,  Back office & management office for support,  Sensormatic system (tag) for products,  Customer service desk,  Easy search option facility (if customers cannot find their size & colour, le West side know, they will make available the same for customers),  Gift certificate (of ₹ 50, 100, 500, 1000 in the denomination),  Trial purchase,  Special sales for regular customers,  Parking,  Restaurants (food café, coffee parlour, gourmet West – food section within th store),  Shopping bags,  Refreshment facilities.  Safety of personal things on baggage counter,  Lift, escalator & staircase facility,  Demonstrations,  Entertainment facilities,  More than 5 billing counters and one express billing counter for members,  CCTV for security,  Back office & management office for support,  Sensormatic system (tag) for products,  Customer service desk,  Easy search option facility (if customers cannot find their size & colour, let West side know, they will make available the same for customers),  Gift certificate (of ₹ 50, 100, 500, 1000 in the denomination),  Trial purchase,  Special sales for regular customers,  Special sales for regular customers,  Parking,  Parking,  Restaurants (food café, coffee parlour, gourmet West – food section within the store),  Shopping bags,  Refreshment facilities. 147 5.2.8 Services introduced by Max Max provides following services to their customers within the store premises. Max provides following services to their customers within the store premises.  Home delivery (if customers could not find their size),  Alteration facilities,  Packaging / gift wrapping,  Complaints & exchange handling (as per the guarantee & exchange policy),  Personal shoppers,  Parking,  Fitting / trial room (separate for men, women & children),  Information,  Safety of personal things on baggage counter,  Lift, escalator & staircase facility,  Demonstrations,  Entertainment facilities,  Acceptance of all international credit & debit cards,  Pulling shopping carts,  Home delivery (if customers could not find their size),  Acceptance of all international credit & debit cards,  Pulling shopping carts,  Pulling shopping carts, 148  Well maintained washrooms & drinking water facilities (outside the store),  Air-condition environment,  Lighting & western music,  4 billing-counters,  CCTV for security,  Fire safety arrangements with exit point,  Mirrors in store,  Sensormatic system (tag) for products,  Customer service associate on desk,  Easy search option facility (if customers cannot find their size & colour, let Max know, they will make available the same for customers),  TV sets at billing counters,  Loudspeaker announcement,  Pass of different activities where Max is the sponsor,  SMS service to know the store location & reward points (SMS ‘TIC’ to 56767),  Gift certificate (of ₹ 100, 250, 500, 1000, 2000 and 5000 in the denomination),  Special sales for regular customers,  Restaurants (outside the store),  Shopping bags,  Refreshment facilities.  Well maintained washrooms & drinking water facilities (outside the store),  Air condition environment  Well maintained washrooms & drinking water facilities (outside the store)  Air-condition environment,  Air-condition environment,  Lighting & western music,  4 billing-counters,  Easy search option facility (if customers cannot find their size & colour, let Max know, they will make available the same for customers),  TV sets at billing counters,  Loudspeaker announcement,  Pass of different activities where Max is the sponsor,  SMS service to know the store location & reward points (SMS ‘TIC’ to 56767),  Gift certificate (of ₹ 100, 250, 500, 1000, 2000 and 5000 in the denomination),  Special sales for regular customers,  Restaurants (outside the store),  Refreshment facilities. 149 5.2.9 Services introduced by Reliance Digital resQ care plan (RCP) takes the hassles and expense out of unexpected product failures, break down and malfunctions long after the  After sales service & support through resQ for 365 days between 10 am to 10 pm,  After sales service & support through resQ for 365 days between 10 am to 10 pm,  Demonstrations (the digital zone),  Entertainment facilities (play station),  Acceptance of all international credit & debit cards,  Game zone (play station),  Washrooms & drinking water facility,  Air-condition environment,  Lighting & music,  4 billing-counters,  CCTV for security,  Good fire safety arrangement with exit point,  The Sensormatic system (tag) for merchandise,  My assist- customer service centre / desk,  Home Theater experience zone,  Customer care service centre (02267276727),  My finance desk,  Back office,  Easy EMI & spot finance facility (through Bajaj FinanceV on credit card of HDFC, City, ICICI, Kotak, Axis, HSBC Bank credit card),  Online green room service,  Online reliance digital solution box,  Maintenance & repair,  Specialised customer training,  One stop solution for repairs of all reconnect brands,  Gift certificate (of ₹ 500 to 10000 in the denomination),  Special sales for regular customers,  Shopping bags,  Health checkup plan – a preventive maintenance,  Enhanced utility through resQ accessories and extended warranty service: a resQ care plan (investing ₹ 2* per day can free you from the worries of repairs and maintenance! 5.2.9 Services introduced by Reliance Digital Reliance Digital provides following services to their customers within the store premises.  Credit (EMI facility - through Bajaj FinanceV on the credit card of HDFC, City, ICICI, Kotak, Axis, HSBC Bank credit card),  Home delivery,  Installation through resQ,  Packaging / gift wrapping,  Complaints & return handling (as per return policy),  Personal shoppers,  Parking,  Restroom,  Babysitting,  Fur storage,  Information,  Safety of personal things on baggage counter,  Staircase facility, 150  After sales service & support through resQ for 365 days between 10 am to 10 pm,  Demonstrations (the digital zone),  Entertainment facilities (play station),  Acceptance of all international credit & debit cards,  Game zone (play station),  Washrooms & drinking water facility,  Air-condition environment,  Lighting & music,  4 billing-counters,  CCTV for security,  Good fire safety arrangement with exit point,  The Sensormatic system (tag) for merchandise,  My assist- customer service centre / desk,  Home Theater experience zone,  Customer care service centre (02267276727),  My finance desk,  Back office,  Easy EMI & spot finance facility (through Bajaj FinanceV on credit card of HDFC, City, ICICI, Kotak, Axis, HSBC Bank credit card),  Online green room service,  Online reliance digital solution box,  Maintenance & repair,  Specialised customer training,  One stop solution for repairs of all reconnect brands,  Gift certificate (of ₹ 500 to 10000 in the denomination),  Special sales for regular customers,  Shopping bags,  Health checkup plan – a preventive maintenance,  Enhanced utility through resQ accessories and extended warranty service: a resQ care plan (investing ₹ 2* per day can free you from the worries of repairs and maintenance! 5.2.9 Services introduced by Reliance Digital resQ care plan (RCP) takes the hassles and expense out of unexpected product failures, break down and malfunctions long after the manufacturer's warranty has expired, RCP includes 5 bonus coupons that offer 151  Demonstrations (the digital zone),  Easy EMI & spot finance facility (through Bajaj FinanceV on credit card of HDFC, City, ICICI, Kotak, Axis, HSBC Bank credit card),  Online green room service Easy EMI & spot finance facility (through Bajaj FinanceV on credit card of HDFC, City, ICICI, Kotak, Axis, HSBC Bank credit card),  Online green room service,  Online reliance digital solution box,  One stop solution for repairs of all reconnect brands,  Gift certificate (of ₹ 500 to 10000 in the denomination),  Special sales for regular customers,  Special sales for regular customers,  Health checkup plan – a preventive maintenance,  Health checkup plan – a preventive maintenance,  Enhanced utility through resQ accessories and extended warranty service: a resQ care plan (investing ₹ 2* per day can free you from the worries of repairs and maintenance! resQ care plan (RCP) takes the hassles and expense out of unexpected product failures, break down and malfunctions long after the manufacturer's warranty has expired, RCP includes 5 bonus coupons that offer 151 you range of benefits i.e. 10% discount coupon on. resQ range of accessories, ₹ 250 discount coupon on labour charges (only for repairs), one free preventive maintenance service (PMS) coupon, 10% discount coupon on purchase of RCP for any other product, 15% discount coupon on renewal of RCP) 5.2.10 Services introduced by NEXT NEXT provides following services to their customers within the store premises.  Credit (EMI facility through credit card),  Home delivery,  Cash on delivery,  Installation,  Fur storage,  Information,  After sales service & support,  Demonstrations,  Entertainment facilities (play station), 152  Acceptance of all international credit & debit cards,  Game zone (play station),  Washrooms & drinking water facility,  Air-condition environment,  Lighting & music,  4 billing-counters,  CCTV for security,  Fire safety arrangements,  Sensormatic system (tag) for merchandise,  Customers service centre / desk,  Customer helpline number service centre during 8 am to 8 pm (0124- 4993500),  Back office,  Easy 0 % interest, EMI & spot finance facility,  Free gifts,  Online information regarding product care,  30 day replacement guarantee policy,  30 days change of mind policy,  Shopping bags,  Gift certificate (of ₹ 500 to 10000 in the denomination),  Special sales for regular customers,  Maintenance & repair,  Extended warranty,  Recharge your mobile phone, DTH & pay mobile & landline bills.  Acceptance of all international credit & debit cards,  Air-condition environment,  Lighting & music,  4 billing-counters,  CCTV for security,  Customer helpline number service centre during 8 am to 8 pm (0124- 4993500),  Gift certificate (of ₹ 500 to 10000 in the denomination),  Special sales for regular customers,  Maintenance & repair,  Extended warranty,  Recharge your mobile phone, DTH & pay mobile & landline bills. 153 Introduction This chapter comprises an analysis of the various data related to retailing strategies introduced by organised retail industry, which have been collected through the institutional questionnaire by personal interaction with responsible personnel of the respective organised retail store. Further, results have been interpreted and discussed to reach on some inferences at the end of the different sections. Retailing strategies introduced by organised retail industry  Retail marketing strategy  Segmentation strategy  Targeting strategy  Positioning strategy  Defensive strategy  Strategic tools  Generic (business-level) strategy  Retail marketing mix  Merchandise planning  Retail assortment strategy  Stock planning  Pricing strategy  Price-adjustment strategy  Promotional strategy  Customer service strategy  Customer relationship management  Store design  etc. 6.1 Retail market segmentation All the organised retailers (included in the study) use demographic, psychographic and behavioural variables in order to segment the retail market. Age, family life cycle, gender, income, occupation and education are most the common variables for demographic segmentation while lifestyle and personality are most common variables for psychographic segmentation and usage rate, loyalty status, occasions, benefits and innovativeness are the most common variables for behavioural segmentation of the retail market. All the organised retailers (included in the study) do demographic segmentation on the basis of gender (male and female); income (low, lower middle, middle, upper middle, high…); age (infants, 2-5 years, 5-10 years, 11-21 years, …., 60+ years) except Reliance Digital and NEXT; family life cycle (young, single; young, married, no children; young, married, child….) except Star Bazaar; occupation (unskilled worker, skilled worker, students, self-employed, homemaker, retired) except West Side and Max; and education (illiterate, school, non-graduate, graduate / postgraduate - general/professional)) except West Side, Max and NEXT. While lifestyle (culture oriented, sport oriented, outdoor oriented….) and personality (compulsive, gregarious, authoritarian, ambitious, aggressive, shy, emotional….) for psychographic segmentation and occasions (regular/special); 155 benefits (quality, service, economy, speed….); usage rate (light, medium & heavy) except Big Megamart, HyperCITY and NEXT; loyalty status (none, medium, strong & absolute) except D Mart and Big Megamart; innovativeness (innovator, early adopter, early majority, late majority, laggard) except D Mart; for behavioural segmentation of retail market. Apart from most common variables for behavioural segmentation, some of the organised retailers use other variables like attitude towards the merchandise (enthusiastic, positive, indifferent, negative & hostile) by Big Bazaar; perceived risk (high, medium and low) by D Mart, Reliance Digital, NEXT, more.MEGAStore; user status (non-user, ex-user, potential user, first time user & regular user) by West Side, Max, more.MEGAStore, Big Bazaar; and readiness stage (unaware, aware, informed, interested, desirous & intending to buy) by Big Megamart, HyperCITY, Star Bazaar, Reliance Digital, more.MEGAStore. 6.2 Segmentation strategy All multi-format organised retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart) adopt full market coverage segmentation strategy in order to segment the retail market. HyperCITY covers the full retail market in the category of food & grocery, health & beauty, home & furniture, fashion, electronics, toys and sports while Star Bazaar covers whole retail market with following merchandise categories, i.e. food & grocery, personal & home care, home & kitchen needs and fashion & accessories. Merchandise categories like grocery, fruits & vegetables, home & personal care, general merchandise, electronics merchandise, computer & accessories, mobile phone & accessories, apparel, footwear and sports are offered by more.MEGAStore in order to cover the full retail market. Big Bazaar covers the full retail market by offering the merchandise in following categories like grocery & staples bazaar, farm fresh, kitchen & dining bazaar, bed & bath bazaar, books, magazines & stationery bazaar, luggage bazaar, home bazaar, personal bazaar, furniture bazaar, electronics bazaar, office bazaar, mobile bazaar, toys, sports & outdoors, fashion bazaar and gifts. And superstore D Mart covers the whole retail market in segments like grocery, foods, toiletries, beauty merchandise, garments, kitchenware, bed & bath linen, footwear, toys & games, water purifiers, baby foods, stationery and electronic items. 156 All the fashion-item organised retailers (i.e. Big Megamart, West Side and Max) prefer selective specialisation segmentation strategy to segment the retail market. Big Megamart offers the merchandise in selective merchandise segments like footwear, apparels, fashion, toys, home and travelling accessories. West Side fulfils the needs of the customers in selective segments like apparel, footwear, cosmetics, home & gift items, luggage items and accessories while Max operates in segments like apparel, footwear and accessories for men, including women, children and infants. Superstore based electronics-item organised retailers (i.e. Reliance Digital and NEXT) adopt market specialisation segmentation strategy while segmenting the retail market. Reliance Digital operates in the electronics market, especially in consumer durables & information technology segments while NEXT concentrates on consumer electronics, small home appliance and mobile phones market segments. 6.3 Targeting strategy Multi-format organised retailers (i.e. Star Bazaar, more.MEGAStore, Big Bazaar and D Mart) adopt an undifferentiating targeting strategy or mass retailing in order to target the retail market. However, HyperCITY prefers multi-segment targeting strategy. Star Bazaar covers mass retail market by offering staples, packaged foods, beverages, dairy, bakery, frozen merchandise, non-vegetation, fruits & vegetables, cleaning aids, health & beauty merchandise, kitchenware, electronics, home decor, luggage items, toys, stationery, men’s wear, women’s wear, kid’s wear, footwear and accessories. more.MEGAStore focuses on mass retailing by offering merchandise like apparels, audio & video, live bakery, beverages, books and audio & video merchandise, computer & accessories, crockery, cookware, do it yourself & auto accessories, electronics, FMCG merchandise, footwear, frozen & dairy merchandise, fresh fruit & vegetables, fresh non-vegetarian, furniture, general merchandise, fitness, home care merchandise, home decor merchandise, home needs & home upkeep, imported food, footwear & accessories, infant & children's apparels, information technology merchandise, large white appliances, luggage, mobile phone & accessories, personal care & cosmetics, prepared food, ready to cook or prepared food, processed food, small white appliances, sporting goods, packaged & loose staples, stationery and toys etc. Big Bazaar focuses on mass retailing by offering something (merchandise) for everyone in a different category of the merchandise. While, D Mart does not offer specific mix for any segment and covers the whole 157 market with their selected merchandise for different segments and HyperCITY offers merchandise and services to the multi-segments of customers according to their age groups, gender, income, occupation, lifestyle, loyalty status, benefits and so on. West Side, Max and NEXT adopt concentrated targeting strategy, which is mostly adopted strategy among superstore organised retailers. West Side carved a niche for its brand of merchandise to create loyal customers. They offer a variety of designs & style. Everything in their stores is exclusively designed & the merchandise range, from stylised clothes, footwear and accessories for men, women, children, too well-coordinated table Lenin, artefact, home accessories and furnishing. Max focuses on niche retailing and concentrates towards few segments like apparel, footwear & accessories for the entire family while NEXT concentrate on mobile phones, home appliances and consumer electronics. Other organised retailers (i.e. Big Megamart and Reliance Digital) adopt differentiating targeting strategy or selective retailing. Big Megamart offers a specific mix to their different customer segments while Reliance Digital offers a complete solution through the specific mix to their targeted customers. 6.3 Targeting strategy 6.4 Positioning strategy Figure 6.1: Positioning strategies of organised retailers 6.4 Positioning strategy Figure 6.1: Positioning strategies of organised retailers 158 More for less is the most popular positioning strategy among both hyper & super store organised retailers and adopted by most of the organised retailers i.e., HyperCITY, more.MEGAStore, Big Bazaar, D Mart, Max and Reliance Digital in order to position their offerings to the targeted customers. Hyper stores like HyperCITY use more for less strategy to position their offerings to the customers with more than 40000 quality merchandise and unmatched values/benefits & price while more.MEGAStore by offering more than 35,000 merchandise & services of great quality with an unbelievable price to keep their promise that Hamesha Extra for customers. Big Bazaar uses this strategy by offering merchandise & services of best quality with the best price guaranteed. Superstore (i.e. D Mart) uses strategy to position their offerings to the customers by offering more benefits for less price while Max by offering international fashion from around the globe & an excellent range of merchandise with a great price in the world-class shopping experience. Moreover, Reliance Digital by offering more than 4000 quality merchandise with more than 200 national & international brands and the lowest price guaranteed. However, some organised retailers (i.e. Star Bazaar and NEXT) adopt same for less positioning strategy while others (Big Megamart and West Side) more for same positioning strategy to position their offerings to the target market. Star Bazaar uses same for less strategy to position their offerings to the customers to keep their promise to their customers “Helping You Spend Less” while NEXT to keep their promise to their customers “Best Brand & Best Bargains”. Moreover, Big Megamart uses more for same strategy to position their offerings to the customers by offering more benefits/quality for the same price in comparison to the competitors while West Side by understanding the customer’s needs, strive to win customers’ confidence and offer best-in-class-merchandise & services at affordable prices. 6.5 Retail marketing strategy Both hyper & superstores / multi-format & specialised organised retailers (like HyperCITY, Star Bazaar, Big Bazaar, Max, Reliance Digital and NEXT) adopt a retail market penetration strategy in order to grave growth opportunities. All these organised retailers penetrate the retail market by displaying merchandise in an attractive way to increase impulse purchase and training to personal shoppers/sales staff for cross & add-on selling. Moreover, these organised retailers also penetrate the 159 retail market by opening the stores for longer hours except HyperCITY and by opening more stores in the same target retail market except HyperCITY and Star Bazaar. Growth Opportunities Target Market Existing New Retail Format Existing Market Penetration HyperCITY, Star Bazaar, Big Bazaar, Max, Reliance Digital and NEXT Market Expansion More.MEGAStore, D Mart and West Side New Format Development Big Megamart Diversification Related/unrelated Figure 6.2: Retail marketing strategies of organised retailers Figure 6.2: Retail marketing strategies of organised retailers While other organised retailers (i.e. more.MEGAStore, D Mart and West Side) adopt a retail market expansion strategy in order to grab the opportunities available in the market. All these retailers hope to expand rapidly with similar format stores in the next few years to grave growth opportunities. However, Big Megamart adopts retail format development strategy by offering merchandise & services with the new retail format, i.e. footwear to same target retail market. 6.6 Defensive strategy The most favourable defensive choice is a preemptive defence strategy among both hyper & super store organised retailers (i.e. Big Megamart, HyperCITY, Star Bazaar, Big Bazaar, Max and Reliance Digital). All these attack on competitors more aggressively by launching a wide range of merchandise and brands in different categories. HyperCITY uses preemptive defence by attacking competitors more aggressively by launching more than 40000 merchandise to their customers with a wide range of local, national, international, exclusive and store label brands. Star 160 Bazaar uses this defensive strategy by spreading their store over a large area 40000- 80000 square feet and offering the entire spectrum of merchandise categories ranging from fresh foods, grocery, apparel, general merchandise and consumer durables with a range of more than 30000 items at great price, modern shopping environment and backed by strong values of TATA group. Big Bazaar uses by offering the entire spectrum of merchandise categories with more than 15 merchandise lines ranging from grocery to gifts and Max, by offering a fresh collection of international designs specially customised to the Indian market in every season (autumn, winter and spring- summer). Moreover, Reliance Digital uses by offering more than 200 international & national brand with 4000 merchandise along private label brand ‘Reconnect’ in the category like mobile phones, tablets, TVs, camera, DVD player, home appliances, kitchen appliances and accessory. Bazaar uses this defensive strategy by spreading their store over a large area 40000- 80000 square feet and offering the entire spectrum of merchandise categories ranging from fresh foods, grocery, apparel, general merchandise and consumer durables with a range of more than 30000 items at great price, modern shopping environment and backed by strong values of TATA group. Big Bazaar uses by offering the entire spectrum of merchandise categories with more than 15 merchandise lines ranging from grocery to gifts and Max, by offering a fresh collection of international designs specially customised to the Indian market in every season (autumn, winter and spring- summer). Moreover, Reliance Digital uses by offering more than 200 international & national brand with 4000 merchandise along private label brand ‘Reconnect’ in the category like mobile phones, tablets, TVs, camera, DVD player, home appliances, kitchen appliances and accessory. The mobile defence strategy is second most favourable defensive choice among some superstores organised retailers (i.e. 6.6 Defensive strategy D Mart, West Side and NEXT) and they hope to expand rapidly by stretching its domain to new territories with the similar format store. However, more.MEGAStore (a multi-format hyper store organised retailer) uses contraction defence strategy as they give up from the weaker territories from where the company is not getting sufficient ROI, for example, Aurangabad’s more.MEGAStore. 6.7 Strategic tools used to perform store-level planning SWOT analysis is the most favourable tool to perform store-level planning and all the organised retailers use it except D Mart. Five force analysis is also a preferred tool and both hyper & super store organised retailers included in the study use it except D Mart, Star Bazaar and NEXT. However, some organised retailers use some more tools to perform store-level planning, i.e. West Side uses ETOP analysis and PEST analysis while Big Bazaar uses a PEST analysis and Reliance Digital uses ETOP analysis in order to perform store level planning. 6.8 Retail store position in Boston matrix 161 BCG Matrix Market Share High Low Market/Business Growth High Big Bazaar, West Side, NEXT Big Megamart, HyperCITY, Star Bazaar, more.MEGAStore, D Mart, Max, Reliance Digital Stars Question Marks Low Cash Caws Dogs Figure 6.3: Retail store position in Boston matrix Figure 6.3: Retail store position in Boston matrix Big Bazaar, West Side and NEXT stand in star category, which is characterised by high market growth & the high market share while others, i.e. Big Megamart, HyperCITY, Star Bazaar, more.MEGAStore, D Mart, Max and Reliance Digital belong to the question mark category that is characterised by high market growth but low market share. 6.9 Generic (business-level) strategy Generic (business-level) strategy Figure 6.4: Generic (business-level) strategies of organised retailers Figure 6.4: Generic (business-level) strategies of organised retailers 162 Cost leadership strategy is a common choice as a business level strategy at store-level among all the multi-format organised retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart), which is characterised by the broad target with lower cost. All these organised retailers adopted this strategy in order to utilise a broad target competitive scope and lower cost competitive advantage. However, Big Megamart (a hyper store fashion-item organised retailer) follows differentiation strategy at store-level that is characterised by the broad target with differentiation. Moreover, West Side (a super store fashion-item organised retailer) follows differentiation focus strategy at store-level by doing niche retailing with differentiation. While other specialised superstore organised retailers (i.e. Max, Reliance Digital and NEXT) prefer cost focus strategy at store-level, which is characterised by the narrow target with lower cost. Max follows this strategy by targeting narrowly in selective areas like apparel, footwear, & accessory with lower cost while Reliance Digital, by targeting a niche segment, i.e. electronics, especially consumer durables & information technology with lower cost and NEXT, targeting niche segment i.e. consumer electronics, home appliance and mobile phones with lower cost. 6.10 Problems faced in framing overall retailing strategy The high price of real estate, lack of preferred locations and retail space, high cost of space and location, trained workforce, multiple & a complex taxation system of India, many entrants as online retailers, high demand for service standards as consumerism is shifting in India are some common problems faced by all the organised retailers. Apart from these most common problems, D Mart faces problems in handling customer complaints while Big Megamart needs bigger premises to operate and Star Bazaar stuck with the inefficiency of the supply chain. Moreover, West Side faces problems due to lack of adequate infrastructure and changing behavioural forces among consumers while Max and Reliance Digital are facing stiff competition and problems due to changing the demand pattern of Indian consumers because of changing demographics of Indian consumers and NEXT is facing stiff competition due to many entrants as online retailers of electronics items. Big Bazaar is facing the problems while framing overall retailing strategies like 163 changing demand pattern of Indian consumers, demand for higher wages by the workforce and stiff competition. 6.11 Core advantages Efficient operations are the core advantages of all the organised retailers included in the study while the image of the retailers is the second most common advantage of organised retailers (i.e. Big Megamart, Star Bazaar, more.MEGAStore, Big Bazaar, West Side, Max, Reliance Digital and NEXT). Moreover, strong private (store-level) brands are also the core advantage of Big Megamart, HyperCITY, more.MEGAStore, Big Bazaar, West Side, Max and Reliance Digital while low cost is also a common core advantage of organised retailers like more.MEGAStore, Big Bazaar, D Mart, Max, Reliance Digital and NEXT. Apart from most common advantages, Big Megamart, West Side and Max have a strong fashion reputation among consumers. Moreover, category dominance is also the core advantage of West Side and NEXT. West Side has dominance in the fashion category, while NEXT, in electronics. 6.12 Retail Marketing Mix (RMM) on importance scale 6.12 Retail Marketing Mix (RMM) on importance scale Table 6.1: Retail marketing mix (RMM) on importance scale Sr. No. Variables Importance Scale* K-S D value calculated 4 3 2 1 1. Product (Merchandise) 10 00 00 00 0.750 2. Price 08 02 00 00 0.550 3. Promotion 09 01 00 00 0.650 4. Premises / Place 07 02 01 00 0.450 5. Personnel / People 08 02 00 00 0.550 6. Presentation 07 01 02 00 0.450 7. Physical Evidence 00 09 01 00 0.400 * 4- extremely important, 3- important, 2- partially important, 1- not important Kolmogorov-Smirnov (K-S) D value tabulated = 0.410; Level of significance (α) = 5 % (0.05) Product (merchandise) is an extremely important element of the retail marketing mix for all the organised retailers included in the study. Price is also an extremely important element of RMM for all the organised retailers except Big Megamart and West Side as it is an important for them. All the organised retailers except Big Megamart consider promotion as an extremely important element of 164 RMM; however, it is important for Big Megamart. The premises / place is an extremely important element of RMM for all the organised retailers except D Mart, Max and NEXT, however, it is an important element for D Mart and Max, while partially important for NEXT. Personnel / people are considered as the extremely important element of RMM by all the organised retailers, except D Mart and Max. For D Mart and Max, it is an important element. 6.11 Core advantages The presentation is an extremely important element of RMM for all the organised retailers, except more.MEGAStore, D Mart and NEXT as it is important for more.MEGAStore and partially important for D Mart and NEXT. Physical evidence is an important element of RMM for all the organised retailers included in the study, except NEXT, as it is partially important. 6.13 Management of retail marketing mix Multi-format organised retailers manage their retail marketing mix as follows. HyperCITY manages its RMM by offering more than 40000 quality merchandise with an unbeatable price to their target customers with a wide range of local, national, international & private brands. Star Bazaar manages its RMM by operating in more than 40000 square feet stores and offering the entire spectrum of merchandise categories ranging from fresh foods, grocery, apparel, general merchandise and consumer durables with a range of more than 30000 items at a great price, modern shopping environment and backed by strong values of TATA. more.MEGAStore manages its RMM by offering Extra quality, Extra choice with Extra savings on merchandise at the same time Extra Care of customers & Extra rewards to customers while Big Bazaar manages its RMM by offering best quality merchandise with the lowest price at any cost for guaranteed customer satisfaction. Superstore, D Mart manages its RMM with a variety of merchandise, less price, significant promotions, proper display and helping personal shoppers. Fashion-item organised retailers manage their retail marketing mix as follows. Big Megamart manages its RMM by offering a specific mix to their target customers while super-stores like West Side manages its RMM by operating in 15,000 - 30,000 square feet stores and offering a differentiated retail marketing mix to their target customers with a variety of designs, styles and exclusively designed. Well-designed interiors, sprawling space, prime locations, strong values of TATA and coffee shops 165 are to enhance the customers’ shopping experience. In addition, Max manages its RMM by bringing to customers the most dynamic offering of the value in fashion merchandise to stringent quality standard and offers latest international trends with great variety & affordable price & a holistic shopping experience. Their sales advisors & visual merchandising team ensure that every Max store represent the spirit of their brand promise “Look Good, Feel Good”. Electronics-item organised retailers manage their retail marketing mix as follows. Reliance Digital manages its RMM by offering the merchandise that best suit to the customer’s lifestyle needs with the latest & widest, digital service, friendly counselling, and best value for money in terms of total cost of ownership and provide total solutions with the lowest price guaranteed. Every Reliance Digital store represents the spirit of their brand promise “We bring technology to life for you”. 6.13 Management of retail marketing mix NEXT manages its RMM by offering multi-brand and multi-merchandise with an entire range of consumer durables with affordable price. 6.14 Brands offered A heterogeneous mix of brands is the first choice among the all the multi- format organised retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart). However, Big Megamart (a fashion store organised retailer) also offers a heterogeneous mix of brands to targeted customers. The offering basket of all these retailers contains generic, local, regional, national, copycat, international and private/store brands. While fashion-item superstores (i.e. West Side and Max) choose a blend of private label & other brands to offer and their offering basket contains private (store label) in all categories offered, exclusive, national (manufacturers) and international brands. Moreover, electronics superstores (i.e. Reliance Digital and NEXT) offer multi-brands in different categories of the merchandise and their offering basket contains all the leading (world’s most popular brands) national and international brands like LG, Samsung, Electrolux, Sony, Philips, Whirlpool, Videocon, Onida, Kelvinator, Kenstar, Sansui and private brands. 6.15 Important factors while planning merchandise 6.15 Important factors while planning merchandise Table 6.2: Important factors while planning merchandise 166 Sr. No. Factors Importance Scale* K-S D value calculated 4 3 2 1 1. Target market 09 01 00 00 0.650 2. Goods & service growth potential 07 03 00 00 0.500 3. Trends (fashion) 08 02 00 00 0.550 4. Image of retailer 09 00 01 00 0.650 5. Competition 00 06 03 01 0.150 6. Customer segments 07 02 01 00 0.450 7. Responsiveness to customers 07 03 00 00 0.500 8. Amount of investment 01 02 07 00 0.250 9. Profitability 01 04 05 00 0.250 10. Risk 04 03 03 00 0.250 * 4- extremely important, 3- important, 2- partially important, 1- not important Kolmogorov-Smirnov (K-S) D value tabulated = 0.410; Level of significance (α) = 5 % (0.05) The target market is an extremely important factor for all the organised retailers included in the study except D Mart, as it is important for them while trends (fashion) are extremely important for all the organised retailers except Reliance Digital and NEXT as it is important for them to plan merchandise for retail store customers. Image (of retailers) is also an extremely important factor for all the organised retailers except D Mart as it is partially important for them while goods & service growth potential is extremely important for all the organised retailers except Big Megamart, HyperCITY and West Side as it is important for them to plan merchandise. Responsiveness to customers and customer segments are extremely important factors for all the organised retailers except HyperCITY, Star Bazaar and D Mart, as these are important factors for HyperCITY and Star Bazaar while responsiveness to customers is important for D Mart and customer segments is a partially important factor in planning store merchandise. The risk is an extremely important for HyperCITY, Star Bazaar, D Mart and West Side while important for Big Megamart, Reliance Digital and NEXT and partially important for more.MEGAStore, Big Bazaar and Max to plan store merchandise. Profitability is partially important for most of the organised retailers (i.e. Star Bazaar, more.MEGAStore, Big Bazaar, West Side and NEXT) while important for others (i.e. D Mart, Big Megamart, Max, and Reliance Digital); however, it is extremely important for HyperCITY. 6.15 Important factors while planning merchandise The amount of investment is a partially important factor for most of the organised except Reliance Digital, NEXT and D Mart because it is important for Reliance Digital and NEXT; while, it is extremely important for D Mart to plan store merchandise. Competition is an important factor for most of the 167 organised retailers except HyperCITY, Star Bazaar, NEXT and D Mart as it is partially important for HyperCITY, Star Bazaar and NEXT while not important for D Mart in order to plan store merchandise. 6.16 Important factors while planning merchandise quality Table 6.3: Important factors while planning merchandise quality Sr. No. Factors Importance Scale* K-S D value calculated 4 3 2 1 1. Target market 09 01 00 00 0.650 2. Competition 01 07 02 00 0.300 3. Retailer’s image 09 01 00 00 0.650 4. Store location 00 01 08 01 0.300 5. Stock turnover 01 07 01 01 0.300 6. Profitability 01 02 07 00 0.250 7. Manufactures V/S private brands 06 03 01 00 0.400 8. The customer service offered 08 01 01 00 0.550 9. Personnel 02 06 02 00 0.300 10. Perceived goods/service benefits 07 03 00 00 0.500 * 4- extremely important, 3- important, 2- partially important, 1- not important Kolmogorov-Smirnov (K-S) D value tabulated = 0.410; Level of significance (α) = 5 % (0.05) 6.16 Important factors while planning merchandise quality Target market is an extremely important factor for all the organised retailers included in the study except D Mart, as it is important to them while image (of retailers) is also an extremely important for all the organised retailers except HyperCITY, it is important for them in order to plan quality of merchandise offered to customers. Perceived goods/service benefits is extremely important for all the organised retailers except HyperCITY, D Mart and West Side, as it is important to them while customer service offered is an extremely important for all the organised retailers except Big Megamart and D Mart, as it is important for D Mart and partially important for Big Megamart. Manufacturers Vs private brands is an extremely important factor for all the organised retailers except Big Megamart, Star Bazaar, more.MEGAStore and D Mart, as it is important for Star Bazaar, more.MEGAStore and D Mart; while partially important for Big Megamart to plan the merchandise quality. 6.15 Important factors while planning merchandise Competition is an important for all the organised retailers except HyperCITY, Star Bazaar and D Mart, as it is partially important for HyperCITY and Star Bazaar while extremely important for D Mart to plan the merchandise quality. Personnel is extremely important for West Side and Reliance 168 Digital, while important for all the organised retailers except HyperCITY and D Mart, as it is partially important to them in order to decide the merchandise quality. Stock turnover is extremely important for more.MEGAStore while important for all other organised retailers except Max and D Mart as it is partially important for Max and not important for D Mart to plan merchandise quality. Profitability is an extremely important factor for HyperCITY and important for Big Megamart and Max, while partially important for all the other organised retailers. Store location is important for West Side, while partially important for all the other organised retailers except D Mart, as it is not important to them for taking a decision about the merchandise quality. 6.17 Retail assortment strategy All the hyper store organised retailers included in the study (i.e. Big Megamart, HyperCITY, Star Bazaar, more.MEGAStore and Big Bazaar) adopt wide variety & a deep assortment strategy, which is characterised by many goods/service categories & a large assortment in each category. While, super store organised retailers included in the study prefer narrow variety & a shallow assortment strategy characterised by few goods/service categories & a limited assortment in each category and narrow variety & a deep assortment strategy characterised by few goods/service categories & a large assortment in each category. D Mart and NEXT opt narrow variety & a shallow assortment strategy while others (i.e. West Side, Max and Reliance Digital) narrow variety & a deep assortment strategy. 6.18 Categories of merchandise offered Multi-format organised retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart) offer staples (basic), fashion, consumer durable and seasonal merchandise categories to their customers. Size, price, quality, brand, customer segments, discounts and schemes are the most common base to categorise the merchandise while displaying in the store by these organised retailers. However, all these organised retailers also categorise the merchandise based on the design except D Mart. Moreover, all these organised retailers categorise the merchandise based on customers’ needs except HyperCITY 169 while colour is also used as a base by more.MEGAStore and Big Bazaar in order to categorise the merchandise while displaying in the store. While, fashion-item organised retailers (i.e. Big Megamart, West Side and Max) offer fashion merchandise categories to their customers and use size, colour, design, price, quality, brand, customer needs, discounts and schemes as a base to categorise the merchandise while displaying in the store. However, West Side and Max also use customer segment as a base for categorization of merchandise. Moreover, electronics superstores (i.e. Reliance Digital and NEXT) offer consumer durables merchandise categories and categorise the merchandise based on size, design, price, quality, brand, customer segments, customer needs, discounts and schemes to display in the store. 6.19 Merchandise lines offered The industry is in for huge challenges in terms of targeting customers because of different and changing consumer behaviour regularly. Therefore, multi-format organised retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart) offer a wide range of merchandise in each category under the different merchandise lines. HyperCITY offers foods & grocery, health & beauty, home & furniture, fashion, electronics, toys and sports while Star Bazaar offers foods & grocery, personal & home care, home & kitchen needs, fashion & accessory merchandise lines to their targeted customers. more.MEGAStore’s offering basket contains grocery, fruits & vegetables, general merchandise, home & personal care, electronics, computer & accessories, mobile phone & accessories, apparels, footwear and sports merchandise lines. While Big Bazaar’s offering basket contains grocery & staples bazaar, farm fresh, kitchen & dining bazaar, bed & bath bazaar, books, magazines & stationery bazaar, luggage bazaar, home bazaar, personal bazaar, furniture bazaar, electronics bazaar, office bazaar, mobile bazaar, toys, sports & outdoors, fashion bazaar and gift merchandise lines. And, D Mart offers foods, toiletries & beauty merchandise, garments, bed & bath linen, kitchenware, toys & games, stationery, home appliances, baby care, 170 footwear, gifts and insecticides merchandise lines to targeted customers in their stores. Fashion-item organised retailers (i.e. Big Megamart, West Side and Max) categorise their offerings under the following different merchandise lines. Hyper store organised retailer, Big Megamart offers footwear, women’s wear, men’s wear, youth wear, kids wear, fashion mart, home mart and travel mart merchandise lines to their target customers in their stores. While, West Side (super store) offers women’s wear, men’s wear, kid’s wear, cosmetics & fragrances, footwear, accessories, home & gift and luggage and Max offers men’s wear, ladies wear, children’s wear, footwear and accessory merchandise lines to their target customers in their stores. Electronics superstores, Reliance Digital offers mobile phones, computers, TVs & audio, camera, home appliances, kitchen appliances, gaming and accessories while NEXT offers home entertainment, home appliances, phones & accessories, camera & camcorder, game & console, computers & tablets, personal care and writing instruments merchandise lines to their target customers in their stores. 6.20 Brands availability in each category of merchandise All the organised retailers make available generic, most of leading local, regional, national & international brands in different categories of the merchandise. Apart from these brands, fashion store organised retailers (i.e. Big Megamart, West Side and Max) also offer exclusive brands to their customers. Big Megamart offer Elitus, Leisha, Auburn Hill, Colt,Bay Island, Mea Casa, Karigari, Donuts, Ruggers, Skinn, and Edge as exclusive brands in their store while West Side offers Lovable, NUON, Gia, Fiorelli from London, Sidney, Frankfurt, Ascot, Richmond, L.O.V., Wardrobe, Westport, Zuba, Meenakari etc. as exclusive brands in their store. Max offers following exclusive brands like mélange, Code, Ginger, Fame, Forever, Forca and Juniors. Moreover, some organised retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar, Reliance Digital and NEXT) also offer store label brands to their customers. HyperCITY offers Fresh Basket, Everyday, Terzo, Avorio, Ebano, Technix, Maxit and Waitrose store label brands while Star Bazaar offers the 171 store label in different categories with name “Star Bazaar”. More.MEGAStore offers Feaster, Kitchenpromise and Selecta, etc. as store label brands while Big Bazaar with name “Such”. Reliance Digital offers store label brand, i.e. ‘Reconnect’ while NEXT offers under the name ‘NEXT’. 6.21 Availability of newly launched merchandise/brands & customisation All the organised retailers included in the study make sure the availability of newly launched merchandise/brands as soon as possible if falls in the categories of the merchandise they offer. Multi-format hyper store organised retailers like, HyperCITY offers the customised merchandise in the categories like clothing, footwear, jewellery, foods, toys, sports and fashion while Star Bazaar offers the customised merchandise in the categories like bakery, staples, non-vegetation, cleaning aids, health & beauty, kitchenware, electronics, home decor, stationary, apparel, footwear. more.MEGAStore offers the customised merchandise in the categories like computers, grocery, general merchandise, staples, non-vegetation, kitchenware, electronics, home decor, apparel while Big Bazaar offers the customised merchandise in the categories like computers, footwear, grocery, kitchen & dining, stationery, home & personal category etc. and multi-format superstore organised retailer (i.e. D Mart) offers the customised merchandise in some categories of the merchandise. Fashion-item organised retailer (i.e. Big Megamart) offers the customised merchandise in the categories like clothing, footwear, jewellery and fashion. While fashion-item superstores like West Side offers the customised merchandise in the categories like accessories; apparels, home & gift under the EASY SEARCH OPTION (If customers cannot find their size & colour, let them know, they will arrange it for customers, contact with sales staff on the floor for the same). Max offers the customised merchandise in the categories like accessories, apparels and footwear under the SEARCH OPTION (If customers cannot find their size & colour, let them know, they will arrange it for customers). Electronics superstores; Reliance Digital offers the customised merchandise in the category like accessories while NEXT offers the customised merchandise in the category like phones & accessories, game & console, computers etc. 6.22 Unique and complementary merchandise / brands 172 Hyper stores organised retailers like, HyperCITY offers some unique brands exclusively available in their store, i.e. Everyday, Fresh Basket, Terzo, Waitrose, Ebano, Avorio, City Style, Joojoobs, River Inc., Technix, Raleish, Maxit and also offers complementary merchandise like shoes, socks, pant & shirts, foam, shaving cream & razor, cookware, pet needs, vegetables, electronics and so on…….while Star Bazaar offers unique brand Star Bazaar’s exclusively available in their store in some categories like food & grocery, cleaning aids, packaged foods, beverages, personal care, Pooja needs, men’s, women’s & kid’s wear and also offers complementary merchandise in categories like cookware, pet needs, vegetables, electronics, cleaning aids, health & beauty, toys, kitchenware and so on……. more.MEGAStore offers some unique brands exclusively available in their store in food category - more., Feasters, Kitchen's Promise, Best of India and in home & personal care category - more., Enriche, 110%, Pestex, Paradise, Germex and also offers complementary merchandise in categories like accessories, shoes, socks, cookware, electronics, cleaning aids, health & beauty, toys, kitchenware and so on…….while Big Bazaar offers some unique brands exclusively available in their store (like Such in personal category) and also offers complementary merchandise in the categories like utensils, cookware, electronics, computers & mobile accessories and so on……. Fashion based hyper store organised retailer, Big Megamart offers some unique brands exclusively available in their store by Arvind Mills like Skinn, Elitus, Colt, Leisha, Donuts, Karigari, Bay Island, Ruggers, Mea Casa, Auburn Hill and Edge and also offers complementary merchandise like shoes, socks, pant & shirts, shaving cream & razor and so on……. While superstores like West Side offers some unique brands like Lovable, NUON, Gia, Fiorelli from London, Sidney, Frankfurt, Ascot, Richmond, L.O.V., Wardrobe, Westport, Zuba, Meenakari etc. and also offers complementary merchandise in categories like accessories, gifts, home accessories and so on……. Max offers some unique brands like mélange, Code, Ginger, Fame, Forever, Forca and also offers complementary merchandise in categories like accessories, footwear and apparels. 173 While electronics based superstore Reliance Digital offers a unique brand ‘Reconnect’ in different categories of the merchandise and offers a host of utility merchandise and accessories that are compatible with customers own merchandise & help customers get more of them. While superstores D Mart and NEXT do not offer any unique merchandise/brand to their customers. 6.23 Management of unprofitable merchandise Some organised retailers (i.e. Big Bazaar, more.MEGAStore, West Side, Max and Reliance Digital) try to find the way to convert unprofitable merchandise into profitable merchandise and then determine that whether unprofitable merchandise relates to profitable business relationships or not; if not then they drop them from the assortment. While, some organised retailers (i.e. HyperCITY, Star Bazaar, D Mart and NEXT) do not try to find the way to make them profitable, but directly determine that whether unprofitable merchandise relates to profitable business relationships or not; if not then they drop them from the assortment. However, Big Megamart usually drops unprofitable merchandise from the assortment. 6.22 Unique and complementary merchandise / brands But, D Mart offers complementary merchandise in some categories of merchandise like foods, toiletries & beauty merchandise, garments, bed & bath linen, kitchenware, toys & games, stationery, home appliances, baby care, footwear while NEXT offers various utility merchandise with core merchandise like utensils with Microwave Owen as complementary merchandise. 6.24 Major suppliers & discounts Manufacturers are first major suppliers of all the organised retailers included in the study while wholesalers are second major suppliers of all the organised retailers except West Side, Max and Reliance Digital and merchandise brokers are third major suppliers of all the organised retailers except Big Megamart, Star Bazaar and D Mart. However, most of the organised retailers also rely on commission agents to supply international brands except Star Bazaar, Big Bazaar and D Mart. Moreover, Star Bazaar also gets supplies for some cooperatively & individually owned offices. All the organised retailers get a trade, quantity and seasonal discounts from their suppliers. Moreover, most of the organised retailers also get promotional discounts except few (i.e. HyperCITY, Big Megamart and D Mart). 174 6.25 Important characteristics while selecting a vendor Table 6.4: Important characteristics while selecting a vendor Sr. No. Characteristics Importance Scale* K-S D value calculated 4 3 2 1 1. Reliability 10 00 00 00 0.750 2. Price-quality 10 00 00 00 0.750 3. Order-processing time 08 02 00 00 0.550 4. Exclusive rights 01 02 05 02 0.050 5. Functions provided 03 05 02 00 0.300 6. Information 03 06 01 00 0.400 7. Ethics 03 05 02 00 0.300 8. Guarantee 08 02 00 00 0.500 9. Credit 02 03 05 00 0.250 10. Long-term relationship 08 02 00 00 0.550 11. Reorders 08 02 00 00 0.550 12. Markup 03 03 04 00 0.250 13. Innovativeness 08 01 01 00 0.550 14. Local advertising 03 03 04 00 0.250 15. Investment 01 04 05 00 0.250 16. Risk 03 05 02 00 0.300 * 4- extremely important, 3- important, 2- partially important, 1- not important Kolmogorov-Smirnov (K-S) D value tabulated = 0.410; Level of significance (α) = 5 % (0.05) Reliability and price-quality are extremely important characteristics for all the organised retailers while selecting a vendor. Apart from reliability and price-quality, for HyperCITY, information, credit, long-term relationship, reorders, markup, investment and risk are extremely important characteristics; order-processing time, functions provided, ethics, guarantee, innovativeness and local advertising are important characteristics while exclusive rights are partially important characteristics of a vendor for selection. For Star Bazaar, order-processing time, long-term relationship, reorders, and innovativeness, are extremely important characteristics; functions provided, information, ethics, guarantee, markup, risk and local advertising are important characteristics while exclusive rights, credit and investment are partially important characteristics of a vendor for selection. 6.26 Collaborate (partner) relationship management (PRM) All the hyper & super store organised retailers show a keen interest in collaborate (partner) relationship management. They pay attention to maintaining the long-lasting relationship with all their partners, especially suppliers/vendors. 6.24 Major suppliers & discounts Order processing time, information, guarantee, long-term relationship and innovativeness are extremely important characteristics; functions provided, ethics, credit, reorders and local advertising are important characteristics while exclusive 175 rights, markup, investment and risk are partially important characteristics for more.MEGAStore to select a vendor. While, for Big Bazaar, order-processing time, ethics, guarantee, long-term relationship and innovativeness are extremely important characteristics; functions provided, information and reorders are important characteristics while credit, markup, local advertising, investment and risk are partially important characteristics and exclusive right is not important characteristics of a vendor to be selected. However, credit, reorders, markup and local advertising are extremely important characteristics; order processing time, functions provided, information, guarantee, long-term relationship, investment, risk are important characteristics; ethics and innovativeness are partially important characteristics and exclusive rights is not important characteristics for D Mart while selecting a vendor. Fashion hyper store organised retailer, i.e. Big Magamart pays importance to order-processing time, ethics, guarantee, reorders, markup, innovativeness and risk extreme level; while to information, credit, long-term relationship and investment are important levels and to exclusive rights, functions provided, local advertising at partial level in order to select a vendor. While, order-processing time, exclusive rights, functions provided, information, ethics, guarantee, long-term relationship, reorders, innovativeness and risk are extremely important characteristics; markup and investment are important characteristics while credit and local advertising are partially important characteristics for fashion based superstore West Side while selecting a vendor. And for another fashion-based superstore organised retailer, i.e. Max, order- processing time, guarantee, long-term relationship, reorders and innovativeness are extremely important characteristics; exclusive rights, ethics, and risk are important characteristics while functions provided, information, credit markup, local advertising and investment are partially important characteristics of a vendor for selection. Superstore based electronics-item organised retailers (i.e. Reliance Digital and NEXT) give importance to different characteristics in this manner. For Reliance Digital, order-processing time, functions provided, guarantee, long-term relationship, reorders, innovativeness and local advertising are extremely important characteristics; exclusive rights, information, ethics and risk are important characteristics while credit, markup and investment are partially important characteristics while selecting a 176 vendor. For NEXT, order-processing time, functions provided, guarantee, long-term relationship, innovativeness and local advertising are extremely important characteristics; information, credit, reorders, markup, investment and risk are important characteristics while exclusive rights and ethics are partially important characteristics of a vendor for selection. 6.27 Stock planning All the organised retailers prefer weekly and monthly stock planning of merchandise. However, multi-format organised retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart) do daily planning in case of perishable merchandise, frozen, vegetables, fruits, other farm merchandise, non- vegetables, etc. Moreover, Max also does quarterly planning for the stock as they make available different merchandise as per the different season. 6.28 Pricing objectives and approaches Most of the organised retailers included in the study (i.e. HyperCITY, Star Bazaar, D Mart, Max, Reliance Digital and NEXT) adopts maximising market share as pricing objective. HyperCITY offers the guarantee of the best price & claims that find a lower price elsewhere & they will double the difference in the case of branded consumer durables, home entertainment & technology merchandise. They also offer the best price & a quality guarantee in case of fruits & vegetables in order to achieve pricing objective. Star Bazaar offers more than 30000 merchandise in different of the merchandise categories with great quality and great price; moreover modern shopping environment backed by the strong value of TATA for achieving pricing objective. Max offers international fashion merchandise at affordable prices while Reliance Digital offers the widest & the latest range of merchandise at the lowest price guaranteed and NEXT offers merchandise & services at affordable price in order to be the market leader in the field of value retailing. 177 Other organised retailers (i.e. more.MEGAStore, Big Bazaar, Big Megamart and West Side) adopt merchandise quality leadership as a pricing objective. more.MEGAStore offers great quality merchandise & services with everyday lowest prices guaranteed while Big Megamart offers mega brands which are known for their quality in order to be the merchandise quality leader in the market. Big Bazaar offers merchandise & services with price & quality guarantee to achieve the pricing objective. Quality guarantee explains that all merchandise are sold at Big Bazaar are guaranteed to be at the best price & of good quality. Price guarantee explains that if within two days of purchase customers find a merchandise of the same brand, the quality level at lesser price return it within a day along with cash memo & Big Bazaar will double the value of the price difference. In order to achieve pricing objective, West Side introduced the policy to satisfy their customer with quality & value of the merchandise. Further, they offer return guarantee, if there is any quality grievance in order to merchandise quality leader in the field of fashion retailing. Most of the organised retailers included in the study use leadership-oriented pricing approach except Big Megamart, D Mart, West Side and NEXT as they use cost-oriented pricing approach in order to price the merchandise. 6.29 Pricing strategy High1-low pricing strategy (a form of value pricing strategy) is the most popular pricing strategy among organised retailers and adopted by HyperCITY, Star Bazaar, D Mart, Big Megamart and NEXT while more.MEGAStore and Big Bazaar adopt another form of value pricing strategy i.e. everyday low pricing strategy. However, perceived value-pricing strategy is also popular among superstore organised retailers (i.e. West Side, Max and Reliance Digital) in order to keep a promise to provide customers well-differenced value & experience in their stores. 6 30 Price adjustment strategy 6.30 Price-adjustment strategy Discount & allowance pricing (reducing the prices to reward customer responses), psychological pricing (adjusting prices for psychological effects) and promotional pricing (temporarily reducing the prices) are most common price- adjustment strategies among all the hyper & super store organised retailers and adopted by all of them in order to adjust or reduce the price of the merchandise and 178 services. However, all these organised retailers also adopt a segmented pricing strategy (adjusting the price to allow for the difference in customers, merchandise or locations) except D Mart. Moreover, some organised retailers (i.e. Star Bazaar, more.MEGAStore, Big Bazaar, West Side, Max and Reliance Digital) also opt dynamic pricing strategy (adjusting prices continually to meet the characteristics and needs of individual customers and situations). 6.31 Tactics followed to fight Price war Value added retailing is most preferred tactics by all the organised retailers except D Mart and NEXT as they follow to maintain and grow tactics to fight a price war. HyperCITY adds value by offering unmatched value (price-quality) merchandise & services while Star Bazaar, by offering great quality with the great price of the merchandise to their customers. more.MEGAStore enhances the value by offering great quality with an unbelievable price to their customers while Big Bazaar, by offering merchandise with price & quality guaranteed to the target market. Big Megamart adds the value by offering best quality while West Side by offering best-in-class merchandise & services at affordable price to their customers. Max enhances the value by the constant focus to develop the best merchandise & value offered to their customers while Reliance Digital follows the customer value proposition strategy in order to ensure value added retailing. 6.32 Customer response before Price change No organised retailers take concern of customer before changing the price of merchandise & services, but all these retailers take the response of customers through feedback and consider for the future course of action except D Mart and NEXT. Even, D Mart and NEXT do not take feedback to know the response of the customers after the price change. 6.34 Methods to develop brand image & customer loyalty (promotional strategy) HyperCITY, more.MEGAStore, Big Bazaar and Reliance Digital appeal to reason; Star Bazaar, D Mart, Max and NEXT appeal to emotion while Big Megamart and West Side appeal to sense in order to create a brand image among targeted customers. All the organised retailers use advertising, sales promotion, direct retailing, personal retailing, public relation / publicity and internet / interactive retailing methods for developing brand image and customer loyalty among existing & potential customers. 6.34.1 Advertising All the organised retailers do both indoor (in-store) as well as outdoors (out- store) advertising. Indoor (in-store) Advertising All the organised retailers included in the study use visual display, signage, point of purchase material, secondary packaging, and cross promotion methods of in-store advertising. However, Star Bazaar, more.MEGAStore, Big Bazaar and Reliance Digital also use pamphlets, while Big Megamart, HyperCITY, Big Bazaar and Max do loud speaker announcement about various discounts & schemes for the day in order to advertise. Moreover, Big Megamart, Big Bazaar and Reliance Digital also use electronic signage to advertise within the store. 6.33 Price charged & maximum retail price (MRP) HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar, Max and Reliance Digital always charge less than the MRP of most of the merchandise due to leadership-oriented approach, but some merchandise are still sold at MRP. While Big Megamart, D Mart, West Side and NEXT charge less than the MRP on most of the 179 merchandise, but due to cost-oriented approach, they charge equal to MRP to recover the margin on the merchandise. Outdoor (out-store) advertising Yellow page advertising is first while the newspaper and hoardings advertising are second most preferred modes of outdoor advertising among all the organised retailers, except Star Bazaar. Moreover, online advertising is also a common method adopted by organised retailers except Big Megamart, HyperCITY and D Mart. However, Big Megamart, Big Bazaar, D Mart, West Side, Max and Reliance Digital also advertise in magazines & journals. Some other methods of outdoor advertising are also used by organised 180 retailers i.e. television ads by Big Bazaar, West Side, Max and Reliance Digital; radio ads by Big Bazaar, D Mart and Reliance Digital; pamphlets / propaganda by HyperCITY, Star Bazaar, more.MEGAStore and NEXT; LCD screening by Big Megamart, Big Bazaar and Reliance Digital; wall paintings by D Mart, Max and Reliance Digital; fair & exhibitions by Max and cinema advertising by Big Bazaar. Advertising Budget No retail store’s management has provided information in this regard as it is not planned at store level. However, all the organised retailers spend less on advertising in comparison to any other industry especially outdoor advertising. Some organised retailers (i.e. Big Megamart, HyperCITY, Star Bazaar, more.MEGAStore, D Mart and NEXT) spend less in comparison to others organised retailers included in the study. 6.34.2 Sales Promotions Schemes, gifts, rebates, discounts, loyalty programs, premiums, prizes, rewards and contests are some popular tools opted by all the organised retailers included in the study. However, some organised retailers i.e. Big Megamart, HyperCITY, Star Bazaar, more.MEGAStore and Big Bazaar also provide a free sample as an integral part of the sales promotion. Schemes Buy & get free, bulk purchase discount, value / combo pack (offer) discount / deal, percentage off (discount) on MRP, up to percentage price off, flat percentage off / discount, price ( ) off / discounts, buy more and pay less (get more), flat price, starting at ₹, only / just for ₹, save ₹, ₹ onward / approx, fixed price scheme, ₹ to ₹, limited offer, offer / deal of the day or today’s offer, store (our) price and special offer schemes are some common sales promotion schemes introduced by all the organised retailers. However, these organised retailers also introduced a membership scheme and exclusive schemes for customers who wish to enrol for membership except D Mart in order to create some loyal customer base. 181 Moreover, hyper stores like HyperCITY introduces schemes like additional percentage off for Discovery Club Members and buy and get a percentage off schemes while Star Bazaar, weekly shopping scheme, best buys of the week schemes, Wow Wednesday - A Exclusive schemes for members, best price schemes, seasonal special schemes, weekend (Saturday & Sunday) offer schemes, mega offers schemes as additional sales promotion schemes. Monthly shopping (saving), best buys of the month, weekly shopping, the offer of the week, best price, super offer, best buys and offer price are some common sales promotion schemes among more.MEGAStore and Big Bazaar. However, more.MEGAStore also introduces schemes like club deals scheme, everyday lowest prices guaranteed scheme, Special occasion (March Mela) scheme, introductory price scheme, compare & save scheme, smart shopper scheme, goodbye MRP scheme (minimum 5% discount for members on selected merchandise), mega buy scheme, more value scheme, cash back offer scheme while Big Bazaar, lowest price scheme, Sabse Saste 5 Din scheme, public holiday sale scheme, the great exchange scheme, percentage extra discount scheme, exchange price scheme, Wednesday Bazaar – Hafte ka sabse sasta din scheme, Bachat Bazaar scheme, great discount scheme, never before price scheme, fashion weekend dhamaka scheme. Free item + Percentage price discount scheme and 2 % assured discount on all private-label brands are other sales promotion schemes introduced by D Mart. Schemes Fashion based organised retailers like Big Megamart introduces buy and get a percentage off scheme, value price scheme while West Side, up to percentage off sale schemes, sale schemes, additional 5 % discount for ClubWest Gold Card Members scheme, additional 5 % discount on ICICI Bank credit & debit card scheme (on purchase of 5000 & more, maximum discount ₹ 500) and Max, end of season sale scheme, hot deals scheme, best buy scheme, shop & get gift voucher, shop & get free gift scheme as additional sales promotion schemes. Electronics-item organised retailers like Reliance Digital introduces saving percentage scheme, best seller scheme, top deals scheme, freebie offer 182 scheme, Monsoon offer scheme, the greatest mismatch exchange sale scheme, easy EMI scheme while NEXT, assured gift scheme, best price scheme, lowest ever price scheme, lowest price guaranteed scheme and easy EMI scheme as additional sales promotion schemes. Free Samples Providing free samples to the customers is not so much popular sales promotion technique among organised retailers. However, some organised retailers (i.e. Big Megamart, HyperCITY, Star Bazaar, more.MEGAStore and Big Bazaar) provide a free sample. Big Megamart provides free samples in some fashion merchandise like perfumes, beauty merchandise, etc. while HyperCITY, in some categories like beauty & personal care. Star Bazaar, more.MEGAStore and Big Bazaar provide free samples of some merchandise in categories like grocery, FMCG etc. Rebates & Discounts All the organised retailers offer bulk purchase discount, value-pack (combo-pack) discount / bundle offer discount, discounts on the purchase of specific items, vouchers and coupons to their customers in order to promote sales. West Side offers gift vouchers in the denomination of ₹ 1000, 500, 100, 50 and a gift card can only be used when loaded with a minimum value of ₹ 250/-. Max offers gift vouchers in convenient denominations of ₹ 100, ₹ 250, ₹ 500, ₹ 1000, ₹ 2000 and ₹ 5000 to their customers. Reliance Digital offers gift vouchers in denominations of ₹ 500 to ₹ 10000 and coupons (like 10 % discount coupon for resQ range of accessories, ₹ 250 discount on labour charges, One free preventive maintenance service, 10 % discount on purchase of RCP for any other merchandise, 15 % discount on renewal of the RCP) to their customers. NEXT offers gift vouchers in denominations of ₹ 500 to ₹ 10000 to their customers. Big Bazaar offers vouchers in the denomination of ₹ 100 to 2000 & coupons to their customers to promote the sale. Moreover, some organised retailers (i.e. Big Bazaar, more.MEGAStore, Reliance Digital and NEXT) also offer the EMI facility to their customers in order to promote sales. HyperCITY offers EMI facility on furniture & electronics items to their customers while Star Bazaar offers EMI facility on durables to their customers. Big Bazaar offers EMI facility through Capital First, while more.MEGAStore offers EMI facility on electronics merchandise such as: Audio & Video and Large White Appliance. Reliance Digital offers EMI facility through Bajaj FinancerV on a credit card, HDFC, City, ICICI, Kotak, Axis HSBC credit card while NEXT offers EMI facility through credit card. Gifts All the organised retailer included in the study give free gifts on the purchase of specific items, assured gifts and premiums to their customers while promotion like West Side gives a free gift on purchase of ₹ 1000 & above of cosmetics. Max gives free gold plated earrings’ set worth ₹ 195 for shopping of ₹ 1200 & above and assured gift & premiums like shop for ₹ 1999 & get a Max gift voucher. Reliance Digital gives free microwave-safe utensils on the shopping of Reconnect Microwave Owen. more.MEGAStore gives free gifts on the purchase of specific items (Brooke Brand Red Label Natural care 50 gm on the purchase of Brooke Brand Red Label 250 gm; Glucose D 100 gm on the purchase of Complain Chocolate 500 gm; Safola Gold 1 lt. pp on the purchase of Safola Gold 5 lt. Jar; Pedigree Denta Stick on the purchase of Pedigree 1.5 Kg and assured gift & premiums like Casserole of ₹ 200 on the purchase of ₹ 2000 & more for members. Big Bazaar gives free gifts on the purchase of specific items and assured gift like shop for ₹ 499, get great prizes. 183 While some organised retailers (i.e. more.MEGAStore, Big Bazaar, West Side, Reliance Digital and NEXT) provide gift hampers on special occasions like Ganesh Chaurthi and Krishna Janmashtami etc. to their customers while promoting. Contests and other activities 184 HyperCITY organises holiday’s activities like Janmashtami, Ganpati Bappa Morya and contests like Ganesh Chaturthi contests for the customers to promote sales while Star Bazaar organises fun activities for the whole family to attract the customers. more.MEGAStore conducts lucky draws like Tata Indica lucky draw, Sweepstakes, holiday activities and contests (name their mascot contest to promote sales among customers. While, Big Bazaar organises lucky draws like bumper draw prizes, sweepstakes, holiday activities like Big Bazaar the great kitchen awards, great Indian kitchen festival and contests (back to school contest, India’s most stylist home contests, cricket contest in order to promote sales among customers. While, D Mart does not organise contests to attract the customers. Fashion-item organised retailers e.g. Big Megamart (a hyper store) organises lucky draws and fashion shows in order to promote sales among customers. While superstore organised retailers like West Side organised in association with KGAF hosted chef Vikas from Sanchos demonstrated easy Mexican cooking for Mumbai Janta!, GourmetWest in association with a Kala Ghoda arts festival brings an event on food with culinary experts in the varied fields! Max organises fashion shows, Max fashion icon contest, Max little icon contest, Max femina miss fashion icon contest, Max Manson cover girl contest, Max kids festivals, Max fluent your style contest, Max Indian kids fashion week, Max corporate style checks contest, Max colour crazy mania contests to promote sales. While Superstore based electronics-item organised retailers as Reliance Digital organises games & activities, holiday activities and film star are invited and the Maha mismatch exchange festival while NEXT organises lucky draws, sweepstakes, holiday activities and contests in order to promote sales among the customers. Customer loyalty programs, premiums, prize & reward All the organised retailers included in the study introduce membership as a part of loyalty programs in order to provide premiums, prizes and rewards to their customers. 185 Multi-format hyper stores like HyperCITY offers the discovery club membership card for their loyal customers. This card & membership facility can be availed free of cost. Members will get discovery points on the purchase of specific items. 1 discovery point = ₹ 1, these points are redeemable on the next or further purchasing. If discovery club member will shop of worth ₹ 3000/- & more with Axis Bank credit card/ debit card, customer/member will get additional 5% cash back and 10 % point back, while, Star Bazaar offers a club card program for their loyal customers. This program offers membership facility to their loyal customers, which can be availed free of cost. Members will get club card points on purchasing. ₹ 200 = 1 Club Card point = ₹ 1, these points are redeemable on the next or further purchasing. Special privileges for Club Card Members on Wednesday, double club card points for all your shopping and special offers for club card members. more.MEGAStore offers clubmore. a membership program for their loyal customers to SHOP more, SAVE more and EARN more. This program offers membership facility to their loyal customers, which can be availed free of cost. ‘clubmore.’ the membership card is valid for a lifetime. Every ₹ 100 worth purchase will give 10 reward points. At 500 reward points, members can redeem them and 1 point = 5 paisa. The points are valid for 2 years. The card can be used at any store of more. As other benefits, customers can enjoy a range of exclusive promotion & offers, get details on the latest clubmore. Events & news, get mouthwatering recipes, sparkling home care tips, special benefits besides the regular offers & promotions, SMS alerts and emails regarding special offers on merchandise & services, special clubmore. member section at more. store, members can know their points by SMS more<space><clubmore.card no.> to 565070 and alternatively members can also login to http/:clubmore.morestore.com. These benefits are subject to change by time to time. While, Big Bazaar introduces membership program by offering payback card & profit club card for their loyal customers. Payback card is free of cost and valid for a lifetime. Customer loyalty programs, premiums, prize & reward Every ₹ 40 worth purchase will give 1 reward points and 200 points = ₹ 40. As other benefits, customers can enjoy a benefit from exclusive as well as bonus point offers, attractive possibilities to redeem your points. Profit club card is one type of fixed deposit 186 with guaranteed rebate paid service (at ₹ 100), by which customers can pay ₹ 10000 and they will get the value card of ₹ 12000. Members can use up to ₹ 1000 for shopping per month. A card will be valid for 18 months. These benefits are subject to change by time to time. Nevertheless, multi-format superstore D Mart does not introduce customer loyalty programs, which is quite unexpected in the contemporary retailing scenario. Fashion-item organised retailers as Big Megamart introduces smart one card for their loyal customers. This card facility can be availed by making purchases of ₹ 1500 on a single invoice and get 26 % additional discount on regular purchase and flat 50 % discount on your birthday. West Side has introduced clubWest membership program for their loyal customers. ClubWest is a two-tier program, which consists clubWest classic and clubWest gold. A purchase of ₹ 2000/- on the same day entitles customers to a complimentary membership into a clubWest classic. Alternatively, customers can enrol by paying a nominal one-time fee of ₹ 150/- & avail ₹ 100 vouchers. For classic members ₹ 100 = 1 reward point = ₹ 1 at the time of redemption. While a purchase of ₹ 5000/- on the same day entitles customers to a complimentary membership into clubWest gold. For gold members ₹ 80 = 1 reward point = ₹ 1 at the time of redemption. Reward points can be viewed by logging on to www.mywestside.com. Other advantages of membership are dedicated clubWest desk for a member's assistance at each of the stores, Exclusive shopping hours only for members during sales, unprecedented access to a host of privileges and services through our exclusive tie-ups from time to time, advance intimation of all in-store promotions and special offers through direct mailers, special offers during promotions, gift vouchers from time to time, special birthday offer, get invited through special events & previews, year-round discount at selected premium restaurant and special offers from their partners. These benefits are subject to change by time to time. Customer loyalty programs, premiums, prize & reward HSBC West Side credit card – Customer can enjoy shopping advantages like 4 reward points for every ₹ 100 spent at West Side, complimentary gift vouchers, priority billing counters, home delivery of altered garments and exclusive sale previews. 187 Max introduces The Inner Circle card program for their loyal customers. This program offers membership facility to their loyal customers, which can be availed free of cost. ‘The Inner Circle’ membership card has a lifetime validity. But, if a member in Silver tier does not shop for 12 months, the membership may be deactivated. Even, the member wishes to reactivate the membership, a fee of ₹ 200 will be levied. The Inner Circle membership program is a three-tire program, i.e. Silver, Gold and Platinum. If any customer spends below ₹ 15000/- will be eligible for Silver Card, between ₹ 15000 to ₹ 6499/- will be eligible for Gold card and above ₹ 65000/- will be eligible for a Platinum card. Cardholders will get 2 reward points on the purchase of ₹ 100, exclusive promotions for them, off the sale preview, out of the store offers (at other Landmark companies) are some common benefits of the cards but gold & a platinum card holder will get 100 & 300 birthday bonus points. While Superstore based electronics-item organised retailers like Reliance Digital introduces a RelianceOne card membership program for their loyal customers. This program offers membership facility to their loyal customers, which can be availed free of cost. ‘RelianceOne’ membership card is valid for a lifetime. If any customer shops for ₹ 100, the customer will get 70 paisa back in his/her account, this money can be used for purchasing at any reliance store. Members also get lifetime reward point’s validity, monthly account e-mail updates, special offer updates via SMS and E-mails, festive & monthly promotion catalogue. These benefits are subject to change by time to time. NEXT introduces My Card membership program for their loyal customers. This program offers membership facility to their loyal customers, which can be availed free of cost. ‘My Card’ membership card is valid for a lifetime. If any customer shops for ₹ 100, the customer will get 1 reward point which is equal to ₹ 1, these points can be redeemed for purchasing at any NEXT store. Table 6.5: Impact of sales promotion Sr. No. 6.34.4 Personal Retailing All the organised retailers (included in the study) adopt the face-to-face retailing technique of personal retailing within the store. 6.34.3 Direct Retailing Direct emails and mail order catalogues are most common direct retailing techniques adopted by all the organised retailers except Big Megamart. While SMS retailing is also most adopted direct retailing technique among organised retailers except Star Bazaar. The customer can SMS WS<Space><City Name> to 5676714 to know about great offers and promotions at West Side and MF<City Name> to 56070 to know about nearer store, great offers and promotions at Max. However, some organised retailers (i.e. HyperCITY, more.MEGAStore, Big Bazaar, Reliance Digital and NEXT) are also involved in tele-retailing / mobile retailing form of direct retailing. Customer loyalty programs, premiums, prize & reward Statements Agreement Scale* K-S D value calculated 4 3 2 1 188 1. Sales Promotions influence customer purchase intention. 05 05 00 00 0.500 2. Sales Promotions increase customer satisfaction level. 00 00 07 03 0.050 3. Sales Promotions affect the frequency to visit the outlet. 04 03 02 01 0.200 4. Sales Promotions play a vital role to spend more. 02 05 02 01 0.200 5. Sales Promotions entice to repurchase. 00 00 03 07 -0.250 6. Customers purchase more quantity due to sales promotion. 03 04 02 01 0.200 7. Your store does adequate sales promotion activities. 07 03 00 00 0.500 8. Your store justifies its punch line. 10 00 00 00 0.750 * 4- extremely agree, 3- agree, 2- partially agree, 1- not agree Kolmogorov-Smirnov (K-S) D value tabulated = 0.410; Level of significance (α) = 5 % (0.05) 6.34.5 Public Relation/Publicity Corporate brochures, annual reports, press release, press conferences, video press conferences are some most common medium to maintain relation with public and adopted by all the organised retailers. An internet release is also a preferred medium for publicity and adopted by all the organised retailers except D Mart. Some organised retailers i.e. Big Bazaar, more.MEGAStore, Reliance Digital and D Mart use posters to make publicity. Some organised retailers (i.e. HyperCITY, more.MEGAStore, Big Bazaar, West Side, Max and Reliance Digital) organise and sponsor special events. HyperCITY organises Ganesh Chaturthi Contests, Janmashtami, Ganpati Bappa Morya as special events while more.MEGAStore organises fruit art activity. Big Bazaar organises cricket contest while Reliance Digital organises events and calls 189 celebrities i.e. Saif Ali Khan, Kareena Kapoor, Ranveer Singh, John Abraham, Arjun Kapoor, Parneeti Chopra, Priyanka Chopra, Shahid Kapoor, Bipasha Bashu, Emraan Hashmi, Esha Gupta and so on… Fashion superstore organised retailers as West Side conducts special events like Light a Diya - Help a Child – Purchase a Diya and Light it at Westside during the Diwali Promotion. The fund collected is donated to NGOs to bring smiles to the faces of underprivileged children and Angels Tree – Purchase a “Silver Star” or a “Gold Star” during our Christmas Promotion, and decorate our Angels Tree. The money collected is donated to various NGOs across the country working with underprivileged children. Max is a fashion partner in Ponds Femina Miss India 2013 and Max was a fashion partner in the Cosmopolitan Fun fearless female & male Award 2012 as special events and sponsorship in order to publicity. Some multi-format organised retailers (i.e. HyperCITY, Big Bazaar, more.MEGAStore) and fashion-item organised retailers (i.e. Big Megamart, West Side, Max) conduct social events & social development programmes for publicity. HyperCITY conducts social development programmes in association with Sathi Foundation, org, blood donation camp in HyperCITY July 16, 2012, collaboration with Nanhi Kali, a well-known girl NGO, HyperCITY associates with Childline India Foundation. Big Bazaar organises social events like The Great Indian Kitchen Awards, International Yoga week. More.MEGAStore organises social events like craft workshops. 6.34.5 Public Relation/Publicity Fashion superstore organised retailers as West Side organises social development programs like Dil Se Dejiye Das Rupees – They ask to add just ₹ 10 to the bill of their customers to help & treat underprivileged cancer patients at the Tata medical centre, Kolkata; www.tmckolkata.com while Max organises social events and social development programs like LIFE Trust (an NGO) to introduce various programs at the school and pre-school level to improve the quality and accountability of the educational system for the underprivileged children, beat diabetes campaign, Max wildlife campaign, and other programs like - blood donation drives, clothes 190 donation drive, go green - an effort to save the environment, Max participated in the Sunfeast Marathon 2010 for publicity among public. 6.34.6 Internet/Interactive Retailing All the organised retailers (included in the study) prefer e-mails and websites as interactive retailing tools in order to create a brand image. Multi-format organised retailers as HyperCITY’s e-mail id is feedback@hypercityindia.com and website is www.hypercityindia.com; Star Bazaar’s e-mail id is starhelp@trent-tata.com and website is https://www.starbazaarindia.com; more.MEGAStore’s e-mail ids are contactus@morestore.com and feedback@morestore.com and the website is www.morestore.com; while Big Bazaar’s e-mail ids are csd.bb@futuregroup.com and writeus@fvrl.in and website is www.bigbazaar.com and D Mart’s e-mail id is suggestion@dmartindia.com and website is www.dmartindia.com. Fashion based organised retailers i.e. Big Megamart’s e-mail id is customercaremegamart@arvindbrands.com and website is http://megamartstores.com; West Side’s e-mail id is mywestside@trent-tata.com and website is http://www.mywestside.com while Max’s e-mail id is customercare@maxretailstores.com and website is www.maxfashionindia.com. Electronics based organised retailers i.e. Reliance Digital’s e-mail id is reliancedigital@ril.com, customersupport@resq.in and website is www.reliancedigital.in while NEXT’s e-mail id is customercare@next.co.in and website is www.next.co.in. Use of social networking sites is also popular among all the organised retailers except D Mart to create a brand image. Multi-format hyper store organised retailers as HyperCITY uses Facebook (www.facebook.com/HyperCITY), twitter (www.twitter.com/HyperCITYindia) & YouTube (www.youtube.com/hypercityindia); Star Bazaar also uses Facebook (www.facebook.com/starbazaarindia), twitter (www.twitter.com/starbazaarindia) and YouTube (www.youtube.com/starbazaarindia) social networking sites; while more. MEGAStore and Big Bazaar use only Facebook i.e. respectively www.facebook.com/morestore and www.facebook.com/BigBazaar. Fashion based superstore organised retailers i.e. Big Megamart uses Facebook (https://www.facebook.com/BigMegamartDapodi), twitter 191 (www.twitter.com/BigMegamartDapodi) and YouTube (www.youtube.com/BigMegamartDapodi), West Side uses only Facebook (www.facebook.com/westsidefanpage) while Max uses Facebook (www.facebook.com/maxfashion), twitter (www.twitter.com/maxfashion), YouTube (www.youtube.com/maxfashionindia), flicker (www.flickr.com/maxfashionindia), and Pinterest (www.pinterest.com/maxfashionindia) as social networking sites. Electronic based superstore organised retailers i.e. Reliance Digital uses Facebook (www.facebook.com/reliancedigital), twitter (https://twitter.com/reliancedigital), YouTube (www.youtube.com/user/reliancedigitaltimes), and foursquare (www.foursquare.com/reliancedigital) while NEXT uses Facebook (www.facebook.com/next.co.in?sk=wall), twitter (www.twitter.com/next_retail), and YouTube (www.youtube.com/nextretail) as social networking sites to create a brand image. Online feedback is also a popular tool of interactive retailing among all the organised retailers except Big Megamart and West Side. Moreover, electronics super store organised retailers (i.e. Reliance Digital and NEXT) take online feedback on Facebook. However, some organised retailers (i.e. HyperCITY, Star Bazaar and Max) use online discussion forum while Superstore based electronics-item organised retailers use e-pay as interactive retailing tools to create a brand image among customers. 6.35 Celebrity services for promotions Use of celebrity services is not so popular among organised retailers. However, Big Bazaar used the services of actors (Hussain), actresses (Asin) and cricketers like MS Dhoni (Indian Cricket Team), Irfan Pathan, Mohd. Kaif, Parthiv Patel, Murli Kartik for promotion. Reliance Digital calls actors and actresses like Saif Ali Khan, Kareena Kapoor, Ranveer Singh, John Abraham, Arjun Kapoor, Parneeti Chopra, Priyanka Chopra, Shahid Kapoor, Bipasha Bashu, Emraan Hashmi, Esha Gupta and so on… to visit their store as part of a promotion. Fashion-item organised retailers i.e. Big Megamart, West Side and Max use celebrity services for promotions. Big Megamart uses the services of models for the 192 promotion of their brands at fashion shows, while, West Side and Max use services of models, actors and actresses in TV commercials. 6.36 Tag line and logo All the organised retailers use customer-centric tagline except HyperCITY, West Side, Reliance Digital and NEXT as these use store-centric tagline. Star Bazaar uses customer-centric tagline, i.e. HELPING YOU SPEND LESS! more.MEGAStore uses customer-centric tagline, i.e. Hamesha Extra. Big Bazaar uses customer-centric tagline, i.e. Naye India Ka Bazaar. D Mart uses customer-centric tagline, i.e. Daily Discounts Daily Savings…!!! Big Megamart uses customer-centric tagline, i.e. Love brands. Love values. Max uses customer-centric tagline, i.e. Look good. Feel good. Star Bazaar uses customer-centric tagline, i.e. HELPING YOU SPEND LESS! more.MEGAStore uses customer-centric tagline, i.e. Hamesha Extra. Max uses customer-centric tagline, i.e. Look good. Feel good. 193 HyperCITY uses store-centric tagline, i.e Big store. Big savings. West Side uses store-centric tagline, i.e. Endless Possibilities Reliance Digital uses store-centric tagline, i.e. Happiness in store NEXT uses store-centric tagline, i.e. Best Brands. Best Bargains. HyperCITY uses store-centric tagline, i.e Big store. Big savings. West Side uses store-centric tagline, i.e. Endless Possibilities Reliance Digital uses store-centric tagline, i.e. Happiness in store NEXT uses store-centric tagline, i.e. Best Brands. Best Bargains. 6.37 Location preference Main street locations are most preferred locations by all the organised retailers except Big Megamart and HyperCITY, as they prefer inner city locations in order to open their stores. However, Max prefers both types of locations. All the organised retailers open their store in shopping malls/shopping centres and strip centres except HyperCITY and D Mart as HyperCITY prefers only shopping malls/shopping centres while D Mart prefers strip centres. Moreover, fashion-item superstores i.e. West Side and Max also prefer to open their stores in outlet centres. 6.38 Important factors while selecting retail store location Table 6.6: Important factors while selecting retail store location Sr. No. Factors Importance Scale* K-S D value calculated 4 3 2 1 1. Population size & characteristics 06 02 02 00 0.350 a. Total size & density 09 01 00 00 0.650 b. Age distribution 03 05 02 00 0.300 c. Average educational level 00 07 03 00 0.250 Table 6.6: Important factors while selecting retail store location 194 d. % of residents owning homes 00 01 08 01 0.300 e. Total disposable income 08 02 00 00 0.550 f. Per capita disposable income 03 07 00 00 0.500 g. Occupation distribution 00 04 06 00 0.250 h. Trends 10 00 00 00 0.750 2. Availability of labour 01 06 02 01 0.200 a. Management 02 08 00 00 0.500 b. Management trainee 01 08 01 00 0.400 c. Clerical 02 06 02 00 0.300 d. Labour 02 06 02 00 0.300 3. Closeness to source of supply 08 02 00 00 0.550 a. Delivery cost 10 00 00 00 0.750 b. Timeliness 09 01 00 00 0.650 c. No of manufacturers & wholesalers 07 03 00 00 0.500 d. Availability & reliability of merchandise lines 10 00 00 00 0.750 4. Promotion facilities 01 03 03 03 -0.050 a. Availability & frequency of media 02 06 02 00 0.300 b. Costs 05 03 02 00 0.300 c. Waste 00 04 06 00 0.250 5. Economic base 04 03 03 00 0.250 a. Dominant industry 06 03 01 00 0.400 b. Extant of diversification 03 04 03 00 0.250 c. Growth projections 08 02 00 00 0.550 d. Freedom from economic & seasonal fluctuations 06 04 00 00 0.500 e. Availability of credit & financial facilities 01 05 04 00 0.250 6. Competitive situation 05 03 02 00 0.300 a. 6.37 Location preference No & size of existing competitors 09 01 00 00 0.650 b. Strength & weakness of competitors 05 05 00 00 0.500 c. Short & long run outlook 01 08 01 00 0.400 d. Level of saturation 02 06 02 00 0.300 7. Availability of store locations 03 06 01 00 0.400 a. No & type of locations 10 00 00 00 0.750 b. Cost of site 09 01 00 00 0.650 c. Access to transportation 01 08 01 00 0.400 d. Owning Vs leasing opportunities 01 01 08 00 0.250 e. Leasing period & conditions 01 02 07 00 0.250 f. Visibility of site 08 02 00 00 0.550 g. Size & shape of trading area 05 05 00 00 0.500 h. Zoning restrictions 00 02 07 01 0.150 i. Parking facilities 09 01 00 00 0.650 j. Size & shape of the building 01 08 01 00 0.40 8. Regulations 00 05 04 01 0.250 a. Taxes 04 03 03 00 0.250 b. Licensing 05 01 04 00 0.250 c. Operations 01 06 03 00 0.250 d. Minimum wages 00 03 07 00 0.250 e. Zoning 00 02 08 00 0.250 * 4- extremely important, 3- important, 2- partially important, 1- not important Kolmogorov-Smirnov (K-S) D value tabulated = 0.410; Level of significance (α) = 5 % (0.05) 195 Closeness to the source of supply is extremely important for all the organised retailers except West Side and Max, as it is important to them while selecting the store location. Population size & characteristics are extremely important for all the organised retailers except Big Bazaar, West Side, Max and Reliance Digital as for West Side and Reliance Digital, it is important while, for Big Bazaar and Max, it is partially important while selecting the store location. Competitive situation is extremely important for all the organised retailers except West Side, Reliance Digital, NEXT, Big Megamart and HyperCITY as it is important for West Side, Reliance Digital and NEXT while partially important for Big Megamart and HyperCITY while selecting the store location. The economic base is extremely important for organised retailers i.e. Big Bazaar, more.MEGAStore, West side and Reliance Digital; important for HyperCITY, Max and NEXT while partially important for Star Bazaar, Big Megamart and D Mart. 6.37 Location preference Availability of store locations is an important factor for all the organised retailers except West Side, Max, Reliance Digital and HyperCITY, as it is extremely important for West Side, Max and Reliance Digital while for HyperCITY, partially important while selecting a store location. Availability of labour is also an important factor for all the organised retailers except Big Megamart, Reliance Digital, NEXT and West Side, as it is partially important for Big Megamart and Reliance Digital while not important for NEXT; however, extremely important for West Side while selecting a store location. Promotion facilities are important factors for fashion-item organised retailers i.e. Big Megamart, West Side and Max; partially important for some multi-format organised retailers i.e. HyperCITY, Big Bazaar and D Mart while not important for some organised retailers i.e. Star Bazaar, more.MEGAStore and NEXT; however, extremely important for Reliance Digital while selecting a store location. Regulations are important for all the organised retailers except HyperCITY, more.MEGAStore, Big Bazaar, D Mart and NEXT as it is partially important for some multi-format organised retailers i.e. HyperCITY, more.MEGAStore, Big Bazaar and D Mart while not important for NEXT while selecting store locations. 196 6.41 Service provided All the organised retailers provide free as well as paid services to their customers. 6.39 Store timings There is no uniformity among organised retailers in store timings. HyperCITY and Big Megamart open their stores between 11.00 am to 9.30 pm on all the days of the week. Star Bazaar opens its stores for 14 hours from 8.30 am to 10.30 pm on all the days of the week. more.MEGAStore opens its stores for 13 hours from 09.30 am to 10.30 pm on all the days of the week. Big Bazaar opens its stores for 12 hours from 10.30 am to 10.30 pm on all the days of the week. While, D Mart also opens its stores for 12 hours from 10.00 am to 10.00 pm on all the days of the week. Max opens its stores for 11 hours from 11.00 am to 10.00 pm on all the days of the week. West Side opens its stores at 11.00 am to 10.30 pm on weekdays and 11.00 am to 10.00 pm on Sunday. Reliance Digital opens its stores between 11.00 am to 10.00 pm on all the days of the week, but on Wednesday for 12 hours from 10.00 am to 10.00 pm. NEXT opens its stores between 10.00 am to 09.30 pm on all the days of the week except Monday as it is closed. 6.40 Customer service strategy All the organised retailers included in the study adopt patronage solidifiers (the low-cost little things that increase loyalty, like courtesy & suggestion selling) as a customer service strategy except more.MEGAStore, Reliance Digital and NEXT as they adopt patronage builders (high-cost activities like transaction speed & credit). Free services Complaints and exchange / return handling (as per the exchange / return policy of the respective retailer), packaging & gift-wrapping, information, safety of personal things, acceptance of all international credit & debit cards, personal shoppers, well maintained washrooms (ladies and gents) & drinking water, air-condition environment, lighting, music, CCTV for security, fire safety, billing counters are some common customer services provided by all the organised retailers included in the study. Moreover, all the organised retailers provide some additional customer services. 197 Multi-format organised retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart) provide following additional free services: Multi-format organised retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart) provide following additional free services: HyperCITY provides credit facility (finance on 0 % interest, EMI on electronics and furniture items), home delivery, alteration facilities, installations, parking, baby strollers, baby-sitting, fitting (trial) room, lift & staircase facility, after sale service, demonstrations, entertainment facilities, special facilities for physically handicapped, shopping carts, warranties, loud- speaker announcements, back office support, ATM terminal, customers care service (by toll free no. 18002097172 & 18001020468) and repairing services as free services. Star Bazaar provides home delivery, alteration facilities, installations, extended store hours, parking, fitting (trial) room, escalator & staircase facility, after sale service, demonstrations, entertainment facilities, shopping carts, warranties, back office support, weight counters of loose grocery, ATM terminal, customers service desk, customer loyalty desk, eye testing service and guarantee on Star brand merchandise as free services. More.MEGAStore provides credit (easy 0 % interest EMI & spot finance facility), home delivery, cash on delivery, alteration, installation, trial purchase, phone orders, parking, baby strollers, fitting rooms, fur storage, lift, escalator & staircase facility, after sales service & support, demonstrations, entertainment facilities, special facilities for physical handicapped, customers support service centre / desk, customer helpline number service centre (08652906676), back office, shopping carts, free testing of the merchandise, free recipes on a store website, SMS facility to find out the nearest store via SMS type more(space)<yourpincode> and sent to 56070, to give feedback SMS more<fdbk><yourcomment> to 56070 as free services. Free services Big Bazaar provides credit (easy 0 % interest EMI & spot finance facility), home delivery, cash on delivery, alteration, installation, trial purchase, phone orders, parking, baby strollers, baby-sitting, fitting rooms, fur storage, lift, escalator & staircase facility, after sales service & support, demonstrations, entertainment facilities, medical facilities, special facility for physical handicapped, exchange counter, TV sets at different places in store, back office with HyperCITY provides credit facility (finance on 0 % interest, EMI on electronics and furniture items), home delivery, alteration facilities, installations, parking, baby strollers, baby-sitting, fitting (trial) room, lift & staircase facility, after sale service, demonstrations, entertainment facilities, special facilities for physically handicapped, shopping carts, warranties, loud- speaker announcements, back office support, ATM terminal, customers care service (by toll free no. 18002097172 & 18001020468) and repairing services as free services. Star Bazaar provides home delivery, alteration facilities, installations, extended store hours, parking, fitting (trial) room, escalator & staircase facility, after sale service, demonstrations, entertainment facilities, shopping carts, warranties, back office support, weight counters of loose grocery, ATM terminal, customers service desk, customer loyalty desk, eye testing service and guarantee on Star brand merchandise as free services. More.MEGAStore provides credit (easy 0 % interest EMI & spot finance facility), home delivery, cash on delivery, alteration, installation, trial purchase, phone orders, parking, baby strollers, fitting rooms, fur storage, lift, escalator & staircase facility, after sales service & support, demonstrations, entertainment facilities, special facilities for physical handicapped, customers support service centre / desk, customer helpline number service centre (08652906676), back office, shopping carts, free testing of the merchandise, free recipes on a store website, SMS facility to find out the nearest store via SMS type more(space)<yourpincode> and sent to 56070, to give feedback SMS more<fdbk><yourcomment> to 56070 as free services. Free services Big Bazaar provides credit (easy 0 % interest EMI & spot finance facility), home delivery, cash on delivery, alteration, installation, trial purchase, phone orders, parking, baby strollers, baby-sitting, fitting rooms, fur storage, lift, escalator & staircase facility, after sales service & support, demonstrations, entertainment facilities, medical facilities, special facility for physical handicapped, exchange counter, TV sets at different places in store, back office with 198 warehouse, weight counter for loose grocery, idli & dosa paste grinding, loudspeaker announcement, mirrors in fashion bazaar section, loan facility, customers service desk, loyalty card desk, foreign currency exchange counter, ATM terminal, customer helpline number service centre (18002002255), shopping carts, free testing of the merchandise, free talk time for TATA Docomo mobile number (shopping of worth ₹ 400 = ₹ 10 talk time free), acceptance of meal vouchers of Sadox & Accor as free services. D Mart provides parking, baby Strollers, fitting (trial) room, shopping carts, warranties, loudspeaker announcements as free services. Fashion-item organised retailers (i.e. Big Megamart, West Side and Max) provide following additional free services. Big Megamart provides alteration facilities, parking, baby Strollers, baby-sitting, Fitting (trial) room, lift, escalator and stairs facility, after sale service, demonstrations, entertainment facilities, game zones, music zone (Karaoke, etc.), special facilities for physically handicapped, shopping carts, warranties, and loudspeaker announcements as free services. West side provides home delivery of altered garments, alteration facilities, installations, staff on late hours for members, extended store hours for members, baby-sitting, Fitting (trial) room, lift, escalator & staircase facility, demonstrations, entertainment facilities, pulling shopping carts, quality assurance, mirror on every floor, good back office & management office for support, Sensormatic system (tag) for merchandise, customers service desk and easy search option facility. While, Max provides home delivery, alteration facilities, parking, fitting (trial) room, lift, escalator & staircase facility, demonstrations, entertainment facilities, pulling shopping carts, mirrors in store, Sensormatic system (tag) for merchandise, customers service associate, easy search option facility, TV sets at billing counters, loudspeaker announcement, pass of different activities where Max is sponsor, and SMS service to know the store location & reward points (SMS ‘TIC’ to 56767). Superstore based electronics-item organised retailers (i.e. Reliance Digital and NEXT) provide following free additional customer services. Reliance Digital provides credit (easy EMI & spot finance facility), home 199 delivery, installation through resQ, parking, restroom, baby-sitting, fur storage, staircase facility, after sales service & support through resQ, demonstrations, entertainment facilities, game. Free services zone, Sensormatic system (tag) for merchandise, customer service centre/desk, home theater experience zone, customer care service centre (02267276727), my finance desk, back office, online green room service, online reliance digital solution box, maintenance & repair, specialised customer training, one stop solution for repairs of all Reconnect brands as free services. While, NEXT provides credit (easy 0 % interest EMI & spot finance facility), home delivery, cash on delivery, installation, mail & phone orders with advance payment, parking, fur storage, staircase facility, after sales service & support, demonstrations, entertainment facilities, game zone, Sensormatic system (tag) for merchandise, customers service centre/desk, customer helpline number service centre (0124-4993500), back office, free gifts, online information regarding merchandise care, 30 days replacement guarantee, 30 days change of mind policy and shopping bags as free services. Paid services Multi-format organised retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart) provide following additional free services: HyperCITY provides gift certificate (of ₹ 50, 100, 250, 500, 1000, 2000), trial purchase, special sales for regular customers, bridal registry, restaurants (food planet within the store), beauty salon (Kapil’s Salon), shopping bags (₹ 1, ₹ 3 and ₹ 4 respectively, for small, medium and large plastic bags), refreshment facilities, extended warranties on electronics and durables as paid services. Star Bazaar provides special sales for regular customers, restaurants (Live Bakery, Dali), fresh grinding service (Chakki), Sodexos, shopping bags (₹ 1, ₹ 2 and ₹ 5 respectively, for small, medium and large plastic bags), refreshment facilities (Coffee Café Day, Confectionary), Tata credit card as paid services. More.MEGAStore provides gift certificates, special sales for regular customers, restaurant & refreshment facility through the live bakery, shopping bags (₹ 1, ₹ 2 and ₹ 3 respectively for small, medium and large plastic bags) and repairing as paid services. Big Bazaar provides gift certificates, special sales for regular customers, restaurants & 200 refreshment facility (at Foodi counter), beauty salon (Javed Habib Hair Expresso), shopping bags (₹ 1, ₹ 2, ₹ 3 and ₹ 4 respectively, for small, medium, large and extra-large plastic bags) Mehndi corner, weight-checking machines, NAVRAS – jewellery bazaar, fixed deposit with guaranteed rebate as paid services. While, D Mart provides alteration, shopping bags (₹ 1, ₹ 2 and ₹ 3 respectively, for small, medium and large plastic bags) and refreshment facilities as paid services. Fashion-item organised retailers (i.e. Big Megamart, West Side and Max) provide following paid services. Big Megamart provides gift certificate, trial purchase, bridal registry, restaurants, beauty salon, shopping bags (₹ 1, ₹ 4 and ₹ 6 respectively, for small, medium and large plastic bags) and refreshment facilities as paid services. West Side provides gift certificate (of ₹ 50, 100, 500, 1000 in the denomination), trial purchase, special sales for regular customers, parking, restaurants (food café, Coffee parlour, Gourmet West – food section), Shopping bags (₹ 3, ₹ 5 and ₹ 7 respectively for small, medium and large plastic bags) and refreshment facilities as paid services. While Max provides gift certificate (of ₹ 100, 250, 500, 1000, 2000 and 5000 in the denomination), special sales for regular customers, restaurants, shopping bags (₹ 3 and ₹ 5 respectively for small and large plastic bags) and refreshment facilities as paid services. Paid services Superstore based electronics-item organised retailers (i.e. Reliance Digital and NEXT) provide following free additional customer services. Reliance Digital provides gift certificate (of ₹ 500 to 10000 in the denomination), special sales for regular customers, shopping bags (₹ 3, ₹ 6 and ₹ 11 respectively, for small, medium and large plastic bags), health checkup plan – a preventive maintenance, enhanced utility through resQ accessories and extended warranty service as paid service. NEXT provides gift certificate (of ₹ 500 to 10000 in the denomination), special sales for regular customers, maintenance & repair, extended warranty, recharge your mobile phone and DTH & pay mobile & landline bills as paid services. (Organised retailers are charging for shopping bags after February 4, 2011, as per the notification of the Ministry of Environment & Forest, government of India). 201 6.42 Exchange / return policy Multi-format organised retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart) offer attractive exchange & return policy. Multi-format organised retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart) offer attractive exchange & return policy. HyperCITY is pleased to offer exchange within 14 days of your purchase on merchandise of the original cash memo. The merchandise should be unused, undamaged, in the original packaging and in saleable condition. Mobile phones, MP3 players, innerwear, swimwear, jewellery, watches, cosmetics and any altered merchandise will not be exchanged. All appliances & electronics items will be exchanged subject to the warranty & guarantee terms of the manufacturer, with a valid receipt, while furniture is not exchanged or refunded in any situation until unless the item has a manufacturing issue within the first year of purchase and if it is not repairable by us once the furniture has been assembled at the customer's address. They issue credit note/s valid for 1 month from the date of issue. All credit notes must be completely used in a single transaction. All cash transactions will be refunded in the form of a credit note. Credit card transactions will be refunded back to the credit card account or credit note as per customer convenience. Under no circumstances, they issue cash refunds. In the case of loss of credit note/s, no duplicate will be issued. Star Bazaar offer exchange within 15 days of your purchase only if sales receipt is presented along with merchandise. An exchange will be done only if merchandise is in its original packaging and in resalable condition. The refund will be done only through the exchange vouchers. No exchange on Saturday & Sunday. Fresh merchandise of bakery, fruits, vegetables, non-vegetarian, undergarments, sale items will not be exchanged. Further, they take utmost care to ensure that only the best quality merchandise are made available to their customers. However, if there is any quality defect in any Star Branded merchandise the same will be replaced with the merchandise of same value within 15 days from the date of purchase if sales receipt is presented with merchandise. more.MEGAStore offers hassle free return & exchange policy. In case, customers change their mind, more.MEGAStore would be pleased to exchange the 202 same within 14 days from the date of purchase - no questions are asked if the same is in original packaging and accompanied by its invoice. 6.42 Exchange / return policy In case, an exchange is not required, customers can get issued a gift card of the full value of the bill, which can be utilised within 15 days at the store as per customer convenience. Big Bazaar guaranteed the exchange of any merchandise that customers have bought from Big Bazaar & not satisfied. Return it along with a cash memo within 7 days of purchase. Merchandise like body care, cosmetics, crockery, jewellery, soft toys, undergarments, hair accessories, CDs, pharmaceuticals, toiletries etc. will not be exchanged. D Mart offers no exchange & return on frozen, dairy, deodorants and seasonal merchandise, no return of gift-wrapped merchandise, 2% deduction on the purchase through debit or credit card if returned, no return on undergarments, shoes, CDs, DVDs. Fashion-item organised retailers (i.e. Big Megamart, West Side and Max) offer exchange & return policy. Big Megamart offers to exchange any merchandise within 15 days after purchasing, and get a credit note of the same value, which will be valid for next 15 days in the same store, but non-returnable unless defective like undergarments, including socks, shoe care and foot care merchandise. Altered, used, damaged and promotional merchandise, personal electronics items like mobile phones, cameras, etc. and sunglasses, watches and writing instruments are also not returnable. Big Megamart adopts no cash refund policy. West side has the policy to satisfy their customers with the range, quality and value of the merchandise they offer. However, if customers are dissatisfied with the product that they have bought, they take required actions to assist them. West side expects that the customers should return unused merchandise within 30 days accompanied by a bill; they would exchange the returned items or give customers a complete refund. In the event that customers do not have the receipt, West side would offer them an exchange or offer a gift card to the selling price. West side has complete confidence in the quality of the products they do offer, however if customers have any complaint, they do address. They would exchange the returned items or give 203 customers a complete refund. In the event that customers do not have the receipt, West side would offer them an exchange or provide them a gift voucher for the current or last known selling price. No exchange or return on watches, jewellery, earrings, sunglasses, toys, cosmetics, disc, and lingerie and undergarments. 6.42 Exchange / return policy Max values their customers and guarantees the merchandise they trade. If a customer wants to change the merchandise bought or find it low in expectation, within 15 days of purchase, their customer relation associates will be happy to assist customers in finding a replacement. However, company policy expects the presentation of the original receipt for settlement. A credit voucher will be issued in value equivalent to a replacement for use in MAX stores in the whole country within 3 months of the date of issue. The above policy excludes the following: altered garments, used merchandise, undergarments including socks, discounted merchandise. Superstore based electronics-item organised retailers (i.e. Reliance Digital and NEXT) offer a return & exchange policy. Reliance Digital offers that if customers are not satisfied with any of the merchandise, they will be happy to accept them for return and issue a credit note subject to the following: request for return should be made within seven days of purchase. The merchandise to be returned should be in saleable condition. Invoice copy and delivery Challan in the original will be required to be submitted at the time of the request. The Credit Note issued can be used for the purchase of any other merchandise within the store within the stipulated validity period. No return of mobile phones and computers. Music, gaming & computer software will be taken back only if its packaging/seal is not broken. No return of any merchandise purchased under a promotional scheme or discount or where invoice specifically mentions "no replacements". NEXT offers its customers a 30 days change of mind policy. If for any reason, a customer wishes to return merchandise, the customer may do so within 30 days from the date of purchase, and may opt for a refund or store credit. Please note that the merchandise should be in factory-sealed condition. All merchandise sold at NEXT are covered under our 30-day replacement guarantee policy. Notify them of any manufacturing defect in the merchandise within 30 days from the date of delivery, and they will issue a replacement to you at no extra cost. In order to return merchandise, 204 fill the returns form that comes with the merchandise. Return will only be accepted if: item(s) are received with their original packaging, along with the original price tags, labels, and invoices. 6.42 Exchange / return policy All free gifts, warranty card, booklets, accessory packaging materials, manuals, like remote controls, cords, cables, etc., and all other items originally included with the merchandise(s) delivered are returned with the merchandise. Returned items are in resalable conditions. 6.43 Use of Gap Model All the organised retailers (included in the study) use gap model to understand knowledge gap, standard gap, delivery gap and communication gap in order to improve customer service quality except Star Bazaar, NEXT and D Mart. 6.44 Customer relationship management (CRM) All the organised retailers (included in the study) have much inclination towards CRM except D Mart as it has partial CRM. HyperCITY has separate discovery club desk / cell dedicated to the customer relationship. Star Bazaar has also a separate cell dedicated to the customer relationship. more.MEGAStore is keen to maintain customer relationship through its dedicated customer support cell while Big Bazaar, through its dedicated customer service cell and loyalty programme cell. Fashion-item organised retailers i.e. Big Megamart maintains customer relations through its dedicated CRM cell; West Side through its clubWest cell while Max through its separate customer service cell and customer relation associates. Superstore based electronics-item organised retailers i.e. Reliance Digital takes care of CRM activities through my assist – a customer service desk while NEXT through the customer care desk. 6.45 Database for CRM All the multi-format hyper store organised retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore and Big Bazaar) collect information for database regarding customer transactions (purchase date, price paid, merchandise purchased, responses to promoted merchandise), customer contacts (record of customer interaction with store by websites, inquiries in-store kiosks, telephone calls), customer preferences (favourite colour, brands, fabrics, flavours, and apparel size), descriptive information (demographic & psychographic data of customers) and customer responses to 205 marketing activity (responsiveness to sales promotion & advertising). While, multi- format superstore organised retailer, i.e. D Mart collects information for database regarding customer transactions (purchase date, the price paid, merchandise purchased, and responses to promoted merchandise) only. Fashion-item organised retailers (i.e. Big Megamart, West Side and Max) collect information for database regarding customer transactions (purchase date, price paid, merchandise purchased, responses to promoted merchandise), customer contacts (record of customer interaction with the store by websites, inquiries in-store kiosks, telephone calls), customer preferences (favourite colour, brands, fabrics, flavours, and apparel size) and descriptive information (demographic & psychographic data of customers). However, West Side also collects the information regarding customer responses to marketing activity (responsiveness to sales promotion & advertising). Electronics-item organised retailers (i.e. Reliance Digital and NEXT) collect information for database regarding customer transactions (purchase date, price paid, merchandise purchased, responses to promoted merchandise), customer contacts (record of customer interaction with the store by websites, inquiries in-store kiosks, telephone calls) and descriptive information (demographic & psychographic data of customers). 6.46 Methods of information collection for database All the organised retailers collect the information for the database by keeping records of the customer transactions. All these organised retailers except D Mart also conduct the in-store survey through printed questionnaire and collect information under membership programme. Moreover, some organised retailers (i.e. more.MEGAStore, Big Bazaar, Reliance Digital and NEXT) conduct an online survey in order to collect information; while Max and Big Bazaar also collect the information by taking customer feedback & suggestions. Big Bazaar also conducts a phone survey while Big Magamart does cross-selling in order to collect information for a database as a part of CRM. 6.47 CRM programmes 206 All the organised retailers use following CRM programmes in order to maintain a relationship and create loyal customers. 6.49 Key customers All the organised retailers included in the study deals with retail clientele only except, Reliance Digital, as it deals with both retail as well as corporate clientele. 6.48 Separate cell to help customers in the store All the organised retailers have a separate cell (customer support cell) to help customers in store and they all use listening and responding approach, respond to the customers with a courtesy, maintain speed and accuracy of service in order to help their customers (c) Programmes to deal with unprofitable customers No organised retailers have programmes to deal with unprofitable customers. (b) Programmes to convert satisfied customers in high lifetime value customers (b) Programmes to convert satisfied customers in high lifetime value customers All the organised retailers (included in the study) do cross-selling and add on selling in order to convert satisfied customers in high lifetime value customers except, D Mart as it does not take efforts to convert satisfied customers in high lifetime value customers. (a) Programmes for customer retention All the organised retailers (included in the study) introduce membership as a part of its frequent shopper program that will help them to personalise their customers and they provide special customer services (free & paid) to retain their customers except, D Mart as it does not pay attention to retaining their customers and do not have any customer loyalty programs. (b) Programmes to convert satisfied customers in high lifetime value customers All the organised retailers (included in the study) do cross-selling and add on selling in order to convert satisfied customers in high lifetime value customers except, D Mart as it does not take efforts to convert satisfied customers in high lifetime value customers. 6.51 Analysis of the profitability of customer retailing relationship All the organised retailers included in the study regularly analyse the profitability of the customer retailing relationship at the account level through department analysis of their customer relationship and through periodic review by a professional cost accountant except D Mart and NEXT; as they do analyse the profitability of the customer retailing relationship at the account level only through periodic review by a professional cost accountant. 6.50 Challenges while dealing with key customers Changing the landscape of Indian consumerism, changing retail trends, changing government role, stiff competition, cultural disparity, high demand for service standards, price war are some major challenges faced by all the organised 207 retailers while dealing with key customers. D Mart also faces the problems due to no membership, no special offers for members, no customised offers and partial CRM in order to deal with key customers. 6.53 Layout while planning store design All the multi-format hyper store organised retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore and Big Bazaar) use a mix of free form or boutique layout & grid layout to display the merchandise while planning store design while multi- format superstore organised retailer i.e. D Mart uses only grid layout to display the merchandise while planning store design. Fashion-item organised retailers (i.e. Big Megamart, West Side and Max) use the free form or boutique layout to display the merchandise while planning store design. Superstore based electronics-item organised retailers (i.e. Reliance Digital and NEXT) use racetrack layout to display the merchandise while planning store design. 6.52 e-retailing No organised retailers included in the study provide e-retailing service to their customers except, Big Bazaar, Reliance Digital and NEXT. Big Bazaar provides e- retailing service to their customers through the group website, i.e. www.futurebazaar.com; Reliance Digital through www.reliancedigitla.in while NEXT through www.next.co.in. 6.56 Fixtures used to display merchandise All the multi-format organised retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart) use straight racks, rounder (capacity fixture), four-way fixture (feature fixture) and gondolas to display the merchandise in stores. Fashion-item organised retailers (i.e. Big Megamart, West Side and Max) use rounder (bulk fixture/capacity fixture), four-way fixture (feature fixture) and gondolas to display the merchandise in stores. Superstore based electronics-item organised retailers (i.e. Reliance Digital and NEXT) use only straight racks and rounder (bulk fixture/capacity fixture) to display the merchandise in the stores. 6.55 Methods to display the merchandise in their stores Freestanding displays, end caps, promotional aisles, wall displays and cash wrap displays (point of purchase counters or check out areas) are most preferred methods of displaying the merchandise in their stores among all the multi-format organised retailers included in the study. Moreover, some multi-format organised retailers (i.e. HyperCITY, more.MEGAStore and Big Bazaar) also use window display and entrance display methods of displaying the merchandise in their stores. Fashion-item organised retailers (i.e. Big Megamart, West Side and Max) use window display, entrance display, free standing display, end caps, promotional aisles and wall display methods of displaying the merchandise in their stores. Moreover, Max also uses a dressing room display method to display the merchandise. Superstore based electronics-item organised retailers (i.e. Reliance Digital and NEXT) use entrance display, freestanding display, promotional aisles and walls display methods of displaying the merchandise in their stores. Moreover, Reliance Digital also uses a cash wrap display method to display the merchandise. 6.54 Signage to locate the merchandise category All the organised retailers included in the study use category, promotional and point of sale signage in order to locate the merchandise category in their stores. 208 Moreover, some organised retailers (i.e. Big Megamart, more.MEGAStore, Big Bazaar and Reliance Digital) also use digital signage in order to locate the merchandise category in their stores. 6.58 Variables used to create an appealing store atmosphere Some multi-format organised retailers (i.e. HyperCITY, more.MEGAStore and Big Bazaar) use lighting, colour, scent and music in order to create an appealing store atmosphere. While, Star Bazaar uses only lighting and music to create an appealing store atmosphere. Moreover, D Mart uses only music to create an appealing store atmosphere. Fashion-item organised retailers (i.e. Big Megamart, West Side and Max) use lighting, colour, scent and music for creating an appealing store atmosphere. Electronics-item organised retailers (i.e. Reliance Digital and NEXT) use lighting, colour, and music to create an appealing store atmosphere. 6.57 Presentation techniques to present the merchandise 209 All the organised retailers included in the study use item & size presentation, price lining, vertical merchandising, tonnage merchandising and frontal presentation techniques to present the merchandise in the store. Moreover, some organised retailers (i.e. Star Bazaar, Big Bazaar, West Side, Max, Reliance Digital and NEXT) also use an idea-oriented presentation technique. 6.59 Store design Visual merchandising is an extremely important component of store design for all the organised retailers included in the study. Store layout is also an extremely important component of store design for all the organised retailers except Star Bazaar and NEXT as it is important to them. Moreover, interior atmospherics is an extremely important component of store design for all the organised retailers except Star Bazaar as it is important. Exterior atmospherics is also an extremely important component of store design for some organised retailers i.e. HyperCITY, Big Bazaar, West Side, Max and Reliance Digital while for others, i.e. Star Bazaar, more.MEGAStore, D Mart, Big Megamart and NEXT, it is important. Table 6.7: Components of store design S. N. Measurement Variables Importance Scale* K-S D value calculated 4 3 2 1 1. Exterior atmospherics 05 05 00 00 0.500 1 Store front 04 06 00 00 0.500 2 Store marquee/sign board 07 03 00 00 0.500 3 Frontage entrance 03 06 01 00 0.400 4 External display window/space 03 03 04 00 0.250 5 Building 04 05 01 00 0.400 210 6 Store architecture 05 04 00 01 0.400 7 Uniqueness of store 04 03 02 01 0.200 8 Visibility of store 09 01 00 00 0.650 9 Location/site of the store 07 03 00 00 0.500 10 Parking facilities 06 04 00 00 0.500 11 Exterior look of store 03 06 00 01 0.400 12 Ease of access 03 07 00 00 0.500 13 Store name plate 04 05 01 00 0.400 14 Health & safety standards 06 03 01 00 0.400 15 Surrounding area 01 03 05 01 0.150 2. Interior atmospherics 09 01 00 00 0.650 1 Flooring & ceiling 05 04 01 00 0.400 2 Lighting 09 01 00 00 0.650 3 Odour/smell/scent 05 05 00 00 0.500 4 Fixtures 02 06 02 00 0.300 5 Wall textures 01 07 02 00 0.300 6 Sounds 08 02 00 00 0.550 7 Colour 02 06 02 00 0.300 8 Temperature 08 02 00 00 0.550 9 Cleanliness 07 03 00 00 0.500 10 Aisles 03 05 02 00 0.300 11 Trial (experience) room/zone 09 00 01 00 0.650 12 Dead area 00 05 04 01 0.150 13 Merchandise 10 00 00 00 0.750 14 Graphics 01 06 02 01 0.200 15 Store size 09 01 00 00 0.650 3. 6.59 Store design Store layout 08 02 00 00 0.550 1 Space planning 09 01 00 00 0.650 2 Floor space allocation 06 04 00 00 0.500 3 E-display 00 01 09 00 0.250 4 Department location 06 03 01 00 0.400 5 Department size 02 08 00 00 0.500 6 Department compatibility 01 06 03 00 0.250 7 Location of various merchandise 07 03 00 00 0.500 8 Accessibility to cash counters 07 02 01 00 0.450 9 Signage 08 02 00 00 0.550 10 Space/merchandise category 07 03 00 00 0.500 4. Visual merchandising 10 00 00 00 0.750 1 Assortment 10 00 00 00 0.750 2 Theme 01 08 01 00 0.400 3 Racks & shelves 05 04 01 00 0.400 4 Ensemble 01 07 02 00 0.300 5 Payment counters 09 00 01 00 0.650 6 Merchandise display 07 03 00 00 0.500 7 The planogram 06 04 00 00 0.500 8 Props & mannequins 01 08 01 00 0.400 9 Point of sale sign 08 02 00 00 0.500 10 Methods of display 08 02 00 00 0.500 211 * 4- extremely important, 3- important, 2- partially important, 1- not important Kolmogorov-Smirnov (K-S) D value tabulated = 0.410; Level of significance (α) = 5 % (0.05) * 4- extremely important, 3- important, 2- partially important, 1- not important Kolmogorov-Smirnov (K-S) D value tabulated = 0.410; Level of significance (α) = 5 % (0.05) Especially, HyperCITY has collaborated with JHP of London for in-store design. JHP is an excellent retail designing service provider with an experience in large format stores around the globe. HyperCITY deploys a good technology enabled infrastructure from JDA that is a leading global retail technology service provider firm. The JDA suite comprises the E3 Advanced Replenishment, Merchandise Management System, WinDSS point of sale and Intactix space planning modules. The system ensures that customers do not have to wait in long queues for billing, or transaction processing for enhancing the overall customer experience at HyperCITY. Johnson Diversey is a world class & trusted the source of cleaning and hygiene merchandise and services. Since food sanitation, store hygiene, safety, etc., are important in any store, Johnson Diversey assists in improving the customer shopping experience. 6.60 Future-plans HyperCITY wants to provide the best merchandise from around the world, intuitive store design & adherence to international standards of service to make HyperCITY an unmatched experience. Star Bazaar looks forward to having the customer over at their stores to experience first-hand their promise and commitment to be the best for their customers at all times. more.MEGAStore’s wants to consistently provide the Indian consumer a complete and differentiated shopping environment and a top retailer of the India by offering superior returns to all stakeholders. Big Bazaar wants to become most admired value retailers in all the sense of retaliating for all the stakeholders. D Mart wants to offer more variety of the merchandise with a lesser price to their customers. Big Megamart wants to offer mega (international) brands at surprisingly low prices and offers a retail experience of a high-end. West Side hopes to expand rapidly with similar format stores that offer a fine balance between style & price retailing. Max hopes to be a market leader in the field of value retailing; wants to provide fashionable merchandise at affordable prices and wants to be innovative, cost effective and globally competitive in order to match with exceeding customer’s expectations. The Reliance Digital hopes to become 212 sustainable value creator for all its stakeholders. The NEXT plans to build a chain of over 1000 showrooms throughout the country with an esteemed annual turnover of ₹ 500 crores in the next 3 years. sustainable value creator for all its stakeholders. The NEXT plans to build a chain of over 1000 showrooms throughout the country with an esteemed annual turnover of ₹ 500 crores in the next 3 years. sustainable value creator for all its stakeholders. The NEXT plans to build a chain of over 1000 showrooms throughout the country with an esteemed annual turnover of ₹ 500 crores in the next 3 years. 213  Analysis of profile of customers in the organised retail industry  Factors influencing buying behaviour of customers in the organised retail industry Analysis of profile of customers in the organised retail industry  Analysis of buying behaviour of customers in the organised retail industry  Analysis of profile of customers in the organised retail industry  Analysis of buying behaviour of customers in the organised retail industry F i fl i b i b h i f i h i d il Chapte r 7  Factors influencing buying behaviour of customers in the organised retail industry Introduction This chapter includes an analysis of the various data pertained to customers of organised retail industry, which have been collected through the individual questionnaire from the customers of the respective organised retailers / stores in a personal interaction. This chapter encompasses the analysis of profiles of customers, changing buying behaviour and factors affecting the buying behaviour. Further, results have been interpreted and discussed to reach on some inferences at the end of the different sections. This chapter has been divided into three sub-sections as follows: 7.1 Analysis of profile of customers in the organised retail industry 7.2 Analysis of buying behaviour of customers in the organised retail industry 7.2 Analysis of buying behaviour of customers in the organised retail industry 7.3 Factors influencing buying behaviour of customers in the organised retail industry 7.3 Factors influencing buying behaviour of customers in the organised retail industry 214 ANALYSIS OF PROFILE OF CUSTOMERS IN THE ORGANISED RETAIL INDUSTRY 7.1.1 Residing area of customers Table 7.1.1: Residing area of customers Retailers Residing Areas Urban Semi-urban Rural f % f % f % HyperCITY 125 62.50 71 35.50 4 2.00 Star Bazaar 98 49.00 80 40.00 22 11.00 more.MEGAStore 92 46.00 78 39.00 30 15.00 Big Bazaar 124 62.00 68 34.00 8 4.00 D Mart 88 44.00 65 32.50 47 23.50 Big Megamart 137 68.50 62 31.00 1 0.50 West Side 107 53.50 81 40.50 12 6.00 Max 123 61.50 62 31.00 15 7.50 Reliance Digital 89 44.50 78 39.00 33 16.50 NEXT 75 37.50 68 34.00 57 28.50 Mean 105.80 52.90 71.30 35.65 22.90 11.45 Standard Deviation 20.466 10.233 7.409 3.705 18.669 9.335 7.1.2 Gender of customers 7.1.2 Gender of customers Organised retail industry taps the male and female customers in equal proportion; however, the footfall of male customers is more in comparison to female customers. But, few retailers (i.e. HyperCITY- a multi-format hyper store; Big Megamart- a fashion-item hyper store, West Side & Max - fashion-item superstores) have more footfalls of female customers in comparison to male customers. All the fashion-item retailers are getting more footfalls from female customers; while, the electronics-item superstores (i.e. NEXT and Reliance Digital) are getting more footfalls from male customers. The multi-format superstore (i.e. D Mart) has an equal footfall from both male and female customers. ANALYSIS OF PROFILE OF CUSTOMERS IN THE ORGANISED RETAIL INDUSTRY ANALYSIS OF PROFILE OF CUSTOMERS IN THE ORGANISED RETAIL INDUSTRY The organised retail industry has its majority of customers from urban areas and semi-urban areas. The organised retailers (included in the study) have the customers from all the areas i.e. urban, semi-urban and rural. But, the first majority of customers of all the retailers is from urban areas and the second majority of customers are from semi-urban areas, while some of the retailers are also able to tap an enough number of customers from rural areas, e.g. NEXT and Reliance Digital (electronics- item superstores), D Mart, Star Bazaar, more.MEGAStore (multi-format organised retailers). However, there are also few organised retailers included in the study have a poor footfall of the customers from rural areas i.e. HyperCITY and Big Bazaar (multi- format organised retailers), Big Megamart (a fashion-item hyper store), West Side and Max (fashion-item superstores). The multi-format hyper stores organised retailers 215 (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar) have a poor footfall from rural areas in comparison to multi-format super-stores organised retailer (i.e. D Mart). 7.1.2 Gender of customers Table 7.1.2: Gender of customers Retailers Genders Male Female f % f % HyperCITY 91 45.50 109 54.50 Star Bazaar 111 55.50 89 44.50 more.MEGAStore 106 53.00 94 47.00 Big Bazaar 100 50.00 100 50.00 D Mart 130 65.00 70 35.00 Big Megamart 76 38.00 124 62.00 West Side 85 42.50 115 57.50 Max 79 39.50 121 60.50 Reliance Digital 162 81.00 38 19.00 NEXT 165 82.50 35 17.50 Mean 110.50 55.25 89.50 44.75 Standard Deviation 32.226 16.113 32.226 16.113 7.1.3 Age of customers The organised retail industry is flourishing with the footfall of youth customers in the stores. Its majority of customers’ age ranges from 21 to 40 years. 216 Table 7.1.3: Age of customers Table 7.1.3: Age of customers Retailers Age (in years) Below 21 21-30 31-40 41-50 51-60 Above 60 f % f % f % f % f % f % HyperCITY 16 8.00 58 29.00 71 35.50 38 19.00 10 5.00 7 3.50 Star Bazaar 12 6.00 47 23.50 68 34.00 47 23.50 16 8.00 10 5.00 more.MEGAStore 13 6.50 43 21.50 67 33.50 46 23.00 19 9.50 12 6.00 Big Bazaar 19 9.50 50 25.00 62 31.00 40 20.00 21 10.50 8 4.00 D Mart 8 4.00 36 18.00 72 36.00 49 24.50 28 14.00 7 3.50 Big Megamart 30 15.00 66 33.00 58 29.00 30 15.00 14 7.00 2 1.00 West Side 22 11.00 67 33.50 62 31.00 31 15.50 12 6.00 6 3.00 Max 18 9.00 62 31.00 68 34.00 32 16.00 15 7.50 5 2.50 Reliance Digital 6 3.00 35 17.50 70 35.00 55 27.50 20 10.00 14 7.00 NEXT 4 2.00 24 12.00 72 36.00 60 30.00 25 12.50 15 7.50 Mean 14.80 7.40 48.80 24.40 67.00 33.50 42.80 21.40 18.00 9.00 8.60 4.30 Standard Deviation 7.913 3.957 14.505 7.253 4.807 2.404 10.337 5.168 5.696 2.848 4.115 2.058 However, the industry is getting the footfall from the customers of all the age groups. The industry is able to tap a greater number of customers from the age group of 41 to 50 years. Some organised retailers (i.e. Big Bazaar, Big Megamart, West Side and Max) included in the study are able to tap the customers from the age group of below 21 years in comparison to others; while a few others (i.e. more.MEGAStore, Big Bazaar, D Mart, Reliance Digital, NEXT) are good enough to attract the footfalls from the age group of 51+ customers in comparison to others. The electronics-item superstores (i.e. NEXT and Reliance Digital) struggling the tap the customers from the age group of below 21 years; while fashion store organised retailers (i.e. Big Megamart, West Side, Max) are struggling the tap the customers from the age group of above 60 years. 7.1.3 Age of customers 7.1.4 Education of customers 217 Table 7.1.4: Education of customers Table 7.1.4: Education of customers Retailers Education Illiterate Below HSC HSC - SSC Graduate Postgraduate Others f % f % f % f % f % f % HyperCITY 0 0.00 14 7.00 28 14.00 67 33.50 86 43.00 5 2.50 Star Bazaar 5 2.50 18 9.00 36 18.00 58 29.00 75 37.50 8 4.00 more.MEGAStore 3 1.50 15 7.50 37 18.50 64 32.00 72 36.00 9 4.50 Big Bazaar 0 0.00 16 8.00 40 20.00 60 30.00 80 40.00 4 2.00 D Mart 10 5.00 21 10.50 68 34.00 71 35.50 28 14.00 2 1.00 Big Megamart 0 0.00 10 5.00 15 7.50 70 35.00 98 49.00 7 3.50 West Side 0 0.00 12 6.00 16 8.00 72 36.00 95 47.50 5 2.50 Max 0 0.00 9 4.50 18 9.00 76 38.00 91 45.50 6 3.00 Reliance Digital 4 2.00 18 9.00 23 11.50 66 33.00 86 43.00 3 1.50 NEXT 6 3.00 17 8.50 39 19.50 57 28.50 75 37.50 6 3.00 Mean 2.80 1.40 15.00 7.50 32.00 16.00 66.10 33.05 78.60 39.30 5.50 2.75 Standard Deviation 3.458 1.729 3.801 1.900 15.944 7.972 6.350 3.175 19.834 9.917 2.173 1.087 The organised retail industry has well-educated customers in the Maharashtra state as the state has 82.34 literacy rate (according to the Census of 2011). Its majority of customers in the industry is graduates or post graduates. However, the industry is getting the footfall from the customers of all education groups. Few organised retailers (Star Bazaar, more.MEGAStore, D Mart, Reliance Digital and NEXT) are also to attract illiterate customers in few numbers; while others (i.e. HyperCITY, Big Bazaar, Big Megamart, West Side, Max) are not having the illiterate customers. 7.1.3 Age of customers 218 7.1.5 Occupation of customers Table 7.1.5: Occupation of customers Retailers Occupations Salaried Self-employed Homemaker Student Retired f % f % f % f % f % HyperCITY 78 39.00 39 19.50 25 12.50 46 23.00 12 6.00 Star Bazaar 68 34.00 32 16.00 52 26.00 31 15.50 17 8.50 more.MEGAStore 70 35.00 28 14.00 42 21.00 50 25.00 10 5.00 Big Bazaar 81 40.50 34 17.00 36 18.00 38 19.00 11 5.50 D Mart 58 29.00 30 15.00 59 29.50 22 11.00 31 15.50 Big Megamart 79 39.50 46 23.00 20 10.00 45 22.50 10 5.00 West Side 80 40.00 42 21.00 16 8.00 50 25.00 12 6.00 Max 76 38.00 39 19.50 20 10.00 51 25.50 14 7.00 Reliance Digital 74 37.00 50 25.00 30 15.00 18 9.00 28 14.00 NEXT 73 36.50 49 24.50 29 14.50 20 10.00 29 14.50 Mean 73.70 36.85 38.90 19.45 32.90 16.45 37.10 18.55 17.40 8.70 Standard Deviation 6.977 3.488 7.852 3.926 14.310 7.155 13.295 6.648 8.514 4.257 The organised retail industry is getting the customers of all the occupational groups. Its majority of customers are salaried and self-employed. The second majority of customers are students and homemakers. D Mart is only retailer getting less footfalls of salaried customers in comparison to others and the industry mean value; while more.MEGAStore is getting less footfall of self-employed customers in comparison to others. The fashion store organised retailers (i.e. Big Megamart, West Side, Max) are getting less footfall from homemakers in comparison to the industry mean value; while, the electronics-item superstores (i.e. NEXT and Reliance Digital) are getting less footfall of student customers in comparison to the industry mean value. Some retailers (i.e. HyperCITY, Big Bazaar, Big Megamart, West Side, Max and Reliance Digital) have more salaried customers in comparison to the industry mean value. Reliance Digital and Big Magamart are able to tap more self-employed customers in comparison to the industry mean value. All the multi-format organised retailers included in the study (except HyperCITY) have more homemaker customers in comparison to the industry mean value. HyperCITY, more.MEGAStore, Big Bazaar, Big Megamart, West Side, and Max are getting more footfalls of student customers in comparison to others and the industry mean value. The electronics-item superstores (i.e. NEXT and Reliance Digital) and multi-format superstore (i.e. D 219 Mart) are getting more footfalls for retired customers in comparison to others and the industry mean value. 7.1.3 Age of customers 7.1.6 Family income (yearly) of customers Table 7.1.6: Family income (yearly) of customers Retailers Family income (in Lacs of ) ₹ < 1 1 to 2 2 to 5 5 to 10 > 10 f % f % f % f % f % HyperCITY 1 0.50 15 7.50 68 34.00 78 39.00 38 19.00 Star Bazaar 12 6.00 22 11.00 78 39.00 62 31.00 26 13.00 more.MEGAStore 15 7.50 25 12.50 85 42.50 56 28.00 19 9.50 Big Bazaar 3 1.50 19 9.50 62 31.00 81 40.50 35 17.50 D Mart 22 11.00 45 22.50 91 45.50 29 14.50 13 6.50 Big Megamart 0 0.00 12 6.00 65 32.50 90 45.00 33 16.50 West Side 2 1.00 14 7.00 70 35.00 87 43.50 27 13.50 Max 3 1.50 17 8.50 64 32.00 88 44.00 28 14.00 Reliance Digital 1 0.50 24 12.00 91 45.50 60 30.00 24 12.00 NEXT 2 1.00 29 14.50 90 45.00 59 29.50 20 10.00 Mean 6.10 3.05 22.20 11.10 76.40 38.20 69.00 34.50 26.30 13.15 Standard Deviation 7.520 3.760 9.647 4.824 11.974 5.987 19.293 9.647 7.718 3.859 7.1.6 Family income (yearly) of customers The organised retail industry is getting the customers of all the income groups. However, its majority of customers are from two income groups, i.e. 2 to 5 LPA and 5 to 10 LPA family income. But, the industry is also attracting and getting the footfall of customers from the high-income group (i.e. more than 10 LPA) as well as low- income group (i.e. 1 to 2 LPA). Moreover, the industry has also customers from the lowest income group, i.e. less than 1 LPA. But, few organised retailers included in this study either do not have customers from less than 1 LPA family income group or having negligible customers from this income group, those are Big Megamart, HyperCITY, Big Bazaar, West Side, Max, Reliance Digital and NEXT. Star Bazaar, more.MEGAStore, D Mart, Reliance Digital and NEXT have more customers from 1 to 2 LPA family income group in comparison to others (included in the study) and the industry mean value. more.MEGAStore and D Mart are getting fewer customers from the 5+ 10 LPA family income group in comparison to others. 7.1.3 Age of customers 7 1 7 M i l f 220 Table 7.1.7: Marital status of customers Table 7.1.7: Marital status of customers Retailers Marital status Single Married f % f % HyperCITY 74 37.00 126 63.00 Star Bazaar 62 31.00 138 69.00 more.MEGAStore 65 32.50 135 67.50 Big Bazaar 72 36.00 128 64.00 D Mart 48 24.00 152 76.00 Big Megamart 88 44.00 112 56.00 West Side 79 39.50 121 60.50 Max 77 38.50 123 61.50 Reliance Digital 52 26.00 148 74.00 NEXT 35 17.50 165 82.50 Mean 65.20 32.60 134.80 67.40 Standard Deviation 16.212 8.106 16.212 8.106 The organised retail industry is getting married as well as single customers. However, the industry has its more customers from the married marital status group in comparison to a single marital status group of customers. The electronics-item superstores (i.e. NEXT and Reliance Digital) and D Mart are getting fewer customers from single marital status group of customers in comparison to others and the industry mean value; while the fashion-item retailers (i.e. Big Megamart, West Side, Max) as well as two multi-format hyper stores (i.e. HyperCITY and Big Bazaar) are tapping more customers from single marital status group of customers in comparison to others and the industry mean value. But, the electronics-item superstores (i.e. NEXT and Reliance Digital) and D Mart are tapping more customers from a married marital status group of customers in comparison to others and the industry mean value. The fashion-item hyper store organised retailer (i.e. Big Megamart) is getting fewer customers from a married marital status group of customers as compared to others and the industry mean value. 7.1.8 Nature of family of customers 221 Table 7.1.8: Nature of family of customers Retailers Nature of family Nuclear Joint f % f % HyperCITY 136 68.00 64 32.00 Star Bazaar 103 51.50 97 48.50 more.MEGAStore 112 56.00 88 44.00 Big Bazaar 129 64.50 71 35.50 D Mart 98 49.00 102 51.00 Big Megamart 146 73.00 54 27.00 West Side 123 61.50 77 38.50 Max 140 70.00 60 30.00 Reliance Digital 105 52.50 95 47.50 NEXT 103 51.50 97 48.50 Mean 119.50 59.75 80.50 40.25 Standard Deviation 17.545 8.773 17.545 8.773 The organised retail industry is getting both nuclear as well as joint families customers. However, the industry is tapping more customers from nuclear families in comparison to joint families. Some retailers included in this study (i.e. 7.1.3 Age of customers Star Bazaar, more.MEGAStore, D Mart, Reliance Digital and NEXT) are getting fewer footfalls from nuclear families in comparison to joint families and industry mean value; few others (i.e. HyperCITY, Big Bazaar, Big Megamart, West Side and Max) are getting fewer footfalls from joint families in comparison to nuclear families and industry mean value. The multi-format superstore included in this study (i.e. D Mart) is having highest customers from joint families in comparison to nuclear families and industry mean value; while the fashion-item hyper store (i.e. Big Megamart) is tapping highest customers from nuclear families in comparison to nuclear families and industry mean value. 7.1.9 Family size of customers 7.1.9 Family size of customers Table 7.1.9: Family size of customers 222 Retailers Family size (Number of Members) < 4 4 to 5 6 to 8 > 8 f % f % f % f % HyperCITY 96 48.00 68 34.00 32 16.00 4 2.00 Star Bazaar 80 40.00 68 34.00 42 21.00 10 5.00 more.MEGAStore 76 38.00 70 35.00 40 20.00 14 7.00 Big Bazaar 92 46.00 65 32.50 35 17.50 8 4.00 D Mart 61 30.50 72 36.00 45 22.50 22 11.00 Big Megamart 102 51.00 71 35.50 25 12.50 2 1.00 West Side 93 46.50 79 39.50 24 12.00 4 2.00 Max 98 49.00 75 37.50 20 10.00 7 3.50 Reliance Digital 72 36.00 80 40.00 36 18.00 12 6.00 NEXT 69 34.50 79 39.50 39 19.50 13 6.50 Mean 83.90 41.95 72.70 36.35 33.80 16.90 9.60 4.80 Standard Deviation 14.091 7.045 5.293 2.646 8.377 4.189 5.967 2.983 Family size (Number of Members) The organised retail industry is tapping the customers from all the family size groups of customers included in the study. But, the first majority of customers are from the family size of below 4 members and the second majority of customers are from 4 to 5 members’ family size. However, the industry is also tapping enough footfalls from 6 to 8 members’ family size. The fashion-item retailers (i.e. Big Megamart, West Side, Max) as well as two multi-format hyper stores (i.e. HyperCITY and Big Bazaar) are tapping more customers from the family size of below 4 members in comparison to others and industry mean value. The electronics- item superstores (i.e. NEXT and Reliance Digital) are getting more customers from the family size of 4 to 5 members in comparison to other retailers included in the study. 7.1.3 Age of customers The multi-format superstore included in this study (i.e. D Mart) is having highest customers from the family size of more than 5 members in comparison others and industry mean value; while the fashion-item retailers (i.e. Big Megamart, West Side and Max) are getting minimum customers from the family size of more than 5 members in comparison to others and industry mean value. 223 7.1.10 Religion of customers Table 7.1.10: Religion of customers g Retailers Religions Hindu Muslim Buddhists Christian Sikh Other f % f % f % f % f % f % HyperCITY 155 77.50 26 13.00 12 6.00 4 2.00 1 0.50 2 1.00 Star Bazaar 150 75.00 40 20.00 7 3.50 1 0.50 0 0.00 2 1.00 more.MEGAStore 161 80.50 18 9.00 15 7.50 0 0.00 1 0.50 5 2.50 Big Bazaar 160 80.00 12 6.00 17 8.50 2 1.00 1 0.50 8 4.00 D Mart 176 88.00 10 5.00 11 5.50 0 0.00 0 0.00 3 1.50 Big Megamart 136 68.00 38 19.00 15 7.50 6 3.00 1 0.50 4 2.00 West Side 135 67.50 48 24.00 10 5.00 4 2.00 0 0.00 3 1.50 Max 165 82.50 15 7.50 14 7.00 2 1.00 1 0.50 3 1.50 Reliance Digital 179 89.50 10 5.00 9 4.50 0 0.00 0 0.00 2 1.00 NEXT 178 89.00 10 5.00 9 4.50 0 0.00 0 0.00 3 1.50 Mean 159.50 79.75 22.70 11.35 11.90 5.95 1.90 0.95 0.50 0.25 3.50 1.75 Standard Deviation 15.953 7.977 14.392 7.196 3.247 1.624 2.132 1.066 0.527 0.264 1.841 0.920 The organised retail industry is getting the customers from all the religions. However, the majority of the customers are Hindus. The second majority is of Muslim customers. Not surprisingly, the third majority is of Buddhists customers, as the Maharashtra state has 73.4% of the total Buddhism followers in India. The retail industry is also tapping the customers from other religions, but they are very less in numbers. 224 ANALYSIS OF BUYING BEHAVIOUR OF CUSTOMERS IN THE ORGANISED RETAIL INDUSTRY 7.2.1 Purchase decision maker in the family 7.2.1: Purchase decision maker in the family of customers Retailers Purchase decision makers Self Spouse Children Parent Others f % f % f % f % f % HyperCITY 82 41.00 65 32.50 10 5.00 38 19.00 5 2.50 Star Bazaar 85 42.50 78 39.00 12 6.00 23 11.50 2 1.00 more.MEGAStore 77 38.50 70 35.00 8 4.00 41 20.50 4 2.00 Big Bazaar 82 41.00 78 39.00 7 3.50 30 15.00 3 1.50 D Mart 96 48.00 66 33.00 20 10.00 16 8.00 2 1.00 Big Megamart 91 45.50 61 30.50 8 4.00 37 18.50 3 1.50 West Side 87 43.50 66 33.00 9 4.50 36 18.00 2 1.00 Max 90 45.00 65 32.50 5 2.50 38 19.00 2 1.00 Reliance Digital 120 60.00 44 22.00 19 9.50 16 8.00 1 0.50 NEXT 118 59.00 40 20.00 21 10.50 18 9.00 3 1.50 Mean 92.80 46.40 63.30 31.65 11.90 5.95 29.30 14.65 2.70 1.35 Standard Deviation 14.808 7.404 12.553 6.276 5.896 2.948 10.078 5.039 1.160 0.580 7.2.1 Purchase decision maker in the family The majority of customers in the organised retail industry take the purchase making decision by self. The second majority of customers relies on their spouse; while the third majority relies on their parents to purchase making decision. Means, the majority of customers in the organised retail industry are either self-purchase decision maker in the family or rely on their spouse and parents. The electronics-item superstores (i.e. NEXT and Reliance Digital) are getting more footfalls from the self-purchase decision-making customers in comparison to other retailers included in this study. The primary reason of the same is that both the stores have more than 80% of customers from the male group of customers. 225 All the multi-format hyper stores included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and Big Bazaar) along with fashion-item retailers (i.e. Big Megamart, West Side and Max) are getting more footfalls of the customers, whose purchase decision makers are their parents in comparison to other retailers. The reason of this more footfalls is that these stores have a good number of student customers, whose purchase decision makers are actually their parents. ANALYSIS OF BUYING BEHAVIOUR OF CUSTOMERS IN THE ORGANISED RETAIL INDUSTRY 7.2.1(a): Collection of information, while purchase decision making by customers Retailers Collection of information for purchase decision-making Yes No f % f % HyperCITY 187 93.50 13 6.50 Star Bazaar 176 88.00 24 12.00 more.MEGAStore 179 89.50 21 10.50 Big Bazaar 184 92.00 16 8.00 D Mart 136 68.00 64 32.00 Big Megamart 190 95.00 10 5.00 West Side 188 94.00 12 6.00 Max 189 94.50 11 5.50 Reliance Digital 193 96.50 7 3.50 NEXT 192 96.00 8 4.00 Mean 181.40 90.70 18.60 9.30 Standard Deviation 16.854 8.427 16.854 8.427 7.2.1(a): Collection of information, while purchase decision making by customers The majority of customers in organised retail industry collects the information about merchandise and services before taking a purchase making decision. In comparison to other retailers included in this study, a large number of customers of a multi-format superstore (i.e. D Mart) are least bothered about the information before taking a purchase making decision. The preferred sources of customers in the organised retail industry for collecting the information before purchase decision making are (in order of most preferred to least preferred): advertising, internet, family members, neighbours / friends / relatives, package labels, in-store display, in-store information, salespersons, telemarketing and other sources. ANALYSIS OF BUYING BEHAVIOUR OF CUSTOMERS IN THE ORGANISED RETAIL INDUSTRY 226 7.2.2: Sources of information, while purchase decision making by customers Retailers Sources of information Salespersons Telemarketing Advertising In-store information In-store display Package labels Family members Neighbours / Friends / Relatives Internet (www) Others f % f % f % f % f % f % f % f % f % f % yperCITY 67 33.50 45 22.50 179 89.50 52 26.00 78 39.00 125 62.50 138 69.00 132 66.00 153 76.50 13 6.50 ar Bazaar 55 27.50 20 10.00 165 82.50 78 39.00 102 51.00 142 71.00 132 66.00 138 69.00 142 71.00 11 5.50 ore.MEGAStore 76 38.00 12 6.00 171 85.50 82 41.00 98 49.00 137 68.50 143 71.50 142 71.00 136 68.00 10 5.00 g Bazaar 72 36.00 13 6.50 175 87.50 90 45.00 89 44.50 150 75.00 146 73.00 139 69.50 150 75.00 15 7.50 Mart 88 44.00 4 2.00 119 59.50 76 38.00 65 32.50 105 52.50 123 61.50 130 65.00 98 49.00 9 4.50 g Megamart 96 48.00 28 14.00 186 93.00 102 51.00 95 47.50 152 76.00 150 75.00 138 69.00 160 80.00 16 8.00 West Side 87 43.50 15 7.50 182 91.00 99 49.50 86 43.00 155 77.50 156 78.00 143 71.50 156 78.00 12 6.00 ax 92 46.00 14 7.00 178 89.00 91 45.50 90 45.00 143 71.50 149 74.50 135 67.50 161 80.50 17 8.50 eliance Digital 104 52.00 23 11.50 180 90.00 103 51.50 98 49.00 89 44.50 167 83.50 169 84.50 178 89.00 13 6.50 EXT 100 50.00 12 6.00 177 88.50 101 50.50 89 44.50 74 37.00 157 78.50 157 78.50 169 84.50 8 4.00 Mean 83.70 41.85 18.60 9.30 171.20 85.60 87.40 43.70 89.00 44.50 127.20 63.60 146.10 73.05 142.30 71.15 150.30 75.15 12.40 6.20 tandard Deviation 15.727 7.864 11.413 5.707 19.240 9.620 15.974 7.987 10.924 5.462 28.401 14.201 12.845 6.422 11.963 5.982 22.046 11.023 2.989 1.494 7.2.2: Sources of information, while purchase decision making by customers Sources of information The customers of multi-format hyper store organised retailers rely on the advertising, internet sources, family members, neighbours / friends / relatives and package labels for collecting the information before the purchase decision making. Good news for the multi-format organised retailers is that there is a large number of their customers also prefer to collect the information from sources like in-store display, salespersons and in-store information. ANALYSIS OF BUYING BEHAVIOUR OF CUSTOMERS IN THE ORGANISED RETAIL INDUSTRY But, the customers of D Mart (a multi-format super store) rely on neighbours / friends / relatives, family members, package labels, advertising and salespersons. The great number of customers from HyperCITY also prefers to collect the information from telemarketing sources before the purchase decision making. The customers of the electronics-item superstores (i.e. NEXT and Reliance Digital) collect massive information from all the sources before the purchase decision making. 7.2.4 Sources of influence for a customer, while purchase decision-making ANALYSIS OF BUYING BEHAVIOUR OF CUSTOMERS IN THE ORGANISED RETAIL INDUSTRY 227 7.2.3 Sources of information from advertising, while purchase decision making 7.2.3: Sources of information from advertising, while purchase decision making by customers Retailers Sources of information from Advertising Television Newspapers Magazines Hoardings Pamphlets Radio Cinema Celebrity endorsement Websites Others f % f % f % f % f % f % f % f % f % f % HyperCITY 167 83.50 49 24.50 25 12.50 13 6.50 15 7.50 7 3.50 5 2.50 152 76.00 153 76.50 12 6.00 Star Bazaar 152 76.00 68 34.00 16 8.00 10 5.00 30 15.00 12 6.00 0 0.00 133 66.50 142 71.00 11 5.50 more.MEGAStore 148 74.00 75 37.50 28 14.00 19 9.50 36 18.00 14 7.00 2 1.00 142 71.00 136 68.00 13 6.50 Big Bazaar 161 80.50 51 25.50 19 9.50 26 13.00 40 20.00 18 9.00 1 0.50 150 75.00 150 75.00 16 8.00 D Mart 102 51.00 86 43.00 8 4.00 28 14.00 56 28.00 16 8.00 0 0.00 100 50.00 98 49.00 8 4.00 Big Megamart 175 87.50 52 26.00 35 17.50 11 5.50 16 8.00 11 5.50 2 1.00 161 80.50 160 80.00 14 7.00 West Side 169 84.50 66 33.00 25 12.50 19 9.50 35 17.50 13 6.50 0 0.00 152 76.00 156 78.00 9 4.50 Max 162 81.00 53 26.50 39 19.50 20 10.00 30 15.00 16 8.00 1 0.50 147 73.50 161 80.50 5 2.50 Reliance Digital 170 85.00 85 42.50 12 6.00 31 15.50 42 21.00 19 9.50 0 0.00 160 80.00 178 89.00 4 2.00 NEXT 166 83.00 90 45.00 10 5.00 39 19.50 48 24.00 20 10.00 0 0.00 157 78.50 169 84.50 2 1.00 Mean 157.20 78.60 67.50 33.75 21.70 10.85 21.60 10.80 34.80 17.40 14.60 7.30 1.10 0.55 145.40 72.70 150.30 75.15 9.40 4.70 Standard Deviation 21.054 10.527 15.911 7.956 10.520 5.260 9.359 4.680 12.891 6.445 4.006 2.003 1.595 0.798 18.038 9.019 22.046 11.023 4.624 2.312 The majority of customers in the organised retail industry, who collect the information from advertising, relies on the following sources (in order of most preferred to least preferred): television, websites, celebrity endorsement, newspapers, pamphlets, magazines, hoardings, radio, other sources and cinema. ANALYSIS OF BUYING BEHAVIOUR OF CUSTOMERS IN THE ORGANISED RETAIL INDUSTRY 7.2.4 Sources of influence for a customer, while purchase decision-making 228 7.2.4: Sources of influence for a customer, while purchase decision-making ailers Sources of influence Self Family members Neighbors / Friends / Colleagues Celebrity endorsement Salesperson in a store Direct-mail piece Websites TV commercials Infomercials Magazines ads TV talk shows Sales Promotion Others f % f % f % f % f % f % f % f % f % f % f % f % f % TY 79 39.50 135 67.50 129 64.50 147 73.50 60 30.00 23 11.50 142 71.00 152 76.00 159 79.50 20 10.00 79 39.50 143 71.50 11 5.50 aar 82 41.00 129 64.50 135 67.50 128 64.00 48 24.00 26 13.00 131 65.50 137 68.50 146 73.00 13 6.50 72 36.00 151 75.50 9 4.50 EGAStore 74 37.00 140 70.00 139 69.50 137 68.50 69 34.50 20 10.00 125 62.50 133 66.50 139 69.50 18 9.00 76 38.00 149 74.50 8 4.00 aar 78 39.00 143 71.50 136 68.00 145 72.50 65 32.50 29 14.50 139 69.50 146 73.00 148 74.00 13 6.50 69 34.50 143 71.50 6 3.00 t 63 31.50 120 60.00 127 63.50 95 47.50 81 40.50 9 4.50 87 43.50 87 43.50 103 51.50 5 2.50 39 19.50 168 84.00 9 4.50 amart 88 44.00 147 73.50 135 67.50 156 78.00 89 44.50 36 18.00 149 74.50 160 80.00 166 83.00 26 13.00 86 43.00 133 66.50 7 3.50 de 84 42.00 153 76.50 140 70.00 146 73.00 80 40.00 40 20.00 145 72.50 154 77.00 160 80.00 17 8.50 82 41.00 139 69.50 9 4.50 87 43.50 146 73.00 132 66.00 142 71.00 85 42.50 35 17.50 160 80.00 147 73.50 159 79.50 30 15.00 80 40.00 140 70.00 8 4.00 Digital 117 58.50 164 82.00 166 83.00 155 77.50 97 48.50 12 6.00 167 83.50 155 77.50 168 84.00 6 3.00 89 44.50 132 66.00 10 5.00 115 57.50 154 77.00 154 77.00 152 76.00 93 46.50 8 4.00 158 79.00 151 75.50 158 79.00 4 2.00 84 42.00 122 61.00 8 4.00 Mean 86.70 43.35 143.10 71.55 139.30 69.65 140.30 70.15 76.70 38.35 23.80 11.90 140.30 70.15 142.20 71.10 150.60 75.30 15.20 7.60 75.60 37.80 142.00 71.00 8.50 4.25 d Deviation 17.023 8.512 12.845 6.422 11.963 5.982 18.000 9.000 15.727 7.864 11.507 5.753 22.799 11.399 21.054 10.527 19.010 9.505 8.779 4.389 14.245 7.123 12.481 6.241 1.434 0.717 The majority of purchase decisions is influenced by different sources. ANALYSIS OF BUYING BEHAVIOUR OF CUSTOMERS IN THE ORGANISED RETAIL INDUSTRY The purchase decisions of customers in the organised retail industry are influenced by (in order of most influencing to least influencing sources): TV talk shows, infomercials, family members, websites, celebrity endorsement, TV commercials, sales promotion, neighbours / friends / colleagues, self, a salesperson in a store, direct-mail piece, magazine ads and others sources. 229 7.2.5 Purchaser (buyer) in the family 7.2.5: Purchaser (buyer) in the family of the customers Retailers Purchaser (buyer) in the family Self Spouse Children Parent Surrogate f % f % f % f % f % HyperCITY 79 39.50 70 35.00 12 6.00 29 14.50 10 5.00 Star Bazaar 81 40.50 81 40.50 13 6.50 18 9.00 7 3.50 more.MEGAStore 73 36.50 74 37.00 10 5.00 34 17.00 9 4.50 Big Bazaar 78 39.00 80 40.00 9 4.50 26 13.00 7 3.50 D Mart 91 45.50 79 39.50 16 8.00 12 6.00 2 1.00 Big Megamart 89 44.50 64 32.00 10 5.00 35 17.50 2 1.00 West Side 85 42.50 68 34.00 11 5.50 32 16.00 4 2.00 Max 88 44.00 70 35.00 8 4.00 31 15.50 3 1.50 Reliance Digital 119 59.50 50 25.00 15 7.50 16 8.00 0 0.00 NEXT 116 58.00 49 24.50 17 8.50 18 9.00 0 0.00 Mean 89.90 44.95 68.50 34.25 12.10 6.05 25.10 12.55 4.40 2.20 Standard Deviation 15.574 7.787 11.433 5.717 3.071 1.536 8.373 4.186 3.627 1.814 The majorities of customers in the organised retail industry are either self- purchase decision maker in the family or rely on their spouse and parents. Similarly, the majority of customers in the organised retail industry purchase the merchandise and services by self. The second majority of customers relies on their spouse; while the third majority relies on their parents to purchase the merchandise and services for them. Moreover, few customers also rely on their children and surrogate to purchase the merchandise and services for them. The electronics-item superstores (i.e. NEXT and Reliance Digital) are getting more footfalls from the self-purchaser in comparison to other retailers included in this study. The primary reason of the same is that both the stores have more than 80% of customers from the male group of customers. All the multi-format organised retailers included in the study are getting more footfalls from the customers, who rely on their spouse to purchase the merchandise and services for them in comparison to other retailers. ANALYSIS OF BUYING BEHAVIOUR OF CUSTOMERS IN THE ORGANISED RETAIL INDUSTRY The fashion store organised retailers included in the study (i.e. Big Megamart, West Side and Max) along with more.MEGAStore, Big Bazaar and HyperCITY are getting more footfalls from the customers, who rely on their parents to purchase the merchandise and services for them in comparison to other retailers and industry mean value. 230 7.2.6 Duration of visiting the store of retailers g 7.2.6: Duration of visiting the store by customers Retailers Durations First time Last 3 months Last 6 months Last 1 year Last 2 year More than 2 years f % f % f % f % f % f % HyperCITY 4 2.00 48 24.00 59 29.50 68 34.00 16 8.00 5 2.50 Star Bazaar 6 3.00 49 24.50 75 37.50 62 31.00 8 4.00 0 0.00 more.MEGAStore 7 3.50 52 26.00 71 35.50 59 29.50 11 5.50 0 0.00 Big Bazaar 3 1.50 28 14.00 72 36.00 56 28.00 29 14.50 12 6.00 D Mart 5 2.50 50 25.00 70 35.00 67 33.50 7 3.50 1 0.50 Big Megamart 9 4.50 29 14.50 46 23.00 64 32.00 30 15.00 22 11.00 West Side 8 4.00 26 13.00 78 39.00 60 30.00 25 12.50 3 1.50 Max 8 4.00 31 15.50 72 36.00 61 30.50 28 14.00 0 0.00 Reliance Digital 12 6.00 44 22.00 79 39.50 65 32.50 0 0.00 0 0.00 NEXT 14 7.00 40 20.00 81 40.50 62 31.00 3 1.50 0 0.00 Mean 7.60 3.80 39.70 19.85 70.30 35.15 62.40 31.20 15.70 7.85 4.30 2.15 Standard Deviation 3.438 1.719 10.253 5.126 10.520 5.260 3.688 1.844 11.470 5.735 7.288 3.644 The majority of customers of HyperCITY (a multi-format hyper store) are visiting their store from last one year. HyperCITY is getting more footfalls of this category customers in comparison to others included in the study. The majority of customers of Star Bazaar (a multi-format hyper store) are visiting their store for last 6 months. Star Bazaar is getting more footfalls of this category customers in comparison to other multi-format organised retailers included in the study. The majority of customers of Big Bazaar, more.MEGAStore and D Mart (all multi- format retailers) are also visiting their store for last 6 months. The majority of customers of the Big Megamart (a fashion-item hyper store) is visiting their store from last one year, while the majority of customers of fashion-item superstores (i.e. ANALYSIS OF BUYING BEHAVIOUR OF CUSTOMERS IN THE ORGANISED RETAIL INDUSTRY West Side and Max) is visiting their store for last 6 months. Moreover, the majority of customers of electronic superstores is also visiting the stores from last 6 months. 231 7.2.7 Visits to the store in a month 7.2.7: Visits of customers to the store in a month Retailers Visits 1 Time 2 Times 3 Times 4 Times More than 4 times f % f % f % f % f % HyperCITY 19 9.50 40 20.00 59 29.50 67 33.50 15 7.50 Star Bazaar 14 7.00 35 17.50 63 31.50 78 39.00 10 5.00 more.MEGAStore 15 7.50 34 17.00 55 27.50 70 35.00 26 13.00 Big Bazaar 20 10.00 42 21.00 38 19.00 81 40.50 19 9.50 D Mart 18 9.00 52 26.00 31 15.50 79 39.50 20 10.00 Big Megamart 89 44.50 68 34.00 24 12.00 14 7.00 5 2.50 West Side 81 40.50 75 37.50 26 13.00 10 5.00 8 4.00 Max 79 39.50 72 36.00 28 14.00 17 8.50 4 2.00 Reliance Digital 102 51.00 69 34.50 18 9.00 9 4.50 2 1.00 NEXT 107 53.50 63 31.50 19 9.50 8 4.00 3 1.50 Mean 54.40 27.20 55.00 27.50 36.10 18.05 43.30 21.65 11.20 5.60 Standard Deviation 40.123 20.061 16.200 8.100 16.895 8.447 33.757 16.879 8.337 4.169 7.2.7: Visits of customers to the store in a month The majority of customers of all the multi-format organised retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart) visits the stores 3 or more times in a month. These retailers get a good footfall of customers from the 1 or 2-time visitors in the month. Moreover, they are also getting the footfalls from more than 4 times visitors in a month. The majority of customers of all the fashion stores organised retailers (i.e. Big Megamart, West Side and Max) as well as the electronics-item superstores (i.e. Reliance Digital and NEXT) visits the stores 1 or 2 times in a month. However, these retailers are also getting enough 3 and 4 times visitors in a month, but the number of more than 4-time visitors is very less and negligible. 7.2.8 Preferred days for visiting the stores The majority of customers in the organised retail industry prefers to visit the stores on weekends. But, it does not mean that the industry is not getting footfall on weekdays. There are a great number of customers, who prefer to visit the stores on weekdays. Moreover, there are also a great number of customers, who actually visit the stores whenever they need to visit. Surprisingly, four multi-format organised retailers (i.e. Star Bazaar, more.MEGAStore, Big Bazaar and D Mart) are getting 232 fewer footfalls from the customers on weekends in comparison to others (included in the study) and the industry mean value. Big Bazaar is highest footfalls of customers on weekdays in comparison to others (included in the study) and the industry mean value. The reason for the same is its attractive schemes on weekdays. 7.2.8: Preferred days of customers for visiting the stores Retailers Days Weekdays Weekends Any day f % f % f % HyperCITY 26 13.00 123 61.50 51 25.50 Star Bazaar 53 26.50 88 44.00 59 29.50 more.MEGAStore 38 19.00 98 49.00 64 32.00 Big Bazaar 68 34.00 79 39.50 53 26.50 D Mart 40 20.00 99 49.50 61 30.50 Big Megamart 29 14.50 148 74.00 23 11.50 West Side 35 17.50 131 65.50 34 17.00 Max 41 20.50 128 64.00 31 15.50 Reliance Digital 30 15.00 132 66.00 38 19.00 NEXT 46 23.00 124 62.00 30 15.00 Mean 40.60 20.30 115.00 57.50 44.40 22.20 Standard Deviation 12.616 6.308 22.405 11.203 14.849 7.424 All multi-format organised retailers included in the study have a great number of customers, who visit the stores on all the days in comparison to others (included in the study) and the industry mean value. 7.2.9 Preferred time for visiting the stores The majority of customers in the organised retail industry prefers to visit the stores in the evening time. But, it does not mean that the industry is not getting footfall in morning and afternoon. The industry has a good number of customers, who prefer to visit their stores in the morning. The industry is also getting the footfalls of the customers in the store in the afternoon, but it is less and negligible in the case of some retailers. 233 7.2.9: Preferred time for visiting the stores 7.2.9: Preferred time for visiting the stores Retailers Timings Morning Afternoon Evening Any time f % f % f % f % HyperCITY 36 18.00 10 5.00 136 68.00 18 9.00 Star Bazaar 16 8.00 2 1.00 152 76.00 30 15.00 more.MEGAStore 28 14.00 16 8.00 128 64.00 28 14.00 Big Bazaar 35 17.50 8 4.00 125 62.50 32 16.00 D Mart 40 20.00 12 6.00 96 48.00 52 26.00 Big Megamart 12 6.00 6 3.00 143 71.50 39 19.50 West Side 18 9.00 5 2.50 137 68.50 40 20.00 Max 20 10.00 10 5.00 143 71.50 27 13.50 Reliance Digital 10 5.00 7 3.50 141 70.50 42 21.00 NEXT 15 7.50 12 6.00 121 60.50 52 26.00 Mean 23.00 11.50 8.80 4.40 132.20 66.10 36.00 18.00 Standard Deviation 10.873 5.437 4.050 2.025 15.796 7.898 11.025 5.513 Moreover, the industry also has a great number of customers, whose timing to visit the store is not fixed and they prefer to visit any point of time. All multi-format organised retailers included in the study except Star Bazaar are getting more footfalls in the morning time in comparison to others (included in the study) and the industry mean value. 7.2.10 Spending amount on purchasing from the stores 7.2.10: Spending amount of purchasing by customers from the stores Retailers Spending amounts < 1000 ₹ 1000 - 2000 ₹ 2001 - 5000 ₹ > 5000 ₹ f % f % f % f % HyperCITY 9 4.50 24 12.00 69 34.50 98 49.00 Star Bazaar 22 11.00 56 28.00 78 39.00 44 22.00 more.MEGAStore 20 10.00 60 30.00 82 41.00 38 19.00 Big Bazaar 16 8.00 36 18.00 75 37.50 73 36.50 D Mart 39 19.50 72 36.00 68 34.00 21 10.50 Big Megamart 8 4.00 30 15.00 65 32.50 97 48.50 West Side 12 6.00 33 16.50 72 36.00 83 41.50 Max 6 3.00 30 15.00 70 35.00 94 47.00 Reliance Digital 35 17.50 41 20.50 52 26.00 72 36.00 NEXT 30 15.00 45 22.50 65 32.50 60 30.00 Mean 19.70 9.85 42.70 21.35 69.60 34.80 68.00 34.00 Standard Deviation 11.691 5.845 15.471 7.735 8.262 4.131 26.733 13.367 234 The majority of customers of the fashion store retailers as well as electronics stores included in this study spends more than 5000 per month for purchasing from ₹ the stores, while the majority of customers of multi-format hyper store retailers except HyperCITY spend 2000 to 5000 per month. The HyperCITY is getting more ₹ ₹ footfalls of customers, who spend more than 5000 per month for purchasing from ₹ the store. 7.2.10 Spending amount on purchasing from the stores 7.2.11 The categories of the products, from which customers purchase products Table 7.2.11: Categories of the products, from which customers purchase Retailers Categories of the products Basic (staples) merchandise Fashion merchandise Seasonal merchandise Consumer durables merchandise f % f % f % f % HyperCITY 139 69.5 104 52.0 46 23.0 98 49.0 Star Bazaar 140 70.0 130 65.0 36 18.0 79 39.5 more.MEGAStore 150 75.0 95 47.5 28 14.0 70 35.0 Big Bazaar 152 76.0 136 68.0 49 24.5 100 50.0 D Mart 164 82.0 142 71.0 50 25.0 69 34.5 Big Megamart 5 2.5 200 100.0 98 49.0 48 24.0 West Side 4 2.0 200 100.0 75 37.5 52 26.0 Max 2 1.0 200 100.0 62 31.0 0 0.0 Reliance Digital 0 0.0 0 0.0 0 0.0 200 100.0 NEXT 4 2.0 12 6.0 0 0.0 200 100.0 Mean 76.0 38.0 121.9 61.0 44.4 22.2 91.6 45.8 Standard Deviation 77.260 38.630 72.226 36.113 30.653 15.326 63.778 31.889 7.2.11 The categories of the products, from which customers purchase products The majority of customers of all the multi-format organised retailers purchases the products from all the categories of merchandise i.e. basic (staples) merchandise, fashion merchandise, seasonal merchandise and consumer durable. However, D Mart is getting more footfalls of customers, who purchase the basic (staple) merchandise, fashion merchandise, seasonal merchandise category products in comparison to other multi-format stores. Big Bazaar and HyperCITY are getting more footfalls of customers, who purchase; the consumer durables merchandise category products in comparison to other multi-format stores. The majority of customers of fashion stores (i.e. Big Megamart, West Side and Max) purchase the products from fashion 235 merchandise category, as these are fashion stores; while the majority of customers of electronics-item superstores (i.e. Reliance Digital and NEXT) purchases the products from consumer durables merchandise category, as these store deals with the durables in large. 7.2.12 Types of products, the customers purchase from the stores Figure 7.2.1 Figure 7.2.2 7.2.12 Types of products, the customers purchase from the stores Figure 7.2.1 7.2.12 Types of products, the customers purchase from the stores Figure 7.2.1 Figure 7.2.2 Figure 7.2.2 236 Figure 7.2.3 Figure 7.2.4 Fi 7 2 3 Figure 7.2.3 Figure 7.2.3 Figure 7.2.3 g Figure 7.2.4 Figure 7.2.4 237 Figure 7.2.5 Figure 7.2.5 The customers of multi-format retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart) purchase all types of products from their stores, they offer. 7.2.10 Spending amount on purchasing from the stores The majority of customers from HyperCITY purchases food & grocery, health & beauty, home & furniture, fashion products as well store level brands. The majority of customers of Star Bazaar purchase food & grocery, fashion & accessories, personal & home care, home & kitchen products as well as store level brands. The majority of customers of more.MEGAStore purchase grocery, home and personal care, apparels, footwear and electronic products. The majority of customers of Big Bazaar purchase grocery & staples, fashion, electronics, personal, kitchen & dining toys, sports, outdoors and home products. The majority of customers of D Mart purchase foods, garments, toiletries, beauty, kitchenware, footwear and home appliances. The food & grocery, fashion, home care, personal care, kitchen, electronics, products are some common types of the products purchased by customers of the organised retail industry. 238 Figure 7.2.6 Figure 7.2.7 Figure 7.2.6 Figure 7.2.6 Figure 7 2 7 Figure 7.2.7 Figure 7.2.7 239 Figure 7.2.8 Figure 7.2.8 Figure 7.2.8 Figure 7.2.8 The customers of fashion store retailers purchase all types of products from the stores, they offer. Women’s wear, men’s wear, kids wear, footwear and accessories are the most common products purchased by the customers of fashion store retailers. The majority of customers of all three fashion store retailers included in the study purchase Women’s wear, men’s wear, footwear and kids wear from the stores. Big Megamart and West Side are also getting a good sale of footwear, cosmetics, travel, toys and home products. Figure 7.2.9 Figure 7.2.9 Figure 7.2.9 240 Figure 7.2.10 Figure 7.2.10 The customers of electronic store retailers purchase all types of products from the stores, they offer. TV, audio, home appliances, kitchen appliances, mobile phones, computer, cameras, gaming products and accessories are the most common products purchased by the customers of electronic store retailers. The majority of customers of both electronics stores retailers included in the study purchase entertainment items, home appliances, cameras, computers, phones, gaming products and accessories from the store. purchase The majority of customers in the organised retail industry is impulse purchasers. The numbers of impulse buyers are high in case of all the multi-format hyper stores (i.e. HyperCITY, Star Bazaar, more.MEGAStore and Big Bazaar) in comparison to the industry mean value. But, D Mart (multi-format superstore) is getting fewer footfalls from the impulse buyers in comparison to the industry mean value as well as multi-format hyper store included in the study. 241 Table 7.2.12: Impulse buyers Retailers Impulse purchasing Yes No f % f % HyperCITY 184 92.00 16 8.00 Star Bazaar 190 95.00 10 5.00 more.MEGAStore 178 89.00 22 11.00 Big Bazaar 185 92.50 15 7.50 D Mart 159 79.50 41 20.50 Big Megamart 188 94.00 12 6.00 West Side 186 93.00 14 7.00 Max 179 89.50 21 10.50 Reliance Digital 128 64.00 72 36.00 NEXT 81 40.50 119 59.50 Mean 165.80 82.90 34.20 17.10 Standard Deviation 35.238 17.619 35.238 17.619 Table 7.2.12: Impulse buyers The numbers of impulse buyers are also high in case of all the fashion stores (i.e. Big Megamart, West Side and Max) in comparison to the industry mean value. However, the electronics retailers are not getting fewer footfalls of the impulse buyers in comparison to the industry mean value. Figure 7.2.11 Figure 7.2.11 242 Figure 7.2.12 Figure 7.2.13 Figure 7.2.12 Figure 7.2.13 Figure 7.2.12 Figure 7.2.13 Figure 7.2.12 243 Figure 7.2.14 Figure 7.2.15 Figure 7.2.14 Figure 7.2.14 Figure 7.2.15 Figure 7.2.15 Figure 7.2.15 244 Food & grocery, beauty, electronics, fashion products are the most common type of products purchased by impulse buyers from the stores of the multi-format organised retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart). Figure 7.2.16 Figure 7.2.17 Figure 7.2.16 Figure 7.2.16 Figure 7.2.16 Figure 7.2.17 Figure 7.2.17 Figure 7.2.17 Figure 7.2.17 245 Figure 7.2.18 Figure 7.2.18 Women’s wear, men’s wear, kid’s wear, foot wears and accessories are the most common type of products purchased by impulse buyers from the stores of fashion store retailers (i.e. Big Megamart, West Side and Max). Figure 7.2.19 246 Figure 7.2.20 Figure 7.2.20 Figure 7.2.20 Accessories are the most common type of products purchased by impulse buyers from the stores of electronics store retailers (i.e. Reliance Digital and Next). However, the NEXT is also getting the impulse buyers in the categories of personal care, phones, games, and computer accessories. 7.2.14 Reasons of impulse purchase by customers According to the customers of organised retail industry, the common and important reasons for impulse purchase are (in order of most important to least important): price reductions, multi-item discounts, the point of purchase display, demonstration and samples, salespeople, store atmosphere, store layout, coupons, packaging and shelving techniques. The majority of customers makes impulse purchases due to price reductions, multi-item discounts, the point of purchase display, demonstration and samples and salespeople. In the case of customers of multi-format organised retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart) included in this study, two most important reasons for their impulse purchase are price reductions and multi-item discounts on products. 247 As per the customers of fashion store retailers (i.e. Big Megamart, West Side and Max), the most important reasons for their impulse purchase are multi-item discounts, the point of purchase display, price reductions, store atmosphere and salespeople. The customers of electronics stores prefer less impulse purchase in comparison to customers of multi-format and fashion store organised retailers. According to the customers of electronics store retailers (i.e. Reliance Digital and Next), the most important reasons for their impulse purchase are multi-item discounts, price reductions and salespeople. 7.2.14 Reasons of impulse purchase by customers Table 7.2.13: Reasons of impulse purchase by customers Retailers Reasons Point of purchase display Price reductions Store layout Coupons Shelving techniques Demonstration and samples Multi-item discounts Packaging Store atmosphere Salespeople f % f % f % f % f % f % f % f % f % f % HyperCITY 58 29.00 82 41.00 15 7.50 22 11.00 16 8.00 38 19.00 69 34.50 28 14.00 42 21.00 26 13.00 Star Bazaar 62 31.00 76 38.00 20 10.00 35 17.50 10 5.00 52 26.00 72 36.00 30 15.00 28 14.00 39 19.50 more.MEGAStore 43 21.50 71 35.50 28 14.00 12 6.00 20 10.00 48 24.00 65 32.50 22 11.00 38 19.00 33 16.50 Big Bazaar 68 34.00 80 40.00 33 16.50 40 20.00 28 14.00 56 28.00 76 38.00 32 16.00 45 22.50 40 20.00 D Mart 12 6.00 88 44.00 8 4.00 28 14.00 3 1.50 22 11.00 79 39.50 12 6.00 5 2.50 12 6.00 Big Megamart 70 35.00 68 34.00 49 24.50 18 9.00 32 16.00 45 22.50 80 40.00 24 12.00 58 29.00 48 24.00 West Side 75 37.50 70 35.00 42 21.00 32 16.00 29 14.50 48 24.00 82 41.00 22 11.00 62 31.00 53 26.50 Max 66 33.00 62 31.00 38 19.00 29 14.50 26 13.00 41 20.50 73 36.50 19 9.50 59 29.50 47 23.50 Reliance Digital 40 20.00 49 24.50 11 5.50 13 6.50 9 4.50 38 19.00 49 24.50 8 4.00 12 6.00 51 25.50 NEXT 32 16.00 38 19.00 5 2.50 7 3.50 3 1.50 17 8.50 39 19.50 9 4.50 9 4.50 43 21.50 Mean 52.60 26.30 68.40 34.20 24.90 12.45 23.60 11.80 17.60 8.80 40.50 20.25 68.40 34.20 20.60 10.30 35.80 17.90 39.20 19.60 Standard Deviation 20.173 10.086 15.320 7.660 15.337 7.668 10.926 5.463 10.967 5.483 12.528 6.264 14.049 7.025 8.553 4.277 21.447 10.723 12.630 6.315 Table 7.2.13: Reasons of impulse purchase by customers 248 7.2.15 The first preference of the customers while purchasing The majority of customers’ first preference is either quality or brand in the organised retail industry. However, the good news for the organised retailers is that there are also a great percentage of customers in the organised retail industry, whose first preference is discounts, schemes, price, quantity and other; while purchasing. The highest number of quality and brand conscious customers are from fashion store retailers (i.e. Big Megamart, West Side and Max). The highest number of price and quantity conscious customers are from multi-format organised retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart); while discounts and schemes conscious customers can be found from all organised retailers. Table 7.2.14: The first preference of the customers while purchasing Retailers Variables for preference Quality Brand Price Quantity Discounts Schemes Others f % f % f % f % f % f % f % HyperCITY 52 26.00 41 20.50 26 13.00 22 11.00 28 14.00 26 13.00 5 2.50 Star Bazaar 37 18.50 30 15.00 32 16.00 29 14.50 34 17.00 32 16.00 6 3.00 more.MEGAStore 28 14.00 33 16.50 35 17.50 26 13.00 40 20.00 34 17.00 4 2.00 Big Bazaar 40 20.00 41 20.50 24 12.00 21 10.50 37 18.50 35 17.50 2 1.00 D Mart 23 11.50 21 10.50 31 15.50 31 15.50 45 22.50 44 22.00 5 2.50 Big Megamart 61 30.50 59 29.50 20 10.00 10 5.00 23 11.50 22 11.00 5 2.50 West Side 63 31.50 61 30.50 18 9.00 8 4.00 23 11.50 21 10.50 6 3.00 Max 60 30.00 57 28.50 21 10.50 9 4.50 24 12.00 25 12.50 4 2.00 Reliance Digital 50 25.00 47 23.50 26 13.00 20 10.00 26 13.00 28 14.00 3 1.50 NEXT 42 21.00 39 19.50 30 15.00 23 11.50 31 15.50 31 15.50 4 2.00 Mean 45.60 22.80 42.90 21.45 26.30 13.15 19.90 9.95 31.10 15.55 29.80 14.90 4.40 2.20 Standard Deviation 13.930 6.965 13.203 6.602 5.638 2.819 8.279 4.140 7.695 3.847 6.925 3.462 1.265 0.632 Table 7.2.14: The first preference of the customers while purchasing Variables for preference 249 7.2.16 Reasons for visiting a particular store by customers The customers in the organised retail industry, visit the particular organised retail stores due to several reasons, most notably, (in order of most important to least important as per the industry mean value calculated from the responses of customers): merchandise (merchandise assortment- variety of merchandise and brands, quality of assortment, in service situations, merchandise offered as part of service, styling or fashion guarantee, warranties and pricing, reasonable prices), promotion (advertising, displays, trading stamps, sales promotion- discount & special schemes, symbol and colours), store atmosphere (atmosphere of congeniality, customers’ feeling of warmth, acceptance or ease, good & spacious ambience, and delighting store environment), convenience (general convenience, parking access, location convenience- proximity to home/office), services (service in general, salesclerk service, presence of self-service, ease of merchandise return, delivery service, phone ordering and credit policy), physical facilities (elevators, lighting, air conditioning, restrooms, store layout, shopping ease, aisle placement, carpeting and architecture), post-transaction satisfaction (merchandise in use, return & adjustment policy), institutional factors (projection of store- conservative or modern, reputation, reliability), clientele (social class appeal, store personnel, fit between self-image and store image- self-image congruence), other (friends suggested, no other store nearby, convenient store hours). The merchandise (merchandise assortment- variety of merchandise and brands, quality of assortment, in service situations, merchandise offered as part of service, styling or fashion guarantee, warranties and pricing, reasonable prices) is the primary reason for visiting the stores among the customers of multi-format retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar except D Mart) as well as electronics-item superstores (i.e. Reliance Digital and NEXT) included in the study. As the promotion (advertising, displays, trading stamps, sales promotion- discount & special schemes, symbol and colours) is the primary reason for visiting the store by customers of D Mart. However, the store atmosphere (atmosphere of congeniality, customers’ feeling of warmth, acceptance or ease, good & spacious ambience, and delighting 250 store environment) is the primary reason for visiting the stores among the customers of fashion store retailers (i.e. Big Megamart, West Side and Max). The convenience (general convenience, parking access, location convenience- proximity to home/office) is a second reason for visiting the stores among the customers of multi-format retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, D Mart except Big Bazaar). As the promotion (advertising, displays, trading stamps, sales promotion- discount & special schemes, symbol and colours) is a second reason for visiting the store by customers of Big Bazaar. 7.2.16 Reasons for visiting a particular store by customers However, the merchandise (merchandise assortment- variety of merchandise and brands, quality of assortment, in service situations, merchandise offered as part of service, styling or fashion guarantee, warranties and pricing, reasonable prices) is a second reason for visiting the stores among the customers of fashion store retailers (i.e. Big Megamart, West Side and Max). Moreover, services (service in general, sales-clerk service, the presence of self-service, ease of merchandise return, delivery service, phone ordering and credit policy) is a second reason for visiting the stores among the customers of Reliance Digital and promotion is a second reason for visiting the store by customers of NEXT. The promotion (advertising, displays, trading stamps, sales promotion- discount & special schemes, symbol and colours) is a third reason for visiting the stores among the customers of two multi-format retailers (i.e. Star Bazaar and more.MEGAStore); while, store atmosphere (atmosphere of congeniality, customers’ feeling of warmth, acceptance or ease, good & spacious ambience, and delighting store environment) is a third reason for visiting the stores by customers of HyperCITY; convenience (general convenience, parking access, location convenience - proximity to home/office) is a third reason for visiting the stores by customers of Big Bazaar (a multi-format hyper store), two fashion-item superstores (i.e. West Side and Max) and Reliance Digital (an electronics superstore), and merchandise (merchandise assortment- variety of merchandise and brands, quality of assortment, in service situations, merchandise offered as part of service, styling or fashion guarantee, warranties and pricing, reasonable prices) is a third reason for visiting the stores by customers of D Mart. 7.2.16 Reasons for visiting a particular store by customers 251 Table 7.2.15: Reasons for visiting a particular store by customers Retailers Reasons for visiting Merchandise Services Clientele Physical Facilities Convenience Promotion Store Atmospheres Institutional Factors Post-transaction Satisfaction Others f % f % f % f % f % f % f % f % f % f % HyperCITY 120 60.00 98 49.00 65 32.50 90 45.00 107 53.50 98 49.00 105 52.50 63 31.50 69 34.50 53 26.50 Star Bazaar 119 59.50 97 48.50 61 30.50 88 44.00 101 50.50 97 48.50 95 47.50 55 27.50 65 32.50 59 29.50 more.MEGAStore 115 57.50 97 48.50 62 31.00 86 43.00 102 51.00 101 50.50 92 46.00 52 26.00 70 35.00 51 25.50 Big Bazaar 114 57.00 104 52.00 64 32.00 92 46.00 109 54.50 110 55.00 108 54.00 63 31.50 72 36.00 55 27.50 D Mart 90 45.00 70 35.00 40 20.00 61 30.50 88 44.00 119 59.50 47 23.50 32 16.00 32 16.00 50 25.00 Big Megamart 117 58.50 103 51.50 66 33.00 93 46.50 102 51.00 93 46.50 118 59.00 79 39.50 73 36.50 61 30.50 West Side 118 59.00 101 50.50 67 33.50 91 45.50 107 53.50 96 48.00 126 63.00 82 41.00 74 37.00 60 30.00 Max 106 53.00 98 49.00 63 31.50 87 43.50 103 51.50 91 45.50 112 56.00 73 36.50 68 34.00 52 26.00 Reliance Digital 114 57.00 105 52.50 68 34.00 79 39.50 101 50.50 96 48.00 92 46.00 69 34.50 66 33.00 53 26.50 NEXT 121 60.50 93 46.50 49 24.50 72 36.00 89 44.50 110 55.00 79 39.50 50 25.00 61 30.50 53 26.50 Mean 113.40 56.70 96.60 48.30 60.50 30.25 83.90 41.95 100.90 50.45 101.10 50.55 97.40 48.70 61.80 30.90 65.00 32.50 54.70 27.35 Standard Deviation 9.264 4.632 10.058 5.029 8.960 4.480 10.311 5.156 7.109 3.555 8.975 4.487 22.530 11.265 15.106 7.553 12.247 6.124 3.917 1.959 Table 7.2.15: Reasons for visiting a particular store by customers However, services (service in general, sales-clerk service, the presence of self-service, ease of merchandise return, delivery service, phone ordering and credit policy) is a third reason for visiting the stores by the customers of Big Megamart (a fashion-item hyper store) and NEXT (an electronics-item superstore). 7.2.16 Reasons for visiting a particular store by customers 252 7.2.17 Customers of organised retailers, visiting other stores / loyal customers of the organised retailers Table 7.2.16: Customers of organised retailers, visiting other stores Retailers Customers visiting other stores Yes No f % f % HyperCITY 117 58.50 83 41.50 Star Bazaar 112 56.00 88 44.00 more.MEGAStore 109 54.50 91 45.50 Big Bazaar 106 53.00 94 47.00 D Mart 121 60.50 79 39.50 Big Megamart 131 65.50 69 34.50 West Side 125 62.50 75 37.50 Max 123 61.50 77 38.50 Reliance Digital 142 71.00 58 29.00 NEXT 146 73.00 54 27.00 Mean 123.20 61.60 76.80 38.40 Standard Deviation 13.348 6.674 13.348 6.674 7.2.17 Customers of organised retailers, visiting other stores / loyal customers of the organised retailers The majority of customers in the organised retail industry visit more than one retail store. However, the industry has a great number of loyal customers. The multi- format organised retailers included in this study (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar, and D Mart) have more loyal customers in comparison to the others as well as industry mean value. However, among the multi- format organised retailers, the hyper stores (i.e. HyperCITY, Star Bazaar, more.MEGAStore and Big Bazaar) have also more loyal customers in comparison to a super store (i.e. D Mart). The fashion store retailers (i.e. Big Megamart, West Side and Max) have less loyal customers and their majority of customers visiting other stores in comparison to the industry mean value. Moreover, the electronics store retailer's situation is poorer in having loyal customers. They have lesser loyal customers and their majority of customers visiting other stores in comparison to the industry mean value. 7.2.16 Reasons for visiting a particular store by customers 253 7.2.18 Type of stores visited by customers of organised retailers other than their stores Table 7.2.17: Type of stores visited by customers of organised retailers other than their stores Retailers Type of stores Organised retail store Unorganised retail store Both f % f % f % HyperCITY 46 39.32 20 17.09 51 43.59 Star Bazaar 36 32.14 27 24.11 49 43.75 more.MEGAStore 37 33.94 25 22.94 47 43.12 Big Bazaar 36 33.96 18 16.98 52 49.06 D Mart 15 12.40 44 36.36 62 51.24 Big Megamart 40 30.53 24 18.32 67 51.15 West Side 36 28.80 21 16.80 68 54.40 Max 28 22.76 25 20.33 70 56.91 Reliance Digital 25 17.61 38 26.76 79 55.63 NEXT 19 13.01 42 28.77 85 58.22 Mean 31.80 26.45 28.40 22.85 63.00 50.71 Standard Deviation 9.750 9.443 9.419 6.373 13.115 5.702 7.2.18 Type of stores visited by customers of organised retailers other than their stores In the organised retail industry, the majority of customers (among, who also visits other stores than the regular one) visit both organised as well as unorganised retail stores. However, there are also a great number of customers, who only visit other organised retail stores or unorganised retail stores. The customers of multi- format retail stores (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar), who visit both organised and unorganised retail stores (other than these stores) are less in comparison to others and especially D Mart(a multi-format super store) & industry mean value. The customers of electronics-item superstores (i.e. Reliance Digital and NEXT), who visit only unorganised retail stores (other than these stores) or more in the comparison industry mean value. The customers of some other retailers (i.e. Star Bazaar, more.MEGAStore and D Mart) who visit only unorganised retail stores (other than these stores) are also more in comparison to the industry mean value and other retailers included in this study except customers of electronics-item superstores. The customers of multi-format retail stores (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar) and two fashion store retailers (i.e. Big Megamart and West Side), who visit only organised retail stores (other than these stores) are more in comparison others and the industry mean value. 7.2.16 Reasons for visiting a particular store by customers 254 7.2.19 Reasons for visiting other stores by customers Table 7.2.18: Reasons for visiting other stores by customers Retailers Reasons Merchandise Service Clientele Physical Facilities Convenience Promotion Store Atmospheres Institutional factors Post-transaction satisfaction Others f % f % f % f % f % f % f % f % f % f % HyperCITY 70 59.829 72 61.538 21 17.949 38 32.479 48 41.026 46 39.316 39 33.333 28 23.932 12 10.256 23 19.658 Star Bazaar 68 60.714 69 61.607 15 13.393 33 29.464 50 44.643 41 36.607 35 31.250 24 21.429 14 12.500 26 23.214 more.MEGAStore 72 66.055 68 62.385 17 15.596 29 26.606 46 42.202 45 41.284 33 30.275 16 14.679 15 13.761 19 17.431 Big Bazaar 54 50.943 67 63.208 14 13.208 25 23.585 41 38.679 52 49.057 32 30.189 18 16.981 9 8.491 25 23.585 D Mart 88 72.727 91 75.207 10 8.264 42 34.711 38 31.405 65 53.719 42 34.711 12 9.917 17 14.050 20 16.529 Big Megamart 67 51.145 71 54.198 23 17.557 29 22.137 34 25.954 35 26.718 27 20.611 32 24.427 9 6.870 31 23.664 West Side 69 55.200 70 56.000 21 16.800 23 18.400 36 28.800 39 31.200 25 20.000 33 26.400 13 10.400 29 23.200 Max 76 61.789 78 63.415 19 15.447 26 21.138 28 22.764 42 34.146 29 23.577 31 25.203 10 8.130 26 21.138 Reliance Digital 75 52.817 81 57.042 18 12.676 49 34.507 48 33.803 62 43.662 56 39.437 46 32.394 28 19.718 23 16.197 NEXT 61 41.781 89 60.959 11 7.534 59 40.411 52 35.616 67 45.890 59 40.411 49 33.562 32 21.918 21 14.384 Mean 70.000 57.300 75.600 61.556 16.900 13.842 35.300 28.344 42.100 34.489 49.400 40.160 37.700 30.379 28.900 22.892 15.900 12.609 24.300 19.900 Standard Deviation 9.068 8.797 8.771 5.763 4.358 3.620 11.691 7.130 7.923 7.279 11.501 8.299 11.653 7.152 12.124 7.439 7.923 4.955 3.860 3.540 Table 7.2.18: Reasons for visiting other stores by customers The customers in the organised retail industry, visit the other retail stores due to several reasons, most notably, (in order of most important to least important as per the industry mean value calculated from the responses of customers): services (service in general, salesclerk service, presence of self-service, ease of merchandise return, delivery service, phone ordering and credit policy), merchandise (merchandise assortment- variety of merchandise and brands, quality of assortment, in service situations, merchandise offered as part of service, styling or fashion guarantee, warranties and pricing, reasonable prices), promotion (advertising, displays, trading stamps, sales promotion- discount & 255 special schemes, symbol and colours), convenience (general convenience, parking access, location convenience- proximity to home/office), store atmosphere (atmosphere of congeniality, customers’ feeling of warmth, acceptance or ease, good & spacious ambience, and delighting store environment), physical facilities (elevators, lighting, air conditioning, restrooms, store layout, shopping ease, aisle placement, carpeting and architecture), institutional factors (projection of store- conservative or modern, reputation, reliability), other (friends suggested, no other store nearby, convenient store hours), clientele (social class appeal, store personnel, fit between self-image and store image- self-image congruence), post-transaction satisfaction (merchandise in use, return & adjustment policy). 7.2.16 Reasons for visiting a particular store by customers The majority visits other stores in the search of better services, merchandise, promotional offers, convenience, store atmospheres and physical facilities. 256 FACTORS INFLUENCING BUYING BEHAVIOUR OF CUSTOMERS IN THE ORGANISED RETAIL INDUSTRY 7.3.1 The influence of sociological / environmental / external factors on buying behaviour of customers 7.3.1 The influence of sociological / environmental / external factors on buying 7.3.1 The influence of sociological / environmental / external factors on buying behaviour of customers Figure 7.3.1 The sociological / environmental / external factors [i.e. cultural, sub-cultural, society, status in society (social class), reference group, family life cycle and role in the family] significantly influence the buying behaviour of customers in the organised retail industry. The calculated weighted average scores (WAS) for the influence of sociological / environmental / external factors on buying behaviour of customers are supporting the statement. The sociological / environmental / external factors have more influence on buying behaviour of customers of HyperCITY and D Mart in comparison to other multi-format organised retailers (i.e. Star Bazaar, more.MEGAStore and Big Bazaar). 257 Table 7.3.1: The influence of sociological / environmental / external factors on buying behaviour of customers Retailers Influence Extremely Influenced Influenced Neither Influenced Nor Uninfluenced Uninfluenced Not at all Influenced Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 60 30.00 86 43.00 34 17.00 13 6.50 7 3.50 0.895 Star Bazaar 58 29.00 78 39.00 39 19.50 15 7.50 10 5.00 0.795 more.MEGAStore 59 29.50 75 37.50 34 17.00 20 10.00 12 6.00 0.745 Big Bazaar 63 31.5 73 36.50 30 15.00 23 11.50 11 5.50 0.770 D Mart 66 33.00 76 38.00 32 16.00 18 9.00 8 4.00 0.870 Big Megamart 65 32.50 76 38.00 28 14.00 20 10.00 11 5.50 0.820 West Side 63 31.50 72 36.00 32 16.00 21 10.50 12 6.00 0.765 Max 63 31.50 71 35.50 32 16.00 22 11.00 12 6.00 0.755 Reliance Digital 62 31.00 69 34.50 33 16.50 24 12.00 12 6.00 0.725 NEXT 59 29.50 66 33.00 32 16.00 27 13.50 16 8.00 0.625 Mean 61.80 30.90 74.20 37.10 32.60 16.30 20.30 10.15 11.10 5.55 0.777 Standard Deviation 2.700 1.350 5.493 2.747 2.875 1.438 4.165 2.082 2.470 1.235 0.076 Table 7.3.1: The influence of sociological / environmental / external factors on buying behaviour of customers The sociological / environmental / external factors also have more influence on buying behaviour of customers of Big Megamart in comparison to other fashion store organised retailers (i.e. West Side and Max). The sociological / environmental / external factors also have more influence on buying behaviour of customers of Big Megamart in comparison to other fashion store organised retailers (i.e. West Side and Max). However, the influence of sociological / environmental / external factors on buying behaviour of customers of electronics store retailers included in the study (i.e. Reliance Digital and NEXT) is less in comparison to other retailers and industry mean value. FACTORS INFLUENCING BUYING BEHAVIOUR OF CUSTOMERS IN THE ORGANISED RETAIL INDUSTRY However, the influence of sociological / environmental / external factors on buying behaviour of customers of electronics store retailers included in the study (i.e. Reliance Digital and NEXT) is less in comparison to other retailers and industry mean value. 258 7.3.2 The influence of psychological / individual / internal factors on buying behaviour of customers 7.3.2 The influence of psychological / individual / internal factors on buying behaviour of customers behaviour of customers The psychological / individual / internal factors have more influence on buying behaviour of customers of Big Megamart in comparison to other fashion store organised retailers (i.e. West Side and Max). However, the influence of psychological / individual / internal factors on buying behaviour of customers of Reliance Digital is significant and insignificant in case of NEXT. Figure 7.3.2 The psychological / individual / internal factors (i.e. self-image, personality, personal values, lifestyle, memory, learning, perception, motivation, emotions, mood, involvement, information processing, beliefs, intention, attitude, communication, and persuasion) significantly influence the buying behaviour of customers in the organised retail industry. The calculated weighted average scores (WAS) for the influence of psychological / individual / internal factors on buying behaviour of customers are supporting the statement. The psychological / individual / internal factors significantly influence the buying behaviour of customers of all the multi-format hyper store organised retailers included in this study (HyperCITY, Star Bazaar, more.MEGAStore and Big Bazaar) except D Mart (a multi-format superstore). The psychological / individual / internal factors have more influence on buying behaviour of customers of HyperCITY in comparison to other multi-format hyper store organised retailers (i.e. Star Bazaar, more.MEGAStore and Big Bazaar). 259 Table 7.3.2: The influence of psychological / individual / internal factors on buying behaviour o Table 7.3.2: The influence of psychological / individual / internal factors on buying behaviour of customers Retailers Influence Extremely Influenced Influenced Neither Influenced Nor Uninfluenced Uninfluenced Not at all Influenced Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 61 30.50 76 38.00 43 21.50 14 7.00 6 3.00 0.860 Star Bazaar 57 28.50 70 35.00 34 17.00 24 12.00 15 7.50 0.650 more.MEGAStore 56 28.00 68 34.00 30 15.00 29 14.50 17 8.50 0.585 Big Bazaar 58 29.00 66 33.00 28 14.00 30 15.00 18 9.00 0.580 D Mart 53 26.50 61 30.50 32 16.00 35 17.50 19 9.50 0.470 Big Megamart 61 30.50 69 34.50 28 14.00 27 13.50 15 7.50 0.670 West Side 57 28.50 65 32.50 25 12.50 31 15.50 22 11.00 0.520 Max 61 30.50 67 33.50 27 13.50 26 13.00 19 9.50 0.625 Reliance Digital 56 28.00 62 31.00 33 16.50 32 16.00 17 8.50 0.540 NEXT 51 25.50 58 29.00 32 16.00 37 18.50 22 11.00 0.395 Mean 57.10 28.55 670 33.10 31.20 15.60 28.50 14.25 17.00 8.50 0.590 Standard Deviation 3.381 1.691 5.116 2.558 5.051 2.525 6.451 3.225 4.570 2.285 0.127 The psychological / individual / internal factors have more influence on buying behaviour of customers of Big Megamart in comparison to other fashion store organised retailers (i.e. West Side and Max). Figure 7.3.2 However, the influence of psychological / individual / internal factors on buying behaviour of customers of Reliance Digital is significant and insignificant in case of NEXT. 260 7.3.3 The influence of store related factors on buying behaviour of customers Figure 7.3.3 Figure 7.3.3 The store related factors (i.e. store image, store atmosphere, store brands, store location, sales personnel, discounts & schemes, and free services) significantly influence the buying behaviour of customers in the organised retail industry. The calculated weighted average scores (WAS) for the influence of store related factors on buying behaviour of customers are supporting the statement. The store related factors significantly influence the buying behaviour of customers of all the multi-format hyper store organised retailers included in this study (HyperCITY, Star Bazaar, more.MEGAStore and Big Bazaar) except D Mart (a multi- format superstore). The store related factors have more influence on buying behaviour of customers of HyperCITY in comparison to other multi-format hyper store organised retailers (i.e. Star Bazaar, more.MEGAStore and Big Bazaar). 261 Table 7.3.3: The influence of store related factors on buying behaviour of customers Table 7.3.3: The influence of store related factors on buying behaviour of customers Retailers Influence Extremely Influenced Influenced Neither Influenced Nor Uninfluenced Uninfluenced Not at all Influenced Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 57 28.50 74 37.00 40 20.00 19 9.50 10 5.00 0.745 Star Bazaar 55 27.50 68 34.00 26 13.00 32 16.00 19 9.50 0.540 more.MEGAStore 56 28.00 68 34.00 25 12.50 32 16.00 19 9.50 0.550 Big Bazaar 58 29.00 68 34.00 25 12.50 31 15.50 18 9.00 0.585 D Mart 51 25.50 56 28.00 32 16.00 37 18.50 24 12.00 0.365 Big Megamart 61 30.50 68 34.00 27 13.50 25 12.50 19 9.50 0.635 West Side 62 31.00 69 34.50 28 14.00 24 12.00 17 8.50 0.675 Max 61 30.50 69 34.50 27 13.50 24 12.00 19 9.50 0.645 Reliance Digital 60 30.00 64 32.00 30 15.00 27 13.50 19 9.50 0.595 NEXT 59 29.50 61 30.50 30 15.00 31 15.50 19 9.50 0.550 Mean 58.00 29.00 66.50 33.25 29.00 14.50 28.20 14.10 18.30 9.15 0.589 Standard Deviation 3.367 1.683 4.994 2.497 4.497 2.249 5.308 2.654 3.434 1.717 0.102 The store related factors also have more influence on buying behaviour of customers of West Side in comparison to other fashion store organised retailers (i.e. Max and Big Megamart). The store related factors also have more influence on buying behaviour of customers of West Side in comparison to other fashion store organised retailers (i.e. Max and Big Megamart). However, the influence of store related factors on buying behaviour of customers of electronics store retailers included in the study (i.e. Reliance Digital and NEXT) is significant.  Customer satisfaction with merchandise (products) and services  Testing of hypotheses Figure 7.3.2 However, the influence of store related factors on buying behaviour of customers of electronics store retailers included in the study (i.e. Reliance Digital and NEXT) is significant. 262 7.3.4 Factors influencing buying behaviour of customers of multi-format organised retailers Figure 7.3.4 Figure 7.3.5 263 7.3.4 Factors influencing buying behaviour of customers of multi-forma organised retailers 7.3.4 Factors influencing buying behaviour of customers of multi-format organised retailers Figure 7.3.4 Figure 7.3.5 263 Figure 7.3.6 Figure 7.3.7 Figure 7.3.7 264 Figure 7.3.8 Figure 7.3.8 The sociological factors (i.e. cultural, sub-cultural, society, status in society (social class), reference group, family life cycle, and role in the family), psychological factors (i.e. self-image, personality, personal values, lifestyle, memory, learning, perception, motivation, emotions, mood, involvement, information processing, beliefs, intention, attitude, communication, and persuasion) and store related factors (i.e. store image, store atmosphere, store brands, store location, sales personnel, discounts & schemes, and free services) significantly influence buying behaviour of customers of multi-format organised retailers. 265 7.3.5 Factors influencing buying behaviour of customers of fashion store organised retailers 7.3.5 Factors influencing buying behaviour of customers of fashion store organised retailers .3.5 Factors influencing buying behaviour of customers of fashion store organised retailers Figure 7.3.9 Figure 7.3.10 Figure 7.3.9 Figure 7.3.10 266 Figure 7.3.11 Figure 7.3.11 The sociological factors (i.e. cultural, sub-cultural, society, status in society (social class), reference group, family life cycle, and role in the family), psychological factors (i.e. self-image, personality, personal values, lifestyle, memory, learning, perception, motivation, emotions, mood, involvement, information processing, beliefs, intention, attitude, communication, and persuasion) and store related factors (i.e. store image, store atmosphere, store brands, store location, sales personnel, discounts & schemes, and free services) significantly influence buying behaviour of customers of fashion store organised retailers. 267 7.3.6 Factors influencing buying behaviour of customers of electronics stor organised retailers organised retailers Figure 7.3.12 Figure 7.3.13 Figure 7.3.13 268 The sociological factors (i.e. cultural, sub-cultural, society, status in society (social class), reference group, family life cycle, and role in the family), psychological factors (i.e. self-image, personality, personal values, lifestyle, memory, learning, perception, motivation, emotions, mood, involvement, information processing, beliefs, intention, attitude, communication, and persuasion) and store related factors (i.e. store image, store atmosphere, store brands, store location, sales personnel, discounts & schemes, and free services) significantly influence buying behaviour of customers of electronics stores organised retailers. 269  Testing of hypotheses Introduction This chapter comprises the analysis of the various data pertained to customer satisfaction with merchandise and services offered by organised retail industry, which have been collected through the individual questionnaire from the customers of the respective stores in a personal interaction. For a better analysis of the customer satisfaction, the customer expectations with merchandise and services have also been taken into the consideration for the study. And the responses pertained to the customer satisfaction have been collected on five point expectation cum satisfaction scale (basically an ordinal scale) ranging from greatly exceeding expectations and delighted (5), exceeding expectations and happy (4), matching expectations and satisfied (3), less than expectations and dissatisfied (2) and much less than expectations and disappointed (1). The weighted average scores (WAS) for all the variables (included in the study for customer satisfaction) and / or all the organised retailers (included in the study) have been computed and presented in the line chart for a comparative analysis. Moreover, the statistics [i.e. frequency of responses (f), the percentage of responses (%), mean, standard deviation, weighted average scores (WAS)] of all the variables (included in the study for customer satisfaction) and / or all the organised retailers included in the study (i.e. multi-format (item) retailers, specialised fashion item retailers and specialised electronics item retailers) have also been presented in the tables for a better comparative analysis. Further, results have been interpreted and discussed to reach on some inferences at the end of the different sections. Further, this chapter also comprises the testing of the hypotheses, which are supposed to be tested. 270 8.1. Customer satisfaction with merchandise assortment Figure 8.1: Weighted average scores (WAS) for customer satisfaction with merchandise assortment Figure 8.1: Weighted average scores (WAS) for customer satisfaction with merchandise assortment The mean WAS of all the organised retailers (included in this study) has been considered as the WAS of the industry for customer satisfaction with merchandise assortment and exhibited with the red line in figure 8.1. This WAS of the industry falls above the satisfied customer line in figure 8.1, which delineates that the organised retail industry satisfies a vast majority of customers with its merchandise assortment. The WAS for customer satisfaction with merchandise assortment of some retailers (i.e. Introduction Big Bazaar, HyperCITY, Big Megamart and West Side) are greater than the WAS of the industry, which indicates that these retailers satisfy more customers with their merchandise assortment than others (included in this study) in the industry. The most (i.e. 31.10%) customers are satisfied with merchandise assortment offered by the organised retailers (included in this study), as it matches with their expectations. The second largest group (i.e. 26.25%) of customers, who perceive that merchandise assortment is exceeding their expectations and they are 271 happy with it. The third major group (i.e. 21.65%) of customers is delighted with merchandise assortment, as it is greatly exceeding their expectations. However, there are also a massive number of customers, who are either dissatisfied (i.e. 13.10%) or disappointed (i.e. 7.90%) with merchandise assortment, because, the same is less than or much less than their expectations. Table 8.1: Comparative statistics of customer satisfaction with merchandise assortment Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 45 22.50 54 27.00 59 29.50 26 13.00 16 8.00 3.430 Star Bazaar 40 20.00 50 25.00 65 32.50 28 14.00 17 8.50 3.340 more.MEGAStore 39 19.50 48 24.00 64 32.00 31 15.50 18 9.00 3.295 Big Bazaar 49 24.50 59 29.50 66 33.00 17 8.50 9 4.50 3.610 D Mart 38 19.00 48 24.00 60 30.00 33 16.50 21 10.50 3.245 Big Megamart 51 25.50 59 29.50 67 33.50 14 7.00 9 4.50 3.645 West Side 47 23.50 55 27.50 64 32.00 24 12.00 10 5.00 3.525 Max 41 20.50 49 24.50 62 31.00 29 14.50 19 9.50 3.320 Reliance Digital 39 19.50 51 25.50 57 28.50 31 15.50 22 11.00 3.270 NEXT 44 22.00 52 26.00 58 29.00 29 14.50 17 8.50 3.385 Mean 43.30 21.65 52.50 26.25 62.20 31.10 26.20 13.10 15.80 7.90 3.407 Standard Deviation 4.596 2.298 4.143 2.072 3.521 1.761 6.233 3.116 4.826 2.413 0.337 Table 8.1: Comparative statistics of customer satisfaction with merchandise assortment Responses on expectation cum satisfaction scale 272 Big Bazaar leads in satisfying more customers with its merchandise assortment in comparison to its competitors or other multi-format retailers included in the study (i.e. Introduction HyperCITY, Star Bazaar, more.MEGAStore and D Mart), while, Big Megamart is a leader in satisfying more customers with its merchandise assortment in comparison to its competitors or other fashion item retailers included in the study (i.e. West Side and Max) and in the case of electronics item retailers included in this study, NEXT satisfies more customers with its merchandise assortment than its competitor included in the study (i.e. Reliance Digital). 8.2. Customer satisfaction with merchandise categories in the assortment Figure 8.2: Weighted average scores (WAS) for customer satisfaction with merchandise categories in the assortment 8.2. Customer satisfaction with merchandise categories in the assortment Figure 8.2: Weighted average scores (WAS) for customer satisfaction with merchandise categories in the assortment The mean WAS of all the organised retailers (included in this study) has been perceived as the WAS of the industry for customer satisfaction with merchandise categories in the assortment and shown with the red line in figure 8.2. This WAS of the industry falls above the satisfied customer line in figure 8.2, which depicts that the organised retail industry satisfies a great majority of customers with its merchandise categories (e.g. basic or staple, fashion, seasonal, consumer durable, etc.) in the assortment. 273 The WAS for customer satisfaction with merchandise categories in the assortment of some retailers (i.e. Big Bazaar, West Side, Big Megamart, NEXT and HyperCITY) are greater than the WAS of the industry, which reveals that these retailers satisfy more customers with their merchandise categories in the assortment than others (included in this study) in the industry. The WAS for customer satisfaction with merchandise categories in the assortment of some retailers (i.e. Big Bazaar, West Side, Big Megamart, NEXT and HyperCITY) are greater than the WAS of the industry, which reveals that these retailers satisfy more customers with their merchandise categories in the assortment than others (included in this study) in the industry. Introduction Table 8.2: Comparative statistics of customer satisfaction with merchandise categories in the assortment Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 41 20.50 48 24.00 58 29.00 32 16.00 21 10.50 3.280 Star Bazaar 38 19.00 49 24.50 58 29.00 33 16.50 22 11.00 3.240 more.MEGAStore 36 18.00 48 24.00 55 27.50 37 18.50 24 12.00 3.175 Big Bazaar 45 22.50 53 26.50 63 31.50 26 13.00 13 6.50 3.455 D Mart 32 16.00 43 21.50 59 29.50 38 19.00 28 14.00 3.065 Big Megamart 43 21.50 51 25.50 60 30.00 31 15.50 15 7.50 3.380 West Side 41 20.50 52 26.00 59 29.50 32 16.00 16 8.00 3.350 Max 37 18.50 47 23.50 54 27.00 36 18.00 26 13.00 3.165 Reliance Digital 37 18.50 46 23.00 55 27.50 36 18.00 26 13.00 3.160 NEXT 42 21.00 50 25.00 53 26.50 33 16.50 22 11.00 3.285 Mean 39.20 19.60 48.70 24.35 57.40 28.70 33.40 16.70 21.30 10.65 3.256 Standard Deviation 3.882 1.941 2.983 1.492 3.098 1.549 3.534 1.767 5.100 2.550 0.264 Table 8.2: Comparative statistics of customer satisfaction with merchandise categories in the mparative statistics of customer satisfaction with merchandise categories in the assortment 274 The maximal (i.e. 28.70%) customers are satisfied with merchandise categories in the assortment offered by the organised retailers (included in this study), as those are matching with their expectations. The second verdant group (i.e. 24.35%) of customers, who deem that merchandise categories in the assortment are exceeding their expectations and they are happy. The industry also delights an ample number (i.e. 19.60%) of customers with merchandise categories in the assortment by greatly exceeding them than customers’ expectations. The next gigantic group (i.e. 16.70%) of customers is dissatisfied with merchandise categories in the assortment because these are deficient to their expectations. Moreover, there is also a non-negligible number (i.e. 10.65%) of customers, those are disappointed with merchandise categories in the assortment as the same is much less than their expectations. Big Bazaar satisfies more customers with its merchandise categories in the assortment in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). Introduction While, Big Megamart leads in satisfying more customers with its merchandise categories in the assortment in comparison to its competitors or other fashion item retailers included in the study (i.e. West Side and Max) and in the case of electronics item retailers included in this study, NEXT satisfies more customers with its merchandise categories in the assortment than its competitor included in the study (i.e. Reliance Digital). 8.3. Customer satisfaction with the variety of merchandise in each category The mean WAS of all the organised retailers (included in this study) has been comprehended as the WAS of the industry for customer satisfaction with the variety of merchandise in each category and displayed with the red line in figure 8.3. This WAS of the industry indulges above the satisfied customer line in figure 8.3, which explicates that the organised retail industry satisfies an umpteenth majority of customers with its the variety of merchandise in each category as per different sizes, colours, designs, price, quality, quantity and brands. The WAS for customer satisfaction with the variety of merchandise in each category of some retailers (i.e. HyperCITY, Big Bazaar, Big Megamart and West Side) are more than the WAS of the industry, which evinces that these retailers satisfy 275 more customers with their the variety of merchandise in each category than others (included in this study) in the industry. Figure 8.3: Weighted average scores (WAS) for customer satisfaction with the variety of merchandise in each category Figure 8.3: Weighted average scores (WAS) for customer satisfaction with the variety of merchandise in each category The maximum (i.e. 28.50%) customers are satisfied with the variety of merchandise in each category offered by the organised retailers (included in this study), as it is matching with their expectations. The second giant group (i.e. 23.45%) of customers, who comprehend that the variety of merchandise in each category are exceeding their expectations and they are happy. Moreover, the industry has also an excessive number (i.e. 20.85%) of delighted customers with the variety of merchandise in each category by greatly exceeding the same than their expectations. The next plenteous group (i.e. 16.00%) of customers is dissatisfied with the variety of merchandise in each category because these are lacking to their expectations. However, there is also a non-negligible disappointed (i.e. 11.20%) customers with the variety of merchandise in each category, as the same is much less than their expectations. 8.3. Customer satisfaction with the variety of merchandise in each category 276 Table 8.3: Comparative statistics of customer satisfaction with the variety of merchandise in each category Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 43 21.50 46 23.00 57 28.50 35 17.50 19 9.50 3.295 Star Bazaar 40 20.00 46 23.00 58 29.00 33 16.50 23 11.50 3.235 more.MEGAStore 39 19.50 44 22.00 55 27.50 37 18.50 25 12.50 3.175 Big Bazaar 49 24.50 57 28.50 61 30.50 20 10.00 13 6.50 3.545 D Mart 35 17.50 43 21.50 58 29.00 37 18.50 27 13.50 3.110 Big Megamart 48 24.00 49 24.50 56 28.00 29 14.50 18 9.00 3.400 West Side 43 21.50 52 26.00 60 30.00 28 14.00 17 8.50 3.380 Max 39 19.50 45 22.50 54 27.00 33 16.50 29 14.50 3.160 Reliance Digital 40 20.00 43 21.50 55 27.50 35 17.50 27 13.50 3.170 NEXT 41 20.50 44 22.00 56 28.00 33 16.50 26 13.00 3.205 Mean 41.70 20.85 46.90 23.45 57.00 28.50 32.00 16.00 22.40 11.20 3.268 Standard Deviation 4.244 2.122 4.533 2.266 2.261 1.130 5.164 2.582 5.317 2.658 0.309 Big Bazaar is also a leader in satisfying more customers with its variety of merchandise in each category in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, Big Megamart leads in satisfying more customers with its variety of merchandise in each category in comparison to its competitors or other fashion item retailers included in the study (i.e. West Side and Max) and in the case of electronics item retailers included in this study, NEXT satisfies more customers with its variety of merchandise in each category than its competitor included in the study (i.e. Reliance Digital). Big Bazaar is also a leader in satisfying more customers with its variety of merchandise in each category in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, Big Megamart leads in satisfying more customers with its variety of merchandise in each category in comparison to its competitors or other fashion item retailers included in the study (i.e. Big Bazaar is also a leader in satisfying more customers with its variety of merchandise in each category in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, Big Megamart leads in satisfying more customers with its variety of merchandise in each category in comparison to its competitors or other fashion item retailers included in the study (i.e. West Side and Max) and in the case of electronics item retailers included in this study, NEXT satisfies more customers with its variety of merchandise in each category than its competitor included in the study (i.e. Reliance Digital). 8.3. Customer satisfaction with the variety of merchandise in each category West Side and Max) and in the case of electronics item retailers included in this study, NEXT satisfies more customers with its variety of merchandise in each category than its competitor included in the study (i.e. Reliance Digital). 277 8.4. Customer satisfaction with availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise 8.4. Customer satisfaction with availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise customised merchandise, unique merchandise and complementary merchandise Figure 8.4: Weighted average scores (WAS) for customer satisfaction with availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise Figure 8.4: Weighted average scores (WAS) for customer satisfaction with availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise Figure 8.4: Weighted average scores (WAS) for customer satisfaction with availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise The mean WAS of all the organised retailers (included in this study) has been perceived as the WAS of the industry for customer satisfaction with availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise and shown with the red line in figure 8.4. This WAS of the industry exists above the satisfied customer line in figure 8.4, which exhorts that the organised retail industry satisfies an adequate majority of customers by making available newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise in each category. The WAS for customer satisfaction with the availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise of some retailers (i.e. Big Bazaar, Big Megamart and West Side) are more than the WAS of the industry, which manifests that these retailers satisfy more customers with the availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise than others (included in this study) in the industry. The most (i.e. 30.15%) customers are satisfied with the availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise by the organised retailers (included in this study). The second exuberant group (i.e. 24.50%) of customers, who understand that the availability of newly launched merchandise, customised merchandise, unique merchandise and 8.3. Customer satisfaction with the variety of merchandise in each category 278 Table 8.4: Comparative statistics of customer satisfaction with the availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 38 19.00 49 24.50 59 29.50 32 16.00 22 11.00 3.245 Star Bazaar 36 18.00 50 25.00 52 26.00 36 18.00 26 13.00 3.170 more.MEGAStore 37 18.50 52 26.00 61 30.50 29 14.50 21 10.50 3.275 Big Bazaar 43 21.50 48 24.00 66 33.00 28 14.00 15 7.50 3.380 D Mart 33 16.50 44 22.00 55 27.50 40 20.00 28 14.00 3.070 Big Megamart 42 21.00 51 25.50 65 32.50 28 14.00 14 7.00 3.395 West Side 41 20.50 49 24.50 64 32.00 29 14.50 17 8.50 3.340 Max 40 20.00 48 24.00 59 29.50 32 16.00 21 10.50 3.270 Reliance Digital 38 19.00 49 24.50 62 31.00 30 15.00 21 10.50 3.265 NEXT 37 18.50 50 25.00 60 30.00 33 16.50 20 10.00 3.255 Mean 38.50 19.25 49.00 24.50 60.30 30.15 31.70 15.85 20.50 10.25 3.267 Standard Deviation 3.028 1.514 2.160 1.080 4.373 2.186 3.860 1.930 4.403 2.202 0.245 The most (i.e. 30.15%) customers are satisfied with the availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise by the organised retailers (included in this study). The second exuberant group (i.e. 24.50%) of customers, who understand that the availability of newly launched merchandise, customised merchandise, unique merchandise and 279 complementary merchandise is exceeding their expectations and they are happy. Moreover, the industry has also an intense number (i.e. 19.25%) of delighted customers with the availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise. The next enormous group (i.e. 20.85%) of customers is dissatisfied with the availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise. However, there is also a significant number (i.e. 10.25%) of disappointed customers with the availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise. Big Bazaar performs better in satisfying more customers with the availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). 8.3. Customer satisfaction with the variety of merchandise in each category While, Big Megamart is also a better performer in satisfying more customers with the availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise as compared to its competitors or other fashion item retailers included in the study (i.e. West Side and Max) and in the case of electronics item retailers included in this study, Reliance Digital perform better in satisfying more customers with the availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise than its competitor included in the study (i.e. NEXT). 8.5. Customer satisfaction with brands in each merchandise category The mean WAS of all the organised retailers (included in this study) has been deemed as the WAS of the industry for customer satisfaction with brands in each merchandise category and demonstrated by the red line in figure 8.5. This WAS of the industry exists above the satisfied customer line in figure 8.5, which addresses that the organised retail industry satisfies a substantial majority of customers with the availability of the brands (i.e. local & regional, national, international, private / store, copycat, exclusive, generic brands) in each merchandise category. 280 Figure 8.5: Weighted average scores (WAS) for customer satisfaction with brands in each merchandise category Figure 8.5: Weighted average scores (WAS) for customer satisfaction with brands in each merchandise category The WAS for customer satisfaction with brands in each merchandise category of some retailers (i.e. Big Bazaar, Big Megamart and West Side) are more than the WAS of the industry, which discloses that these retailers satisfy more customers with the brands in each merchandise category than others (included in this study) in the industry. The most (i.e. 29.40%) customers are satisfied with brands in each merchandise category offered by the organised retailers (included in this study). The second bounteous group (i.e. 24.75%) of customers, who ponder that brands in each merchandise category are exceeding their expectations and they are happy. The third flagrant group (i.e. 20.20%) of customers is delighted with brands in each merchandise category because the same is greatly exceeding their expectations. However, there are also a vast number of customers, who are either dissatisfied (i.e. 16.10%) or disappointed (i.e. 9.55%) with brands in each merchandise category, because, the same is less than or much less than their expectations. Big Bazaar leads in satisfying more customers with brands in each merchandise category in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, Big Megamart leads in satisfying more customers with brands in each merchandise category as compared to its competitors or other fashion item retailers included in the study (i.e. West Side and Max) and in the case of electronics item retailers included in this study, NEXT satisfies more customers with brands in each merchandise category than its competitor included in the study (i.e. Reliance Digital). 8.5. Customer satisfaction with brands in each merchandise category 281 Table 8.5: Comparative statistics of customer satisfaction with the brands in each merchandise category Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 39 19.50 47 23.50 58 29.00 33 16.50 23 11.50 3.230 Star Bazaar 42 21.00 42 21.00 61 30.50 35 17.50 20 10.00 3.255 more.MEGAStore 33 16.50 52 26.00 54 27.00 38 19.00 23 11.50 3.170 Big Bazaar 44 22.00 47 23.50 60 30.00 34 17.00 15 7.50 3.355 D Mart 33 16.50 52 26.00 53 26.50 39 19.50 23 11.50 3.165 Big Megamart 53 26.50 55 27.50 61 30.50 18 9.00 13 6.50 3.585 West Side 45 22.50 51 25.50 60 30.00 30 15.00 14 7.00 3.415 Max 41 20.50 46 23.00 59 29.50 35 17.50 19 9.50 3.275 Reliance Digital 38 19.00 49 24.50 60 30.00 31 15.50 22 11.00 3.250 NEXT 36 18.00 54 27.00 62 31.00 29 14.50 19 9.50 3.295 Mean 40.40 20.20 49.50 24.75 58.80 29.40 32.20 16.10 19.10 9.55 3.300 Standard Deviation 6.077 3.039 4.035 2.017 3.011 1.506 5.940 2.970 3.872 1.936 0.357 While, Big Megamart leads in satisfying more customers with brands in each merchandise category as compared to its competitors or other fashion item retailers included in the study (i.e. West Side and Max) and in the case of electronics item retailers included in this study, NEXT satisfies more customers with brands in each merchandise category than its competitor included in the study (i.e. Reliance Digital). 282 8.6 Customer satisfaction with quality assurance and quality of merchandise Figure 8.6: Weighted average scores (WAS) for customer satisfaction with quality assurance and quality of merchandise Figure 8.6: Weighted average scores (WAS) for customer satisfaction with quality assurance and quality of merchandise The mean WAS of all the organised retailers (included in this study) has acceded as the WAS of the industry for customer satisfaction with quality assurance and quality of merchandise and presented with the red line in figure 8.6. This WAS of the industry lies above the satisfied customer line in figure 8.6, which expresses that the organised retail industry satisfies a constitutive majority of customers with its quality assurance and quality of merchandise. The WAS for customer satisfaction with quality assurance and quality of merchandise of some retailers (i.e. 8.5. Customer satisfaction with brands in each merchandise category Big Bazaar, Big Megamart and West Side) are greater than the WAS of the industry, which unfolds that these retailers satisfy more customers with their quality assurance and quality of merchandise than others (included in this study) in the industry. The majority (i.e. 31.10%) customers are satisfied with quality assurance and quality of merchandise offered by the organised retailers (included in this study), as the same matches with their expectations. The second plenteous group (i.e. 26.55%) of customers, who ponder that quality assurance and quality of merchandise are exceeding their expectations and they are happy. The third adjunct group (i.e. 283 19.50%) of customers is delighted with quality assurance and quality of merchandise because the same is greatly exceeding their expectations. However, there are also a pervasive number of customers, who are either dissatisfied (i.e. 14.50%) or disappointed (i.e. 8.35%) with quality assurance and quality of merchandise, because, the same is less than or much less than their expectations. 8.5. Customer satisfaction with brands in each merchandise category Table 8.6: Comparative statistics of customer satisfaction with quality assurance and quality of merchandise Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 43 21.50 47 23.50 61 30.50 32 16.00 17 8.50 3.335 Star Bazaar 39 19.50 48 24.00 70 35.00 30 15.00 13 6.50 3.350 more.MEGAStore 36 18.00 55 27.50 61 30.50 31 15.50 17 8.50 3.310 Big Bazaar 45 22.50 54 27.00 67 33.50 23 11.50 11 5.50 3.495 D Mart 33 16.50 65 32.50 58 29.00 27 13.50 17 8.50 3.350 Big Megamart 43 21.50 51 25.50 59 29.50 28 14.00 19 9.50 3.355 West Side 41 20.50 53 26.50 61 30.50 29 14.50 16 8.00 3.370 Max 38 19.00 49 24.50 62 31.00 33 16.50 18 9.00 3.280 Reliance Digital 37 18.50 52 26.00 62 31.00 29 14.50 20 10.00 3.285 NEXT 35 17.50 57 28.50 61 30.50 28 14.00 19 9.50 3.305 Mean 39.00 19.50 53.10 26.55 62.20 31.10 29.00 14.50 16.70 8.35 3.344 Standard Deviation 3.916 1.958 5.238 2.619 3.615 1.807 2.828 1.414 2.791 1.395 0.299 mparative statistics of customer satisfaction with quality assurance and quality of merchandise Responses on expectation cum satisfaction scale 284 Big Bazaar is the champ in satisfying more customers with its quality assurance and quality of merchandise in comparison to its competitors or other multi- format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, West Side leads in satisfying more customers with its quality assurance and quality of merchandise as compared to its competitors or other fashion item retailers included in the study (i.e. Big Megamart and Max) and in the case of electronics item retailers included in this study, NEXT leads in satisfying more customers with quality assurance and quality of merchandise than its competitor included in the study (i.e. Reliance Digital). 8.7 Customer satisfaction with the services Figure 8.7: Weighted average scores (WAS) for customer satisfaction with the services 8.7 Customer satisfaction with the services 8.7 Customer satisfaction with the services Figure 8.7: Weighted average scores (WAS) for customer satisfaction with the services The mean WAS of all the organised retailers (included in this study) has been recognised as the WAS of the industry for customer satisfaction with the services and tendered with the red line in figure 8.7. This WAS of the industry alights above the satisfied customer line in the figure 8.7, which describes that the organised retail industry satisfies a robust majority of customers with its services (e.g. Credit, home delivery, alteration, installation, packaging (gift wrapping), complaints & returns handling, gift certificates, trial purchase, special sales for regular customers, staff on 285 late hours, extended store hours, mail & phone orders, restrooms, restaurants, bridal registry, baby strollers, interior designers, personal shoppers, ticket outlets, parking, water fountains, pay phones, fitting rooms, beauty salon, fur storage, shopping bags, information, safety of personal things, lift & escalator facility, babysitting, after sale services, demonstrations, mobile and laptop charger points, entertainment facilities, refreshment facilities, medical facility, the internet billing facility, game zones, music zone (karaoke, etc.), special facilities for physically handicapped, other in-store services, other out-store services, other pre-purchase or post-purchase services, complementary or adjunct services - inventory support, installation, repair, etc.). The WAS for customer satisfaction with the services of some retailers (i.e. HyperCITY, Star Bazaar, Big Bazaar, Big Megamart and West Side) are greater than the WAS of the industry, which divulges that these retailers satisfy more customers with their services than others (included in this study) in the industry. The dignified number (i.e. 30.65%) of customers are satisfied with the services offered by the organised retailers (included in this study), as the same matches with their expectations. The second signified group (i.e. 24.75%) of customers, who consider that the services are exceeding their expectations and they are happy. The third coherent group (i.e. 20.40%) of customers is delighted with the services because the same is greatly exceeding their expectations. However, there are also an extensive number of customers, who are either dissatisfied (i.e. 15.65%) or disappointed (i.e. 8.55%) with the services, because, the same is less than or much less than their expectations. Big Bazaar satisfies more customers with its services in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). 8.7 Customer satisfaction with the services While, West Side leads in satisfying more customers with its services in comparison to its competitors or other fashion item retailers included in the study (i.e. Big Megamart and Max) and in the case of electronics item retailers included in this study, Reliance Digital satisfies more customers with its services than its competitor included in the study (i.e. NEXT). 286 Table 8.7: Comparative statistics of customer satisfaction with the services Table 8.7: Comparative statistics of customer satisfaction with the services Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 48 24.00 59 29.50 61 30.50 21 10.50 11 5.50 3.560 Star Bazaar 42 21.00 51 25.50 67 33.50 24 12.00 16 8.00 3.395 more.MEGAStore 39 19.50 48 24.00 67 33.50 31 15.50 15 7.50 3.325 Big Bazaar 51 25.50 58 29.00 65 32.50 18 9.00 8 4.00 3.630 D Mart 27 13.50 43 21.50 60 30.00 46 23.00 24 12.00 3.015 Big Megamart 43 21.50 49 24.50 57 28.50 34 17.00 17 8.50 3.335 West Side 47 23.50 51 25.50 61 30.50 26 13.00 15 7.50 3.445 Max 31 15.50 43 21.50 58 29.00 45 22.50 23 11.50 3.070 Reliance Digital 41 20.50 47 23.50 59 29.50 33 16.50 20 10.00 3.280 NEXT 39 19.50 46 23.00 58 29.00 35 17.50 22 11.00 3.225 Mean 40.80 20.40 49.50 24.75 61.30 30.65 31.30 15.65 17.10 8.55 3.328 Standard Deviation 7.406 3.703 5.503 2.751 3.743 1.872 9.381 4.691 5.216 2.608 0.471 287 8.8 Customer satisfaction with the quality of the services Figure 8.8: Weighted average scores (WAS) for customer satisfaction with the quality of the services Figure 8.8: Weighted average scores (WAS) for customer satisfaction with the quality of the services The mean WAS of all the organised retailers (included in this study) has been supposed as the WAS of the industry for customer satisfaction with the quality of the services and proposed with the red line in figure 8.8. 8.7 Customer satisfaction with the services This WAS of the industry rests above the satisfied customer line in the figure 8.8, which represents that the organised retail industry satisfies a sufficient majority of customers with their quality of the services by managing service tangibles (tangibility), having consistency and dependability in service performance (reliability), committing to provide services in time (responsiveness), having competence, courtesy to its customers and security of service operations (assurance), and putting itself in its customers place (empathy). The WAS for customer satisfaction with the quality of the services of some retailers (i.e. HyperCITY, Big Bazaar, Big Megamart and West Side) are greater than the WAS of the industry, which promulgates that these retailers satisfy more customers with their quality of the services than others (included in this study) in the industry. The most (i.e. 29.10%) customers are satisfied with the quality of the services offered by the organised retailers (included in this study), as the same is as per their expectations. The second plenteous group (i.e. 23.70) of customers, who 288 think that the quality of the services is exceeding their expectations and they are happy. Moreover, the industry has also a thundering number (i.e. 19.90%) of delighted customers with the quality of the services. The next giant group (i.e. 17.45%) of customers is dissatisfied with the quality of the services because the same is lacking to their expectations. However, there is also a non-negligible disappointed (i.e. 9.85%) customers with the quality of the services. 8.7 Customer satisfaction with the services Table 8.8: Comparative statistics of customer satisfaction with the quality of the services Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 41 20.50 49 24.50 58 29.00 33 16.50 19 9.50 3.300 Star Bazaar 38 19.00 48 24.00 59 29.50 35 17.50 20 10.00 3.245 more.MEGAStore 39 19.50 46 23.00 57 28.50 37 18.50 21 10.50 3.225 Big Bazaar 48 24.00 52 26.00 63 31.50 24 12.00 13 6.50 3.490 D Mart 30 15.00 41 20.50 55 27.50 48 24.00 26 13.00 3.005 Big Megamart 47 23.50 51 25.50 58 29.00 29 14.50 15 7.50 3.430 West Side 46 23.00 53 26.50 60 30.00 27 13.50 14 7.00 3.450 Max 34 17.00 44 22.00 57 28.50 41 20.50 24 12.00 3.115 Reliance Digital 39 19.50 46 23.00 58 29.00 36 18.00 21 10.50 3.230 NEXT 36 18.00 44 22.00 57 28.50 39 19.50 24 12.00 3.145 Mean 39.80 19.90 47.40 23.70 58.20 29.10 34.90 17.45 19.70 9.85 3.264 Standard Deviation 5.846 2.923 3.893 1.947 2.150 1.075 7.078 3.539 4.473 2.237 0.349 Table 8.8: Comparative statistics of customer satisfaction with the quality of the services 289 Big Bazaar is also a champ in satisfying more customers with its quality of the services in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, West Side a champ in satisfying more customers with its quality of the services in comparison to its competitors or other fashion item retailers included in the study (i.e. Big Megamart and Max) and in the case of electronics item retailers included in this study, Reliance Digital leads in satisfying more customers with its quality of the services than its competitor included in the study (i.e. NEXT). 8.9 Customer satisfaction with the pricing of merchandise & services 8.9 Customer satisfaction with the pricing of merchandise & services Figure 8.9: Weighted average scores (WAS) for customer satisfaction with the pricing of merchandise & services Figure 8.9: Weighted average scores (WAS) for customer satisfaction with the pricing of merchandise & services The mean WAS of all the organised retailers (included in this study) has been reckoned as the WAS of the industry for customer satisfaction with the pricing of merchandise & services and put forwarded with the red line in figure 8.9. This WAS of the industry reclines above the satisfied customer line in figure 8.9, which signifies that the organised retail industry satisfies an ample majority of customers with its customer expected pricing of merchandise & services. The WAS for customer satisfaction with the pricing of merchandise & services of some retailers (i.e. Big Bazaar, D Mart and NEXT) are more than the 290 WAS of the industry, which declares that these retailers satisfy more customers with their pricing of merchandise & services than others (included in this study) in the industry. WAS of the industry, which declares that these retailers satisfy more customers with their pricing of merchandise & services than others (included in this study) in the industry. Table 8.9: Comparative statistics of customer satisfaction with the pricing of merchandise & services Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 40 20.00 44 22.00 59 29.50 34 17.00 23 11.50 3.220 Star Bazaar 39 19.50 47 23.50 61 30.50 32 16.00 21 10.50 3.255 more.MEGAStore 38 19.00 49 24.50 60 30.00 31 15.50 22 11.00 3.250 Big Bazaar 43 21.50 48 24.00 69 34.50 25 12.50 15 7.50 3.395 D Mart 46 23.00 53 26.50 59 29.50 24 12.00 18 9.00 3.425 Big Megamart 37 18.50 47 23.50 57 28.50 36 18.00 25 12.50 3.205 West Side 39 19.50 45 22.50 56 28.00 37 18.50 23 11.50 3.200 Max 41 20.50 44 22.00 58 29.00 35 17.50 22 11.00 3.235 Reliance Digital 36 18.00 52 26.00 58 29.00 34 17.00 20 10.00 3.250 NEXT 47 23.50 55 27.50 61 30.50 23 11.50 14 7.00 3.490 Mean 40.60 20.30 48.40 24.20 59.80 29.90 31.10 15.55 20.30 10.15 3.293 Standard Deviation 3.688 1.844 3.836 1.918 3.615 1.807 5.216 2.608 3.592 1.796 0.293 291 The maximum (i.e. 8.9 Customer satisfaction with the pricing of merchandise & services 29.90%) customers are satisfied with the pricing of merchandise & services decided by the organised retailers (included in this study), as it matches with their expectations. The second exuberant group (i.e. 24.20%) of customers, who perceive that the pricing of merchandise & services is exceeding their expectations and they are happy. Moreover, the industry also delights an adequate number (i.e. 20.30%) of customers with the pricing of merchandise & services by greatly exceeding the same than customers’ expectations. The next enormous group (i.e. 15.55%) of customers is dissatisfied with the pricing of merchandise & services because it is deficient to their expectations. There is also a non-negligible number (i.e. 10.15%) of customers, those are disappointed with the pricing of merchandise & services, as the same is much less than their expectations. D Mart leads in satisfying more customers with its pricing of merchandise & services in comparison to its competitors or other multi-format retailers included in the study (i.e. Big Bazaar, HyperCITY, Star Bazaar and more.MEGAStore). While, Max leads in satisfying more customers with its pricing of merchandise & services in comparison to its competitors or other fashion item retailers included in the study (i.e. Big Megamart and West Side) and in the case of electronics item retailers included in this study, NEXT leads in satisfying more customers with its pricing of merchandise & services than its competitor included in the study (i.e. Reliance Digital). services The mean WAS of all the organised retailers (included in this study) has been presumed as the WAS of the industry for customer satisfaction with the overall value Vs price of merchandise & services and tendered with the red line in figure 8.10. This WAS of the industry slouches above the satisfied customer line in figure 8.10, which evinces that the organised retail industry satisfies a considerable majority of customers with its overall value (i.e. product value, service value, personal value, image value) offered as compared to the price/cost (i.e. monetary cost, time cost, energy cost, psychic cost) of merchandise & services. 292 Figure 8.10: Weighted average scores (WAS) for customer satisfaction with the overall value Vs price of merchandise & services Figure 8.10: Weighted average scores (WAS) for customer satisfaction with the overall value Vs price of merchandise & services The WAS for customer satisfaction with the overall value offered compared to the price/cost of some retailers (i.e. Big Bazaar and HyperCITY) are greater than the WAS of the industry, which manifests that these retailers satisfy more customers with their overall value offered compared to the price/cost than others (included in this study) in the industry. The utmost (i.e. 28.30%) customers are satisfied with the overall value offered compared to the price/cost by the organised retailers (included in this study), as it matches with their expectations. The second plenteous group (i.e. 23.15%) of customers, who comprehend that the overall value offered compared to the price/cost is exceeding their expectations and they are happy. Moreover, the industry also delights a sufficient number (i.e. 20.25%) of customers with the overall value offered compared to the price/cost by greatly exceeding the value than customers’ expectations. The next giant group (i.e. 16.80%) of customers is dissatisfied with the overall value offered compared to the price/cost because it is lacking to their expectations. There is also a non-negligible number (i.e. 11.50%) of customers, those are disappointed with the overall value offered compared to the price/cost, as the same is much less than their expectations. Big Bazaar is a leader in satisfying more customers with its overall value offered compared to the price/cost in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, Max services 293 Table 8.10: Comparative statistics of customer satisfaction with the overall value Vs price of merchandise & services Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 40 20.00 46 23.00 59 29.50 33 16.50 22 11.00 3.245 Star Bazaar 39 19.50 45 22.50 57 28.50 36 18.00 23 11.50 3.205 more.MEGAStore 37 18.50 52 26.00 53 26.50 33 16.50 25 12.50 3.215 Big Bazaar 46 23.00 53 26.50 61 30.50 25 12.50 15 7.50 3.450 D Mart 38 19.00 50 25.00 54 27.00 34 17.00 24 12.00 3.220 Big Megamart 39 19.50 46 23.00 57 28.50 34 17.00 24 12.00 3.210 West Side 40 20.00 44 22.00 58 29.00 35 17.50 23 11.50 3.215 Max 43 21.50 39 19.50 57 28.50 41 20.50 20 10.00 3.220 Reliance Digital 41 20.50 43 21.50 56 28.00 33 16.50 27 13.50 3.190 NEXT 42 21.00 45 22.50 54 27.00 32 16.00 27 13.50 3.215 Mean 40.50 20.25 46.30 23.15 56.60 28.30 33.60 16.80 23.00 11.50 3.239 Standard Deviation 2.635 1.318 4.270 2.135 2.459 1.229 3.950 1.975 3.528 1.764 0.245 Big Bazaar is a leader in satisfying more customers with its overall value offered compared to the price/cost in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, Max Big Bazaar is a leader in satisfying more customers with its overall value offered compared to the price/cost in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, Max 294 leads in satisfying more customers with its overall value offered compared to the price/cost in comparison to its competitors or other fashion item retailers included in the study (i.e. Big Megamart and West Side) and in the case of electronics item retailers included in this study, NEXT is a leader in satisfying more customers with its overall value offered compared to the price/cost than its competitor included in the study (i.e. Reliance Digital). services 8.11 Customer satisfaction with the exterior atmospherics of the store Figure 8.11: Weighted average scores (WAS) for customer satisfaction with the exterior atmospherics of the store 8.11 Customer satisfaction with the exterior atmospherics of the store Figure 8.11: Weighted average scores (WAS) for customer satisfaction with the exterior atmospherics of the store The mean WAS of all the organised retailers (included in this study) has been envisaged as the WAS of the industry for customer satisfaction with the exterior atmospherics of the store and furnished with the red line in figure 8.11. This WAS of the industry crouches above the satisfied customer line in figure 8.11, which connotes that the organised retail industry satisfies a cogitable majority of customers with the exterior atmospherics (i.e. storefront, store marquee / sign board, frontage entrance, external display window / space, building, store architecture, uniqueness of the store, visibility of the store, location / site of store, parking facilities, exterior look of the store, ease of access, store nameplate, health & safety standards, surrounding area, etc.) of the stores. 295 The WAS for customer satisfaction with the exterior atmospherics of the store of some retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar, Big Megamart, West Side and Reliance Digital) are greater than the WAS of the industry, which unfolds that these retailers satisfy more customers with their exterior atmospherics of the store than others (included in this study) in the industry. The WAS for customer satisfaction with the exterior atmospherics of the store of some retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar, Big Megamart, West Side and Reliance Digital) are greater than the WAS of the industry, which unfolds that these retailers satisfy more customers with their exterior atmospherics of the store than others (included in this study) in the industry. services Table 8.11: Comparative statistics of customer satisfaction with the exterior atmospherics of the store Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 41 20.50 46 23.00 67 33.50 29 14.50 17 8.50 3.325 Star Bazaar 43 21.50 47 23.50 63 31.50 31 15.50 16 8.00 3.350 more.MEGAStore 37 18.50 53 26.50 55 27.50 33 16.50 22 11.00 3.250 Big Bazaar 49 24.50 50 25.00 68 34.00 21 10.50 12 6.00 3.515 D Mart 27 13.50 49 24.50 61 30.50 37 18.50 26 13.00 3.070 Big Megamart 52 26.00 49 24.50 61 30.50 24 12.00 14 7.00 3.505 West Side 43 21.50 48 24.00 65 32.50 26 13.00 18 9.00 3.360 Max 31 15.50 42 21.00 59 29.50 40 20.00 28 14.00 3.040 Reliance Digital 42 21.00 48 24.00 66 33.00 26 13.00 18 9.00 3.350 NEXT 23 11.50 48 24.00 65 32.50 37 18.50 27 13.50 3.015 Mean 38.80 19.40 48.00 24.00 63.00 31.50 30.40 15.20 19.80 9.90 3.278 Standard Deviation 9.319 4.660 2.828 1.414 4.028 2.014 6.293 3.146 5.633 2.817 0.441 296 The industry satisfies the utmost (i.e. 31.50%) customers with the exterior atmospherics of the store by matching the same with their expectations. The industry also has happy customers with the exterior atmospherics of the store, as a second major group (i.e. 24.00%), which feels that the exterior atmospherics of the store is exceeding their expectations. Moreover, the industry also delights a substantial number (i.e. 19.40%) of customers with the exterior atmospherics of the store by greatly exceeding the same than customers’ expectations. However, the industry carries the next major group (i.e. 15.20%) of dissatisfied customers with the exterior atmospherics of the store, as lacking to meet expectations of customers. At the same time, the industry hauls a non- negligible number (i.e. 9.90%) of disappointed customers with the exterior atmospherics of the store, as offers much less than expectations of customers. Big Bazaar satisfies more customers with its exterior atmospherics of the store in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, Big Megamart satisfies more customers with its exterior atmospherics of the store in comparison to its competitors or other fashion item retailers included in the study (i.e. services Max and West Side) and in the case of electronics item retailers included in this study, Reliance Digital leads in satisfying more customers with its exterior atmospherics of the store than its competitor included in the study (i.e. NEXT). 8.12 Customer satisfaction with the interior atmospherics of the store The mean WAS of all the organised retailers (included in this study) has been conceived as the WAS of the industry for customer satisfaction with the interior atmospherics of the store and decorated with the red line in figure 8.12. This WAS of the industry inclines above the satisfied customer line in figure 8.12, which enlightens that the organised retail industry satisfies a volumetric majority of customers with the interior atmospherics (i.e. flooring & ceiling, lighting, odour / smell / scent, fixtures, wall textures, sounds, colour, temperature, cleanliness, aisles, trial room, dead area, graphics, store size, etc.) of the stores. 297 Figure 8.12: Weighted average scores (WAS) for customer satisfaction with the interior atmospherics of the store Figure 8.12: Weighted average scores (WAS) for customer satisfaction with the interior atmospherics of the store The WAS for customer satisfaction with the interior atmospherics of the store of some retailers (i.e. HyperCITY, Star Bazaar, moreMEGAStore, Big Bazaar, Big Megamart, West Side and Reliance Digital) are more than the WAS of the industry, which announces that these retailers satisfy more customers with the interior atmospherics of the store than others (included in this study) in the industry. The industry satisfies the exhaustive (i.e. 30.85%) customers with the interior atmospherics of the store by matching the same with their expectations. The industry also has happy customers with the interior atmospherics of the store, as a second key group (i.e. 22.80%), which feels that the interior atmospherics of the store is exceeding their expectations. Hereto, the industry also delights a competent number (i.e. 19.10%) of customers with the interior atmospherics of the store by greatly exceeding the same than customers’ expectations. However, the industry totes the next large group (i.e. 16.90%) of dissatisfied customers with the interior atmospherics of the store, as imperfect to meet expectations. Moreover, the industry haulages a non-negligible number (i.e. 10.35%) of disappointed customers with the interior atmospherics of the store, as tender much less than expectations of customers. Big Bazaar also satisfies more customers with its interior atmospherics of the store in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, Big Megamart also satisfies more customers with its interior atmospherics of the store in comparison to its competitors or other fashion item retailers included in the study (i.e. 8.12 Customer satisfaction with the interior atmospherics of the store 298 Table 8.12: Comparative statistics of customer satisfaction with the interior atmospherics of the store Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 44 22.00 47 23.50 64 32.00 28 14.00 17 8.50 3.365 Star Bazaar 45 22.50 45 22.50 65 32.50 27 13.50 18 9.00 3.360 more.MEGAStore 39 19.50 47 23.50 62 31.00 31 15.50 21 10.50 3.260 Big Bazaar 48 24.00 53 26.50 67 33.50 21 10.50 11 5.50 3.530 D Mart 19 9.50 34 17.00 61 30.50 57 28.50 29 14.50 2.785 Big Megamart 49 24.50 52 26.00 61 30.50 23 11.50 15 7.50 3.485 West Side 41 20.50 49 24.50 62 31.00 28 14.00 20 10.00 3.315 Max 38 19.00 46 23.00 55 27.50 37 18.50 24 12.00 3.185 Reliance Digital 41 20.50 48 24.00 58 29.00 31 15.50 22 11.00 3.275 NEXT 18 9.00 35 17.50 62 31.00 55 27.50 30 15.00 2.780 Mean 38.20 19.10 45.60 22.80 61.70 30.85 33.80 16.90 20.70 10.35 3.234 Standard Deviation 10.983 5.491 6.363 3.182 3.401 1.700 12.506 6.253 5.926 2.963 0.608 Big Bazaar also satisfies more customers with its interior atmospherics of the store in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, Big Megamart also satisfies more customers with its interior atmospherics of the store in comparison to its competitors or other fashion item retailers included in the study (i.e. 299 Max and West Side) and in the case of electronics item retailers included in this study, Reliance Digital also leads in satisfying more customers with its interior atmospherics of the store than its competitor included in the study (i.e. NEXT). Max and West Side) and in the case of electronics item retailers included in this study, Reliance Digital also leads in satisfying more customers with its interior atmospherics of the store than its competitor included in the study (i.e. NEXT). 8.12 Customer satisfaction with the interior atmospherics of the store 8.13 Customer satisfaction with the store layout Figure 8.13: Weighted average scores (WAS) for customer satisfaction with the store layout 8.13 Customer satisfaction with the store layout 8.13 Customer satisfaction with the store layout Figure 8.13: Weighted average scores (WAS) for customer satisfaction with the store layout The mean WAS of all the organised retailers (included in this study) has been visualised as the WAS of the industry for customer satisfaction with the store layout and embellished with the red line in figure 8.13. This WAS of the industry bows above the satisfied customer line in figure 8.13, which communicates that the organised retail industry satisfies an illustrious majority of customers with the store layout (i.e. space planning, floor space allocation, e-display, department location, department size, department compatibility, location of various products, accessibility to cash/billing counters, signage, etc.). The WAS for customer satisfaction with the store layout of some retailers (i.e. HyperCITY, Big Bazaar, Big Megamart, West Side, Max and Reliance Digital) are greater than the WAS of the industry, which divulges that these retailers satisfy more customers with their store layout than others (included in this study) in the industry. 300 Table 8.13: Comparative statistics of customer satisfaction with the store layout Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 38 19.00 48 24.00 59 29.50 35 17.50 20 10.00 3.245 Star Bazaar 36 18.00 46 23.00 57 28.50 34 17.00 27 13.50 3.150 more.MEGAStore 34 17.00 46 23.00 55 27.50 38 19.00 27 13.50 3.110 Big Bazaar 40 20.00 45 22.50 60 30.00 35 17.50 20 10.00 3.250 D Mart 18 9.00 36 18.00 59 29.50 56 28.00 31 15.50 2.770 Big Megamart 53 26.50 51 25.50 54 27.00 25 12.50 17 8.50 3.490 West Side 45 22.50 52 26.00 58 29.00 26 13.00 19 9.50 3.390 Max 38 19.00 49 24.50 57 28.50 34 17.00 22 11.00 3.235 Reliance Digital 37 18.50 45 22.50 57 28.50 33 16.50 28 14.00 3.150 NEXT 20 10.00 37 18.50 59 29.50 53 26.50 31 15.50 2.810 Mean 35.90 17.95 45.50 22.75 57.50 28.75 36.90 18.45 24.20 12.10 3.160 Standard Deviation 10.429 5.215 5.318 2.659 1.900 0.950 10.137 5.069 5.181 2.591 0.523 Table 8.13: Comparative statistics of customer satisfaction with the store layout The industry satisfies the extensive group (i.e. 8.12 Customer satisfaction with the interior atmospherics of the store 28.75%) of customers with the store layout by matching the same with their expectations. The industry also has happy customers with the store layout, as a second fundamental group (i.e. 22.75%), which feels that the store layout is exceeding their expectations. However, the industry bears the next large group (i.e. 18.45%) of dissatisfied customers with the store layout, as 301 immature to meet expectations. Herein, the industry also delights an effectual number (i.e. 17.95%) of customers with the store layout by greatly exceeding the same than customers’ expectations. Moreover, the industry ferriages a non-negligible number (i.e. 12.10%) of disappointed customers with the store layout, as maintains much less than expectations of customers. Big Bazaar also champs in satisfying more customers with its store layout in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, Big Megamart champs in satisfying more customers with its store layout in comparison to its competitors or other fashion item retailers included in the study (i.e. Max and West Side) and in the case of electronics item retailers included in this study, Reliance Digital also champs in satisfying more customers with its store layout than its competitor included in the study (i.e. NEXT). 8.14 Customer satisfaction with the visual merchandising Figure 8.14: Weighted average scores (WAS) for customer satisfaction with the visual merchandising 8.14 Customer satisfaction with the visual merchandising 8.14 Customer satisfaction with the visual merchandising Figure 8.14: Weighted average scores (WAS) for customer satisfaction with the visual merchandising The mean WAS of all the organised retailers (included in this study) has been imagined as the WAS of the industry for customer satisfaction with the visual merchandising and adorned with the red line in figure 8.14. This WAS of the industry 302 reclines above the satisfied customer line in figure 8.14, which heralds that the organised retail industry satisfies an exuberant majority of customers with the visual merchandising (i.e. theme, racks & shelves, ensemble, payment counters, merchandise display, the planogram, props & mannequins, point of sale sign, methods of display, etc.). The WAS for customer satisfaction with the visual merchandising of some retailers (i.e. HyperCITY, Big Bazaar, Big Megamart, West Side, Max and Reliance Digital) are greater than the WAS of the industry, which declares that these retailers satisfy more customers with their visual merchandising than others (included in this study) in the industry. The industry satisfies the massive (i.e. 28.80%) customers with the visual merchandising by matching the same with their expectations. The industry also has happy customers with the visual merchandising, as a second basic group (i.e. 22.60%), which feels that the visual merchandising is exceeding their expectations. Hereto, the industry also delights a great number (i.e. 19.45%) of customers with the visual merchandising by greatly exceeding the same than customers’ expectations. However, the industry endures the next large group (i.e. 17.80%) of dissatisfied customers with the visual merchandising, as infantile to meet expectations. Moreover, the industry hauls a non-negligible number (i.e. 11.35%) of disappointed customers with the visual merchandising, as serve much less than expectations of customers. Big Bazaar also leads in satisfying more customers with its visual merchandising in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, Big Megamart also leads in satisfying more customers with its visual merchandising in comparison to its competitors or other fashion item retailers included in the study (i.e. Max and West Side) and in the case of electronics item retailers included in this study, Reliance Digital also champs in satisfying more customers with its visual merchandising than its competitor included in the study (i.e. NEXT). 8.14 Customer satisfaction with the visual merchandising 303 Table 8.14: Comparative statistics of customer satisfaction with the visual merchandising Table 8.14: Comparative statistics of customer satisfaction with the visual merchandising Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 41 20.50 47 23.50 59 29.50 31 15.50 22 11.00 3.270 Star Bazaar 40 20.00 42 21.00 56 28.00 37 18.50 25 12.50 3.175 more.MEGAStore 38 19.00 45 22.50 56 28.00 36 18.00 25 12.50 3.175 Big Bazaar 42 21.00 46 23.00 60 30.00 32 16.00 20 10.00 3.290 D Mart 25 12.50 39 19.50 59 29.50 46 23.00 31 15.50 2.905 Big Megamart 49 24.50 53 26.50 56 28.00 26 13.00 16 8.00 3.465 West Side 47 23.50 49 24.50 57 28.50 29 14.50 18 9.00 3.390 Max 41 20.50 47 23.50 58 29.00 35 17.50 19 9.50 3.280 Reliance Digital 40 20.00 44 22.00 57 28.50 37 18.50 22 11.00 3.215 NEXT 26 13.00 40 20.00 58 29.00 47 23.50 29 14.50 2.935 Mean 38.90 19.45 45.20 22.60 57.60 28.80 35.60 17.80 22.70 11.35 3.210 Standard Deviation 7.810 3.905 4.211 2.106 1.430 0.715 6.769 3.385 4.809 2.404 0.393 304 8.15 Customer satisfaction with the methods of promotion and communication Figure 8.15: Weighted average scores (WAS) for customer satisfaction with the methods of promotion and communication Figure 8.15: Weighted average scores (WAS) for customer satisfaction with the methods of promotion and communication The mean WAS of all the organised retailers (included in this study) has been envisioned as the WAS of the industry for customer satisfaction with the methods of promotion and communication and shown with the red line in figure 8.15. This WAS of the industry crouches above the satisfied customer line in figure 8.15, which reports that the organised retail industry satisfies an affluent majority of customers with the methods of promotion and communication (i.e. advertising, sales promotion, direct retailing, personal retailing, public relation / publicity and the internet / interactive retailing). The WAS for customer satisfaction with the methods of promotion and communication of some retailers (i.e. Big Bazaar, D Mart, Max and NEXT) are greater than the WAS of the industry, which publicises that these retailers satisfy more customers with their methods of promotion and communication than others (included in this study) in the industry. The most (i.e. 8.14 Customer satisfaction with the visual merchandising 29.40%) customers are satisfied with the methods of promotion and communication adopted by the organised retailers (included in this study), as those are matching with their expectations. The second largest group (i.e. 23.45%) of 305 customers, who perceive that the methods of promotion and communication are exceeding their expectations and they are happy. The third major group (i.e. 20.55%) of customers is delighted with the methods of promotion and communication, as the same are greatly exceeding their expectations. However, there are also a massive number of customers, who are either dissatisfied (i.e. 16.30%) or disappointed (i.e. 10.33%) with the methods of promotion and communication, because, those are less than or much less than their expectations. Table 8.15: Comparative statistics of customer satisfaction with the methods of promotion and communication Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 41 20.50 44 22.00 55 27.50 37 18.50 23 11.50 3.215 Star Bazaar 39 19.50 48 24.00 61 30.50 33 16.50 19 9.50 3.275 more.MEGAStore 41 20.50 44 22.00 59 29.50 35 17.50 21 10.50 3.245 Big Bazaar 47 23.50 51 25.50 62 31.00 25 12.50 15 7.50 3.450 D Mart 40 20.00 47 23.50 66 33.00 27 13.50 20 10.00 3.300 Big Megamart 39 19.50 49 24.50 58 29.00 33 16.50 21 10.50 3.260 West Side 40 20.00 45 22.50 56 28.00 35 17.50 24 12.00 3.210 Max 43 21.50 50 25.00 57 28.50 31 15.50 19 9.50 3.335 Reliance Digital 39 19.50 45 22.50 56 28.00 36 18.00 24 12.00 3.195 NEXT 42 21.00 46 23.00 58 29.00 34 17.00 20 10.00 3.280 Mean 41.10 20.55 46.90 23.45 58.80 29.40 32.60 16.30 20.60 10.30 3.277 Standard Deviation 2.470 1.235 2.514 1.257 3.360 1.680 3.893 1.947 2.716 1.358 0.215 306 Big Bazaar is a leader in satisfying more customers with its methods of promotion and communication in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, Max leads in satisfying more customers with its methods of promotion and communication in comparison to its competitors or other fashion item retailers included in the study (i.e. 8.14 Customer satisfaction with the visual merchandising Big Megamart and West Side) and in the case of electronics item retailers included in this study, NEXT champs in satisfying more customers with its methods of promotion and communication than its competitor included in the study (i.e. Reliance Digital). 8.16 Customer satisfaction with the advertising methods Figure 8.16: Weighted average scores (WAS) for customer satisfaction with the advertising methods 8.16 Customer satisfaction with the advertising methods Figure 8.16: Weighted average scores (WAS) for customer satisfaction with the advertising methods The mean WAS of all the organised retailers (included in this study) has been speculated as the WAS of the industry for customer satisfaction with the advertising methods and exhibited with the red line in figure 8.16. This WAS of the industry reclines above the satisfied customer line in figure 8.16, which apprises that the organised retail industry satisfies an immense majority of customers with its advertising methods (i.e. in-store advertising methods and out-store advertising methods). The in-store advertising methods comprise the visual display, signage, pamphlets, the point of purchase material, secondary packaging and cross 307 promotions. And out-store advertising methods comprise the television ads, radio ads, newspaper ads, journals & magazine ads, yellow pages, hoarding, wall paintings, pamphlets/propaganda, van advertising, online advertising, mobile billboard, LCD screening, cinema, fairs & exhibitions. The WAS for customer satisfaction with the advertising methods of some retailers (i.e. HyperCITY, more.MEGAStore, Big Bazaar, Big Megamart, West Side and Reliance Digital) are more than the WAS of the industry, which annunciates that these retailers satisfy more customers with their advertising methods than others (included in this study) in the industry. 8.14 Customer satisfaction with the visual merchandising Table 8.16: Comparative statistics of customer satisfaction with the advertising methods Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 40 20.00 47 23.50 56 28.00 36 18.00 21 21 3.245 Star Bazaar 38 19.00 42 21.00 58 29.00 39 19.50 29 23 3.195 more.MEGAStore 39 19.50 45 22.50 57 28.50 37 18.50 23 22 3.215 Big Bazaar 46 23.00 52 26.00 62 31.00 24 12.00 16 16 3.440 D Mart 32 16.00 40 20.00 56 28.00 44 22.00 30 28 3.030 Big Megamart 42 21.00 45 22.50 57 28.50 34 17.00 22 22 3.255 West Side 41 20.50 44 22.00 59 29.50 36 18.00 21 20 3.255 Max 40 20.00 44 22.00 57 28.50 35 17.50 25 24 3.210 Reliance Digital 40 20.00 46 23.00 63 31.50 30 15.00 21 21 3.270 NEXT 34 17.00 39 19.50 54 27.00 43 21.50 30 30 3.020 Mean 39.20 19.60 44.40 22.20 57.90 28.95 35.80 17.90 23.80 11.90 3.214 Standard Deviation 3.938 1.969 3.688 1.844 2.767 1.383 5.846 2.923 4.638 2.319 0.295 Table 8.16: Comparative statistics of customer satisfaction with the advertising methods Responses on expectation cum satisfaction scale 308 The industry satisfies the massive (i.e. 28.95%) customers with the advertising methods by matching the same with their expectations. The industry also has happy customers with the visual merchandising, as a second basic group (i.e. 22.20%), which feels that the advertising methods are exceeding their expectations. Hereto, the industry also delights a great number (i.e. 19.60%) of customers with the advertising methods by greatly exceeding the same than customers’ expectations. However, the industry endures the next large group (i.e. 17.90%) of dissatisfied customers with the advertising methods, as infantile to meet expectations. Moreover, the industry hauls a non-negligible number (i.e. 11.90%) of disappointed customers with the advertising methods, as perform much less than expectations of customers. Big Bazaar is also a leader in satisfying more customers with its advertising methods in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, Big Megamart and West Side stand at an equal position to satisfy more customers with its advertising methods in comparison to its competitors or other fashion item retailers included in the study (i.e. 8.14 Customer satisfaction with the visual merchandising Max) and in the case of electronics item retailers included in this study, Reliance Digital leads in satisfying more customers with its advertising methods than its competitor included in the study (i.e. NEXT). 8.17 Customer satisfaction with the sales promotion methods Figure 8.17: Weighted average scores (WAS) for customer satisfaction with the sales promotion methods 8.17 Customer satisfaction with the sales promotion methods Figure 8.17: Weighted average scores (WAS) for customer satisfaction with the sales promotion methods 309 The mean WAS of all the organised retailers (included in this study) has been figured as the WAS of the industry for customer satisfaction with the sales promotion methods and documented with the red line in figure 8.17. This WAS of the industry inclines above the satisfied customer line in figure 8.17, which notifies that the organised retail industry satisfies a hulking majority of customers with its sales promotion methods (i.e. schemes, free samples, gifts, rebates & discounts, contests and other activities, customer loyalty programs, premiums, prize & rewards). The WAS for customer satisfaction with the sales promotion methods of some retailers (i.e. Big Bazaar and NEXT) are greater than the WAS of the industry, which pronounces that these retailers satisfy more customers with their sales promotional methods than others (included in this study) in the industry. The maximal (i.e. 29.30%) customers are satisfied with sales promotion methods adopted by the organised retailers (included in this study), as those are matching with their expectations. The second verdant group (i.e. 23.80%) of customers, who deem that sales promotion methods are exceeding their expectations and they are happy. The industry also delights an ample number (i.e. 20.85%) of customers with sales promotion methods by greatly exceeding them than customers’ expectations. The next gigantic group (i.e. 16.15%) of customers is dissatisfied with sales promotion methods because those are deficient to their expectations. Moreover, there is also a non-negligible number (i.e. 9.80%) of customers, those are disappointed with sales promotion methods as the same is much less than their expectations. Big Bazaar also leads in satisfying more customers with its sales promotion methods in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, West Side leads in satisfying more customers with its sales promotion methods in comparison to its competitors or other fashion item retailers included in the study (i.e. Big Megamart and Max) and in the case of electronics item retailers included in this study, NEXT leads in satisfying more customers with its sales promotion methods than its competitor included in the study (i.e. Reliance Digital). 8.14 Customer satisfaction with the visual merchandising 310 Table 8.17: Comparative statistics of customer satisfaction with the sales promotion methods Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 42 21.00 45 22.50 60 30.00 35 17.50 18 9.00 3.290 Star Bazaar 40 20.00 46 23.00 62 31.00 32 16.00 20 10.00 3.270 more.MEGAStore 42 21.00 47 23.50 58 29.00 34 17.00 19 9.50 3.295 Big Bazaar 49 24.50 56 28.00 63 31.50 20 10.00 12 6.00 3.550 D Mart 42 21.00 48 24.00 58 29.00 30 15.00 22 11.00 3.290 Big Megamart 40 20.00 46 23.00 55 27.50 35 17.50 24 12.00 3.215 West Side 41 20.50 47 23.50 58 29.00 36 18.00 18 9.00 3.285 Max 39 19.50 47 23.50 59 29.50 35 17.50 20 10.00 3.250 Reliance Digital 39 19.50 46 23.00 57 28.50 36 18.00 22 11.00 3.220 NEXT 43 21.50 48 24.00 58 29.00 30 15.00 21 10.50 3.310 Mean 41.70 20.85 47.60 23.80 58.80 29.40 32.30 16.15 19.60 9.80 3.298 Standard Deviation 2.908 1.454 3.098 1.549 2.348 1.174 4.877 2.439 3.273 1.636 0.235 Table 8.17: Comparative statistics of customer satisfaction with the sales promotion methods While, West Side leads in satisfying more customers with its sales promotion methods in comparison to its competitors or other fashion item retailers included in the study (i.e. Big Megamart and Max) and in the case of electronics item retailers included in this study, NEXT leads in satisfying more customers with its sales promotion methods than its competitor included in the study (i.e. Reliance Digital). 311 8.18 Customer satisfaction with the direct retailing methods Figure 8.18: Weighted average scores (WAS) for customer satisfaction with the direct retailing methods Figure 8.18: Weighted average scores (WAS) for customer satisfaction with the direct retailing methods The mean WAS of all the organised retailers (included in this study) has been described as the WAS of the industry for customer satisfaction with the direct retailing methods and demonstrated by the red line in figure 8.18. This WAS of the industry curves underneath the satisfied customer line in figure 8.18, which denotes that the organised retail industry do not satisfy a giant majority of customers with its direct retailing methods (i.e. direct mail, tele-retailing, direct response on TV/radio, mail order catalogues, SMS retailing and mobile retailing). 8.14 Customer satisfaction with the visual merchandising The WAS for customer satisfaction with the direct retailing methods of some retailers (i.e. HyperCITY, Big Bazaar, West Side, Reliance Digital and NEXT) are more than the WAS of the industry, which notifies that these retailers satisfy more customers with their direct retailing methods than others (included in this study) in the industry. The industry satisfies the extensive group (i.e. 27.15%) of customers with the direct retailing methods by matching the same with their expectations. However, the industry bears the next large group (i.e. 21.25%) of dissatisfied customers with the direct retailing methods, as immature to meet expectations. The industry also has 312 happy customers with the direct retailing methods, as a third fundamental group (i.e. 20.70%), which feels that the direct retailing methods are exceeding their expectations. Moreover, the industry ferriages a non-negligible number (i.e. 16.00%) of disappointed customers with the direct retailing methods, as maintains much less than expectations of customers. Herein, the industry also delights an effectual number (i.e. 14.90%) of customers with the direct retailing methods by greatly exceeding the same than customers’ expectations. Table 8.18: Comparative statistics of customer satisfaction with the direct retailing methods Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 38 19.00 47 23.50 55 27.50 36 18.00 24 12.00 3.195 Star Bazaar 19 9.50 36 18.00 55 27.50 50 25.00 40 20.00 2.720 more.MEGAStore 20 10.00 34 17.00 53 26.50 51 25.50 42 21.00 2.695 Big Bazaar 40 20.00 46 23.00 60 30.00 33 16.50 21 10.50 3.255 D Mart 18 9.00 35 17.50 49 24.50 55 27.50 43 21.50 2.650 Big Megamart 27 13.50 45 22.50 50 25.00 46 23.00 32 16.00 2.945 West Side 40 20.00 44 22.00 58 29.00 36 18.00 22 11.00 3.220 Max 17 8.50 37 18.50 48 24.00 51 25.50 47 23.50 2.630 Reliance Digital 39 19.50 46 23.00 57 28.50 33 16.50 25 12.50 3.205 NEXT 40 20.00 44 22.00 58 29.00 34 17.00 24 12.00 3.210 Mean 29.80 14.90 41.40 20.70 54.30 27.15 42.50 21.25 32.00 16.00 2.973 Standard Deviation 10.475 5.238 5.211 2.606 4.165 2.082 8.860 4.430 10.044 5.022 0.567 313 Big Bazaar satisfies more customers with its direct retailing methods in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). 8.14 Customer satisfaction with the visual merchandising While, West Side also satisfies more customers with its direct retailing methods in comparison to its competitors or other fashion item retailers included in the study (i.e. Big Megamart and Max) and in the case of electronics item retailers included in this study, NEXT also satisfies more customers with its direct retailing methods than its competitor included in the study (i.e. Reliance Digital). 8.19 Customer satisfaction with the personal retailing methods Figure 8.19: Weighted average scores (WAS) for customer satisfaction with the personal retailing methods 8.19 Customer satisfaction with the personal retailing methods Figure 8.19: Weighted average scores (WAS) for customer satisfaction with the personal retailing methods The mean WAS of all the organised retailers (included in this study) has been reckoned as the WAS of the industry for customer satisfaction with the personal retailing methods and scripted with the red line in figure 8.19. This WAS of the industry falls above the satisfied customer line in figure 8.19, which enlightens that the organised retail industry satisfies a gigantic majority of customers with its personal retailing methods (i.e. face to face retailing in store, person to person telephonic conversation, etc.). 314 The WAS for customer satisfaction with the personal retailing methods of some retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar, Big Megamart, West Side and Reliance Digital) are greater than the WAS of the industry, which registers that these retailers satisfy more customers with their personal retailing methods than others (included in this study) in the industry. The WAS for customer satisfaction with the personal retailing methods of some retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar, Big Megamart, West Side and Reliance Digital) are greater than the WAS of the industry, which registers that these retailers satisfy more customers with their personal retailing methods than others (included in this study) in the industry. 8.14 Customer satisfaction with the visual merchandising Table 8.19: Comparative statistics of customer satisfaction with the personal retailing methods Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 43 21.50 43 21.50 58 29.00 34 17.00 22 11.00 3.255 Star Bazaar 39 19.50 47 23.50 60 30.00 33 16.50 21 10.50 3.250 more.MEGAStore 38 19.00 46 23.00 54 27.00 38 19.00 24 12.00 3.180 Big Bazaar 44 22.00 46 23.00 57 28.50 31 15.50 22 11.00 3.295 D Mart 27 13.50 41 20.50 55 27.50 47 23.50 30 15.00 2.940 Big Megamart 41 20.50 46 23.00 53 26.50 35 17.50 25 12.50 3.215 West Side 43 21.50 48 24.00 55 27.50 33 16.50 21 10.50 3.295 Max 29 14.50 37 18.50 53 26.50 48 24.00 33 16.50 2.905 Reliance Digital 40 20.00 46 23.00 59 29.50 33 16.50 22 11.00 3.245 NEXT 29 14.50 39 19.50 59 29.50 43 21.50 30 15.00 2.970 Mean 37.30 18.65 43.90 21.95 56.30 28.15 37.50 18.75 25.00 12.50 3.155 Standard Deviation 6.482 3.241 3.725 1.863 2.627 1.313 6.258 3.129 4.397 2.198 0.361 Table 8.19: Comparative statistics of customer satisfaction with the personal retailing me 315 The maximum (i.e. 28.15%) customers are satisfied with the personal retailing methods adopted by the organised retailers (included in this study), as those are matching with their expectations. The second giant group (i.e. 21.95%) of customers, who comprehend that the personal retailing methods are exceeding their expectations and they are happy. The next plenteous group (i.e. 18.75%) of customers is dissatisfied with the personal retailing methods because these are lacking to their expectations. Moreover, the industry has also an excessive number (i.e. 18.65%) of delighted customers with the personal retailing methods by greatly exceeding the same than their expectations. However, there is also a non-negligible disappointed (i.e. 12.50%) customers with the personal retailing methods, as those are much less than their expectations. Big Bazaar also satisfies more customers with its personal retailing methods in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, West Side is also a better performer in satisfying more customers with its personal retailing methods in comparison to its competitors or other fashion item retailers included in the study (i.e. 8.14 Customer satisfaction with the visual merchandising Big Megamart and Max) and in the case of electronics item retailers included in this study, Reliance Digital is a better performer in satisfying more customers with its personal retailing methods than its competitor included in the study (i.e. NEXT). 8.20 Customer satisfaction with the public relation/publicity methods The mean WAS of all the organised retailers (included in this study) has been depicted as the WAS of the industry for customer satisfaction with the public relation/publicity methods and pointed by the red line in the figure 8.20. This WAS of the industry furls beneath the satisfied customer line in the figure 8.20, which denotes that the organised retail industry do not satisfy a significant majority of customers with its public relation / publicity methods (i.e. corporate brochures, posters, annual reports, press release, press conference, video press conference, internet release, store tours for media, meeting with customers, special events & sponsorships, social events, social development programs). 316 Figure 8.20: Weighted average scores (WAS) for customer satisfaction with the public relation/publicity methods Figure 8.20: Weighted average scores (WAS) for customer satisfaction with the public relation/publicity methods The WAS for customer satisfaction with the public relation/publicity methods of some retailers (i.e. HyperCITY, more.MEGAStore, Big Bazaar, Big Megamart, West Side, Max and Reliance Digital) are more than the WAS of the industry, which records that these retailers satisfy more customers with their public relation/publicity methods than others (included in this study) in the industry. The industry satisfies the exhaustive (i.e. 26.35%) customers with the public relation / publicity methods by matching the same with their expectations. However, the industry totes the next large group (i.e. 21.00%) of dissatisfied customers with public relation / publicity methods, as imperfect to meet expectations. The industry also has happy customers with the public relation / publicity methods, as a third key group (i.e. 20.40%), which feels that the public relation / publicity methods are exceeding their expectations. Hereto, the industry also delights a competent number (i.e. 16.25%) of customers with the public relation / publicity methods by greatly exceeding the same than customers’ expectations. Moreover, the industry haulages a non-negligible high number (i.e. 16.00%) of disappointed customers with public relation / publicity methods, as tender much less than expectations of customers. 8.20 Customer satisfaction with the public relation/publicity methods 317 Table 8.20: Comparative statistics of customer satisfaction with the public relation/publicity methods Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 38 19.00 48 24.00 57 28.50 35 17.50 22 11.00 3.225 Star Bazaar 22 11.00 33 16.50 59 29.50 47 23.50 39 19.50 2.760 more.MEGAStore 40 20.00 46 23.00 58 29.00 34 17.00 22 11.00 3.240 Big Bazaar 42 21.00 47 23.50 61 30.50 32 16.00 18 9.00 3.315 D Mart 14 7.00 21 10.50 34 17.00 69 34.50 62 31.00 2.280 Big Megamart 39 19.50 47 23.50 54 27.00 36 18.00 24 12.00 3.205 West Side 38 19.00 48 24.00 55 27.50 33 16.50 26 13.00 3.195 Max 37 18.50 50 25.00 57 28.50 34 17.00 22 11.00 3.230 Reliance Digital 40 20.00 44 22.00 59 29.50 33 16.50 24 12.00 3.215 NEXT 15 7.50 24 12.00 33 16.50 67 33.50 61 30.50 2.325 Mean 32.50 16.25 40.80 20.40 52.70 26.35 42.00 21.00 32.00 16.00 2.999 Standard Deviation 10.977 5.489 10.737 5.369 10.318 5.159 14.353 7.176 16.499 8.250 0.870 Big Bazaar also leads in satisfying more customers with its public relation / publicity methods in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, Big Megamart is a better performer in satisfying more customers with its public relation / publicity methods in comparison to its competitors or other fashion item retailers included in the study (i.e. West Side and Max) and in the case of electronics item retailers included in this study, Reliance Digital is also a better performer in satisfying more customers with its public relation / publicity methods than its competitor included in the study (i.e. NEXT). 318 8.21 Customer satisfaction with the internet/interactive retailing methods Figure 8.21: Weighted average scores (WAS) for customer satisfaction with the internet/interactive retailing methods Figure 8.21: Weighted average scores (WAS) for customer satisfaction with the internet/interactive retailing methods The mean WAS of all the organised retailers (included in this study) has been considered as the WAS of the industry for customer satisfaction with the internet / interactive retailing methods and graphed with the red line in figure 8.21. 8.20 Customer satisfaction with the public relation/publicity methods This WAS of the industry inclines beyond the satisfied customer line in figure 8.21, which intimates that the organised retail industry satisfies a large majority of customers with its internet / interactive retailing methods (i.e. e-mails, online discussion forums, weblogs, websites, e-pay, social networking sites). The WAS for customer satisfaction with the internet / interactive retailing methods of some retailers (i.e. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar, Big Megamart, West Side, Max and Reliance Digital) are greater than the WAS of the industry, which inscribes that these retailers satisfy more customers with their internet / interactive retailing methods than others (included in this study) in the industry. The utmost (i.e. 27.90%) customers are satisfied with the internet / interactive retailing methods adopted by the organised retailers (included in this study). The second exuberant group (i.e. 22.80%) of customers, who understand that the 319 internet / interactive retailing methods are exceeding their expectations and they are happy. Moreover, the industry has also an intense number (i.e. 19.90%) of delighted customers with the internet / interactive retailing methods. The next enormous group (i.e. 17.60%) of customers is dissatisfied with the internet / interactive retailing methods. However, there is also a significant number (i.e. 11.80%) of disappointed customers with the internet / interactive retailing methods adopted by industry. 8.20 Customer satisfaction with the public relation/publicity methods Table 8.21: Comparative statistics of customer satisfaction with the internet/interactive retailing methods Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 44 22.00 47 23.50 53 26.50 36 18.00 20 10.00 3.295 Star Bazaar 40 20.00 48 24.00 56 28.00 38 19.00 18 9.00 3.270 more.MEGAStore 39 19.50 48 24.00 59 29.50 34 17.00 20 10.00 3.260 Big Bazaar 46 23.00 48 24.00 62 31.00 28 14.00 16 8.00 3.400 D Mart 18 9.00 31 15.50 44 22.00 58 29.00 49 24.50 2.555 Big Megamart 44 22.00 45 22.50 53 26.50 35 17.50 23 11.50 3.260 West Side 40 20.00 47 23.50 53 26.50 36 18.00 24 12.00 3.215 Max 43 21.50 49 24.50 61 30.50 29 14.50 18 9.00 3.350 Reliance Digital 45 22.50 47 23.50 62 31.00 26 13.00 20 10.00 3.355 NEXT 39 19.50 46 23.00 55 27.50 32 16.00 28 14.00 3.180 Mean 39.80 19.90 45.60 22.80 55.80 27.90 35.20 17.60 23.60 11.80 3.214 Standard Deviation 8.080 4.040 5.254 2.627 5.554 2.777 8.917 4.458 9.571 4.785 0.527 320 Big Bazaar leads in satisfying more customers with its internet / interactive retailing methods in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, Max is a better performer in satisfying more customers with its internet / interactive retailing methods in comparison to its competitors or other fashion item retailers included in the study (i.e. West Side and Big Megamart) and in the case of electronics item retailers included in this study, Reliance Digital is also a better performer in satisfying more customers with its internet / interactive retailing methods than its competitor included in the study (i.e. NEXT). 8.22 Customer satisfaction with the sales personnel Figure 8.22: Weighted average scores (WAS) for customer satisfaction with the sales personnel 8.22 Customer satisfaction with the sales personnel Figure 8.22: Weighted average scores (WAS) for customer satisfaction with the sales personnel The mean WAS of all the organised retailers (included in this study) has been presumed as the WAS of the industry for customer satisfaction with the sales personnel and displayed with the red line in figure 8.22. 8.20 Customer satisfaction with the public relation/publicity methods This WAS of the industry leans heavily beyond the satisfied customer line in figure 8.22, which furnishes that the organised retail industry satisfies a great majority of customers with its sales personnel (i.e. sales boys and sales girls). 321 The WAS for customer satisfaction with the sales personnel of some retailers (i.e. HyperCITY, Star Bazaar, Big Bazaar, Big Megamart, West Side and Reliance Digital) are more than the WAS of the industry, which limns that these retailers satisfy more customers with their sales personnel than others (included in this study) in the industry. The WAS for customer satisfaction with the sales personnel of some retailers (i.e. HyperCITY, Star Bazaar, Big Bazaar, Big Megamart, West Side and Reliance Digital) are more than the WAS of the industry, which limns that these retailers satisfy more customers with their sales personnel than others (included in this study) in the industry. Table 8.22: Comparative statistics of customer satisfaction with the sales personnel Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 45 22.50 49 24.50 56 28.00 31 15.50 19 9.50 3.350 Star Bazaar 42 21.00 48 24.00 55 27.50 33 16.50 22 11.00 3.275 more.MEGAStore 40 20.00 45 22.50 57 28.50 36 18.00 22 11.00 3.225 Big Bazaar 48 24.00 52 26.00 61 30.50 26 13.00 13 6.50 3.480 D Mart 21 10.50 33 16.50 51 25.50 56 28.00 39 19.50 2.705 Big Megamart 44 22.00 50 25.00 58 29.00 28 14.00 20 10.00 3.350 West Side 42 21.00 52 26.00 60 30.00 29 14.50 17 8.50 3.365 Max 29 14.50 42 21.00 56 28.00 42 21.00 31 15.50 2.980 Reliance Digital 43 21.50 48 24.00 58 29.00 31 15.50 20 10.00 3.315 NEXT 40 20.00 46 23.00 54 27.00 36 18.00 24 12.00 3.210 Mean 39.40 19.70 46.50 23.25 56.60 28.30 34.80 17.40 22.70 11.35 3.226 Standard Deviation 8.168 4.084 5.662 2.831 2.914 1.457 8.779 4.389 7.394 3.697 0.486 Table 8.22: Comparative statistics of customer satisfaction with the sales personnel 322 The industry satisfies the utmost (i.e. 28.30%) customers with the sales personnel by matching the same with their expectations. The industry also has happy customers with the sales personnel, as a second major group (i.e. 23.25%), which feels that the sales personnel are exceeding their expectations. 8.20 Customer satisfaction with the public relation/publicity methods Moreover, the industry also delights a substantial number (i.e. 19.70%) of customers with the sales personnel by greatly exceeding the performance than customers’ expectations. However, the industry carries the next major group (i.e. 17.40%) of dissatisfied customers with the sales personnel, as lacking to meet expectations of customers. At the same time, the industry hauls a non-negligible number (i.e. 11.35%) of disappointed customers with the sales personnel, as performed much less than expectations of customers. Big Bazaar also leads in satisfying more customers with its sales personnel in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, West Side is a better performer in satisfying more customers with its sales personnel in comparison to its competitors or other fashion item retailers included in the study (i.e. Max and Big Megamart) and in the case of electronics item retailers included in this study, Reliance Digital is also a better performer in satisfying more customers with its sales personnel than its competitor included in the study (i.e. NEXT). 8.23 Customer satisfaction with the contact personnel 8.23 Customer satisfaction with the contact personnel 8.23 Customer satisfaction with the contact personnel Figure 8.23: Weighted average scores (WAS) for customer satisfaction with the sales personnel Figure 8.23: Weighted average scores (WAS) for customer satisfaction with the sales personnel 323 The mean WAS of all the organised retailers (included in this study) has been recognised as the WAS of the industry for customer satisfaction with the contact personnel and demonstrated by the red line in figure 8.23. This WAS of the industry bows beyond the satisfied customer line in figure 8.23, which imparts that the organised retail industry satisfies a huge majority of customers with its contact personnel (i.e. help desk people, service desk people, billing counter people, etc.). The WAS for customer satisfaction with the contact personnel of some retailers (i.e. HyperCITY, Big Bazaar, Big Megamart, West Side and Reliance Digital) are greater than the WAS of the industry, which pens that these retailers satisfy more customers with their contact personnel than others (included in this study) in the industry. The majority (i.e. 28.30%) customers are satisfied with the contact personnel of the organised retailers (included in this study), as those are matching with their expectations. The second plenteous group (i.e. 23.45%) of customers, who ponder that the contact personnel are exceeding their expectations and they are happy. The third adjunct group (i.e. 20.20%) of customers is delighted with the contact personnel because the performance of the same is greatly exceeding their expectations. However, there are also a pervasive number of customers, who are either dissatisfied (i.e. 16.70%) or disappointed (i.e. 11.35%) with the contact personnel, because, those are less than or much less than their expectations. Big Bazaar also leads in satisfying more customers with its contact personnel in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). While, West Side is also a better performer in satisfying more customers with its contact personnel in comparison to its competitors or other fashion item retailers included in the study (i.e. Max and Big Megamart) and in the case of electronics item retailers included in this study, Reliance Digital also leads in satisfying more customers with its contact personnel than its competitor included in the study (i.e. NEXT). 8.23 Customer satisfaction with the contact personnel 324 Table 8.23: Comparative statistics of customer satisfaction with the contact personnel Table 8.23: Comparative statistics of customer satisfaction with the contact personnel Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 43 21.50 45 22.50 54 27.00 35 17.50 23 11.50 3.250 Star Bazaar 40 20.00 46 23.00 57 28.50 35 17.50 22 11.00 3.235 more.MEGAStore 37 18.50 44 22.00 57 28.50 37 18.50 25 12.50 3.155 Big Bazaar 46 23.00 53 26.50 60 30.00 27 13.50 14 7.00 3.450 D Mart 31 15.50 42 21.00 54 27.00 40 20.00 33 16.50 2.990 Big Megamart 47 23.50 51 25.50 59 29.50 25 12.50 18 9.00 3.420 West Side 46 23.00 53 26.50 57 28.50 28 14.00 16 8.00 3.425 Max 40 20.00 46 23.00 58 29.00 33 16.50 23 11.50 3.235 Reliance Digital 42 21.00 49 24.50 57 28.50 30 15.00 22 11.00 3.295 NEXT 32 16.00 40 20.00 53 26.50 44 22.00 31 15.50 2.990 Mean 40.40 20.20 46.90 23.45 56.60 28.30 33.40 16.70 22.70 11.35 3.245 Standard Deviation 5.641 2.821 4.483 2.242 2.271 1.135 6.022 3.011 6.001 3.000 0.355 325 8.24 Customer satisfaction with the factors pertained to personnel Figure 8.24: Weighted average scores (WAS) for customer satisfaction with the factors pertained to personnel Figure 8.24: Weighted average scores (WAS) for customer satisfaction with the factors pertained to personnel The mean WAS of all the organised retailers (included in this study) has been admitted as the WAS of the industry for customer satisfaction with the factors pertained to personnel and striated with the red line in figure 8.24. This WAS of the industry slouches beyond the satisfied customer line in the figure 8.24, which informs that the organised retail industry satisfies an enormous majority of customers with factors pertained to personnel (i.e. mood, efforts, commitment, attitude, knowledge, skills, behaviour, approach to deliver the service, competence, courtesy, credibility, reliability, responsiveness, communication style, friendliness, helpfulness, discipline & punctuality, customer counselling ability). The WAS for customer satisfaction with the factors pertained to personnel of some retailers (i.e. HyperCITY, Big Bazaar, Big Megamart, West Side and Reliance Digital) are more than the WAS of the industry, which shapes that these retailers satisfy more customers with the factors pertained to personnel than others (included in this study) in the industry. The maximum (i.e. 8.23 Customer satisfaction with the contact personnel 28.75%) customers are satisfied with the factors pertained to personnel of the organised retailers (included in this study), as those are matching 326 with their expectations. The second exuberant group (i.e. 23.55%) of customers, who perceive that the factors pertained to personnel are exceeding their expectations and they are happy. with their expectations. The second exuberant group (i.e. 23.55%) of customers, who perceive that the factors pertained to personnel are exceeding their expectations and they are happy. Table 8.24: Comparative statistics of customer satisfaction with the factors pertained to personnel Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 46 23.00 53 26.50 60 30.00 27 13.50 14 7.00 3.450 Star Bazaar 40 20.00 42 21.00 57 28.50 38 19.00 23 11.50 3.190 more.MEGAStore 39 19.50 43 21.50 56 28.00 37 18.50 25 12.50 3.170 Big Bazaar 48 24.00 55 27.50 61 30.50 25 12.50 11 5.50 3.520 D Mart 29 14.50 42 21.00 56 28.00 42 21.00 31 15.50 2.980 Big Megamart 48 24.00 54 27.00 60 30.00 24 12.00 14 7.00 3.490 West Side 44 22.00 52 26.00 59 29.50 28 14.00 17 8.50 3.390 Max 31 15.50 41 20.50 51 25.50 45 22.50 32 16.00 2.970 Reliance Digital 40 20.00 47 23.50 61 30.50 31 15.50 21 10.50 3.270 NEXT 30 15.00 42 21.00 54 27.00 41 20.50 33 16.50 2.975 Mean 39.50 19.75 47.10 23.55 57.50 28.75 33.80 16.90 22.10 11.05 3.241 Standard Deviation 7.307 3.653 5.782 2.891 3.308 1.654 7.700 3.850 8.075 4.038 0.465 327 Moreover, the industry also delights an adequate number (i.e. 19.75%) of customers with the factors pertained to personnel by greatly exceeding the same than customers’ expectations. The next enormous group (i.e. 16.90%) of customers are dissatisfied with the factors pertained to personnel because those are deficient to their expectations. There is also a non-negligible number (i.e. 11.05%) of customers, those are disappointed with the factors pertained to personnel, as those are much less than their expectations. Big Bazaar champs in satisfying more customers with the factors pertained to personnel in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). 8.23 Customer satisfaction with the contact personnel While, Big Megamart is a better performer in satisfying more customers with the factors pertained to personnel in comparison to its competitors or other fashion item retailers included in the study (i.e. Max and West Side) and in the case of electronics item retailers included in this study, Reliance Digital also champs in satisfying more customers with the factors pertained to personnel than its competitor included in the study (i.e. NEXT). 8.25 Customer satisfaction with the other tangible evidence Figure 8.25: Weighted average scores (WAS) for customer satisfaction with the other tangible evidence 8.25 Customer satisfaction with the other tangible evidence Figure 8.25: Weighted average scores (WAS) for customer satisfaction with the other tangible evidence 328 The mean WAS of all the organised retailers (included in this study) has been conceded as the WAS of the industry for customer satisfaction with the other tangible evidence and drawn with the red line in the figure 8.25. This WAS of the industry crouches beyond the satisfied customer line in the figure 8.25, which confabulates that the organised retail industry satisfies a giant majority of customers with other tangible evidence (i.e. employee appearance, uniforms, billing statements, stationery, business cards, signage, exterior design, reports, brochures, the internet or web pages). The WAS for customer satisfaction with the other tangible evidences of some retailers (i.e. HyperCITY, Star Bazaar, Big Bazaar, Big Megamart, West Side and Reliance Digital) are greater than the WAS of the industry, which indites that these retailers satisfy more customers with the other tangible evidence than others (included in this study) in the industry. The industry satisfies the utmost (i.e. 28.90%) customers with the other tangible evidence by matching the same with their expectations. The industry also has happy customers with the other tangible evidence, as a second major group (i.e. 24.25%), which feels that the other tangible evidence are exceeding their expectations. Moreover, the industry also delights a substantial number (i.e. 19.15%) of customers with the other tangible evidence by greatly exceeding the same than customers’ expectations. However, the industry carries the next major group (i.e. 16.50%) of dissatisfied customers with the other tangible evidence, as lacking to meet expectations of customers. At the same time, the industry hauls a non-negligible number (i.e. 11.20%) of disappointed customers with the other tangible evidence, as provided much less than expectations of customers. 8.23 Customer satisfaction with the contact personnel Big Bazaar also champs in satisfying more customers with the other tangible evidence in comparison to its competitors or other multi-format retailers included in the study (i.e. HyperCITY, Star Bazaar, more.MEGAStore and D Mart). 329 Table 8.25: Comparative statistics of customer satisfaction with the other tangible evidence Retailers Responses on expectation cum satisfaction scale 5 4 3 2 1 Weighted Average Score (WAS) f % f % f % f % f % HyperCITY 41 20.50 52 26.00 57 28.50 30 15.00 20 10.00 3.320 Star Bazaar 42 21.00 53 26.50 55 27.50 31 15.50 19 9.50 3.340 more.MEGAStore 37 18.50 45 22.50 56 28.00 37 18.50 25 12.50 3.160 Big Bazaar 40 20.00 57 28.50 59 29.50 26 13.00 18 9.00 3.375 D Mart 27 13.50 40 20.00 58 29.00 45 22.50 30 15.00 2.945 Big Megamart 45 22.50 55 27.50 59 29.50 26 13.00 15 7.50 3.445 West Side 40 20.00 54 27.00 61 30.50 26 13.00 19 9.50 3.350 Max 40 20.00 42 21.00 57 28.50 36 18.00 25 12.50 3.180 Reliance Digital 43 21.50 45 22.50 59 29.50 33 16.50 20 10.00 3.290 NEXT 28 14.00 42 21.00 57 28.50 40 20.00 33 16.50 2.960 Mean 38.30 19.15 48.50 24.25 57.80 28.90 33.00 16.50 22.40 11.20 3.237 Standard Deviation 6.075 3.037 6.311 3.156 1.751 0.876 6.481 3.240 5.700 2.850 0.398 While, Big Megamart is also a better performer in satisfying more customers with the other tangible evidence in comparison to its competitors or other fashion item retailers included in the study (i.e. Max and West Side) and in the case of electronics item retailers included in this study, Reliance Digital also leads in satisfying more customers with the other tangible evidences than its competitor included in the study (i.e. NEXT). 330 Introduction The mentioned below null hypotheses were formulated based on gaps in the literature review and objectives of the study. These hypotheses were formulated to study whether any significance exists of various strategies of organised retailers for significantly satisfying customers with various products & services. H01: Retailing strategies introduced by organised retailers are not significant to sustain in the hyper-competitive retail market. H01: Retailing strategies introduced by organised retailers are not significant to sustain in the hyper-competitive retail market. H02: Customers are not significantly satisfied with various products & services introduced by organised retailers. H02: Customers are not significantly satisfied with various products & services introduced by organised retailers. For the purpose of statistical analyses, Kolmogorov-Smirnov one-sample test and one sample t-test have been identified to test the hypotheses. Testing of hypothesis H01 The conventional method for testing the hypothesis is to formulate the two hypotheses instead of one (Beri, 2000). The first is null hypothesis and second is an alternate hypothesis. Hence, the alternate hypothesis of H01 has also been formulated as mentioned below. HA1: Retailing strategies introduced by organised retailers are significant to sustain in the hyper-competitive retail market. HA1: Retailing strategies introduced by organised retailers are significant to sustain in the hyper-competitive retail market. This hypothesis was formulated under certain assumptions for which no validation required as it is a common noticeable phenomenon that “retailing strategies introduced by organised retailers are significant to sustain in hyper-competitive retail market”. However, this statement is supported by product portfolio analysis (presented in chapter 5) where the significant majority of merchandise (products) of all the organised retailers (included in this study) belongs to Star category in the BCG matrix. Moreover, this alternate hypothesis (HA1) is also supported by the customer satisfaction with various merchandise (products) and services offered by all the 331 organised retailers (included in this study). Therefore, the null hypothesis H01 i.e. retailing strategies introduced by organised retailers are not significant to sustain in the hyper-competitive retail market is rejected and the alternate hypothesis HA1 is accepted. Testing of hypothesis H02 The alternate hypothesis of H02 has also been formulated as mentioned below. HA2: Customers are significantly satisfied with various products & services introduced by organised retailers. HA2: Customers are significantly satisfied with various products & services introduced by organised retailers. For testing this hypothesis, the data has been collected from the customers of 10 organised retailers (included in this study) through the question no. 30 of the individual questionnaire. This question no. 30 of the individual questionnaire comprises the 25 variables and some sub-variables of the organised retailers. The responses of customers pertained to the satisfaction have been collected on five point expectation cum satisfaction scale (basically an ordinal scale) ranging from greatly exceeding expectations and delighted (5), to much less than expectations and disappointed (1) with the middle of the scale identified by the response alternative matching expectations and satisfied (3). These responses of the customers about all 25 variables have been utilised to test this hypothesis (H02) with the comprehension that if the customers are significantly satisfied with these 25 variables of organised retailers, therefore, they are significantly satisfied with various products & services introduced by organised retailers. Moreover, the means of the responses on all the 5 parameters of the scale have been considered as the representative responses of the industry for testing the hypothesis in the case of the overall industry. For the purpose of statistical analyses, Kolmogorov-Smirnov one-sample test and one sample t-test have been identified to test the hypothesis. Kolmogorov- Smirnov one-sample test was used to study that whether there is a difference in the level of satisfaction of the respondents (i.e. customers of organised retailers). Therefore, Kolmogorov-Smirnov ‘D’ values (K-S ‘D’ Value) for all the 25 variables were calculated and compared with the critical values. The critical K-S ‘D’ value at 332 the 5% level of significance (α = 0.05) is 0.096 and at the 10% level of significance (α = 0.10) is 0.86. And then, one sample t-test has been used to test the hypothesis H02. The degree of freedom (DF) for each organised retailer will be 199 as the sample size is 200 customers of each organised retailers. The mean of the scale (μ) is 3, which is the test value. Only right-tailed values have been considered to test the hypothesis H02. Testing of hypothesis H02 At the 5% level of significance (α = 0.05), the critical value of t is 1.645 and at the 10% level of significance (α = 0.10), the critical value of t is 1.282. The results of both tests for all the 25 variables have been presented in the tables 8.26 to 8.50. Table 8.26: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with merchandise assortment Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? HyperCITY 0.190 0.05 Yes 2.337 0.010 0.05 Yes Star Bazaar 0.175 0.05 Yes 1.819 0.035 0.05 Yes more.MEGAStor e 0.155 0.05 Yes 1.699 0.045 0.05 Yes Big Bazaar 0.270 0.05 Yes 2.389 0.009 0.05 Yes D Mart 0.130 0.05 Yes 1.654 0.050 0.05 Yes Big Megamart 0.285 0.05 Yes 2.417 0.008 0.05 Yes West Side 0.230 0.05 Yes 2.344 0.010 0.05 Yes Max 0.160 0.05 Yes 1.906 0.029 0.05 Yes Reliance Digital 0.135 0.05 Yes 1.890 0.030 0.05 Yes NEXT 0.170 0.05 Yes 2.287 0.012 0.05 Yes Overall Industry 0.190 0.05 Yes 2.145 0.017 0.05 Yes Table 8.26: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with merchandise assortment The calculated K-S ‘D’ values exceed the critical values (at the 5% level of significance) for customer satisfaction with merchandise assortment of all the organised retailers (included in this study) as well the overall industry, which signifies that there is a significant difference in the level of satisfaction of customers with merchandise assortment of all the organised retailers. 333 The calculated ‘t’ values exceed the critical values (at the 5% level of significance) for customer satisfaction with merchandise assortment of all the organised retailers (included in this study) as well as the overall industry. Hence, the null hypothesis H02 i.e. “customers are not significantly satisfied with various products & services introduced by organised retailers” is rejected in case of customer satisfaction with merchandise assortment of all the organised retailers (included in this study) as well as the overall industry and the alternate hypothesis HA2 is accepted. Testing of hypothesis H02 Table 8.27: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with merchandise categories in the assortment Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? HyperCITY 0.135 0.05 Yes 1.963 0.026 0.05 Yes Star Bazaar 0.125 0.05 Yes 1.716 0.044 0.05 Yes more.MEGAStor e 0.095 0.10 Yes 1.466 0.072 0.10 Yes Big Bazaar 0.205 0.05 Yes 2.242 0.013 0.05 Yes D Mart 0.070 0.10 No 0.539 0.295 0.10 No Big Megamart 0.170 0.05 Yes 2.162 0.016 0.05 Yes West Side 0.160 0.05 Yes 2.068 0.020 0.05 Yes Max 0.090 0.10 Yes 1.529 0.064 0.10 Yes Reliance Digital 0.090 0.10 Yes 1.458 0.073 0.10 Yes NEXT 0.125 0.05 Yes 2.243 0.013 0.05 Yes Overall Industry 0.127 0.05 Yes 1.839 0.034 0.05 Yes Table 8.27: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with merchandise categories in the assortment The computed K-S ‘D’ values exceed the critical values (at an α of 5% or 10%) for customer satisfaction with merchandise categories in the assortment of all the organised retailers (included in this study) as well the overall industry except D Mart, which evinces that there is a significant difference in the level of satisfaction of customers with merchandise categories in the assortment of all the organised retailers except D Mart. The calculated ‘t’ values also exceed the critical values (at an α of 5% or 10%) for customer satisfaction with merchandise categories in the assortment of all the 334 organised retailers (included in this study) as well as the overall industry except D Mart. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with merchandise categories in the assortment of all the organised retailers (included in this study) as well as the overall industry except D Mart and the alternate hypothesis HA2 is accepted. But, the null hypothesis H02 is accepted in case of merchandise categories in the assortment of D Mart. D Mart is a superstore based retailer and offers limited merchandise categories in the assortment to its customers. Testing of hypothesis H02 Table 8.28: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the variety of merchandise in each category Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? HyperCITY 0.130 0.05 Yes 2.086 0.026 0.05 Yes Star Bazaar 0.120 0.05 Yes 1.779 0.044 0.05 Yes more.MEGAStor e 0.090 0.10 Yes 1.604 0.072 0.10 Yes Big Bazaar 0.235 0.05 Yes 2.475 0.013 0.05 Yes D Mart 0.080 0.10 No 0.950 0.295 0.10 No Big Megamart 0.165 0.05 Yes 2.522 0.016 0.05 Yes West Side 0.175 0.05 Yes 2.171 0.020 0.05 Yes Max 0.090 0.10 Yes 1.616 0.064 0.10 Yes Reliance Digital 0.090 0.10 Yes 1.643 0.073 0.10 Yes NEXT 0.105 0.05 Yes 1.801 0.013 0.05 Yes Overall Industry 0.128 0.05 Yes 2.003 0.034 0.05 Yes Table 8.28: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the variety of merchandise in each category The calculated K-S ‘D’ values exceed the critical values (at α = 0.05 or 0.10) for customer satisfaction with the variety of merchandise in each category of all the organised retailers (included in this study) as well the overall industry except D Mart, which evinces that there is a significant difference in the level of satisfaction of customers with the variety of merchandise in each category of all the organised retailers except D Mart. The computed ‘t’ values also exceed the critical values (at an α of 5% or 10%) for customer satisfaction with the variety of merchandise in each category of all the 335 organised retailers (included in this study) as well as the overall industry except D Mart. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with the variety of merchandise in each category of all the organised retailers (included in this study) as well as the overall industry except D Mart and the alternate hypothesis HA2 is accepted. But, the null hypothesis H02 is accepted in the case of the variety of merchandise in each category of D Mart. D Mart is a superstore based multi-item retailer and offers a limited variety of merchandise in each category for its customers. organised retailers (included in this study) as well as the overall industry except D Mart. Testing of hypothesis H02 Hence, the null hypothesis H02 is rejected in case of customer satisfaction with the variety of merchandise in each category of all the organised retailers (included in this study) as well as the overall industry except D Mart and the alternate hypothesis HA2 is accepted. But, the null hypothesis H02 is accepted in the case of the variety of merchandise in each category of D Mart. D Mart is a superstore based multi-item retailer and offers a limited variety of merchandise in each category for its customers. Table 8.29: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? HyperCITY 0.130 0.05 Yes 1.697 0.046 0.05 Yes Star Bazaar 0.090 0.10 Yes 1.565 0.060 0.10 Yes more.MEGAStor e 0.150 0.05 Yes 1.677 0.048 0.05 Yes Big Bazaar 0.185 0.05 Yes 1.951 0.026 0.05 Yes D Mart 0.060 0.10 No 0.672 0.251 0.10 No Big Megamart 0.190 0.05 Yes 1.994 0.024 0.05 Yes West Side 0.170 0.05 Yes 1.880 0.031 0.05 Yes Max 0.135 0.05 Yes 1.852 0.033 0.05 Yes Reliance Digital 0.145 0.05 Yes 1.651 0.050 0.05 Yes NEXT 0.135 0.05 Yes 1.648 0.050 0.05 Yes Overall Industry 0.139 0.05 Yes 1.734 0.042 0.05 Yes Table 8.29: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise The computed K-S ‘D’ values exceed the critical values (at an α of 5% or 10%) for customer satisfaction with the availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise by all the organised retailers (included in this study) as well the overall industry except D Mart, which reveals that there is a significant difference in the level of satisfaction of customers with the availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise by all the organised retailers except D Mart. Testing of hypothesis H02 336 The calculated ‘t’ values also exceed the critical values (at an α of 5% or 10%) for customer satisfaction with the availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise by all the organised retailers (included in this study) as well as the overall industry except D Mart. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with the availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise from all the organised retailers (included in this study) as well as the overall industry except D Mart and the alternate hypothesis HA2 is accepted. But, the null hypothesis H02 is accepted in the case of the availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise by D Mart. D Mart is a superstore based multi-item retailer and makes sure the limited availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise to its customers. Table 8.30: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the brands in each merchandise category Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? HyperCITY 0.120 0.05 Yes 1.724 0.043 0.05 Yes Star Bazaar 0.125 0.05 Yes 1.725 0.043 0.05 Yes more.MEGAStor e 0.095 0.10 Yes 1.302 0.097 0.10 Yes Big Bazaar 0.155 0.05 Yes 2.116 0.018 0.05 Yes D Mart 0.090 0.10 Yes 1.292 0.099 0.10 Yes Big Megamart 0.245 0.05 Yes 2.585 0.005 0.05 Yes West Side 0.180 0.05 Yes 2.283 0.012 0.05 Yes Max 0.130 0.05 Yes 1.871 0.031 0.05 Yes Reliance Digital 0.135 0.05 Yes 1.676 0.048 0.05 Yes NEXT 0.160 0.05 Yes 1.663 0.049 0.05 Yes Overall Industry 0.144 0.05 Yes 1.952 0.026 0.05 Yes Table 8.30: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the brands in each merchandise category 337 The calculated K-S ‘D’ values exceed the critical values (at α = 0.05 or 0.10) for customer satisfaction with the brands in each merchandise category of all the organised retailers (included in this study) as well the overall industry, which manifests that there is a significant difference in the level of satisfaction of customers with the brands in each merchandise category of all the organised retailers. Testing of hypothesis H02 The computed ‘t’ values also exceed the critical values (at an α of 5% or 10%) for customer satisfaction with the brands in each merchandise category of all the organised retailers (included in this study) as well as the overall industry. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with the brands in each merchandise category of all the organised retailers (included in this study) as well as the overall industry and the alternate hypothesis HA2 is accepted. Table 8.31: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with quality assurance and quality of merchandise Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? HyperCITY 0.155 0.05 Yes 2.028 0.022 0.05 Yes Star Bazaar 0.185 0.05 Yes 1.653 0.050 0.05 Yes more.MEGAStor e 0.160 0.05 Yes 1.725 0.043 0.05 Yes Big Bazaar 0.230 0.05 Yes 2.171 0.016 0.05 Yes D Mart 0.180 0.05 Yes 1.700 0.045 0.05 Yes Big Megamart 0.165 0.05 Yes 2.164 0.016 0.05 Yes West Side 0.175 0.05 Yes 2.046 0.021 0.05 Yes Max 0.145 0.05 Yes 1.687 0.047 0.05 Yes Reliance Digital 0.155 0.05 Yes 1.675 0.048 0.05 Yes NEXT 0.165 0.05 Yes 1.666 0.049 0.05 Yes Overall Industry 0.172 0.05 Yes 1.884 0.031 0.05 Yes Table 8.31: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with quality assurance and quality of merchandise The computed K-S ‘D’ values exceed the critical values (at the 5% level of significance) for customer satisfaction with quality assurance and quality of merchandise by all the organised retailers (included in this study) as well the overall 338 industry, which proves that there is a significant difference in the level of satisfaction of customers with quality assurance and quality of merchandise of all the organised retailers. The calculated ‘t’ values also exceed the critical values (at the 5% level of significance) for customer satisfaction with quality assurance and quality of merchandise by all the organised retailers (included in this study) as well as the overall industry. Testing of hypothesis H02 Hence, the null hypothesis H02 is rejected in case of customer satisfaction with quality assurance and quality of merchandise of all the organised retailers (included in this study) as well as the overall industry and the alternate hypothesis HA2 is accepted. Table 8.32: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the services Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? HyperCITY 0.240 0.05 Yes 2.463 0.007 0.05 Yes Star Bazaar 0.200 0.05 Yes 1.924 0.028 0.05 Yes more.MEGAStor e 0.170 0.05 Yes 1.678 0.047 0.05 Yes Big Bazaar 0.270 0.05 Yes 2.482 0.007 0.05 Yes D Mart 0.080 0.10 No 0.102 0.459 0.10 No Big Megamart 0.145 0.05 Yes 2.181 0.015 0.05 Yes West Side 0.195 0.05 Yes 2.352 0.010 0.05 Yes Max 0.085 0.10 No 0.519 0.302 0.10 No Reliance Digital 0.135 0.05 Yes 1.910 0.029 0.05 Yes NEXT 0.115 0.05 Yes 1.689 0.046 0.05 Yes Overall Industry 0.158 0.05 Yes 1.939 0.027 0.05 Yes Table 8.32: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the services The calculated K-S ‘D’ values exceed the critical values (at α = 0.05) for customer satisfaction with the services of all the organised retailers (included in this study) as well the overall industry except D Mart and Max, which endorses that there is a significant difference in the level of satisfaction of customers with the services of all the organised retailers except D Mart and Max. 339 The computed ‘t’ values also exceed the critical values (at an α of 5%) for customer satisfaction with the services of all the organised retailers (included in this study) as well as the overall industry except D Mart and Max. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with the services of all the organised retailers (included in this study) as well as the overall industry except D Mart and Max, and the alternate hypothesis HA2 is accepted. But, the null hypothesis H02 is accepted in the case of the services of D Mart and Max. D Mart is a superstore based multi-item retailer and Max is a superstore based fashion-item retailer. Both offer limited services to their customers in comparison to their competitors. Testing of hypothesis H02 Table 8.33: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the quality of the services Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? HyperCITY 0.140 0.05 Yes 2.004 0.023 0.05 Yes Star Bazaar 0.125 0.05 Yes 1.677 0.048 0.05 Yes more.MEGAStor e 0.110 0.05 Yes 1.706 0.045 0.05 Yes Big Bazaar 0.215 0.05 Yes 2.362 0.010 0.05 Yes D Mart 0.070 0.10 No 0.041 0.484 0.10 No Big Megamart 0.180 0.05 Yes 2.442 0.008 0.05 Yes West Side 0.195 0.05 Yes 2.364 0.010 0.05 Yes Max 0.080 0.10 No 0.941 0.174 0.10 No Reliance Digital 0.115 0.05 Yes 1.693 0.046 0.05 Yes NEXT 0.085 0.10 No 1.206 0.115 0.10 No Overall Industry 0.127 0.05 Yes 1.836 0.034 0.05 Yes Table 8.33: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the quality of the services The computed K-S ‘D’ values exceed the critical values (at an α of 5%) for customer satisfaction with the quality of the services of all the organised retailers (included in this study) as well the overall industry except D Mart, Max and NEXT, which substantiates that there is a significant difference in the level of satisfaction of customers with the quality of the services of all the organised retailers except D Mart, Max and NEXT. 340 The calculated ‘t’ values also exceed the critical values (at an α of 5%) for customer satisfaction with the quality of the services of all the organised retailers (included in this study) as well as the overall industry except D Mart, Max and NEXT. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with the quality of the services of all the organised retailers (included in this study) as well as the overall industry except D Mart, Max and NEXT and the alternate hypothesis HA2 is accepted. But, the null hypothesis H02 is accepted in the case of the quality of the services of D Mart, Max and NEXT. D Mart is a superstore based multi-item retailer, Max is a superstore based fashion-item retailer and NEXT is a superstore based electronics-item retailer. According to their customers, the quality of the services of these three retailers falls short of their expectations. Testing of hypothesis H02 Table 8.34: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the pricing of merchandise & services Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? HyperCITY 0.115 0.05 Yes 1.661 0.049 0.05 Yes Star Bazaar 0.135 0.05 Yes 1.685 0.047 0.05 Yes more.MEGAStor e 0.135 0.05 Yes 1.676 0.048 0.05 Yes Big Bazaar 0.200 0.05 Yes 1.881 0.031 0.05 Yes D Mart 0.190 0.05 Yes 2.352 0.010 0.05 Yes Big Megamart 0.105 0.05 Yes 1.691 0.046 0.05 Yes West Side 0.100 0.05 Yes 1.661 0.049 0.05 Yes Max 0.115 0.05 Yes 1.789 0.038 0.05 Yes Reliance Digital 0.130 0.05 Yes 1.648 0.050 0.05 Yes NEXT 0.215 0.05 Yes 2.391 0.009 0.05 Yes Overall Industry 0.144 0.05 Yes 1.918 0.028 0.05 Yes Table 8.34: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the pricing of merchandise & services The calculated K-S ‘D’ values exceed the critical values (at the 5% level of significance) for customer satisfaction with the pricing of merchandise & services of 341 all the organised retailers (included in this study) as well the overall industry, which reasserts that there is a significant difference in the level of satisfaction of customers with the pricing of merchandise & services of all the organised retailers. The computed ‘t’ values also exceed the critical values (at the 5% level of significance) for customer satisfaction with the pricing of merchandise & services of all the organised retailers (included in this study) as well as the overall industry. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with the pricing of merchandise & services of all the organised retailers (included in this study) as well as the overall industry and the alternate hypothesis HA2 is accepted. Table 8.35: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the overall value V/S price of merchandise & services Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? Testing of hypothesis H02 HyperCITY 0.125 0.05 Yes 1.766 0.039 0.05 Yes Star Bazaar 0.105 0.05 Yes 1.647 0.050 0.05 Yes more.MEGAStor e 0.110 0.05 Yes 1.761 0.040 0.05 Yes Big Bazaar 0.200 0.05 Yes 2.327 0.010 0.05 Yes D Mart 0.110 0.05 Yes 1.808 0.036 0.05 Yes Big Megamart 0.110 0.05 Yes 1.689 0.046 0.05 Yes West Side 0.110 0.05 Yes 1.681 0.047 0.05 Yes Max 0.100 0.05 Yes 1.663 0.049 0.05 Yes Reliance Digital 0.100 0.05 Yes 1.727 0.043 0.05 Yes NEXT 0.105 0.05 Yes 2.009 0.023 0.05 Yes Overall Industry 0.117 0.05 Yes 1.878 0.031 0.05 Yes Table 8.35: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the overall value V/S price of merchandise & services The computed K-S ‘D’ values exceed the critical values (at α = 0.05) for customer satisfaction with the overall value V/S price of merchandise & services of all the organised retailers (included in this study) as well the overall industry, which confirms that there is a significant difference in the level of satisfaction of customers with the overall value V/S price of merchandise & services of all the organised retailers. 342 The calculated ‘t’ values also exceed the critical values (at α = 0.05) for customer satisfaction with the overall value V/S price of merchandise & services of all the organised retailers (included in this study) as well as the overall industry. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with the overall value V/S price of merchandise & services of all the organised retailers (included in this study) as well as the overall industry and the alternate hypothesis HA2 is accepted. Table 8.36: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the exterior atmospherics of the store Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? Testing of hypothesis H02 HyperCITY 0.170 0.05 Yes 1.727 0.043 0.05 Yes Star Bazaar 0.165 0.05 Yes 1.985 0.024 0.05 Yes more.MEGAStor e 0.125 0.05 Yes 1.795 0.037 0.05 Yes Big Bazaar 0.235 0.05 Yes 2.242 0.013 0.05 Yes D Mart 0.085 0.10 No 0.468 0.320 0.10 No Big Megamart 0.210 0.05 Yes 2.527 0.006 0.05 Yes West Side 0.180 0.05 Yes 1.940 0.027 0.05 Yes Max 0.060 0.10 No 0.329 0.371 0.10 No Reliance Digital 0.180 0.05 Yes 1.855 0.033 0.05 Yes NEXT 0.080 0.10 No 0.088 0.465 0.10 No Overall Industry 0.149 0.05 Yes 1.681 0.047 0.05 Yes Table 8.36: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the exterior atmospherics of the store The computed K-S ‘D’ values exceed the critical values (at an α of 5%) for customer satisfaction with the exterior atmospherics of the store of all the organised retailers (included in this study) as well the overall industry except D Mart, Max and NEXT, which ratifies that there is a significant difference in the level of satisfaction of customers with the exterior atmospherics of the store of all the organised retailers except D Mart, Max and NEXT. The calculated ‘t’ values also exceed the critical values (at an α of 5%) for customer satisfaction with the exterior atmospherics of the store of all the organised 343 retailers (included in this study) as well as the overall industry except D Mart, Max and NEXT. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with the exterior atmospherics of the store of all the organised retailers (included in this study) as well as the overall industry except D Mart, Max and NEXT and the alternate hypothesis HA2 is accepted. But, the null hypothesis H02 is accepted in the case of the exterior atmospherics of the store of D Mart, Max and NEXT. D Mart is a superstore based multi-item retailer, Max is a superstore based fashion-item retailer and NEXT is a superstore based electronics-item retailer. As per their customers, the exterior atmospherics of the store of these three retailers are less than their expectations. Table 8.37: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the interior atmospherics of the store Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Testing of hypothesis H02 Calculated ‘t’ value p-value α Calculated value exceeds critical value? HyperCITY 0.175 0.05 Yes 2.014 0.023 0.05 Yes Star Bazaar 0.175 0.05 Yes 1.976 0.025 0.05 Yes more.MEGAStor e 0.140 0.05 Yes 1.664 0.049 0.05 Yes Big Bazaar 0.240 0.05 Yes 2.279 0.012 0.05 Yes D Mart 0.055 0.10 No -1.180 0.120 0.10 No Big Megamart 0.210 0.05 Yes 2.440 0.008 0.05 Yes West Side 0.160 0.05 Yes 1.891 0.030 0.05 Yes Max 0.095 0.10 Yes 1.607 0.055 0.10 Yes Reliance Digital 0.135 0.05 Yes 1.952 0.026 0.05 Yes NEXT 0.050 0.10 No -1.212 0.113 0.10 No Overall Industry 0.128 0.10 Yes 1.546 0.062 0.10 Yes Table 8.37: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the interior atmospherics of the store The calculated K-S ‘D’ values exceed the critical values (at an α of 5% or 10%) for customer satisfaction with the interior atmospherics of the store of all the organised retailers (included in this study) as well the overall industry except D Mart and NEXT, which establishes that there is a significant difference in the level of satisfaction of customers with the interior atmospherics of the store of all the organised retailers except D Mart and NEXT. 344 The computed ‘t’ values also exceed the critical values (at an α of 5% or 10%) for customer satisfaction with the interior atmospherics of the store of all the organised retailers (included in this study) as well as the overall industry except D Mart and NEXT. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with the interior atmospherics of the store of all the organised retailers (included in this study) as well as the overall industry except D Mart and NEXT and the alternate hypothesis HA2 is accepted. But, the null hypothesis H02 is accepted in the case of the interior atmospherics of the store of D Mart and NEXT. D Mart is a superstore based multi-item retailer and NEXT is a superstore based electronics-item retailer. As per their customers, the interior atmospherics of the store of these retailers are less than their expectations. Table 8.38: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the store layout Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Testing of hypothesis H02 Calculated ‘t’ value p-value α Calculated value exceeds critical value? HyperCITY 0.125 0.05 Yes 1.677 0.048 0.05 Yes Star Bazaar 0.095 0.10 Yes 1.284 0.100 0.10 Yes more.MEGAStore 0.075 0.10 No 1.015 0.156 0.10 No Big Bazaar 0.125 0.05 Yes 1.715 0.044 0.05 Yes D Mart 0.045 0.10 No -1.329 0.093 0.10 No Big Megamart 0.190 0.05 Yes 2.783 0.003 0.05 Yes West Side 0.175 0.05 Yes 2.320 0.011 0.05 Yes Max 0.120 0.05 Yes 1.735 0.042 0.05 Yes Reliance Digital 0.095 0.10 Yes 1.321 0.094 0.10 Yes NEXT 0.045 0.10 No -1.190 0.118 0.10 No Overall Industry 0.095 0.10 Yes 1.293 0.099 0.10 Yes Table 8.38: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the store layout The computed K-S ‘D’ values exceed the critical values (at an α of 5% or 10%) for customer satisfaction with the store layout of all the organised retailers (included in this study) as well the overall industry except more.MEGAStore, D 345 Mart and NEXT, which ratifies that there is a significant difference in the level of satisfaction of customers with the store layout of all the organised retailers except more.MEGAStore, D Mart and NEXT. The calculated ‘t’ values also exceed the critical values (at an α of 5% or 10%) for customer satisfaction with the store layout of all the organised retailers (included in this study) as well as the overall industry except more.MEGAStore, D Mart and NEXT. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with the store layout of all the organised retailers (included in this study) as well as the overall industry except more.MEGAStore, D Mart and NEXT and the alternate hypothesis HA2 is accepted. But, the null hypothesis H02 is accepted in the case of the store layout of more.MEGAStore, D Mart and NEXT. more.MEGAStore is a hyper store-based retailer, D Mart is a superstore based multi-item retailer and NEXT is a superstore based electronics-item retailer. According to their customers, the store layout of these retailers is poorer than their expectations. Table 8.39: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the visual merchandising Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? Testing of hypothesis H02 HyperCITY 0.135 0.05 Yes 1.890 0.026 0.05 Yes Star Bazaar 0.090 0.10 Yes 1.575 0.044 0.10 Yes more.MEGAStore 0.095 0.10 Yes 1.526 0.072 0.10 Yes Big Bazaar 0.140 0.05 Yes 1.929 0.013 0.05 Yes D Mart 0.045 0.10 No -0.716 0.295 0.10 No Big Megamart 0.190 0.05 Yes 2.601 0.016 0.05 Yes West Side 0.165 0.05 Yes 2.438 0.020 0.05 Yes Max 0.130 0.05 Yes 1.932 0.064 0.05 Yes Reliance Digital 0.105 0.05 Yes 1.702 0.073 0.05 Yes NEXT 0.055 0.10 No -0.495 0.013 0.10 No Overall Industry 0.109 0.05 Yes 1.639 0.034 0.10 Yes Table 8.39: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the visual merchandising The computed K-S ‘D’ values exceed the critical values (at an α of 5% or 10%) for customer satisfaction with the visual merchandising of all the organised retailers (included in this study) as well the overall industry except D Mart and 346 NEXT, which confirms that there is a significant difference in the level of satisfaction of customers with the visual merchandising of all the organised retailers except D Mart and NEXT. The calculated ‘t’ values also exceed the critical values (at an α of 5% or 10%) for customer satisfaction with the visual merchandising of all the organised retailers (included in this study) as well as the overall industry except D Mart and NEXT. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with the visual merchandising of all the organised retailers (included in this study) as well as the overall industry except D Mart and NEXT and the alternate hypothesis HA2 is accepted. But, the null hypothesis H02 is accepted in the case of the visual merchandising of D Mart and NEXT. D Mart is a superstore based multi-item retailer and NEXT is a superstore based electronics-item retailer. According to their customers, the visual merchandising of these retailers is poorer than their expectations. Table 8.40: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the methods of promotion and communication Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? Testing of hypothesis H02 HyperCITY 0.100 0.05 Yes 1.850 0.033 0.05 Yes Star Bazaar 0.140 0.05 Yes 1.743 0.041 0.05 Yes more.MEGAStore 0.120 0.05 Yes 1.773 0.039 0.05 Yes Big Bazaar 0.200 0.05 Yes 2.321 0.011 0.05 Yes D Mart 0.165 0.05 Yes 1.668 0.048 0.05 Yes Big Megamart 0.130 0.05 Yes 1.820 0.035 0.05 Yes West Side 0.105 0.05 Yes 1.772 0.039 0.05 Yes Max 0.150 0.05 Yes 2.209 0.014 0.05 Yes Reliance Digital 0.100 0.05 Yes 1.657 0.050 0.05 Yes NEXT 0.130 0.05 Yes 1.980 0.025 0.05 Yes Overall Industry 0.134 0.05 Yes 1.916 0.028 0.05 Yes Table 8.40: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the methods of promotion and communication The calculated K-S ‘D’ values exceed the critical values (at the 5% level of significance) for customer satisfaction with the methods of promotion and communication of all the organised retailers (included in this study) as well the 347 overall industry, which establishes that there is a significant difference in the level of satisfaction of customers with the methods of promotion and communication of all the organised retailers. The computed ‘t’ values also exceed the critical values (at the 5% level of significance) for customer satisfaction with the methods of promotion and communication of all the organised retailers (included in this study) as well as the overall industry. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with the methods of promotion and communication of all the organised retailers (included in this study) as well as the overall industry and the alternate hypothesis HA2 is accepted. Table 8.41: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the advertising methods Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? Testing of hypothesis H02 HyperCITY 0.115 0.05 Yes 1.876 0.031 0.05 Yes Star Bazaar 0.090 0.10 Yes 1.323 0.094 0.10 Yes more.MEGAStor e 0.105 0.05 Yes 1.650 0.050 0.05 Yes Big Bazaar 0.200 0.05 Yes 2.275 0.012 0.05 Yes D Mart 0.060 0.10 No 0.183 0.427 0.10 No Big Megamart 0.120 0.05 Yes 1.959 0.026 0.05 Yes West Side 0.120 0.05 Yes 1.774 0.039 0.05 Yes Max 0.105 0.05 Yes 1.694 0.046 0.05 Yes Reliance Digital 0.145 0.05 Yes 1.686 0.047 0.05 Yes NEXT 0.050 0.10 No 0.216 0.415 0.10 No Overall Industry 0.108 0.05 Yes 1.623 0.053 0.10 Yes Table 8.41: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the advertising methods The computed K-S ‘D’ values exceed the critical values (at an α of 5% or 10%) for customer satisfaction with the advertising methods of all the organised retailers (included in this study) as well the overall industry except D Mart and NEXT, which confirms that there is a significant difference in the level of satisfaction of customers with the advertising methods of all the organised retailers except D Mart and NEXT. 348 The calculated ‘t’ values also exceed the critical values (at an α of 5% or 10%) for customer satisfaction with the advertising methods of all the organised retailers (included in this study) as well as the overall industry except D Mart and NEXT. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with the advertising methods of all the organised retailers (included in this study) as well as the overall industry except D Mart and NEXT and the alternate hypothesis HA2 is accepted. But, the null hypothesis H02 is accepted in the case of the advertising methods of D Mart and NEXT. D Mart is a superstore based multi-item retailer and NEXT is a superstore based electronics-item retailer. According to their customers, advertising methods of these retailers are not as per the expectations of them. Table 8.42: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the sales promotion methods Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? Testing of hypothesis H02 HyperCITY 0.135 0.05 Yes 1.894 0.030 0.05 Yes Star Bazaar 0.140 0.05 Yes 1.721 0.043 0.05 Yes more.MEGAStor e 0.135 0.05 Yes 2.019 0.022 0.05 Yes Big Bazaar 0.240 0.05 Yes 2.429 0.008 0.05 Yes D Mart 0.140 0.05 Yes 2.030 0.022 0.05 Yes Big Megamart 0.105 0.05 Yes 1.847 0.033 0.05 Yes West Side 0.130 0.05 Yes 1.928 0.028 0.05 Yes Max 0.125 0.05 Yes 1.729 0.043 0.05 Yes Reliance Digital 0.110 0.05 Yes 1.705 0.045 0.05 Yes NEXT 0.145 0.05 Yes 2.117 0.018 0.05 Yes Overall Industry 0.141 0.05 Yes 1.995 0.024 0.05 Yes Table 8.42: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the sales promotion methods The calculated K-S ‘D’ values exceed the critical values (at α = 0.05) for customer satisfaction with the sales promotion methods of all the organised retailers (included in this study) as well the overall industry, which homologates that there is a significant difference in the level of satisfaction of customers with the sales promotion methods of all the organised retailers. 349 The computed ‘t’ values also exceed the critical values (at α = 0.05) for customer satisfaction with the sales promotion methods of all the organised retailers (included in this study) as well as the overall industry. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with the sales promotion methods of all the organised retailers (included in this study) as well as the overall industry and the alternate hypothesis HA2 is accepted. Table 8.43: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the direct retailing methods Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? Testing of hypothesis H02 HyperCITY 0.100 0.05 Yes 1.663 0.049 0.05 Yes Star Bazaar 0.000 0.10 No -2.003 0.023 0.10 No more.MEGAStor e 0.000 0.10 No -2.258 0.013 0.10 No Big Bazaar 0.130 0.05 Yes 1.753 0.041 0.05 Yes D Mart 0.000 0.10 No -2.439 0.008 0.10 No Big Megamart 0.040 0.10 No -0.554 0.290 0.10 No West Side 0.110 0.05 Yes 1.687 0.047 0.05 Yes Max 0.000 0.10 No -2.663 0.004 0.10 No Reliance Digital 0.110 0.05 Yes 1.674 0.048 0.05 Yes NEXT 0.110 0.05 Yes 1.671 0.048 0.05 Yes Overall Industry 0.040 0.10 No -0.282 0.389 0.10 No Table 8.43: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the direct retailing methods The calculated K-S ‘D’ values are less than the critical values (at α = 0.10) for customer satisfaction with the direct retailing methods of all the organised retailers (included in this study) as well the overall industry except HyperCITY, Big Bazaar, D Mart, West Side, Reliance Digital and NEXT, which establishes that there is no significant difference in the level of satisfaction of customers with the direct retailing methods of all the organised retailers except HyperCITY, Big Bazaar, D Mart, West Side, Reliance Digital and NEXT. 350 The computed ‘t’ values are also less than the critical values (at α = 0.10) for customer satisfaction with the direct retailing methods of all the organised retailers (included in this study) as well as the overall industry except HyperCITY, Big Bazaar, D Mart, West Side, Reliance Digital and NEXT. Hence, the null hypothesis H02 is accepted in case of customer satisfaction with the direct retailing methods of all the organised retailers (included in this study) as well as the overall industry except HyperCITY, Big Bazaar, D Mart, West Side, Reliance Digital and NEXT and the alternate hypothesis HA2 is rejected. But, the alternate hypothesis HA2 is accepted in the case of the direct retailing methods of HyperCITY, Big Bazaar, D Mart, West Side, Reliance Digital and NEXT. HyperCITY and Big Bazaar are a hyper store- based multi-item retailers, D Mart is a superstore based multi-item retailer, West Side is a superstore based fashion-item retailers and NEXT is a superstore based electronics-item retailer. According to their customers, the direct retailing methods adopted by these retailers are as per the expectations of them. Testing of hypothesis H02 Table 8.44: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the personal retailing methods Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? HyperCITY 0.120 0.05 Yes 1.925 0.028 0.05 Yes Star Bazaar 0.130 0.05 Yes 1.705 0.045 0.05 Yes more.MEGAStor e 0.090 0.10 Yes 1.616 0.054 0.10 Yes Big Bazaar 0.135 0.05 Yes 2.160 0.016 0.05 Yes D Mart 0.050 0.10 No -0.514 0.304 0.10 No Big Megamart 0.100 0.05 Yes 2.014 0.023 0.05 Yes West Side 0.130 0.05 Yes 2.217 0.014 0.05 Yes Max 0.035 0.10 No -0.936 0.175 0.10 No Reliance Digital 0.125 0.05 Yes 1.766 0.039 0.05 Yes NEXT 0.050 0.10 No -0.247 0.403 0.10 No Overall Industry 0.088 0.10 Yes 1.360 0.088 0.10 Yes Table 8.44: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the personal retailing methods The computed K-S ‘D’ values exceed the critical values (at an α of 5% or 10%) for customer satisfaction with the personal retailing methods of all the 351 organised retailers (included in this study) as well the overall industry except D Mart, Max and NEXT, which indicates that there is a significant difference in the level of satisfaction of customers with the personal retailing methods of all the organised retailers except D Mart, Max and NEXT. The calculated ‘t’ values also exceed the critical values (at an α of 5% or 10%) for customer satisfaction with the personal retailing methods of all the organised retailers (included in this study) as well as the overall industry except D Mart, Max and NEXT. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with personal retailing methods of all the organised retailers (included in this study) as well as the overall industry except D Mart, Max and NEXT and the alternate hypothesis HA2 is accepted. But, the null hypothesis H02 is accepted in the case of the personal retailing methods of D Mart, Max and NEXT. D Mart is a superstore based multi-item retailer, Max is a superstore based fashion-item retailer and NEXT is a superstore based electronics-item retailer. According to their customers, the personal retailing methods adopted by these retailers are not as per the expectations of them. Testing of hypothesis H02 Table 8.45: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the public relation/publicity methods Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? HyperCITY 0.115 0.05 Yes 1.694 0.046 0.05 Yes Star Bazaar 0.005 0.10 No -1.714 0.044 0.10 No more.MEGAStor e 0.120 0.05 Yes 1.789 0.038 0.05 Yes Big Bazaar 0.150 0.05 Yes 1.952 0.026 0.05 Yes D Mart 0.000 0.10 No -2.941 0.002 0.10 No Big Megamart 0.100 0.05 Yes 1.801 0.037 0.05 Yes West Side 0.105 0.05 Yes 1.681 0.047 0.05 Yes Max 0.120 0.05 Yes 1.671 0.048 0.05 Yes Reliance Digital 0.115 0.05 Yes 1.647 0.050 0.05 Yes NEXT 0.000 0.10 No -2.946 0.002 0.10 No Overall Industry 0.040 0.10 No -0.012 0.495 0.10 No Table 8.45: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the public relation/publicity methods 352 The computed K-S ‘D’ values exceed the critical values (at an α of 5%) for customer satisfaction with the public relation/publicity methods of all the organised retailers (included in this study) except Star Bazaar, D Mart, NEXT and the overall industry, which evinces that there is a significant difference in the level of satisfaction of customers with the public relation/publicity methods of all the organised retailers except Star Bazaar, D Mart, NEXT. The calculated ‘t’ values also exceed the critical values (at an α of 5%) for customer satisfaction with the public relation/publicity methods of all the organised retailers (included in this study) except Star Bazaar, D Mart, NEXT and the overall industry. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with the public relation/publicity methods of all the organised retailers (included in this study) except Star Bazaar, D Mart, NEXT and the overall industry and the alternate hypothesis HA2 is accepted. But, the null hypothesis H02 is accepted in the case of the public relation/publicity methods of Star Bazaar, D Mart, NEXT and the overall industry. Star Bazaar is a hyper store based multi-item retailer, D Mart is a superstore based multi-item retailer and NEXT is a superstore based electronics-item retailer. According to their customers, the public relation/publicity methods adopted by these retailers are not as per the expectations of them. Testing of hypothesis H02 Table 8.46: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the internet/interactive retailing methods Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? HyperCITY 0.120 0.05 Yes 2.314 0.011 0.05 Yes Star Bazaar 0.120 0.05 Yes 1.900 0.029 0.05 Yes more.MEGAStore 0.130 0.05 Yes 1.771 0.039 0.05 Yes Big Bazaar 0.180 0.05 Yes 2.215 0.014 0.05 Yes D Mart 0.000 0.10 No -2.834 0.003 0.10 No Big Megamart 0.110 0.05 Yes 2.272 0.012 0.05 Yes West Side 0.100 0.05 Yes 1.943 0.027 0.05 Yes Table 8.46: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the internet/interactive retailing methods 353 Max 0.165 0.05 Yes 2.077 0.020 0.05 Yes Reliance Digital 0.170 0.05 Yes 2.090 0.019 0.05 Yes NEXT 0.100 0.05 Yes 1.661 0.049 0.05 Yes Overall Industry 0.106 0.05 Yes 1.788 0.038 0.05 Yes The calculated K-S ‘D’ values exceed the critical values (at α = 0.05) for customer satisfaction with the internet/interactive retailing methods of all the organised retailers (included in this study) as well the overall industry except D Mart, which evinces that there is a significant difference in the level of satisfaction of customers with the internet/interactive retailing methods of all the organised retailers except D Mart. Max 0.165 0.05 Yes 2.077 0.020 0.05 Yes Reliance Digital 0.170 0.05 Yes 2.090 0.019 0.05 Yes NEXT 0.100 0.05 Yes 1.661 0.049 0.05 Yes Overall Industry 0.106 0.05 Yes 1.788 0.038 0.05 Yes The calculated K-S ‘D’ values exceed the critical values (at α = 0.05) for The calculated K-S ‘D’ values exceed the critical values (at α = 0.05) for customer satisfaction with the internet/interactive retailing methods of all the organised retailers (included in this study) as well the overall industry except D Mart, which evinces that there is a significant difference in the level of satisfaction of customers with the internet/interactive retailing methods of all the organised retailers except D Mart. Testing of hypothesis H02 HyperCITY 0.150 0.05 Yes 2.354 0.010 0.05 Yes Star Bazaar 0.125 0.05 Yes 2.131 0.017 0.05 Yes more.MEGAStor e 0.110 0.05 Yes 1.760 0.040 0.05 Yes Big Bazaar 0.205 0.05 Yes 2.420 0.008 0.05 Yes D Mart 0.005 0.10 No -2.102 0.018 0.10 No Big Megamart 0.160 0.05 Yes 2.232 0.013 0.05 Yes Table 8.47: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the sales personnel 354 West Side 0.170 0.05 Yes 2.109 0.018 0.05 Yes Max 0.045 0.10 No -0.185 0.427 0.10 No Reliance Digital 0.145 0.05 Yes 2.126 0.017 0.05 Yes NEXT 0.100 0.05 Yes 1.871 0.031 0.05 Yes Overall Industry 0.113 0.05 Yes 1.776 0.039 0.05 Yes The calculated K-S ‘D’ values exceed the critical values (at α = 0.05) for customer satisfaction with the sales personnel of all the organised retailers (included in this study) as well the overall industry except D Mart and Max, which reasserts that there is a significant difference in the level of satisfaction of customers with the sales personnel of all the organised retailers except D Mart and Max. West Side 0.170 0.05 Yes 2.109 0.018 0.05 Yes Max 0.045 0.10 No -0.185 0.427 0.10 No Reliance Digital 0.145 0.05 Yes 2.126 0.017 0.05 Yes NEXT 0.100 0.05 Yes 1.871 0.031 0.05 Yes Overall Industry 0.113 0.05 Yes 1.776 0.039 0.05 Yes The calculated K-S ‘D’ values exceed the critical values (at α = 0.05) for The computed ‘t’ values also exceed the critical values (at an α of 5%) for customer satisfaction with the sales personnel of all the organised retailers (included in this study) as well as the overall industry except D Mart and Max. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with the sales personnel of all the organised retailers (included in this study) as well as the overall industry except D Mart and Max and the alternate hypothesis HA2 is accepted. But, the null hypothesis H02 is accepted in the case of the sales personnel of D Mart and Max. D Mart is a superstore based multi-item retailer and Max is a superstore based fashion- item retailer. According to their customers, the sales personnel of these retailers perform below the expectations of them. Testing of hypothesis H02 The calculated K-S ‘D’ values exceed the critical values (at α = 0.05) for customer satisfaction with the internet/interactive retailing methods of all the organised retailers (included in this study) as well the overall industry except D Mart, which evinces that there is a significant difference in the level of satisfaction of customers with the internet/interactive retailing methods of all the organised retailers except D Mart. The computed ‘t’ values also exceed the critical values (at an α of 5%) for customer satisfaction with the internet/interactive retailing methods of all the organised retailers (included in this study) as well as the overall industry except D Mart. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with the internet/interactive retailing methods of all the organised retailers (included in this study) as well as the overall industry except D Mart and the alternate hypothesis HA2 is accepted. But, the null hypothesis H02 is accepted in the case of the internet/interactive retailing methods of D Mart. D Mart is a superstore based multi-item retailer and not good at meeting the expectations of today's generation, as poor at the internet/interactive retailing. Table 8.47: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the sales personnel Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? Testing of hypothesis H02 Table 8.48: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the contact personnel Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? HyperCITY 0.110 0.05 Yes 2.144 0.017 0.05 Yes Star Bazaar 0.115 0.05 Yes 1.810 0.036 0.05 Yes more.MEGAStore 0.090 0.10 Yes 1.324 0.094 0.10 Yes Big Bazaar 0.195 0.05 Yes 2.364 0.010 0.05 Yes D Mart 0.035 0.10 No -0.110 0.456 0.10 No Big Megamart 0.185 0.05 Yes 2.385 0.009 0.05 Yes Table 8.48: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the contact personnel 355 West Side 0.180 0.05 Yes 2.440 0.008 0.05 Yes Max 0.120 0.05 Yes 1.779 0.038 0.05 Yes Reliance Digital 0.140 0.10 Yes 2.089 0.019 0.10 Yes NEXT 0.045 0.10 No -0.110 0.456 0.10 No Overall Industry 0.120 0.05 Yes 1.895 0.030 0.05 Yes The calculated K-S ‘D’ values exceed the critical values (at an α of 5% or 10%) for customer satisfaction with the contact personnel of all the organised retailers (included in this study) as well the overall industry except D Mart and NEXT, which reasserts that there is a significant difference in the level of satisfaction of customers with the contact personnel of all the organised retailers except D Mart and NEXT. West Side 0.180 0.05 Yes 2.440 0.008 0.05 Yes Max 0.120 0.05 Yes 1.779 0.038 0.05 Yes Reliance Digital 0.140 0.10 Yes 2.089 0.019 0.10 Yes NEXT 0.045 0.10 No -0.110 0.456 0.10 No Overall Industry 0.120 0.05 Yes 1.895 0.030 0.05 Yes The calculated K S ‘D’ values exceed the critical values (at an α of 5% or The calculated K-S ‘D’ values exceed the critical values (at an α of 5% or 10%) for customer satisfaction with the contact personnel of all the organised retailers (included in this study) as well the overall industry except D Mart and NEXT, which reasserts that there is a significant difference in the level of satisfaction of customers with the contact personnel of all the organised retailers except D Mart and NEXT. Testing of hypothesis H02 The computed ‘t’ values also exceed the critical values (at an α of 5% or 10%) for customer satisfaction with the contact personnel of all the organised retailers (included in this study) as well as the overall industry except D Mart and NEXT. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with the contact personnel of all the organised retailers (included in this study) as well as the overall industry except D Mart and NEXT and the alternate hypothesis HA2 is accepted. But, the null hypothesis H02 is accepted in the case of the contact personnel of D Mart and NEXT. D Mart is a superstore based multi-item retailer and NEXT is a superstore based electronics-item retailer. According to their customers, the contact personnel of these retailers perform below the expectations of them. Table 8.49: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the factors pertained to personnel Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? HyperCITY 0.195 0.05 Yes 2.364 0.010 0.05 Yes Star Bazaar 0.095 0.10 Yes 1.570 0.059 0.10 Yes more.MEGAStore 0.090 0.10 Yes 1.521 0.065 0.10 Yes Big Bazaar 0.220 0.05 Yes 2.454 0.007 0.05 Yes D Mart 0.045 0.10 No -0.185 0.427 0.10 No Big Megamart 0.210 0.05 Yes 2.456 0.007 0.05 Yes Table 8.49: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the factors pertained to personnel 356 West Side 0.175 0.05 Yes 2.257 0.013 0.05 Yes Max 0.040 0.10 No -0.351 0.363 0.10 No Reliance Digital 0.140 0.05 Yes 1.769 0.039 0.05 Yes NEXT 0.035 0.10 No -0.267 0.395 0.10 No Overall Industry 0.121 0.05 Yes 1.798 0.037 0.05 Yes h d l d h i i l l ( f The computed K-S ‘D’ values exceed the critical values (at an α of 5% or 10%) for customer satisfaction with the factors pertained to personnel of all the organised retailers (included in this study) as well the overall industry except D Mart, Max and NEXT, which proves that there is a significant difference in the level of satisfaction of customers with the factors pertained to personnel of all the organised retailers except D Mart, Max and NEXT. Testing of hypothesis H02 The calculated ‘t’ values also exceed the critical values (at an α of 5% or 10%) for customer satisfaction with the factors pertained to personnel of all the organised retailers (included in this study) as well as the overall industry except D Mart, Max and NEXT. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with the factors pertained to personnel of all the organised retailers (included in this study) as well as the overall industry except D Mart, Max and NEXT and the alternate hypothesis HA2 is accepted. But, the null hypothesis H02 is accepted in the case of the factors pertained to personnel of D Mart, Max and NEXT. D Mart is a superstore based multi-item retailer, Max is a superstore based fashion-item retailer and NEXT is a superstore based electronics-item retailer. According to their customers, personnel of these retailers perform below the expectations of them pertained to mood, efforts, commitment, attitude, knowledge, skills, behaviour, approach to delivering the service, competence, courtesy, credibility, reliability, responsiveness, communication style, friendliness, helpfulness, discipline & punctuality, customer counselling ability of personnel. Table 8.50: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the other tangible evidence Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? Table 8.50: Statistics of Kolmogorov-Smirnov one-sample test and one sample t-test for customer satisfaction with the other tangible evidence Retailers Statistics of Kolmogorov- Smirnov one sample test Statistics of one sample t-test Calculated K-S ‘D’ Value α Calculated value exceeds critical value? Calculated ‘t’ value p-value α Calculated value exceeds critical value? Testing of hypothesis H02 357 HyperCITY 0.150 0.05 Yes 2.094 0.019 0.05 Yes Star Bazaar 0.150 0.05 Yes 2.242 0.013 0.05 Yes more.MEGAStor e 0.090 0.10 Yes 1.398 0.082 0.10 Yes Big Bazaar 0.180 0.05 Yes 2.057 0.020 0.05 Yes D Mart 0.050 0.05 No -0.442 0.329 0.05 No Big Megamart 0.195 0.05 Yes 2.352 0.010 0.05 Yes West Side 0.175 0.05 Yes 1.961 0.026 0.05 Yes Max 0.095 0.10 Yes 1.558 0.060 0.10 Yes Reliance Digital 0.135 0.05 Yes 1.996 0.024 0.05 Yes NEXT 0.035 0.05 No -0.363 0.358 0.05 No Overall Industry 0.123 0.05 Yes 1.726 0.043 0.05 Yes The computed K-S ‘D’ values exceed the critical values (at an α of 5% or 10%) for customer satisfaction with the other tangible evidences of all the organised retailers (included in this study) as well the overall industry except D Mart and NEXT, which confirms that there is a significant difference in the level of satisfaction of customers with the other tangible evidence of all the organised retailers except D Mart and NEXT. The calculated ‘t’ values also exceed the critical values (at an α of 5% or 10%) for customer satisfaction with the other tangible evidence of all the organised retailers (included in this study) as well as the overall industry except D Mart and NEXT. Hence, the null hypothesis H02 is rejected in case of customer satisfaction with the other tangible evidence of all the organised retailers (included in this study) as well as the overall industry except D Mart and NEXT and the alternate hypothesis HA2 is accepted. But, the null hypothesis H02 is accepted in the case of the other tangible evidence of D Mart and NEXT. D Mart is a superstore based multi-item retailer and NEXT is a superstore based electronics-item retailer. According to their customers, these retailers do not match the expectations of them pertained to the other tangible evidence. 358 Conclusion and recommendation s  Introduction  Conclusions  Recommendations  Areas for further research Introduction The retail industry of India is one among the most contributing sectors to the GDP and employment in the country, considered as a vibrant industry and a pillar of the Indian economy. The growth estimations of the industry are quite lucrative and estimated that the Indian retail market will be one of the top five markets (in terms of economic value) in the world. The business in the retail sector is divided between the organised and unorganised retailers. The major share in the retailing business is with unorganised retailers. However, the organised retail business is growing considerably and estimated to grow with much faster rate in the future. The organised retail sector is passing through the transformation from the last one decade due to the entry of corporate groups like Tata Group, Reliance Group, Bharti Enterprises, Aditya Birla Group, Future Group, RPG Enterprises, Piramal Group, Vishal Group, K Raheja Group, Vivek Group, etc. The organised retail sector is shifting towards the adoption of modern retailing concepts due to these new entrants, who are keen to revolutionise the industry. The retailing sector is witnessing the drastic changes in the introduction of the different formats of retailing e.g. shopping malls, supermarkets, hypermarkets, department stores, convenience store, etc. The organised retailers are not only revolutionising the sector in the urban areas but also moving towards smaller cities and even rural areas of the country. The organised retailers offer the world-class assortment of merchandise and services, affordable pricing and an appealing shopping environment to meet the expectations of customers. It is a known fact that the customer is the king. Mahatma Gandhi also told the importance of the customer to a business and said that the customer is the most important visitor in the business premises. The customer is not dependent on your business. The business is dependent on him. The customer is not an interruption in the business. The customer is the purpose of the business. The customer is not an outsider in the business. The customer is part of the business. Businesses are not doing a favour by serving the customer. The customer is doing a favour by giving an opportunity to do 359 so. 9.2 Recommendations 9.2 Recommendations 9.2 Recommendations 9.3 Areas for further research Introduction By taking the inspiration from this great thought, this current study has been pursued to analyse the products (merchandise) & services and retailing strategies by considering the point of views of both the organised retailers as well as customers. This chapter comprises the conclusions driven from the analyses of the data collected (which have been presented in chapter 5, 6, 7 and 8) from the organised retailers and customers of them. This chapter also comprises the recommendations based on the analyses of the data collected from the organised retailers and customers of them and areas of future research. This chapter has been segregated into the sections as follows: 9.1 Conclusions  Conclusions drawn from the analysis of merchandise (products) & customer services in the organised retail industry  Conclusions drawn from the analysis of retailing strategies introduced by organised retail industry  Conclusions drawn from the analysis of profile of customers in the organised retail industry  Conclusions drawn from the analysis of buying behaviour of customers in the organised retail industry  Conclusions drawn from the analysis factors influencing buying behaviour of customers in the organised retail industry  Conclusions drawn from the analysis of customer satisfaction with the merchandise (products) and services offered by organised retail industry  Conclusions drawn from testing of hypotheses  Conclusions drawn from testing of hypotheses 9.1 CONCLUSIONS 360 9.1.1 Conclusions drawn from the analysis of merchandise (products) & customer services in the organised retail industry 9.1.1 Conclusions drawn from the analysis of merchandise (products) & customer services in the organised retail industry The merchandise lines of the organised retailers were considered in the analysis of merchandise in the organised retail industry, as it was not possible to include all the merchandise offered by some multi-format retailers, offering more than 30000 merchandise to their customers. The players in the organised retail industry offer very much similar merchandise lines to their customers. However, the variety of merchandise under different merchandise lines slightly differs with each other. The multi-format (item) organised retailers (e.g. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart) offer very much similar merchandise lines to each other and the most common merchandise lines are food & grocery, health, beauty / personal care, home care, furniture, fashion, electronics, toys, sports, accessories, private / store levels. However, D Mart does not offer the private / store levels to its customers. Some multi-item organised retailers (e.g. Star Bazaar and Big Bazaar) offer more merchandise lines as compared to their competitors. Star Bazaar offers some international brand under the special merchandise line International zone and Big Bazaar offers some gift items (e.g. cakes & chocolates, gift vouchers, gift sets, valentine's day, Diwali, charity, flowers, greeting cards, seasonal gifts, sweets & dry fruits) for special occasions under the gift merchandise line. The multi-format (item) organised retailers offer a wide variety of merchandise in each category as per different sizes, colours, designs, price, quality, quantity and brands, which differs slightly as compared to each other. However, the hyper-store based organised retailers (e.g. HyperCITY, Star Bazaar, more. MEGAStore and Big Bazaar) offer more variety of merchandise as compared to the super - store based organised retailers (e.g. D Mart). The fashion item organised retailers offer a very much similar merchandise line with each other and the most common merchandise line are women's wear, men's wear, youth wear, kid's wear, footwear and fashion accessories. However, few (e.g. Big Magamart and West Side) offers more merchandise line, i.e. home items and travel 361 items. Moreover, Big Magamart also offers toys to its customers as compared to its other for differentiating itself. The hyper store based fashion-item organised retailers (e.g. Big Megamart) offer a wide variety of merchandise in each category as per different sizes, colours, designs, price, quality, quantity and brands while superstore based fashion-item organised retailers (e.g. West Side and Max) offer a narrow variety of merchandise in each category. The electronics-item organised retailers (e.g. 9.1.1 Conclusions drawn from the analysis of merchandise (products) & customer services in the organised retail industry Reliance Digital and NEXT) also offer a very much similar merchandise lines to each other and the most common merchandise lines are phones, computers, TVs & audio, cameras, home appliances, kitchen appliances, gaming and accessories. However, some retailers (e.g. NEXT) offers few more merchandise lines (e.g. writing instruments and fragrances) to its customers. The electronics-item organised retailers also offer a narrow variety of merchandise as per different sizes, colours, designs, price, quality, quantity and brands. In conclusion, food & grocery, beauty / personal care, home care and fashion are the most contributing merchandise lines in the growth of the multi-format (item) organised retailers, while women's wear, men's wear and kid's wear are the most contributing merchandise lines in the growth of the fashion item organised retailers and kitchen appliances and home appliances are the most contributing merchandise lines in the growth of the electronics item organised retailers The players in the organised retail industry also offer a very much similar services to their customers and the most common services are home delivery, alteration facilities (in case of apparel), packaging / gift wrapping, installations in case of electronics item, complaints & exchange handling, personal shoppers, parking, fitting / trial rooms, information, safety of personal things, lift / escalator & staircase facility, after sale service for electronics and durables, demonstrations of merchandise, entertainment facilities, acceptance of all international credit & debit cards, shopping carts, warranties on merchandise, washrooms & drinking water facilities, air-condition environment, music, enough billing-counters, CCTV for security, fire safety arrangement, back office for support, customers care service, repairing services in case 362 of durable, gift certificates, special sales for regular customers, restaurants, shopping bags, refreshment facilities etc. However, the hyper store based organised retailers (e.g. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and Big Megamart) provide more services to their customers as compared to superstore based organised retailers (e.g. D Mart, West Side, Max, Reliance Digital and NEXT). organised retail industry The organised retailers use similar retailing strategies with little difference in the execution. However, organised retailers employ different retailing strategies in a different retailing format, e.g. retailing strategies of hyper store retailers differ from superstore retailers. Nevertheless, it does not mean that hyper store or superstore retailers are not adopting the successful strategies of each other. Moreover, organised retailers also employ different retailing strategies in the different areas of operations, e.g. retailing strategies of multi-item organised retailers differ from the fashion item organised retailers and electronics item organised retailers. Despite, there is also similarity among the strategies of multi-item organised retailers, fashion-item organised retailers and electronics-item organised retailers. The organised retailers use demographic, psychographic and behavioural variables for segmenting the retail market. Age, family life cycle, gender, income, occupation and education are most the common variables for demographic segmentation while lifestyle and personality are most common variables for psychographic segmentation and usage rate, loyalty status, occasions, benefits and innovativeness are the most common variables for behavioural segmentation of the retail market. The multi-format organised retailers (e.g. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart) adopt full market coverage segmentation strategy while the fashion item organised retailers (e.g. Big Megamart, West Side and Max) prefer selective specialisation segmentation strategy and the electronics item organised retailers (e.g. Reliance Digital and NEXT) adopt market specialisation segmentation strategy for segmenting the retail market. The multi-format organised retailers (e.g. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart) adopt an undifferentiating targeting 363 strategy or mass retailing to target the retail market. The concentrated targeting strategy is mostly adopted strategy among superstore organised retailers (e.g. West Side, Max and NEXT) while other organised retailers (e.g. Big Megamart and Reliance Digital) prefers differentiating targeting strategy or selective retailing. More for less is most popular positioning strategy among both hyper & superstore based organised retailers (e.g. HyperCITY, more.MEGAStore, Big Bazaar, D Mart, Max and Reliance Digital) for positioning their offerings among the targeted customers. However, some organised retailers (e.g. Star Bazaar and NEXT) adopt same for less positioning strategy while others (Big Megamart and West Side) more for same positioning strategy. Both hyper & superstores / multi-format & specialised organised retailers (e.g. HyperCITY, Star Bazaar, Big Bazaar, Max, Reliance Digital and NEXT) adopt a retail market penetration strategy in order to grab the growth opportunities, while some other organised retailers (e.g. organised retail industry more.MEGAStore, D Mart and West Side) adopt a retail market expansion strategy and others (e.g. Big Megamart) adopt a retail format development strategy by offering merchandise & services with the new retail format. The most favourable defensive choice is a preemptive defence strategy among both hyper and superstore organised retailers (e.g. Big Megamart, HyperCITY, Star Bazaar, Big Bazaar, Max and Reliance Digital). All these retailers attack competitors more aggressively by launching a wide range of merchandise and brands in different categories. The mobile defence strategy is second most favourable defensive choice among superstores organised retailers (e.g. D Mart, West Side and NEXT) and they hope to expand rapidly by stretching its domain to new territories with the similar format store. However, some (e.g. more.MEGAStore – a multi-format hyper store organised retailer) uses contraction defence strategy by giving up from the weaker territories from where the company is not getting sufficient ROI. Cost leadership strategy is a common choice as a business level strategy at store-level among the multi-format organised retailers (e.g. HyperCITY, Star Bazaar, 364 more.MEGAStore, Big Bazaar and D Mart). However, hyper store based fashion-item organised retailers (e.g. Big Megamart) follow differentiation strategy at store-level while some super store fashion-item organised retailers (e.g. West Side) follow differentiation focus strategy at store-level and other specialised superstore organised retailers (e.g. Max, Reliance Digital and NEXT) prefer cost focus strategy at store- level. The organised retailers face some common problems in framing overall retailing strategies are the high price of real estate, lack of preferred locations and retail space, high cost of space and location, trained workforce, multiple & a complex taxation system of India, many entrants as online retailers, high demand for service standards as consumerism is shifting in India. The organised retailers consider the product (merchandise) as an extremely important element of their retail marketing mix. Price is also an extremely important element of RMM for the organised retailers except few (e.g. Big Megamart and West Side). However, these organised retailers consider it as an important element of RMM. The organised retailers also consider promotion as an extremely important element of RMM except few (e.g. Big Megamart); however, it is important for Big Megamart. The premises / place is an extremely important element of RMM for the organised retailers except few (e.g. organised retail industry D Mart, Max and NEXT), however, it is an important element for D Mart and Max, while partially important for NEXT. Personnel / people are considered as an extremely important element of RMM by the organised retailers, except few (e.g. D Mart and Max). For them, it is an important element. The presentation is an extremely important element of RMM for all the organised retailers, except few (e.g. more.MEGAStore, D Mart and NEXT) as it is important for more.MEGAStore and partially important for D Mart and NEXT. Physical evidence is an important element of RMM for the organised retailers except few (e.g. NEXT), as it is partially important. A heterogeneous mix of brands is the first choice among the multi-format organised retailers (e.g. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart), however, a hyper store based fashion store organised retailers (e.g. Big 365 Megamart) also offer a heterogeneous mix of brands to targeted customers. The offering basket of all these retailers contains generic, local, regional, national, copycat, international and private/store brands, while fashion-item superstores (e.g. West Side and Max) choose a blend of private label & other brands to offer and their offering basket contains private (store level) in all categories offered, exclusive, national (manufacturers) and international brands. Moreover, electronics item superstores (e.g. Reliance Digital and NEXT) offer multi-brands in different categories of the merchandise and their offering basket contains all the leading (world’s most popular brands) national and international brands. Megamart) also offer a heterogeneous mix of brands to targeted customers. The offering basket of all these retailers contains generic, local, regional, national, copycat, international and private/store brands, while fashion-item superstores (e.g. West Side and Max) choose a blend of private label & other brands to offer and their offering basket contains private (store level) in all categories offered, exclusive, national (manufacturers) and international brands. Moreover, electronics item superstores (e.g. Reliance Digital and NEXT) offer multi-brands in different categories of the merchandise and their offering basket contains all the leading (world’s most popular brands) national and international brands. The target market, goods & service growth potential, trends (fashion), image of retailer, customer segments and responsiveness to customers are some important factors for organised retailers, while planning merchandise and target market, retailer’s image, manufactures Vs private brands, the customer service offered and perceived goods/service benefits are some important factors for organised retailers while planning merchandise quality. organised retail industry The hyper store organised retailers (e.g. Big Megamart, HyperCITY, Star Bazaar, more.MEGAStore and Big Bazaar) adopt wide variety & a deep assortment strategy as a retail assortment strategy, which is characterised by many goods/service categories & a large assortment in each category, while, some super store organised retailers (e.g. D Mart and NEXT) opt narrow variety & a shallow assortment strategy characterised by few goods/service categories & a limited assortment in each category and a few others (e.g. West Side, Max and Reliance Digital) adopt narrow variety & a deep assortment strategy characterised by few goods/service categories & a large assortment in each category. Multi-format organised retailers (e.g. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart) offer staples (basic), fashion, consumer durable and seasonal merchandise category to their customers. Size, price, quality, brand, customer segments, discounts and schemes are the most common base to categorise the merchandise while displaying in the store. Fashion-item organised retailers (e.g. Big Megamart, West Side and Max) offer fashion merchandise category 366 to their customers and use size, colour, design, price, quality, brand, customer needs, discounts and schemes as a base to categorise the merchandise while displaying in the store. And electronics-item superstores (e.g. Reliance Digital and NEXT) offer consumer durables merchandise category and categorise the merchandise based on size, design, price, quality, brand, customer segments, customer needs, discounts and schemes to display in the store. The organised retailers make sure the availability of newly launched merchandise/brands as soon as possible if falls in the category of the merchandise they offer. Moreover, the retailers (e.g. HyperCITY, Star Bazaar, more. MEGAStore, Big Bazaar, Big Megamart, West Side, Max and Reliance Digital) also offer customised, unique and complementary merchandise in a few categories. As a management of unprofitable merchandise, some organised retailers (e.g. Big Bazaar, more.MEGAStore, West Side, Max and Reliance Digital) try to find the way to convert unprofitable merchandise into profitable merchandise and then determine that whether unprofitable merchandise relates to profitable business relationships or not; if not then they drop them from the assortment. While, other organised retailers (e.g. HyperCITY, Star Bazaar, D Mart and NEXT) do not try to find the way to make them profitable, but directly determine that whether unprofitable merchandise relates to profitable business relationships or not; if not then they drop them from the assortment. However, some other (e.g. Big Megamart) usually drops unprofitable merchandise from the assortment. organised retail industry Manufacturers are first major suppliers of all the organised retailers while wholesalers are second major suppliers of all the organised retailers except few (e.g. West Side, Max and Reliance Digital) and merchandise brokers are third major suppliers of the organised retailers except few (e.g. Big Megamart, Star Bazaar and D Mart). However, most of the organised retailers also rely on commission agents to supply international brands except few (e.g. Star Bazaar, Big Bazaar and D Mart). Moreover, few other (e.g. Star Bazaar) also gets supplies for some cooperatively & individually owned offices. All the organised retailers get a trade, quantity and seasonal discounts from their suppliers. Moreover, most of the organised retailers also get 367 promotional discounts except few (e.g. HyperCITY, Big Megamart and D Mart). The organised retailers consider reliability, price-quality, order-processing time, guarantee, long-term relationship, reorders and innovativeness as the most important characteristics for selecting a vendor for the supply of the items of their assortment. Both hyper and superstore organised retailers show a keen interest in the partner relationship management. They maintain the long-lasting relationship with all their partners, especially suppliers / vendors. The organised retailers prefer weekly and monthly stock planning of merchandise. However, multi-format organised retailers (e.g. HyperCITY, Star Bazaar, more. MEGAStore, Big Bazaar and D Mart) also do the daily planning in case of perishable merchandise, frozen, vegetables, fruits, other farm merchandise, non-vegetables, etc. Most of the organised retailers (e.g. HyperCITY, Star Bazaar, D Mart, Max, Reliance Digital and NEXT) adopt maximising market share as pricing objective, while, other organised retailers (e.g. more.MEGAStore, Big Bazaar, Big Megamart and West Side) adopt merchandise quality leadership as a pricing objective. Most of the organised retailers (HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar, Max, and Reliance Digital) adopt leadership-oriented pricing approach and others (e.g. Big Megamart, D Mart, West Side and NEXT) adopt cost-oriented pricing approach in order to price the merchandise. High-low pricing strategy (a form of value pricing strategy) is the most popular pricing strategy among organised retailers (e.g. HyperCITY, Star Bazaar, D Mart, Big Megamart and NEXT) while other retailers (e.g. more.MEGAStore and Big Bazaar) adopt another form of value pricing strategy, i.e. everyday low pricing strategy. However, perceived value-pricing strategy is also popular among superstore organised retailers (e.g. West Side, Max and Reliance Digital) in order to keep a promise to provide customers well-difference value & experience in their stores. organised retail industry Discount & allowance pricing (i.e. reducing the prices to reward customer responses), psychological pricing (i.e. adjusting prices for psychological effects) and promotional pricing (i.e. temporarily reducing the prices) are most common price-adjustment strategies among both hyper and superstore organised retailers to adjust or reduce the price of the merchandise and services. Value added retailing is most preferred tactics by 368 all the organised retailers except few (e.g. D Mart and NEXT) as they follow to maintain and grow tactics to fight a price war. No organised retailers take concern of customer before changing the price of merchandise & services, but all these retailers take the response of customers through feedback and consider for the future course of action except few (e.g. D Mart and NEXT). The organised retailers (e.g. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar, Max and Reliance Digital) always charge less than the MRP of most of the merchandise due to leadership-oriented approach, however, some merchandise is still sold at MRP. While some other organised retailers (e.g. Big Megamart, D Mart, West Side and NEXT) charge less than the MRP on most of the merchandise, but due to cost-oriented approach, they charge equal to MRP to recover the margin on the merchandise. The organised retailers (e.g. HyperCITY, more.MEGAStore, Big Bazaar and Reliance Digital) appeal to reason; some others (e.g. Star Bazaar, D Mart, Max and NEXT) appeal to emotion while others (e.g. Big Megamart and West Side) appeal to sense in order to create a brand image among targeted customers. All the organised retailers use advertising, sales promotion, direct retailing, personal retailing, public relation / publicity and internet / interactive retailing methods for developing brand image and customer loyalty among existing and potential customers. The organised retailers do both indoor (in-store) as well as outdoors (out-store) advertising to communicate and promote their offerings and include visual display, signage, the point of purchase material, secondary packaging and cross promotion methods of in-store advertising. However, some retailers (e.g. Star Bazaar, more.MEGAStore, Big Bazaar and Reliance Digital) also use pamphlets, while some others (e.g. Big Megamart, HyperCITY, Big Bazaar and Max) do loud speaker announcement about various discounts & schemes for the day in order to advertise. Moreover, others (e.g. Big Megamart, Big Bazaar and Reliance Digital) also use electronic signage to advertise within the store. organised retail industry Yellow page advertising, the newspaper and hoardings advertising are most preferred modes of outdoor advertising among the organised retailers. Moreover, online advertising is also a common method adopted by organised retailers except a few. However, some retailers (e.g. Big Megamart, Big Bazaar, D Mart, West Side, Max and Reliance Digital) also advertise in magazines & 369 journals. Some other methods of outdoor advertising are also used by organised retailers i.e. television ads by few (e.g. Big Bazaar, West Side, Max and Reliance Digital); radio ads by few (e.g. Big Bazaar, D Mart and Reliance Digital). The organised retailers spend less on advertising in comparison to any other industry especially outdoor advertising. Schemes, gifts, rebates, discounts, loyalty programmes, premiums, prizes, rewards and contests are some popular tools opted by the organised retailers for sales promotion. However, some organised retailers (e.g. Big Megamart, HyperCITY, Star Bazaar, more.MEGAStore and Big Bazaar) also provide a free sample as an integral part of the sales promotion. Buy & get free, bulk purchase discount, value / combo pack (offer) discount / deal, percentage off (discount) on MRP, up to percentage price off, flat percentage off / discount, price ( ) off / discounts, buy more and pay less (get more), flat price, starting at ₹, only / just for ₹, save ₹, ₹ onward / approx, fixed price scheme, ₹ to ₹, limited offer, offer / deal of the day or today’s offer, store (our) price and special offer schemes are some common sales promotion schemes introduced by the organised retailers. Moreover, organised retailers also introduced a membership scheme and exclusive schemes for customers who wish to enrol for membership except few (e.g. D Mart) in order to create some loyal customer base. The organised retailers give free gifts on the purchase of specific items, assured gifts and premiums to their customers while promotion. Moreover, some organised retailers (e.g. Big Bazaar, more.MEGAStore, Reliance Digital and NEXT) also offer the EMI facility to their customers in order to promote sales. The organised retailers think that the sales promotion influence customer purchase intention, therefore they do adequate sales promotional activities in the store to influence the customer purchase intention. Direct emails and mail order catalogues and SMS retailing are most common direct retailing techniques adopted by the organised retailers except a few. However, some organised retailers (e.g. organised retail industry HyperCITY, more.MEGAStore, Big Bazaar, Reliance Digital and NEXT) are also involved in tele-retailing / mobile retailing form of direct retailing. The organised retailers adopt the face-to-face retailing technique of personal retailing within the store. Corporate brochures, annual reports, press release, press conferences, video press conferences are some most common medium to maintain relationships with the public. However, some organised retailers (e.g. HyperCITY, 370 more.MEGAStore, Big Bazaar, West Side, Max and Reliance Digital) organise and sponsor special events. Moreover, some multi-format organised retailers (e.g. HyperCITY, Big Bazaar, more.MEGAStore) and fashion item organised retailers (e.g. Big Megamart, West Side, Max) conduct social events & social development programmes for publicity. The organised retailers use e-mails and websites as interactive retailing tools in order to create a brand image among the internet users. The organised retailers also utilise the social media platforms to interact with their techno savvy customers except few (e.g. D Mart) to create a brand image. Online feedback is also a popular tool of interactive retailing among the organised retailers except few (e.g. Big Megamart and West Side). Moreover, electronics super store organised retailers (e.g. Reliance Digital and NEXT) take online feedback on Facebook. However, some other organised retailers (e.g. HyperCITY, Star Bazaar and Max) use the online discussion forum as interactive retailing tools to create a brand image among customers. Use of celebrity services is not so popular among organised retailers to create a brand image. However, some organised retailers (e.g. Big Bazaar, Big Megamart, West Side and Max) use celebrity services for promotions. The organised retailers use the tagline and logo for creating a unique brand image among the customers. The most of the organised retailers use customer-centric tagline except few (e.g. HyperCITY, West Side, Reliance Digital and NEXT) as these others use the store-centric tagline. The organised retailers prefer main street locations to open their stores except few (e.g. Big Megamart and HyperCITY), as these others prefer inner city locations. However, some others (e.g. Max) prefer both types of locations. The organised retailers open their store in shopping malls / shopping centres and / or strip centres. However, some fashion item superstores (e.g. West Side and Max) also prefer to open their stores in outlet centres. organised retail industry Total size & density of population, total disposable income of the population, per capita disposable income of the population, trends among the population, delivery cost of the supply, timeliness for supply, number of manufacturers & wholesalers in the area, availability & reliability of merchandise lines, the growth projections in the area, freedom from economic & seasonal fluctuations, number and size of existing competitors in the area, strength & weakness of competitors, number and type of 371 locations in the area, cost of site, visibility of site, size & shape of trading area and parking facilities are some important factors for selecting a retail store location by the organised retailers. There is no uniformity among organised retailers in store timings. Some organised retailers (e.g. HyperCITY and Big Megamart) keep open their stores 10.30 hours, some others (e.g. Star Bazaar) keep open stores for 14 hours, few others (e.g. more.MEGAStore) keep open stores for 13 hours, some retailers (e.g. Big Bazaar, D Mart) keep open stores for 12 hours, other retailers (e.g. Max, Reliance Digital) keep open stores for 11 hours on all the days of the week, while others (e.g. West Side) keep open stores at 11.30 hours on weekdays and 11.00 hours on Sunday, moreover, few others (e.g. NEXT) keep open stores for 11.30 hours on all the days of the week except Monday as it is closed. The most of the organised retailers adopts patronage solidifiers customers service strategy and involved in the low-cost little things that increase loyalty, like courtesy & suggestion selling) except some (e.g. more.MEGAStore, Reliance Digital and NEXT) as these others adopt patronage builders customers service strategy and involved in high-cost activities like transaction speed & credit. The organised retailers (whether multi-format, fashion item or electronics item organised retailers) offer attractive exchange & a return policy to protect their customers and gain their confidence. The organised retailers use gap model to understand knowledge, standard, delivery and communication gaps to improve customer service quality except few retailers (e.g. Star Bazaar, NEXT and D Mart). The organised retailers interested to have a CRM and maintain the same through the customer service desk or customer loyalty desk at the store level. The multi-format hyper store organised retailers (e.g. HyperCITY, Star Bazaar, more.MEGAStore and Big Bazaar) and fashion item organised retailers (e.g. organised retail industry Big Megamart, West Side and Max) collect information for database regarding customer transactions (purchase date, price paid, merchandise purchased, responses to promoted merchandise), customer contacts (record of customer interaction with store by websites, inquiries in-store kiosks, telephone calls), customer preferences (favourite colour, brands, fabrics, flavours, and apparel size), descriptive information (demographic & 372 psychographic data of customers) and customer responses to marketing activity (responsiveness to sales promotion & advertising), while electronics-item organised retailers (e.g. Reliance Digital and NEXT) collect information for database regarding customer transactions, customer contacts and descriptive information only and multi- format superstore organised retailers (e.g. D Mart) collect information for database regarding customer transactions only. The organised retailers collect the information for the database by keeping records of the customer transactions. These organised retailers also conduct the in-store survey through printed questionnaire and collect information under a membership programme except few (e.g. D Mart). Moreover, some organised retailers (e.g. more.MEGAStore, Big Bazaar, Reliance Digital and NEXT) also conduct an online survey in order to collect information; while some others (e.g. Max and Big Bazaar) collect the information by taking customer feedback & suggestions. The organised retailers use CRM programmes in order to maintain a relationship and create loyal customers, except few (e.g. D Mart). The organised retailers also have a separate cell (customer support cell) to help customers in store and they all use listening and responding approach, respond to the customers with a courtesy, maintain speed and accuracy of service in order to help their customers The most of the organised retailers deals with retail clientele only except few (e.g. Reliance Digital), who deal with both retail as well as corporate clientele. The organised retailers face the problems like changing the landscape of Indian consumerism, changing retail trends, changing government role, stiff competition, cultural disparity, high demand for service standards, the price war while dealing with key customers. Some of the store based organised retailers (e.g. Big Bazaar, Reliance Digital and NEXT) also provide e-retailing service to their customers. The multi-format hyper store organised retailers (e.g. HyperCITY, Star Bazaar, more.MEGAStore and Big Bazaar) use a mix of free form or boutique layout and grid layout to display the merchandise while planning store design, while multi-format superstore organised retailers (e.g. D Mart) use only grid layout to display the merchandise while planning store design. Fashion-item organised retailers (e.g. organised retail industry Big Megamart, West Side and Max) use the free form or boutique layout to display the 373 merchandise while planning store design and superstore based electronics-item organised retailers (e.g. Reliance Digital and NEXT) use racetrack layout to display the merchandise while planning store design. The organised retailers use category, promotional and point of sale signage for locating the merchandise category in their stores. Moreover, some organised retailers (e.g. Big Megamart, more.MEGAStore, Big Bazaar and Reliance Digital) also use digital signage. Freestanding displays, end caps, promotional aisles, wall displays and cash wrap displays (point of purchase counters or check out areas) are most preferred methods of displaying the merchandise in the stores among the multi-format organised retailers. However, some multi-format organised retailers (e.g. HyperCITY, more.MEGAStore and Big Bazaar) also use window display and entrance display methods of displaying the merchandise in their stores. Fashion-item organised retailers (e.g. Big Megamart, West Side and Max) use window display, entrance display, free standing display, end caps, promotional aisles and wall display methods of displaying the merchandise in their stores. However, some fashion item organised retailers (e.g. Max) also use a dressing room display method. Superstore based electronics-item organised retailers (e.g. Reliance Digital and NEXT) use entrance display, freestanding display, promotional aisles and walls display methods of displaying the merchandise in their stores. However, some other (e.g. Reliance Digital) also uses a cash wrap display method. The multi-format organised retailers (e.g. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart) use straight racks, rounder (capacity fixture), four-way fixture (feature fixture) and gondolas, while fashion-item organised retailers (e.g. Big Megamart, West Side and Max) use rounder (bulk fixture/capacity fixture), four-way fixture (feature fixture) and gondolas and superstore based electronics-item organised retailers (e.g. Reliance Digital and NEXT) use only straight racks and rounder (bulk fixture/capacity fixture) to display the merchandise in the stores. The organised retailers use item & size presentation, price lining, vertical merchandising, tonnage merchandising and frontal presentation techniques, however, some organised retailers (e.g. Star Bazaar, Big Bazaar, West Side, Max, Reliance 374 Digital and NEXT) also use an idea-oriented presentation technique to present the merchandise in the store. Some hyper store based multi-format organised retailers (e.g. HyperCITY, more.MEGAStore and Big Bazaar) and fashion item organised retailers (e.g. Big Megamart, West Side and Max) use lighting, colour, scent and music, while some other hyper store based multi-format organised retailers (e.g. organised retail industry Star Bazaar) use only lighting and music for creating an appealing store atmosphere. Moreover, multi-format superstore organised retailers (e.g. D Mart) use only music to create an appealing store atmosphere and electronics items organised retailers (i.e. Reliance Digital and NEXT) use lighting, colour, and music to create an appealing store atmosphere. Visual merchandising, store layout, interior atmospherics and exterior atmospherics are important components of the store design for the organised retailers. Storefront, store marquee/sign board, visibility of the store, location/site of the store, parking facilities and ease of access are some important component of the exterior atmospherics of the store. Lighting, odour/smell/scent, sounds, temperature, cleanliness, trial (experience) room/zone, merchandise and store size are the important component of the interior atmospherics of the store. Space planning, floor space allocation, department size, the location of various merchandise, signage and space/merchandise categories are the important component of the store layout. Assortment, payment counters, merchandise display, the planogram, point of sale sign and methods of display are the important component of visual merchandising. retail industry Data pertained to residing area, gender, age, education, occupation, family income (yearly), marital status, nature of family, family size and religion of customers was accumulated for analysis of the profile of customers in the organised retail industry. The analysis of the data concludes the following: The organised retail industry is getting the customers from all the areas i.e. urban, semi-urban and rural areas. However, the majority of customers of organised retailers is from urban areas and semi-urban areas. Nevertheless, some retailers (e.g. 375 NEXT and Reliance Digital, D Mart, Star Bazaar, more.MEGAStore), locating their stores near to the rural areas are also able to tap an enough number of customers from rural areas. The organised retail industry is attracting the male and female customers in equal proportion with its offerings. However, the footfall of male customers is more in comparison to female customers. It is the fact the females are more fashion conscious in India. Not surprising, the fashion item retailers (e.g. Big Megamart, West Side & Max) attract more female customers; while, the electronics item retailers (e.g. NEXT and Reliance Digital) attract more male customers, which reflects that the purchase of an expensive item is still in the hand of males. The organised retail industry has something to offer to all the age groups. But, it is flourishing with the footfall of youth customers, whose age ranges from 21 to 40 years. The organised retailers are also attracting a great number of customers from the age group of 41 to 50 years with their offerings. However, the electronics item retailers (e.g. NEXT and Reliance Digital) are getting their majority of customers from the age group of 31 to 60 years. The industry is attracting the customers of all education groups and has well- educated customers in the Maharashtra state as the state has 82.34 literacy rate. And, the majority of customers in the industry are graduates or post graduates. The organised retailers are attracting the customers from all the occupational groups. However, the majority of customers are salaried and self-employed and another majority is of students and homemakers. The organised retailers have something to offer to the customers of all the income groups. However, the majority of customers are from two income groups, i.e. 2 to 5 LPA and 5 to 10 LPA. Nevertheless, the retailers are also attracting the customers from the high-income group (i.e. retail industry more than 10 LPA) as well as low-income group (i.e. 1 to 2 LPA). 376 The organised retail industry is attracting both married as well as single customers. However, the industry has its more customers from the married marital status group. Nevertheless, the electronics item superstores (e.g. NEXT and Reliance Digital) and multi-format superstores (e.g. D Mart) are getting more customers from a married marital status group, while the fashion item retailers (e.g. Big Megamart, West Side, Max) as well as some multi-format hyper stores (e.g. HyperCITY and Big Bazaar) are attracting more customers from single marital status group. The offering basket of organised retailers has something for both nuclear as well as joint families customers. However, the industry is getting more customers from nuclear families. Nevertheless, some retailers (e.g. Star Bazaar, more.MEGAStore, D Mart, Reliance Digital and NEXT) are getting fewer footfalls from nuclear families. The retailers are also attracting the customers from all the family size groups of customers. However, the majority of customers are from the family size of below 5 members’ family size. The organised retailers are getting the customers from all the religions due to the diversity of the country. However, the majority of the customers are Hindus. The second majority is of Muslim customers, nevertheless, not surprisingly, the third majority is of Buddhists customers, as the Maharashtra state has 73.4% of the total Buddhism followers in India. the organised retail industry Data pertained to various factors / elements of customer behaviour were accumulated for analysis of the buying behaviour of customers in the organised retail industry. The analysis of the data concludes the following: The majority of customers in the organised retail industry is either self-purchase decision maker in the family or relies on their spouse and parents. The electronics-item superstores (e.g. NEXT and Reliance Digital) are getting more footfalls from the self- purchase decision-making customers. The primary reason of the same is that such retailers have more than 80% of customers are male. 377 The customers in organised retail industry collect the information about merchandise and services before taking a purchase making decision and their preferred sources for collecting the information are advertising, the internet, family members, neighbours / friends / relatives, package labels, in-store display, in-store information, salespersons and telemarketing. Good news for the retailers is that the customers also prefer to collect the information from sources like in-store display, salespersons and in- store information. The customers, who collect the information from advertising, rely on the television, websites, celebrity endorsement, newspapers, pamphlets, magazines, hoardings, radio and cinema and the purchase decisions TV talk shows, infomercials, family members, websites, celebrity endorsement, TV commercials, sales promotion, neighbours / friends / colleagues, self, a salesperson in a store, direct-mail piece and magazine ads. The customers of the organised retailers prefer to purchase the merchandise and service, either by self or rely on their spouse and parents. The customers of the multi- format retailers (e.g. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar and D Mart) visit the stores 3 or more times in a month, while customers of the fashion stores organised retailers (e.g. Big Megamart, West Side and Max) as well as the electronics item superstores (e.g. Reliance Digital and NEXT) visit the stores 1 or 2 times in a month. The reason for more visits in the multi-format stores is that all these offer staples categories merchandise to their customers. The customers prefer to visit the stores on weekends. But, it does not mean that the industry is not getting footfall on weekdays. The multi-format organised retailers get great footfalls of customers on all the seven days. Usually, customers prefer to visit the stores in the evening time. But, it does not mean that the industry is not getting footfall in morning and afternoon. the organised retail industry The customers of the fashion store retailers as well as electronics stores spend more than 5000 per month for purchasing from the ₹ stores while the customers of multi-format retailers spend 2000 to 5000 per month. The customers of the multi- ₹ ₹ format organised retailers purchases the products from all the categories of merchandise i.e. basic (staples) merchandise, fashion merchandise, seasonal merchandise and consumer durable, while customers of fashion stores (e.g. Big Megamart, West Side and Max) purchase the products from fashion merchandise 378 categories and customers of electronics item superstores (e.g. Reliance Digital and NEXT) purchases the products from consumer durables merchandise category. The good news for the industry is that the customers in the organised retail industry are impulse buyers. The multi-format hyper stores (e.g. HyperCITY, Star Bazaar, more.MEGAStore and Big Bazaar) and fashion stores (e.g. Big Megamart, West Side and Max) are attracting a great number of impulse buyers. Food & grocery, beauty, electronics, fashion products are the most common type of products purchased by impulse buyers from the stores of the multi-format organised retailers, while women’s wear, men’s wear, kid’s wear, footwears and accessories are the most common type of products purchased by impulse buyers from the stores of fashion store retailers and accessories are the most common type of products purchased by impulse buyers from the stores of electronics item retailers. The common reasons for impulse purchase are price reductions, multi-item discounts, the point of purchase display, demonstration and samples, salespeople, store atmosphere, store layout, coupons, packaging and shelving techniques. The customers’ first preference is either quality or brand while purchasing merchandise. However, the good news for the organised retailers is that there are also a great percentage of customers in industry, whose first preference is discounts, schemes, price, quantity and other while purchasing. the organised retail industry The customers of organised retailers visit their stores due to several reasons, most notably merchandise (merchandise assortment- variety of merchandise and brands, quality of assortment, in service situations, merchandise offered as part of service, styling or fashion guarantee, warranties and pricing, reasonable prices), promotion (advertising, displays, trading stamps, sales promotion- discount & special schemes, symbol and colours), store atmospheres (atmosphere of congeniality, customers’ feeling of warmth, acceptance or ease, good & spacious ambience, and delighting store environment), convenience (general convenience, parking access, location convenience- proximity to home/office), services (service in general, salesclerk service, presence of self-service, ease of merchandise return, delivery service, phone ordering and credit policy), physical facilities (elevators, lighting, air conditioning, restrooms, store layout, shopping ease, aisle placement, carpeting and 379 architecture), post-transaction satisfaction (merchandise in use, return & adjustment policy), institutional factors (projection of store- conservative or modern, reputation, reliability), clientele (social class appeal, store personnel, fit between self-image and store image- self-image congruence), other (friends suggested, no other store nearby, convenient store hours). The merchandise is the primary reason for visiting the stores among the customers of multi-format hyper store retailers (e.g. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar) as well as electronics-item superstores (e.g. Reliance Digital and NEXT). The promotion is the primary reason for visiting the store by customers of multi-format superstore retailers (e.g. D Mart). However, the store atmosphere is the primary reason for visiting the stores among the customers of fashion store retailers (e.g. Big Megamart, West Side and Max). The majority of customers in the organised retail industry, visit more than one retail store. However, the industry has a great number of loyal customers, who do not visit any other stores. The multi-format organised retailers (e.g. HyperCITY, Star Bazaar, more.MEGAStore, Big Bazaar, and D Mart) have more loyal customers. However, among the multi-format organised retailers, the hyper stores (e.g. HyperCITY, Star Bazaar, more.MEGAStore and Big Bazaar) have more loyal customers in comparison to a superstore (e.g. D Mart). The customers, who visit more than the regular one store, visit both organised as well as unorganised retail stores. However, those customers first preference is both organised as well as unorganised retail stores. the organised retail industry The customers visit the other retail stores due to several reasons, most notably for better services, finding the merchandise of their choice, better promotional offers, convenience, better store atmospheres and physical facilities, due to institutional factors like projection of store, reputation, reliability etc., due to other reasons like friends suggested, no other store nearby, convenient store hours etc., better post-transaction satisfaction like merchandise in use, return & adjustment policy. 9.1.5 Conclusions drawn from the analysis factors influencing buying behaviour of customers in the organised retail industry 380 Data pertained to various factors influencing the buying behaviour of customers was accumulated for analysis of factors influencing the buying behaviour of customers in the organised retail industry. The analysis of the data concludes the following: The sociological / environmental / external factors (e.g. cultural, sub-cultural, society, status in society, reference group, family life cycle and role in the family) have a significant influence on the buying behaviour of customers in the organised retail industry. However, the influence of sociological / environmental / external factors on buying behaviour of customers of electronics store retailers is less in comparison to customers of other retailer formats. This depicts that the buying decisions pertained to the consumer durables particularly electronics item and other technological items are less influenced by sociological factors. Nevertheless, the buying decisions pertained to the staples (basic) and fashion merchandise are more influenced by the sociological factors. The psychological / individual / internal factors (e.g. self-image, personality, personal values, lifestyle, memory, learning, perception, motivation, emotions, mood, involvement, information processing, beliefs, intention, attitude, communication, and persuasion) also have significant influence on the buying behaviour of customers in the organised retail industry and these factors influence the buying behaviour of customers of different formats (superstores or hyper stores) and type of retailers (e.g. multi-item, fashion item or electronics item). The store related factors (e.g. store image, store atmosphere, store brands, store location, sales personnel, discounts & schemes and free services) also significantly influence the buying behaviour of customers in the organised retail industry and these factors also influence the buying behaviour of customers of different formats (superstores or hyper stores) and type of retailers (e.g. multi-item, fashion item or electronics item). 9.1.6 Conclusions drawn from the analysis of customer satisfaction with the merchandise (products) and services offered by organised retail industry Data pertained to customer satisfaction with various merchandise and services was accumulated for analysis of customer satisfaction with the merchandise (products) 381 and services offered by organised retail industry. The analysis of the data concludes the following: The organised retail industry satisfies the customers with its customer expected merchandise assortment; merchandise categories (e.g. basic or staple, fashion, seasonal, consumer durable, etc.) in the assortment; the variety of merchandise in each category as per different sizes, colours, designs, price, quality, quantity and brands; availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise in each category; availability of the brands (i.e. local & regional, national, international, private / store, copycat, exclusive, generic brands) in each merchandise category; quality assurance and quality of merchandise; services and quality of the services; pricing of merchandise & services; overall value offered as compared to the price/cost of merchandise & services; exterior atmospherics, the interior atmospherics of the stores; store layout and visual merchandising; methods of promotion and communication (e.g. advertising, sales promotion, personal retailing and the internet or interactive retailing); sales personnel, contact personnel and factors pertained to personnel; and other tangible evidence (e.g. employee appearance, uniforms, billing statement, internet or web pages etc.). The hyper-stores based organised retailers leads in satisfying the customers as compared superstore based organised retailers. However, superstore retailers also perform better than hyper-stores in satisfying customers in some areas. The superstore retailers (e.g. D Mart, West Side and Max) satisfy more customers with their pricing of merchandise & services as compared to their hyper store counterparts. The fashion item superstore based retailers (e.g. West Side and Max) performs better in satisfying customers with their sales promotion methods, public relation/publicity methods, internet/interactive retailing methods as compared to their hyper store counterparts (e.g. Big Megamart). The electronics-item superstore retailers (e.g. Reliance Digital and NEXT) also perform better than the hyper store similar item retailers (e.g. HyperCITY, Star Bazaar, more.MEGAStore and Big Bazaar) in satisfying customer with the availability of newly launched merchandise, customised merchandise, unique merchandise and complimentary merchandise, brands in each category, pricing of merchandise & services and direct retailing. 382 Some superstore retailers (e.g. 9.1.6 Conclusions drawn from the analysis of customer satisfaction with the merchandise (products) and services offered by organised retail industry D Mart, Max, Reliance Digital and Next) are lacking to match the industry standards in satisfying their customers in an area like the quality of services, exterior and interior atmospherics of the store, store layout, personal retailing methods, factors pertained to personnel, other tangible evidence, etc.. Moreover, even some hyper store retailers (e.g. Star Bazaar and more.MEGAStore) are also not matching the industry standards in satisfying their customers in an area like store layout, visual merchandising, factors pertained to personnel, etc. Therewith, adequate individual customers of different retail stores are dissatisfied or disappointed with the merchandise and services offered. Those products & services have been discussed in the recommendation section for better understanding and recommendation to the concerns. 9.1.6 Conclusions drawn from testing of hypotheses The first null hypothesis stated that “retailing strategies introduced by organised retailers are not significant to sustain in the hyper-competitive retail market”. This hypothesis was framed under certain assumptions for which no validation required as it is a common noticeable phenomenon that “retailing strategies introduced by organised retailers are significant to sustain in hyper-competitive retail market”. However, this statement is supported by product portfolio analysis (presented in chapter 5) where the significant majority of merchandise (products) of all the organised retailers belongs to Star category in the BCG matrix. Moreover, this statement is also supported by the customer satisfaction with various merchandise (products) and services offered by all the organised retailers. Therefore, the first null hypothesis is rejected and the alternate hypothesis, i.e. “retailing strategies introduced by organised retailers are significant to sustain in hyper-competitive retail market” is accepted. However, the retailers have to travel a long distance to meet the international standards in organised retailing. The second null hypothesis stated that “customers are not significantly satisfied with various products & services introduced by organised retailers”. The results of hypothesis testing depict that customers are significantly satisfied with various products & services introduced by organised retailers except direct retailing methods and public relation/publicity methods. Therefore, the second null hypothesis is also rejected, which 383 signifies that the organised retailers offer various products and services as per the expectation of their targeted customers. Therewith, it would be better to mention that the customers of some retailers (e.g. D Mart, Max, Reliance Digital, NEXT, Star Bazaar and more.MEGAStore) are not significantly satisfied with some products & services introduced by them. Those products & services have been discussed in the recommendation section for better understanding and recommendation to the concerns. 9.2 RECOMMENDATIONS This sub-section comprises the recommendations based on the analyses of the data collected from the organised retailers and customers of them for the betterment of the industry and to improve the level of customer satisfaction. 1) Some merchandise lines of the organised retailers fall in the category of Question Mark in the BCG matrix, e.g. HyperCITY’s electronics and toys; 1) Some merchandise lines of the organised retailers fall in the category of Question Mark in the BCG matrix, e.g. HyperCITY’s electronics and toys; 384 more.MEGAStore’s general merchandise, mobile phones & accessories and computers & accessories; Big Bazaar’s bed & bath bazaar, home bazaar, mobile bazaar and furniture bazaar; D Mart’s home appliances and bags & trolleys; Big Megamart’s home mart and travel mart; West Side’s home & gift; Max’s children’s wear and footwear, Reliance Digital’s mobile phones and computers; and NEXT’s computers & tablets and personal care. The concern retailer should select few types of merchandise to convert them into the Star category, or they may also divest them from the assortment. 2) Few more merchandise lines of the organised retailers also fall in the category of Dog in the BCG matrix, e.g. HyperCITY’s sports; Star Bazaar’s international zone; more.MEGAStore’s fruits & vegetables and sports; Big Bazaar’s farm fresh and books, magazines & stationery bazaar; D Mart’s frozen, toys & games and stationery; Big Megamart’s toys; West Side’s luggage; Reliance Digital’s gaming; and NEXT’s fragrances and writing Instruments. The concern retailer should liquidate these merchandise lines from the assortment. This action will assist them to introduce new merchandise lines on the place of the existing one, which are not attracting the enough customers. In this way, the retailers can increase the sales and the satisfaction of the customers. 2) 3) The industry is not getting an adequate number of the customers from rural areas. The retailers should offer the product and service needed by the rural area people. E.g. the organised retailer can introduce the farm and agriculture merchandise line for the rural area farm background customers, under this line can offer the merchandise like seeds, insecticides, pesticides, fertilisers, small farm equipment, etc. to attract the rural area customers. Similarly, retailers can also offer the merchandise line for pet animals of rural area people. 3) 4) The organised retail industry is also getting fewer customers from the age group of below 21. 9.2 RECOMMENDATIONS The retailers should update their assortment by adding books, study materials; other items needed by that group, etc. to attract this age group customer. The organised retail industry is also getting fewer customers from the family income group of less than 2 LPA. The retailers should communicate and 385 promote their merchandise and service among them for winning their confidence and attracting them to the stores. 5) The customers in the organised retail industry collect the information about the merchandise and services from various sources, e.g. advertising, internet, family members, neighbours / friends / relatives, package labels, in-store display, in- store information, salespersons, telemarketing, etc. The retailers need to make sure the availability of the adequate information through all the sources. The customers in the organised retail industry prefer to visit the stores on weekends. The retailers need to entice them on weekdays too by offering better promotional schemes, eat, play and shop environment. 6) The customers are also impulse buyers. The retailers need to cash their impulse buying habits by offering better services, sales promotion, visual merchandising and employing helping sales people. The customers in the organised retail industry also visit more than one store for better merchandise and services. The retailers should make them loyal by matching their expectation. 6) The customers are also impulse buyers. The retailers need to cash their impulse buying habits by offering better services, sales promotion, visual merchandising and employing helping sales people. The customers in the organised retail industry also visit more than one store for better merchandise and services. The retailers should make them loyal by matching their expectation. 7) The sociological, psychological and store related factors significantly influence the buying behaviour of the customers. The retailers should offer something unique for every customer according to their expectations to in cash the influence of the various factors on buying behaviour. 8) The industry needs to ensure the full support to their customers and customers should realise the same. It will help the industry to gain the confidence of the customers and they will not only be loyal but also spread positive word of mouth. 8) The industry needs to ensure the full support to their customers and customers should realise the same. It will help the industry to gain the confidence of the customers and they will not only be loyal but also spread positive word of mouth. 9.2 RECOMMENDATIONS 9) The industry needs to reward good customers through the special promotional schemes e.g. deposit schemes with guaranteed rebates, special discounts to loyal customers, the gift for special occasions, etc. This will also assist the retailers to create a great number of loyal customers. 9) The industry needs to reward good customers through the special promotional schemes e.g. deposit schemes with guaranteed rebates, special discounts to loyal customers, the gift for special occasions, etc. This will also assist the retailers to create a great number of loyal customers. 386 10) The problems of the customers need to be handled efficiently. The contact and sales personnel should be capable enough to understand the questions and concerns of the customers as well able to answer questions. 10) The problems of the customers need to be handled efficiently. The contact and sales personnel should be capable enough to understand the questions and concerns of the customers as well able to answer questions. 11) Today’s customers want a quick response to their problems and issues through all the mediums. The retailers should ensure the same through all the mediums like phone, email, social media, etc. The stores must be reachable through all the mediums. 12) The industry is revolutionising the organised retail sector of India. But, many players in the industry adopt only some traditional promotional methods for communication with their customers. For better promotion and communication, the industry should integrate the retail communication mix and need to include the various tools from advertising, sales promotion, personal retailing, direct retailing, internet retailing and public relation methods. 13) Although organised retailers consider the customer relationship as the prime goal to attract the customers and maintain good customer relationship. However, many players, especially superstore based retailers (e.g. D Mart and NEXT) lack to maintain the customer relation (CR). These players need to maintain a customer relationship management department for analysing and applying customer relationship at the individual level. 14) The most organised retailers in the industry concentrate only on the retail clientele. These retailers can also target of the corporate / business clientele to grow their business in this segment. 15) The retail marketing mix is the key element or vital role player for satisfying customers. Many players in the industry (e.g. D Mart and NEXT) not giving equal importance to every element of the mix. 9.2 RECOMMENDATIONS The retailers need to give equal importance to every element of mix whether it is merchandise or price, promotion, premises, presentation, personnel. 387 16) Store level planning is the important decision to increase the efficiency of the particular store of an organised retailer. Many small players (e.g. D Mart, Max and NEXT) do not do much planning at store-level. The retailers need to use the different and diversified tools available for store-level planning to increase the efficiency and effectiveness of the store. The BCG matrix should be used for merchandise portfolio analysis. ETOP and PEST analysis for better understanding of the internal as well as external environmental factors should be used while five force analysis for understanding the competition and its intensity. 16) Store level planning is the important decision to increase the efficiency of the particular store of an organised retailer. Many small players (e.g. D Mart, Max and NEXT) do not do much planning at store-level. The retailers need to use the different and diversified tools available for store-level planning to increase the efficiency and effectiveness of the store. The BCG matrix should be used for merchandise portfolio analysis. ETOP and PEST analysis for better understanding of the internal as well as external environmental factors should be used while five force analysis for understanding the competition and its intensity. 16) 17) Although, the organised retailers offer good merchandise categories to their customers, but the customers of some retailers (e.g. more.MEGAStore, D Mart, Max and NEXT) are not happy with the merchandise categories and a variety of merchandise in each category offered by them. Therefore, these retailers first need to revise their merchandise categories as per the expectations of the customers and then should offer the variety of merchandise in each category as per different sizes, colours, designs, price, quality, quantity and brands. 17) Although, the organised retailers offer good merchandise categories to their customers, but the customers of some retailers (e.g. more.MEGAStore, D Mart, Max and NEXT) are not happy with the merchandise categories and a variety of merchandise in each category offered by them. Therefore, these retailers first need to revise their merchandise categories as per the expectations of the customers and then should offer the variety of merchandise in each category as per different sizes, colours, designs, price, quality, quantity and brands. 9.2 RECOMMENDATIONS 18) The industry also makes sure availability of the newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise in each category. However, few retailers (e.g. Star Bazaar and D Mart) lack to match the industry standards, hence need to improve the availability of the newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise in each category offered by them to enhance the customer satisfaction level. 18) The industry also makes sure availability of the newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise in each category. However, few retailers (e.g. Star Bazaar and D Mart) lack to match the industry standards, hence need to improve the availability of the newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise in each category offered by them to enhance the customer satisfaction level. 18) 19) Although, the organised retailers offer a wide range of brands in each category, but the few (e.g. more.MEGAStore and D Mart) are lacking to satisfy the customers with their brand offerings. These retailers strongly need to re-shuffle their brand offerings from local & regional brands, national brands, international brands, private (store) brands and others, i.e. copycat, exclusive, generic brands. 19) 388 20) Although, the organised retailers offer a wide variety of services (e.g. credit, home delivery, alteration, installation, packaging, complaints & returns handling and may more) to their customers. But, the customers are not happy with some services of the few retailers (e.g. D Mart and Max). These retailers should re-check their services and the quality of their services. The retailers must ensure the standards of the services and make sure that their personnel are able to manage tangibility of the services, the reliability of services, responsiveness in the services, able to assure the quality and empathy while providing the services. 20) Although, the organised retailers offer a wide variety of services (e.g. credit, home delivery, alteration, installation, packaging, complaints & returns handling and may more) to their customers. But, the customers are not happy with some services of the few retailers (e.g. D Mart and Max). These retailers should re-check their services and the quality of their services. The retailers must ensure the standards of the services and make sure that their personnel are able to manage tangibility of the services, the reliability of services, responsiveness in the services, able to assure the quality and empathy while providing the services. 9.2 RECOMMENDATIONS 20) 20) 21) The organised retail industry is trying very hard to enhance the shopping experience of the customers and also getting the success in the same. However, few retailers (e.g. D Mart, Max and NEXT) are inefficient to satisfy their customers with the exterior and interior environment of the stores. The need of the hour is to provide an appealing store environment to keep enticing the customers. Therefore, these retailers need to rethink and recreate the exterior and interior environment of their stores. First, retailers should create an appealing exterior environment through an attracting storefront, sign board, store nameplate, frontage entrance and external display window. The uniqueness, visibility and location of the store also play an important role in creating an appealing exterior environment. Then, retailers should make sure an appealing interior environment through an attracting lighting, odour / smell / scent, sounds, colour, temperature, fixtures, wall textures, flooring & ceiling, cleanliness, aisles, trial room, and store size. 21) The store layout and visual merchandising play an important role in attracting the customers; easily locate merchandise, space management and presentation of the merchandise. Although, the industry inclined to match the international standards in visual merchandising, but some retailers (e.g. Star Bazaar, more.MEGAStore, D Mart, Max and NEXT) still least bother about layouts and visual merchandising planning, as their customers are not much satisfied. Therefore, these retailers need to improve their store layout planning by doing proper space planning, floor space allocation, e-display, managing a department 22) 389 location, department size, department compatibility, the location of various products, accessibility to cash/billing counters and signage. The visual merchandising needs to comprise the attractive themes, racks & shelves, ensemble, merchandise display, the point of sale sign and methods of display. 23) Although, the organised retailers are satisfying their customers through the adoption of promotional and communication methods. Nevertheless, retailers need to improve the some communication and promotional methods. E.g. the advertising is not considered much important in this industry as compared to other. But, the retailers should not forget the power of the communication through the advertising methods and should adopt some in store as well as out- store methods of communication. 9.2 RECOMMENDATIONS The in-store methods are visual display, signage, pamphlets, point of purchase material, secondary packaging and cross- promotion, while out-store methods can be selected from television, radio, newspapers, journals & magazine, yellow pages, hoardings, wall paintings, pamphlets/propaganda, van advertising, online advertising, mobile billboard, LCD screening, cinema, fairs & exhibitions. 23) 24) The many customers are dissatisfied with the direct retailing and public relation methods of communication adopted by the industry. The retailers need to adopt some new methods of direct retailing (e.g. direct mail, tele-retailing, direct response on TV/radio, mail order catalogues, SMS retailing and mobile retailing) and public relations (e.g. corporate brochures, posters, annual reports, press release, press conference, video press conference, internet release, store tours for media, meeting with customers, special events & sponsorships, social events, social development programs). 24) 25) Today, the people spend much time on the world wide webs. The organised retailers also understand the fact and making their presence on the internet through the website, social media platforms, weblogs, discussion forums and emails. Nevertheless, some retailers (e.g. D Mar) are poor in making their effective presence on the world wide webs and social media platforms like 25) 390 Facebook, Twitter, Youtube, etc. These retailers need to match their steps with the need of the hours to be successful in the long-run. Facebook, Twitter, Youtube, etc. These retailers need to match their steps with the need of the hours to be successful in the long-run. 26) The people are an integral part of any business and play an important role in satisfying customers. In the retail business, the two categories of people have direct contact with the customers, i.e. sales people and contact people (e.g. help desk people, service desk people, billing counter people etc.). These people’s mood, efforts, commitment, attitude, knowledge, skills, behaviour, approach to delivering the service, competence, courtesy, credibility, reliability, responsiveness, communication style, friendliness, helpfulness, discipline & punctuality, counselling ability has the direct relation with the satisfaction of the customers. Although, the retailers employ capable and trained the people to deal with the customers. But, some retailers (e.g. Star Bazaar, more.MEGAStore, D Mart, Max and NEXT) lack to so and need to improve this area. 26) 27) Some organised retailers (e.g. Star Bazaar, more.MEGAStore, D Mart, Max and NEXT) also need to improve the management of other tangible evidence like employee appearance, uniforms, billing statements, the internet or web pages, etc. 9.2 RECOMMENDATIONS 28) The customers suggested some facilities required to introduce by retailers like deposit schemes with guaranteed rebate, mail & phone orders, internet facility (wifi), mobile and laptop charger points, product on demand, information about discounts & schemes through SMS and Whatsapp, internet billing facility, medical facility and insurance facility. 29) The customers also think that there is an urgent need to improve the following services: parking facility is very poor in case of some retailers, especially superstore retailers, and even it is also paid service. The retailers charge an amount for shopping bags after February 4, 2011, as per the notification of the Ministry of Environment & Forest, Government of India, however, the customer are not happy with it. The customers also face the problems at baggage counters for the safety of personal things due to large queues. The after sale services on durable purchased from the retailers are provided by the 391 manufacturers. The customers complain that they do not get adequate mediator support from the retailers. The customers also face long queues at cash/billing counters, which really annoy them sometimes and expect less than five minutes waiting time in the queue. The female customers are not happy with the trial rooms’ management. Where, they are allowed to carry only a few items to try at a time. The customers also found that the attrition rate of the sale and contact personnel is high in the industry. Therefore, the customers have to listen from the personnel that they are new and recently joined the store and not aware about the different policies of the retailers. 30) Organised retailers face the closest competition from the unorganised retailers of the Industry. To beat the competitions effectively, organised retailers need to offer the unmatched merchandise assortment at an unmatched price and continue to improve the shopping experience of the customers through the adoption of international standards in retailing. 30) 31) Online retailing is the biggest threat for the store based organised retailers. To face this threat, the retailers need to continue with their shop, eat and play concept of retailing. The organised retailers need to continuously improve the refreshment and entertainment facilities in their stores along with the merchandise and services. Moreover, organised retailers should also introduce and improve (those already introduced) phone, email and web order facility for their customers. 9.2 RECOMMENDATIONS The organised retailer should also launch their free mobile app, through which the customers can not only place their orders, but also get the information about the latest promotional schemes and discounts. 31) The organised retailers face many problems like the high price of real estate, lack of preferred locations and retail space, high cost of space and location, trained workforce, multiple & a complex taxation system, many entrants as online retailers, high demand for service standards, inefficiency of the supply chains, lack of adequate infrastructure and demand for higher wages by the workforce, etc. The stakeholders in the industry like Government, real state suppliers, workforce suppliers, other supply chains and others should understand the issues of the organised retailers because the support of all the 32) 392 stockholders will lead to a solid foundation towards the growth of the organised retail sector. 9.3 AREAS FOR FURTHER RESEARCH Future research should consider this study for other sectors like banking, insurance, telecom, automobile, etc. The researchers can study the marketing / retailing strategies for products and services in these suggested areas. Another area of research would be a cross-comparison of retailing strategies of unorganised retailers and organised retailers. 393 This is the study of store-based retailers. The future research should carry this study with the e-retailers. The comparative study of store-based and e-retailers would also be carried. In this study, only two format retailers (i.e. hyper store and superstore) have been selected, in future research another formats (e.g. convenience store, discount stores, speciality stores, department stores) can be considered. In this study, three different item retailers (multi-item, fashion item and electronics item) were taken into the consideration, in future research another kind of item retailers can be taken into the consideration. 394 Supportive Information  Bibliography  Appendix  Bibliography  Appendix BIBLIOGRAPHY 1. Admin (2011), “PEST analysis”, Crackmba, available at: http://crackmba.com/pest-analysis. at: 2. Altintas, Murat Hakan; Kiliç, Serkan; Senol, Gokhan and Isin, Feride Bahar (2010), "Strategic objectives and competitive advantages of private label products: Manufacturers' perspective", International Journal of Retail & Distribution Management, Vol. 38 Iss: 10, pp. 773-788. 3. Arbuckle, Dani (2015), “Defensive Strategies in Strategic Management”, Demand Media, available at: http://yourbusiness.azcentral.com/defensive-strategies- strategic-management-4676.html. 4. Asikhia, Olalekan (2010), “Customer Orientation and Firm Performance among Nigerian Small and Medium Scale Businesses”, International Journal of Marketing Studies, Vol. 2 No. 1, pp. 197-212. 5. Auh, Seigyoung; Menguc, Bulent; Fisher, Michelle and Haddad, Abeer (2011), “The perceived autonomy–perceived service climate relationship: The contingency effect of store-level tenure diversity”, Journal of Retailing and Consumer Services, Vol. 18 Iss: 6, pp. 509-520. 6. Avlonitis, George J. and Indounas, Kostis A. (2007), “Service pricing: An empirical investigation”, Journal of Retailing and Consumer Services, Vol. 14 Iss: 1, pp. 83-94. 7. Bajaj, Chetan; Tuli, Rajnish and Srivastava, Nidhi Varma (2010), Retail Management, Oxford. 8. Barber, Clifford S and Tietje, Brian C (2004), “A distribution services approach for developing effective competitive strategies against “big box” retailers”, Journal of Retailing and Consumer Services, Vol. 11 Iss: 2, pp. 95-107. 9. Barone, Michael J.; Norman, Andrew T. and Miyazaki, Anthony D. (2007), “Consumer response to retailer use of cause-related marketing: Is more fit better?”, Journal of Retailing, Vol. 83 Iss: 4, pp. 437-445. 10. Barry, B. J. and Jill, A. S. (2000), “Atmospheric affect as a tool for creating value and gaining share of customer”, Journal of Business Research, 49, pp. 91-101. 11. Baskaran, Kamaladevi (2011), “Success of retail in India: the customer experience management scenario” International Journal of Electronic Marketing and Retailing, Vol. 4 No. 2/3, pp. 206-223. 395 12. Batra, S. K. and Kazmi S. H. H. (2004), Consumer Behaviour – Text and Cases, Excel Books, India, 1st Ed. 13. Beard, Ross (2014), “Why Customer Satisfaction is important”, Client Heartbeat, available at: http://blog.clientheartbeat.com/why-customer-satisfaction-is- important/. 14. Beri, G. C. (2000), Marketing Research, Tata McGraw-Hill Education Pvt. Ltd., New Delhi, India. 15. Bodet, Guillaume (2008), “Customer satisfaction and loyalty in service: Two concepts, four constructs, several relationships”, Journal of Retailing and Consumer Services, Vol. 15 Iss: 3, pp. 156-162. 16. Brown, Steven P. and Lam, Son K. (2008), “A Meta-Analysis of Relationships Linking Employee Satisfaction to Customer Responses”, Journal of Retailing, Vol. 84 Iss: 3, pp. 243-255. 17. BIBLIOGRAPHY Burt, Steve; Johansson, Ulf and Thelander, Åsa (2011), “Standardised marketing strategies in retailing? IKEA’s marketing strategies in Sweden, the UK and China”, Journal of Retailing and Consumer Services, Vol. 18 Iss: 3, pp. 183-193. 18. Cacho-Elizondo, Silvia and Loussaïef, Leïla (2010), “The Influence of Sustainable Development on Retail Store Image”, International Business Research, Vol. 3 No. 3, pp. 100-110. 19. Campo, Katia and Gijsbrechts, Els (2004) “Should retailers adjust their micro- marketing strategies to the type of outlet? An application to location-based store space allocation in limited and full-service grocery stores”, Journal of Retailing and Consumer Services, Vol. 11 Iss: 6, pp. 369-383. 20. Cassab, Harold (2009), “Investigating the dynamics of service attributes in multi- channel environments”, Journal of Retailing and Consumer Services, Vol. 16 Iss: 1, pp. 25-30. 21. Chakraborty, Anasua (2013), “Retail Industry”, available at: http://info.shine.com/industry/retail/7.html. 22. Chen, Shih-Mei and Huddleston, Patricia (2009), “A comparison of four strategies to promote fair trade products”, International Journal of Retail & Distribution Management, Vol. 37 Iss: 4, pp. 336-345. 23. Ciotti, Gregory (2013), “15 Customer Service Skills that Every Employee needs”, February, available at: http://www.helpscout.net/blog/customer-service-skills/. 396 24. Coca-Stefaniak, J. Andres; Parker, Cathy and Rees, Patricia (2010), "Localisation as a marketing strategy for small retailers", International Journal of Retail & Distribution Management, Vol. 38 Iss: 9, pp. 677-697. 25. Cohen, Heidi (2011), “30 Branding Definitions”, Branding, available at: http://heidicohen.com/30-branding-definitions/. 26. Copper, D. and Schindler, P. S. (2006), Marketing Research: Concept & Case, Tata McGraw Hill Publishing Co. Ltd., Delhi. 27. Corporate Catalyst India (2012), “A report on Indian Retail Industry”, available at http://www.rasci.in/downloads/2012/Retail_Industry_India_2012.pdf, pp. 13-14 28. Cronbach, L. (1951), “Coefficient alpha and the internal structure of tests”, Psychometrika, 16, pp. 297-334 29. D&B (2014), “Indian Retail Industry: Challenges, Opportunities & Outlook”, available at https://www.dnb.co.in/IndianRetailIndustry/overview.asp. 30. Darley, William K.; Luethge, Denise J. and Thatte, Ashish (2008), “Exploring the relationship of perceived automotive salesperson attributes, customer satisfaction and intentions to automotive service department patronage: The moderating role of customer gender”, Journal of Retailing and Consumer Services, Vol. 15 Iss: 6, pp. 469-479. 31. Datta, Palto Ranjan; Akwensivie, Divine Mawuli and Akwensivie, Gad (2008), “Store design: the effect of Iceland’s refit strategy on food consumers”, Journal of Business & Retail Management Research, Vol.1 Issue1, October. 32. Davidson, William R.; Sweeney, Daniel J. and Stampfl, Ronald W. (1988), Retailing Management, John Wiley & Sons. 33. Demoulin, Nathalie T. M. BIBLIOGRAPHY and Zidda, Pietro (2009), “Drivers of Customers’ Adoption and Adoption Timing of a New Loyalty Card in the Grocery Retail Market”, Journal of Retailing, Vol. 85 Iss: 3, pp. 391-405. 34. Demoulin, Nathalie T.M. and Zidda, Pietro (2008), “On the impact of loyalty cards on store loyalty: Does the customers’ satisfaction with the reward scheme matter?”, Journal of Retailing and Consumer Services, Vol. 15 Iss: 5, pp 386-398. 35. Dharni, Khushdeep and Chawla, Madhur (2012), “Antecedents of IT effectiveness in Indian retail sector”, International Journal of Electronic Marketing and Retailing, Vol. 5, No.1, pp. 32-47. 397 36. Dhillon, Jaskaran Singh; Joshi, Madhur and Verma, Ramita (2012), “Emergence of Retailing Sector in India: Challenges and Opportunities”, IJMBS, Vol. 2, Issue 4, pp. 02-03. 37. Dhume, Sudheer (2012), “Service Quality Dimensions In Retail Settings: An Empirical Study at Selected Apparel Specialty Stores of Mumbai”, International Journal of Research in Commerce, Economics & Management, Vol. 2 Iss: 2, pp. 01-05. 38. Diallo, Mbaye Fall (2012), “Effects of store image and store brand price-image on store brand purchase intention: Application to an emerging market”, Journal of Retailing and Consumer Services, Vol. 19 Iss: 3, pp. 360-367. 39. Dikshit, Anand (2011), “The Uneasy Compromise - Indian Retail”, The Wall Street Journal, August, 12, pp. 05-22. 40. Diona, Delphine and Arnould, Eric (2011), “Retail Luxury Strategy: Assembling Charisma through Art and Magic”, Journal of Retailing, Vol. 87 Iss: 4, pp. 502- 520. 41. Esbjerg, Lars; Jensen, Birger Boutrup; Bech-Larsen, Tino; Barcellos, Marcia Dutra de; Boztug, Yasemin and Grunert, Klaus G. (2012), “An integrative conceptual framework for analysing customer satisfaction with shopping trip experiences in grocery retailing”, Journal of Retailing and Consumer Services, Vol. 19 Iss: 4, pp. 445-456. 42. Etgar, Michael and Rachman-Moore, Dalia (2011), “The relationship between national cultural dimensions and retail format strategies”, Journal of Retailing and Consumer Services, Vol. 18 Iss: 5, pp. 397-404. 43. Evans, Jody; Bridson, Kerrie; Byrom, John and Medway, Dominic (2008), "Revisiting retail internationalisation: Drivers, impediments and business strategy", International Journal of Retail & Distribution Management, Vol. 36 Iss: 4, pp. 260- 280. 44. FICCI (2011), “Sector Profile”, Federation of Indian Chambers of Commerce and Industry, available at http://www.ficci.com/sector/33/Project_docs/Sector- prof.pdf. 45. Fonseca, Jaime R. S. (2009), “Customer satisfaction study via a latent segment model”, Journal of Retailing and Consumer Services, Vol. 16 Iss: 5, pp. 352-359. 46. Fornell, C. BIBLIOGRAPHY (1992), “A national customer satisfaction barometer: the Swedish experience”, Journal of Marketing, Vol. 56, No. 1, pp. 6-21. 398 47. Fullerton, Gordon (2005), “The service quality–loyalty relationship in retail services: does commitment matter?”, Journal of Retailing and Consumer Services, Vol. 12 Iss: 2, pp. 99-111. 48. Gauri, Dinesh Kumar; Trivedi, Minakshi and Grewal, Dhruv (2008), “Understanding the Determinants of Retail Strategy: An Empirical Analysis”, Journal of Retailing, Vol. 84 Iss: 3, pp. 256-267. 49. Glueck, W. F. and Lauch, L. R. (1984), Business Policy and Strategic Management, McGraw-Hill, New York. 50. González-Benito, Óscar and Martos-Partal, Mercedes (2012), “Role of Retailer Positioning and Product Category on the Relationship between Store Brand Consumption and Store Loyalty”, Journal of Retailing, Vol. 88 Iss: 2, pp. 236- 249. 51. Goodrich, Kendall and Ramsey, Rosemary (2012), “Are consumers with disabilities receiving the services they need?”, Journal of Retailing and Consumer Services, Vol. 19 Iss: 1, pp. 88-97. 52. Grewal, Dhruv; Roggeveen, Anne L. and Tsiros, Michael (2008), The Effect of Compensation on Repurchase Intentions in Service Recovery”, Journal of Retailing, Vol 84 Iss: 4, pp. 424-434. 53. Haemoon, Oh (1999), “Service quality, customer satisfaction and customer value: A holistic perspective”, International Journal of Hospitality Management, Vol. 18 Iss: 1, pp. 67-82. 54. Hamburg, C. and Koschate, N. (2004), “How do consumers react to price increases”, International Journal of Research in Marketing, Vol. 26 No. 4. 55. Handa, Vidushi and Grover, Navneet (2012), “Retail Sector in India: Issues & Challenges” ZENITH International Journal of Multidisciplinary Research, Vol. 2, Issue 5, pp. 6. 56. Hansen, Torben; Jensen, Jan Møller and Solgaard, Hans Stubbe (2011), "When supermarket consumers get stocked in the middle", International Journal of Retail & Distribution Management, Vol. 39 Iss: 11, pp. 836-850. 57. Harper, Douglas (2015), “Retail”, Online Etymology Dictionary, available at: http://www.etymonline.com/index.php?term=retail. 58. Hoffman, K. Douglas and Bateson, John E. G. (2007), Services Marketing: Concepts, Strategies, & Cases, Thomson South-Western, India, 3rd Ed. 399 59. Huddleston, Patricia; Whipple, Judith; Mattick, Rachel Nye and Lee, So Jung (2009), "Customer satisfaction in food retailing: comparing specialty and conventional grocery stores", International Journal of Retail & Distribution Management, Vol. 37 Iss: 1, pp. 63-80. 60. Humphrey, Albert (2005), “SWOT Analysis for Management Consulting”, SRI Alumni Newsletter, December. 61. Hyper One Center (2011), Hyper Market, Malls Guide, available at http://www.mallsguide.com/ShoppingCenter.aspx?sc_id=1572&title=Hyper %20One%20Center. 62. IBEF (2015), “Retail Industry in India”, India Brand Equity Foundation, available at: http://www.ibef.org/industry/retail-india.aspx. 63. BIBLIOGRAPHY Ilonen, Laura; Wren, Jody; Gabrielsson, Mika and Salimäki, Markku (2011), "The role of branded retail in manufacturers' international strategy", International Journal of Retail & Distribution Management, Vol. 39 Iss: 6, pp. 414-433. 64. Jayawardhena, Chanaka and Farrell, Andrew M. (2011), "Effects of retail employees' behaviours on customers' service evaluation", International Journal of Retail & Distribution Management, Vol. 39 Iss: 3, pp. 203-217. 65. Jhureley, Ashutosh (2011), “6P's of Retail leading to 7th P – Profit”, Multi- channel Retail & Supply Chain, available at: http://ashutoshjhureley.blogspot.in/2010/09/6ps-of-retail-leading-to-7th-p- profit.html. 66. Joo, Mingyu; Mazumdar, Tridib and Raj, S. P. (2012) “Bidding Strategies and Consumer Savings in NYOP Auctions”, Journal of Retailing, Vol. 88 Iss: 1, pp. 180-188. 67. Kantabutra, Sooksan (2011), “Examining store manager effects in consumer and staff satisfaction: Evidence from Thailand”, Journal of Retailing and Consumer Services, Vol. 18 Iss: 1, pp. 46-57. 68. Kazmi, A. (2007), Business Policy and Strategic Management, Tata McGraw Hill Publishing Company Limited, New Delhi. 69. Khraim, Hamza Salim; Khraim, Aymen Salim; Al-Kaidah, FirasMuslam and AL- Qurash, Daher (2011), “Jordanian Consumer’s Evaluation of Retail Store Attributes: The Influence of Consumer Religiosity”, International Journal of Marketing Studies, Vol. 3 No. 4, pp. 105-116. 400 70. Kleijnen, Mirella; Ruyter, Ko de and Wetzels, Martin (2007), “An assessment of value creation in mobile service delivery and the moderating role of time consciousness”, Journal of Retailing, Vol. 83 Iss: 1, pp. 33-46. 71. Kohijoki, Anna-Maija (2011), “The effect of aging on consumer disadvantage in grocery retail services among the Finnish elderly”, Journal of Retailing and Consumer Services, Vol. 18 Iss: 4, pp. 370-377. 72. Koistinen, Katri and Järvinen, Raija (2009), “Consumer observations on channel choices—Competitive strategies in Finnish grocery retailing”, Journal of Retailing and Consumer Services, Vol. 16 Iss: 4, pp. 260-270. 73. Kukar-Kinney, Monika; Ridgwaya, Nancy M. and Monroea, Kent B. (2012), “The Role of Price in the Behaviour and Purchase Decisions of Compulsive Buyers”, Journal of Retailing, Vol. 88 Iss: 1, pp. 63-71. 74. Kukar-Kinney, Monika; Xia, Lan and Monroe, Kent B. (2007), “Consumers’ perceptions of the fairness of price-matching refund policies”, Journal of Retailing, Vol. 83, Iss: 3, pp. 325-337. 75. Levy, M.; Weitz, B. and Pandit, A. (2012), Retailing Management, Tata McGraw Hill Education Private Limited, New Delhi. 76. Likert, R. (1970), “A Technique for the Measurement of Attitude”, In Gene F. summer ed., Attitude Measurement, Chicago: Rand McNally. 77. Lindquist, Jay D. and Sirgy, M. 81. Madaan, KVS (2011), Fundamental of Retailing, Tata McGraw Hill Education Pvt. Ltd., New Delhi. BIBLIOGRAPHY Joseph (2003), Shopper, Buyer and Consumer Behaviour – Theory and Marketing Applications, Atomic Dog Publishing, USA, 2nd Edition. 78. Lonarkar, Pramod Pandurangrao and Gore, Parmeshwar (2012), “Change in Retail Shopping Behaviour: Why and for Whom?”, International Journal of Research in Commerce, Economics & Management, Vol. 2 Iss: 1, pp. 108-111. 79. Loundon, David L. and Della Bitta, Albert J. (2002), Consumer Behaviour, Tata McGraw-Hill Publishing Company Limited, New Delhi, 4th Ed. 80. Lu, Yan and Seock, Yoo-Kyoung (2008), "The influence of grey consumers' service quality perception on satisfaction and store loyalty behaviour", International Journal of Retail & Distribution Management, Vol. 36 Iss: 11, pp. 901-918. 81. Madaan, KVS (2011), Fundamental of Retailing, Tata McGraw Hill Education Pvt. Ltd., New Delhi. 401 82. Mahajan, A. C. (2015), “Retail eCommerce Sales in India 2014 – 2018”, Dazeinfo, January 7. 83. Mai, Enping (Shirley); Yang, Jun and Chen, Haozhe (2011), "Primary product network size on complementary product sales: Moderating effects of customer characteristics", International Journal of Retail & Distribution Management, Vol. 39 Iss: 11, pp. 851-866. 84. Majumder, Sanjoy (2011), “Changing the way Indians shop”, BBC News, November, 25, 85. Manning, Kenneth C. and Sprott, David E. (2007) “Multiple unit price promotions and their effects on quantity purchase intentions”, Journal of Retailing, Vol. 83 Iss: 4, pp. 411-421. 86. Martenson, Rita (2007), "Corporate brand image, satisfaction and store loyalty: A study of the store as a brand, store brands and manufacturer brands", International Journal of Retail & Distribution Management, Vol. 35 Iss: 7, pp. 544-555. 87. Martínez-Ruiz, María Pilar; Jiménez-Zarco, Ana Isabel and Cascio, Robert (2011), "Assessing the maximum level of customer satisfaction in grocery stores: A comparison between Spain and the USA", International Journal of Retail & Distribution Management, Vol. 39 Iss: 7, pp. 504-521. 88. Mazaheri, Ebrahim; Basil, Debra Z.; Yanamandram, Venkata and Daroczi, Zoltan (2011), “The impact of pre-existing attitude and conflict management style on customer satisfaction with service recovery”, Journal of Retailing and Consumer Services, Vol. 18 Iss: 3, pp. 235-245. 89. McDaniel, C. D. and Gates, R. H. (2001), Marketing Research Essential, Thomson South-Western. 90. Medina, Cayetano and Rufín, Ramón (2009), "The mediating effect of innovation in the relationship between retailers' strategic orientations and performance", International Journal of Retail & Distribution Management, Vol. 37 Iss: 7, pp. 629- 655. 91. BIBLIOGRAPHY Meyer-Waarden, Lars (2007), “The effects of loyalty programs on customer lifetime duration and share of wallet”, Journal of Retailing, Vol. 83, Iss: 2, pp. 223-236. 92. Mohan, Geetha; Sivakumaran, Bharadhwaj and Sharma, Piyush (2012), “Store environment's impact on variety seeking behaviour” Journal of Retailing and Consumer Services, Vol. 19 Iss: 4, pp. 419-428. 402 93. Moriarty, Mike; Peng, Althea; Rhim, Helen and Salman, Fabiola (2013), “Global Retailers: Cautiously Aggressive or Aggressively Cautious?”, The 2014 Global Retail Development Index, A.T. Kearney Consultants, USA. 94. Moriarty, Mike; Rhim, Helen; Salman, Fabiola and Ben-Shabat, Hana (2014), “Full Steam Ahead for Global Retailers”, The 2014 Global Retail Development Index, A.T. Kearney Consultants, USA. 95. MSEDCL (2015) “MSEDCL locations”, available at: http://www.mahadiscom.in/maps/map_index.shtm. 96. Murthi, B.P.S. and Rao, Ram C. (2012), “Price Awareness and Consumers’ Use of Deals in Brand Choice”, Journal of Retailing, Vol. 88 Iss: 1, pp. 34-46. 97. Muzondo, Noeland and Mutandwa, Edward (2011), “The Seven Ps of Marketing and Choice of main Grocery Store in a Hyperinflationary Economy”, Contemporary Marketing Review, Vol. 1 (9), pp. 01-18. 98. Nair, Suja R. (2011), Retailing Management, Himalaya Publishing House, India. 99. Nath, Rachna and Gupt, Akash (2012), “Winds of change Retail reforms in India”, PWC, available at https://www.pwc.in/en_IN/in/assets/pdfs/industries/retail- andconsumer/ Winds_of_change_Retail_Reforms_in_India_final_Sept_2012_(with_RM).pdf. 100. Neelankavil, J. P. (2007), International Business Research, M. E. Sharpe Inc., New York. 101. Netemeyer, Richard G. and Maxham III, James G. (2007), “Employee versus supervisor ratings of performance in the retail customer service sector: Differences in predictive validity for customer outcomes”, Journal of Retailing, Vol. 83 Iss: 1, pp. 131-145. 102. Ngobo, Paul-Valentin (2011), “Private label share, branding strategy and store loyalty”, Journal of Retailing and Consumer Services, Vol. 18 Iss: 4, pp. 259-270. 103. Ogden, James R. and Ogden, Denise T. (2004), Retailing: Integrated Retail Management, Houghton Mifflin. 104. Omar, Ogenyi and Abisoye, Kareem O. (2008), “The Role of Price as a Determining Factor in Grocery Retail Store Patronage”, Journal of Business & Retail Management Research, Vol. 2, Issue 2. 105. Omholt, Tore (2002), “Strategic Reconfiguration and Management of Retail Centers Creating Competitive Advantage through Communicative Competence”, Asian Academy of Management Journal, Vol. 7 No. 2, pp. 87-101. 403 106. Orth, Ulrich R. and Green, Mark T. (2009), “Consumer loyalty to family versus non-family business: The roles of store image, trust and satisfaction”, Journal of Retailing and Consumer Services, Vol. 16 Iss: 4, pp. 248-259. 107. BIBLIOGRAPHY Padilla, Charlette and Eastlick, Mary Ann (2009), "Exploring urban retailing and CBD revitalisation strategies", International Journal of Retail & Distribution Management, Vol. 37 Iss: 1, pp. 7-23. 108. Parasuraman, A., Grewal, D. and Krishnan, R. (2007), Marketing Research, Biztantra, New Delhi, India. 109. Perner, Lars (2008), “Consumer Behaviour”, Department of Marketing, Marshall School of Business, University of Southern California, USA. 110. Phillips, B S (1966), Social Research Strategy & Tactics, McMillan Publishing Company Inc., US. 111. Phulia, Anita Bindal and Sharma, Monika (2014), “Role of Retail sector in India”, International Journal of Research, Vol. 1, Issue 8, pp. 152-158. 112. Porter, M. E. (1985), Competitive Advantage: Creating and Sustaining Superior Performance, Free Press, New York. 113. Pradhan, S. (2009), Retailing Management – Text & Cases, Tata McGraw Hill Education Private Limited, New Delhi. 114. Pradhan, Swapna (2011), Retail Merchandising, Tata Mcgraw Hill Education Private Ltd., New Delhi, India. 115. Prasad, B. V. S. (2013), “E-tailing in India”, Abhinav- National Monthly Refereed Journal of Research in Commerce & Management, Vol. 2, Issue 7, pp. 20-28. 116. Prasad, Cherukuri Jayasankara and Aryasri, Ankisetti Ramachandra (2011) "Effect of shopper attributes on retail format choice behaviour for food and grocery retailing in India", International Journal of Retail & Distribution Management, Vol. 39 Iss: 1, pp. 68-86. 117. Prasad, Priya and Harikumar, P. N. (2013), “Indian Retail Sector reforms since liberation”, Caarmal Journal of Management Research, Vol. 1, Issue 2, pp. 94. 118. PTI (2014), “Indian retail market to reach 47 lakh crore by 2016-17: study”, The Economic Times, 11, February. 119. Rahman, Tazyn (2012), “Organised Retail Industry in India – Opportunities and Challenges”, International Journal of Research in Finance & Marketing, Volume 2, Issue 2, pp. 82-94. 404 120. Rahman, Tazyn (2012), “Organised Retail Industry in India – Opportunities and Challenges”, International Journal of Research in Finance & Marketing, Vol. 2, Iss: 2, pp. 82-94. 121. Rai, Manish and Gopal, R. (2014), “A study on the feasibility of retail store location for TATA Tesco convenience store in Navi Mumbai”, IOSR Journal of Business and Management, International Business Research Conference, IES Management College and Research Centre, Mumbai, India, Special Issue, pp. 35- 39. 122. Rajagopal (2011), "Impact of radio advertisements on buying behaviour of urban commuters", International Journal of Retail & Distribution Management, Vol. 39 Iss: 7, pp. 480-503. 123. Ranganath, N. Santosh; Kama, Raju T. and Rao, P. BIBLIOGRAPHY Trinadha (2011), “Emerging Role of Technology in Retail Sector”, International Journal of Research in Computer Application & Management, Vol. 1 Iss. 5, pp. 71-75. 124. Rani, A. Sandhya (2012), “Retailing Strategies for Customer Satisfaction: Comparative Study of More and Food World”, International Journal of Research in Commerce & Management, Vol. 3 Iss: 5, pp. 135-136. 125. Rao, K. Rama Mohana and Manikyam, K. Ratna (2012), “Customers’ Experience with Small Scale Retail Stores - An Empirical Study”, International Journal of Research in Commerce & Management, Vol. 3 Iss. 5, pp. 86-89. 126. Richards, Timothy J.; Gómez, Miguel I. and Pofahl Geoffrey (2012), “A Multiple- discrete/Continuous Model of Price Promotion”, Journal of Retailing, Vol. 88 Iss: 2, pp. 206–225. 127. Ripley, Louise M. (2015), “Selecting an Overall Positioning Strategy”, Introductory Marketing, available at: http://www.yorku.ca/lripley/imUsegment.htm. at: 128. Ruiz, María Pilar Martínez and Descals, Alejandro Mollá (2008), "Retail price promotion influences for product varieties in grocery stores: evidence from Spain", International Journal of Retail & Distribution Management, Vol. 36 Iss: 6, pp. 494-517. 129. Sánchez-Fernández, Raquel and Iniesta-Bonillo, M. Ángeles (2009), “Efficiency and quality as economic dimensions of perceived value: Conceptualisation, measurement, and effect on satisfaction”, Journal of Retailing and Consumer Services, Vol. 16 Iss: 6, pp. 425-433. 405 130. Sayed, Md Abu (2015), “Merchandising: The magic word”, Textile Apex, available at: http://textileapex.blogspot.in/2014/09/merchandising-magic- word.html. 131. Schau, Hope Jensen; Dellande, Stephanie and Gilly, Mary C. (2007), “The impact of code switching on service encounters”, Journal of Retailing, Vol. 83 Iss: 1, pp. 65-78. 132. Sekaran, U. (2003), Research Methods for Business: A Skill Building Approach, John Wiley & Sons Inc., New York. 133. Shaikh, Anwar; Alvi, Sajid and Jagtap, Manisha (2012), “A Study of the Factors Influencing the Buyer’s Decision to Buy a Particular Product with reference to selected Supermarkets in Pune City”, Indian Journal of Commerce & Management Studies, Special Issue, Jan., pp. 7-15. 134. Shankar, Venkatesh; Inman, J. Jeffrey; Mantrala, Murali; Kelley, Eileen and Rizley, Ross (2011), “Innovations in Shopper Marketing: Current Insights and Future Research Issues”, Journal of Retailing, Vol. 87, Supplement 1, pp. S29- S42. 135. Sheng, Shibin and Pan, Yue (2009), “Bundling as a new product introduction strategy: The role of brand image and bundle features”, Journal of Retailing and Consumer Services, Vol. 16 Iss: 5, pp. 367-376. 136. BIBLIOGRAPHY Sikri, Sunita and Wadhwa, Dipti (2012), “Growth and Challenges of Retail Industry in India: An Analysis”, Asia Pacific Journal of Marketing and Management Review, Vol. 1, Issue 1, pp. 01-14. 137. Singh, Harvinder; Chadha, S.K. and Rishi, Bikramjit (2010), “Service quality measurement and its implications: a case study of Vishal Mega Mart”, International Journal of Services and Standards, Vol. 6 No. 2, pp. 150-169. 138. Smith, Rhonda L. (2006), “The Australian grocery industry: a competition perspective”, The Australian Journal of Agricultural and Resource Economics, Vol. 50, pp. 33-50. 139. Söderlund, Magnus and Julander, Claes-Robert (2009), “Physical attractiveness of the service worker in the moment of truth and its effects on customer satisfaction”, Journal of Retailing and Consumer Services, Vol. 16 Iss: 3, pp. 216-226. 140. Söderlund, Magnus and Rosengren, Sara (2007), “Receiving word-of-mouth from the service customer: An emotion-based effectiveness assessment”, Journal of Retailing and Consumer Services, Vol. 14 Iss: 2, pp. 123-136. 406 141. Söderlund, Magnus and Rosengren, Sara (2010), “The happy versus unhappy service worker in the service encounter: Assessing the impact on customer satisfaction”, Journal of Retailing and Consumer Services, Vol. 17 Iss: 2, pp. 161- 169. 142. Standop, Dirk and Grunwald, Guido (2009), "How to solve product-harm crises in retailing?: Empirical insights from service recovery and negative publicity research", International Journal of Retail & Distribution Management, Vol. 37 Iss: 11, pp. 915-932. 143. Stanton, A. (2015), “Store Design Layout”, available at: http://www.radford.edu/~astanton/mktg343/store_design_layout_visual_merchand ising.ppt&usd=2&usg=AFQjCNHbo1zz4IPbL3qzykLjaXvPVeUbBw. 144. Sun, Jing and Ma, Jingting (2009), “The Study of China Retail Business Development Strategy”, International Journal of Marketing Studies, Vol. 1 No. 1, pp. 19-22. 145. Suwelack, Thomas; Hogreve, Jens and Hoyer, Wayne D. (2011), “Understanding Money-Back Guarantees: Cognitive, Affective, and Behavioural Outcomes”, Journal of Retailing, Vol. 87 Iss: 4, pp. 462-478. 146. Swinyard, William R (2003), “The effects of salesperson mood, shopper behaviour, and store type on customer service”, Journal of Retailing and Consumer Services, Vol. 10 Iss: 6, pp. 323-333. 147. Tanabe, Mario; Angelo, Claudio Felisoni De and Alexander, Nicholas (2004), “The effectiveness of strategic planning: competitiveness in the Brazilian supermarket sector”, Journal of Retailing and Consumer Services, Vol. 11 Iss: 1, pp. 51-59. 148. Thakor, Mrugank V.; Suri, Rajneesh and Saleh, Katayoun (2008), “Effects of service setting and other consumers’ age on the service perceptions of young consumers”, Journal of Retailing, Vol. 84 Iss: 2, pp. 137-149. 149. BIBLIOGRAPHY The Hindu (2014), “Indian retail market to reach 47 lakh crore by 2016-17: study”, 11, February. 150. The Hindu Business Line (2014), “Indian retail market set to touch $865 billion by 2023”, 27 February. 151. The planning Factory Limited (2014) “What is Merchandise Planning?”, available at: http://www.planfact.co.uk/what_is_merchandise_planning.htm. 407 152. Thiruvenkadam, T. and Panchanatham, N. (2011), “Influence Of Demography on Store Patronage Behaviour of Chennai Shoppers”, International Journal of Research in Computer Application & Management, Vol. 1 Iss. 5, pp. 39-42. 153. Torres-Moraga, Eduardo; Jara-Sarrua, Luis and Moneva, Jose M. (2008) “Measuring supermarket service quality: proposal for a scale”, International Journal of Services and Standards, Vol. 4 No. 1, pp. 81-96. 154. Tyagi, Rahul (2008), “Supply Chain Implications of Customer-Centric Merchandising Strategy”, Journal of Business & Retail Management Research, Vol. 2 Issue. 2. 155. Vázquez-Carrasco, Rosario and Foxall, Gordon R. (2006), “Influence of personality traits on satisfaction, the perception of relational benefits, and loyalty in a personal service context”, Journal of Retailing and Consumer Services, Vol. 13 Iss: 3, pp. 205-219. 156. Vesel, Patrick and Zabkar, Vesna (2009), “Managing customer loyalty through the mediating role of satisfaction in the DIY retail loyalty program”, Journal of Retailing and Consumer Services, Vol.16 Iss: 5, pp. 396-406. 157. Vetrivel, T. (2011), “Customer Relationship Management in Retailing with Special Reference to Fast Moving Consumer Goods In Erode District, Tamilnadu, India”, International Journal of Research in Computer Application & Management, Vol. 1 Iss. 2, pp. 47-52. 158. Walker, James L. (1995), “Service encounter satisfaction: conceptualised”, Journal of Services Marketing, Vol. 9 No.1, pp. 5-14. 159. Watchravesringkan, Kittichai (Tu) and Punyapiroje, Chompunuch (2011), "A comparative investigation of consumers' attitudes toward marketing practices of hypermarket retailers in Thailand", International Journal of Retail & Distribution Management, Vol. 39 Iss: 9, pp. 702-720. 160. Waters, Shari (2015), “Retailing”, About Money, available at: http://retail.about.com/od/glossary/g/retailing.htm. 161. Watson, Stevie (2012), “Consumer responses to service situations: Tests for main and interaction effects”, Journal of Retailing and Consumer Services, Vol. 19 Iss: 3, pp. 287-296. 162. White, Allyn; Breazeale, Michael and Collier, Joel E. (2012), “The Effects of Perceived Fairness on Customer Responses to Retailer SST Push Policies”, Journal of Retailing, Vol. 88 Iss: 2, pp. 250-261. 408 163. Wiles, Michael A. (2007), “The effect of customer service on retailers’ shareholder wealth: The role of availability and reputation cues”, Journal of Retailing, Vol. 83, Iss: 1, pp. 19-31. 164. INSTITUTIONAL QUESTIONNAIRE Name ____________________________________________ Designation ____________________________________________ Name of Store ____________________________________________ Place of Store ____________________________________________ BIBLIOGRAPHY Yan, Ruiliang and Pei, Zhi (2009), “Retail services and firm profit in a dual- channel market”, Journal of Retailing and Consumer Services, Vol. 16 Iss: 4, pp. 306-314. 165. Yi, Y. (1991), “A critical review of customer satisfaction”, Review of Marketing, American Marketing Association, Chicago II, pp. 68-123. 166. Zielke, Stephan (2008), “Exploring asymmetric effects in the formation of retail price satisfaction”, Journal of Retailing and Consumer Services, Vol. 15 Iss: 5, pp. 335-347. 167. Zolfagharian, Mohammad Ali and Paswan, Audhesh (2009), “Perceived service innovativeness, consumer trait innovativeness and patronage intention”, Journal of Retailing and Consumer Services, Vol. 16 Iss: 2, pp. 155-162. 409 2. Which segmentation strategy do you use & why? Single Segment Concentration Strategy Selective Specialisation Strategy Product Specialisation Strategy Market Specialisation Strategy Full Market Coverage Strategy Single Segment Concentration Strategy Product Specialisation Strategy Market Specialisation Strategy Selective Specialisation Strategy Full Market Coverage Strategy 1. What are your variables to segment the market? 1. What are your variables to segment the market? What are your variables to segment the market?  Geographic Segmentation  Demographic Segmentation  Psychographic Segmentation  Behavioural Segmentation  Geographic Segmentation  Psychographic Segmentati  Geographic Segmentation  Demographic Segmentation  Behavioural Segmentation  Demographic Segmentation If Geographic-  Region- North, West, South, East, Central  Area- Urban, Suburban, Rural If Demographic-  Age- infants, below 5 years, 5-10, 11-20, 21-34, 35-49, 50-60, 60+  Family Life Cycle- young, single; young, married, no children; young, married, child……  Gender- male and female  Gender- male and female  Income- low, lower middle, middle, upper middle, and high  Occupation- unskilled worker, skilled worker, students, self-employed, homemaker, retired  Education- illiterate, school SSC/HSC, non-graduate, graduate / postgraduate (general/professional)  Religion- Hinduism, Islam, Catholic, Sikh f Psychographic- If Psychographic-  Life Style- culture oriented, sport oriented, outdoor oriented  Personality- compulsive, gregarious, authoritarian, ambitious, aggressive, shy, emotional 410 If Behavioural- If Behavioural- User Status- non-user, ex-user, potential user, first time user, regular user Usage Rate- light, medium, heavy Loyalty Status- none, medium, strong, absolute Readiness Stage- unaware, aware, informed, interested, desirous, intending to buy Attitude towards Product- enthusiastic, positive, indifferent, negative, hostile Occasions- regular, special Benefits- quality, service, economy, speed Perceived Risk- high, medium, low Innovativeness- innovator, early adopter, early majority, late majority, laggard Loyalty Status- none, medium, strong, absolute Readiness Stage- unaware, aware, informed, interested, desirous, intending to buy Attitude towards Product- enthusiastic, positive, indifferent, negative, hostile Occasions- regular, special Innovativeness- innovator, early adopter, early majority, late majority, laggard 3. Which targeting strategy do you use & why?  Undifferentiating targeting strategy or mass retailing (no specific mix)  Differentiated targeting strategy or selective retailing (specific mix)  Concentrated targeting strategy or niche retailing (small but profitable)  Micro (local or individual) targeting strategy or customised retailing  Multi-segment targeting strategy  Multi-segment targeting strategy 4. Which Positioning Strategy do you use & why?  More for more  Less for much less  More for more  More for the same  More for less  The same for less 5. What kind of retail marketing strategy do you adopt?  Market penetration  Retail format development  Market penetration  Market expansion  Retail format development  Diversification  Market expansion  Diversification 6. Which defensive strategy do you use? 6. Which defensive strategy do you use? 411  Preemptive defense (more aggressive launch an attack on enemy)  Counteroffensive defense (provides an opportunity to attacker to attack frontally)  Flanking defense (serve as a defensive corner for weak front)  Position defense (strategy to retain position)  Mobile defense (stretches its domain to new territories)  Contraction defense (giving up weaker territories or activities) 7. With which tool you perform Store-Level Planning.  SWOT analysis  PEST analysis  Five force analysis  ETOP analysis Is SWOT analysis the tool? Yes/No If ‘Yes’ please, specify SWOT Analysis- Strengths_____________________________________________________________ Weaknesses___________________________________________________________ Opportunities__________________________________________________________ Threats_______________________________________________________________ Is PEST analysis the tool? Yes/No If ‘Yes’ please, specify PEST Analysis- Political & Legal Forces_________________________________________________ EconomicForces_______________________________________________________ Socio-culturalForces____________________________________________________  Preemptive defense (more aggressive launch an attack on enemy)  Counteroffensive defense (provides an opportunity to attacker to attack frontally)  Flanking defense (serve as a defensive corner for weak front)  Position defense (strategy to retain position)  Mobile defense (stretches its domain to new territories)  Contraction defense (giving up weaker territories or activities)  Counteroffensive defense (provides an opportunity to attacker to attack frontally)  Flanking defense (serve as a defensive corner for weak front)  Position defense (strategy to retain position)  Mobile defense (stretches its domain to new territories)  Contraction defense (giving up weaker territories or activities) 7. With which tool you perform Store-Level Planning.  SWOT analysis  PEST analysis  Five force analysis  ETOP analysis Is SWOT analysis the tool? Yes/No If ‘Yes’ please, specify SWOT Analysis- Strengths_____________________________________________________________ Weaknesses___________________________________________________________ Opportunities__________________________________________________________ Threats_______________________________________________________________ Is PEST analysis the tool? Yes/No If ‘Yes’ please, specify PEST Analysis- Political & Legal Forces_________________________________________________ EconomicForces_______________________________________________________ Socio-culturalForces____________________________________________________ TechnologicalForces____________________________________________________ Is Five Forces analysis the tool? Yes/No 7. With which tool you perform Store-Level Planning. 412 If ‘Yes’ please, specify Five Force Analysis- Forces of Competition created by Rivalry___________________________________ Potential Threats from New Entrants_______________________________________ Potential Threats from the Firms, which provide Substitute Products & Services_____ Supplier’s Bargaining Powers_____________________________________________ Buyer’s Bargaining Powers_______________________________________________ Is ETOP analysis the tool? Yes/No If ‘Yes’ please, specify- Market_______________________________________________________________ Technological_________________________________________________________ Suppliers_____________________________________________________________ Economic_____________________________________________________________ Regulatory____________________________________________________________ Political______________________________________________________________ Socio-cultural_________________________________________________________ International__________________________________________________________ _ 413 413 8. With the Boston Matrix where you see your retail store at present? 8. With the Boston Matrix where you see your retail store at present? 8. With the Boston Matrix where you see your retail store at present?  Star (i.e. high market share, high rate of market growth)  Cash Cow (i.e. high market share, low rate of market growth)  Dog (i.e. low market share, low rate of market growth)  Question Mark (i.e. low market share, high rate of market growth) 9. Which Generic Strategy do you follow at the retail store level?  Cost Leadership (i.e. broadly target with lower cost)  Differentiation (i.e. broadly target with differentiation)  Cost Focus (i.e. narrowly target with lower cost)  Differentiation Focus (i.e. narrowly target with differentiation) 10. What problems do you face in framing overall retailing strategy- 11. What are your core advantages?  Exclusive brands  Mix of private label & other brands  Mix of private label & other brands  Exclusive brands 15. Which of the following brands are offered by your store?  Generic brands  Local brands  National (manufacturer) brands  Copycat brands  National (manufacturer) brands  International brands  International brands 11. What are your core advantages?  Low cost  Efficient operations  Strong private brands  Fashion reputation  Category dominance  Image 12. Please rate the following variables of the retail marketing mix on the importance scale. Sr. No. Variables Importance Scale* 4 3 2 1 1. Product (Merchandise) 2. Price 3. Promotion 4. Premises / Place 5. Personnel / People 6. Presentation 7. Physical Evidence * 4- extremely important, 3- important, 2- partially important, 1- not important 13. How do you manage your retail marketing-mix? __________________________________________________________ 14. Which types of brands are available in your store? 12. Please rate the following variables of the retail marketing mix on the importance 12. Please rate the following variables of the retail marketing mix on the importan scale. Sr. No. Variables Importance Scale* 4 3 2 1 1. Product (Merchandise) 2. Price 3. Promotion 4. Premises / Place 5. Personnel / People 6. Presentation 7. Physical Evidence * 4- extremely important, 3- important, 2- partially important, 1- not important 13. How do you manage your retail marketing-mix? 13. How do you manage your retail marketing-mix? 14. Which types of brands are available in your store? 414  Heterogeneous mix of brands  Multi-brands  Multi-brands  Exclusive brands  Mix of private label & other brands 15. Which of the following brands are offered by your store?  Generic brands  Local brands  National (manufacturer) brands  International brands  Private (store) brands  Exclusive brands  Copycat brands 16. What are the factors important when planning merchandise? Target market Goods & service growth potential Fashion trends Your image Competition  Customer segments  Responsiveness to customers  Amount of investment  Profitability  Risk Please rate the factors on the importance scale. Sr. No. Factors Importance Scale* 4 3 2 1 1. Target market 2. Goods & service growth potential 3. Trends (fashion) 4. Your image 5. Competition 6. Customer segments 7. Responsiveness to customers 8. Amount of investment 9. Profitability 10. Risk * 4- extremely important, 3- important, 2- partially important, 1- not important 17. What are the factors important to you when planning merchandise quality?  Target market  Competition  Your image  Store location  Stock turnover  Profitability  The customer service offered  Personnel  Perceived goods/service benefits  Manufactures vs. private brands Rate the factors on the importance scale. 18. Which of the following product assortment (retail assortment strategy) is adopted by your store? 18. Which of the following product assortment (retail assortment strategy) is adopted by your store?  A narrow variety & a shallow assortment (few goods/service categories & a limited assortment in each category)  A wide variety & a shallow assortment (many goods/service categories & a limited assortment in each category)  A narrow variety & a deep assortment (few goods/service categories & a large assortment in each category)  A wide variety & a deep assortment (many goods/service categories & a large assortment in each category) 19. How do you categorise the products/brands?  Size  Colour  Design  Price  Quality  Brand  Customer Segment  Customer Needs  Discounts  Schemes 16. What are the factors important when planning merchandise? Target market Goods & service growth potential Fashion trends Your image Competition  Customer segments  Responsiveness to customers  Amount of investment  Profitability  Risk Target market Goods & service growth potential Fashion trends Your image Competition  Customer segments  Responsiveness to customers  Amount of investment  Profitability  Risk Target market Please rate the factors on the importance scale. Please rate the factors on the importance scale. Sr. No. Factors Importance Scale* 4 3 2 1 1. Target market 2. Goods & service growth potential 3. Trends (fashion) 4. Your image 5. Competition 6. Customer segments 7. Responsiveness to customers 8. Amount of investment 9. Profitability 10. Risk * 4- extremely important, 3- important, 2- partially important, 1- not important 17. What are the factors important to you when planning merchandise quality? 17. What are the factors important to you when planning merchandise quality?  Target market  Competition  Your image  Store location  Stock turnover  Profitability  The customer service offered  Personnel  Perceived goods/service benefits  Manufactures vs. private brands Rate the factors on the importance scale. Rate the factors on the importance scale. 415 Sr. No. Factors Importance Scale* 4 3 2 1 1. Target market 2. Competition 3. Your image 4. Store location 5. Stock turnover 6. Profitability 7. Manufacturers Vs private brands 8. The customer service offered 9. Personnel 10. Perceived goods/service benefits * 4- extremely important, 3- important, 2- partially important, 1- not important Sr. No. Factors Importance Scale* 4 3 2 1 1. Target market 2. Competition 3. Your image 4. Store location 5. Stock turnover 6. Profitability 7. Manufacturers Vs private brands 8. The customer service offered 9. Personnel 10. Perceived goods/service benefits * 4- extremely important, 3- important, 2- partially important, 1- not important Sr. No. Factors Importance Scale* 4 3 2 1 1. Target market 2. Competition 3. Your image 4. Store location 5. Stock turnover 6. Profitability 7. Manufacturers Vs private brands 8. The customer service offered 9. Personnel 10. Perceived goods/service benefits * 4- extremely important, 3- important, 2- partially important, 1- not important 25. Do you provide newly launched products/brands? If ‘Yes’ please, specify- 25. Do you provide newly launched products/brands? If ‘Yes’ please, specify- 27. Do you provide complementary products (merchandise) / items? If ‘Yes’ please, specify- 28. How do you handle unprofitable products & services-  Find a way to make them profitable  Drop them from the assortment  Determine if they are related to profitable business retailing relationships  Others  None 20. Which categories of merchandise does your store offer?  Basic (Staples)  Fashion  Consumer Durable  Seasonal 21. Do you provide a wide range of products (merchandise) in each category? 21. Do you provide a wide range of products (merchandise) in each category? Yes/No . o you p ov de w de ge o p oduc s ( e c d se) e c c ego y? Yes/No 416 If ‘Yes’ please, specify- 22. Do you provide different product (merchandise) lines in each category? Yes/No If ‘Yes’ please, specify- ___________________________________________________________________ 23. Do you provide different brands in each category of merchandise? Yes/No If ‘Yes’ please, specify- ___________________________________________________________________ 24. Do you provide customised products (merchandise) to customers? Yes/No If ‘Yes’ please, specify- __________________________________________________________________ 25. Do you provide newly launched products/brands? Yes/No If ‘Yes’ please, specify- __________________________________________________________________ 26. Do you provide unique products (merchandise) to the customers? Yes/No If ‘Yes’ please, specify- _____________________________________________________________________ 27. Do you provide complementary products (merchandise) / items? Yes/No If ‘Yes’ please, specify- _____________________________________________________________________ 28. How do you handle unprofitable products & services-  Find a way to make them profitable  Determine if they are related to profitable business retailing relationships  Drop them from the assortment  Others  None 29. Who are your main suppliers? M f Wh l l 29. Who are your main suppliers?  Wholesalers  Manufacturers 417  Merchant Brokers  Commission Agents  Trade Shows  The Cooperatively Owned Offices  The Independent Owned Offices  The Individually Owned Offices  Merchant Brokers 30. What type of discounts your stores get from suppliers? 30. What type of discounts your stores get from suppliers?  Trade  Quantity  Promotional  Promotional  Cash  Trade  Quantity  Promotional  Promotional If ‘Yes’, for how much time- 36. Which pricing strategy do you use?  Market-skimming pricing  Market-penetration pricing  Variable (zone) pricing  Markup pricing  Target return pricing  Perceived value pricing  Going rate pricing  Value pricing Everyday low pricing (EDLP) High-low pricing  Target return pricing 31. Which characteristics are important for you while selecting a vendor? 31. Which characteristics are important for you while selecting a vendor  Reliability  Price-quality  Order processing time  Exclusive rights  Functions provided  Information  Ethics  Guarantee  Credit  Long-term relationship  Reorders  Markup  Innovativeness  Local advertising  Investment  Risk Please rate the characteristics on the importance scale. Please rate the characteristics on the importance scale. Sr. No. Characteristics Importance Scale* 4 3 2 1 1. Reliability 2. Price-quality 3. Order processing time 4. Exclusive rights 5. Functions provided 6. Information 7. Ethics 8. Guarantee 9. Credit 10. Long-term relationship 11. Reorders 12. Markup 13. Innovativeness 14. Local advertising 15. Investment 16. Risk * 4- extremely important, 3- important, 2- partially important, 1- not important 32. Do you make stock planning? Yes/No If ‘Yes’, for how much time- 418  Weekly planning  Yearly planning  Monthly planning  Long term planning  Monthly planning  Long term planning 33. Do you work on partner relationship with your suppliers/vendors? Yes/No 33. Do you work on partner relationship with your suppliers/vendors? Yes/No 35. Which pricing approach do you follow to set the price?  Demand-oriented pricing  Cost-oriented pricing  Leadership-oriented pricing  Cost-oriented pricing 34. What is your main objective of Pricing?  Maximising market skimming  Product quality leadership  Survival  Maximising current profit  Maximising market share 37. Which price-adjustment strategy do you use?  Discount & allowance pricing- reducing prices to reward customer responses such as paying early or promoting the product.  Segmented pricing- adjusting the price to allow for differences in customers, products, or locations.  Psychological pricing- adjusting prices for psychological effect.  Promotional pricing- temporarily reducing prices to increase short-run sales.  Geographical pricing- adjusting prices to account for the geographic location of the customer.  Dynamic pricing- adjusting prices continually to meet the characteristics and needs of individual customers and situations.  International pricing- adjusting prices for the international market. 419 41. How do you create Brand Image?  Appeal to emotion  Appeal to reason  Appeal to senses 38. Which tactics do you follow to fight with price war? 38. Which tactics do you follow to fight with price war?  Value added retailing  Maintain & grow  Customer education  Quit 39. Do you ask the customers before price change? Yes/No 40. Do you charge always less than MRP? Yes/No 41. How do you create Brand Image?  Appeal to reason  Appeal to senses  Appeal to emotion b) Free samples: please specify_ c) Gifts c) Gifts 1. Free gift with purchase of specific items 2. Gift hampers for special occasions 3. Assured gift & premiums 4. Others, please specify_____________ d) Rebates & discounts 1. Discounts on purchase of specific items 2. Discount on bulk purchase 3. Combo-pack discounts 4. EMI 5. Vouchers & coupons 6. Others, please specify ____________________ e) Customer loyalty programs, prizes & rewards: Please specify_________________________________________ f) Contest 1. Lucky draws, sweepstakes 2. Holiday activities 3. Fashion shows, art exhibits 4. Others, please specify ________________________ ________________________ Please rate the following statements on agreement scale. Statements Agreement Scale* 4 3 2 1 Sales promotions influence customer purchase intention. 421 1. Free gift with purchase of specific items 2. Gift hampers for special occasions 3. Assured gift & premiums 4. Others, please specify___ 4. Others, please specify_____________ 42. Do you use any method to develop a brand image & customer loyalty 42. Do you use any method to develop a brand image & customer loyalty- Yes/No 42. Do you use any method to develop a brand image & customer loyalty- Yes/No If yes, please specify which promotional tools are you using to entice customers?  Advertising  Sales promotion  Direct retailing  Personal retailing  Public relation/publicity  Internet/interactive retailing  Advertising  Sales promotion  Direct retailing If advertising, a) In-store 1. Visual display 2. Signage 3. Pamphlets 4. Point of purchase material b) Out-Store 8. Van advertising 9. Online advertising 10.Mobile billboard 11.LCD screening 12.Cinema 13.Fairs & exhibitions 14.Others, please specify___ 1. Television 3. Newspapers 4. Journals & magazine 6. Hoardings, wall paintings 7. Pamphlets/propaganda 420 If, sales promotions a) Schemes 1. Membership 2. Monthly shopping 3. Limited offers 4. Best buys of the month 5. Buy & get free 6. Discount on bulk purchase 7. Combo packs discount 8. Exclusive schemes for members 9. Others, please specify__________ If, sales promotions a) Schemes 1. Membership 2. Monthly shopping 3. Limited offers 4. Best buys of the month 5. Buy & get free 6. Discount on bulk purchase 7. Combo packs discount 8. Exclusive schemes for members 9. Others, please specify__________ d) Rebates & discounts * 4- extremely agree, 3- agree, 2- partially agree, 1- not agree If, direct retailing a) Direct mail b) Tele-retailing c) Direct response on TV/radio d) Mail order catalogues e) SMS retailing f) Mobile retailing g) Others, please specify ___________ ___________________________ If, personal retailing a) Face to Face Retailing b) Personal online chats c) Person to person video conference d) Person to person telephonic conversation e) Others, please specify ___________ If, internet / interactive retailing a) E-mails b) Online discussion forums c) Weblogs d) Web sites e) e-pay f) Social networking sites g) Others, please specify ___________ ___________________________ If, public relation / publicity retailing a) Corporate Brochures b) Posters c) Annual Reports d) Press Release e) Press conference f) Video press conference g) Internet release h) Stores tours for media i) Meeting with customers j) Special events & sponsorships k) Social events l) Social development programs 422 If, direct retailing a) Direct mail b) Tele-retailing c) Direct response on TV/radio d) Mail order catalogues e) SMS retailing f) Mobile retailing g) Others, please specify ___________ _______________________ d) Person to person telephonic conversation e) Others, please specify d) Person to person telephonic conversation e) Others, please specify d) Rebates & discounts 1. Discounts on purchase of specific items 2. Discount on bulk purchase 3. Combo-pack discounts 4. EMI 5. Vouchers & coupons 6. Others, please specify ____________________ 1. Discounts on purchase of specific items 2. Discount on bulk purchase 3. Combo-pack discounts 4. EMI 5. Vouchers & coupons 6. Others, please specify ____________________ 3. Combo-pack discounts e) Customer loyalty programs, prizes & rewards: Please specify_________________________________________ f) Contest 1. Lucky draws, sweepstakes 2. Holiday activities 3. Fashion shows, art exhibits 4. Others, please specify ________________________ ________________________ Please rate the following statements on agreement scale. Statements Agreement Scale* 4 3 2 1 Sales promotions influence customer purchase intention. e) Customer loyalty programs, prizes & rewards: Please specify_________________________________________ f) Contest 1. Lucky draws, sweepstakes 2. Holiday activities 3. Fashion shows, art exhibits 4. Others, please specify _______________________ _______________________ g g Statements Agreement Scale* 4 3 2 1 Sales promotions influence customer purchase intention. 421 Sales promotions increase customer satisfaction level. Sales promotions affect the frequency to visit the store. Sales promotions play a vital role to spend more. Sales promotions entice to repurchase. Customers purchase more quantity due to sales promotion. Your store does adequate sales promotion activities. Your store justifies its punch line. * 4- extremely agree, 3- agree, 2- partially agree, 1- not agree Your store does adequate sales promotion activities. Your store justifies its punch line. 43. How much is spent on advertising- 44. Are you using celebrity services for adverting? Yes/No If Yes, Please Specify the Name- , p y _________________________________________ 45. Do you use a tag line and logo? Yes/No If Yes, Please Specify- ________________________________________________ 45. Do you use a tag line and logo? Yes/No If Yes, Please Specify- ________________________________________________ If yes, which of the following category of tagline do you use? If yes, which of the following category of tagline do you use?  Business-Centric  Relationship-Centric  Other, specify  Other, specify  Relationship-Centric  Customer-Centric __________________  Customer-Centric 46. Which location do you prefer to open the stores? 46. Which location do you prefer to open the stores?  Downtown location  Main street location  Downtown location  Downtown location  Inner city location  Main street location  Inner city location  Inner city location If, public relation / publicity retailing a) Corporate Brochures b) Posters c) Annual Reports d) Press Release e) Press conference f) Video press conference g) Internet release h) Stores tours for media i) Meeting with customers j) Special events & sponsorships k) Social events l) Social development programs g) Internet release h) Stores tours for media i) Meeting with customers j) Special events & sponsorships k) Social events l) Social development programs 422 m) Others, please specify 43. How much is spent on advertising- 47. Generally, where do you prefer to open your stores?  Shopping malls/centres  Strip centres  Unplanned centres  Community centres  Neighbourhood centres  Regional centres  Outlet centres  Theme/festival centres  Power centres  Lifestyle centres  Shopping malls/centres  Strip centres  Unplanned centres  Community centres  Neighbourhood centres  Regional centres  Outlet centres  Theme/festival centres  Power centres  Lifestyle centres 48. Which are the important factors, you consider while selecting a retail location. 48. Which are the important factors, you consider while selecting a retail locat  Credit  Delivery  Alterations  Installations  Packaging (gift wrapping)  Complaints & returns handling  Gift certificates  Trial purchase  Special sales for regular customers  Staff on late hours  Extended store hours  Mail & phone orders  Bridal registry  Interior designers  Personal shoppers  Ticket outlets  Parking  Water fountains  Payphones  Baby strollers  Restrooms  Restaurants  Babysitting  Fitting rooms  Beauty salon  Fur storage  Shopping bags  Information  Safety of personal things  Lift & escalator facility  After sale services  Demonstrations  Mobile and laptop charger points  Entertainment facilities  Refreshment facilities  Insurance cover  Medical facility  The internet billing facility 48. Which are the important factors, you consider while selecting a retail locat 48. Which are the important factors, you consider while selecting a retail location  Population size & characteristics  Closeness to source of supply  Promotion facilities  Promotion facilities  Availability of labour  Availability of labour  Economic base  Economic base 423 Please rate the following factors on the importance scale. Sr. No. Factors Importance Scale* 4 3 2 1 1. Population size & characteristics a. Total size & density b. Age distribution c. Average educational level d. % of residents owning homes e. Total disposable income f. Per capita disposable income g. Occupation distribution h. Trends 2. Availability of labour a. Management b. Management trainee c. Clerical d. Labour 3. Closeness to source of supply a. Delivery cost b. Timeliness c. No of manufacturers & wholesalers d. Availability & reliability of merchandise lines 4. Promotion facilities a. Availability & frequency of media b. Costs c. Waste 5. Economic base a. Dominant industry b. Extent of diversification c. Growth projections d. Freedom from economic & seasonal fluctuations e. Availability of credit & financial facilities 6. Competitive situation a. Number & size of existing competitors b. Strength & weakness of competitors c. Short & long run outlook d. Level of saturation 7. Availability of store locations a. Number & type of locations b. Cost of site c. Access to transportation d. Owning V/S leasing opportunities e. Leasing period & conditions Please rate the following factors on the importance scale. 424 f. Visibility of site g. Size & shape of trading area h. Zoning restrictions i. Parking facilities j. Size & shape of the building 8. Regulations a. Taxes b. Licencing c. Operations d. Minimum wages e. Zoning * 4- extremely important, 3- important, 2- partially important, 1- not important 0. What customer services are provided by your store? 52. Does your store use GAP model for improving retail customer services? If, yes  Knowledge gap  Standard gap 53. Does your retail store have a CRM?  Yes  No  Partial  Yes 51. Which customer service strategy do you adopt?  Patronage builders (high-cost activities that are the primary factors behind customer loyalties, like transaction speed & credit)  Patronage solidifiers (the “low-cost little things” that increase loyalty, like courtesy & suggestion selling)  Disappointers (expensive activities that do no real good, like weekday deliveries for two-earner families)  Basics (low-cost activities that are “naturally expected”. They don’t build patronage, but their absence could reduce patronage, like free parking, in-store directories) 52. Does your store use GAP model for improving retail customer services? 50. What customer services are provided by your store? 425  Internet facility (Wi-Fi)  Game zones  Music zone (karaoke, etc.)  Special facilities for physically handicapped  Others, please specify  Internet facility (Wi-Fi)  Game zones  Music zone (karaoke, etc.)  Special facilities for physically handicapped  Others, please specify  Special facilities for physically handicapped  Special facilities for physically handicapped  Internet facility (Wi-Fi)  Game zones  Others, please specify 51. Which customer service strategy do you adopt? 54. What information do you collect in the customer database?  Customer Transactions- Purchase date, price paid, product purchased, Responses to the promoted products  Customer contacts- A record of customer interaction with the store by websites, inquiries in-store kiosks, telephone calls  Customer Preference- Favourite colour, brands, fabrics, flavours, apparel size  Descriptive information-Demographic & psychographic data of customers  Customer Responses to Marketing Activity-Responsiveness to sales promotion, advertising 426 55. What CRM programs do you use?  Programs for Customer Retention Frequent Shoppers Programs Special Customer Services Personalisation of Customers To develop the sense of community among customers  Programs to Convert Satisfied Customers in High Life Time Value Customers Cross Selling Add on Selling  Programs to Deal with Unprofitable Customers 56. Is there a separate cell, to help the customers in your store- Yes/No If yes, how do you solve the problems of the customers? __________________________________________________ 57. On what basis and by what method do you receive information about your customers- __________________________________________________ 58. What do you consider your weak spots with key customers- ______________________________________________________________ 59. With which customer from following category you deal most-  Retail clientele Corporate clientele  Correspondent clientele 60. Does your store regularly analyse the profitability of consumer retailing relationship- Yes/No If yes, how do you analyse them?  At household level  At the account level  At the individual level If yes, how do you analyse them?  At household level  At the account level 58. What do you consider your weak spots with key customers- 59. With which customer from following category you deal most-  Retail clientele Corporate clientele  Correspondent clientele 60. Does your store regularly analyse the profitability of consumer retailing relationship- Yes/No If yes, how do you analyse them?  At household level  At the account level  At the individual level How do you perform the analysis? 59. With which customer from following category you deal most-  Retail clientele Corporate clientele  Correspondent clientele 60. Does your store regularly analyse the profitability of consumer retailing relationship- Yes/No 59. With which customer from following category you deal most  Retail clientele Corporate clientele  Correspondent clientele 60. Does your store regularly analyse the profitability of consumer retailing relationship- Yes/No 60. Does your store regularly analyse the profitability of consumer retailing relationship- Yes/No 56. Is there a separate cell, to help the customers in your store- If h d l h bl f h ? 57. On what basis and by what method do you receive information about your customers- 55. What CRM programs do you use?  Programs for Customer Retention Frequent Shoppers Programs Frequent Shoppers Programs Special Customer Services Personalisation of Customers To develop the sense of community among customers  Programs to Convert Satisfied Customers in High Life Time Value Customers Cross Selling Add on Selling  Programs to Deal with Unprofitable Customers 56. Is there a separate cell, to help the customers in your store- Yes/No If yes, how do you solve the problems of the customers? __________________________________________________ 57. On what basis and by what method do you receive information about your customers- __________________________________________________ 58. What do you consider your weak spots with key customers- ______________________________________________________________ 59. With which customer from following category you deal most-  Retail clientele Corporate clientele  Correspondent clientele 60. Does your store regularly analyse the profitability of consumer retailing relationship- Yes/No If yes, how do you analyse them?  At household level  At the account level  At the individual level How do you perform the analysis? 427 Special Customer Services Personalisation of Customers Personalisation of Customers To develop the sense of community among customers  Programs to Convert Satisfied Customers in High Life Time Value Customers Cross Selling Add on Selling  Programs to Deal with Unprofitable Customers 56. Is there a separate cell, to help the customers in your store- Yes/No If yes, how do you solve the problems of the customers? __________________________________________________ 57. On what basis and by what method do you receive information about your customers- __________________________________________________ 58. What do you consider your weak spots with key customers- ______________________________________________________________ 59. With which customer from following category you deal most-  Retail clientele Corporate clientele  Correspondent clientele 60. Does your store regularly analyse the profitability of consumer retailing relationship- Yes/No If yes how do you analyse them? To develop the sense of community among customers  Programs to Convert Satisfied Customers in High Life Time Value Customers Cross Selling Add on Selling  Programs to Deal with Unprofitable Customers 56. Is there a separate cell, to help the customers in your store- Yes/No If yes, how do you solve the problems of the customers? 56. Is there a separate cell, to help the customers in your store- Yes/No If yes, how do you solve the problems of the customers? 64. Which methods does your store adopt to display the products?  Window displays  Entrance display  Free standing display  End caps  Promotional aisles  Walls display  Dressing room display  Cash wraps display  Window displays  Dressing room display  Cash wraps display  End caps 63. Which types of Signage are used by your store to locate the product category? 63. Which types of Signage are used by your store to locate the product category?  Category  Point of sale  Point of sale  Category  Promotional  Digital signage  Promotional  Digital signage 61. Are you providing e-retailing service to customers-  Yes  No  Yes 62. Which layout does your store adopt while planning store design? How do you perform the analysis? 427  Vendor provided CRM software  Departmental analysis of their CR  “Homegrown” CRM software  Periodic review by a professional cost accountant  Other 65. Which type of fixtures does your store use to display products?  Straight racks  Rounder (bulk fixture/capacity fixture)  Four-way fixture (feature fixture)  Gondolas  Four-way fixture (feature fixture)  Straight racks  Gondolas 66. Which presentation techniques are used by your store to present the products? 66. Which presentation techniques are used by your store to present the products? 62. Which layout does your store adopt while planning store design?  Free flow layout  Grid layout  Loop layout  Spine layout  Free form or boutique layout  Racetrack layout INDIVIDUAL QUESTIONNAIRE Name of the city: _ 66. Which presentation techniques are used by your store to present the products What are your plans in the new future at store-level? ________________________________________________________________ ________________________________________________________________ ________________________________________________________________ ________________________________________________________________ ________________________________________________________________ Thank you so much. INDIVIDUAL QUESTIONNAIRE Name of the city: _______________ Name of the Store: __________________ CUSTOMER PROFILE 69. What are your plans in the new future at store-level? ________________________________________________________________ ________________________________________________________________ ________________________________________________________________ ________________________________________________________________ ________________________________________________________________ 66. Which presentation techniques are used by your store to present the products  Idea-oriented presentation  Item & size presentation  Colour presentation  Price lining  Vertical merchandising  Tonnage merchandising  Frontal presentation  Item & size presentation 428 Which variables are used by your store to create an appealing store atmosphere? 68. Please rate the following variables of store atmospherics on the importance scale, which entice the customers. S. N. Measurement Variables Importance Scale* 4 3 2 1 1. Exterior atmospherics 1 Storefront 2 Store marquee/sign board 3 Frontage entrance 4 External display window/space 5 Building 6 Store architecture 7 Uniqueness of store 8 Visibility of store 9 Location/site of the store 10 Parking facilities 11 Exterior look of store 12 Ease of access 13 Store nameplate 14 Health & safety standards 15 Surrounding area 2. Interior atmospherics 1 Flooring & ceiling 2 Lighting 3 Odour/smell/scent 4 Fixtures 5 Wall textures 6 Sounds 7 Colour 8 Temperature 9 Cleanliness 10 Aisles 11 Trial (experience) room/zone 12 Dead area 13 Merchandise 14 Graphics 15 Store size 3. Store layout 1 Space planning 2 Floor space allocation 3 E-display 4 Department location 5 Department size 68. 429 6 Department compatibility 7 Location of various merchandise 8 Accessibility to cash counters 9 Signage 10 Space/merchandise category 4. Visual merchandising 1 Assortment 2 Theme 3 Racks & shelves 4 Ensemble 5 Payment counters 6 Merchandise display 7 The planogram 8 Props & mannequins 9 Point of sale sign 10 Methods of display * 4- extremely important, 3- important, 2- partially important, 1- not important 69. What are your plans in the new future at store-level? ________________________________________________________________ ________________________________________________________________ ________________________________________________________________ ________________________________________________________________ ________________________________________________________________ 6 Department compatibility 7 Location of various merchandise 8 Accessibility to cash counters 9 Signage 10 Space/merchandise category 4. Visual merchandising 1 Assortment 2 Theme 3 Racks & shelves 4 Ensemble 5 Payment counters 6 Merchandise display 7 The planogram 8 Props & mannequins 9 Point of sale sign 10 Methods of display * 4- extremely important, 3- important, 2- partially important, 1- not important 6 Department compatibility 7 Location of various merchandise 8 Accessibility to cash counters 9 Signage 10 Space/merchandise category 4. Visual merchandising 1 Assortment 2 Theme 3 Racks & shelves 4 Ensemble 5 Payment counters 6 Merchandise display 7 The planogram 8 Props & mannequins 9 Point of sale sign 10 Methods of display * 4- extremely important, 3- important, 2- partially important, 1- not important 69. CUSTOMER PROFILE 3. Contact Number & Email: 430 430 4. Residing area: Urban Semi-urban Rural 5. Gender:  Male  Female 6. Age (in years):  Below 21  21-30  31-40  41-50  51-60  Above 60 7. Education:  Illiterate  Below 10th (HSC)  10th -12th (HSC - SSC)  Graduate  Postgraduate  Other Plz., Specify 8. Occupation:  Salaried  Self-employed  Homemaker  Student  Retired 9. Family Income (Yearly):  Less than 1 Lakh  1 to 2 Lakhs  2 to 5 Lakhs  5 to 10 Lakhs  More than 10 Lakhs 10. Marital Status:  Single  Married 11. Nature of Family:  Nuclear  Joint 12. Family Size  Less than 4 Members  4-5 Members  6-8 Members  More than 8 Members 13. Religion:  Hindu  Muslim  Buddhists Semi-urban  6-8 Members  Buddhists 431  Other plz., specify Christian BUYING BEHAVIOUR 14. Generally, who take the purchase making decisions in your family?  Self  Spouse  Children  Parents  Others 15. Do you collect the information, while purchase decision making? If yes, from where you collect information regarding the products & services?  Salespersons  Telemarketing  Advertising  In-store information  In-store display  Package labels  Family members  Neighbours/Friends/Relatives  Through Internet (www)  Others, please specify _______________________ -If advertising, by which source?  Television  Newspapers  Magazines  Hoardings  Pamphlets  Radio  Cinema  Celebrity endorsement  Websites  Others -If advertising, by which source?  Television  Newspapers  Magazines  Hoardings  Pamphlets  Radio  Cinema  Celebrity endorsement  Websites  Others 15. Do you collect the information, while purchase decision making?  Yes  No 23. Which category of the products do you usually purchase?  Basic (staples) merchandise  Seasonal merchandise  Fashion merchandise  Consumer durable merchandise 21. When you prefer to visit this store?  Morning  Afternoon  Evening  Any time 22. How much do you spend for purchasing from this store in a month?  Less than 1000 ₹  1000-2000 ₹  2001-5000 ₹  More than 5000 ₹ 19. How many times do you visit the store in a month? 19. How many times do you visit the store in a month?  1 Time  2 Times  3 Times  4 Times  More than 4 times 18. For how long, you have been visiting this retail store.  First time Last 3 months Last 1 year More than 2 years 19. How many times do you visit the store in a month?  1 Time  2 Times  3 Times  4 Times  More than 4 times 20. On which day of the week, you prefer to visit this store.  Weekdays (Monday to Friday)  Any day 16. How are you influenced to buy a product? 16. How are you influenced to buy a product?  Self  Family members  Neighbours / Friends / Colleagues  Celebrity endorsement  Salesperson in a store  Direct-mail piece  Websites  TV commercials  Infomercials  Magazines ads  TV talk shows  News  Sales Promotion  Others 432 17. Generally, who do the purchasing in your family?  Self  Spouse  Children  Parents  Surrogate 18. For how long, you have been visiting this retail store.  First time  Last 3 months  Last 6 months  Last 1 year  Last 2 year  More than 2 years 19. How many times do you visit the store in a month?  1 Time  2 Times  3 Times  4 Times  More than 4 times 20. On which day of the week, you prefer to visit this store.  Weekdays (Monday to Friday)  Weekends (Saturday & Sunday)  Any day 21. When you prefer to visit this store?  Morning  Afternoon  Evening  Any time 22. How much do you spend for purchasing from this store in a month?  Less than 1000 ₹  1000-2000 ₹  2001-5000 ₹  More than 5000 ₹ 23. Which category of the products do you usually purchase?  Basic (staples) merchandise  Fashion merchandise  Seasonal merchandise  Consumer durable merchandise 24. What type of products do you usually purchase? 17. Generally, who do the purchasing in your family? Reason for unplanned purchase Reason for unplanned purchase  Point of purchase display  Price reductions  Store layout  Coupons  Shelving techniques  Demonstration and samples  Multi-item discounts  Packaging  Store atmosphere  Salespeople  Any other, please specify __________________________ 26. While purchasing, you give the preference to- (Put number 1, 2, 3, 4, 5, 6 & 7 in order of preference) (Put number 1, 2, 3, 4, 5, 6 & 7 in order of preference)  Brand  Price  Quality  Quantity  Discounts  Schemes  Others  Brand  Price  Quality  Quantity  Discounts  Schemes  Others a) Merchandise a) Merchandise Merchandise assortment (variety of merchandise and brands), quality of assortment, in service situations, merchandise offered as part of service, styling or fashion guarantee, warranties and pricing, reasonable prices b) Service Service in general, ease of merchandise return, delivery service, salesclerk service, phone ordering, presence of self-service, and credit policy 25. Do you make unplanned purchasing from this store? 25. Do you make unplanned purchasing from this store?  Yes  No If yes, what type of products_____________________________________? Reason for unplanned purchase  Point of purchase display  Price reductions  Store layout  Coupons  Shelving techniques  Demonstration and samples  Multi-item discounts  Packaging  Store atmosphere  Salespeople  Any other, please specify ___________________________  Yes  No If yes, what type of products_____________________________________? _? 24. What type of products do you usually purchase?  Food & Grocery  Health & Beauty  Home & Furniture  Fashion  Electronics  Toys  Sports  Other, please specify  Food & Grocery  Health & Beauty  Home & Furniture  Fashion  Electronics  Toys  Sports  Other, please specify  Food & Grocery  Health & Beauty  Home & Furniture  Fashion  Electronics  Toys  Sports  Other, please specify 433 433 g) Store Atmospheres g) Store Atmospheres Atmosphere of congeniality, customers’ feeling of warmth, acceptance or ease, good & spacious ambience, and delighting store environment g) Store Atmospheres Atmosphere of congeniality, customers’ feeling of warmth, acceptance or ease, good & spacious ambience, and delighting store environment h) Institutional factors Projection of store (conservative or modern), reputation, reliability i) Post-transaction satisfaction Merchandise in use and return & adjustment policy j) Others Friends suggested, no other store nearby, convenient store hours 28. Do you visit any other store apart from this?  Yes  No If yes,  Organised retail store  Unorganised retail store  Both If yes,  Organised retail store  Unorganised retail store  Both Reasons for visiting this other store Reasons for visiting this other store e) Convenience f) Promotion Advertising, displays, trading stamps, sales promotion (discount & special schemes), symbol and colours c) Clientele c) Clientele Social class appeal, store personnel, Fit between self-image and store image (self-image congruence) 434 Elevators, lighting, air conditioning, restrooms, store layout, shopping ease, aisle placement, carpeting and architecture. Elevators, lighting, air conditioning, restrooms, store layout, shopping ease, aisle placement, carpeting and architecture. e) Convenience General convenience, parking access, location convenience (proximity to home/office) e) Convenience General convenience, parking access, location convenience (proximity to home/office) c) Clientele d) Physical Facilities Elevators, lighting, air conditioning, restrooms, store layout, shopping ease, aisle placement, carpeting and architecture. a) Merchandise Merchandise assortment (variety of merchandise and brands), quality of assortment, in service situations, merchandise offered as part of service, styling or fashion guarantee, warranties and pricing, reasonable prices Merchandise assortment (variety of merchandise and brands), quality of assortment, in service situations, merchandise offered as part of service, styling or fashion guarantee, warranties and pricing, reasonable prices 435 b) Service Service in general, ease of merchandise return, delivery service, salesclerk service, phone ordering, presence of self-service, and credit policy f) Promotion g) Store Atmospheres Atmosphere of congeniality, customers’ feeling of warmth, acceptance or ease, good & spacious ambience, and delighting store environment i) Post-transaction satisfaction Merchandise in use and return & adjustment policy j) Others Friends suggested, no other store nearby, convenient store hours d) Physical Facilities Elevators, lighting, air conditioning, restrooms, store layout, shopping ease, aisle placement, carpeting and architecture. e) Convenience General convenience, parking access, location convenience (proximity to home/office) f) Promotion Advertising, displays, trading stamps, sales promotion (discount & special schemes), symbol and colours b) Service c) Clientele Social class appeal, store personnel, Fit between self-image and store image (self-image congruence) j) Others 436 FACTORS INFLUENCING BUYING BEHAVIOUR 29. How is your buying behaviour influenced by following Factors? Please rate them on a scale ranging from +2 to -2 (by Tick Mark). S. N. Factors 2 1 0 -1 -2 1 Sociological / Environmental / External factors Cultural, sub-cultural, society, status in society (social class), reference group, family life cycle, and role in the family 2 Psychological / Individual / Internal factors Self-image, personality, personal values, lifestyle, memory, learning, perception, motivation, emotions, mood, involvement, information processing, beliefs, intention, attitude, communication, and persuasion 3 Store related factors Store image, store atmosphere, store brands, store location, sales personnel, discounts & schemes, and free services. *2 = Extremely Influenced, 1 = Influenced, 0 = Neither Influenced nor uninfluenced, -1 = Uninfluenced, -2 = Not at all influenced CUSTOMER SATISFACTION 30. How do you rate the variables of the store or retailers mentioned below describing the products (merchandise) & services, prices, promotion, premises / place, presentation, personnel / people, physical evidence on the expectation cum satisfaction scale? Please rate these variables on a five-point scale (by tick mark). The scales are ranging like following- 30. How do you rate the variables of the store or retailers mentioned below describing the products (merchandise) & services, prices, promotion, premises / place, presentation, personnel / people, physical evidence on the expectation cum satisfaction scale? Please rate these variables on a five-point scale (by tick mark). The scales are ranging like following- Expectation cum satisfaction scale-  A rating of 5 indicates that the variable is greatly exceeding your expectations and you are delighted.  A rating of 4 indicates that the variable is exceeding your expectations and you are happy.  A rating of 3 indicates that the variable is matching your expectations and you are satisfied.  A rating of 2 indicates that the variable is less than your expectations you are dissatisfied. 2 Merchandise categories in the assortment Basic or staple, fashion, seasonal, consumer durable etc. 3 Variety of merchandise in each category As per different sizes, colours, designs, price, quality, quantity and brands 4 Availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise 5 Brands in each merchandise category Local & regional brands, national brands, international brands, private (store) brands and others, i.e. copycat, exclusive, generic brands 6 Quality assurance and quality of merchandise Services offered 7 Services Credit, home delivery, alteration, installation, packaging (gift wrapping), complaints & returns handling, gift certificates, trial purchase, special sales for regular customers, staff on late hours, extended store hours, mail & phone orders, bridal registry, babysitting, interior designers, personal shoppers, baby strollers, ticket outlets, parking, water fountains, pay phones, restaurants, fitting rooms, beauty salon, fur storage, shopping bags, restrooms, information, safety of personal things, lift & escalator facility, after sale services, demonstrations, mobile and laptop charger points, entertainment facilities, refreshment facilities, the internet billing facility, game zones, music zone (karaoke, etc.), special facilities for physically handicapped, other in-store services, other out-store services, other pre-purchase or post-purchase services, complementary (adjunct) services (i.e. inventory support, installation, repair) 8 Quality of services As per the ability to manage its tangibles (tangibility), consistency and dependability in service performance (reliability), commitment to providing services in time (responsiveness), competence, courtesy to its customers and the security of its operations (assurance), ability to put itself in its customers place (empathy) Price 9 The pricing of merchandise & services 10 The overall value (product value, service value, personal value, image value) Vs price (overall cost- monetary cost, time cost, energy cost, psychic cost) of merchandise & services Premises / Place 11 Exterior atmospherics of the store 2 Merchandise categories in the assortment Basic or staple, fashion, seasonal, consumer durable etc. Expectation cum satisfaction scale- 3 Variety of merchandise in each category As per different sizes, colours, designs, price, quality, quantity and brands 4 Availability of newly launched merchandise, customised merchandise, unique merchandise and complementary merchandise 5 Brands in each merchandise category Local & regional brands, national brands, international brands, private (store) brands and others, i.e. copycat, exclusive, generic brands 6 Quality assurance and quality of merchandise Services offered 7 Services Credit, home delivery, alteration, installation, packaging (gift wrapping), complaints & returns handling, gift certificates, trial purchase, special sales for regular customers, staff on late hours, extended store hours, mail & phone orders, bridal registry, babysitting, interior designers, personal shoppers, baby strollers, ticket outlets, parking, water fountains, pay phones, restaurants, fitting rooms, beauty salon, fur storage, shopping bags, restrooms, information, safety of personal things, lift & escalator facility, after sale services, demonstrations, mobile and laptop charger points, entertainment facilities, refreshment facilities, the internet billing facility, game zones, music zone (karaoke, etc.), special facilities for physically handicapped, other in-store services, other out-store services, other pre-purchase or post-purchase services, complementary (adjunct) services (i.e. Expectation cum satisfaction scale- inventory support, installation, repair) 8 Quality of services As per the ability to manage its tangibles (tangibility), consistency and dependability in service performance (reliability), commitment to providing services in time (responsiveness), competence, courtesy to its customers and the security of its operations (assurance), ability to put itself in its customers place (empathy) Price 9 The pricing of merchandise & services 10 The overall value (product value, service value, personal value, image value) Vs price (overall cost- monetary cost, time cost, energy cost, psychic cost) of merchandise & services Premises / Place 11 Exterior atmospherics of the store Storefront, store marquee / sign board, frontage entrance, external display window / space, building, store architecture, uniqueness of the store, visibility of the store, location / site of store, parking facilities, exterior look of the store, ease of access, store nameplate, health & safety standards, surrounding area 12 Interior atmospherics of the store Flooring & ceiling, lighting, odour / smell / scent, fixtures, wall textures, sounds, colour, temperature, cleanliness, aisles, trial room, dead area, graphics, store size Presentation 13 Store layout Space planning, floor space allocation, e-display, department location, department size, department compatibility, location of various products, accessibility to cash/billing counters, signage 14 Visual merchandising Theme, racks & shelves, ensemble, payment counters, merchandise display, the planogram, props & mannequins, point of sale sign, methods of display Promotion 15 Methods of promotion and communication 16 Advertising methods In-store methods (i.e. Expectation cum satisfaction scale- visual display, signage, pamphlets, point of purchase material, secondary packaging, cross-promotion) and out-store methods (television, radio, newspapers, journals & magazine, yellow pages, hoardings, wall paintings, pamphlets/propaganda, van advertising, online advertising, mobile billboard, LCD screening, cinema, fairs & exhibitions) 17 Sales promotion methods Schemes, free samples, gifts, rebates & discounts, contests and other activities, customer loyalty programs, premiums, prize & rewards 18 Direct retailing methods Direct mail, tele-retailing, direct response on TV/radio, mail order catalogues, SMS retailing, mobile retailing 19 Personal retailing methods Face to face retailing in store, person to person telephonic conversation 20 Public relation / publicity methods Corporate brochures, posters, annual reports, press release, press conference, video press conference, internet release, store tours for media, meeting with customers, special events & sponsorships, social events, social development programs 21 Internet / interactive retailing methods E-mails, online discussion forums, weblogs, websites, e-pay, social networking sites Personnel / people 22 Sales personnel Storefront, store marquee / sign board, frontage entrance, external display window / space, building, store architecture, uniqueness of the store, visibility of the store, location / site of store, parking facilities, exterior look of the store, ease of access, store nameplate, health & safety standards, surrounding area Sales boys and sales girls 23 Contact personnel Help desk people, service desk people, billing counter people, etc. 24 Factors pertained to personnel Mood, efforts, commitment, attitude, knowledge, skills, behaviour, approach to delivering the service, competence, courtesy, credibility, reliability, responsiveness, communication style, friendliness, helpfulness, discipline & punctuality, customer counselling ability Physical (tangible) evidence 25 Other tangible evidence Employee appearance, uniforms, billing statements, stationery, business cards, signage, exterior design, reports, brochures, internet or web pages Sales boys and sales girls 23 Contact personnel Help desk people, service desk people, billing counter people, etc. 24 Factors pertained to personnel Mood, efforts, commitment, attitude, knowledge, skills, behaviour, approach to delivering the service, competence, courtesy, credibility, reliability, responsiveness, communication style, friendliness, helpfulness, discipline & punctuality, customer counselling ability Physical (tangible) evidence 25 Other tangible evidence Employee appearance, uniforms, billing statements, stationery, business cards, signage, exterior design, reports, brochures, internet or web pages CUSTOMER SUGGESTIONS 31. If, any variable or sub-variable (mentioned in question 30) falls short of your expectation and you want an improvement in it. Please mention here. S. N. Variable S. N. Variable 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 31. If, any variable or sub-variable (mentioned in question 30) falls short of your expectation and you want an improvement in it. Please mention here. 39 40 41 42 43 44 45 46 47 48 49 50 32. Any other facility, which you want from your store, please mention here. 1 6 2 7 3 8 4 9 5 10 33. Any other suggestion, please. _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ Thank you so much. MAP OF MAHARASHTRA STATE IN INDIA 39 40 41 42 43 44 45 46 47 48 49 50 32. Any other facility, which you want from your store, please mention here. 1 6 2 7 3 8 4 9 5 10 33. Any other suggestion, please. _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ Thank you so much. MAP OF MAHARASHTRA STATE IN INDIA Source: (MSEDCL, 2015) ADDITIONAL INFORMATION Source: (MSEDCL, 2015) ADDITIONAL INFORMATION Participation in seminars / conferences and papers presented  Participated in international conference on “Recontouring Business”, organised by Sadhu Vaswani Institute of Management Studies For Girls, Pune on January 22-23, 2016.  Participated in international conference on “Recontouring Business”, organised by Sadhu Vaswani Institute of Management Studies For Girls, Pune on January 22-23, 2016.  Participated in international conference on “Ongoing Research in the fields of Management Studies and Information Technology”, organised by Audyogik Shikshan Mandal, Pune on January 8-9, 2016.  Participated in international conference on “Ongoing Research in the fields of Management Studies and Information Technology”, organised by Audyogik Shikshan Mandal, Pune on January 8-9, 2016.  Presented a paper titled “NGO Management through Efficacious Human Resources” at a national conference organised by Siddhant Institute of Business Management, Pune on March 13 & 14, 2015.  Presented a paper titled “Assessing the Demographic Orientation towards Store Design” at a national conference on “Emerging Trends in Information Technology and Business Management” organised by Dr D. Y. Patil Institute of Management, Pune, on March 9, 2012.  Presented a paper titled “Corporate Social Responsibility Practices in India” at a national conference on “Recent Amendments in Corporate Laws inclusive of Direct & Indirect Laws” organised by Indrayani Mahavidyalaya, Talegaon on February 16 to 18, 2012.  Presented a paper titled “Islamic Banking: Strategic Issues and Challenges in India” at a national conference on “Strategic Management for Today’s Business” organised by Dept. of Business Management, Siddhant College of Engineering, Pune on October 7, 2011.  Presented a paper titled “Store Image - A Critical Success Factor in Dynamic Retailing Environment”, at a national seminar on “Economic Environment of Business” organised by Sir M Visvesvaraya Institute of Management Studies & Research, Mumbai on September 24, 2011. Publications Publications Publications  Atul Kumar and Vinaydeep, “Intrinsic Reward System & Motivation: A Study of Management Teachers Perspective” in International Journal of Human Resource Management and Research (IJHRMR), Vol. 2, Issue 4, December 2012, pp. 33-44, ISSN: 2249-6874 (Print) & 2249-6874 (Online), Impact Factor (JCC): 2.9966.  Atul Kumar and Vinaydeep Brar, “Measurement of the Demographic Inclination towards Store Image” in Vishwakarma Business Review, Vol. 2, Issue 2, July 2012, pp. 88-93, ISSN: 2229-6514 (Print) & 2230-8237 (Online).  Atul Kumar and Vinaydeep Brar, “Measurement of the Demographic Inclination towards Store Image” in Vishwakarma Business Review, Vol. 2, Issue 2, July 2012, pp. 88-93, ISSN: 2229-6514 (Print) & 2230-8237 (Online).  Atul Kumar, “Store Image - A Critical Success Factor in Dynamic Retailing Environment: An Investigation” in Asian Journal of Management, Vol. 3, Issue 1, January-March 2012, pp. 01-05, ISSN 0976- 495X (Print).  Atul Kumar, “Store Image - A Critical Success Factor in Dynamic Retailing Environment: An Investigation” in Asian Journal of Management, Vol. 3, Issue 1, January-March 2012, pp. 01-05, ISSN 0976- 495X (Print).  Atul Kumar, “The Changing Buying Behaviour of Customers in Organised Retail Sector of Pune City” in International Journal of Research in Social Sciences, Vol. 2, Issue 1, February 2012, pp. 242-263, ISSN: 2249-2496 (Online).  Atul Kumar, “The Changing Buying Behaviour of Customers in Organised Retail Sector of Pune City” in International Journal of Research in Social Sciences, Vol. 2, Issue 1, February 2012, pp. 242-263, ISSN: 2249-2496 (Online).  Atul Kumar, “Measuring the Women’s Involvement in Purchase Making Decisions” in International Journal of Marketing and Technology, Vol. 2, Issue 2, February 2012, pp. 255-276, ISSN: 2249-1058 (Online), IC Value: 5.9.  Atul Kumar, “Measuring the Women’s Involvement in Purchase Making Decisions” in International Journal of Marketing and Technology, Vol. 2, Issue 2, February 2012, pp. 255-276, ISSN: 2249-1058 (Online), IC Value: 5.9.  Atul Kumar and Vinaydeep, “Purchase Decisions: An Empirical Study of Women’s Role in Family” in KBSCMR’s Journal of Management Research, Vol. 2, December 2011, pp. 43-54, ISSN 0973-1513 (Print).  Atul Kumar and Vinaydeep, “Purchase Decisions: An Empirical Study of Women’s Role in Family” in KBSCMR’s Journal of Management Research, Vol. 2, December 2011, pp. 43-54, ISSN 0973-1513 (Print).  Atul Kumar, “Demographic & Geographic Inclination towards Store Design: a Study of Shopping Mall Customers in Maharashtra-State, India” in Asian Journal of Management, Vol. 2, Issue 3, July-Sept. 2011, pp. Publications 87-93, ISSN 0976- 495X (Print).  Atul Kumar, “Demographic & Geographic Inclination towards Store Design: a Study of Shopping Mall Customers in Maharashtra-State, India” in Asian Journal of Management, Vol. 2, Issue 3, July-Sept. 2011, pp. 87-93, ISSN 0976- 495X (Print).  Atul Kumar, “Islamic Banking: Strategic Issues and Challenges in India” in proceedings of a national conference on Strategic Management for Today’s Business organised by Siddhant College of Engineering, Pune, ISBN 978-81- 8465-811-8.  Atul Kumar, “Islamic Banking: Strategic Issues and Challenges in India” in proceedings of a national conference on Strategic Management for Today’s Business organised by Siddhant College of Engineering, Pune, ISBN 978-81- 8465-811-8.  Vinaydeep and Atul Kumar, “Talent Management for Effective Customer Relationship Management” in proceedings of a national conference on Strategic Management for Today’s Business organised by Siddhant College of Engineering, ISBN 978-81-8465-811-8.  Vinaydeep and Atul Kumar, “Talent Management for Effective Customer Relationship Management” in proceedings of a national conference on Strategic Management for Today’s Business organised by Siddhant College of Engineering, ISBN 978-81-8465-811-8.
https://openalex.org/W2045944541	http://seer.tupa.unesp.br/index.php/BIOENG/article/download/42/43	English	null	COD REMOVAL OF PARBOILIZED RICE WASTEWATER BY CYANOBACTERIA APHANOTHECE MICROSCOPICA NÄGELI	Revista Brasileira de Engenharia de Biossistemas	2,009	cc-by	2,789	COD REMOVAL OF PARBOILIZED RICE WASTEWATER BY CYANOBACTERIA APHANOTHECE MICROSCOPICA NÄGELI REMOÇÃO DE DQO DO EFLUENTE DA PARBOILIZAÇÃO DO ARROZ PELA CIANOBACTÉRIA APHANOTHECE MICROSCOPICA NÄGELI BASTOS, R.G.1 QUEIROZ, M.I.2 ZEPKA, L.Q.3 VOLPATO, G.3 GARCIA, M.S.3 JACOB-LOPES, E.3 245 245 BioEng, Campinas, v.3 n.3, p.245-250, Set/Dez., 2009 1 Center of Agricultural Sciences (CCA) - Federal University of São Carlos (UFSCar), Araras, SP- Brazil. e-mail: reinaldo@cca.ufscar.br 2 Chemical Department - University of Rio Grande (FURG) – Rio Grande, RS. Brazil. 3 Food Engineering – University of Rio Grande (FURG) – Rio Grande, RS. Brazil. Keywords: Cyanobacteria, Aphanothece, Organic matter removal, Parboilized rice effluent. 3 Food Engineering – University of Rio Grande (FURG) – Rio Grande, RS. Brazil. 2 Chemical Department - University of Rio Grande (FURG) – Rio Grande, RS. Brazil. ABSTRACT Cyanobacteria are photosynthetic prokaryotes organisms used on single-cell protein production and wastewater treatment for nitrogen and phosphorus removal. Moreover, some strains have the particular characteristic of growing in the dark on simple molecules such as acetate, glucose and organic acids, and consequently reduction of Chemical Oxygen Demand (COD) from nutrient medium. Aphanothece microscopica Nägeli is a cyanobacteria frequently found in south of Brazil, which has been studied with respect to residues valorization in the single-cell protein production. Wastewater from rice parboilization process presents characteristics that suggest the removal of nutrients and organic matter by incorporation into a biomass. The aim of this paper is to extend the findings of the previous studies about heterotrophic metabolism application of these cyanobacteria on wastewater treatment, considering influence of temperature and inoculum amount. Experiments indicated that Aphanothece microscopica Nägeli presented high potential of organic matter removal from parboilized rice effluent by 300 mgL-1 inoculum at temperature range 25-35oC. Sciences (CCA) - Federal University of São Carlos (UFSCar), Araras, SP- Brazil. e-mail: reinaldo@cca.ufscar.br Center of Agricultural Sciences (CCA) - Federal University of São Carlos (UFSCar), Araras, SP- Brazil. e-mail: re RESUMO Cianobactérias são microrganismos fotossintetizantes usados na produção de proteínas unicelulares e na remoção de nitrogênio e fósforo de águas residuárias. Algumas linhagens apresentam a característica particular de crescerem no escuro a partir de moléculas orgânicas simples como acetato, glicose e ácidos orgânicos, com conseqüente redução da DQO. Aphanothece microscopica Nägeli é uma cianobactéria de grande ocorrência no sul do Brasil, tendo sido estudada quanto a valoração de resíduos agroindustriais e na produção de proteínas unicelulares. O efluente da parboilização do arroz apresenta características que sugerem a remoção de nutrientes e matéria orgânica por incorporação em biomassa microbiana. Assim, o objetivo deste trabalho é prosseguir os estudos a respeito da aplicação do metabolismo heterotrófico desta cianobactéria no tratamento de águas residuárias, considerando a influência da temperatura e inoculo. Os resultados indicaram alto potencial de remoção de DQO do efluente da parboilização do arroz com 300mg/L de inoculo na faixa de 25 a 35ºC. Palavras-chaves: Cianobactéria, Aphanothece, Remoção de matéria orgânica, Efluente da parboilização do arroz. Keywords: Cyanobacteria, Aphanothece, Organic matter removal, Parboilized rice effluent. 246 BASTOS, R.G.,QUEIROZ, M.I.,ZEPKA, L.Q.VOLPATO, G,GARCIA, M.S.,JACOB-LOPES, E. Palavras-chaves: Cianobactéria, Aphanothece, Remoção de matéria orgânica, Efluente da parboilização do arroz. INTRODUCTION The use of cyanobacteria cultures in waste- water treatment is suggested since many years for nitrogen and phosphorus removal from industrial and urban effluents (ARDELEAN & ZARNEA, 1998; ASLAN & KAPDAN, 2006; LINCOLN et al. 1996; NUÑEZ et al. 2001; QUEIROZ et al. 2007; TAM & WONG, 1996). Microalgae are prokaryotic, as green algae (Chlorophyta), or eukaryotic photo- synthetic microorganisms, as cyanobacteria (Cy- anophyceae) (MATA et al. 2010). Cyanobacteria or blue-green algae are distinguished from other prokaryotes by their generally low rates of endog- enous respiration and by their limited ability to uti- lize organic substances as a source of carbon and energy (FAY, 1983). These microorganisms were always regarded to be typical photosynthetic in which the light-dependent fixation of CO2 is the dominant mode of nutrition. This concept, however, was obscured by observation that in nature blue- greens are most abundant in habitats rich in or- ganic matter. Thus, some strains of cyanobacterial are capable of using simple organic molecules (glu- cose and organic acid) as a source of carbon and energy for continuous slow growth in dark. Disac- charides could be converted readily to the compo- nent sugars in the heterocysts, degraded along the glycolytic pathway, and oxidized through the hexose monophosphate shunt (oxidative pentose phosphate pathway), providing energy and reduc- tant for N2 fixation. There is a great deal of evi- dence that the catabolic pathways remain opera- tional after heterocyst differentiation (FAY, 1992). This heterotrophic potential could be used in biotreatment of agro-industrial wastewaters for si- multaneous organic matter and nitrogen removal. BICH et al. (1999) indicated the possibility of us- ing microalgae in a treatment system for the nitro- gen removal from an agro-industrial wastewater with Chemical Oxygen Demand (COD) about 500 mg/L. BASHAN & BASHAN (2004) reported the application of Chlorella vulgaris in the biological treatment of recalcitrant anaerobic industrial efflu- ent, mainly for phosphorus removal. Aphanothece microscopica Nägeli is a cyanobacterium that has been studied with respect to residues valorization in the single-cell protein production, due to high specific growth rate (0.24h-1) on wastewater (QUEIROZ et al. 2002). QUEIROZ et al. (2007) evaluated nitrogen and organic matter consump- tion from parboiled rice effluent by with 100mg/L of Aphanothece inoculum and 20oC on stirred-batch reactor. Results indicated that maximum efficiency COD removal occurred at 15 hours of cultivation, approximately 83.4%, suggesting the consumption of organic matter components on dark. Organic matter removal Removal of matter organic was evaluated through soluble COD consumption, express in per- centage (CONTRERAS et al. 2000). COD con- sumption rates were calculated through slope of COD profile by differential approach (LEVENSPIEL, 2000). Average specific rates were calculated by ratio of COD consumption rate and biomass con- centration during the experiment, in terms of COD per cyanobacterial concentration per time (CONTRERAS et al. 2000). MATERIAL AND METHODS MATERIAL AND METHODS inoculum was centrifuged at 3000xg for 15minutes and cell concentration was measured by gravimetric method, through filtration of a given volume of culture medium in a 0.45 µm filter and drying at 105°C for 24 hours (APHA 1998). Experiemtal reactor and sampling A stirred-batch reactor was filled with 4L of the effluent with 100 and 300mg/L of cyanobacterial concentration. Initial pH of effluent must be adjusted to 7.6 for Aphanothece’ growth (QUEIROZ & KOETZ, 1997). The cyanobacteria were cultivated without light source on 25 and 35°C, with constant aeration of 1VVM during all experimental period. Samples for cell concentration and wastewater assay were collected every 3 hour during 24 hours. RESULTS AND DISCUSSION Table 1 presents characterization of waste- water from rice parboilization processing. All pa- rameters showed the high variation coefficient (%), except pH (10.7%), due mainly to variations on the process and type of rice grains. Nowadays, rice is one of the more important cereals in the world, being the basic constituent of the Brazilian diet. Brazil is among the 10 largest rice produc- ers, and parboiling one of the most important im- provement processes. As this process involves sig- nificant water amount (average 4m3 per ton of rice processed grain), resulting in a considerable wastewater amount, with characteristics that sug- gest the possibility of biotreatment for the incorpo- ration of these nutrients into the biomass (QUEIROZ & KOETZ, 1997). Wastewater characterization Some samples of effluent were collected di- rectly from tanks of rice parboilization processing at six different days. Wastewater was character- ized through pH, Kjeldahl Total Nitrogen (KTN), Chemical Oxygen Demand (COD) and Total Vola- tile Acids (TVA), according to APHA, (1998). Total Sugars (TS) were obtaining for Felhing Method (AOAC, 1995). Data were analyzing by variation coefficient in percentage, i.e., average per stan- dard deviation. INTRODUCTION dence that the catabolic pathways remain opera- tional after heterocyst differentiation (FAY, 1992). This heterotrophic potential could be used in biotreatment of agro-industrial wastewaters for si- multaneous organic matter and nitrogen removal. BICH et al. (1999) indicated the possibility of us- ing microalgae in a treatment system for the nitro- gen removal from an agro-industrial wastewater with Chemical Oxygen Demand (COD) about 500 mg/L. BASHAN & BASHAN (2004) reported the application of Chlorella vulgaris in the biological treatment of recalcitrant anaerobic industrial efflu- ent, mainly for phosphorus removal. Aphanothece microscopica Nägeli is a cyanobacterium that has been studied with respect to residues valorization in the single-cell protein production, due to high specific growth rate (0.24h-1) on wastewater (QUEIROZ et al. 2002). QUEIROZ et al. (2007) evaluated nitrogen and organic matter consump- tion from parboiled rice effluent by with 100mg/L of Aphanothece inoculum and 20oC on stirred-batch reactor. Results indicated that maximum efficiency COD removal occurred at 15 hours of cultivation, approximately 83.4%, suggesting the consumption of organic matter components on dark. These data motivated new researches to extend the findings of the previous studies consid- ering influence of temperature and inoculum amount, according to improve the wastewater treat- ment by cyanobacterial. BioEng, Campinas, v.3 n.3, p.245-250, Set/Dez., 2009 247 Inoculum Stock suspensions of cyanobacterium Aphanothece microscopica Nägeli were maintained in complete BGN medium (Braun– Grunow Medium), according to RIPPKA et al. (1979), for at least 48 hours, light period of 12 hours at 30°C (QUEIROZ et al. 2002). Log phase TABLE 1 - Characterization of wastewater from rice parbolization processing (units in mg/L, except pH) pH COD KTN TVA TS 4,5 2821 45 924 919 4,2 2578 49 975 722 4,5 3286 93 624 1154 4,3 4514 88 1275 2085 4,3 6480 79 780 2554 5,5 5022 70 1290 2032 Average 4,6 4177 71 978 1578 Variation (%) 10,7 33,3 28,6 24,8 47,1 COD: chemical demand; KTN: Kjeldahl total nitrogen; TVA: total volatile acid; TS: total sugar TABLE 1 - Characterization of wastewater from rice parbolization processing (units in mg/L, except pH) 248 BASTOS, R.G.,QUEIROZ, M.I.,ZEPKA, L.Q.VOLPATO, G,GARCIA, M.S.,JACOB-LOPES, E. COD average for wastewaters samples was 4117 mg/L, with 33.3% of variation, occurred due to variations on maceration time, temperature and type of grain (QUEIROZ & KOETZ, 1997). cyanobacterial and microalgae could be used as simple molecules as glucose and acids organic or acetate for growth heterotrophic on dark cultivation (FAY, 1983; TAM & WONG, 1996). Figure 1 presents soluble COD profile in dif- ferent experimental conditions of inoculum and tem- perature. It can be observed in high consumption with 300mg/L inoculum for 25 and 35oC, which sug- gests that the extent of biotreatment of wastewa- ters are proportional to the amount of microorgan- ism (LINCOLN et al. 1996). According to CONTRERAS et al. (2000), the effluents generated for food industries present variable discharges concentrations of COD, solids and nitrogen on different forms. Volatile acids and total sugars are represented 62% of COD, calculated with parameters express in Table. 1. It is very important because some strains of FIGURE 1 - COD soluble consumption from rice parboilized effluent by Aphanothece microscopica Nägeli at different temperatures 25oC ( ) and 35oC (ο) with inoculums 100mg/L (a) and 300mg/L (b) FIGURE 1 - COD soluble consumption from rice parboilized effluent by Aphanothece microscopica Nägeli at different temperatures 25oC ( ) and 35oC (ο) with inoculums 100mg/L (a) and 300mg/L (b) effluent at 20oC and inoculum of 100mg/L, about 0.08h-1. Moreover, the rates are similar to obtained by BELTRAN-HEREDIA et al. CONCLUSIONS BASHAN, L.; BASHAN, Y.Recent advances in re- moving phosphorus from wastewater and its fu- ture use as fertilizer (1997–2003). Water Research - Elsevier n. 38, p.4222–4246, 2004. Based on the experimental data, maximum COD removal by Aphanothece microscopica Nägeli suggest the application of this cyanobacterium on biotreatment of wastewater from rice parboilization processing for inoculus of 300 mgL-1 at temperature range 25-35oC. BELTRAN-HEREDIA, J.; TORREGROSA, J.; DOMINGUEZ, J.; GARCIA, J. Aerobic biological treatment ofblack table olive washing wastewaters: effect of an ozonation stage. Process Biochemis- try , Elsevier, n. 35, p.1183-1190. 2000. BELTRAN-HEREDIA, J.; TORREGROSA, J.; DOMINGUEZ, J.; GARCIA, J. Aerobic biological treatment ofblack table olive washing wastewaters: effect of an ozonation stage. Process Biochemis- try , Elsevier, n. 35, p.1183-1190. 2000. ACKNOWLEDGMENTS BICH, N.; YAZIZ, M.; KAKTI, N. Combination of Chlorella vulgaris and Eichhornia crassipes for wastewater nitrogen removal. Water Research – Elsevier, n. 33: 2357–2362. 1999. This work was supported by FAPERGS, CNPq and CAPES, Brazil. Inoculum (2000) in your study of oxidative degradation of the organic matter present in the washing waters from the black table olive industry by activated sludge, i.e., conventional processes of wastewater treatment. Thus, Aphanothece microscopica Nägeli presented het- erotrophic metabolism and it could be used in stirred tank-reactors on dark to COD removal in wastewater from rice parboilization processing in the experimental conditions tested on this paper. COD consumption rate and efficiency of removal are express in Table 2. High specific COD consumption rates (0.25mg/mg.h) suggest high or- ganic matter consumption with 300mg/L of inocu- lum, reflecting the profiles presented in Fig. 1, with practically COD depletion at 15 hours for 25 and 35oC. The results are important for possible appli- cation this cyanobacterium on rice wastewater with stirred tank in the temperature range, typical of subtropical climate. Specific rates found in this pa- per was higher than obtained by QUEIROZ et al. (2007) working Aphanothece in rice parboiled rice BioEng, Campinas, v.3 n.3, p.245-250, Set/Dez., 2009 249 TABLE 2 - Comparison of average specific consumption rates and COD removal for inoculum and temperature experimental conditions TABLE 2 - Comparison of average specific consumption rates and COD removal for inoculum and temperature experimental conditions Temperature Inoculum Specific COD consumption rate Maximum COD removal and time [oC] [mg/l] [mg COD/mg biomass.h] [%] 25 100 -0,046 66,7 at 24 hours 25 300 -0,250 97,7 at 15 hours 35 100 -1,160 84,8 at 12 hours 35 300 -0,250 97,7 at 15 hours REFERENCES CONTRERAS, E.; GIANNUZZI, L.; ZARITZKY, N. Growth kinetics of the filamentous microorganism Sphaerotilus natans in a model system of a food industry wastewater. Water Research – Elsevier, n 34, p. 4455-4463, 2000. AMERICAN PUBLIC HEALTH ASSOCIATION. Standard methods for the examination of water and wastewater. 20.ed. Washington, DC. - USA : APHA, 1998. AMERICAN PUBLIC HEALTH ASSOCIATION. Standard methods for the examination of water and wastewater. 20.ed. Washington, DC. - USA : APHA, 1998. ARDELEAN I.; ZARNEA, G. Photosynthesis respi- ration Interplay in cyanobacteria: fundamentals and aplication. In: SUBRAMANIAN G.; KAUSHIK B. VENKATARAMAN G. Cyanobacterial biotechnol- ogy. USA: Science Publishers, 1998. p.103-107. FAY, P. The blue-greens (Cyanophyta- cyanobacteria), 5.ed. London: Edward Arnold, 1983. FAY, P. Oxygen relations of nitrogen fixation in cyanobacteria. Clinical Microbiology Reviews, Washington, D.C., n.56, p. 340-373. 1992. ASLAN S.; KAPDAN, I. Batch kinetics of nitrogen and phosphorus removal from synsthetic wastewater by algae. Ecological Engineering, Elsevier, n.28 p. 64-70, 2006. LEVENSPIEL, O. Engenharia das reações químicas. São Paulo: Edgard Blücher 2000. LINCOLN, E.; WILKIE, A.; FRENCH, B. Cyanobacterial process for renovating dairy waste- water. Biomass & Bioenergy , Elsevier, n. 10 p. 63-68,1996. LEVENSPIEL, O. Engenharia das reações químicas. São Paulo: Edgard Blücher 2000. q g LINCOLN, E.; WILKIE, A.; FRENCH, B. Cyanobacterial process for renovating dairy waste- water. Biomass & Bioenergy , Elsevier, n. 10 p. 63-68,1996. ASSOCIATION OF OFFICIAL ANALYTICAL CHESMISTER – AOAC. Official methods of analy- sis. Washington., D.C.: AOAC, 1995. ASSOCIATION OF OFFICIAL ANALYTICAL CHESMISTER – AOAC. Official methods of analy- sis. Washington., D.C.: AOAC, 1995. BioEng, Campinas, v.3 n.3, p.245-250, Set/Dez., 2009 250 BASTOS, R.G.,QUEIROZ, M.I.,ZEPKA, L.Q.VOLPATO, G,GARCIA, M.S.,JACOB-LOPES, E. MATA, T. M.; MARTINS, A. A.; CAETANO, N. S. Microalgae for biodiesel production and others applications: a review. Renewable and Sustain- able Energy Reviews, Elsevier, n.14, p.212-232, 2010. NUÑEZ, V.; VOLTOLINA, D.; NIEVES, M.; PIÑA, P. MEDINA A, GUERRERO M Nitrogen budget in Scenedesmus obliquus cultures with artificial wastewater. Bioresource Technology , Elsevier, n.78 p.161-164, 2001. QUEIROZ, M.; KOETZ, P.R. Caracterização do efluente da parbolização do arroz. Revista Brasileira de Agrociência , Pelotas, RS, n. 3 p. 139-143, 1997. QUEIROZ, M.I.; BASTOS, R.; BENERI, R. ALMEIDA, R. Evaluación del crecimiento de la Aphanothece microscopica Nägeli en las aguas residuales de la parbolización del arroz. Revista Internacional Información Tecnológica, Chile, n.13 p. 61-65, 2002. QUEIROZ, M.I.; JACOB-LOPES, E.; ZEPKA, L.Q.; BASTOS, R.G.; GOLDBECK, R. BioEng, Campinas, v.3 n.3, p.245-250, Set/Dez., 2009 REFERENCES The kinetics of the removal of nitrogen and organic matter from parboiled rice effluent by cyanobacteria in a stirred batch reactor. Bioresource Technology, Elsevier, n.98, p 2163-2169, 2007. RIPPKA, R.; DERUELLES, J.; WATERBURY, J.; HERDMAN, M.; STANIER, R. Generic assign- ments strain histories and properties of pure cul- tures of cyanobacteria. Journal of General Micro- biology, New Jersey, n. 111, p. 1-61, 1979. SUBRAMANIAN G.; KAUSHIK B. VENKATARAMAN G. Cyanobacterial biotechnol- ogy. USA: Science Publishers, 1998. p. 103-107 TAM, N.; WONG, Y. Effect of ammonia concentra- tion on growth of Chlorella vulgaris and nitrogen removal from media. Bioresource Technology, Elsevier, n. 57, p. 45-50, 1996. BioEng, Campinas, v.3 n.3, p.245-250, Set/Dez., 2009
https://openalex.org/W2995361249	https://researchonline.ljmu.ac.uk/id/eprint/12517/1/Behavioural%20responses%20in%20a%20congested%20sea%20an%20observational%20study%20on%20a%20coastal%20nest%20guarding%20fish.pdf	English	null	Behavioural responses in a congested sea: an observational study on a coastal nest-guarding fish	The European zoological journal	2,019	cc-by	12,129	Mascolino, S, Mariani, S and Benvenuto, C Behavioural responses in a congested sea: an observational study on a coastal nest-guarding fish http://researchonline.ljmu.ac.uk/id/eprint/12517/ Article LJMU has developed LJMU Research Online for users to access the research output of the University more effectively. Copyright © and Moral Rights for the papers on this site are retained by the individual authors and/or other copyright owners. Users may download and/or print one copy of any article(s) in LJMU Research Online to facilitate their private study or for non-commercial research. You may not engage in further distribution of the material or use it for any profit-making activities or any commercial gain. The version presented here may differ from the published version or from the version of the record. Please see the repository URL above for details on accessing the published version and note that Citation (please note it is advisable to refer to the publisher’s version if you intend to cite from this work) Mascolino, S, Mariani, S and Benvenuto, C (2019) Behavioural responses in a congested sea: an observational study on a coastal nest-guarding fish. European Zoological Journal, 86 (1). pp. 504-518. ISSN 2475-0263 LJMU Research Online Mascolino, S, Mariani, S and Benvenuto, C Behavioural responses in a congested sea: an observational study on a coastal nest-guarding fish http://researchonline.ljmu.ac.uk/id/eprint/12517/ Article LJMU has developed LJMU Research Online for users to access the research output of the University more effectively. Copyright © and Moral Rights for the papers on this site are retained by the individual authors and/or other copyright owners. Users may download and/or print one copy of any article(s) in LJMU Research Online to facilitate their private study or for non-commercial research. You may not engage in further distribution of the material or use it for any profit-making activities or any commercial gain. The version presented here may differ from the published version or from the version of the record. Please see the repository URL above for details on accessing the published version and note that Citation (please note it is advisable to refer to the publisher’s version if you intend to cite from this work) Mascolino, S, Mariani, S and Benvenuto, C (2019) Behavioural responses in a congested sea: an observational study on a coastal nest-guarding fish. European Zoological Journal, 86 (1). pp. 504-518. ISSN 2475-0263 LJMU Research Online LJMU Research Online LJMU Research Online Mascolino, S, Mariani, S and Benvenuto, C Article Citation (please note it is advisable to refer to the publisher’s version if you intend to cite from this work) Mascolino, S, Mariani, S and Benvenuto, C (2019) Behavioural responses in a congested sea: an observational study on a coastal nest-guarding fish. European Zoological Journal, 86 (1). pp. 504-518. ISSN 2475-0263 LJMU has developed LJMU Research Online for users to access the research output of the University more effectively. Copyright © and Moral Rights for the papers on this site are retained by the individual authors and/or other copyright owners. Users may download and/or print one copy of any article(s) in LJMU Research Online to facilitate their private study or for non-commercial research. You may not engage in further distribution of the material or use it for any profit-making activities or any commercial gain. The version presented here may differ from the published version or from the version of the record. Please see the repository URL above for details on accessing the published version and note that access may require a subscription. For more information please contact researchonline@ljmu.ac.uk http://researchonline.ljmu.ac.uk/ Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=tizo21 The European Zoological Journal ISSN: (Print) 2475-0263 (Online) Journal homepage: https://www.tandfonline.com/loi/tizo21 S. Mascolino, S. Mariani & C. Benvenuto To cite this article: S. Mascolino, S. Mariani & C. Benvenuto (2019) Behavioural responses in a congested sea: an observational study on a coastal nest-guarding fish, The European Zoological Journal, 86:1, 504-518, DOI: 10.1080/24750263.2019.1699611 Abstract Abstract The deleterious effects of anthropogenic noise on animal communication are nowadays recognised, not only in urban environ- ments but also in terrestrial habitats and along coasts and in open waters. Yet, the assessment of short- and long-term exposure consequences of anthropogenic noise in marine organisms remains challenging, especially in ﬁsh and invertebrates. Males of the Mediterranean damselﬁsh Chromis chromis vocalise and perform visual displays (multimodal communication) to attract mates. The frequency-range of courtship vocalisations overlaps with low-frequency noise generated by maritime activities, resulting in a reduced detection distance among conspeciﬁcs. We quantiﬁed the number of courtship-related visual displays performed by males living in areas with different levels of maritime trafﬁc. We also tried to manipulate ambient noise in the ﬁeld to test male short-term response to increased noise levels. Males living in busier areas (near to a harbour) performed signiﬁcantly more visual displays than those living in less congested areas. When exposed to artiﬁcially-increased ambient noise level (playback of boat noise), males did not adjust the number of visual displays accordingly. Yet, we note how assessing the actual effect of maritime trafﬁc in marine populations in their natural environments is particularly difﬁcult, as the effects of boat noise cannot be easily disentangled from a variety of other intrinsic or environmental factors, discussed in the paper. We thus present suggestions to obtain more robust analyses of variations of courtship behaviours in territorial ﬁshes. We hope this will facilitate a further understanding of the potential long-term effects of anthropogenic noise, whose analyses should be prioritised in the context of environmental impact assessment, resource management and biodiversity conservation. Keywords: Anthropogenic noise, Chromis chromis, mating behavioural plasticity, visual display, aco To link to this article: https://doi.org/10.1080/24750263.2019.1699611 To link to this article: https://doi.org/10.1080/24750263.2019.1699611 © 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. View supplementary material Published online: 17 Dec 2019. Submit your article to this journal Article views: 107 View related articles View Crossmark data Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=tizo21 The European Zoological Journal, 2019, 504–518 Vol. 86, No. 1, https://doi.org/10.1080/24750263.2019.1699611 © 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Behavioural responses in a congested sea: an observational study on a coastal nest-guarding ﬁsh S. MASCOLINO1, S. MARIANI 1,2,3, & C. BENVENUTO 3* S. MASCOLINO1, S. MARIANI 1,2,3, & C. BENVENUTO 3* 1UCD School of Biology and Environmental Science, University College Dublin, Dublin, Ireland, 2School of Biological & Environmental Sciences, Liverpool John Moores University, Liverpool, UK, and 3School of Science, Engineering and Environment, University of Salford, Salford, UK 1UCD School of Biology and Environmental Science, University College Dublin, Dublin, Ireland, 2School of Biological & Environmental Sciences, Liverpool John Moores University, Liverpool, UK, and 3School of Science, Engineering and Environment, University of Salford, Salford, UK 1UCD School of Biology and Environmental Science, University College Dublin, Dublin, Ireland, 2School of Biological & Environmental Sciences, Liverpool John Moores University, Liverpool, UK, and 3School of Science, Engineering and Environment, University of Salford, Salford, UK (Received 31 May 2019; accepted 22 November 2019) Correspondence: C. Benvenuto, School of Science, Engineering and Environment, University of Salford, Salford, M5 4WT, UK. Tel: 0044(0)1612955141. Email: c.benvenuto@salford.ac.uk Introduction 2010; Parmentier et al. 2010; Fakan & McCormick 2019) and in the Mediterranean (Picciulin et al. 2002, 2010; Codarin et al. 2009; Bracciali et al. 2012). Damselﬁshes are an abundant component of coral and rocky reefs, where they play a key role in the trophic functioning of the ecosystem (Bracciali et al. 2012; Pinnegar 2018). Chromis chromis Linnaeus, 1758, is the only native pomacentrid in the Mediterranean Sea (Allen 1991); it is a small damselﬁsh commonly found in shoals near rocky reefs or above seagrass meadows at depths between 3 and 30 m (Quignard & Pras 1986). Mediterranean damselﬁsh are characterised by loca- lised spawning areas and habitat use. Their life history traits include the production of benthic eggs and short- lived pelagic larvae (18–19 days; Raventós & Macpherson 2001). These features and the marked territoriality of adults are expected to favour population isolation and geographic genetic structure (Domingues et al. 2005). During the summer, males colonise nests and guard them until the end of the reproductive sea- son. They court females by vocalising and performing speciﬁc behavioural patterns, i.e., signal jumps (“the male swims rapidly upwards for 1–3 m before rotating downwards and returning to the nest”; Laglbauer et al. 2017) as visual displays, in order to receive egg deposi- tion (Picciulin et al. 2002, 2010 and references therein). The frequency range of their vocalisations overlaps with low frequency noise generated by boats and ships, resulting in a marked reduction (up to 100 times) of the detection distance among conspeciﬁcs (Mann & Lobel 1997; Codarin et al. 2009; Wysocki et al. 2009). This study aimed to investigate variations in the courtship strategies of the Mediterranean damselﬁsh In light of the results obtained from studies per- formed in terrestrial environments (e.g., Barber et al. 2009; Francis & Barber 2013), researchers suggest that similar effects should be expected in aquatic envir- onments, especially with regards to ﬁshes (Dooling et al. 2015). However, caution should be taken, given the differences existing between terrestrial and aquatic media in terms of physical properties (McGregor et al. 2013) and the different adaptations to sound detection and acoustic communication in different taxa (Wong 2014). Even when sound pollu- tion affects ﬁsh communication, the main question is “how much does it matter” (Brumm 2014)? In other words: can ﬁsh respond in a phenotypically plastic way to noise stressors? Introduction focus is broadening from an initial concern targeted mainly to marine mammals (Weilgart 2007; Southall et al. 2008; Ellison et al. 2011; Miller et al. 2015), investigations on acoustic behaviours of aquatic organ- isms in general and the detrimental effects of man- made noise in particular remain challenging in marine and freshwater ecosystems (Lobel et al. 2010; Bolgan et al. 2016; Mickle & Higgs 2017; Hasan et al. 2018; Linke et al. 2018). Studies on the effects of anthropo- genic noise on invertebrates are scant (Wale et al. 2013; Williams et al. 2015; Aguilar de Soto 2016; André et al. 2016; Tidau & Briffa 2016) and more research is also needed on ﬁsh (see the recent review by Popper & Hawkins 2019), especially on wild populations in nat- ural habitat, which is extremely difﬁcult (Slabbekoorn 2016). Finally, as in many behavioural responses Coastal and offshore human activities often lead to local increments of ambient noise levels, polluting the marine soundscape (Slabbekoorn et al. 2010; Radford et al. 2014; Whitﬁeld & Becker 2014; Dooling et al. 2015; Hawkins & Popper 2018; Putland et al. 2018). To better understand the short and long-term effects of marine acoustic pollution, not only at the individual but also at the population and ecosystem level (Kunc et al. 2016; Nabe-Nielsen et al. 2018), more under- water acoustic ecological studies should be performed (Slabbekoorn et al. 2010; Brumm 2014; Radford et al. 2014; Slabbekoorn 2016). Indeed, even though these topics are gaining interest and conservation relevance (Williams et al. 2015; Aguilar de Soto et al. 2016; Greggor et al. 2016; Gordon et al. 2018), and the Behavioural responses in a congested sea 505 (Siegenthaler et al. 2018), a certain level of plasticity to address anthropogenic noise is expected not only at the interspeciﬁc and interpopulation level, but also at the intrapopulation level based on inter and intra- individual variability and time of exposure (Radford et al. 2015; Harding et al. 2018). mates: indeed, a very important context during which teleosts vocalise is courtship (e.g., Amorim & Neves 2007; Parmentier et al. 2010; Amorim et al. 2012; Maruska et al. 2012; Casaretto et al. 2015). Given its vocal behaviours, the family Pomacentridae (damselﬁshes) has been an interesting study system for acoustic studies (Myrberg & Spires 1972; Mann & Lobel 1997; Amorim 2006; Kasumyan 2008; Ladich 2013; Weimann et al. 2018), in coral reefs (Lobel et al. Introduction How and how well can ﬁsh adjust their communication in response to anthropogenic impacts? Can short-term effects, at the individual level, impact population dynamics in the long term? p p p y g In the water, anthropogenic noise can be generated intentionally, for example to prevent harmful interac- tions between marine mammals and ﬁshery activities (Waples et al. 2013) or during seismic studies (Popper et al. 2005) and sonar applications (Hildebrand 2009). More often, noise is the by-product of various activities, including wind farm plants (Wahlberg & Westerberg 2005), underwater explosions, pile- driving operations (Haelters et al. 2013) or, most commonly, maritime trafﬁc (Popper & Hastings 2009). Noise generated by ships and boats mainly falls in the lower frequency range of the acoustic spec- trum (1–1000 Hz) which is also the range at which all ﬁsh studied to date are able to hear (<50 to 1000–1500 Hz; Popper 2003; Popper & Fay 2011; Radford et al. 2014; Popper & Hawkins 2019) even though some ﬁsh reach 4000–8000 Hz (Lobel et al. 2010; Ladich & Fay 2013). This study aimed to investigate variations in the courtship strategies of the Mediterranean damselﬁsh in three different areas in Sicily, Italy. Literature sug- gests that the high levels of ambient noise (sensu Hildebrand 2009: “the sound ﬁeld against which sig- nals must be detected”) caused by navigation make it more difﬁcult for females to hear male vocalisations (Codarin et al. 2009). Thus, we predicted that males living in a more congested (thus possibly noisier) envir- onment would on average perform more visual displays (VD; signal jumps) to overcome the loss of efﬁciency of vocal cues, adjusting from a multimodal acoustic and visual communication to a mainly visual display (Partan 2017). To explore the role of short-term Fish use sounds (production and/or detection) in multiple ways (Bass & Ladich 2008; Ladich 2013; Popper & Hawkins 2019): speciﬁc sounds can be used during migrations and as directional orientation clues for larvae to settle (Tolimieri et al. 2004) but mostly sounds are received and produced to acquire and transmit information. Many ﬁsh species actively produce sounds to communicate the presence of pre- dators or food to their conspeciﬁcs (Bessey & Heithaus 2013), during aggressive behaviours (Millot et al. 2011), to defend their territories (Myrberg 1997; Pereira et al. 2014) and to attract S. Mascolino et al. Introduction 506 behavioural plasticity in accounting for the patterns observed, we also tested whether a relative change in courtship strategy (visual displays vs vocal cues) would occur after a few hours of increased ambient noise level in populations not normally exposed to high levels of boat/ship noise. This observational study performed on wild individuals in their natural habitat provides initial evidence on the evolutionary adaptation and beha- vioural ﬂexibility of ﬁsh communication in different areas. A more robust analysis should be performed with acoustic recordings of the vocalizations and hydrophones able to record sound pressure levels com- bined with particle velocity levels as well as sound exposure levels (Spiga et al. 2012). We thus highlight the challenges for pure behavioural ecologists and evo- lutionary biologists to master fast-evolving technologies and suggest a stronger collaboration from the planning stage of projects with acousticians for a more effective characterization of the soundscape and recording of animal vocalizations (see also Popper & Hawkins 2019). ﬁrst area (A) was located just outside the main com- mercial harbour of Palermo, the ﬁfth largest city in Italy. The intense maritime trafﬁc was due to commer- cial ships, cruise ships, as well as smaller boats used for various purposes around the dock areas (see http:// www.marinetrafﬁc.com/it/ais/details/ports/635/Italy_ port:PALERMO). Two smaller docks were also pre- sent in proximity of the main harbour, one hosting several sailing and motor boats (also affecting ambient noise levels; Haviland-Howell et al. 2007) and the other used as a shelter for small ﬁshing boats. The second area, Zingaro (B) was located between two small harbours (the distance between the sam- pling area and each of the two harbours was approxi- mately 10 km) mostly accessed by recreational boats. The third area (C) was located in a zone where boat trafﬁc is prohibited (Vendicari Nature Reserve). Even though long-term records were not registered, due to logistics, the trafﬁc data should represent the average situation for each area (but we are aware of changes on short term – e.g., weekends vs week days, days vs night). “Marine roads” (i.e., shipping routes; Pirotta et al. 2019) and maritime trafﬁc can be used as a “quick and dirty” metric to identify more acousti- cally impacted areas. Introduction Having classiﬁed the three areas on the basis of trafﬁc data (Figure 1), background/ ambient noise was measured during each behavioural recording session (see details below) using an uncali- brated hydrophone (H2a-XLR, operational frequency range 10 Hz −100 KHz, sensitivity −180dB re: 1V/ µPa, Aquarian Audio Products, Washington, USA) Experimental set-up An ethogram was prepared using the detailed descrip- tion of C. chromis behaviour by Verginella et al. (1999), grouping all described behaviours in three main cate- gories: courtship related visual displays (VD; Figure 2); mating (MT: time spent by a pair, male and female, in the nest while spawning); non courtship related beha- viours (NC: all other behaviours, ranging from cleaning the nest to feeding). The time spent by the focal male outside of the camera view (OV) was also taken into account. From the original recording session (approxi- mately 90 min long) the initial and ﬁnal parts were removed to minimise the possible disturbance created by the diver accessing the cameras and the scoring was focused on the central 75 min of the recordings. Each video was scored in full length to determine the number of VD performed and the amount of time spent in MT, NC and OV. Then, each focal male was assigned to either the “parental care” (number of VD and MT equal to zero) or “courtship” (number of VD and/or MT greater than zero) groups. The latter group was the focus of the analyses. Because of the slightly asynchro- nous reproductive cycles among individuals within the same colony, the majority of the males were in parental care, so the available data on courtship behaviours was reduced. Videos were scored using the software JWatcher 1.0 (Blumstein & Daniel 2007). For standar- disation purposes, the number of visual displays per hour (VD h−1) was quantiﬁed considering only the time spent by focal males in sight of the camera view (i.e., VD * (total time in min – out of site in min)−1 * 60). Two types of experiments (one observational and one manipulative) were performed to assess the difference in courtship strategies performed by nesting males in different areas. In the ﬁrst phase, data were collected in the three selected sites (A, B, C). Subsequently, a manipulative ﬁeld experiment was conducted to assess whether nesting males from the natural reserve of Vendicari (C) were able to adjust their courtship strategy when the ambient noise level was artiﬁcially increased with playback of boat noise (D). Behavioural observations - The three study sites (5–12 m from the coast) were reached by boat and the nesting areas (7–12 m deep) were identiﬁed by scuba diving. Study areas Behavioural observations were performed along the Sicilian coast, Italy, in summer 2011 and 2012. Three areas characterised by different maritime trafﬁc (used as proxy of possible different ambient noise levels) were identiﬁed as study sites (Figure 1). The Figure 1. Global shipping density map of Sicily, Italy as recorded by Maritime Trafﬁc (https://www.marinetrafﬁc.com/). Trafﬁc density is colour coded, moving from low to high trafﬁc (from blue to red). The sampling sites are identiﬁed in the map. Imagery©2015 TerraMetrics; Map data ©2015 Google. Figure 1. Global shipping density map of Sicily, Italy as recorded by Maritime Trafﬁc (https://www.marinetrafﬁc.com/). Trafﬁc density is colour coded, moving from low to high trafﬁc (from blue to red). The sampling sites are identiﬁed in the map. Imagery©2015 TerraMetrics; Map data ©2015 Google. 507 Behavioural responses in a congested sea suspended from a boat (3 m deep), connected to a digital recorder (DR-100mkII, recording at 48 KHz 24-bit, Tascam Teac Professional, USA). Gain settings at all locations were kept constant. The methodology used solely relies on sound pressure levels, without including analyses of particle motion and velocity component (Radford et al. 2012). Also, the methodology used cannot discriminate the overall background noise in more speciﬁc components, such as biological noise (produced by organisms), inﬂu- ences of currents and sediment type, and trafﬁc noise (Robinson et al. 2014; Nedelec et al. 2015). Nevertheless, the measurements obtained, combined with trafﬁc data, can be used as a proxy of in-situ noise. was hung 3 m deep from the boat and the 5-min noise track was played continuously for 1 h to let the damselﬁsh colony acclimate to the new disturbance. After this period, behavioural videos and ambient noise were recorded as described in the previous section, while the noise-generating system was play- ing. The courtship behaviour of the focal males from the observational study was used as the control to this experiment. Thus, videos were monitored for 19 focal males (12 in C and 7 in D). A total of 54 videos (approximately 81 h of obser- vation in total) were recorded among the three study sites (A, B, C) and the manipulative treatment (D, artiﬁcial noise given by playback of boat noise). Experimental set-up Video cameras (XTC-200, Midland, China) were installed in front of nests, at a distance of 80–120 cm. Cameras recorded the interactions occurring around the nests between 9:00 am and 11:30 am, in sunny conditions. During that time the hydrophone was activated, to match visual beha- viours with background noise. Courtship behaviours were monitored in this way for a total of 47 focal males (20 in A, 15 in B and 12 in C). Manipulative observation - In order to alter the background noise in the natural reserve (C), a sound track was prepared recording the noise generated by several boats and ships transiting in the commercial harbour of Palermo with the same hydrophone and digital recorder used throughout the behavioural observations. These recordings were pooled to cre- ate a 5-min noise track using the software Audacity 2.0.0 (http://audacity.sourceforge.net/). This track was reproduced underwater in Vendicari Nature Reserve using a sound generating system composed by an underwater speaker (LL916C, Lubell Labs, Ohio, USA), an ampliﬁer (HPB-210, Monacor, UK) and a music player, all powered by two 12 Volts 12000 mAh batteries. The underwater speaker Damselﬁsh abundance analyses Damselﬁsh abundance (number of individuals within the colony) varied across populations. Chromis chromis are well known to be very abundant 508 S. Mascolino et al. Figure 2. Signal jump (the most common courtship-related visual display performed by Chromis chromis) sequential snapshots. Nesting male leaps toward a female (a, b), performs an arching movement (c, d) and goes back to his nest (e, f). Figure 2. Signal jump (the most common courtship-related visual display performed by Chromis chromis) sequential snapshots. Nesting male leaps toward a female (a, b), performs an arching movement (c, d) and goes back to his nest (e, f). which the Mediterranean damselﬁsh hear and voca- lise (Wysocki et al. 2009; Picciulin et al. 2010). Subsample tracks were generated with Audacity, and Leq values were estimated using MATLAB (Brandt et al. 2011). Variability in ambient noise level between A, B and C and between C and D was compared by examining mean Leq for each narrowband and 95% C.I. All sound tracks were also visualised using the R packages seewave (Sueur et al. 2008) and tuneR, following Villanueva- Rivera et al. (2011). in the Mediterranean Sea (Pinnegar 2018). The variability in adult density across study sites might affect the reproductive behaviour of the ﬁsh (Soucy & Travis 2003). At the end of randomly chosen sampling sessions, video recordings (n = 6 for A, n = 3 for B, n = 2 for C) were used to estimate relative ﬁsh abundance. The diver, while ﬂoating at the centre of sampling area, collected a 360° panora- mic video-recording of approximately 30 s (Colin et al. 2003). Three random snapshots from each recording were used to count the number of damsel- ﬁsh using the software ImageJ 1.46r (Schneider et al. 2012). The three abundance values from each observation were averaged and this value was associated to the behavioural observations obtained during the same sampling session. For the beha- vioural observations during which abundance values were not collected we used the average value of the corresponding study area. Statistical analyses Average Leq and 95% conﬁdence intervals (C.I.) were calculated for each frequency narrowband and for each treatment. The difference in the num- ber of VD * h−1 performed by focal males between A (n = 12), B (n = 6) and C (n = 4) were investi- gated. Also, visual displays were compared between the manipulative treatment D (n = 6) and C (n = 4). For this purpose, two generalised linear models (GLM) were ﬁtted to the data (software R 2.13; R Core Team 2018). Due to data overdispersion, a GLM with a negative-binomial distribution and log link function was employed (O’Hara & Kotze 2010). The rate of visual displays (VD * h−1) was the response variable and area/treatment the inde- pendent variable. Due to marked differences in terms of number of individuals in each colony, Ambient noise The differences in noise level among the four areas/ treatments were empirically quantiﬁed (Table I). As expected, due to the prohibition of boat trafﬁc, the Vendicari Nature Reserve (C) was the area with lower noise levels, while Zingaro (B), surprisingly, exhibited overlap with the noisy harbour of Palermo (A) across most frequency intervals, and even proved to be “louder” at 400–630 Hz (Figure 3). In the manipulative treatment (D), the noise- generating system was successful in increasing the ambient noise to levels similar to the ones recorded in A (Table I; Figure 3). Measuring underwater sounds: lesson learned This observational study was performed to investi- gate variations in courtship strategies used by male damselﬁsh to attract females to their nests, in differ- ent areas, in the ﬁeld. Acoustic recordings for ﬁsh vocalization in open water are methodologically chal- lenging (Benoit-Bird & Lawson 2016; Slabbekoorn 2016) and often ecologists do not have easy access to the correct instruments or the required expertise from acousticians, from the early planning stages of the experiment. As the techniques advance and allow more relevant underwater measurements (e.g., from sound pressure to particle motion and sound expo- sure levels; Radford et al. 2012; Spiga et al. 2012; Hawkins & Popper 2018; Popper & Hawkins 2019) not only can we use eco-acoustic (Sueur & Farina 2015) to monitor biodiversity (Pieretti et al. 2017; Akamatsu et al. 2018; Doray et al. 2018; Linke et al. 2018; Gibb et al. 2019) and the effects of Ambient noise analyses Ambient noise levels were estimated for each of the three areas (A, B, C) and for the manipulative treat- ment (D). For each one, ten 1-min subsample tracks were randomly selected from the recordings col- lected in the ﬁeld and equivalent continuous sound level (Leq) were estimated for the one-third octave band centre frequencies (1/3 OBCF) 100, 125, 160, 200, 250, 315, 400, 500 and 630 Hz, the range at 509 Behavioural responses in a congested sea back transformed from logarithmic to linear scale: A = 27.9 [15.2, 51.1], B = 13.3 [4.7, 37.4], C = 5.3 [1.6, 17.1]). Speciﬁcally, a signiﬁcant dif- ference was found when comparing A and C: males in the area near the main commercial harbour of Palermo performed signiﬁcantly more VD than males in the natural reserve (z = 2.786, P = 0.0053). No signiﬁcant difference was found between B and the other two areas (B vs C, z = 1.121, P = 0.2622; B vs A, z = 1.227, P = 0.2197). After being subjected to an artiﬁcially increased ambient noise level (manipulative treat- ment with playback of boat noise, D), nesting males from C did not change their VD (z = −0.215, P = 0.829; Figure 5(b)). Note that in this case damselﬁsh abundance was not used as a covariate, as the two treatments (C and D) occurred in the same study area (Vendicari). relative abundance estimates were included as a covariate in the GLM and visualised and inter- preted using the effect display (“effects” package in R; Fox & Hong 2009). Due to the use of three areas, we do not have enough independent replicates to statistically infer differences due to nautical trafﬁc (Hurlbert 1984; Jordan 2018; Lazic et al. 2018). Nevertheless, we offer a biological interpretation (Davies & Gray 2015) for behavioural differences among sites, dis- cussing the various factors potentially involved (including boat trafﬁc and corresponding noise). Behavioural responses During the ﬁeld observations, the abundance of individuals per colony varied across the three areas (Figure 4(a–c)). A negative relationship was found between conspeciﬁc abundance and the rate of visual displays (z = −2.204, P = 0.0275; Figure 4(d)), thus abundance needed to be incorporated in the analyses. Nesting males from the three study areas (A, B and C) showed different numbers of VD per hour (Figure 5(a); mean and 95% C.I., Table I. Relative dB uncalibrated ambient noise level (mean, with 95% C.I. in brackets) for each 1/3 OBCF tested for the three study sites and the manipulative treatment. dB values were normalised by setting ‘0’ as the highest value reported. Table I. Relative dB uncalibrated ambient noise level (mean, with 95% C.I. in brackets) for each 1/3 OBCF tested for the three study sites and the manipulative treatment. dB values were normalised by setting ‘0’ as the highest value reported. 1/3 OBCF Palermo (A) Zingaro (B) Vendicari (C) Vendicari (D) 100 −7 (−12, −2) −5 (−12, 2) −19 (−20, −17) −13 (−18, −8) 125 −10 (−15, −5) −6 (−13, 1) −22 (−23, −21) −13 (−18, −9) 160 −12 (−16, −7) −6 (−11, −1) −22 (−24, −20) −12 (−17, −7) 200 −12 (−16, −8) −7 (−11, −2) −20 (−21, −18) −16 (−21, −12) 250 −12 (−16, −8) −6 (−11, −1) −20 (−21, −19) −14 (−19, −9) 315 −12 (−15, −9) −5 (−10, 0) −20 (−21, −20) −10 (−16, −5) 400 −13 (−17, −9) −5 (−9, 0) −21 (−22, −20) −14 (−18, −9) 500 −13 (−17, −10) −3 (−8, 1) −20 (−21, −19) −8 (−13, −3) 630 −11 (−15, −7) 0 (−5, 5) −17 (−18, −17) 0 (−6, 6) 510 S. Mascolino et al. Figure 3. Examples of selected spectrograms and oscillograms of the three study areas caracherised by different level of noise: a) Palermo; b) Zingaro; c) Vendicari; and d) the artiﬁcially altered area Vendicari. These are representative (most common visual representation) of the sound recorded in the different areas. Figure 3. Examples of selected spectrograms and oscillograms of the three study areas caracherised by different level of noise: a) Palermo; b) Zingaro; c) Vendicari; and d) the artiﬁcially altered area Vendicari. These are representative (most common visual representation) of the sound recorded in the different areas. Figure 4. Behavioural responses They are colonial and defend territories (e.g., Ferguson & Cleary 2001), so their sound might be prevalent in some areas but not others. Visual courtship behaviours, in B, were intermediate on average but not statistically different from either A or C. Signiﬁcant differences might not have been detected due to the great inter-individual variability, typical of animal beha- viour (Dall et al. 2012; Siegenthaler et al. 2018): when comparing two situations extremely different from one another (at least in terms of noise), as A and C, despite individual variability we can still detect signiﬁcant dif- ferences; this is not the case for intermediate situations, as B. Biological components (e.g., the sound produced by snapping shrimp) should interfere less than anthro- pogenic sounds with acoustic signalling, since during evolutionary time different taxa have utilised different “acoustic windows” (at different frequencies), to avoid interspeciﬁc acoustic overlapping (Brumm & Slabbekoorn 2005; Lugli 2010; Radford et al. 2014; Ruppé et al. 2015). Figure 5. (a) Number of VD * h-1 (back-transformed to linear scale and adjusted for abundance of individuals within the col- ony) performed by focal nesting males (mean and 95% C.I.) in the three areas, Palermo (A); Zingaro (B); Vendicari (C). (b) Number of VD * h-1 (back-transformed to linear scale) per- formed by focal nesting males for the two treatments (C and D) in Vendicari (mean and 95% C.I.). Despite the most evident difference among the three sites being maritime trafﬁc, other biological and envir- onmental factors might have played an important role in the trade-off between a more visual or vocal strategy. An increase in visual displays was the hypothesised response to the masking of vocal cues, but a visual strategy can also be risky if detected by predators. Presence and abundance of predators were not assessed in this study; nevertheless the high noise area is likely to be the one with the lowest presence of natural predators of C. chromis while the low noise area, being a natural reserve, is likely to be also the area with highest biodi- versity and higher presence of natural predators (Madin et al. 2015). The observed reduction of visual displays in the latter might not only depend on the low noise level but also on predation pressure (birds and piscivor- ous ﬁshes; Pinnegar 2018). Behavioural responses (a-c) Examples of snapshots of the number of individuals in the colonies in the three study locations: a) Palermo; b) Zingaro; c) Vendicari. (d) Effect of number of individuals in the colony on the number of visual displays (mean and 95% C.I.) performed by focal nesting males (plotted with the effect display in R). Figure 4. (a-c) Examples of snapshots of the number of individuals in the colonies in the three study locations: a) Palermo; b) Zingaro; c) Vendicari. (d) Effect of number of individuals in the colony on the number of visual displays (mean and 95% C.I.) performed by focal nesting males (plotted with the effect display in R). the hydrophone (as sometime still reported in litera- ture; Merchant et al. 2015) and the difﬁculty to record ﬁsh vocalizations, resulted in proxy values for ambient noise and the impossibility to analyse male calls. A more multidisciplinary framework to connect ecologists with acousticians should be encouraged to produce stronger analyses. anthropogenic inﬂuences on its distribution, but also we can perform evolutionary and behavioural ecolo- gical studies on the effect of noise pollution (Benoit- Bird & Lawson 2016) and move from the individual to the population and ecosystem level (Kunc et al. 2016; Nabe-Nielsen et al. 2018; Lowerre-Barbieri et al. 2019). In this study, a lack of calibration of 511 Behavioural responses in a congested sea Figure 5. (a) Number of VD * h-1 (back-transformed to linear scale and adjusted for abundance of individuals within the col- ony) performed by focal nesting males (mean and 95% C.I.) in the three areas, Palermo (A); Zingaro (B); Vendicari (C). (b) Number of VD * h-1 (back-transformed to linear scale) per- formed by focal nesting males for the two treatments (C and D) in Vendicari (mean and 95% C.I.). soundscape. By listening to the recordings collected in B, and by looking at the spectrograms of the sound ﬁles (Figure 3), it is reasonable to assume that this result was due to a high baseline noise of non-anthropogenic ori- gin (such as the biophony produced by snapping shrimps, family Alpheidae; Radford et al. 2010; Nedelec et al. 2015; Pieretti et al. 2017; Putland et al. 2017; audio ﬁles in supplementary material). Snapping shrimp produce a very broad frequency spectrum, with a low frequency start and mean peak levels between 2 and 5 kHz but up to 200 kHz and beyond (Au & Banks 1998). Behavioural responses Moreover, while the males are performing their visual displays (signal jumps), nest predators (i.e., egg eating predators such as the ornate wrasse Thalassoma pavo; Milazzo et al. 2006) might take advantage of the absence of the male from the nest to attack the eggs. Thus, visually conspicuous dis- plays might pose a higher risk for nesting males A comparison of sites characterised by differing levels of boat trafﬁc A comparison of sites characterised by differing levels of boat trafﬁc We can hypothesize that, when vocalisations fail to attract females to the nest (because of the masking effect of anthropogenic noise), males possibly rely more on visual displays, which are not affected by ambient noise (Neenan et al. 2016). Such a multimodal shift from acoustic to visual signalling has been described in a variety of other animals, includ- ing frogs, birds and mammals (Caldart et al. 2016; Partan 2017) and reported in ﬁsh (under lab condi- tions; de Jong et al. 2018). In this observational study, a statistically signiﬁcant difference between two of the three areas under investigation was recorded. Male damselﬁsh from the Palermo commercial harbour (A) performed more courtship-related visual displays (over ﬁve times) than individuals from a natural reserve (C). One measured difference between these two sites is ambient noise (which reﬂects boat trafﬁc). Interestingly, the other area (B), where maritime trafﬁc is considerably less prevalent and less regular than in Palermo harbour, presented a different overall S. Mascolino et al. 512 short-term plasticity. Behavioural plasticity can facili- tate the adjustment to anthropogenic pressures or environmental changes. For instance, male guppies are able to quickly adjust their mating behaviour from visual courtship displays in higher light conditions to a sneaky mating strategy under low light and visibility conditions (Chapman et al. 2009). In our case a lack of short-term immediate behavioural ﬂexibility might result in more durable adaptive changes in the long term (Swaddle et al. 2015; Harding et al. 2018). Also, a recent study on Pomacentrus amboinensis juveniles (Holmes et al. 2017) showed a short response of ﬁsh to small boat noise (20 min only) before ﬁsh resume normal behaviours: the used one-hour acclimation time might have allowed a desensitization of the ﬁsh by the beginning of the recordings (but this would not be the case if the vocalization would be masked and males had to rely more on visual displays). Collectively, our ﬁnd- ings suggest that courtship behaviour in C. chromis may be a locally adapted complex trait, shaped by long-term (multi-generational) exposure to ambient noise levels. The visual vs vocalisation trade-off may represent an important biocomplexity that allows individuals to attain optimal reproductive success in divergent noise environments. employing them, making predation pressure another likely explanation for the difference in courtship strat- egy observed between areas characterised by different levels of maritime trafﬁc. A comparison of sites characterised by differing levels of boat trafﬁc The high number of individuals in one colony seemed to play a role in the investment in visual displays. High abundance of conspeciﬁc means high presence of females (thus possibly reduced effort to attract them to the nest) as well as higher presence of sneakers (Picciulin et al. 2004; Mascolino et al. 2016), conspeciﬁc males who can “sneak” in another male’s nest to release sperm while the female is spawning, parasitically fertilising eggs at the expenses of the nesting male. The visual display is a signal that both females (the intended target) and sneakers (the opportunistic receiver) can detect. Again, it is important to underline that noise is not the only driver able to modulate courtship behaviours. In this study, damselﬁsh abundance was included in the analyses, but predation was not, and these two factors seem to play a combined role in other ﬁsh species, where in the presence of predators, sneakers were chased less than in the absence of predators (Järvi-Laturi et al. 2011). Visual displays might not be effective in turbid water. Even though turbidity was not measured, the most turbid area of the study was the noisier, Palermo (Mascolino personal observation; see also Figure 4(a– c)), where more visual displays were recorded. Studies on visual communication in ﬁshes have focused on colouration and colour perception and indeed turbid- ity can mask this important signal to attract females (Seehausen et al. 1997). Nevertheless, colour is not important in C. chromis mate choice, and at the low distances at which C. chromis interact, movement can still be easily detected. In many species, differential ﬁtness of “courtship genotypes” may be an important factor to be consid- ered for population viability and management, similar to life history traits, such as growth and maturation, which are more commonly considered in a ﬁshery per- spective (van Wijk et al. 2013). Molecular studies can offer the framework to address this aspect more in detail. In this regards, C. chromis is typically charac- terised by localised habitat use and marked territoriality as well as a relatively short pelagic larval stage (Raventós & Macpherson 2001). Domingues et al. (2005) detected a certain extent of genetic structure among C. chromis population across the Mediterranean Sea, but small-scale empirical evidence for substructure and demographic independence in this species is miss- ing. A comparison of sites characterised by differing levels of boat trafﬁc The two populations from Palermo and Zingaro are characterised by reduced genetic connectivity: a high degree of population structure was recorded (FST = 0.024, 95% BootStrap C.I. 0.02, 0.03), despite being separated by only 70 Km of coastline. If ambient noise proves to select for certain traits over others (i.e., visual display vs vocalisation), it is reasonable to infer that anthropogenic noise-induced evolutionary change will be particularly rapid in sedentary species with low dispersal. Given the limitations associated with the methodol- ogy adopted to collect data, the fact that acoustic voca- lisations produced by male C. chromis were not recorded and the potential interaction of multiple fac- tors, our results might not be conclusive. Still, on the basis of our evidence it is reasonable to infer that boat trafﬁc, with the concomitant anthropogenic noise, is one of the drivers of the differences seen in the three areas. Short-term plasticity in response to increased noise levels Short-term plasticity in response to increased noise levels When faced with artiﬁcially increased ambient noise (D), focal nesting males accustomed to a low-noise environment were not able to adjust their courtship strategy to the new noisier condition after one-hour acclimation and they did not increase their number of visual displays. This result suggests the absence of Male C. chromis use both visual and acoustic dis- plays to communicate with conspeciﬁcs and attract mates (Picciulin et al. 2010). In particular, female reproductive responses are triggered by both visual 513 Behavioural responses in a congested sea and auditory stimuli and this multimodal commu- nication is quite common in ﬁshes (Brumm & Slabbekoorn 2005; Bolgan et al. 2013; Pedroso et al. 2013). The fact that there might be the possi- bility to modulate the signalling modality (Radford et al. 2014 and references therein) has been consid- ered as a potential mitigation of the disturbance produced by anthropogenic noise, thus minimizing ﬁtness consequences (Brumm 2014). On the other hand, the two modes of communication could be complementary more than redundant, as seen in gobies (Torricelli et al. 1986) and the effects of anthropogenic stressors could affect more beha- viours at once (McLaughlin & Kunc 2015). Indeed, the dynamics of multiple signalling are com- plex (Bro-Jørgensen 2010) and it makes sense that they would evolve in heterogeneous environments, such as the soundscape of coastal areas (Nedelec et al. 2015). relation to short-term disruption, but also as a potential driver of evolutionary change. Our results suggest that underwater noise pollu- tion could signiﬁcantly impact the evolutionary tra- jectories of marine organisms’ communication systems at local scales. More speciﬁc studies should be designed to assess this more in detail (Underwood 2000): our analysis has pointed out a variety of other possible factors (including preda- tor pressure, population density, water turbidity, courtship genotypes) that can affect changes in courtship displays from acoustic to visual, in the ﬁeld. There is still relatively little understanding of the ecological responses of marine organisms to the increase in ambient sound (Slabbekoorn et al. 2010; Kunc et al. 2016) and this preliminary investigation paves the way towards ﬁlling this gap together with other studies on the same spe- cies (Codarin et al. 2009; Picciulin et al. 2010; Bracciali et al. 2012). However, the potential long-term effects of these processes warrant an increased awareness in the context of environ- mental impact assessment and resource manage- ment. Short-term plasticity in response to increased noise levels Special concern should be given to marine protected areas where noise mitigation strategies, such as navigation-speed restrictions or rerouting of shipping lanes (as suggested by Merchant et al. 2012), and speciﬁc guidelines (Popper & Hawkins 2019) are overall necessary to ensure the protec- tion of ecological/evolutionary processes. Furthermore, anthropogenic noise could have an effect, not only on adult damselﬁsh, but also on juveniles: it has been shown that juveniles of other pomacentrid species are able to use acoustic cues to discriminate among habitats and to be guided dur- ing the homing process (Radford et al. 2011). This mechanism seems to be plastic (Simpson et al. 2010) and can be affected by boat noise (Holles et al. 2013). Considering that all ﬁsh studied to date are able to hear and that maritime trafﬁc is a frequent, widespread phenomenon at a global scale, it can be expected that every-day human activ- ities could have a pervasive impact on ecological and evolutionary processes of coastal ﬁsh species. References Bolgan M, Pedroso SS, Picciulin M, Fonseca PJ, Amorim MCP. 2013. Differential investment in acoustic communication dur- ing social interactions in two closely-related sand goby species. Behaviour 150:133–152. DOI: 10.1163/1568539X- 00003041. Aguilar de Soto N. 2016. Peer-reviewed studies on the effects of anthropogenic noise on marine invertebrates: From scallop larvae to giant squid. In: Popper AN, Hawkins A, editors. The effects of noise on aquatic life II. New York: Springer. pp. 17–26. DOI: 10.1007/978-1-4939-2981-8_3. Bracciali C, Campobello D, Giacoma C, Sarà G. 2012. Effects of nautical trafﬁc and noise on foraging patterns of Mediterranean damselﬁsh (Chromis chromis). PLoS ONE 7: e40582. DOI: 10.1371/journal.pone.0040582. Aguilar de Soto N, Gkikopoulou K, Hooker S, Isojunno S, Johnson M, Miller P, Tyack P, Wensveen P, Donovan C, Harris CM, Harris D, Marshall L, Oedekoven C, Prieto R, Thomas L. 2016. From physiology to policy: A review of physiological noise effects on marine fauna with implications for mitigation. Proceedings of Meetings on Acoustics 27:040008. DOI: 10.1121/2.0000299. Brandt MJ, Diederichs A, Betke K, Nehls G. 2011. Responses of harbour porpoises to pile driving at the Horns Rev II offshore wind farm in the Danish North Sea. Marine Ecology Progress Series 421:205–216. DOI: 10.3354/meps08888. Akamatsu T, Lin TH, Tsao Y, Sinniger F, Harii S. 2018. Coral reef soundscape to measure the species distribution and bio- diversity. 2018. Kobe: OCEANS-MTS/IEEE Kobe Techno- Oceans (OTO). pp. 1–4. DOI: 10.1109/ OCEANSKOBE.2018.8559475 Bro-Jørgensen J. 2010. Dynamics of multiple signalling systems: Animal communication in a world in ﬂux. Trends in Ecology & Evolution 25:292–300. DOI: 10.1016/j.tree.2009.11.003. Brumm H. 2014. Fish struggle to be heard-but just how much ﬁn waving is there? A comment on Radford et al. Behavioral Ecology 25:1033–1034. DOI: 10.1093/beheco/aru029. Allen GR. 1991. Damselﬁsh of the World. Melle, Germany: Mergus Press. Brumm H, Slabbekoorn H. 2005. Acoustic communication in noise. Advances in the Study of Behavior 35:151–209. DOI: 10.1016/S0065-3454(05)35004-2. Amorim MCP. 2006. Diversity of sound production in ﬁsh. In: Ladich F, Collin SP, Moller P, Kapoor BG, editors. Communication in ﬁshes. Enﬁeld: Science Publishers. pp. 71–104. Caldart VM, Iop S, Lingnau R, Cechin SZ. 2016. Communication in a noisy environment: Short-term acoustic adjustments and the underlying acoustic niche of a Neotropical stream-breeding frog. Acta Ethologica 19:151–162. DOI: 10.1007/s10211-016-0235-2. Amorim MCP, Neves A. 2007. Acoustic signalling during court- ship in the painted goby, Pomatoschistus pictus. Journal of Marine Biological Association United Kingdom 87:1017–1023. DOI: 10.1017/S0025315407056822. Casaretto L, Picciulin M, Hawkins AD. Conclusions We would like to thank: Devin Lyons and Julius Piercy for their assistance with data analysis; Mark Avis and Andy Moorhouse for their advice on acoustics; Donal Lennon for the guidance pro- vided to build the underwater sound-generating system; Natalia Niceta, Jennifer Coughlan, and Siobhán Bradley, Kelly Hickey, Emma Lawlor, Meabh Nic Mhathúna, Claire Morton and Martina O’Brien for their help in the ﬁeld. Access to the Vendicari no-navigation zone was kindly granted by the crew of R.N.O. Oasi Faunistica di Vendicari. The research reported here follows the Association for the Study of Animal Behaviour and Animal Behavior Society Guidelines for the Use of Animals in Research and does not include any collection of samples (just video recordings). Ethical approval was obtained by UCD, University College Dublin (AREC-P-10-33). There is an increased awareness of the detrimental impact of anthropogenic noise on biodiversity, at multiple levels (Francis & Barber 2013; Shannon et al. 2016). Recent studies have targeted ﬁtness effects (McGregor et al. 2013), especially in birds (e.g., Halfwerk et al. 2011) and marine mammals (e.g., New et al. 2014). This is one of few ﬁeld studies observing different behaviours of ﬁsh in dif- ferent areas characterised by different boat trafﬁc and thus noise levels. The overall preliminary ﬁnd- ings of this study suggest that investment in visual versus acoustic displays can be affected by ambient noise levels. The association between courtship behaviour and anthropogenic noise does not appear to have short-term plasticity and may result from local adaptation (Swaddle et al. 2015); hence ambi- ent noise may represent a key factor in conservation biology and environmental management, not only in 514 S. Mascolino et al. Au WW, Banks K. 1998. The acoustics of the snapping shrimp Synalpheus parneomeris in Kaneohe Bay. Journal of the Acoustical Society of America 103:41–47. DOI: 10.1121/ 1.423234. ORCID Blumstein DT, Daniel JC. 2007. Quantifying behavior the JWatcher Way. Massachusetts, USA: Sinauer Associates. DOI: 10.1093/icb/icn005. S. Mariani http://orcid.org/0000-0002-5329-0553 C Benvenuto http://orcid org/0000-0002-8378-8168 S. Mariani http://orcid.org/0000-0002-5329-0553 S. Mariani http://orcid.org/0000-0002-5329-0553 C. Benvenuto http://orcid.org/0000-0002-8378-8168 C. Benvenuto http://orcid.org/0000-0002-8378-8168 Bolgan M, Chorazyczewska E, Winﬁeld IJ, Codarin A, O’Brien J, Gammell M. 2016. First observations of anthropogenic underwater noise in a large multi-use lake. Journal of Limnology 75:644–651. Funding Bass AH, Ladich F. 2008. Vocal-acoustic communication: From neurons to behavior. In: Webb JF, Fay RR, Popper AN, editors. Fish bioacoustics. Springer Handbook of Auditory Research 32. New York, NY: Springer. pp. 253–278. This work was supported by the Irish Research Council (IRC) [Embark Initiative (RS/2010/2106)]. Benoit-Bird KJ, Lawson GL. 2016. Ecological insights from pelagic habitats acquired using active acoustic techniques. Annual Review Marine Science 8:463–490. DOI: 10.1146/ annurev-marine-122414-034001. Supplementary material Supplemental data for this article can be accessed here. Bessey C, Heithaus MR. 2013. Alarm call production and tem- poral variation in predator encounter rates for a facultative teleost grazer in a relatively pristine seagrass ecosystem. Journal of Experimental and Marine Biology and Ecology 449:135–141. DOI: 10.1016/j.jembe.2013.09.008. Disclosure statement No potential conﬂict of interest was reported by the authors. Barber JR, Crooks KR, Fristrup KM. 2009. The costs of chronic noise exposure for terrestrial organisms. Trends in Ecology & Evolution 25:180–189. DOI: 10.1016/j.tree.2009.08.002. References Don’t let spurious accusations of pseu- doreplication limit our ability to learn from natural experiments (and other messy kinds of ecological monitoring). Ecology and Evolution 5:5295–5304. DOI: 10.1002/ece3.1782. de Jong K, Amorim M, Fonseca PJ, Heubel KU. 2018. Noise affects multimodal communication during courtship in a marine ﬁsh. Frontiers in Ecology and Evolution 6:113. DOI: 10.3389/fevo.2018.00113. Halfwerk W, Holleman LJM, Lessels CM, Slabbekoorn H. 2011. Negative impact of trafﬁc noise on avian reproductive success. Journal of Applied Ecology 48:210–219. DOI: 10.1111/ j.1365-2664.2010.01914.x. Domingues VS, Bucciarelli G, Almada V, Bernardi G. 2005. Historical colonization and demography of the Mediterranean damselﬁsh, Chromis chromis. Molecular Ecology 14:4051–4063. DOI: 10.1111/j.1365-294X.2005.02723.x. Harding HR, Gordon TAC, Hsuan RE, Mackaness ACE, Radford AN, Simpson SD. 2018. Fish in habitats with higher motorboat disturbance show reduced sensitivity to motorboat noise. Biology Letters 14:20180441. DOI: 10.1098/ rsbl.2018.0441. Dooling RJ, Leek MR, Popper AN. 2015. Effects of noise on ﬁshes: What we can learn from humans and birds. Integrative Zoology 10:29–37. DOI: 10.1111/1749-4877.12094. Hasan MR, Crane AL, Ferrari MC, Chivers DP. 2018. A cross-modal effect of noise: The disappearance of the alarm reaction of a freshwater ﬁsh. Animal Cognition 21:419–424. DOI: 10.1007/s10071-018-1179-x. Doray M, Petitgas P, Romagnan JB, Huret M, Duhamel E, Dupuy C, Spitz J, Authier M, Sanchez F, Berger L, Dorémus G, Bourriau P, Grellier P, Massé J. 2018. The PELGAS survey: Ship-based integrated monitoring of the Bay of Biscay pelagic ecosystem. Progress in Oceanography 166:15–29. DOI: 10.1016/j.pocean.2017.09.015. Haviland-Howell G, Frankel AS, Powell CM, Bocconcelli A, Herman RL, Sayigh LS. 2007. Recreational boating trafﬁc: A chronic source of anthropogenic noise in the Wilmington, North Carolina Intracoastal Waterway. Journal of Acoustic Society of America 122:151–160. DOI: 10.1121/1.2717766. Ellison W, Southall B, Clark C, Frankel A. 2011. A new context-based approach to assess marine mammal behavioral responses to anthropogenic sounds. Conservation Biology 26:21–28. DOI: 10.1111/j.1523-1739.2011.01803.x. Hawkins AD, Popper AN. 2018. Effects of man-made sound on ﬁshes. In: Slabbekoorn H, Dooling R, Popper AN, Fay RR, editors. Effects of anthropogenic noise on animals. New York, NY: Springer. pp. 145–177. Fakan E, McCormick M. 2019. Boat noise affects the early life history of two damselﬁshes. Marine Pollution Bulletin 141:493–500. DOI: 10.1016/j.marpolbul.2019.02.054. Hildebrand JA. 2009. Anthropogenic and natural sources of ambient noise in the ocean. Marine Ecology Progress Series 395:5–20. DOI: 10.3354/meps08353. Ferguson BG, Cleary JL. 2001. References 2015. Seasonal patterns and individual differences in the calls of male haddock Melanogrammus aegleﬁnus. Journal of Fish Biology 87:579–603. DOI: 10.1111/jfb.12740. Amorim MCP, Pedroso SS, Bolgan M, Jordao JM, Caiano M, Fonseca PJ. 2012. Painted gobies sing their quality out loud: Acoustic rather than visual signals advertise male quality and contribute to mating success. Functional Ecology 27:289–298. DOI: 10.1111/1365-2435.12032. Chapman BB, Morrell LJ, Krause J. 2009. Plasticity in male courtship behaviour as a function of light intensity in guppies. Behavioral Ecology and Sociobiology 63:1757–1763. DOI: 10.1007/s00265-009-0796-4. André M, Kaifu K, Solé M, van der Schaar M, Akamatsu T, Balastegui A, Sánchez AM, Castell JV. 2016. Contribution to the understanding of particle motion perception in marine invertebrates. In: Popper AN, Hawkins A, editors. The effects of noise on aquatic life II. New York: Springer. pp. 47–55. DOI: 10.1007/978-1-4939-2981-8_6. Codarin A, Wysocki LE, Ladich F, Picciulin M. 2009. Effects of ambient and boat noise on hearing and communication in three ﬁsh species living in a marine protected area 515 Behavioural responses in a congested sea (Miramare, Italy). Marine Pollution Bulletin 58:1880–1887. DOI: 10.1016/j.marpolbul.2009.07.011. a changing world: Learning from the past to promote sustain- ability of ﬁsh populations. Journal of Fish Biology 92:804–827. DOI: 10.1111/jfb.2018.92.issue-3. Colin PL, Sadovy YJ, Domeier ML. 2003. Manual for the study and conservation of reef ﬁsh spawning aggregations. Society for the conservation of reef ﬁsh aggregations special publication No. 1 (Version 1.0). pp. 1–98+iii. https://www.researchgate. net/proﬁle/Michael_Domeier/publication/228436345_Manual_ for_the_Study_and_Conservation_of_Reef_Fish_Spawning_ Aggregations/links/0deec528510e82dc7b000000.pdf. Greggor AL, Berger-Tal O, Blumstein DT, Angeloni L, Bessa- Gomes C, Blackwell BF, St Clair CC, Crooks K, De Silva S, Fernández-Juricic E, Goldenberg SZ, Mesnick SL, Owen M, Price CJ, Saltz D, Schell CJ, Suarez AV, Swaisgood RR, Winchell CS, Sutherland WJ. 2016. Research priorities from animal behaviour for maximising conservation progress. Trends in Ecology & Evolution 31:953–964. DOI: 10.1016/j. tree.2016.09.001. Dall SRX, Bell AM, Bolnick DI, Ratnieks FLW. 2012. An evolu- tionary ecology of individual differences. Ecology Letters 15:1189–1198. DOI: 10.1111/j.1461-0248.2012.01846.x. Haelters J, Debusschere E, Botteldooren D, Dulière V, Hostens K, Norro A. 2013. The effects of pile driving on marine Mammals and ﬁsh in Belgian waters. In: Degraer S, Brabant R, Rumes B, editors. Environmental impacts of off- shore wind farms in the Belgian part of the North Sea. pp. 70–77. DOI: 10.1088/1748-9326/9/3/034012. Davies GM, Gray A. 2015. References Effects of noise on sound detection and acoustic communication in ﬁshes. In: Brumm H, editor. Animal com- munication and noise. Berlin Heidelberg: Springer-Verlag. pp. 65–90. DOI: 10.1007/978-3-642-41494-7_4. Mickle MF, Higgs DM. 2017. Integrating techniques: A review of the effects of anthropogenic noise on freshwater ﬁsh. Canadian Journal of Fisheries and Aquatic Sciences 75:1534–1541. DOI: 10.1139/cjfas-2017-0245. Ladich F, Fay RR. 2013. Auditory evoked potential audiometry in ﬁsh. Reviews in Fish Biology and Fisheries 23:317–364. DOI: 10.1007/s11160-012-9297-z. Laglbauer BJ, Afonso P, Donnay A, Santos RS, Fontes J. 2017. Reproductive synchrony in a temperate damselﬁsh, Chromis limbata. Acta Ethologica 20:297–311. DOI: 10.1007/s10211- 017-0269-0. Milazzo M, Anastasi I, Willis TJ. 2006. Recreational ﬁsh feeding affects coastal ﬁsh behavior and increases frequency of preda- tion on damselﬁsh Chromis chromis nests. Marine Ecology Progress Series 310:165–172. DOI: 10.3354/meps310165. Lazic SE, Clarke-Williams CJ, Munafò MR. 2018. What exactly is ‘N’ in cell culture and animal experiments? PLoS Biology 16:e2005282. DOI: 10.1371/journal.pbio.2005282. Miller P, Kvadsheim P, Lam F, Tyack P, Curé C, DeRuiter S, Wensveen PJ, von Benda-beckmann AM, MartínLópez LM, Narazaki T, Hooker SK. 2015. First indications that northern bottlenose whales are sensitive to behavioural disturbance from anthropogenic noise. Royal Society Open Science 2:140484. DOI: 10.1098/rsos.140484. Linke S, Gifford T, Desjonquères C, Tonolla D, Aubin T, Barclay L, Karaconstantis C, Kennard MJ, Rybak F, Sueur J. 2018. Freshwater ecoacoustics as a tool for contin- uous ecosystem monitoring. Frontiers in Ecology and the Environment 16:231–238. DOI: 10.1002/fee.2018.16.issue-4. Millot S, Vandewalle P, Parmentier E. 2011. Sound production in red-bellied piranhas (Pygocentrus nattereri, Kner): An acous- tical, behavioural and morphofunctional study. Journal of Experimental Biology 214:3613–3618. DOI: 10.1242/ jeb.061218. Lobel PS, Kaatz IM, Rice AN. 2010. Acoustical behaviour of coral reef ﬁshes. In: Cole KS, editor. Reproduction and sexu- ality in marine ﬁshes. Patterns and processes. Berkeley, CA, USA: University of California Press. pp. 307–385. DOI: 10.1525/california/9780520264335.001.0001. Myrberg AA. 1997. Sound production by a coral reef ﬁsh (Pomacentrus partitus): Evidence for a vocal, territorial “Keep Out” signal. Bulletin Marine Science 60:1017–1025. Lowerre-Barbieri SK, Catalán IA, Frugård Opdal A, Jørgensen C. 2019. Preparing for the future: Integrating spa- tial ecology into ecosystem-based management. ICES Journal of Marine Science 76:467–476. DOI: 10.1093/icesjms/fsy209. Myrberg AA, Spires JY. 1972. Sound discrimination by the bico- lor damselﬁsh, Eupomacentrus partitus. Journal of Experimental Biology 57:727–735. Lugli M. 2010. Sounds of shallow water ﬁshes pitch within the quiet window of the habitat ambient noise. References In situ source level and source position estimates of biological transient signals produced by snapping shrimp in an underwater environment. The Journal of the Acoustical Society of America 109:3031–3037. DOI: 10.1121/1.1339823. Holles S, Simpson SD, Radford AN, Berten L, Lecchini D. 2013. Boat noise disrupts orientation behaviour in a coral reef ﬁsh. Marine Ecology Progress Series 485:295–300. DOI: 10.3354/ meps10346. Fox J, Hong J. 2009. Effect displays in R for multinomial and proportional-odds logit models: Extensions to the effects package. Journal of Statistical Software 32:1–24. DOI: 10.1111/j.1467-9531.2006.00180.x. Holmes LJ, McWilliam J, Ferrari MC, McCormick MI. 2017. Juvenile damselﬁsh are affected but desensitize to small motor boat noise. Journal of Experimental Marine Biology and Ecology 494:63–68. DOI: 10.1016/j.jembe.2017.05.009. Francis CD, Barber JR. 2013. A framework for understanding noise impacts on wildlife: An urgent conservation priority. Frontiers in Ecology and the Environment 11:305–314. DOI: 10.1890/120183. Hurlbert SH. 1984. Pseudoreplication and the design of ecologi- cal ﬁeld experiments. Ecological Monographs 54:187–211. DOI: 10.2307/1942661. Gibb R, Browning E, Glover-Kapfer P, Jones KE. 2019. Emerging opportunities and challenges for passive acoustics in ecological assessment and monitoring. Methods in Ecology and Evolution 10:169–185. DOI: 10.1111/2041-210X.13101. Järvi-Laturi M, Lindstrӧm K, Kvarnemo C, Svensson O. 2011. Sand goby males trade off between defence against egg pre- dators and sneak intrusions. Journal of Zoology 283:269–275. DOI: 10.1111/j.1469-7998.2011.00788.x. Gordon TAC, Harding HR, Clever FK, Davidson IK, Davison W, Montgomery DW, Weatherhead RC, Windsor FM, Armstrong JD, Bardonnet A, Bergman E, Britton JR, Côté IM, D’agostino D, Greenberg LA, Harborne AR, Kahilainen KK, Metcalfe NB, Mills SC, Milner NJ, Mittermayer FH, Montorio L, Nedelec SL, Prokkola JM, Rutterford LA, Salvanes AGV, Simpson SD, Vainikka A, Pinnegar JK, Santos EM. 2018. Fishes in Jordan CY. 2018. Population sampling affects pseudoreplication. PLoS Biology 16:e2007054. DOI: 10.1371/journal. pbio.2007054. Kasumyan A. 2008. Sounds and sound production in ﬁshes. Journal of Ichthyology 48:981–1030. DOI: 10.1134/ S0032945208110039. Kunc HP, McLaughlin KE, Schmidt R. 2016. Aquatic noise pollution: Implications for individuals, populations, and 516 S. Mascolino et al. ecosystems. Proceedings of the Royal Society B: Biological Sciences 283:20160839. DOI: 10.1098/rspb.2016.0839. ecosystems. Proceedings of the Royal Society B: Biological Sciences 283:20160839. DOI: 10.1098/rspb.2016.0839. Merchant ND, Witt MJ, Blondel P, Godley BJ, Smith GH. 2012. Assessing sound exposure from shipping in coastal waters using a single hydrophone and Automatic Identiﬁcation System (AIS) data. Marine Pollution Bulletin 64:1320–1329. DOI: 10.1016/j.marpolbul.2012.05.004. Ladich F. 2013. References Journal of Comparative Physiology 196:439–451. DOI: 10.1007/ s00359-010-0528-2. Nabe-Nielsen J, van Beest FM, Grimm V, Sibly RM, Teilmann J, Thompson PM. 2018. Predicting the impacts of anthropo- genic disturbances on marine populations. Conservation Letters 11:e12563. DOI: 10.1111/conl.12563. Madin EM, Dill LM, Ridlon AD, Heithaus MR, Warner RR. 2015. Human activities change marine ecosystems by altering predation risk. Global Change Biology 22:44–60. DOI: 10.1111/gcb.13083. Nedelec SL, Simpson SD, Holderied M, Radford AN, Lecellier G, Radford C, Lecchini D. 2015. Soundscapes and living communities in coral reefs: Temporal and spatial variation. Marine Ecology Progress Series 524:125–135. DOI: 10.3354/meps11175. Mann DA, Lobel PS. 1997. Propagation of damselﬁsh (Pomacentridae) courtship sounds. The Journal of the Acoustical Society of America 101:3783–3791. DOI: 10.1121/1.418425. Neenan ST, Piper R, White PR, Kemp P, Leighton TG, Shaw PJ. 2016. Does masking matter? Shipping noise and ﬁsh vocalizations. In: Popper AN, Hawkins A, editors. The effects of noise on aquatic life II. New York: Springer. pp. 747–753. Maruska KP, Ung US, Fernald RD. 2012. The African cichlid ﬁsh Astatotilapia burtoni uses acoustic communication for reproduction: Sound production, hearing, and behavioral signiﬁcance. PLoS One 7:e37612. DOI: 10.1371/journal. pone.0037612. New LF, Clark JS, Costa DP, Fleishman E, Hindell MA, Klanjšček T, Lusseau D, Kraus S, McMahon CR, Robinson PW, Schick RS, Schwarz LK, Simmons SE, Thomas L, Tyack P, Harwood J. 2014. Using short-term measures of behaviour to estimate long-term ﬁtness of south- ern elephant seals. Marine Ecology Progress Series 496:99–108. DOI: 10.3354/meps10547. Mascolino S, Benvenuto C, Gubili C, Sacchi C, Boufana B, Mariani S. 2016. The ART of mating: Alternative reproduc- tive tactics and mating success in a nest-guarding ﬁsh. Journal of Fish Biology 89:2643–2657. DOI: 10.1111/jfb.13130. McGregor PK, Horn AG, Leonard ML, Thomsen F. 2013. Anthropogenic noise and conservation. In: Brumm H, editor. Animal communication and noise. Berlin Heidelberg: Springer-Verlag. pp. 409–444. DOI: 10.1007/978-3-642- 41494-7_1. O’Hara RB, Kotze DJ. 2010. Do not log-tranform count data. Methods in Ecology and Evolution 1:118–122. DOI: 10.1111/j.2041-210X.2010.00021.x. Parmentier E, Kever L, Casadevall M, Lecchini D. 2010. Diversity and complexity in the acoustic behaviour of Dacyllus ﬂavicaudus (Pomacentridae). Marine Biology 157:2317–2327. DOI: 10.1007/s00227-010-1498-1. McLaughlin KE, Kunc HP. 2015. Changes in the acoustic envir- onment alter the foraging and sheltering behaviour of the cichlid Amititlania nigrofasciata. Behavioural Processes 116:75–79. DOI: 10.1016/j.beproc.2015.04.012. Partan SR. 2017. Multimodal shifts in noise: Switching channels to communicate through rapid environmental change. Animal Behaviour 124:325–337. DOI: 10.1016/j.anbehav.2016.0 8.003. References Merchant ND, Fristrup KM, Johnson MP, Tyack PL, Witt MJ, Blondel P, Parks SE. 2015. Measuring acoustic habitats. Methods in Ecology and Evolution 6:257–265. DOI: 10.1111/2041-210X.12330. Pedroso SS, Barber I, Svensson O, Fonseca PJ, Amorim MCP. 2013. Courtship sounds advertise species identity and male Behavioural responses in a congested sea 517 quality in sympatric Pomatoschistus spp. gobies. PLoS ONE 8: e64620. DOI: 10.1371/journal.pone.0064620. Computing. Available: https://www.R-project.org/. Accessed Nov 2019 27. Pereira R, Rismondo S, Caiano M, Pedroso SS, Fonseca PJ, Amorim MCP. 2014. The role of agonistic sounds in male nest defence in the painted goby Pomatoschistus pictus. Ethology 120:53–63. DOI: 10.1111/eth.12180. Radford AN, Kerridge E, Simpson SD. 2014. Acoustic commu- nication in a noisy world: Can ﬁsh compete with anthropo- genic noise? Behavioral Ecology 25:1022–1030. DOI: 10.1093/beheco/aru029. Picciulin M, Costantini M, Hawkins AD, Ferrero EA. 2002. Sound emissions of the Mediterranean damselﬁsh Chromis chromis (Pomacentridae). Bioacoustics 12:236–238. DOI: 10.1080/09524622.2002.9753707. Radford AN, Purser J, Bruintjes R, Voellmy IK, Everley KA, Wale MA, Holles S, Simpson SD. 2015. Beyond a simple effect: Variable and changing responses to anthropogenic noise. In: Popper AN, Hawkins A, editors. The effects of noise on aquatic life II. New York: Springer. pp. 901–907. DOI: 10.1007/978-1-4939-2981-8_111. Picciulin M, Sebastianutto L, Codarin A, Farina A, Ferrero EA. 2010. In situ behavioural responses to boat noise exposure of Gobius cruentatus (Gmelin, 1789; fam. Gobiidae) and Chromis chromis (Linnaeus, 1758; fam. Pomacentridae) living in a Marine Protected Area. Journal of Experimental Marine Biology and Ecology 386:125–132. DOI: 10.1016/j. jembe.2010.02.012. Radford CA, Montgomery JC, Caiger P, Higgs DM. 2012. Pressure and particle motion detection thresholds in ﬁsh: A re-examination of salient auditory cues in teleosts. Journal of Experimental Biology 215:3429–3435. DOI: 10.1242/ jeb.073320. Picciulin M, Verginella L, Spoto M, Ferrero EA. 2004. Colonial nesting and the importance of the brood size in male parasitic reproduction of the Mediterranean damselﬁsh Chromis chromis (Pisces: Pomacentridae). Environmental Biology of Fishes 70:23–30. DOI: 10.1023/B:EBFI.0000022851.49302.df. Radford CA, Stanley JA, Simpson SD, Jeffs AG. 2011. Juvenile coral reef ﬁsh use sound to locate habitats. Coral Reefs 30:295–305. DOI: 10.1007/s00338-010-0710-6. Radford CA, Stanley JA, Tindle CT, Montgomery JC, Jeffs AG. 2010. Localised coastal habitats have distinct underwater sound signatures. Marine Ecology Progress Series 401:21–29. DOI: 10.3354/meps08451. Pieretti N, Martire ML, Farina A, Danovaro R. 2017. Marine soundscape as an additional biodiversity monitoring tool: A case study from the Adriatic Sea (Mediterranean Sea). Ecological Indicators 83:13–20. DOI: 10.1016/j. ecolind.2017.07.011. References Raventós N, Macpherson E. 2001. Planktonic larval duration and settlement marks on the otoliths of Mediterranean littoral ﬁshes. Marine Biology 138:1115–1120. DOI: 10.1007/ s002270000535. Pinnegar J. 2018. Why the damselﬁsh Chromis chromis is a key species in the Mediterranean rocky littoral-a quantitative perspective. Journal of Fish Biology 92:851–872. DOI: 10.1111/jfb.13551. Robinson SP, Lepper PA, Hazelwood RA. 2014. Good practice guide for underwater noise measurement. National Measurement Ofﬁce, Marine Scotland, The Crown Estate, NPL Good Practice Guide 133. ISSN: 1368-6550. https:// www.npl.co.uk/special-pages/guides/gpg133underwater. Pirotta V, Grech A, Jonsen ID, Laurance WF, Harcourt RG. 2019. Consequences of global shipping trafﬁc for marine giants. Frontiers in Ecology and Environment 17:39–47. DOI: 10.1002/fee.1987. Ruppé L, Clément G, Herrel A, Ballesta L, Décamps T, Kéver L, Parmentier E. 2015. Environmental constraints drive the par- titioning of the soundscape in ﬁshes. Proceedings of the National Academy of Sciences 112:6092–6097. DOI: 10.1073/pnas.1424667112. Popper AN. 2003. Effects of anthropogenic sounds on ﬁshes. Fisheries 28:24–31. DOI: 10.1111/j.1095-8649.2009.02319.x. Popper AN, Fay RR. 2011. Rethinking sound detection by ﬁshes. Hearing Research 273:25–36. DOI: 10.1016/j. heares.2009.12.023. Schneider CA, Rasband WS, Eliceiri KW. 2012. NIH Image to ImageJ: 25 years of image analysis. Nature Methods 9:671–675. DOI: 10.1038/nmeth.2089. Popper AN, Hastings MC. 2009. The effects of human-generated sound on ﬁsh. Integrative Zoology 4:43–52. DOI: 10.1111/ j.1749-4877.2008.00134.x. Seehausen O, Van Alphen JJ, Witte F. 1997. Cichlid ﬁsh diversity threatened by eutrophication that curbs sexual selection. Science 277:1808–1811. DOI: 10.1126/science.277.5333.1808. Popper AN, Hawkins AD. 2019. An overview of ﬁsh bioacoustics and the impacts of anthropogenic sounds on ﬁshes. Journal of Fish Biology 94:692–713. DOI: 10.1111/jfb.13948. Shannon G, McKenna MF, Angeloni LM, Crooks KR, Fristrup KM, Brown E, Warner KA, Nelson MD, White C, Briggs J, Briggs J, McFarland S, Wittemyer G. 2016. A synthesis of two decades of research documenting the effects of noise on wildlife. Biological Reviews 91:982–1005. DOI: 10.1111/brv.12207. Popper AN, Smith ME, Cott PA, Hanna BW, MacGillivray AO, Austin ME, Mann DA. 2005. Effects of exposure to seismic airgun use on hearing of three ﬁsh species. Journal of the Acoustical Society of America 117:3958–3971. DOI: 10.1121/1.1904386. Siegenthaler A, Mastin A, Dufaut C, Mondal D, Benvenuto C. 2018. Background matching in the brown shrimp Crangon cran- gon: Adaptive camouﬂage and behavioural-plasticity. Scientiﬁc Reports 8:3292. DOI: 10.1038/s41598-018-21412-y. Putland R, Constantine R, Radford C. 2017. Exploring spatial and temporal trends in the soundscape of an ecologically signiﬁcant embayment. Scientiﬁc Reports 7:5713. DOI: 10.1038/s41598-017-06347. References Noise negatively affects foraging and antipredator behaviour in shore crabs. Animal Behaviour 86:111e118. DOI: 10.1016/j. anbehav.2013.05.001. Sueur J, Aubin T, Simonis C. 2008. Seewave: A free modular tool for sound analysis and synthesis. Bioacoustics 18:213–226. DOI: 10.1080/09524622.2008.9753600. Waples DM, Thorne LH, Hodge LEW, Burke EK, Urian KW, Read AJ. 2013. A ﬁeld test of acoustic deterrent devices used to reduce interactions between bottlenose dolphins and a coastal gillnet ﬁshery. Biological Conservation 157:163–171. DOI: 10.1016/j.biocon.2012.07.012. Sueur J, Farina A. 2015. Ecoacoustics: The ecological investiga- tion and interpretation of environmental sound. Biosemiotics 8:493–502. DOI: 10.1007/s12304-015-9248-x. Swaddle JP, Francis CD, Barber JR, Cooper CB, Kyba CC, Dominoni DM, Shannon G, Aschehoug E, Goodwin SE, Kawahara AY, Luther D, Spoelstra K, Voss M, Longcore T. 2015. A framework to assess evolutionary responses to anthropogenic light and sound. Trends in Ecology & Evolution 30:550–560. DOI: 10.1016/j.tree.2015.06.009. Weilgart LS. 2007. The impacts of anthropogenic ocean noise on cetaceans and implications for management. Canadian Journal of Zoology 85:1091–1116. DOI: 10.1139/Z07-101. Weimann SR, Black AN, Leese J, Richter ML, Itzkowitz M, Burger RM. 2018. Territorial vocalization in sympatric dam- selﬁsh: Acoustic characteristics and intruder discrimination. Bioacoustics 27:87–102. DOI: 10.1080/ 09524622.2017.1286263. Tidau S, Briffa M. 2016. Review on behavioral impacts of aquatic noise on crustaceans. In: Proceedings of Meetings on Acoustics 4ENAL. Volume 27: Fourth International Conference on the Effects of Noise on Aquatic Life, Dublin, Ireland, 10-16 July. https://asa.scitation.org/doi/pdf/10.1121/ 2.0000302?class=pdf. Whitﬁeld A, Becker A. 2014. Impacts of recreational motorboats on ﬁshes: A review. Marine Pollution Bulletin 83:24–31. DOI: 10.1016/j.marpolbul.2014.03.055. Williams R, Wright AJ, Ashe E, Blight L, Bruintjes R, Canessa R, Clark C, Cullis-Suzuki S, Dakin D, Erbe C, Hammond PS, Merchant ND, O’Hara PD, Purser J, Radford AN, Simpson SD, Thomas L, Wale MA. 2015. Impacts of anthro- pogenic noise on marine life: Publication patterns, new dis- coveries, and future directions in research and management. Ocean & Coastal Management 115:17–24. DOI: 10.1016/j. ocecoaman.2015.05.021. Tolimieri N, Haine O, Jeffs A, McCauley R, Montgomery J. 2004. Directional orientation of pomacentrid larvae to ambi- ent reef sound. Coral Reefs 23:184–191. DOI: 10.1007/ s00338-004-0383-0. Torricelli P, Lugli M, GandolﬁG. 1986. A quantitative analysis of the occurrence of visual and acoustic displays during the courtship in the freshwater goby, Padogobius martensi (Günther, 1961) (Pisces, Gobiidae). Italian Journal of Zoology 53:85–89. DOI: 10.1080/11250008609355488. Wong BB. 2014. Animal communication in a human-dominated world: A comment on Radford et al. Behavioral Ecology 25:1033. DOI: 10.1093/beheco/aru139. References Simpson SD, Meekan MG, Larsen NJ, McCauley RD, Jeffs A. 2010. Behavioral plasticity in larval reef ﬁsh: Orientation is inﬂuenced by recent acoustic experiences. Behavioral Ecology 21:1098–1105. DOI: 10.1093/beheco/arq117. Putland RL, Merchant ND, Farcas A, Radford CA. 2018. Vessel noise cuts down communication space for vocalizing ﬁsh and marine mammals. Global Change Biology 24:1708–1721. DOI: 10.1111/gcb.13996. Slabbekoorn H. 2016. Aiming for progress in understanding underwater noise impact on ﬁsh: Complementary need for indoor and outdoor studies. In: Popper AN, Hawkins A, edi- tors. The effects of noise on aquatic life II. New York: Springer. pp. 1057–1065. Quignard JP, Pras A. 1986. Pomacentridae. In: Whitehead PJP, Bauchot ML, Hureau JC, Nielsen J, Tottonese E, editors. Fishes of the North-eastern Atlantic and Mediterranean (Volume II, pp. 916–918). UK: UNESCO. R Core Team. 2018. R: A language and environment for statis- tical computing. Vienna, Austria: R Foundation for Statistical Slabbekoorn H, Bouton N, van Opzeeland I, Coers A, Ten Cate C, Popper AN. 2010. A noisy spring: The impact of globally rising S. Mascolino et al. 518 based shifts in body size and maturation. Frontiers in Ecology and Environment 11:181–187. DOI: 10.1890/120229. underwater sound levels on ﬁsh. Trends in Ecology & Evolution 25:419–427. DOI: 10.1016/j.tree.2010.04.005. Soucy S, Travis J. 2003. Multiple paternity and population genetic structure in natural populations of the poeciliid ﬁsh, Heterandria formosa. Journal of Evolutionary Biology 16:1328–1336. DOI: 10.1046/j.1420-9101.2003.00608.x. Verginella L, Spoto M, Ciriaco S, Ferrero EA. 1999. Ethogram of the reproductive territorial male of the Mediterranean damsel- ﬁsh Chromis chromis L. (Pisces: Pomacentridae). Bollettino della Società Adriatica di Scienze Naturali LXXVIII:437–454. Southall BL, Bowles AE, Ellison WT, Finneran JJ, Gentry RL, Greene JCR, Kastak D, Ketten DR, Miller JH, Nachtigal PE, Richardson WJ, Thomas JA, Tyack PL. 2008. Marine mammal noise-exposure criteria: Initial scientiﬁc recommendations. Bioacoustics 17:273–275. DOI: 10.1578/AM.33.4.2007.411. Villanueva-Rivera L, Pijanowski BC, Doucette J, Pekin B. 2011. A primer of acoustic analysis for landscape ecologists. Landscape Ecology 9:1233–1246. DOI: 10.1007/s10980-011- 9636-9. Wahlberg M, Westerberg H. 2005. Hearing in ﬁsh and their reactions to sounds from offshore wind farms. Marine Ecology Progress Series 288:295–309. DOI: 10.3354/ meps288295. Spiga I, Cheesman S, Hawkins A, Perez-Dominguez R, Roberts L, Hughes D, Elliott M, Nedwell J, Bentley M. 2012. Understanding the scale and impacts of anthropogenic noise upon ﬁsh and invertebrates in the marine environment. Defra: SoundWaves Consortium Technical Review (ME5205). Wale MA, Simpson SD, Radford AN. 2013. References Underwood A. 2000. Importance of experimental design in detecting and measuring stresses in marine populations. Journal of Aquatic Ecosystem Stress and Recovery 7:3–24. DOI: 10.1023/A:1009983229076. Wysocki LE, Codarin A, Ladich F, Picciulin M. 2009. Sound pressure and particle acceleration audiograms in three marine ﬁsh species from the Adriatic Sea. Journal of the Acoustical Society of America 126:2100–2107. DOI: 10.1121/ 1.3203562. van Wijk SJ, Taylor MI, Creer S, Dreyer C, Rodrigues FM, Ramnarine IW, van Oosterhout C, Carvalho GR. 2013. Experimental harvesting of ﬁsh populations drives genetically
https://openalex.org/W4389393103	https://www.frontiersin.org/articles/10.3389/fnins.2023.1252850/pdf?isPublishedV2=False	English	null	Characterization of dynamic patterns of human fetal to neonatal brain asymmetry with deformation-based morphometry	Frontiers in neuroscience	2,023	cc-by	11,518	TYPE Original Research PUBLISHED 06 December 2023 DOI 10.3389/fnins.2023.1252850 TYPE Original Research PUBLISHED 06 December 2023 DOI 10.3389/fnins.2023.1252850 OPEN ACCESS OPEN ACCESS EDITED BY Alessandro Crimi, Sano Center for Computational Medicine, Poland REVIEWED BY Cemal Koba, Sano Centre for Computational Personalised Medicine, Poland Gesa Berretz, Ruhr University Bochum, Germany CORRESPONDENCE Andras Jakab andras.jakab@kispi.uzh.ch 1Center for MR Research, University Children’s Hospital Zurich, University of Zurich, Zürich, Switzerland, 2Children’s Research Center, University Children’s Hospital Zurich, University of Zurich, Zurich, Switzerland, 3Pediatric Heart Center, Division of Pediatric Cardiology, University Children’s Hospital Zurich, University of Zurich, Zurich, Switzerland, 4Neuroscience Center Zurich, University of Zurich and Swiss Federal Institute of Technology, Zurich, Switzerland, 5Department of Information Technology and Electrical Engineering, ETH Zurich, Zurich, Switzerland, 6Newborn Research, Department of Neonatology, University of Zurich and University Hospital Zurich, Zurich, Switzerland, 7Larsson-Rosenquist Foundation Center for Neurodevelopment, Growth and Nutrition of the Newborn, Department of Neonatology, University Hospital Zurich, University of Zurich, Zurich, Switzerland, 8Department of Diagnostic Imaging, University Children’s Hospital Zurich, University of Zurich, Zurich, Switzerland RECEIVED 04 July 2023 ACCEPTED 03 November 2023 PUBLISHED 06 December 2023 RECEIVED 04 July 2023 ACCEPTED 03 November 2023 PUBLISHED 06 December 2023 RECEIVED 04 July 2023 ACCEPTED 03 November 2023 PUBLISHED 06 December 2023 CITATION Steger C, Moatti C, Payette K, De Silvestro A, Nguyen TD, Coraj S, Yakoub N, Natalucci G, Kottke R, Tuura R, Knirsch W and Jakab A (2023) Characterization of dynamic patterns of human fetal to neonatal brain asymmetry with deformation-based morphometry. Front. Neurosci. 17:1252850. doi: 10.3389/fnins.2023.1252850 Steger C, Moatti C, Payette K, De Silvestro A, Nguyen TD, Coraj S, Yakoub N, Natalucci G, Kottke R, Tuura R, Knirsch W and Jakab A (2023) Characterization of dynamic patterns of human fetal to neonatal brain asymmetry with deformation-based morphometry. Front. Neurosci. 17:1252850. doi: 10.3389/fnins.2023.1252850 COPYRIGHT © 2023 Steger, Moatti, Payette, De Silvestro, Nguyen, Coraj, Yakoub, Natalucci, Kottke, Tuura, Knirsch and Jakab. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Introduction: Despite established knowledge on the morphological and functional asymmetries in the human brain, the understanding of how brain asymmetry patterns change during late fetal to neonatal life remains incomplete. Frontiers in Neuroscience 1 Introduction Many studies report brain asymmetry, but quantitative MRI data on gestational time-dependent brain morphology changes primarily come from cross-sectional studies, which are constrained by signiﬁcant inter-individual brain surface variability. Consequently, inferring developmental trends from cross-sectional data can be confounded by this variability and may not accurately represent a within-individual dynamic asymmetry evolution. Furthermore, technical challenges may have contributed to the lack of studies quantifying the dynamic patterns of brain asymmetry over the fetal to neonatal period. Popular neuroscience software and toolboxes are very limited for use with fetal or neonatal MRI data due to diﬀerent MR image contrasts and not yet developed brain structures, highlighting the urgent need for specialized tools to examine early brain development. While brain volumetry is seen as the gold standard for analyzing brain growth, segmentation of the brain into consistent subregions throughout development remains challenging. Alternative techniques such as deformation-based morphometry (DBM) have been applied before (Jakab et al., 2019; Ng et al., 2020), relying on non-linear image registration. Using DBM, no consistent segmentation of subregions is required to detect local asymmetries of the brain. Brain asymmetry is likely not a static or linearly emerging property during development; instead, it might involve a complex dynamic process where patterns of left-right asymmetry may appear or diminish due to variations in the expression of developmental genetic programs that control the formation of the characteristic cortical surface patters and brain specialization. Inter-hemispheric asymmetry already develops in the embryonic stage and is induced not only by genetic components but also dependent on environmental factors (Schmitz et al., 2019; Güntürkün et al., 2020). Post-mortem studies have indicated that right hemispheric structures and a leftward planum temporale (PT) asymmetry have an early origin in fetal brain development (Wada et al., 1975; Chi et al., 1977). While some works found no morphological asymmetries in the fetus using post-mortem analysis (Zhang et al., 2013), asymmetrical cortical folding patterns were described in-vivo as early as the 22nd gestational week (Habas et al., 2012). Several studies described volumetric changes throughout late fetal development for each hemisphere separately to assess their asymmetry (Cai et al., 2020; Vasung et al., 2020; Kienast et al., 2021; Machado-Rivas et al., 2022; Yun et al., 2022). For example, temporal lobe asymmetries are known to change dynamically during the second half of gestation (Kasprian et al., 2011). 1 Introduction until adulthood in healthy human brains (Dubois et al., 2009; Hill et al., 2010). Asymmetry of the brain manifests on various levels, including genetic, morphological, and functional dimensions (Wan et al., 2022). From the neuroscience perspective, studying anatomical asymmetries and their development can reveal insights into the emergence of brain functional specialization, using macro- morphological inter-hemispheric asymmetry as a proxy. The lateralization of sensory-motor functions, behavioral and cognitive processes in the brain is a well-described phenomenon, which is not exclusive to humans. It is widely believed that this lateralization provides evolutionary advantages at the behavioral level, including decreased reaction time and parallel processing in both hemispheres (Rogers et al., 2004; Dongen, 2006; Güntürkün et al., 2020). On a morphological level, post-mortem studies in humans provide evidence for cytoarchitectonic asymmetries (Galaburda et al., 1978) and macroscopic asymmetries such as the ‘Yakovlevian-Torque’ can be observed by pure visual inspection. From the clinical perspective, there is emerging evidence for the association of altered or disrupted hemispheric asymmetry in neurological disorders and neuropsychological conditions (Postema et al., 2019; Lubben et al., 2021; Sone et al., 2022). In addition to the neuroscientiﬁc signiﬁcance of inter- hemispheric asymmetries throughout cerebral development, the ﬁnal stage of this development holds considerable biological and clinical importance, warranting further investigation in this domain. During the last 8 weeks of gestation and early newborn phase, the brain undergoes rapid growth and morphological features, such as secondary and tertiary gyri emerge. In addition, birth takes place, undoubtedly a crucial moment, where a lot of mechanical stress is experienced (Ami et al., 2022) and the entire organism needs to adapt to new conditions and is no longer supported by the mother. Furthermore, this timeframe is of interest, as this is a realistic timeframe to acquire longitudinal data in patients which show altered brain development and risk for brain injuries (Peyvandi et al., 2021) and alternatives to tissue speciﬁc volume measurements could provide new insights. In recent years, advances in fetal and infant MRI have improved access to studying in vivo human brain development. However, limited literature exists on longitudinally acquired data and there is a clear limitation of these studies because of their cross- sectional design (that is, each developmental stage is represented by a diﬀerent individual). OPEN ACCESS The goal of this study was to characterize the dynamic patterns of inter- hemispheric brain asymmetry over this critically important developmental stage using longitudinally acquired MRI scans. Introduction: Despite established knowledge on the morphological and functional asymmetries in the human brain, the understanding of how brain asymmetry patterns change during late fetal to neonatal life remains incomplete. The goal of this study was to characterize the dynamic patterns of inter- hemispheric brain asymmetry over this critically important developmental stage using longitudinally acquired MRI scans. Methods: Super-resolution reconstructed T2-weighted MRI of 20 neurotypically developing participants were used, and for each participant fetal and neonatal MRI was acquired. To quantify brain morphological changes, deformation- based morphometry (DBM) on the longitudinal MRI scans was utilized. Two registration frameworks were evaluated and used in our study: (A) fetal to neonatal image registration and (B) registration through a mid-time template. Developmental changes of cerebral asymmetry were characterized as (A) the inter-hemispheric differences of the Jacobian determinant (JD) of fetal to neonatal morphometry change and the (B) time-dependent change of the JD capturing left-right differences at fetal or neonatal time points. Left-right and fetal-neonatal differences were statistically tested using multivariate linear models, corrected for participants’ age and sex and using threshold-free cluster enhancement. Results: Fetal to neonatal morphometry changes demonstrated asymmetry in the temporal pole, and left-right asymmetry differences between fetal and neonatal timepoints revealed temporal changes in the temporal pole, likely to go from right dominant in fetal to a bilateral morphology in neonatal timepoint. Furthermore, the analysis revealed right-dominant subcortical gray matter in neonates and three clusters of increased JD values in the left hemisphere from fetal to neonatal timepoints. 01 Frontiers in Neuroscience frontiersin.org Steger et al. 10.3389/fnins.2023.1252850 Discussion: While these ﬁndings provide evidence that morphological asymmetry gradually emerges during development, discrepancies between registration frameworks require careful considerations when using DBM for longitudinal data of early brain development. KEYWORDS brain asymmetry, magnetic resonance imaging (MRI), fetal brain, neonatal brain, DBM = deformation-based morphometry, longitudinal Frontiers in Neuroscience frontiersin.org 2.2 MRI acquisition Fetal MRI was performed on a 1.5T clinical whole-body MRI scanner (GE Signa Discovery MR450) using a 32-channel cardiac coil or an 8-channel body array coil. The MRI scanner was upgraded to GE Signa Artist during the data collection. Post- upgrade fetal scans were performed with an AIR coil, with the anterior array (blanked coil) having 30 channels and table coil 40 channels. T2-weighted single shot Fast Spin Echo (ssFSE) sequences were acquired in axial, sagittal and coronal orientations, relative to the fetal brain. In-plane resolution was 0.5 mm × 0.5 mm and a slice thickness of 3–5 mm was applied, depending on the covered organ (placenta, whole fetus or brain only). The sequence parameters were the following: TR: 2,000–3,500 ms, TE: 120 ms (minimum), ﬂip angle: 90. During the scan, the quality of the ssFSE scans was checked by the attending pediatric radiologist and were repeated in case excessive movement or other artifacts were detected. As a rule of thumb, at least two image sequences per orientation were acquired. 2 Materials and methods The MR images were super-resolution (SR) reconstructed using the slice-to-volume reconstruction toolkit (Kuklisova-Murgasova et al., 2012). Fetal ssFSE images were ﬁrst checked visually for quality, and at least one axial, sagittal and coronal image was selected for SR reconstruction. Similarly, for the neonatal data, one image from each of the three orientations were used. A reference image was chosen and a mask covering the area of the brain was manually drawn using 3D Slicer Software. Images were pre- processed with N4ITK ﬁlter in the 3D Slicer Software (Fedorov et al., 2012) and denoising BTK Toolkit (Rousseau et al., 2013). Both fetal and neonatal images were reconstructed to 0.75 mm isotropic voxel resolution. SR images were reoriented to an age-matched template using a 6 degrees-of-freedom transformation to ensure consistent anatomical orientation. In a ﬁnal step, brains were skull- stripped using a binary mask based on brain segmentation labels (described below). 2.1 Participants Data from twenty healthy participants recruited as participants of two ongoing, prospective studies (BrainDNIU study and BrainCHD study) between 2017 and 2022 were taken for this analysis. Based on patient history, ultrasound (US) and MR scans, all fetuses and newborns were classiﬁed as having normal brain development. The female/male ratio was 14/6. All participants were scanned twice: during fetal life and as a newborn. The gestational week (GW) at scan was calculated based on clinical US records. Mean ± standard deviation (SD) age at scan was 32.4 ± 1.1 GW for the fetal timepoint and 41.7 ± 1.4 GW for neonatal. All babies were born around term at a mean age of 39.7 ± 1.2 GW. The captured time interval between the two MRIs was 65 ± 12 days for each participant. As part of the pre-processing for non-linear image registration, we explored whether replacing the black background voxels with bright voxels would aﬀect the performance of the deformable image registration. The aim of this procedure was to reduce the confounding eﬀect of the interface between dark voxels, such as the skull and background air, and the bright voxels of extracerebral ﬂuid spaces, as the latter appear very diﬀerent between the fetal and neonatal MR images. This step was performed by an in- house developed script with image processing steps that rely on FSL (Jenkinson et al., 2012) and AFNI (Cox, 1996). Brain and background classes were identiﬁed using 3dkmeans and the background was replaced by the 99th percentile value of the image. See Figure 1 for a fetal and neonatal image example with and without a “bleached” background. Parents gave written informed consent for the participation of the study they were enrolled in as well as consent for data sharing between the studies. BrainCHD and brainDNIU study were approved by the local ethics committee (BASEC IDs: 2019- 01993, 2017-00885). 1 Introduction Asymmetries in the PT and superior temporal sulcus are believed to persist throughout infancy and remain relatively stable The goal of our study was twofold: ﬁrst, to improve our understanding of the neuroanatomical changes from late fetal to neonatal stage by describing the age-related changes of 02 frontiersin.org Steger et al. 10.3389/fnins.2023.1252850 were the following: TR: 5,900 ms, TE: 97 ms, ﬂip angle: 90◦. In- plane resolution was 0.35 × 0.35 mm, slice thickness: 2.5 mm with a 0.2 mm slice gap. For the neonatal MRI, only one image sequence per orientation was acquired. During data collection time, the MR scanner was upgraded to GE Signa Premier. Post-upgrade neonatal scans were performed with a 48-channel head coil. brain asymmetry based on real longitudinal MRI data acquired at two timepoints. Second, to address outstanding challenges in the deformation-based morphometry analysis of fetal-to- newborn longitudinal development. Speciﬁcally, by conducting our analysis using two plausible registration frameworks and evaluating the eﬀects of image post processing options on the image registration accuracy. Frontiers in Neuroscience frontiersin.org 2.4.1.1 Anatomical overlap and volume differences All registration steps were done using the ANTs toolbox (Tustison et al., 2021). Figure 2C shows an overview of the steps, more details can be found in the Supplementary material. The fetal and neonatal SR images and the mid-age atlas MRI template were segmented into the following structures: cerebral spinal ﬂuid (CSF), grey matter (GM), white matter (WM), deep grey matter (dGM), brain stem (BS), and cerebellum (Cer). To automate this step, two in-house trained nnUNets (Isensee et al., 2021) were used, corresponding to fetal and neonatal MRI segmentation tasks. The fetal brain segmentation network was trained using the FeTA dataset (Payette et al., 2021), which provides manually annotated ground truth data. For the neonatal data, the ground truth data was generated using the developing human Frontiers in Neuroscience FIGURE 2 Overview of the two frameworks used to reveal asymmetry patterns and their temporal dynamics. (A) Framework A (violet), calculating the inter-hemispheric asymmetry of fetal to neonatal regional brain growth maps. (B) Framework B (orange), calculating fetal and neonatal inter-hemispheric asymmetry at the given timepoint, registering these to a mid-timepoint template and then calculating the differences between the inter-hemispheric asymmetry. (C) Flowchart showing steps required to extract JD maps and perform asymmetry analysis for framework A and B and steps required to extract metrics for the registration evaluation. Indicates registration step was evaluated (see * registration evaluation). Overview of the two frameworks used to reveal asymmetry patterns and their temporal dynamics. (A) Framework A (violet), calculating the inter-hemispheric asymmetry of fetal to neonatal regional brain growth maps. (B) Framework B (orange), calculating fetal and neonatal inter-hemispheric asymmetry at the given timepoint, registering these to a mid-timepoint template and then calculating the differences between the inter-hemispheric asymmetry. (C) Flowchart showing steps required to extract JD maps and perform asymmetry analysis for framework A and B and steps required to extract metrics for the registration evaluation. Indicates registration step was evaluated (see registration evaluation). In framework B (Figure 2B), we ﬁrst deﬁned static inter- hemispheric asymmetries at the fetal and neonatal time points individually, and then captured their time-dependent changes after registering the asymmetry maps to a mid-age template. For the asymmetry analysis in framework B, fetal and neonatal images were ﬁrst ﬂipped along the midline axis for each participant. Each ﬂipped image was then registered to the non- ﬂipped image and the JD capturing left-right diﬀerences were extracted. Next, the images and JD maps corresponding to the fetal and neonatal timepoints were registered to a mid- age template. Asymmetry analysis was performed by statistically testing for diﬀerences between the fetal and neonatal JD maps. section we provide an overview of the diﬀerent registration parameters tried. 2.4 Characterization of the temporal dynamics of morphological brain asymmetry To capture time-dependent changes of the inter-hemispheric asymmetry, multiple image registration frameworks are viable choices. Figure 2 gives an overview of the two frameworks used in our study to perform the asymmetry analysis. In framework A (Figure 2A), we ﬁrst deﬁned brain growth based on image registrations that map the fetal to the neonatal image within the same participant, directly capturing dynamic morphometric changes from fetal to neonatal timepoints. Similar to Jakab et al. (2019), longitudinal MR image pairs were registered for each participant using subsequent linear and non-linear image registrations. A Jacobian determinant (JD) map was calculated from the nonlinear registration component of this step. To In the neonates, all MR scans were performed in natural sleep, without sedation. Neonates received protective earplugs and earmuﬀs. T2-weighted fast spin-echo (FSE) sequences in all three planes were acquired on a 3.0 T MRI scanner using an 8-channel head coil (GE Signa Discovery MR750). The sequence parameters 03 frontiersin.org Steger et al. 10.3389/fnins.2023.1252850 FIGURE 2 Overview of the two frameworks used to reveal asymmetry patterns and their temporal dynamics. (A) Framework A (violet), calculating the inter-hemispheric asymmetry of fetal to neonatal regional brain growth maps. (B) Framework B (orange), calculating fetal and t l i t h i h i t t th i ti i t FIGURE 1 Axial view of fetal and neonatal image of a participant. Top: fetal, bottom: neonatal. Left original image, right image background replaced with bright values. Axial view of fetal and neonatal image of a participant. Top: fetal, bottom: neonatal. Left original image, right image background replaced with bright values. Axial view of fetal and neonatal image of a participant. Top: fetal, bottom: neonatal. Left original image, right image background replaced with bright values. statistically test inter-hemispheric asymmetry, as a next step, the JD maps and neonatal MRI were then ﬂipped along the midline axis. The ﬂipped and non-ﬂipped brain and JD pairs were registered to a neonatal template where the asymmetry analysis was performed. Inter-hemispheric diﬀerence in morphometric changes were tested as the diﬀerences between the ﬂipped and non-ﬂipped JD in the shared template space. frontiersin.org 2.5 Statistical analysis of inter-hemispheric brain asymmetry FIGURE 3 Visualization of the registration evaluation metrics used in our study. (A) Coronal, sagittal and axial view of segmentation overlaid on a T2-weighted SR image of a fetus. (B) Coronal, sagittal and axial view of a glass brain rendering showing all sixteen, manually placed landmarks that were used during the registration evaluation. As deﬁned in Section “2.4 Characterization of the temporal dynamics of morphological brain asymmetry,” both frameworks used JD maps to capture brain morphometry changes or brain asymmetry. Following this logic, the statistical evaluations of brain asymmetry and the time-dependent change of brain asymmetry were performed as follows. First, JD used for the given analysis were smoothed [using 3dBlurInMask (Cox, 1996)] and stacked into a 4D format using fslmerge (Jenkinson et al., 2012). Statistical brain asymmetry analysis was carried out using the randomize tool of FSL. Randomize is an implementation of randomized non-parametric permutation testing (Winkler et al., 2014). Paired sample designs were set up. Threshold-free cluster enhancement (TFCE) (Smith and Nichols, 2009) was used to correct for multiple comparisons, using TFCE with 5,000 permutations. In framework A the linear regression was performed contrasting two groups: ﬂipped and non- ﬂipped JD maps. Demeaned GW at fetal and neonatal scan were added as covariates. Similarly, in framework B the asymmetry analysis for the fetal and neonatal timepoint was performed. The two groups ﬂipped and non-ﬂipped were contrasted and the demeaned GW at scan and sex were added as covariate. To capture the temporal changes in asymmetry from fetal to neonatal in framework B the (non-ﬂipped) fetal and (non-ﬂipped) neonatal JD were tested for signiﬁcant diﬀerences while including sex and the age variable of both timepoints as covariates. Visualization of the registration evaluation metrics used in our study. (A) Coronal, sagittal and axial view of segmentation overlaid on a T2-weighted SR image of a fetus. (B) Coronal, sagittal and axial view of a glass brain rendering showing all sixteen, manually placed landmarks that were used during the registration evaluation. connectome project (dHCP) structural pipeline (Makropoulos et al., 2018). To improve the consistency between the anatomical deﬁnition of the segmentations in the fetal and neonatal timepoints, several modiﬁcations of the tissue label maps were necessary. We combined the cerebral spinal ﬂuid and ventricle labels. The hippocampus label generated by the neonatal network was assigned to gray matter. All segmentations were visually checked by the ﬁrst author of this study. 2.4.1 Evaluation of registration accuracy for large registration steps Registration accuracy was tested using two volumetric (image segmentation-based) accuracy metrics and one anatomical landmark-based metric. The rationale for this was to obtain a set of quantitative measurements of accuracy. In the Supplementary 04 frontiersin.org Steger et al. 10.3389/fnins.2023.1252850 2.5 Statistical analysis of inter-hemispheric brain asymmetry The segmented anatomical structures are illustrated in Figure 3A. The two segmentation-based evaluation metrics, Dice overlap coeﬃcient (Dice, 1945) and volume diﬀerence were evaluated using the co-registered segmentations (framework A: fetal, framework B: fetal and neonatal) and the target segmentation (framework A: participant speciﬁc neonatal segmentation, framework B: mid- age template). The Dice coeﬃcient and volume diﬀerence were calculated using their implementation in nipype (Gorgolewski et al., 2011). Volume diﬀerence is described using the following deﬁnition: 2.4.1.2 Anatomical landmark distances The following landmarks were annotated in each image following Lau (Lau et al., 2019): posterior commissure, infracollicular sulcus, pontomesencephalic junction (PMJ). In addition, the most caudal point of the 4th ventricle and the optic chiasm were annotated and for left and right hemisphere, the rostral end of: hemisphere, temporal lobe and lateral ventricles; and the caudal end of: lateral ventricles and hemisphere were annotated. The landmarks were manually placed in the fetal and neonatal SR images as well as the mid-age template using the Markup module in 3D Slicer, and their coordinates were exported for the registration evaluation. After transforming each landmark with the given registration method, the distance between the ground truth and transformed points were calculated using the Euclidean distance. The anatomical landmarks are illustrated in Figure 3B. 2.6 Visualization and clusters For visualizing the results of the statistical tests, the anatomical clusters were shown after thresholding them at a value of 0.975, corresponding to p < 0.025. FSL Cluster reported the cluster size and generated corresponding labels. Mean, minimum and maximum JD value for each cluster were extracted using the labels. All visualizations were done in fsleyes in the FSL tool, images in the axial view were mirrored to show then in neurological orientation (left hemisphere corresponding to left side in the image). Statistical images were overlaid on the corresponding T2- weighted template images. Volume Diﬀerence = (segmentimage1 −segmentimage2)/ For the evaluation the absolute value of this metric was used. Frontiers in Neuroscience A) In addition, a pattern within the individual metrics seems to be visible. Landmarks located around the brainstem seem to have achieved consistently high alignment, whereas other landmarks had larger errors. For the segmentation-based metrics, the cerebellum, brainstem and deep grey matter achieved the highest dice score and there was lower accuracy for cortical grey matter and CSF. While the mean volume diﬀerence was close to zero, the individual labels had larger errors in both negative and positive directions. We ﬁrst quantiﬁed inter-hemispheric diﬀerences in regional fetal to neonatal morphometric change using the framework A. A repeated measures linear model (paired ﬂipped vs. non- ﬂipped measurements) revealed two anatomical clusters in which inter-hemispheric diﬀerences in the regional fetal to neonatal morphometry change maps were signiﬁcant (Figure 5). The larger cluster (Table 1, FN 1) was localized to the left temporal pole region extending into cortical gray matter of the temporal lobe as well as the surrounding CSF. The mean values of the JD maps for the ﬂipped and non-ﬂipped hemispheres are visualized within the cluster in the axial view (Figure 5, left panel). This analysis revealed that the regional fetal to neonatal morphometry change map in the left hemisphere (non-ﬂipped) is dominated by positive JD values, referring to regional expansion, however, due to the smoothing applied to the JD maps, it’s not possible to determine whether voxels in the CSF or GM are driving these results. Results of a complementary analysis, where we tested diﬀerent settings for the registration can be found in the Supplementary material. 3.1 Registration accuracy In Figure 4 the achieved registration accuracy for the three registrations fetal to neonatal, fetal to mid-template and neonatal to mid-template are visualized. Post-hoc Wilcox test between registration showed signiﬁcant diﬀerence between mean landmarks accuracy of fetal to neonatal and neonatal to template registration (p = 0.001). For the mean dice there was a signiﬁcant diﬀerence for all comparisons (fet2neo vs. fetal: p = 0.03, fet2neo vs. neonatal: p < 0.001, fetal vs. neonatal: p < 0.001). Finally, mean volume diﬀerence was signiﬁcantly diﬀerent for fetal to neonatal 05 frontiersin.org Steger et al. 10.3389/fnins.2023.1252850 FIGURE 4 Achieved metric accuracy for each participant for each registration step. (A) Landmark accuracy as Euclidean distance in mm. Overall mean landmark distance per registration step (Mean, top row) and individual landmark distance. (B) Overall mean Dice score and Dice per segmentation label. (C) Overall mean volume difference and volume difference per segmentation label. fet2neo = fetal to neonatal registration, fetal = fetal to template registration, neonatal = neonatal to template registration. PMJ, pontomesencephalic junction. ns = p-value > 0.05, * = p-value < 0.05, = p-value < 0.01, ** = p-value < 0.0001. Achieved metric accuracy for each participant for each registration step. (A) Landmark accuracy as Euclidean distance in mm. Overall mean landmark distance per registration step (Mean, top row) and individual landmark distance. (B) Overall mean Dice score and Dice per segmentation label. (C) Overall mean volume difference and volume difference per segmentation label. fet2neo = fetal to neonatal registration, fetal = fetal to template registration, neonatal = neonatal to template registration. PMJ, pontomesencephalic junction. ns = p-value > 0.05, * = p-value < 0.05, = p-value < 0.01, ** = p-value < 0.0001. frontiersin.org Frontiers in Neuroscience 3.2.1 Inter-hemispheric asymmetry of regional fetal to neonatal brain growth maps (framework A) registration compared to fetal (p < 0.001) and neonatal (p < 0.001) to template registration. 3.2 Dynamic patterns of brain morphometry and asymmetry In all experiments, overall scaling, translation, and rotation from the linear registration component has been factored out since our analyses relied on the deformation component of the fetal to neonatal or mid-age template registration. Therefore, we refer to inter-hemispheric asymmetry as the asymmetry of morphometric change, meaning relative expansion or contraction of a given anatomical structure. FIGURE 6 Fetal and neonatal inter-hemispheric asymmetry and dynamic, age-related changes of inter-hemispheric asymmetry (framework B). (Fetal/neonatal) Clusters corresponding to anatomical regions showing (static) inter-hemispheric asymmetry in either the fetal or neonatal time points. (Fetal to neonatal) Clusters corresponding to anatomical regions where inter-hemispheric asymmetry showed differences between the neonatal and fetal timepoints. We only illustrate clusters where neonatal JD was higher than fetal JD, since the opposite statistical test (fetal JD > neonatal) would reveal very similar ﬁndings in the homotopic ipsilateral anatomical regions. F1. . .N: clusters representing regional asymmetry in the fetal images (signiﬁcant TFCE-corrected clusters where ﬂipped JD > non-ﬂipped JD). N1. . .N: clusters representing regional asymmetry in the neonatal images (signiﬁcant TFCE-corrected clusters where ﬂipped JD > non-ﬂipped JD). C1. . .N: clusters where inter-hemispheric asymmetry changed from fetal to neonatal age. the JD maps to a mid-age template as shown in Figure 6. This approach enabled the quantiﬁcation of inter-hemispheric asymmetry individually at the fetal and neonatal timepoint as well as the characterization of age-dependent changes. Using a repeated measures linear model (fetal vs. neonatal measurements) regions with larger JD values in neonates were identiﬁed. This statistical analysis revealed three anatomical regions where inter-hemispheric asymmetry changed from fetal to neonatal timepoints. Table 2 summarizes the anatomical locations, cluster sizes, minimum p-values as well as JD (fetal and neonatal timepoints) of each cluster. Furthermore, we describe the localization of these clusters and link them to the previously described static asymmetries at the fetal and neonatal age to better characterize the dynamic changes from fetal to neonatal age. Cluster IDs are numbered in descending order of size. toward leftwards asymmetry of the deep white matter of the post- central gyrus and superior temporal lobule (or a right to left shift during fetal to neonatal development). In the parietal lobe, a cluster was found in the deep cerebral white matter, extending into to the body of the left lateral ventricle and extending superiorly into the parietal white matter of the post- central and angular gyri (ID in Table 2 and Figure 6: C2). Within the parietal aspect of this cluster, no fetal asymmetry was found (Figure 6), while strong, left dominant neonatal asymmetry was found in the static analysis (N1 in Figure 6). This refers to a trend toward more lateralized morphology in neonatal age in the left hemisphere. Frontiers in Neuroscience 3.2.2 Fetal to neonatal changes of inter-hemispheric asymmetry (framework B) We illustrated the anatomical clusters in which ﬂipped and non-ﬂipped morphometry change maps were signiﬁcantly different, referring to an inter-hemispheric difference in the regional fetal to neonatal morphometry change (TFCE-corrected p < 0.025). Left panel: population mean, voxel-level JD values in the left and right hemispheres characterizing regional brain growth (ﬂipped to illustrate anatomical match). Right panel: anatomical clusters overlaid on a neonatal T2-weighted template. Inter-hemispheric asymmetry of the regional fetal to neonatal morphometry change maps (framework A). We illustrated the anatomical clusters in which ﬂipped and non-ﬂipped morphometry change maps were signiﬁcantly different, referring to an inter-hemispheric difference in the regional fetal to neonatal morphometry change (TFCE-corrected p < 0.025). Left panel: population mean, voxel-level JD values in the left and right hemispheres characterizing regional brain growth (ﬂipped to illustrate anatomical match). Right panel: anatomical clusters overlaid on a neonatal T2-weighted template. 3.2.2 Fetal to neonatal changes of inter-hemispheric asymmetry (framework B) In framework B, inter-hemispheric asymmetry at each timepoint (fetal or neonatal) was tested after transforming 06 frontiersin.org Steger et al. 10.3389/fnins.2023.1252850 TABLE 1 Anatomical regions demonstrating inter-hemispheric morphometric asymmetry: fetal to neonatal asymmetry analysis (Framework A). Cluster Hemisphere Anatomical location Size (# voxels) Lowest p-value JD values ipsilateral Min–max Mean FN1 Left Temporal pole 1,462 0.006 −1.00 to 0.82 0.09 Clusters of a statistical threshold of TFCE-corrected p < 0.025 were included in this table. The clusters represent areas, where the JD values of the non-ﬂipped images were signiﬁcantly larger compared to the ﬂipped images. TABLE 1 Anatomical regions demonstrating inter-hemispheric morphometric asymmetry: fetal to neonatal asymmetry analysis (Framework A). Cluster Hemisphere Anatomical location Size (# voxels) Lowest p-value JD values ipsilateral Min–max Mean FN1 Left Temporal pole 1,462 0.006 −1.00 to 0.82 0.09 Clusters of a statistical threshold of TFCE-corrected p < 0.025 were included in this table. The clusters represent areas, where the JD values of the non-ﬂipped images were signiﬁcantly larger compared to the ﬂipped images. FIGURE 5 Inter-hemispheric asymmetry of the regional fetal to neonatal morphometry change maps (framework A). We illustrated the anatomical clusters in which ﬂipped and non-ﬂipped morphometry change maps were signiﬁcantly different, referring to an inter-hemispheric difference in the regional fetal to neonatal morphometry change (TFCE-corrected p < 0.025). Left panel: population mean, voxel-level JD values in the left and right hemispheres characterizing regional brain growth (ﬂipped to illustrate anatomical match). Right panel: anatomical clusters overlaid on a neonatal T2-weighted template. FIGURE 6 Fetal and neonatal inter-hemispheric asymmetry and dynamic, age-related changes of inter-hemispheric asymmetry (framework B). (Fetal/neonatal) Clusters corresponding to anatomical regions showing (static) inter-hemispheric asymmetry in either the fetal or neonatal time points. (Fetal to neonatal) Clusters corresponding to anatomical regions where inter-hemispheric asymmetry showed differences between the neonatal and fetal timepoints. We only illustrate clusters where neonatal JD was higher than fetal JD, since the opposite statistical test (fetal JD > neonatal) would reveal very similar ﬁndings in the homotopic ipsilateral anatomical regions. F1. . .N: clusters representing regional asymmetry in the fetal images (signiﬁcant TFCE-corrected clusters where ﬂipped JD > non-ﬂipped JD). N1. . .N: clusters representing regional asymmetry in the neonatal images (signiﬁcant TFCE-corrected clusters where ﬂipped JD > non-ﬂipped JD). C1. . .N: clusters where inter-hemispheric asymmetry changed from fetal to neonatal age. Inter-hemispheric asymmetry of the regional fetal to neonatal morphometry change maps (framework A). frontiersin.org 4 Discussion referring to a trend toward more lateralized morphology in neonatal age in the right hemisphere. Prenatal human brain development is a complex and precisely orchestrated process inﬂuenced by genetic programs and environmental factors. The appearance of macroscopic asymmetries between the brain’s hemispheres likely arises from a combination of various developmental events. To gain a deeper understanding of this phenomenon, our study examined the changing patterns of inter-hemispheric brain asymmetry from late fetal development to the neonatal stage. We employed an analytical approach with multiple tasks to optimize the image processing steps, all aimed at answering the fundamental neurobiological question: How does inter-hemispheric asymmetry change in the relatively short period from before birth to after birth? Out of the four areas showing signiﬁcant diﬀerences in inter-hemispheric asymmetry between the two developmental stages, two were predominantly localized to the left temporal lobe and left parietal and post-central region, one in the right deep gray matter and white matter structures, and the smallest cluster found in the left inferior frontal gyrus. Contrary to our expectation of obtaining concordant results with the two frameworks, only one anatomical region (temporal lobe) survived the TFCE-based correction for multiplicity in two diﬀerent frameworks of analysis. While the static asymmetry pattern provided insights into what could have driven the observed dynamic patterns, they also revealed regions that did not correspond to the temporal analysis. In the analysis of the static neonatal asymmetry, cluster N3, located in the right frontal lobe, may correspond to parts of F1 of the fetal asymmetry analysis. The presence of these two clusters could indicate, that asymmetry is present in both timepoints to a similar degree. For F3 located in the occipital lobe and N6 (not visualized) located in the right cerebellum we could not ﬁnd any correspondence between the time points. As these clusters were only identiﬁed in one either fetal or neonatal static asymmetry analysis, but not in the temporal dynamics, their temporal dynamics remain unclear. 3.2.3 Common ﬁndings in framework A and framework B Our working hypothesis was that the two methodological frameworks would yield comparable descriptions of the temporal dynamics of brain asymmetry, this way, cross-validating the results by using two diﬀerent pipelines. The trend toward more bilateral morphology or a left to right shift of the pole of the left temporal lobe was the only consistent ﬁnding across the two image analysis frameworks in our study. However, in both approaches that investigated the temporal dynamics we found anatomical regions that showed diﬀerent asymmetry patterns in late fetal than newborn age. In framework A the left temporal pole had higher JD values than the right, similarly, in framework B the left temporal pole revealed an asymmetry diﬀerence between fetal and neonatal timepoint. When looking at the static asymmetry results, this temporal dynamic seems to be driven by the smaller fetal left temporal pole, which is less asymmetric in the neonatal brains. FIGURE 6 In the left temporal lobe, a large cluster (ID in Table 2 and Figure 6: C1) was localized to the temporal pole, extending superiorly into the temporo-occipital cortex as well as post- central gyrus in the parietal cortex. Within this cluster, a right- dominant temporal pole was shown at the fetal stage, while no right dominance was found in the neonatal age, therefore the neonatal JD > fetal JD was signiﬁcant in the left hemisphere for the temporal pole region (Figure 6 C1). In the parietal cortex, extending into the deep white matter of the post-central gyrus and superior parietal lobule, more leftwards inter-hemispheric asymmetry was demonstrated in the neonates (Figure 6, N1/N4/N7). This marks a trend toward more bilateral morphology of the temporal lobes and temporo-occipital cortex in newborns compared to fetuses (or a left to right shift during fetal to neonatal development), and a trend In the deep white matter and subcortical gray matter, a cluster was revealed with its central aspect in the right thalamus, basal ganglia, claustrum, internal and external capsules, extending laterally to the posterior insula / posterior aspect of the central sulcus (ID in Table 2 and Figure 6: C3). In the fetal images, this cluster did not show inter-hemispheric asymmetry (Figure 6), 07 frontiersin.org Steger et al. 10.3389/fnins.2023.1252850 egions demonstrating inter-hemispheric morphometric asymmetry: fetal and neonatal MRI to template analysis (Framework B). TABLE 2 Anatomical regions demonstrating inter-hemispheric morphometric asymmetry: fetal and neonatal MRI to template analysis (Framework B). Cluster Hemisphere Anatomical location Most likely corresponding clusters Size (# voxels) Lowest p-value JD values fetal JD values neonatal Min– max Mean Min– max Mean C1 Left Temporal pole, temporo-occipital cortex, post-central gyrus, white matter of the parietal cortex F1, F2, N2 27,513 <0.001 −1.33 to 0.67 −0.19 −0.71 to 0.46 −0.02 C2 Left Centrum semiovale, white matter of the post-central and angular gyri, left ventricle frontal horn N1 8,825 <0.001 −0.66 to 0.60 −0.07 −0.67 to 0.37 0.06 C3 Right Thalamus, basal ganglia, claustrum, internal and external capsules, posterior most aspect of the lateral sulcus N2 5,639 0.006 −0.32 to 0.24 −0.05 −0.15 to 0.29 0.05 The clusters represent areas, where the JD values of the neonatal group were signiﬁcantly larger compared to the fetal group. Clusters were ordered according to descending size. Frontiers in Neuroscience 4.1 The dynamic patterns of inter-hemispheric brain asymmetry over the fetal to neonatal developmental period We have identiﬁed multiple brain regions that exhibit an increasing inter-hemispheric morphometric asymmetry from the fetal to neonatal timepoints. In our analysis, such a region could be detected either as a signiﬁcant cluster found in both the late fetal and neonatal periods, which gradually became more lateralized 08 frontiersin.org Steger et al. 10.3389/fnins.2023.1252850 TABLE 3 Anatomical regions demonstrating inter-hemispheric morphometric asymmetry, static asymmetry in the fetal and neonatal timepoints. cal regions demonstrating inter-hemispheric morphometric asymmetry, static asymmetry in the fetal and neonatal timepoints. Cluster Hemisphere with larger JD values Anatomical location Most likely corresponding clusters Size (# voxels) Lowest p-value JD values ipsilateral Min– max Mean Fetal inter-hemispheric morphometric asymmetry F1 Right Superior frontal gyrus, temporal pole N2, C1 13,191 <0.001 −0.31 to 1.03 0.15 F2 Right Occipital pole, cerebellar hemisphere N5 6,654 0.003 −0.52 to 0.74 0.13 F3 Right Occipital cortex, cuneus C1 1,995 0.013 −0.64 to 0.62 0.13 F4 Right Superior frontal gyrus N3 1,515 0.013 −0.27 to 0.57 0.11 F5 Right Superior frontal gyrus N3 273 0.02 −0.26 to 0.65 0.8 F6 Left Occipital cortex, primary visual N1 178 0.023 −0.13 to 0.28 0.09 Neonatal inter-hemispheric morphometric asymmetry N1 Left White matter, post-central gyrus, superior parietal lobule, superior temporal gyrus / PT region C2 22,230 <0.001 −0.61 to 0.40 0.05 N2 Right Thalamus, basal ganglia, claustrum, internal and external capsules, posteriormost aspect of the lateral sulcus F1, C1 18,602 <0.001 −0.54 to 0.49 0.06 N3 Right Superior frontal gyrus F5, F4 5,606 0.007 −0.22 to 0.60 0.04 N4 Left Olfactory cortex, white matter of the superior frontal gyrus 4,230 0.005 −0.18 to 0.26 0.04 N5 Right Occipital pole F2 815 0.017 −0.66 to 0.49 0.12 N6 Right Cerebellum 506 0.013 −0.07 to 0.29 0.07 N7 Left Deep white matter of the frontal lobe 407 0.023 −0.13 to 0.22 0.05 The clusters represent areas, where the JD values on the ipsilateral side (non-ﬂipped image) were larger than on the contralateral side (ﬂipped image). Clusters were ordered according to descending size. ere the JD values on the ipsilateral side (non-ﬂipped image) were larger than on the contralateral side (ﬂipped image). Clusters were ordered according to temporal lobe and Sylvian ﬁssure length (Lyttelton et al., 2009; Mallela et al., 2020), but a rightwards dominance in terms of gyral and sulcal development (Chi et al., 1977; Kasprian et al., 2011; Habas et al., 2012; Rajagopalan et al., 2011, 2012). Frontiers in Neuroscience frontiersin.org 4.1 The dynamic patterns of inter-hemispheric brain asymmetry over the fetal to neonatal developmental period These results were achieved by increasing the number of iterations, reducing the radius for the cross-correlation to two and replacing the background of the fetal images with a bright value (See Supplementary section Registration testing). The smallest cluster showing dynamically changing inter- hemispheric asymmetry was localized to the left frontal lobe, around the region of the left inferior frontal gyrus. Data from adults in the ENIGMA consortium conﬁrmed rightwards asymmetry of the surface area of the inferior frontal gyrus, however, there is a scarcity of developmental studies conﬁrming this observation. While this is the least signiﬁcant ﬁnding in our analysis, it is possible that it marks an emerging leftwards dominance of the inferior frontal gyrus. When using segmentation and landmark based metrics, certain errors have to be expected. Oishi et al. (2011) performed registration accuracy tests for their atlas based anatomical segmentation of neonatal brains. The average Dice measure was 0.82, which was similar to the achieved accuracy using manual segmentation. However, the segmented structures were anatomically detailed which makes a direct comparison to our metrics not possible. Similarly, for the landmark-based metric, Lau et al. (2019) developed a landmark based framework to evaluate the anatomical correspondence between to brain images in adults. They reported a landmark localization error of less than 1 mm for landmarks such as PMJ, culmen and infracollicular sulcus and a mean error after nonlinear registration of 0.68, 2.7, and 0.93. For our worst performing registration setting, fetal to neonatal, we could report registration for the PMJ of 1.07 after linear and 0.82 after nonlinear registration. While some of the landmarks were inspired by the paper of Lau et al. (2019), we also identiﬁed our own landmarks that were further away from the brain stem which might have suﬀered from greater inaccuracy during the annotation. The motivation behind this was to capture the registration accuracy in the entire brain and not just core structures. There are several ambiguities in our ﬁndings. For instance, there are regions that were statistically signiﬁcantly lateralized in the late fetal images but not in the neonatal images (such as F3, F5, F6, F7, see Table 2 and Figure 6) or regions that were only detected in the neonatal images (such as N4, N6 and N7, see Table 3 and Figure 6). There are several reasons for these changes. 4.1 The dynamic patterns of inter-hemispheric brain asymmetry over the fetal to neonatal developmental period 10.3389/fnins.2023.1252850 10.3389/fnins.2023.1252850 documented early sulcation development in the right hemisphere (De Vareilles et al., 2023). While visual inspection of the results is often advised, it still can be challenging to assess the best registration strategy for the given problem. Here, we used a set of landmarks and segmentation-based metrics to quantify the registration accuracy. Segmentation based metrics, similarly to volumetric studies, rely on consistent segmentation across the two timepoints. Therefore, we used tissue labels, which are easier to segment consistently across fetal and neonatal brains but allow for less local evaluation of the accuracy than ﬁner segmentation labels. The distance between two landmarks on the other hand is a single point and therefore describes the accuracy on a very local level. However, it is challenging to establish a set of suitable landmarks, as they need to be visible in the images in both timepoints. We identiﬁed a cluster with right sided (increasing) neonatal morphometric lateralization located in the subcortical gray matter, internal and external capsules, extending ventrally into the hippocampus or the temporal horn of the right lateral ventricle. Our ﬁnding is concordant with one of others, suggesting that morphological features of subcortical gray matter in neonates are right dominant (Ratnarajah et al., 2013; Dean et al., 2018). However, in the largest performed study of this kind using the data of the ENIGMA consortium, the thalamus and basal ganglia were overall left dominant, while the caudate nucleus, hippocampus and amygdala were right dominant (Guadalupe et al., 2017). While the cluster C3/N2 reached ventrally to the hippocampus, it is too early to conclude whether this is a ﬁnding concordant with the data from the ENIGMA study, since we also revealed a strong rightwards dominance in the neonatal brain (N2) in a region that encompasses the lateral aspects of the thalamus. It is highly likely that asymmetries in the thalamus reﬂect asymmetry of the white matter pathways they relay, and therefore might not be consistently left or right dominant for all parts of the thalamus and basal ganglia. After nonlinear registration, the achieved accuracy was moderate (Mean Dice of 0.75, 0.85, and 0.74 and mean landmark distance of 2.4, 2.0, and 2.8 mm, respectively for fetal to template, neonatal to template and fetal to neonatal registration). 4.1 The dynamic patterns of inter-hemispheric brain asymmetry over the fetal to neonatal developmental period First, technical aspects such as the presence of clusters that are smaller than the registration inaccuracy or noise diﬀered between fetal and neonatal images. Secondly, there is the possibility that very subtle changes did not survive the statistical correction for multiple comparisons. Our results imply that static asymmetry in the neonatal stage may be more easily and reliably detected. This is likely due to the larger size, better image quality, increased development and maturity of the neonatal brain, which makes it easier to identify morphometric features even with less precise registration. 4.1 The dynamic patterns of inter-hemispheric brain asymmetry over the fetal to neonatal developmental period The results of multiple studies in infants are consistent with those found in adult brains (Kong et al., 2018), indicating that the structural diﬀerences responsible for hemispheric speech and language dominance may be formed during the ﬁnal trimester of fetal development (Chi et al., 1977; Wada and Davis, 1977; Gilmore et al., 2007; Hill et al., 2010; Li et al., 2014). The literature regarding the temporal lobe and its development remains ambiguous since an overall rightwards asymmetry for the temporal lobe in newborns has been reported (Lehtola et al., 2019), and there is converging evidence for rightwards superior temporal sulcus asymmetry (Hill et al., 2010; Leroy et al., 2015; Specht and Wigglesworth, 2018). Our study, however, used longitudinal data, therefore removing some of the ambiguity of inter-individual variability that may hinder studies that concluded developmental changes of asymmetry based on cross-sectional data. A similar study using a longitudinal design and a versatile, surface-based analysis in preterm infants replicated well- known asymmetries in the Sylvian ﬁssure and a more signiﬁcant opercularization in the left hemisphere, in contrast to the previously as indicated by larger JD values in the ipsilateral hemisphere, or alternatively, a region that is not (signiﬁcantly) detected in the fetal, but in the neonatal stage. Speciﬁcally, the left-sided deep white matter of the post- central gyrus, superior parietal lobule, superior temporal gyrus, and planum temporale region present in the dynamic analysis (Figure 6, C2 cluster) was present in both the fetal (Figure 6, F4 cluster) and neonatal brain (Figure 6, N1 cluster). The most concordant ﬁnding in our study in terms of across- method reproducibility, cluster size or signiﬁcance was the inter- hemispheric morphometric asymmetries of the temporal lobe. A cluster in the temporal pole in framework B (C1) was conﬂuent with the post-central gyrus as well as deep white matter while the temporal pole was the only region to be consistently found in framework A and B. The parietal lobe is known to display global leftwards dominance, which might be concordant with the leftwards localization of our clusters (C1, C2) in the deep white matter of the superior parietal lobule and angular gyrus (Lehtola et al., 2019). Several studies reported inter-hemispheric asymmetries in the temporal lobe during development, predominantly revealing an emerging left dominance for the planum temporale region and 09 frontiersin.org Steger et al. 4.2 Selection of registration methods for DBM in longitudinal fetal to neonatal cohorts For DBM, accurate registration is critical for obtaining reliable results. A lower degree of anatomical correspondence would require larger spatial smoothing, which is a crucial confounding factor in DBM, VBM and related studies (Scarpazza et al., 2015). Evaluation of registration accuracy was crucially important for our study as the size and morphology of the brain changes signiﬁcantly from late fetal to newborn stage. Diﬀerent frameworks may oﬀer varying levels of precision in the registration process. In the next section, we provide a critical review and highlight the potential limitations of our methodology to select the best registration method for morphometry. Frontiers in Neuroscience Data availability statement The datasets presented in this article are not readily available because of restrictions in consent. Non-participant speciﬁc data supporting the conclusions of this article will be made available by the authors, without undue reservation. Code is available on GitHub: https://github.com/CST333/fet2neo_ asymmetry. Requests to access the datasets should be directed to andras.jakab@kispi.uzh.ch. 5 Conclusion In conclusion, our study complements existing neuroscientiﬁc studies on inter-hemispheric neuroanatomical asymmetry by providing evidence that morphological asymmetry is not static during late fetal development, and dynamic processes take place during late fetal development. Importantly, the neuroscientiﬁc value of our work is to characterize the dynamic patterns of asymmetry within-subject, therefore reducing some of the confounding eﬀects of inter-individual variability that is present in cross-sectional studies on human brain development. Our results underscore the importance of the perinatal period for brain development, encompassing signiﬁcant milestones such as tertiary gyri emergence and maturation, myelination, and cortical adaptation in response to external stimuli. The study also implies that a suitable image registration is critical when using DBM and that results might vary when using diﬀerent frameworks, which must be considered when interpreting results. In our study, we utilized image registrations in DBM, to infer inter-hemispheric asymmetry. However, this approach has some limitations with regard to the interpretability of our ﬁndings. We referred to our statistically signiﬁcant “clusters” as anatomical regions where the morphometric features of the brain diﬀer between homotopic regions of the two hemispheres. This approach is somewhat restrictive, as it relies on information contained within the JD maps to infer morphology. It is worth noting that the JD maps are based on image-based registration (and not surface or landmark-based), which aims to map visual features between a moving and target image. Consequently, the expansion or contraction of a brain region is only detected if there are features that can be matched and which expand or contract. In many brain regions, this is not the case during development. For instance, for the cortex, we primarily detected shrinkage (negative JD values) as the general tendency from fetal to neonatal age, which may be attributed to the increasing convolutional pattern (gyriﬁcation) with age after factoring out the majority of global brain scaling that occurred during the linear registration step. Furthermore, our non-linear registration methods are not likely to capture emerging structures such as two gyri appearing instead of one, or structures rotating, etc. As a result, while we employ a robust statistical approach and a dual-framework analysis, we cannot deﬁnitively conclude whether the asymmetries are driven by (asymmetric) isotropic growth, folding, rotation, or other types of morphometric diﬀerences between the two hemispheres. 4.3 Methodological limitations Our ﬁndings on inter-hemispheric brain asymmetry are limited by the following methodological factors. In framework A, the large registration from the fetal to neonatal timepoints is challenging, however the JD which describes morphometric changes within each participant already captures the longitudinal aspect of the data. It describes the change between the two timepoints and simply by ﬂipping the JD maps it is possible to perform longitudinal 10 frontiersin.org Steger et al. 10.3389/fnins.2023.1252850 10.3389/fnins.2023.1252850 substantiate our ﬁndings, we suggest that validation of our results using comparable datasets or a database comprising preterm newborns, where clinically indicated longitudinal MRI scans are available. asymmetry testing. In framework B, the asymmetry at each timepoint is captured by registering the ﬂipped to the non-ﬂipped brain. By registration to a shared template the asymmetry then can be tested for diﬀerences. The registration to a shared template could be seen as less challenging as the timeframe and therefore diﬀerence between participant and template might be smaller. However, it is important to consider that the accuracy of the neonatal to template registration is higher than that for the fetal to template registration. Another inherent aspect of this framework is that the generation of JD maps based on ﬂipped fetal to non-ﬂipped fetal and ﬂipped and non-ﬂipped neonatal brains might produce subtle systematic diﬀerences due to the diﬀerence in the images. Ethics statement A possible solution to these problems would be to perform landmark, sulcal, surface-based analyses (Lefèvre et al., 2016; Bozek et al., 2018; Lebenberg et al., 2018; Ahmad et al., 2019; Oishi et al., 2019), however, there the ﬁdelity and topological correctness of meshes and segmentation accuracy in fetal brain remains a critical challenge (de Dumast et al., 2022; Payette et al., 2023). A newer approach might be to learn registrations by deep learning (Wei et al., 2020), however, it is not yet clear how eﬃcient these networks are for fetal and neonatal development. Alternatively, a future approach could be to re-visit each region individually in each participant and provide an observer based manual annotations [such as in Kasprian et al. (2011)] to reveal concrete morphometric changes. The studies involving humans were approved by the Ethics Committee Zurich (Kantonale Ethikkommission Zürich). The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation in this study was provided by the participants’ legal guardians/next of kin. Frontiers in Neuroscience frontiersin.org References Ahmad, S., Wu, Z., Li, G., Wang, L., Lin, W., Yap, P., et al. (2019). Surface- constrained volumetric registration for the early developing brain. Med. Image Anal. 58:101540. doi: 10.1016/j.media.2019.101540 Dongen, S. (2006). Fluctuating asymmetry and developmental instability in evolutionary biology: Past, present and future. J. Evol. Biol. 19, 1727–1743. doi: 10. 1111/j.1420-9101.2006.01175.x Dubois, J., Hertz-Pannier, L., Cachia, A., Mangin, J. F., Le Bihan, D., and Dehaene- Lambertz, G. (2009). Structural asymmetries in the infant language and sensori-motor networks. Cerebral. Cortex 19, 414–423. Ami, O., Maran, J., Musset, D., Dubray, C., Mage, G., and Boyer, L. (2022). Using magnetic resonance imaging during childbirth to demonstrate fetal head moldability and brain compression: Prospective cohort study. JMIR Form. Res. 6:e27421. doi: 10.2196/27421 Fedorov, A., Beichel, R., Kalpathy-Cramer, J., Finet, J., Fillion-Robin, J., Pujol, S., et al. (2012). 3D Slicer as an image computing platform for the Quantitative Imaging Network. Magn. Reson. Imaging 30, 1323–1341. doi: 10.1016/j.mri.2012.05.001 Avants, B., Epstein, C., Grossman, M., and Gee, J. (2008). Symmetric diﬀeomorphic image registration with cross-correlation: Evaluating automated labeling of elderly and neurodegenerative brain. Med. Image Anal. 12, 26–41. doi: 10.1016/j.media.2007.06. 004 Galaburda, A., Sanides, F., and Geschwind, N. (1978). Human brain. Cytoarchitectonic left-right asymmetries in the temporal speech region. Arch. Neurol. 35, 812–817. doi: 10.1001/archneur.1978.00500360036007 Bozek, J., Makropoulos, A., Schuh, A., Fitzgibbon, S., Wright, R., Glasser, M., et al. (2018). Construction of a neonatal cortical surface atlas using multimodal surface matching in the developing human connectome project. Neuroimage 179, 11–29. doi: 10.1016/j.neuroimage.2018.06. 018 Gilmore, J., Lin, W., Prastawa, M., Looney, C., Vetsa, Y., Knickmeyer, R., et al. (2007). Regional gray matter growth, sexual dimorphism, and cerebral asymmetry in the neonatal brain. J. Neurosci. 27, 1255–1260. doi: 10.1523/jneurosci.3339-06. 2007 Cai, S., Zhang, G., Zhang, H., and Wang, J. (2020). Normative linear and volumetric biometric measurements of fetal brain development in magnetic resonance imaging. Childs Nerv. Syst. 36, 2997–3005. doi: 10.1007/s00381-020- 04633-3 Gorgolewski, K., Burns, C., Madison, C., Clark, D., Halchenko, Y., Waskom, M., et al. (2011). Nipype: A ﬂexible, lightweight and extensible neuroimaging data processing framework in python. Front. Neuroinform. 5:13. doi: 10.3389/fninf.2011. 00013 Chi, J., Dooling, E., and Gilles, F. (1977). Left-right asymmetries of the temporal speech areas of the human fetus. Arch. Neurol. 34, 346–348. doi: 10.1001/archneur. 1977.00500180040008 Guadalupe, T., Mathias, S., vanErp, T., Whelan, C., Zwiers, M., Abe, Y., et al. (2017). Author contributions GN, WK, AJ, and RT contributed to the conception and design of the studies. GN, WK, and AJ contributed to the funding acquisition. CS, AS, TN, NY, and SC recruited participants and collected the dataset. KP implemented the reconstruction tools. CM implemented the segmentation network. CS processed the dataset and performed The present study is subject to limitations due to the relatively small sample size (n = 20), limiting the statistical power of our analyses. Nevertheless, this challenge of achieving substantial sample sizes is present in many studies using fetal or neonatal MRI of healthy pregnancies. To 11 frontiersin.org Steger et al. 10.3389/fnins.2023.1252850 Acknowledgments We would like to thank all families that participated in the BrainCHD and BrainDNIU study and clinical staﬀ involved in the data collection. We especially thank the radiology and neonatology department of the University Children’s Hospital Zürich. A preprint version of this article was published on bioarchiv (https://doi.org/10.1101/2023.10.30.564508). Supplementary material The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fnins.2023. 1252850/full#supplementary-material Conﬂict of interest the entire analysis. CS and AJ wrote the ﬁrst draft of the manuscript. All authors contributed to manuscript revision and read and approved the submitted version. The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Publisher’s note This work was supported by the Swiss National Science Foundation (SNSF 320030_184932, PZOOP3_161146, and 200886), the Anna Mueller Grocholski Foundation, the EMDO Foundation, the Prof. Max Cloetta Foundation, and the Vontobel Foundation. All claims expressed in this article are solely those of the authors and do not necessarily represent those of their aﬃliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. References Neuroimage 17, 825–841. doi: 10.1016/s1053-8119(02)91132-8 Postema, M., van Rooij, D., Anagnostou, E., Arango, C., Auzias, G., Behrmann, M., et al. (2019). Altered structural brain asymmetry in autism spectrum disorder in a study of 54 datasets. Nat. Commun. 10:4958. doi: 10.1038/s41467-019-13005-8 Jenkinson, M., Beckmann, C., Behrens, T., Woolrich, M., and Smith, S. (2012). FSL. Neuroimage 62, 782–790. doi: 10.1016/j.neuroimage.2011.09.015 Rajagopalan, V., Scott, J., Habas, P., Kim, K., Corbett-Detig, J., Rousseau, F., et al. (2011). Local tissue growth patterns underlying normal fetal human brain gyriﬁcation quantiﬁed in utero. J. Neurosci. 31, 2878–2887. doi: 10.1523/JNEUROSCI.5458-10. 2011 Kasprian, G., Langs, G., Brugger, P., Bittner, M., Weber, M., Arantes, M., et al. (2011). The prenatal origin of hemispheric asymmetry: An in utero neuroimaging study. Cereb. Cortex 21, 1076–1083. doi: 10.1093/cercor/bhq179 Kienast, P., Schwartz, E., Diogo, M., Gruber, G., Brugger, P., Kiss, H., et al. (2021). The prenatal origins of human brain asymmetry: Lessons learned from a cohort of fetuses with body lateralization defects. Cereb. Cortex 31, 3713–3722. doi: 10.1093/ cercor/bhab042 Rajagopalan, V., Scott, J., Habas, P., Kim, K., Rousseau, F., Glenn, O., et al. (2012). Mapping directionality speciﬁc volume changes using tensor based morphometry: An application to the study of gyrogenesis and lateralization of the human fetal brain. Neuroimage 63, 947–958. doi: 10.1016/j.neuroimage.2012.03.092 Kong, X., Mathias, S. R., Guadalupe, T., Group Enigma Laterality Working, Glahn, D. C., Franke, B., et al. (2018). Mapping cortical brain asymmetry in 17,141 healthy individuals worldwide via the ENIGMA Consortium. Proc. Natl. Acad. Sci. U. S. A. 115, E5154–E5163. doi: 10.1073/pnas.1718418115 Ratnarajah, N., Rifkin-Graboi, A., Fortier, M., Chong, Y., Kwek, K., Saw, S., et al. (2013). Structural connectivity asymmetry in the neonatal brain. Neuroimage 75, 187–194. doi: 10.1016/j.neuroimage.2013.02.052 Rogers, L., Zucca, P., and Vallortigara, G. (2004). Advantages of having a lateralized brain. Proc. Biol. Sci. 271(Suppl. 6), S420–S422. doi: 10.1098/rsbl.2004. 0200 Kuklisova-Murgasova, M., Quaghebeur, G., Rutherford, M., Hajnal, J., and Schnabel, J. (2012). Reconstruction of fetal brain MRI with intensity matching and complete outlier removal. Med. Image Anal. 16, 1550–1564. doi: 10.1016/j.media.2012.07.004 Rousseau, F., Oubel, E., Pontabry, J., Schweitzer, M., Studholme, C., Koob, M., et al. (2013). BTK: An open-source toolkit for fetal brain MR image processing. Comput. Methods Programs Biomed. 109, 65–73. doi: 10.1016/j.cmpb.2012.08.007 Lau, J., Parrent, A., Demarco, J., Gupta, G., Kai, J., Stanley, O., et al. (2019). A framework for evaluating correspondence between brain images using anatomical ﬁducials. Hum. Brain Mapp. 40, 4163–4179. References doi: 10.1002/hbm.24693 Scarpazza, C., Tognin, S., Frisciata, S., Sartori, G., and Mechelli, A. (2015). False positive rates in Voxel-based Morphometry studies of the human brain: Should we be worried? Neurosci. Biobehav. Rev. 52, 49–55. doi: 10.1016/j.neubiorev.2015. 02.008 Lebenberg, J., Labit, M., Auzias, G., Mohlberg, H., Fischer, C., Rivière, D., et al. (2018). A framework based on sulcal constraints to align preterm, infant and adult human brain images acquired in vivo and post mortem. Brain Struct. Funct. 223, 4153–4168. doi: 10.1007/s00429-018-1735-9 Schmitz, J., Güntürkün, O., and Ocklenburg, S. (2019). Building an asymmetrical brain: The molecular perspective. Front. Psychol. 10:982. doi: 10.3389/fpsyg.2019. 00982 Lefèvre, J., Germanaud, D., Dubois, J., Rousseau, F., de Macedo Santos, I., Angleys, H., et al. (2016). Are developmental trajectories of cortical folding comparable between cross-sectional datasets of fetuses and preterm newborns? Cereb. Cortex 26, 3023– 3035. doi: 10.1093/cercor/bhv123 Smith, S., and Nichols, T. (2009). Threshold-free cluster enhancement: Addressing problems of smoothing, threshold dependence and localisation in cluster inference. Neuroimage 44, 83–98. doi: 10.1016/j.neuroimage.2008.03.061 Lehtola, S., Tuulari, J., Karlsson, L., Parkkola, R., Merisaari, H., Saunavaara, J., et al. (2019). Associations of age and sex with brain volumes and asymmetry in 2- 5-week-old infants. Brain Struct. Funct. 224, 501–513. doi: 10.1007/s00429-018-1787- x Sone, M., Koshiyama, D., Zhu, Y., Maikusa, N., Okada, N., Abe, O., et al. (2022). Structural brain abnormalities in schizophrenia patients with a history and presence of auditory verbal hallucination. Transl. Psychiatry 12:511. doi: 10.1038/s41398-022- 02282-5 Leroy, F., Cai, Q., Bogart, S., Dubois, J., Coulon, O., Monzalvo, K., et al. (2015). New human-speciﬁc brain landmark: The depth asymmetry of superior temporal sulcus. Proc. Natl. Acad. Sci. U. S. A. 112, 1208–1213. doi: 10.1073/pnas.1412389112 Specht, K., and Wigglesworth, P. (2018). The functional and structural asymmetries of the superior temporal sulcus. Scand. J. Psychol. 59, 74–82. doi: 10.1111/sjop.12410 Li, G., Nie, J., Wang, L., Shi, F., Lyall, A., Lin, W., et al. (2014). Mapping longitudinal hemispheric structural asymmetries of the human cerebral cortex from birth to 2 years of age. Cereb. Cortex 24, 1289–1300. doi: 10.1093/cercor/bhs413 Tustison, N., Cook, P., Holbrook, A., Johnson, H., Muschelli, J., Devenyi, G., et al. (2021). The ANTsX ecosystem for quantitative biological and medical imaging. Sci. Rep. 11:9068. doi: 10.1038/s41598-021-87564-6 Lubben, N., Ensink, E., Coetzee, G., and Labrie, V. (2021). The enigma and implications of brain hemispheric asymmetry in neurodegenerative diseases. Brain Commun. 3:fcab211. References Human subcortical brain asymmetries in 15,847 people worldwide reveal eﬀects of age and sex. Brain Imaging Behav. 11, 1497–1514. doi: 10.1007/s11682-016- 9629-z Cox, R. (1996). AFNI: Software for analysis and visualization of functional magnetic resonance neuroimages. Comput. Biomed. Res. 29, 162–173. doi: 10.1006/cbmr.1996. 0014 Güntürkün, O., Ströckens, F., and Ocklenburg, S. (2020). Brain lateralization: A comparative perspective. Physiol. Rev. 100, 1019–1063. doi: 10.1152/physrev.00006. 2019 de Dumast, P., Hamza, K., Vincent, D., Mériam, K., and Meritxell, B. C. (2022). Multi-dimensional topological loss for cortical plate segmentation in fetal brain MRI. ArXiv [Preprint]. Available online at: http://arxiv.org/abs/2208.07566 (accessed Auguest 16, 2022). Habas, P., Scott, J., Roosta, A., Rajagopalan, V., Kim, K., Rousseau, F., et al. (2012). Early folding patterns and asymmetries of the normal human brain detected from in utero MRI. Cereb. Cortex 22, 13–25. doi: 10.1093/cercor/bhr053 De Vareilles, H., Rivière, D., Pascucci, M., Sun, Z., Fischer, C., Leroy, F., et al. (2023). Exploring the emergence of morphological asymmetries around the brain’s Sylvian ﬁssure: A longitudinal study of shape variability in preterm infants. Cereb. Cortex 33, 6667–6680. doi: 10.1093/cercor/bhac533 Hill, J., Dierker, D., Neil, J., Inder, T., Knutsen, A., Harwell, J., et al. (2010). A surface- based analysis of hemispheric asymmetries and folding of cerebral cortex in term- born human infants. J. Neurosci. 30, 2268–2276. doi: 10.1523/JNEUROSCI.4682-09. 2010 Dean, D., Planalp, E., Wooten, W., Schmidt, C., Kecskemeti, S., Frye, C., et al. (2018). Investigation of brain structure in the 1-month infant. Brain Struct. Funct. 223, 1953–1970. doi: 10.1007/s00429-017-1600-2 Isensee, F., Jaeger, P., Kohl, S., Petersen, J., and Maier-Hein, K. (2021). nnU-Net: A self-conﬁguring method for deep learning-based biomedical image segmentation. Nat. Methods 18, 203–211. doi: 10.1038/s41592-020-01008-z Dice, L. R. (1945). Measures of the amount of ecologic association between Species. Ecology 26, 297–302. Jakab, A., Meuwly, E., Feldmann, M., Rhein, M., Kottke, R., O’Gorman Tuura, R., et al. (2019). Left temporal plane growth predicts language development in 12 Frontiers in Neuroscience frontiersin.org Steger et al. 10.3389/fnins.2023.1252850 10.3389/fnins.2023.1252850 newborns with congenital heart disease. Brain 142, 1270–1281. doi: 10.1093/brain/ awz067 Peyvandi, S., Lim, J., Marini, D., Xu, D., Reddy, V., Barkovich, A., et al. (2021). Fetal brain growth and risk of postnatal white matter injury in critical congenital heart disease. J. Thorac. Cardiovasc. Surg. 162, 1007–1014.e1. doi: 10.1016/j.jtcvs.2020.09. 096 Jenkinson, M., Bannister, P., Brady, M., and Smith, S. (2002). Improved optimization for the robust and accurate linear registration and motion correction of brain images. References doi: 10.1093/braincomms/fcab211 Vasung, L., Rollins, C., Yun, H., Velasco-Annis, C., Zhang, J., Wagstyl, K., et al. (2020). Quantitative In vivo MRI assessment of structural asymmetries and sexual dimorphism of transient fetal compartments in the human brain. Cereb. Cortex 30, 1752–1767. doi: 10.1093/cercor/bhz200 Lyttelton, O., Karama, S., Ad-Dab’bagh, Y., Zatorre, R., Carbonell, F., Worsley, K., et al. (2009). Positional and surface area asymmetry of the human cerebral cortex. Neuroimage 46, 895–903. doi: 10.1016/j.neuroimage.2009.03.063 Wada, J., and Davis, A. (1977). Fundamental nature of human infant’s brain asymmetry. Can. J. Neurol. Sci. 4, 203–207. doi: 10.1017/s031716710002 5233 Machado-Rivas, F., Gandhi, J., Choi, J., Velasco-Annis, C., Afacan, O., Warﬁeld, S., et al. (2022). Normal growth, sexual dimorphism, and lateral asymmetries at fetal brain MRI. Radiology 303, 162–170. doi: 10.1148/radiol.211222 Wada, J. A., Clarke, R., and Hamm, A. (1975). Cerebral hemispheric asymmetry in humans. Cortical speech zones in 100 adults and 100 infant brains. Archives. of Neurology. 32, 239–246. Makropoulos, A., Robinson, E., Schuh, A., Wright, R., Fitzgibbon, S., Bozek, J., et al. (2018). The developing human connectome project: A minimal processing pipeline for neonatal cortical surface reconstruction. Neuroimage 173, 88–112. doi: 10.1016/j. neuroimage.2018.01.054 Wan, B., Bayrak, ¸S, Xu, T., Schaare, H., Bethlehem, R., Bernhardt, B., et al. (2022). Heritability and cross-species comparisons of human cortical functional organization asymmetry. Elife 11:e77215. doi: 10.7554/eLife.77215 Mallela, A., Deng, H., Brisbin, A., Bush, A., and Goldschmidt, E. (2020). Sylvian ﬁssure development is linked to diﬀerential genetic expression in the pre-folded brain. Sci. Rep. 10:14489. doi: 10.1038/s41598-020-71535-4 Wei, D., Sahar, A., Yunzhi, H., Lei, M., Zhengwang, W., Gang, L., et al. (2020). An auto-context deformable registration network for infant brain MRI. ArXiv [Preprint]. Available online at: http://arxiv.org/abs/2005.09230 (accessed May 19, 2020). Ng, I., Bonthrone, A., Kelly, C., Cordero-Grande, L., Hughes, E., Price, A., et al. (2020). Investigating altered brain development in infants with congenital heart disease using tensor-based morphometry. Sci. Rep. 10:14909. Wiesenfarth, M., Reinke, A., Landman, B., Eisenmann, M., Saiz, L., Cardoso, M., et al. (2021). Methods and open-source toolkit for analyzing and visualizing challenge results. Sci. Rep. 11:2369. doi: 10.1038/s41598-021-82017-6 Oishi, K., Chang, L., and Huang, H. (2019). Baby brain atlases. Neuroimage 185, 865–880. doi: 10.1016/j.neuroimage.2018.04.003 Winkler, A., Ridgway, G., Webster, M., Smith, S., and Nichols, T. (2014). Permutation inference for the general linear model. Neuroimage 92, 381–397. doi: 10.1016/j.neuroimage.2014.01.060 Oishi, K., Mori, S., Donohue, P., Ernst, T., Anderson, L., Buchthal, S., et al. (2011). References Multi-contrast human neonatal brain atlas: Application to normal neonate development analysis. Neuroimage 56, 8–20. doi: 10.1016/j.neuroimage.2011.01.051 Yun, H., Lee, H., Lee, J., Tarui, T., Rollins, C., Ortinau, C., et al. (2022). Quantiﬁcation of sulcal emergence timing and its variability in early fetal life: Hemispheric asymmetry and sex diﬀerence. Neuroimage 263:119629. doi: 10.1016/j. neuroimage.2022.119629 Payette, K., de Dumast, P., Kebiri, H., Ezhov, I., Paetzold, J., Shit, S., et al. (2021). An automatic multi-tissue human fetal brain segmentation benchmark using the Fetal Tissue Annotation Dataset. Sci. Data 8:167. doi: 10.1038/s41597-021-00946-3 Zhang, Z., Hou, Z., Lin, X., Teng, G., Meng, H., Zang, F., et al. (2013). Development of the fetal cerebral cortex in the second trimester: Assessment with 7T postmortem MR imaging. AJNR Am. J. Neuroradiol. 34, 1462–1467. doi: 10.3174/ajnr .A3406 Payette, K., Li, H., de Dumast, P., Licandro, R., Ji, H., Siddiquee, M., et al. (2023). Fetal brain tissue annotation and segmentation challenge results. Med. Image Anal. 88:102833. doi: 10.1016/j.media.2023.102833 13 Frontiers in Neuroscience frontiersin.org
https://openalex.org/W2962715917	https://europepmc.org/articles/pmc6377270?pdf=render	English	null	In-memory computing on a photonic platform	Science advances	2,019	cc-by	10,896	1Department of Materials, University of Oxford, Parks Road, Oxford OX1 3PH, UK. 2IBM Research–Zurich, Säumerstrasse 4, 8803 Rüschlikon, Switzerland. 3Institute of Physics, University of Muenster, Heisenbergstr, 11, 48149 Muenster, Germany. 4Department of Engineering, University of Exeter, Exeter EX4 QF, UK. Present address: Department of Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, MA 02139, USA. †These authors contributed equally to this work. ‡Corresponding author. Email: ase@zurich.ibm.com (A.S.); harish.bhaskaran@ materials.ox.ac.uk (H.B.) S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E In-memory computing on a photonic platform Carlos Ríos1†, Nathan Youngblood1†, Zengguang Cheng1, Manuel Le Gallo2, Wolfram H. P. Pernice3, C. David Wright4, Abu Sebastian2‡, Harish Bhaskaran1‡ Collocated data processing and storage are the norm in biological computing systems such as the mammalian brain. As our ability to create better hardware improves, new computational paradigms are being explored beyond von Neumann architectures. Integrated photonic circuits are an attractive solution for on-chip computing which can leverage the increased speed and bandwidth potential of the optical domain, and importantly, remove the need for electro-optical conversions. Here we show that we can combine integrated optics with collocated data storage and processing to enable all-photonic in-memory computations. By employing nonvolatile photonic elements based on the phase-change material, Ge2Sb2Te5, we achieve direct scalar and matrix-vector multiplication, featuring a novel single-shot Write/Erase and a drift-free process. The output pulse, carrying the information of the light-matter interaction, is the result of the computation. Our all-optical approach is novel, easy to fabricate and operate, and sets the stage for development of entirely photonic computers. INTRODUCTION tion of systems of linear equations (22) and a host of other modern- day computational areas, such as machine learning [e.g., for sparse coding (17) and compressed sensing (20)] and deep learning [e.g., for forward and backward propagation in deep neural networks (23, 24)]. Our work thus represents a key milestone for optical processing in memory. Integrated photonics offers attractive solutions for using light to carry out computational tasks on a chip (1–6), and phase-change materials are emerging as functional materials of choice on photonic platforms (7–13). On-chip nonvolatile memories that can be written, erased, and accessed optically are rapidly bridging a gap toward all-photonic chip-scale information processing (13–15). However, breaking the processor-memory dichotomy would vastly transform the computing landscape by allowing processing directly on the memory elements— so-called in-memory computing. Electronic implementations of such systems, able to carry out complex tasks such as scalar multiplication, bulk-bitwise operations, correlation detection, and compressed sensing recovery, are now emerging (16–20). Photonic implementations of in- memory computing on an integrated photonic chip have the potential to further transform the computing landscape, by providing, ultimately, increased speeds and bandwidths that can come from working directly in the optical domain, leveraging both inherent wavelength division multiplexing capabilities and the rapid technological advances of the Si photonics “revolution” (21). Although integrated photonic mem- ories have been showcased in recent years (11–13), carrying out com- putational tasks in the same device that implements the memory function provides crucial challenges in terms of switching energy, speed, and single-shot programming and recovery. Here, we overcome such challenges and demonstrate the first instance of a photonic computational memory for direct scalar multiplication of two numbers. Specifically, we demonstrate the multiplication a × b = c, with a, b, c ∈[0, 1], using a single integrated photonic phase-change memory cell. Subsequently, we extend this idea to demonstrate matrix- vector operations using multiple phase-change cells. Note that, matrix- vector multiplication operations underpin key processing operations in the area of “big data” analytics and artificial intelligence (AI) (16). For example, such an architecture can be used for the efficient solu- C O M P U T E R S C I E N C E Copyright © 2019 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of Science. No claim to originalU.S.Government Works. Distributed under a Creative Commons Attribution License 4.0 (CC BY). In-memory computing on a photonic platform Carlos Ríos1†, Nathan Youngblood1†, Zengguang Cheng1, Manuel Le Gallo2, Wolfram H. P. Pernice3, C. David Wright4, Abu Sebastian2‡, Harish Bhaskaran1‡ In-memory computing on a photonic platform Carlos Ríos1†, Nathan Youngblood1†, Zengguang Cheng1, Manuel Le Gallo2, Wolfram H. P. Pernice3, C. David Wright4, Abu Sebastian2‡, Harish Bhaskaran1‡ Copyright © 2019 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of Science. No claim to originalU.S.Government Works. Distributed under a Creative Commons Attribution License 4.0 (CC BY). Ríos et al., Sci. Adv. 2019;5:eaau5759 15 February 2019 RESULTS To Erase (recrystallize), we used two approaches: a train of decreasing-energy pulses as used in (10, 13) and, as explained later in this paper, a single two-step pulse. The former allowed us to reach any of the lower transmission levels, while the latter allowed us to re- crystallize down to the baseline. While using the train of pulses, there are no restrictions in intralevel transitions, i.e., any level can be reached starting from any other and the number of levels is limited by the geo- metry of the GST cell, the available signal-to-noise ratio (SNR), and the inherent randomness associated with phase transition in GST (27). We first optimized the parameters of our devices, particularly to de- crease the energy consumption and to improve the speed of operation. For a chip with a 1-mm-long cell of GST, the pump pulse width was varied while keeping a constant power to reach different transmission levels, as shown in Fig. 2A. For pulses with widths longer than 45 ns, there is a saturation due to the finite size of our memory cell, in which no more amorphous material can be obtained without using higher powers (30, 31). Therefore, longer pulses represent a waste in energy as the GST will not further amorphize and could even be ablated. From these results, it can also be observed that 25-ns pulses induce a change equivalent to ~75% of the maximum achievable transmission. We found that for cells between 1 and 4 mm long, 25-ns pulses were enough to achieve clear transmission contrast among levels, limited only by the SNR of the measurement. We further reduced the energy consumption by using single-shot Erase pulses from any arbitrary level to the fully crystalline level (baseline). This was achieved by using the double-step pulse sequence, shown in Fig. 2B, using a total energy of approximately 917 pJ, of which only ~577 pJ was absorbed by GST (also known as switching energy). The double-step pulse is sufficient to heat the material over the crystallization temperature and of a duration (125 ns) sufficient to keep it above the glass transition temperature, but below the melting temperature, until full recrystallization occurs. This double- step Erase pulse replaces the multiple-pulse erase scheme in previous works [e.g., the 19 × 100–ns pulse scheme in Ríos et al. (13)] where each individual pulse induced partial recrystallization. RESULTS We used the photonic memory device with phase-change materials shown in Fig. 1A as the functional element to demonstrate scalar mul- tiplication of two numbers. To do so, we mapped the numbers to the power of an input pulse Pin and the transmittance T of the device, which is set by a Write pulse PWrite, as sketched in Fig. 1 (A to D). Our device relies on the near-field coupling between the propagating electromagnetic mode inside the waveguide and a phase-change material segment (cell) placed on top (of the waveguide) to absorb enough energy to crystallize (anneal over 150°C) or amorphize (melt-quench over ~600°C) (25). Because most phase-change materials have a nonnegligible imagi- nary refractive index in the visible and near-infrared wavelength range, light is attenuated in different amounts depending on the phase configuration of the material, which gives rise to differentiable transmitted signals, thus encoding information. The Write pulse PWrite modulates the transmittance state of the photonic memory device. In particular, we used pulse energiesEPWrite and EPin, such that EPWrite > ETh > EPin, where ETh is the threshold energy to partially amorphize the fully crystalline phase-change material. Pulses with energies below such threshold will not change the transmittance because the temperature reached by the GST is lower than the melting point, resulting in no amorphization. Moreover, the recrystallization process requires pulses with energies in the same range as those used to amorphize; therefore, pulses with EPin < ETh also do not have any effect on intermediate partial-amorphous states (13). As we demonstrate, this simple yet powerful architecture is a promising solution for optical information processing applications, given that light is attenuated without directly blocking the transmis- sion. In our experiments, we use the phase-change material alloy Ge2Sb2Te5 (GST) in combination with Si3N4/SiO2 photonic wave- guides at telecom wavelengths, as sketched in Fig. 1A (see Materials and Methods and the Supplementary Materials). To characterize our devices and condition our material for multilevel operation (i.e., the 1 of 9 Ríos et al., Sci. Adv. 2019;5:eaau5759 15 February 2019 S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E Fig. 1. GST-based photonic memory cell and the scalar multiplication concept. (A) Scheme of an on-chip bidirectional pump-probe to switch, read, and monitor the transmission in a GST-based nanophotonic memory device. RESULTS Write and Erase consisted of pulses at l = 1590 nm and the probe of a low-power CW signal at l = 1598 nm. LGST = 1 or 2 mm, W = 1.3 mm, and H = 165 nm (etched down from a 330-nm-thick Si3N4). (B) The pulse PWrite is used to switch the GST from a low to a high transmission level, thus defining the transmittance T of the device. (C) Scheme of the multiplication of two scalars a and b, codified in the device transmittance T and in the energy of the read pulse Pin. (D) The low-energy read pulse Pin, which propagates through without inducing phase change, is measured at the output with an amplitude modu- lated by the transmittance T. Fig. 1. GST-based photonic memory cell and the scalar multiplication concept. (A) Scheme of an on-chip bidirectional pump-probe to switch, read, and monitor the transmission in a GST-based nanophotonic memory device. Write and Erase consisted of pulses at l = 1590 nm and the probe of a low-power CW signal at l = 1598 nm. LGST = 1 or 2 mm, W = 1.3 mm, and H = 165 nm (etched down from a 330-nm-thick Si3N4). (B) The pulse PWrite is used to switch the GST from a low to a high transmission level, thus defining the transmittance T of the device. (C) Scheme of the multiplication of two scalars a and b, codified in the device transmittance T and in the energy of the read pulse Pin. (D) The low-energy read pulse Pin, which propagates through without inducing phase change, is measured at the output with an amplitude modu- lated by the transmittance T. ably more efficient than previous demonstrations of multiplication by sequential addition (15, 28, 29). process of programming the device to specific levels), we used the counter-propagating pump-probe measurement configuration we had previously utilized (13). Pulses were used to control the phase configuration of GST, while a low-power continuous-wave (CW) probe was used to monitor the transmission state. Multiple and non- volatile levels of transmission were reached, as a result of the mixture between amorphous and crystalline GST (26), by controlling the power of the pump pulse to Write (amorphize) up to any level of higher trans- mission. Ríos et al., Sci. Adv. 2019;5:eaau5759 15 February 2019 S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E (D) Multilevel conditioning of a 2-mm- long GST cell. Thirteen distinguishable levels are demonstrated and accessed randomly; the number of levels is limited only by the SNR and the confidence interval as shown in (E). (E) Error of the multilevel operation calculated from subtracting the measured transmission level from the programmed level in 600 transitions or switching events. The red line corresponds to a normal distribution fitting curve. The beginning of such saturation can be observed in Fig. 2C for pulse energies approaching 354 pJ. For this reason, only energies of up to 354 pJ were considered here. To study the error in reaching the programmed transmission level, we sent combinations of Write and Erase pulses to reach 10 different transmission levels in 20 consecutive sequences. This experiment was repeated three times. The results of subtracting the actual transmission level after a particular transition from the originally programmed level are plotted in the Fig. 2E. The programmed levels were taken as the average level after several back- and-forth switching during a conditioning process (13). With an SD of 0.35% (for the total change in transmission for DT), a confidence interval for each transmission level can be established so that each level is uniquely distinguished. This value is determined not only by SNR but also by the variation in power of the pump pulses due to fluctuations in the electro-optical conversion. for example, in Ríos et al. (13)]; the total energy used during the Erase process was 9.5 nJ, and the duration was 3.8 ms, figures that are im- proved upon by factors of 10 and 25, respectively, in the current work. In terms of operational speed for a Write/Erase cycle, it can be observed in Fig. 2B that approximately 200 ns (including the pulse duration) is required to obtain a stable transmission level for the Write pulse and ~600 ns for the Erase pulse. However, this time varies with the pulse width: The shorter the pulse, the lower the time for the material to cool down, and for the refractive index to stabilize, the timescale of which is governed by the thermo-optical effect (32). Further im- provements in terms of speed and energy can be achieved by decreasing the pulse width to even shorter times, especially the Write pulse, which could be a picosecond and even femtosecond pulse (10, 33, 34). S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E ig. 2. Device optimization. (A) Write pulses with varying widths achieve different transmission levels in a 1-mm-long GST cell before saturating at around 45 ns and heir equivalent pulse energy. Shorter pulses can be chosen, thus saving energy and time during memory operations. DT = (T −Tmin)/Tmin is the change in transmission f the level T with respect to the baseline Tmin. (B) Amorphization to a higher transmission level is achieved with a single square Write pulse (13), while a new Erase ouble-step pulse is used to fully recrystallize, i.e., bring the transmission to the baseline. The latter pulse, at the input, consisted of 14.1 mW (or peak power) for 25 ns nd 5.64 mW (0.4 times the peak power) for 100 ns, where peak power is equal to the power used to reach the maximum transmission level. (C) Write pulse energies sed for the multilevel conditioning shown in (D) with a linear response by the GST memory cell beyond the threshold energy. (D) Multilevel conditioning of a 2-mm- ong GST cell. Thirteen distinguishable levels are demonstrated and accessed randomly; the number of levels is limited only by the SNR and the confidence interval as hown in (E). (E) Error of the multilevel operation calculated from subtracting the measured transmission level from the programmed level in 600 transitions or witching events. The red line corresponds to a normal distribution fitting curve. Fig. 2. Device optimization. (A) Write pulses with varying widths achieve different transmission levels in a 1-mm-long GST cell before saturating at around 45 ns and their equivalent pulse energy. Shorter pulses can be chosen, thus saving energy and time during memory operations. DT = (T −Tmin)/Tmin is the change in transmission of the level T with respect to the baseline Tmin. (B) Amorphization to a higher transmission level is achieved with a single square Write pulse (13), while a new Erase double-step pulse is used to fully recrystallize, i.e., bring the transmission to the baseline. The latter pulse, at the input, consisted of 14.1 mW (or peak power) for 25 ns and 5.64 mW (0.4 times the peak power) for 100 ns, where peak power is equal to the power used to reach the maximum transmission level. (C) Write pulse energies used for the multilevel conditioning shown in (D) with a linear response by the GST memory cell beyond the threshold energy. RESULTS The double-step pulse approach results in a considerable reduction in energy con- sumption and increase in overall speed during the Erase process [again, Figure 1B shows the Write pulse, PWrite, used to program a spe- cific level of transmittance T of the device, which relies on the multilevel conditioning of the material. The portion of the PWrite pulse that is transmitted was not taken into account (in the calculation of the result of the multiplication). A second low-energy read pulse Pin propagates through the photonic memory device, experiencing a transmittance given by the current phase configuration of the GST cell (as conditioned by the pulse PWrite), but does not induce any change to the material. The power of the pulse of Pin at the output port, Pout = T(PWrite) × Pin, is the result of the multiplication a × b by mapping the multiplicand a to T and the multiplier b to Pin, as shown schematically in Fig. 1C and in more detail in Fig. 1D. Such a direct multiplication process is consider- 2 of 9 Ríos et al., Sci. Adv. 2019;5:eaau5759 15 February 2019 S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E Ríos et al., Sci. Adv. 2019;5:eaau5759 15 February 2019 S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E This would also reduce the “dead” time between the two pulses used to per- form the scalar multiplication (see below). For the proper execution of scalar multiplication operations using multilevel memory states, it will also be important for such states to be stable over time. Thus, we show in Fig. 3 the optical transmission traces measured for a duration of over 3 × 104 s (8.8 hours), with the aim of studying the temporal variations of programmed levels (drift). Transmission levels in an arbitrary order were written and erased in two 2-mm-long GST memory cells. Subsequently, the devices were kept in an intermediate transmission level for a prolonged period of time, one having a 0.1-mW probe ON and the other OFF, as shown in Fig. 3 (A and B), respectively. Mechanical drift of the sample stage Since the ability to program our integrated phase-change photonic cells into multilevel states is key to the implementation of direct scalar multiplication, we demonstrate in Fig. 2 (C and D) multilevel behavior in a 2-mm-long GST cell using single-shot Write and Erase with the same pulse parameters of Fig. 2B. Thirteen transmission levels were programmed, limited only by the noise of the photodetector (see the Supplementary Materials). We found a linear response of the atten- uation caused by the GST cell as the pulse energy is increased beyond the threshold energy. For large energies, the linearity is lost due to saturation, which implies that the GST is in its most amorphous state. 3 of 9 Ríos et al., Sci. Adv. 2019;5:eaau5759 15 February 2019 S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E levels in our optical memory devices exhibit such excellent temporal stability is not fully understood and needs further investigation. How- ever, the temporal stability can no doubt be partially attributed to the fact that transmission is dominated by the more stable crystalline phase [even the less absorptive state of our memory cell is composed of an amorphous area embedded within crystalline material (31)]. was observed because of the relaxation of the picomotors over time; however, once the chip was placed in a stable position, the transmission level remained constant for the case in which the probe was kept ON. S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E Therefore, we do not find measurable drift for up to 104 s of constant measurement. This drift-free process represents a big advantage for photonic computational memory over its electrical analog, which un- dergoes notable resistance drift over time, thus preventing (or at least making it very difficult) to achieve and maintain reliable levels (35–38). The electrical resistance is sensitive to the Fermi level and trap states, which, for a highly stressed amorphous GST after a quick melt quenching, are sensitive to the structural relaxation that takes place over time (38). There is also some experimental evidence that the refractive index and the bandgap of amorphous phase-change material are also influenced by structural relaxation (37, 39). The crystalline state, on the other hand, is stable and, thus, has a small or negligible resistance drift (35). To date, the reason the programmed Moreover, the same specific transmission levels were retrieved after the measurement; hence, the multilevel conditioning is also preserved over time. On the other hand, for the case when the probe was turned OFF, a drift of nearly 9% was observed once the probe was turned on again. This drift is due to the relaxation of the material when the probe is removed, given that the probe itself heats up the material to a constant temperature to cause a thermo-optical effect that modifies the values of the complex refractive index without crystallizing (32). However, once the probe was turned ON, the drift is easily corrected for by sending a write pulse with the same energy as that required to reach the originally Drift in transmission. (A) Multilevel operation of a 2-mm-long GST memory cell using 50-ns pulses with energies in the range of 350 to 600 pJ to ons) and a train of power-decreasing pulses to Erase (downward transitions) level (13). The blue highlighted area corresponds to ~8.5 hours of c nt with a CW probe of 0.1 mW (inside the waveguide). (B) Same as (A), but turning the probe OFF for around 1.5 hours. A drift is observed betw which is corrected by sending the pulse energy of the level where the memory was originally set. (C) Multilevel operation of a different 2-mm-long ng 25-ns pulses with energies in the range of 200 to 354 pJ (see Fig. 2D). The Write and Erase were done in the same way as for (A) and (B). S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E As result, we are able to reliably retrieve both the original level and any other level, demonstrating that the multilevel conditioning was preserved. We were also able to avoid the drift entirely (i.e., whether the probe remains ON or OFF) by simply decreasing the probe power to 0.05 mW—half of its previous value. In this case, the material relaxation after the probe is turned OFF and then ON is negligible, given that the temperature is lower at the GST, as shown in Fig. 3C. We now demonstrate the multiplication a × b = c, with a, b, c ∈[0, 1]. We used a photonic memory device with a 2-mm-long GST cell and the single-shot Write and Erase scheme. We mapped the multiplicand a to the transmittance of the device, T = DT + T0, where DT corresponds to the linear response of the change in transmission as function of the PWrite pulse (shown in Fig. 2C), and T0 is the baseline transmission level (fully crystalline). Subsequently, the multiplier b is mapped to the energy of a second pulse Pin. The result of the multiplication is calculated from the output of this latter pulse, which is equivalent to Pout = T × Pin. Wegenerated both PWrite and Pin in the same manner by tuning the pulse power at the electro-optical modulator (see Materials Weuse energiesforPWritein the [180pJ,354pJ] range orequivalently, DT ∈[0, 0.143] (see Fig. 2C), and energies for Pin in the [0 pJ, 112.8 pJ] range. In Fig. 4B, we show the experimental realization of three multi- plications and the actual output pulses, Pout. In particular, we used the minimum and the maximum of the multiplicand: DTmin = 0 (i.e., a = 0) and DTmax = 0.143(i.e., a = 1), and the maximum (P1 in) and an intermediate (P2 in) energy values for the multiplier, 112.8 pJ (i.e., b = 1) and 38.9 pJ (b = 0.4), respectively. In this figure, it can be observed that the output powers have distinct energies, which can be measured Fig. 4. Scalar multiplication. (A) Read pulse Pin energies and corresponding measured Pout energy (past the crystalline GST cell and a grating coupler). All the mea- surements were carried out with the memory cell in the transmission baseline (i.e., DT = 0) of the multilevel conditioning shown in Fig. 2C. S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E After each switching, an Erase pulse was used to bring the device to the baseline before the following pulse. A projection of the values for Pout for the case DT = 0.143 is plotted for comparison. (B) Demonstration of three multiplications by applying two different Pin, P1 in = 112.8 pJ (~4.51 mW for 25 ns) and P2 in = 38.9 pJ (~1.56 mW for 25 ns), to memories programmed in the maximum and minimum transmittance levels, DT = 0.143 and DT = 0, respectively. The result when multiplying by T = 0 corresponds to the level- specific offset. The short time delay between pulses is due to the difference in optical path between the reference pulse and the pulse that is coupled into the photonic chip, both obtained from the pump pulse using a 90/10 beam splitter. (C) Realization of 429, c = a × b multiplications equivalent to 13 different, equally spaced, values for T (reached with PWrite ∈[180 pJ, 354 pJ]), thus creating 13 values for the multiplicand a, and 33 different values for Pin ∈[0 pJ, 112.8 pJ] for 33 scalars corresponding to b (note: a, b, c ∈[0, 1]). PWrite was sent first to establish a level followed by the 33 Pin pulses before changing levels again. (D) Error of the scalar multiplication calculated from subtracting the measured value from the exact. The red line corresponds to a normal distribution fit whose mean and SD values are shown on the plot. The inset corresponds to the error as function of the values of a and b; this shows how for values closer to 1, the error grows linearly, which, in turn, explains the spread for values close to 1 in (C). Fig. 4. Scalar multiplication. (A) Read pulse Pin energies and corresponding measured Pout energy (past the crystalline GST cell and a grating coupler). All the mea- surements were carried out with the memory cell in the transmission baseline (i.e., DT = 0) of the multilevel conditioning shown in Fig. 2C. After each switching, an Erase pulse was used to bring the device to the baseline before the following pulse. A projection of the values for Pout for the case DT = 0.143 is plotted for comparison. S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E The pro a time of ~2 hours, but in this case, the CW probe power was reduced to 0.05 mW, which is enough to avoid drift. The colored traces are adde on of the levels. Fig. 3. Drift in transmission. (A) Multilevel operation of a 2-mm-long GST memory cell using 50-ns pulses with energies in the range of 350 to 600 pJ to Write (upward transitions) and a train of power-decreasing pulses to Erase (downward transitions) level (13). The blue highlighted area corresponds to ~8.5 hours of constant mea- surement with a CW probe of 0.1 mW (inside the waveguide). (B) Same as (A), but turning the probe OFF for around 1.5 hours. A drift is observed between 4600 and 4800 s, which is corrected by sending the pulse energy of the level where the memory was originally set. (C) Multilevel operation of a different 2-mm-long GST memory cell using 25-ns pulses with energies in the range of 200 to 354 pJ (see Fig. 2D). The Write and Erase were done in the same way as for (A) and (B). The probe was turned off for a time of ~2 hours, but in this case, the CW probe power was reduced to 0.05 mW, which is enough to avoid drift. The colored traces are added to track the evolution of the levels. 4 of 9 Ríos et al., Sci. Adv. 2019;5:eaau5759 15 February 2019 S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E S C I E N C E A D V A N C E S and Methods). Figure 4A shows the read pulse energies used to map the multiplier and the corresponding output pulse energy after pro- pagating past the fully crystalline GST cell and through one grating coupler. The resulting relation shows that the response of the material (i.e., its transmittance) is linear for all the Pin energies. This result, together with the linear changes in transmittance as function of PWrite, demonstrated in Fig. 2C, is highly enabling because the mapping of both scalar numbers does not rely on fitting functions, which is the case of in-memory computing with electrical PCMs following a “pseudo” Ohm’s law (16). programmed level (i.e., the level set immediately prior to the probe be- ing turned off). S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E G11 G12 ½ P1 P3 P5 P2 P4 P6 ¼ 1 2 P1G11 þ P2G12 P3G11 þ P4G12 P5G11 þ P6G12 2 4 3 5 T where the factor of ½ is due to the 50% loss at the waveguide splitter. This approach will be a time-consuming serial process if the same wave- lengths for P1and P2, P3 and P4, etc., are used. A distinct advantage of photonics over the electrical domain is the ability to wavelength multiplex on-chip, so it is possible to send matrix elements P1 through P6 simultaneously and separate the respective output vector elements via wavelength filtering before summing on multiple photodetectors. In our memory cell, the nonvolatility of the transmission levels implies that the multiplicand is fixed until the next pulse excitation changes its state. This property, in an architecture comprising arrays of scalar multiplying units, could be exploited for efficient calculation of matrix-vector multiplications. To demonstrate that this type of matrix-vector multiplication is well suited to our photonic framework, we propose an architecture to implement matrix-vector multiplication using two memory cells in parallel. In such an architecture, shown in Fig. 5A, the output signals containing the result of each multiplication— carriedoutintwodifferentGSTcells—are combinedusingabeamsplit- ter. Initially, the transmittance values equivalent to two numbers, G11 and G12, are preprogrammed using two write pulses of power P1 Write and P2 Write. Given that these values are fixed, the two numbers together represent a 1 × 2 matrix [G11 G12], which can be switched to any other levelat agiven time, asshown in Fig. 5B. To multiply this matrix by a 2× 1 vector, pairs of input pulses representing any two numbers, P1 and P2 for instance, are sent simultaneously to the left and right grating cou- plers of the device shown in Fig. 5A. After propagating through the two respective GST memory cells, the pulses are weighted by the transmit- tance function and, therefore, carry the information of each individual multiplication (i.e., P1 × G11 and P2 × G12). Figure 5C shows the exper- imental demonstration for three pairs of numbers (input pulses), which all together represent three multiplications between a constant matrix and a vector with varying elements, before and after the multiplication by the GST transmittance. Each pair of pulses is then combined using a beam splitter and summed via a photodetector with a linear response. S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E (B) Demonstration of three multiplications by applying two different Pin, P1 in = 112.8 pJ (~4.51 mW for 25 ns) and P2 in = 38.9 pJ (~1.56 mW for 25 ns), to memories programmed in the maximum and minimum transmittance levels, DT = 0.143 and DT = 0, respectively. The result when multiplying by T = 0 corresponds to the level- specific offset. The short time delay between pulses is due to the difference in optical path between the reference pulse and the pulse that is coupled into the photonic chip, both obtained from the pump pulse using a 90/10 beam splitter. (C) Realization of 429, c = a × b multiplications equivalent to 13 different, equally spaced, values for T (reached with PWrite ∈[180 pJ, 354 pJ]), thus creating 13 values for the multiplicand a, and 33 different values for Pin ∈[0 pJ, 112.8 pJ] for 33 scalars corresponding to b (note: a, b, c ∈[0, 1]). PWrite was sent first to establish a level followed by the 33 Pin pulses before changing levels again. (D) Error of the scalar multiplication calculated from subtracting the measured value from the exact. The red line corresponds to a normal distribution fit whose mean and SD values are shown on the plot. The inset corresponds to the error as function of the values of a and b; this shows how for values closer to 1, the error grows linearly, which, in turn, explains the spread for values close to 1 in (C). 5 of 9 Ríos et al., Sci. Adv. 2019;5:eaau5759 15 February 2019 S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E S C I E N C E A D V A N C E S the measured and the expected values, which are calculated from the GST transmittance and the measured input pulse powers, with results normalized for simplicity. This process can be generalized to a multipli- cation between a k × N matrix and N × 1 vector using multiple cascaded power splitters (see fig. S6). S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E This signal, Pout, is the linear combination of the transmitted P1 and P2, provided they are optical pulses with unique wavelengths to prevent un- wanted interference. The resulting scalar number is the matrix-vector product, whose outputs are shown in Fig. 5D for two different matrices (i.e., [G11 G12] and [G′11 G′12]). Good agreement is observed between S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E Using such methods, an input matrix A could be mapped to GST nanophotonic memory cells, and the solution x of Ax = b with A∈RNN and b; x ∈RN can be computed from iterative algorithms, as demonstrated for electrically switchable phase- change memory devices (16). Moreover, this approach can be extended to a matrix-matrix multiplication by sequentially sending the column values of the second matrix as input vectors. The implementation of this approach can already be visualized from the results of Fig. 5, in which combining the results of the three input vectors, the following matrix- matrix multiplication can be reconstructed and then rescaled to obtain the result of the operations: 1 × 0 = 0, 1 × 1 = 1, and 1 × 0.4 = 0.4. Note that the lowest transmission level of the three, which in theory corresponds to 0.4, is actually smaller than the level for 0. This is due to a combination of two reasons: (i) to calculate the former, we used the relatively low power pulseP2 in, whereas for the latter we use the higher power pulseP1 in, and (ii) even in the fully crys- talline state, GST will not absorb all the light propagating down the waveguide [unless a much longer GST cell is used to substantially in- crease the optical attenuation (12)]. Therefore, there is an offset given by the transmission baseline T0, which has to be subtracted from every multiplication to enable exact linear rescaling to the [0, 1] results (see the Supplementary Materials). In Fig. 4C, we demonstrate 429 multi- plications choosing arbitrary values for a and b with the associated error (difference between the exact and the measured value of c) shown in Fig. 4D. We found good agreement between the exact and the measured value of the multiplication, having an error that spreads as a and b get larger, as shown in the inset of Fig. 4D (see the Supplemen- tary Materials). The exact value was calculated from the linear fits in the characterization of the device and the subsequent mapping to [0, 1] for both multiplicand and multiplier. The measured value corresponds to the average of the output pulse energy, correcting the offset and normal- izing to [0, 1]. While the results of the multiplication are not exact due to factors such as the fluctuations in the values of T, as shown above in Fig. S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E 2E, this kind of multiplication operation has proved useful in application areas such as machine learning (17). Moreover, in application domains where arbitrarily high accuracy is required, ideas such as mixed- precision computing can be used where the low precision multiply unit is used in conjunction with a high precision unit (16). and then rescaled to obtain the result of the operations: 1 × 0 = 0, 1 × 1 = 1, and 1 × 0.4 = 0.4. Note that the lowest transmission level of the three, which in theory corresponds to 0.4, is actually smaller than the level for 0. This is due to a combination of two reasons: (i) to calculate the former, we used the relatively low power pulseP2 in, whereas for the latter we use the higher power pulseP1 in, and (ii) even in the fully crys- talline state, GST will not absorb all the light propagating down the waveguide [unless a much longer GST cell is used to substantially in- crease the optical attenuation (12)]. Therefore, there is an offset given by the transmission baseline T0, which has to be subtracted from every multiplication to enable exact linear rescaling to the [0, 1] results (see the Supplementary Materials). In Fig. 4C, we demonstrate 429 multi- plications choosing arbitrary values for a and b with the associated error (difference between the exact and the measured value of c) shown in Fig. 4D. We found good agreement between the exact and the measured value of the multiplication, having an error that spreads as a and b get larger, as shown in the inset of Fig. 4D (see the Supplemen- tary Materials). The exact value was calculated from the linear fits in the characterization of the device and the subsequent mapping to [0, 1] for both multiplicand and multiplier. The measured value corresponds to the average of the output pulse energy, correcting the offset and normal- izing to [0, 1]. While the results of the multiplication are not exact due to factors such as the fluctuations in the values of T, as shown above in Fig. 2E, this kind of multiplication operation has proved useful in application areas such as machine learning (17). Moreover, in application domains where arbitrarily high accuracy is required, ideas such as mixed- precision computing can be used where the low precision multiply unit is used in conjunction with a high precision unit (16). S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E The different wavelengths in the two arms (1590 and 1598 nm) and linear response etector ensure that the two multiplied pulses are added together. A factor of ½ is due to the 50% loss at the waveguide splitter. E N C E A D V A N C E S \| R E S E A R C H A R T I C L E Fig. 5. Photonic architecture for matrix-vector multiplication. (A) An optical image of the device used to implement multiplication between a 1 × 2 matrix and a 2 × 1vector is shown on the left. The matrixelementsare preprogrammedin the GST(G11and G12) using two write pulses asdiscussedin the main text. To implementmultiple vectors, pairs of Pin pulses (P1and P2, P3 and P4, etc.) are sent simultaneously and recorded as Pout (P1G11 + P2G12, P3G11 + P4G12, etc.). (B) Change in transmission of the two GST cells as recorded by the probe. Highlighted blue region indicates when PWrite pulses were used to switch one or more GST cells, thus reprogramming the matrix. (C) Recorded traces of input pulses before and after the GST cells. Transmitted power of the optical pulses (dashed lines) is the product of the input pulse and GST’s transmission. (D) Calculated and measured output pulses after combining pulses from the two arms on a photodetector. The different wavelengths in the two arms (1590 and 1598 nm) and linear response of the photodetector ensure that the two multiplied pulses are added together. A factor of ½ is due to the 50% loss at the waveguide splitter. and current exist). This eliminates the need for nonlinear fitting functions and additional postprocessing [see, e.g., Le Gallo et al. (16)] that are needed when implementing multiplication electronically. To summarize, we have demonstrated direct multiplication of scalar numbers using photonic in-memory computing. This is achieved by using the distinct interaction of two pulses, each of which represents a number to be multiplied—one with energy above the switching threshold to induce structural changes in the phase-change material, and another one below—in an integrated photonic waveguide. We also extended this concept to matrix-vector multiplication operations that form a keycomputationalprimitiveforthevastmajorityofAIalgorithms. These resultsconfirmthepotentialofphase-change materialsin photon- ic hardware computational paradigms—including the ability to perform computations with memory in the same physical location using light. DISCUSSION h Using the new mechanism introduced in this paper, we have shown that single sub-nanojoule energy pulses are sufficient for both Write and Erase operations in our integrated phase-change photonic devices. However, the energy consumption associated with programming matrix elements into the phase-change cells remains relatively high. It shouldbe notedthat in applications where the same matrix elementsare repeatedly used, the programming step needs to be done only once, given the nonvolatile nature of information storage in the photonic memory device. In several applications such as iterative solution of linear equa- tions, compressed sensing, and deep learning inference, the matrix elements remain fixed. The entries of the vector that is used to multiply with the fixed matrix, on the other hand, are codified into the energy of the read pulses. Hence, read pulses of energies on the order of femto- joule (pulse width of tens of picoseconds and powers of tens-hundreds of microwatts, for instance) can be used as long as the resulting output pulse can be resolved by the detector. Moreover, the collocation of pro- cessing and storage means that there is little energy expended on data transfer. Hence, the energy consumption associated with the matrix- vector multiply operation can be substantially low. The energy consumption associated with our approach is compara- ble to its electronic computational memory analogs (16), but there are severalkeyadvantagesassociated withphotonic in-memorycomputing. The most notable one is the ability to process transmitted signals op- tically without having to use highly inefficient electro-optical con- verters. The experimental results presented in this paper also point to other important advantages as follows: (i) The output pulse varies linearly with the input (read) pulse, un- like in electrical phase-change devices (where nonlinear relationships between resistance level and excitation voltage and between voltage 6 of 9 Ríos et al., Sci. Adv. 2019;5:eaau5759 15 February 2019 S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E hotonic architecture for matrix-vector multiplication. (A) An optical image of the device used to implement multiplication between a 1 × 2 matrix and shown on the left. The matrixelementsare preprogrammedin the GST(G11and G12) using two write pulses asdiscussedin the main text. To implementmultiple ve n pulses (P1and P2, P3 and P4, etc.) are sent simultaneously and recorded as Pout (P1G11 + P2G12, P3G11 + P4G12, etc.). (B) Change in transmission of the two GST c by the probe. Highlighted blue region indicates when PWrite pulses were used to switch one or more GST cells, thus reprogramming the matrix. (C) Recorded tra ses before and after the GST cells. Transmitted power of the optical pulses (dashed lines) is the product of the input pulse and GST’s transmission. (D) Calculate output pulses after combining pulses from the two arms on a photodetector. The different wavelengths in the two arms (1590 and 1598 nm) and linear response ector ensure that the two multiplied pulses are added together. A factor of ½ is due to the 50% loss at the waveguide splitter. N C E A D V A N C E S \| R E S E A R C H A R T I C L E Photonic architecture for matrix-vector multiplication. (A) An optical image of the device used to implement multiplication between a 1 × 2 matrix and r is shown on the left. The matrixelementsare preprogrammedin the GST(G11and G12) using two write pulses asdiscussedin the main text. To implementmultiple v Pin pulses (P1and P2, P3 and P4, etc.) are sent simultaneously and recorded as Pout (P1G11 + P2G12, P3G11 + P4G12, etc.). (B) Change in transmission of the two GST c ed by the probe. Highlighted blue region indicates when PWrite pulses were used to switch one or more GST cells, thus reprogramming the matrix. (C) Recorded tra ulses before and after the GST cells. Transmitted power of the optical pulses (dashed lines) is the product of the input pulse and GST’s transmission. (D) Calculate ed output pulses after combining pulses from the two arms on a photodetector. Measurement setup 7. M. Wuttig, H. Bhaskaran, T. Taubner, Phase-change materials for non-volatile photonic applications. Nat. Photonics 11, 465–476 (2017). The pump-probe experimental setup in fig. S2 was used to carry out the multilevel, time-resolved, and scalar multiplication measurements. The setup was then extended to that in fig. S3 to carry out the matrix-vector multiplications within balanced splitters with two GST memory cells. In both setups, the optical signals were confined to the photonic circuit, thatis, Write,Erase, read-out, and multiplicationswereallrealizedwith- in the integrated chip. To avoid interference, two different C + L CW tunable laser sources were used; wavelengths of 1598 nm (TSL-550, Santec) and 1590 nm (N7711A, Keysight) were chosen for the probe and pump, respectively. The pump pulses—as well as the multiplicand pulse Pin—were subsequently generated with an electro-optical mod- ulation (Lucent Technologies, 2623NA), which was controlled by a 100-MHz electrical pulse generator (AFG 3102C, Tektronix). The pulse was further power amplified by a low-noise erbium-doped fiber ampli- fier (AEDFA-CL-23, Amonics). Both the pump pulses and the probe were coupled into the photonic device using integrated grating couplers with transmission peak at 1598 nm and coupling efficiency of ~20%. The counter-propagating scheme was used to ease the separation of the signals, and tunable optical filters (OTF-320, Santec) were intro- duced to the optical lines to further suppress noise resulting from reflec- tions. At the probe output of the device, the CW signal was divided into two beams using a 90/10 beam splitter to measure the time-resolved and the long-term transmission with a 200-kHz low-noise photoreceiver (NewFocus, 2011) and a 125 MHz photodetector (NewFocus, 1811), respectively. At the other output, the transmitted pulses were monitored using a 1-GHz photodetector (NewFocus, 1611). 8. Y. Ikuma, Y. Shoji, M. Kuwahara, X. Wang, K. Kintaka, H. Kawashima, D. Tanaka, H. Tsuda, Small-sized optical gate switch using Ge2Sb2Te5 phase-change material integrated with silicon waveguide. Electron. Lett. 46, 368–369 (2010). with silicon waveguide. Electron. Lett. 46, 368–369 (2010). 9. K. Kato, M. Kuwahara, H. Kawashima, T. Tsuruoka, H. Tsuda, Current-driven phase- change optical gate switch using indium-tin-oxide heater. Appl. Phys. Express. 10, 072201 (2017). 10. M. Stegmaier, C. Ríos, H. Bhaskaran, C. D. Wright, W. H. P. Pernice, Nonvolatile all-optical 1 × 2 switch for chipscale photonic networks. Adv. Opt. Mater. 5, 1600346 (2017). 11. Y. Lu, M. Stegmaier, P. Nukala, M. A. Giambra, S. Ferrari, A. Busacca, W. H. P. Pernice, R. S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E While there is plenty of room for improvement, these results are at the vanguard of collocated memory and processing on a photonic platform. p g p y (ii) There is no drift of any programmed transmission level in our approach (within the test parameters presented in this paper, including continuous measurements over periods extending to 8.5 hours). This compares favorably to electrical phase-change devices, where resistance drift necessitates the use of additional processing and hardware over- heads (e.g., periodic refreshing of values or the use of calibration cells). (iii) We are able to set levels deterministically, a process that is challenging and has been the subject of much effort in electronic phase-change memories [which use iterative level-setting schemes— see, e.g., Papandreou et al. (40)]. (ii) There is no drift of any programmed transmission level in our approach (within the test parameters presented in this paper, including continuous measurements over periods extending to 8.5 hours). This compares favorably to electrical phase-change devices, where resistance drift necessitates the use of additional processing and hardware over- heads (e.g., periodic refreshing of values or the use of calibration cells). (iii) We are able to set levels deterministically, a process that is challenging and has been the subject of much effort in electronic phase-change memories [which use iterative level-setting schemes— see, e.g., Papandreou et al. (40)]. (ii) There is no drift of any programmed transmission level in our approach (within the test parameters presented in this paper, including continuous measurements over periods extending to 8.5 hours). This compares favorably to electrical phase-change devices, where resistance drift necessitates the use of additional processing and hardware over- heads (e.g., periodic refreshing of values or the use of calibration cells). (iii) We are able to set levels deterministically, a process that is challenging and has been the subject of much effort in electronic phase-change memories [which use iterative level-setting schemes— see, e.g., Papandreou et al. (40)]. (iii) We are able to set levels deterministically, a process that is challenging and has been the subject of much effort in electronic phase-change memories [which use iterative level-setting schemes— see, e.g., Papandreou et al. (40)]. 7 of 9 Ríos et al., Sci. Adv. 2019;5:eaau5759 15 February 2019 SUPPLEMENTARY MATERIALS Meeting (IEDM) (2017), San Francisco, CA, USA, 2 to 6 December 2017 Supplementary material for this article is available at http://advances.sciencemag.org/cgi/ content/full/5/2/eaau5759/DC1 Section S1. Device characterizations using balanced splitters Section S2. Experimental setup Section S3. Noise Section S4. Offset correction Section S5. Error propagation Section S6. Proposed matrix-vector multiplication architecture Fig. S1. Balanced splitter characterization. Fig. S2. Diagram of the experimental pump-probe setup. Fig. S3. Diagram of the experimental setup for matrix-vector multiplication. Fig. S4. Noise for up to 20-ms measurements using a 125-MHz (New Focus, 1811) photodetector for different case scenarios. Fig. S5. Noise for up to 2-min measurements using a 200-kHz (New Focus, 2011) photodetector. Fig. S6. Example design of [2 × 2] matrix-vector multiplication. Supplementary material for this article is available at http://advances.sciencemag.org/cgi/ content/full/5/2/eaau5759/DC1 Section S1. Device characterizations using balanced splitters Section S2. Experimental setup Section S3. Noise Section S4. Offset correction Section S5. Error propagation Section S6. Proposed matrix-vector multiplication architecture Fig. S1. Balanced splitter characterization. Fig. S2. Diagram of the experimental pump-probe setup. Fig. S3. Diagram of the experimental setup for matrix-vector multiplication. Fig. S4. Noise for up to 20-ms measurements using a 125-MHz (New Focus, 1811) photodetector for different case scenarios. Fig. S5. Noise for up to 2-min measurements using a 200-kHz (New Focus, 2011) photodetector. Fig. S6. Example design of [2 × 2] matrix-vector multiplication. Supplementary material for this article is available at http://advances.sciencemag.org/cgi/ content/full/5/2/eaau5759/DC1 21. K. Bergman, J. Shalf, T. Hausken, Optical interconnects and extreme computing. Opt. Photonics News 27, 32–39 (2016). Section S1. Device characterizations using balanced splitters 22. Y. Saad, Iterative Methods for Sparse Linear Systems (Society for Industrial and Applied Mathermatics, 2003). 23. M. Hu, J. P. Strachan, Z. Li, E. M. Grafals, N. Davila, C. Graves, S. Lam, N. Ge, J. J. Yang, R. S. Williams, Dot-Product Engine for Neuromorphic Computing: Programming 1T1M Crossbar to Accelerate Matrix-Vector Multiplication, in 2016 53nd ACM/EDAC/IEEE Design Automation Conference (DAC) (2016), Austin, TX, USA, 5 to 9 June 2016. Section S5. Error propagation Section S6. Proposed matrix-vector multiplication architecture 24. G. W. Burr, R. M. Shelby, S. Sidler, C. di Nolfo, J. Jang, I. Boybat, R. S. Shenoy, P. Narayanan, K. Virwani, E. U. Giacometti, B. N. Kurdi, H. Hwang, Experimental demonstration and tolerancing of a large-scale neural network (165,000 synapses), using phase-change memory as the synaptic weight element. IEEE Trans. Electron Devices 62, 3498–3507 (2015). Fig. S1. Balanced splitter characterization. 25. SUPPLEMENTARY MATERIALS M. Wuttig, N. Yamada, Phase-change materials for rewriteable data storage. Nat. Mater. 6, 824–832 (2007). Fig. S5. Noise for up to 2-min measurements using a 200-kHz (Ne Fig. S5. Noise for up to 2-min measurements using a 200-kHz (New Focus, 2011) photodetector. Fig. S6. Example design of [2 × 2] matrix-vector multiplication. 26. E. Kuramochi, M. Notomi, Phase-change memory. Nat. Photonics 9, 712–714 (2015). MATERIALS AND METHODS Sample fabrication 1. C. Sun, M. T. Wade, Y. Lee, J. S. Orcutt, L. Alloatti, M. S. Georgas, A. S. Waterman, J. M. Shainline, R. R. Avizienis, S. Lin, B. R. Moss, R. Kumar, F. Pavanello, A. H. Atabaki, H. M. Cook, A. J. Ou, J. C. Leu, Y.-H. Chen, K. Asanović, R. J. Ram, M. A. Popović, V. M. Stojanović, Single-chip microprocessor that communicates directly using light. Nature 528, 534–538 (2015). The nanophotonic memory cells were fabricated on 330-nm Si3N4/ 3.3-mm SiO2 wafers. A JEOL JBX-5500ZD 50-kV electron-beam lithog- raphy (EBL) was used to write the photonic circuitry using MaN-2403 negative resist, followed by a reflow process of 90 s at 100°C. Subse- quently, reactive ion etching in CHF3/Ar/O2 was carried out to etch 165 nm of the Si3N4 and thus obtain the bare photonic device. A second EBL writing step using poly(methyl methacrylate), followed by a lift-off process, was used to pattern the phase-change materials. A stack of 10 nm of GST with a 10-nm indium tin oxide capping (to avoid oxida- tion) was deposited in an argon environment using a homebuilt radio frequency sputtering system (Nordiko). Before the measurements, the GST was crystallized on a hot plate following a 5-min annealing at 250°C. Figure S1 shows an optical microscope image of the balanced splitters, the photonic device used in this work. 2. K. Vandoorne, P. Mechet, T. Van Vaerenbergh, M. Fiers, G. Morthier, D. Verstraeten, B. Schrauwen, J. Dambre, P. Bienstman, Experimental demonstration of reservoir computing on a silicon photonics chip. Nat. Commun. 5, 3541 (2014). 3. J. M. Shainline, S. M. Buckley, R. P. Mirin, S. W. Nam, Superconducting optoelectronic circuits for neuromorphic computing. Phys. Rev. Appl. 7, 034013 (2017). 4. Y. Shen, N. C. Harris, S. Skirlo, M. Prabhu, T. Baehr-Jones, M. Hochberg, X. Sun, S. Zhao, H. Larochelle, D. Englund, M. Soljačić, Deep learning with coherent nanophotonic circuits. Nat. Photonics. 11, 441–446 (2017). 5. D. Kielpinski, R. Bose, J. Pelc, T. Van Vaerenbergh, G. Mendoza, N. Tezak, R. G. Beausoleil, Information processing with large-scale optical integrated circuits, in 2016 IEEE International Conference on Rebooting Computing (ICRC) (2016), San Diego, CA, USA, 17 to 19 October 2016. 6. B. G. Lee, N. Dupuis, P. Pepeljugoski, L. Schares, R. Budd, J. R. Bickford, C. L. Schow, Silicon photonic switch fabrics in computer communications systems. J. Lightwave Technol. 33, 768–777 (2014). Measurement setup Agarwal, Mixed-mode operation of hybrid phase-change nanophotonic circuits. Nano Lett. 17, 150–155 (2017). 12. C. Ríos, P. Hosseini, C. D. Wright, H. Bhaskaran, W. H. P. Pernice, On-chip photonic memory elements employing phase-change materials. Adv. Mater. 26, 1372–1377 (2014). 13. C. Ríos, M. Stegmaier, P. Hosseini, D. Wang, T. Scherer, C. D. Wright, H. Bhaskaran, 13. C. Ríos, M. Stegmaier, P. Hosseini, D. Wang, T. Scherer, C. D. Wright, H. Bhaskaran, W. H. P. Pernice, Integrated all-photonic non-volatile multi-level memory. Nat. Photonics 9, 725–732 (2015). W. H. P. Pernice, Integrated all-photonic non-volatile multi-level memory. Nat. Photonics 9, 725–732 (2015). 14. H. J. Caulfield, S. Dolev, Why future supercomputing requires optics. Nat. Photonics 4, 261–263 (2010). 15. J. Feldmann, M. Stegmaier, N. Gruhler, C. Ríos, H. Bhaskaran, C. D. Wright, W. H. P. Pernice, Calculating with light using a chip-scale all-optical abacus. Nat. Commun 1256 (2017). 16. M. Le Gallo, A. Sebastian, R. Mathis, M. Manica, H. Giefers, T. Tuma, C. Bekas, A. Curioni, E. Eleftheriou, Mixed-precision in-memory computing. Nat. Electron. 1, 246–253 (2018). 17. P. M. Sheridan, F. Cai, C. Du, W. Ma, Z. Zhang, W. D. Lu, Sparse coding with memristor networks. Nat. Nanotechnol. 12, 784–789 (2017). 18. V. Seshadri, D. Lee, T. Mullins, H. Hassan, A. Boroumand, J. Kim, M. A. Kozuch, O. Mutlu, P. B. Gibbons, T. C. Mowry, Buddy-RAM: Improving the Performance and Efficiency of Bulk Bitwise Operations Using DRAM. arXiv:1611.09988 (2016). 19. A. Sebastian, T. Tuma, N. Papandreou, M. L. Gallo, L. Kull, T. Parnell, E. Eleftheriou, Temporal correlation detection using computational phase-change memory. Nat. Commun. 8, 1115 (2017). 20. M. Le Gallo, A. Sebastian, G. Cherubini, H. Giefers, E. Eleftheriou, Compressed Sensing Recovery using Computational Memory, in 2017 IEEE International Electron Devices Meeting (IEDM) (2017), San Francisco, CA, USA, 2 to 6 December 2017. S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E Symposium of Circuits and Systems (ISCAS) (2011), Rio de Janeiro, Brazil, 15 to 18 May 2011, pp. 329–332. 27. M. Le Gallo, T. Tuma, F. Zipoli, A. Sebastian, E. Eleftheriou, Inherent stochasticity in phase-change memory devices, 2016 46th European Solid-State Device Research Conference (ESSDERC) (2016), Lausanne, Switzerland, 12 to 15 September 2016, pp. 373–376. 28. C. D. Wright, P. Hosseini, J. A. V. Diosdado, Beyond von-Neumann computing with nanoscale phase-change memory devices. Adv. Funct. Mater. 23, 2248–2254 (2013). Acknowledgments: We thank the collaborative nature of European Science for allowing this work to be carried out and A. Ne for the helpful discussions. Funding: This research was supported by EPSRC via grants EP/J018694/1, EP/M015173/1, and EP/M015130/1 in the United Kingdom and the Deutsche Forschungsgemeinschaft (DFG) grant PE 1832/2-1 in Germany. We acknowledge funding for this work from the European Union’s Horizon 2020 Research and Innovation Program (Fun-COMP project, #780848). A.S. acknowledges support from the European Research Council through the European Union’s Horizon 2020 Research and Innovation Program under grant number 682675. Author contributions: C.R., N.Y., and Z.C. built the setup, worked on device fabrication, performed the measurements, and analyzed the data. H.B., A. S., M.L.G., C.R., and N.Y conceived and planned the project. H.B., A.S., W.H.P.P., and C.D.W. led the project. All authors discussed the results and wrote the manuscript. Competing interests: H.B., W.H.P.P., and C.R. are inventors on several patents related to this work filed by Oxford University Innovation [application no. PCT/GB2016/052871 (lapsed); application no. US15/760 945 (pending); application no. EP16770973.2 (pending); application no. GB1607345.4 (lapsed); and application no. PCT/GB2017/051143, publication no. WO 2017/ 187151 (A1)]. N.Y., C.R., Z.C., H.B., W.H.P.P., and C.D.W. are inventors on several additional patents also related to this work filed by Oxford University Innovation [application no. GB1719346.7 (lapsed); application no. PCT/GB2018/051815; and application no. GB1719347.5]. All the other authors declare that they have no competing interests. Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Additional data related to this paper may be requested from the authors or Oxford Research Archive for Data (https://ora.ox.ac.uk). 29. P. Hosseini, A. Sebastian, N. Papandreou, C. D. Wright, H. Bhaskaran, Accumulation-based computing using phase-change memories with FET access devices. IEEE Electron Device Lett. 36, 975–977 (2015). 30. Z. Cheng, C. Ríos, N. Youngblood, C. D. Wright, W. H. P. Pernice, H. Ríos et al., Sci. Adv. 2019;5:eaau5759 15 February 2019 8 of 9 Ríos et al., Sci. Adv. 2019;5:eaau5759 15 February 2019 S C I E N C E A D V A N C E S \| R E S E A R C H A R T I C L E Bhaskaran, Device-level photonic memories and logic applications using phase-change materials. Adv. Mater. 30, 1802435 (2018). 31. C. Ríos, M. Stegmaier, Z. Cheng, N. Youngblood, C. D. Wright, W. H. P. Pernice, H. Bhaskaran, Controlled switching of phase-change materials by evanescent-field coupling in integrated photonics [Invited]. Opt. Mater. Express 8, 2455–2470 (2018). 32. M. Stegmaier, C. Ríos, H. Bhaskaran, W. H. P. Pernice, Thermo-optical effect in phase-change nanophotonics. ACS Photonics 3, 828–835 (2016). 33. L. Waldecker, T. A. Miller, M. Rudé, R. Bertoni, J. Osmond, V. Pruneri, R. E. Simpson, R. Ernstorfer, S. Wall, Time-domain separation of optical properties from structural transitions in resonantly bonded materials. Nat. Mater. 14, 991–995 (2015). 34. D. Loke, T. H. Lee, W. J. Wang, L. P. Shi, R. Zhao, Y. C. Yeo, T. C. Chong, S. R. Elliott, Breaking the speed limits of phase-change memory. Science 336, 1566–1569 (2012). 35. D. Ielmini, S. Lavizzari, D. Sharma, A. L. Lacaita, Physical interpretation, modeling and impact on phase change memory (PCM) reliability of resistance drift due to chalcogenide structural relaxation (IEDM Technical Digest, 2007), pp. 939–942. 36. I. V. Karpov, M. Mitra, D. Kau, G. Spadini, Y. A. Kryukov, V. G. Karpov, Fundamental drift of parameters in chalcogenide phase change memory. J. Appl. Phys. 102, 124503 (2007). Submitted 25 June 2018 Accepted 7 January 2019 Published 15 February 2019 10.1126/sciadv.aau5759 37. P. Fantini, S. Brazzelli, E. Cazzini, A. Mani, Band gap widening with time induced by structural relaxation in amorphous Ge2Sb2Te5 films. Appl. Phys. Lett. 100, 013505 (2012). 38. M. Le Gallo, D. Krebs, F. Zipoli, M. Salinga, A. Sebastian, Collective structural relaxation in phase-change memory devices. Adv. Electron. Mater. 4, 1700627 (2018). 39. J. Y. Raty, W. Zhang, J. Luckas, C. Chen, R. Mazzarello, C. Bichara, M. Wuttig, Aging mechanisms in amorphous phase-change materials. Nat. Commun. 6, 7467 (2015). Citation: C. Ríos, N. Youngblood, Z. Cheng, M. Le Gallo, W. H. P. Pernice, C. D. Wright, A. Sebastian, H. Bhaskaran, In-memory computing on a photonic platform. Sci. Adv. 5, eaau5759 (2019). 40. N. Papandreou, H. Pozidis, A. Pantazi, A. Sebastian, M. Breitwisch, C. Lam, E. Eleftheriou, Programming algorithms for multilevel phase-change memory, in 2011 IEEE International 9 of 9
https://openalex.org/W1931489379	https://hal-univ-rennes1.archives-ouvertes.fr/hal-01187426/document	English	null	Spinal cord hemorrhage in a patient with neurosarcoidosis on long-term corticosteroid therapy: case report	BMC neurology	2,015	cc-by	3,005	To cite this version: Benoit Pegat, Sophie Drapier, Xavier Morandi, Gilles Edan. Spinal cord hemorrhage in a patient with neurosarcoidosis on long-term corticosteroid therapy: case report. BMC Neurology, 2015, 15 (1), pp.123. 10.1186/s12883-015-0373-6. hal-01187426 * Correspondence: benoit.pegat@gmail.com 1Department of Neurology, Pontchaillou Hospital, Rennes University Hospital, Rennes, France Full list of author information is available at the end of the article Background g Sarcoidosis is a multisystemic granulomatosis charac- terized by non-caseating granulomas [1]. In 5–7 % cases, it may involve the neurological system [2–4]. Spinal cord sarcoidosis occurs in about 6 % of all neu- rosarcoidosis cases [3, 5, 6]. Although several patho- logical studies have reported microvascular changes in neurosarcoidosis [7, 8], bleeding is rare. Only 18 cases of cerebral hemorrhage have been reported [9–12], and four of these concerned patients on corticosteroid therapy [10, 13–15]. We present the first recorded case of spinal cord hemorrhage in a 48-year-old Caucasian woman who had been treated with corticosteroids for 5 years for relapsing neurosarcoidosis. HAL Id: hal-01187426 https://univ-rennes.hal.science/hal-01187426v1 Submitted on 10 Oct 2018 L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. Pegat et al. BMC Neurology (2015) 15:123 DOI 10.1186/s12883-015-0373-6 Spinal cord hemorrhage in a patient with neurosarcoidosis on long-term corticosteroid therapy: case report Benoit Pegat1, Sophie Drapier1, Xavier Morandi2 and Gilles Edan1 © 2015 Pegat et al. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http:// creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Abstract Background: Central nervous system bleeding is a rare complication of neurosarcoidosis: only 18 cases of spontaneous cerebral hematoma have been reported. We present the first recorded case of spinal cord hemorrhage in neurosarcoidosis. Case presentation: A 48-year-old Caucasian woman had relapsing neurosarcoidosis for 5 years, with inflammatory spinal and cerebral lesions. While on 20 mg corticosteroids, she experienced subacute paraparesia with right leg pain. A spine MRI revealed a low thoracic hematomyelia at the T10-T11 level. Despite high doses of corticosteroids, her condition continued to worsen. Surgical evacuation of the hematoma was performed 10 days after the onset of bleeding, and she partially recovered. Conclusion: This report highlights the possibility of spinal cord hemorrhage secondary to sarcoid vasculitis. The patient improved after surgical evacuation of the intramedullary hematoma. Immuno-modulating agents must be envisaged in severe neurosarcoidosis, to prevent complications. Keywords: Hematomyelia, Hemorrhage, Spinal cord, Neurosarcoidosis, Magnetic resonance imaging, Corticosteroid : Hematomyelia, Hemorrhage, Spinal cord, Neurosarcoidosis, Magnetic resonance imaging, Corticoste when the patient presented with erythema nodosum and bilateral hilar adenopathy. Onset of neurosarcoidosis occurred 13 years later, characterized by cauda equina syndrome with a sensorimotor deficit in the right leg. Magnetic resonance imaging (MRI) revealed diffuse en- largement of the spinal cord, with high T2-weighted signal intensity from T8 to the L1-L2 level (Fig. 1a). There was no enhancement on gadolinium-enhanced T1-weighted images of the spine. Cerebral MRI was normal. Cerebrospinal fluid (CSF) tests revealed aseptic lymphocytic meningitis, with 540 cells/mm3 and a pro- tein concentration of 375 mg/dL. Oligoclonal bands were absent, and the blood level of angiotensin- converting enzyme was normal. The patient’s condition improved after high doses of intravenous steroids, followed by oral corticosteroids (starting at 1 mg/kg/ day, then tapering). CSF returned to normal after 4 months’ treatment. The leg dysesthesia persisted, as did bladder disturbances, treated by intermittent self- catheterization. (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http:// creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Case presentation Sarcoidosis was histologically proven at 30 years, with pulmonary biopsy revealing non caseating granulomas Two years later, the patient had a relapse, while on steroids (4 mg/day), characterized by gait ataxia and Page 2 of 4 Pegat et al. BMC Neurology (2015) 15:123 Fig. 1 Spinal MRI images of the first inflammatory event (a) and the hematomyelia (b and c). a. Sagittal T2-weighted MRI showing a high-intensity signal and enlargement of the spinal cord extending from T8 to the L1-L2 level. b. Sagittal T2-weighted turbo spin-echo MRI showing a heterogeneous low-intensity signal at the T10-T11 level. c. Transversal T2-weighted gradient-echo MRI at the T10–T11 level confirming the intramedullary low-intensity signal: it is an intramedullary hemorrhage (hematomyelia) Fig. 1 Spinal MRI images of the first inflammatory event (a) and the hematomyelia (b and c). a. Sagittal T2-weighted MRI showing a high-intensity signal and enlargement of the spinal cord extending from T8 to the L1-L2 level. b. Sagittal T2-weighted turbo spin-echo MRI showing a heterogeneous low-intensity signal at the T10-T11 level. c. Transversal T2-weighted gradient-echo MRI at the T10–T11 level confirming the intramedullary low-intensity signal: it is an intramedullary hemorrhage (hematomyelia) cognitive impairment, inflammatory activity in the CSF (40 cells/mm3, protein concentration of 137 mg/dL) and multifocal lesions on a brain MRI (Fig. 2a-b). T2- fluid attenuation inversion recovery (FLAIR) and T2- weighted gradient-echo imaging showed a low-intensity frontoparietal lesion identified as blood deposits. Ste- roids were increased to 1 mg/kg/day and she showed a marked improvement. Follow-up CSF analysis indicated fewer than 10 cells/mm3, and oligoclonal bands were found this time. worsening of gait difficulties (unable to walk alone owing to a major motor deficit in the right leg) and urinary re- tention. Sensory examination revealed a sensitive level at L3, with lower-limb hypoesthesia, and decreased pelvic sensation. Spinal MRI revealed a large heterogeneous intramedullary lesion at T10-T11, with a low-intensity T2-weighted gradient-echo signal, diagnosed as hema- tomyelia (Fig. 1b-c). Part of the medullary cone was enhanced by gadolinium in a T1-weighted sequence. Coagulation blood tests and platelets were normal. Des- pite high doses of intravenous corticosteroids, she be- came paraplegic within 10 days, with a sensory deficit Three years later, while on corticosteroids (20 mg/day), the patient experienced paroxysmal right leg pain, acute Fig. 2 Cranial MRI images of the second inflammatory episode. a. Case presentation Axial T2 fluid attenuation inversion recovery (FLAIR)-weighted MRI indicating large, high-intensity periventricular lesions, and a low-intensity left frontoparietal lesion. b. Coronal T2-weighted gradient-echo MRI confirming the presence of a low-intensity lesion, suggestive of hemosiderin deposits, that had not been found on previous MRIs Fig. 2 Cranial MRI images of the second inflammatory episode. a. Axial T2 fluid attenuation inversion recovery (FLAIR)-weighted MRI indicating large, high-intensity periventricular lesions, and a low-intensity left frontoparietal lesion. b. Coronal T2*-weighted gradient-echo MRI confirming the presence of a low-intensity lesion, suggestive of hemosiderin deposits, that had not been found on previous MRIs Pegat et al. BMC Neurology (2015) 15:123 Page 3 of 4 at the T12 level. A neurosurgical intervention was per- formed to evacuate the hematoma, without any additional complications. The neurosurgeon observed numerous abnormally enlarged blood vessels on the back of the spinal cord. Histological analysis of spinal cord tissue did not reveal any abnormal cells. directly caused the bleeding here. Indeed, we decided to increase the corticosteroid dose, as adding an immuno- modulating agent to control the neurosarcoid vasculitis (and spare corticosteroids) was not possible because of an infectious complication. There have not been any randomized controlled trials of pharmacological treat- ments for neurosarcoidosis, but the algorithm developed by Nozaki and Judson, has proved useful for severe neu- rosarcoidosis [24]. After yearlong intensive rehabilitation, the patient remained paraparetic, with a severe deficit in the right leg (3/5 on motor scale), but was nonetheless able to walk with a walking frame. Adding an immune-modulating agent was discussed but not done, because of a severe infectious complication of a pressure ulcer that required 6 weeks’ hospitalization in intensive care. Four years after the hematomyelia, the patient was on a daily dose of just 4 mg corticosteroids and remained relapse free, with no new lesions on follow-up MRI. Kreppel and colleagues’ meta-analysis [25] is the largest review of reports of spinal hematoma published before 1996. These authors reviewed 613 patients belonging to four etiological groups of spinal cord hemorrhage: intra- medullary (hematomyelia), subarachnoidal, subdural and epidural. Hematomyelia was found in only 0.82 % of patients, possibly owing to the dearth of MRI-based diag- nosis prior to 1996. More recently, Leep and colleagues [19] and Heldner and colleagues [20] reviewed the diagno- sis, cause and treatment of hematomyelia. Hematomyelia can present as an acute, subacute, step-wise or chronic myelopathy [19]. Abbreviations Abbreviations CNS: Central nervous system; CSF: Cerebrospinal fluid; MRI: Magnetic resonance imaging; FLAIR: Fluid attenuation inversion recovery. CNS: Central nervous system; CSF: Cerebrospinal fluid; MRI: Magnetic resonance imaging; FLAIR: Fluid attenuation inversion recovery. Long-term corticosteroids also increase vascular fragil- ity, owing to reduced collagen formation in vessel walls [23], but there was no evidence to suggest that steroids Discussion h d In the case described here, the diagnosis of neurosarcoi- dosis was definite according to Judson’s criteria [16] and probable according to Zajicek’s [17]. This first report highlights the possibility of spinal cord hemorrhage in patients with neurosarcoidosis, as suggested by Waubant and colleagues, who described hemosiderin deposits in spinal-cord sarcoidosis [18]. Common causes of atraumatic hematomyelia [19, 20] include vascular malformations (arteriovenous fistula, cavernoma, capillary telangiectasia, venous angioma) [21], neoplasms, Gowers’intrasyringal hemorrhage, spinal radi- ation, anticoagulation and bleeding disorders. None of these applied to our patient, suggesting that her hemato- myelia was caused by bleeding of a spinal cord granuloma, given the proximity of the first inflammatory lesion (T8-L1) to the hemorrhage (T10-T11). Case presentation As it did in our case, neurological deteri- oration can occur after the initial hemorrhage, owing to a secondary tissue response [19]. It is essential to monitor vital signs and neurological status in the intensive care unit, to prevent complications [20]. However, there have not been any clinical trials comparing different ways of managing hematomyelia [19, 20], notably the timing of hematoma evacuation. In our case, the decision to operate was made because of a progressive, rapid and inexorable neurological deterioration. There was a definite, but not total, clinical improvement following treatment for this rare and severe complication of neurosarcoidosis. Conclusion Despite the lack of histological confirmation of the source of bleeding, there is sufficient evidence from cere- bral cases [9] to suggest that hemorrhage in neurosarcoi- dosis may be caused by vascular pathologies (venous, arteriolar, or micro-arterio-venous malformations) in- duced by the disease itself. Sarcoid granulomas in the central nervous system (CNS) follow perivascular spaces and penetrate brain parenchyma [22]. Post mortem exam- inations of intracranial hemorrhage in neurosarcoidosis have revealed perivascular inflammatory infiltrate and vessel wall destruction by granulomas [7, 9]. The enlarged blood vessels noticed by the neurosurgeon in our patient seem to support the hypothesis of neurosarcoid vasculitis. The frontoparietal hemorrhage that occurred during the second episode, 3 years before the hematomyelia (Fig. 2a-b), is also suggestive of the fragility of blood ves- sels in the CNS damaged by the neurosarcoid vasculitis. This report presents the first case of intramedullary spinal cord hemorrhage complicating neurosarcoidosis, which improved following surgical evacuation. Neuro- sarcoid vasculitis may have favored this rare complica- tion. Immuno-modulating agents could be envisaged as a means of preventing the complications of long- term corticosteroid therapy and vasculitis in severe CNS neurosarcoidosis. Consent Written informed consent was obtained from the patient for publication of this Case report and any accompany- ing images. A copy of the written consent is available for review by the Editor of this journal. Abbreviations CNS: Central nervous system; CSF: Cerebrospinal fluid; MRI: Magnetic resonance imaging; FLAIR: Fluid attenuation inversion recovery. 21. Anson JA, Spetzler RF. Interventional neuroradiology for spinal pathology. Clin Neurosurg. 1992;39:388. 22. Herring AB, Urich H. Sarcoidosis of the central nervous system. J Neurol Sci. 1969;9:405–22. 23. Hein R, Mauch C, Hatamochi A, Krieg T. Influence of corticosteroids on chemotactic response and collagen metabolism of human skin fibroblasts. Biochem Pharmacol. 1988;37:2723–9. 24. Nozaki K, Judson MA. Neurosarcoidosis: clinical manifestations, diagnosis and treatment. Presse Médicale. 2012;41:e331–48. 25. Kreppel D, Antoniadis G, Seeling W. Spinal hematoma: a literature survey with meta-analysis of 613 patients. Neurosurg Rev. 2003;26:1–49. Authors’ contributions All the authors interviewed and examined the patient. XM operated on the patient. The draft was written by BP, but SD, XM and GE contributed references and undertook a critical review. All authors read and approved the final manuscript. Competing interests The authors declare that they have no competing interests. Page 4 of 4 Page 4 of 4 Pegat et al. BMC Neurology (2015) 15:123 Acknowledgements We are grateful to Drs Hélène Raoult and Jean-Christophe Ferré for their assistance in collecting and interpreting the MRI data. Author details 1 1Department of Neurology, Pontchaillou Hospital, Rennes University Hospital, Rennes, France. 2Department of Neurosurgery, Pontchaillou Hospital, Rennes University Hospital, Rennes, France. Received: 17 November 2014 Accepted: 6 July 2015 22. Herring AB, Urich H. Sarcoidosis of the central nervous system. J Neurol Sci. 1969;9:405–22. References 1. Iannuzzi MC, Rybicki BA, Teirstein AS. Sarcoidosis. N Engl J Med. 2007;357:2153–65. 2. Stern BJ, Krumholz A, Johns C, Scott P, Nissim J. Sarcoidosis and its neurological manifestations. Arch Neurol. 1985;42:909–17. 3. Gascon-Bayarri J, Mana J, Martinez-Yelamos S, Murillo O, Rene R, Rubio F. Neurosarcoidosis: report of 30 cases and a literature survey. Eur J Intern Med. 2011;22:e125–32. 4. Morimoto T, Azuma A, Abe S, Usuki J, Kudoh S, Sugisaki K, et al. Epidemiology of sarcoidosis in Japan Eur Respir J 2008;31:372–9 5. Duhon BS, Shah L, Schmidt MH. Isolated intramedullary neurosarcoidosis of the thoracic spine: case report and review of the literature. Eur Spine J. 2012;21 Suppl 4:S390–5. 6. Junger SS, Stern BJ, Levine SR, Sipos E, Marti-Masso JF. Intramedullary spinal sarcoidosis: clinical and magnetic resonance imaging characteristics. Neurology. 1993;43:333–7. gy 7. Libman RB, Sharfstein S, Harrington W, Lerner P. Recurrent intracerebral hemorrhage from sarcoid angiitis. J Stroke Cerebrovasc Dis. 1997;6:373–5. 8. Caplan L, Corbett J, Goodwin J, Thomas C, Shenker D, Schatz N. Neuro- ophthalmologic signs in the angiitic form of neurosarcoidosis. Neurology. 1983;33:1130–5. 8. Caplan L, Corbett J, Goodwin J, Thomas C, Shenker D, Schatz N. Neuro- ophthalmologic signs in the angiitic form of neurosarcoidosis. Neurology. 1983;33:1130–5. 9. O’Dwyer JP, Al-Moyeed BA, Farrell MA, Pidgeon CN, Collins DR, Fahy A, et al. Neurosarcoidosis-related intracranial haemorrhage: three new cases and a systematic review of the literature. Eur J Neurol. 2013;20:71–8. 10. Spencer TS, Campellone JV, Maldonado I, Huang N, Usmani Q, Reginato AJ. Clinical and magnetic resonance imaging manifestations of neurosarcoidosis. Semin Arthritis Rheum. 2005;34:649–61. 11. Tsappidi S, Hui F, Turan TN, Hunter S. Intracerebral hemorrhage: an unusual presentation of neurosarcoidosis. Neurologist. 2011;17:292–6. 11. Tsappidi S, Hui F, Turan TN, Hunter S. Intracerebral hemorrhage: an unusual presentation of neurosarcoidosis. Neurologist. 2011;17:292–6. 12. Lettau M, Laible M, Treier M. Infarction with subarachnoid hemorrhage from vasculitis with neurosarcoidosis. Rofo. 2011;183:573–5. 12. Lettau M, Laible M, Treier M. Infarction with subarachnoid hemorrhage from vasculitis with neurosarcoidosis. Rofo. 2011;183:573–5. 13. Dakdouki GK, Kanafani ZA, Ishak G, Hourani M, Kanj SS. Intracerebral bleeding in a patient with neurosarcoidosis while on corticosteroid therapy. South Med J. 2005;98:492–4. 13. Dakdouki GK, Kanafani ZA, Ishak G, Hourani M, Kanj SS. Intracerebral bleeding in a patient with neurosarcoidosis while on corticosteroid therapy. South Med J. 2005;98:492–4. 14. Eid H, O’Connor CR, Catalano E, Reginato AJ. Life-threatening vasculitis associated with sarcoidosis. J Clin Rheumatol. 1998;4:338–43. 14. 25. Kreppel D, Antoniadis G, Seeling W. Spinal hematoma: a literature survey with meta-analysis of 613 patients. Neurosurg Rev. 2003;26:1–49. 21. Anson JA, Spetzler RF. Interventional neuroradiology for spinal pathology. Clin Neurosurg. 1992;39:388. References Eid H, O’Connor CR, Catalano E, Reginato AJ. Life-threatening vasculitis associated with sarcoidosis. J Clin Rheumatol. 1998;4:338–43. 15. Lavrnic DV, Vidakovic AB, Apostoloski S, Stefanova ED, Stevic-Marinkovic Z. Intracerebral hemorrhage in a female patient with pulmonary sarcoidosis. Srp Arh Celok Lek. 1992;120:363–6. Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: Submit your next manuscript to BioMed Central and take full advantage of: 16. Judson MA, Baughman RP, Teirstein AS, Terrin ML, Yeager Jr H. Defining organ involvement in sarcoidosis: the ACCESS proposed instrument. ACCESS Research Group. A Case Control Etiologic Study of Sarcoidosis. Sarcoidosis Vasc Diffuse Lung Dis. 1999;16:75–86. • Convenient online submission 17. Zajicek JP, Scolding NJ, Foster O, Rovaris M, Evanson J, Moseley IF, et al. Central nervous system sarcoidosis–diagnosis and management. QJM. 1999;92:103–17. • Thorough peer review 18. Waubant E, Manelfe C, Bonafe A, Berry I, Clanet M. MRI of intramedullary sarcoidosis: follow-up of a case. Neuroradiology. 1997;39:357–60. 19. Leep Hunderfund AN, Wijdicks EF. Intramedullary spinal cord hemorrhage (hematomyelia). Rev Neurol Dis. 2009;6:E54–61. 20. Heldner MR, Arnold M, Nedeltchev K, Gralla J, Beck J, Fischer U. Vascular diseases of the spinal cord: a review. Curr Treat Options Neurol. 2012;14:509–20.
https://openalex.org/W4239588467	https://biblio.ugent.be/publication/8752693/file/8752694.pdf	English	null	Practice Recommendations for Lung Cancer Radiotherapy During the COVID-19 Pandemic: An ESTRO-ASTRO Consensus Statement	International journal of radiation oncology, biology, physics	2,020	cc-by	8,208	COVID-19 Rapid Communication COVID-19 Rapid Communication Int J Radiation Oncol Biol Phys, Vol. 107, No. 4, pp. 631e640, 2020 0360-3016/ 2020 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/license by/4.0/). Int J Radiation Oncol Biol Phys, Vol. 107, No. 4, pp. 631e640, 2020 0360-3016/ 2020 The Author(s). Published by Elsevier B.V. This is an o by/4.0/). Practice Recommendations for Lung Cancer Radiotherapy During the COVID-19 Pandemic: An ESTRO-ASTRO Consensus Statement Matthias Guckenberger,* Claus Belka,y Andrea Bezjak,z Jeffrey Bradley,x Megan E. Daly,k Dirk DeRuysscher,{ Rafal Dziadziuszko,# Corinne Faivre-Finn, Michael Flentje,yy Elizabeth Gore,zz Kristin A. Higgins,xx Puneeth Iyengar,kk Brian D. Kavanagh,{{ Sameera Kumar,## Cecile Le Pechoux,* Yolande Lievens,yyy Karin Lindberg,zzz Fiona McDonald,xxx Sara Ramella,kkk Ramesh Rengan,{{{ Umberto Ricardi,### Andreas Rimner,** George B. Rodrigues,yyyy Steven E. Schild,zzzz Suresh Senan,xxxx Charles B. Simone II,kkkk Ben J. Slotman,{{{{ Martin Stuschke,#### Greg Videtic,*** Joachim Widder,yyyyy Sue S. Yom,zzzzz and David Palma yyyy https://doi.org/10.1016/j.ijrobp.2020.05.012 Corresponding Author: Matthias Guckenberger; E-mail: Matthias. Guckenberger@usz.ch , , pp , Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/ Matthias Guckenberger,* Claus Belka,y Andrea Bezjak,z Jeffrey Bradley,x Megan E. Daly,k Dirk DeRuysscher,{ Rafal Dziadziuszko,# Corinne Faivre-Finn, Michael Flentje,yy Elizabeth Gore,zz Kristin A. Higgins,xx Puneeth Iyengar,kk Brian D. Kavanagh,{{ Sameera Kumar,## Cecile Le Pechoux,* Yolande Lievens,yyy Karin Lindberg,zzz Fiona McDonald,xxx Sara Ramella,kkk Ramesh Rengan,{{{ Umberto Ricardi,### Andreas Rimner,** George B. Rodrigues,yyyy Steven E. Schild,zzzz Suresh Senan,xxxx Charles B. Simone II,kkkk Ben J. Slotman,{{{{ Martin Stuschke,#### Greg Videtic,* Joachim Widder,yyyyy Sue S. Yom,zzzzz and David Palma yyyy Oncology, Ghent University Hospital and Ghent University, Belgium; zzzDepartment of Head, Neck, Lung and Skin Cancer, Theme Cancer, Karolinska University Hospital, Stockholm, Sweden; xxxRoyal Marsden Hospital, London, UK; kkkRadiation Oncology, Campus Bio-Medico University of Rome, Italy; {{{Department of Radiation Oncology, University of Washington School of Medicine, Seattle, USA; ###Department of Oncology, University of Turin, Italy; Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, USA; yyyyDivision of Radiation Oncology, Western University, London, Canada; zzzzDepartment of Radiation Oncology, Mayo Clinic, Phoenix, USA; xxxxDepartment of Radiation Oncology, Amsterdam UMC, The Netherlands; kkkkDepartment of Radiation Oncology, New York Proton Center and Memorial Sloan Kettering Cancer Center, USA; {{{{Department of Radiation Oncology, Amsterdam UMC, The Netherlands; ####Department of Radiotherapy, University Duisburg-Essen, Germany; **Department of Radiation Oncology, Cleveland Clinic Lerner College of Medicine of Case Western University, USA; yyyyyDepartment of Radiation Oncology, Comprehensive Cancer Center, Medical University of Vienna, Austria; zzzzzDepartment of Radiation Oncology, University of California San Francisco, USA Received Apr 1, 2020. Accepted for publication Apr 1, 2020. Background: The COVID-19 pandemic has caused radiotherapy resource pressures and led to increased risks for lung cancer patients and healthcare staff. An international group of experts in lung cancer radiotherapy established this practice recom- mendation pertaining to whether and how to adapt radiotherapy for lung cancer in the COVID-19 pandemic. Methods: For this ESTRO & ASTRO endorsed project, 32 experts in lung cancer radiotherapy contributed to a modiﬁed Del- phi consensus process. We assessed potential adaptations of radiotherapy in two pandemic scenarios. The ﬁrst, an early pandemic scenario of risk mitigation, is characterized by an altered riskebeneﬁt ratio of radiotherapy for lung cancer patients due to their increased susceptibility for severe COVID-19 infection, and minimization of patient travelling and exposure of radiotherapy staff. The second, a later pandemic scenario, is characterized by reduced radiotherapy resources requiring pa- tient triage. Six common lung cancer cases were assessed for both scenarios: peripherally located stage I NSCLC, locally advanced NSCLC, postoperative radiotherapy after resection of pN2 NSCLC, thoracic radiotherapy and prophylactic cranial irradiation for limited stage SCLC and palliative thoracic radiotherapy for stage IV NSCLC. Background: The COVID-19 pandemic has caused radiotherapy resource pressures and led to increased risks for lung cancer patients and healthcare staff. An international group of experts in lung cancer radiotherapy established this practice recom- mendation pertaining to whether and how to adapt radiotherapy for lung cancer in the COVID-19 pandemic. Matthias Guckenberger, Claus Belka,y Andrea Bezjak,z Jeffrey Bradley,x Megan E. Daly,k Dirk DeRuysscher,{ Rafal Dziadziuszko,# Corinne Faivre-Finn, Michael Flentje,yy Elizabeth Gore,zz Kristin A. Higgins,xx Puneeth Iyengar,kk Brian D. Kavanagh,{{ Sameera Kumar,## Cecile Le Pechoux,* Yolande Lievens,yyy Karin Lindberg,zzz Fiona McDonald,xxx Sara Ramella,kkk Ramesh Rengan,{{{ Umberto Ricardi,### Andreas Rimner,** George B. Rodrigues,yyyy Steven E. Schild,zzzz Suresh Senan,xxxx Charles B. Simone II,kkkk Ben J. Slotman,{{{{ Martin Stuschke,#### Greg Videtic,*** Joachim Widder,yyyyy Sue S. Yom,zzzzz and David Palma yyyy Department of Radiation Oncology, University Hospital Zurich, University of Zurich, Switzerland; yDepartment of Radiation Oncology, LMU Klinikum, LMU Munich, Germany; zDepartment of Radiation Oncology, Princess Margaret Cancer Center, University of Toronto, Canada; xDepartment of Radiation Oncology, Winship Cancer Institute of Emory University, Atlanta; kDepartment of Radiation Oncology, University of California Davis Comprehensive Cancer Center, Sacramento, USA; {Department of Radiation Oncology (Maastro Clinic), Maastricht University Medical Centerþ, GROW Research Institute, The Netherlands; #Dept. of Oncology and Radiotherapy, Medical University of Gdansk, Poland; Division of Cancer Sciences, University of Manchester and The Christie NHS Foundation Trust, Manchester, UK; yyDepartment of Radiation Oncology, University Hospital Wu¨rzburg, JMU Wu¨rzburg, Germany; zzDepartment of Radiation Oncology, Medical College of Wisconsin and Zablocki VAMC, Milwaukee; xxDepartment of Radiation Oncology, Winship Cancer Institute of Emory University, Atlanta; kkDepartment of Radiation Oncology, UT Southwestern Medical Center, Dallas; {{Department of Radiation Oncology, University of Colorado, Aurora; ##Department of Radiation Oncology, Fox Chase Cancer Center, Philadelphia, USA; *Department of Radiation Oncology, Gustave Roussy Cancer Campus, Villejuif, France; yyyDepartment of Radiation Corresponding Author: Matthias Guckenberger; E-mail: Matthias. Guckenberger@usz.ch statements or opinions expressed by the authors of these articles. Practi- tioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds or experiments described herein. Because of rapid advances in the medical sciences, in particular, independent veriﬁcation of diagnoses and drug dosages should be made. For more information see the editorial “Radio- therapy & Oncology during the COVID-19 pandemic”, Vol. 146, 2020. Reprinted from Radiotherapy and Oncology, published by Elsevier B.V. M. Guckenberger, C. Belka, A. Bezjak et al., Practice recommen- dations for lung cancer radiotherapy during the COVID-19 pandemic: An ESTRO-ASTRO consensus statement. Radiotherapy and Oncology 146 (2020) 222-228, https://doi.org/10.1016/j.radonc.2020.04.001. The Editors of the Journal, the Publisher and the European Society for Radiotherapy and Oncology (ESTRO) cannot take responsibility for the https://doi.org/10.1016/j.ijrobp.2020.05.012 International Journal of Radiation Oncology Biology Physics 632 Guckenberger et al. Received Apr 1, 2020. Accepted for publication Apr 1, 2020. Methods On March 22nd, 32 experts in lung cancer radiotherapy were invited to participate in this project, 16 European and 16 US/Canadian experts. By March 24th, 97% had agreed, and a replacement was found for the single invitee who was unavailable, to keep the total at 32 with a balance between groups. All invited participants are co-authors of recent national and international lung cancer practice guidelines or principal investigators of lung cancer clinical trials. A modiﬁed Delphi process was used to establish consensus about whether and how to adapt radiotherapy for lung cancer in the COVID-19 pandemic.16 Surveys were circu- lated to all individual participants using the online survey tool SurveyMonkey. All respondents agreed to participate in a rapid Delphi process, with 24 h to complete each round and successive rounds starting 24 h after the closure of the previous round. Two scenarios of the COVID-19 pandemic were assessed, both of which were already occurring in some geographical regions: In a second (later) pandemic scenario, radiotherapy re- sources may not be available in sufﬁcient quantity for treatment of all patients. A severe shortage of radiotherapy resources may result from sickness or home-quarantine of our department stafﬁng. Service or repair of radiotherapy software and hardware might be restricted or unavailable by radiotherapy vendors. These issues would require the allocation of resources and triage of patients,14,15 in addi- tion to the potential need to make changes to lung cancer radiotherapy prescriptions. This phase is sometimes referred to as the “crisis standard of care”.10 ●Early pandemic scenario 1 e risk mitigation: In this scenario, we asked if respondents would recommend changes to standard practice during the early phase, considering these challenges: The altered riskebeneﬁt ratio of radiotherapy for lung cancer patients due to their increased susceptibility for severe COVID-19 infection, and minimization of patient traveling and exposure of our radiotherapy staff. ●Early pandemic scenario 1 e risk mitigation: In this scenario, we asked if respondents would recommend changes to standard practice during the early phase, considering these challenges: The altered riskebeneﬁt ratio of radiotherapy for lung cancer patients due to their increased susceptibility for severe COVID-19 infection, and minimization of patient traveling and exposure of our radiotherapy staff. In this setting of the COVID-19 pandemic, hypo- fractionation is an attractive treatment option, one that is actively being discussed within the radiotherapy commu- nity on social media platforms such as Twitter and the- Mednet. Matthias Guckenberger, Claus Belka,y Andrea Bezjak,z Jeffrey Bradley,x Megan E. Daly,k Dirk DeRuysscher,{ Rafal Dziadziuszko,# Corinne Faivre-Finn, Michael Flentje,yy Elizabeth Gore,zz Kristin A. Higgins,xx Puneeth Iyengar,kk Brian D. Kavanagh,{{ Sameera Kumar,## Cecile Le Pechoux,* Yolande Lievens,yyy Karin Lindberg,zzz Fiona McDonald,xxx Sara Ramella,kkk Ramesh Rengan,{{{ Umberto Ricardi,### Andreas Rimner,** George B. Rodrigues,yyyy Steven E. Schild,zzzz Suresh Senan,xxxx Charles B. Simone II,kkkk Ben J. Slotman,{{{{ Martin Stuschke,#### Greg Videtic,*** Joachim Widder,yyyyy Sue S. Yom,zzzzz and David Palma yyyy These dramatic developments related to COVID-19 are associated with challenges for the practice of radiation oncology,7,8 especially for radiotherapy of lung cancer patients, who represent one of the highest-risk ESTRO-ASTRO practice recommendations for lung cancer radiotherapy in COVID-19 pandemic 633 Volume 107 Number 4 2020 groups, with high risks of death from both cancer and COVID-19 illness. Oncology (ASTRO), an international group of experts in lung cancer radiotherapy aims to rapidly provide guidance about the potential need to adapt the practice and frac- tionation of radiotherapy for lung cancer in the current COVID-19 pandemic. It may be challenging for radiation oncologists to continue to follow accepted practice guidelines, given these limitations, and delivering standard therapies may even become inappropriate. There are two potential scenarios that may unfold, with different radiation practice patterns.9 y p p In a ﬁrst (early) pandemic scenario, sufﬁcient resources are still available to deliver radiation or multimodality treatment. This is sometimes referred to as the “contin- gency standard of care”.10 However, suppression strategies aiming to slow down the virus spread may also impact the practice of lung cancer radiotherapy due to: a) the need for suppression of the coronavirus and, therefore, the need to minimize the travel of patients11 and exposure of our radiotherapy staff;12 b) an altered riskebeneﬁt ratio of radiotherapy for lung cancer patients due to their increased susceptibility for severe COVID-19 infection when repeatedly leaving their home and traveling to radiotherapy treatment and simultaneously being treated with thoracic radiotherapy. For example, a patient who contracts COVID- 19 during a visit for radiation is at a high risk of morbidity and death due to that visit. In this scenario of the COVID- 19 pandemic, standard-of-care practice of curative or palliative radiotherapy for lung cancer might require ad- aptations and lead to treatment recommendations that are outside current guidelines.13 Matthias Guckenberger,* Claus Belka,y Andrea Bezjak,z Jeffrey Bradley,x Megan E. Daly,k Dirk DeRuysscher,{ Rafal Dziadziuszko,# Corinne Faivre-Finn, Michael Flentje,yy Elizabeth Gore,zz Kristin A. Higgins,xx Puneeth Iyengar,kk Brian D. Kavanagh,{{ Sameera Kumar,## Cecile Le Pechoux,* Yolande Lievens,yyy Karin Lindberg,zzz Fiona McDonald,xxx Sara Ramella,kkk Ramesh Rengan,{{{ Umberto Ricardi,### Andreas Rimner,** George B. Rodrigues,yyyy Steven E. Schild,zzzz Suresh Senan,xxxx Charles B. Simone II,kkkk Ben J. Slotman,{{{{ Martin Stuschke,#### Greg Videtic,*** Joachim Widder,yyyyy Sue S. Yom,zzzzz and David Palma yyyy g p py g Results: In a risk-mitigation pandemic scenario, efforts should be made not to compromise the prognosis of lung cancer pa- tients by departing from guideline-recommended radiotherapy practice. In that same scenario, postponement or interruption of radiotherapy treatment of COVID-19 positive patients is generally recommended to avoid exposure of cancer patients and staff to an increased risk of COVID-19 infection. In a severe pandemic scenario characterized by reduced resources, if pa- tients must be triaged, important factors for triage include potential for cure, relative beneﬁt of radiation, life expectancy, and performance status. Case-speciﬁc consensus recommendations regarding multimodality treatment strategies and fractionation of radiotherapy are provided. Conclusion: This joint ESTRO-ASTRO practice recommendation established pragmatic and balanced consensus recommen- dations in common clinical scenarios of radiotherapy for lung cancer in order to address the challenges of the COVID-19 pandemic. 2020 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). After the outbreak of the coronavirus SARS-CoV-2 (COVID-19) in Wuhan, China, in December 2019,1 the disease rapidly became a global pandemic. Infection rates peaked and began to decline in some Asian countries by March 2020, but Europe and the US are now among the most affected regions.2 Most COVID-19 infections are characterized by only mild symptoms of fever and cough; however, there is a high risk of severe pulmonary infection and death, in particular for the elderly and populations with comorbidities such as diabetes, hyper- tension and cardiopulmonary diseases.3,4 Cancer patients have been reported to be at increased risk of mortality.5 Therefore, many countries have implemented strategies to reduce the risk of spread, aiming to slow-down or “ﬂatten” the infection rate of the coronavirus and to stay within the capacity of the healthcare services, especially intensive care units.6 The pandemic mitigation strategies of most countries also apply to medical care in general and to oncology in particular, and include reduction of elective services, a focus on remote visits (e.g. telemedicine and video), and use of personal protective equipment. However, most health authorities maintain emergency services (e.g. for accidents) and services for diagnosis and treatment of severe diseases such as cancer. Methods 634 Table 1 The six lung cancer cases described, including the diagnosis and the presumed standard guideline-recommended therapy Case 1: Stage I NSCLC New diagnosis of stage I, inoperable, peripherally located NSCLC Institutional standard fractionation of SBRT according to NCCN: 3e4 Fx total dose 45e54 Gy Case 2: Stage III NSCLC Locally advanced stage IIIA (bulky N2) NSCLC Standard fractionation of radiochemotherapy: 30e33 Fx over 6e6.5 weeks, total dose 60e66 Gy Case 3: PORT NSCLC Resected N2 (multi-station and extra nodal spread) NSCLC Standard fractionation of radiotherapy: 27 Fx over 5.5 weeks, total dose 54 Gy Case 4: LS SCLC SCLC, limited stage Standard fractionation of radiochemotherapy: 30 Fx over 3 weeks, BID, total dose 45 Gy, OR 33 Fx over 6.5 weeks, total dose 66 Gy Case 5: PCI LS SCLC PCI for SCLC limited stage after good response to radiochemotherapy Standard fractionation of radiotherapy: 10 Fx over 2 weeks, total dose 25 Gy Case 6: palliative NSCLC Palliative metastatic NSCLC with failure after ﬁrst-line chemo-IO combination and symptoms due to mediastinal/hilar disease progression and severe cough and moderate dyspnea Standard fractionation of radiotherapy: 10 Fx over 2 weeks, total dose 30 Gy Standard fractionation of radiotherapy: 10 Fx over 2 weeks, total dose 25 Gy Palliative metastatic NSCLC with failure after ﬁrst-line chemo-IO combination and symptoms due to mediastinal/hilar disease progression and severe cough and moderate dyspnea Standard fractionation of radiotherapy: 10 Fx over 2 weeks, total dose 30 Gy The questions for the ﬁrst round of the Delphi process are shown in Table 2. All responses were analyzed and consolidated by two investigators (MG and DAP). A threshold of 66% for agreement or disagreement was required for each item to reach consensus and a threshold of 80% for strong consensus. For questions voting on prioritizing the cases, the results of the vote are presented without necessarily achieving consensus. In the second and third rounds, participants received the results and summary of comments from prior rounds, and were asked to vote again on items that had not reached 66% agreement. New questions were constructed (by MG and DAP) in order to gain clariﬁcation or to raise issues noted in the comments from participants. Following the third round, any items still lacking consensus were not considered a recommendation, but some important issues raised by a large minority of respondents are reﬂected below. Methods However, the results of such ad hoc discussions do not address the needs of our radiotherapy community in an optimal way. Individual opinions may not be clinically appropriate and might expose cancer patients to potential harm from suboptimal radiotherapy practice. On the other end of the spectrum, adoption of appropriate hypofractio- nation might be low if that practice is outside of interna- tional guidelines and not endorsed by recognized experts and professional societies, yet such guidelines often take many months to develop. ●Later pandemic scenario 2 e reduced radiotherapy re- sources: In this scenario, we asked respondents to consider how their above recommendations from sce- nario 1 would change if a lack of radiotherapy resources prevented some patients at their centre from receiving radiation treatment. Six common lung cancer cases were assessed for both pandemic scenarios (Table 1). For all six, we assumed a patient with average / standard characteristics for the lung cancer population. The standard treatment for each case was provided, consistent with guideline-recommended radiotherapy according to current versions of the National Comprehensive Cancer Network (NCCN), ESTRO, ASTRO and European Organization for Research and Treatment of Cancer (EORTC) guidelines. In this practice recommendation, which is endorsed by the European Society for Radiotherapy and Oncology (ESTRO) and the American Society for Radiation International Journal of Radiation Oncology Biology Physics Guckenberger et al. Results A total of three Delphi rounds were conducted. Surveys remained open for 24 h and response rates were 29/32 (March 23rd, round 1), 31/32 (March 25th, round 2) and 30/ 32 (March 27th, round 3). If a decision was made for hypofractionation beyond standard fractionations, there was strong consensus for using a single fraction SBRT of 30e34 Gy in case 1 (stage I NSCLC). In case 6 (palliative NSCLC), palliative regimens in 5 fractions, 2 fractions and a 1 fraction all had similar support. Question: Would you recommend hypofractionating beyond your usual fractionation? In the early phase of the COVID-19 pandemic, there was consensus not to universally change radiotherapy practice to more hypofractionated regimens (Table 4). There was consensus or strong consensus not to change to more hypofractionated approaches in case 3 (PORT NSCLC), case 4 (LS SCLC) and case 5 (PCI SCLC). In contrast, there was strong consensus to change to more hypo- fractionation in case 6 (palliative NSCLC). Methods For case 1 (stage I NSCLC), answers on postponement were balanced and we asked for factors inﬂuencing the decision whether or not to postpone. There was strong consensus that tumor growth rate (87%) should be used in the decision-making process and some support for these other factors (33e66%): patient preference, solid compo- nent vs GGO, patient performance status, T1 vs T2, current and future status of pandemic. For case 5 (PCI SCLC) we asked about regular contrast- enhanced cranial MRI follow up as an alternative to PCI: this strategy was supported by 46% without reaching consensus. Question: Would you recommend hypofractionating beyond your usual fractionation? Early phase of the COVID-19 pandemic: risk mitigation Case 1 Case 1B: An operable patient with stage I NSCLC is referred to you by a thoracic surgeon because timely access to surgery is not available due to surgical capacity issues. Would you treat with SABR/SBRT? Case 2 Would you recommend starting with induction chemotherapy to postpone the start of radiation? Later pandemic scenario 2 e reduced radiotherapy resources All cases How highly would you prioritize this patient’s treatment compared to all other cancer patients in your centre? All cases If there was a critical shortage of RT capacity, would you recommend further hypofractionation beyond what you have described above? All cases If you answered yes to the question above, what would be the maximum degree of hypofractionation you would propose to a patient in your clinical service? Specify the total dose, number of fractions, total treatment time, and provide any pertinent references if available All cases In the setting of reduced RT capacity, if this patient was COVID-19 positive before the start of treatment, what would be the maximum duration to postpone the initiation of radiotherapy (in weeks)? All cases In the setting of reduced RT capacity, if this patient became COVID-19 positive after starting treatment, would you recommend interrupting RT until the patient becomes asymptomatic and the test for COVID-19 negative? Overall Please rank the six cases in order of priority, starting with the highest-priority case, in the setting of reduced resources Overall If you were to triage patients for treatment, in the setting of reduced RT resources, please provide up to 5 factors that you would use to decide who gets treatment, in order of importance Table 2 Questions in the ﬁrst round of the Delphi process Early pandemic scenario 1 e risk mitigation All cases Do you recommend that physicians change their radiotherapy practice to address the challenges in this early phase of the COVID-19 pandemic? (i.e. risks due to multiple visits, susceptibility of lung cancer patients to COVID-19 morbidity/mortality) All cases Would you recommend postponing the initiation of treatment by 4e6 weeks? All cases Would you recommend hypofractionating beyond your usual fractionation? Case 1e3 Would your answers to questions #2 and #3 above change if the tumor was mutation positive (EGFR or ALK) or PD- L1 positive (i.e. >50%)? Case 2 Would you recommend induction therapy in this case? Early phase of the COVID-19 pandemic: risk mitigation Question: Would you recommend postponing the initiation of treatment by 4e6 weeks? For case 2 (stage III NSCLC), we further differentiated fractionations based on whether the patient was treated with radiotherapy only, with sequential radiochemotherapy or concomitant radiochemotherapy. There was strong consensus that hypofractionated radiotherapy is appropriate in radiotherapy alone or sequential radiochemotherapy; however, there was consensus against hypofractionation in concomitant radiochemotherapy (Table 5). Various frac- tionations were considered as appropriate, with total doses between 50 Gy and 66 Gy delivered in 15e30 fractions. In the early phase of the COVID-19 pandemic, decisions on delay of treatment depended on the clinical case (Table 3). There was a strong consensus not to postpone curative treatment for case 2 (stage III NSCLC), case 4 (LS SCLC) and case 6 (palliative NSCLC). In contrast, there was a strong consensus to postpone treatment for case 3 (PORT NSCLC) and a consensus to postpone for case 5 (PCI SCLC). Volume 107 Number 4 2020 ESTRO-ASTRO practice recommendations for lung cancer radiotherapy in COVID-19 pandemic ESTRO-ASTRO practice recommendations for lung cancer radiotherapy in COVID-19 pandemic Volume 107 Number 4 2020 635 Table 2 Questions in the ﬁrst round of the Delphi process Early pandemic scenario 1 e risk mitigation All cases Do you recommend that physicians change their radiotherapy practice to address the challenges in this early phase of the COVID-19 pandemic? (i.e. risks due to multiple visits, susceptibility of lung cancer patients to COVID-19 morbidity/mortality) All cases Would you recommend postponing the initiation of treatment by 4e6 weeks? All cases Would you recommend hypofractionating beyond your usual fractionation? Case 1e3 Would your answers to questions #2 and #3 above change if the tumor was mutation positive (EGFR or ALK) or PD- L1 positive (i.e. >50%)? Case 2 Would you recommend induction therapy in this case? All cases If you recommended hypofractionation, what would be the maximum degree of hypofractionation you would propose to a patient in your clinical service? Specify the total dose, number of fractions, total treatment time, and provide any pertinent references if available. All cases If this patient was COVID-19 positive before starting treatment, would you postpone RT until the patient becomes asymptomatic and the test for COVID-19 negative? All cases If this patient became COVID-19 positive after starting treatment, would you recommend interrupting RT until the patient becomes asymptomatic and the test for COVID-19 negative? Early phase of the COVID-19 pandemic: risk mitigation All cases If you recommended hypofractionation, what would be the maximum degree of hypofractionation you would propose to a patient in your clinical service? Specify the total dose, number of fractions, total treatment time, and provide any pertinent references if available. All cases If this patient was COVID-19 positive before starting treatment, would you postpone RT until the patient becomes asymptomatic and the test for COVID-19 negative? All cases If this patient became COVID-19 positive after starting treatment, would you recommend interrupting RT until the patient becomes asymptomatic and the test for COVID-19 negative? Case 1 Case 1B: An operable patient with stage I NSCLC is referred to you by a thoracic surgeon because timely access to surgery is not available due to surgical capacity issues. Would you treat with SABR/SBRT? Case 2 Would you recommend starting with induction chemotherapy to postpone the start of radiation? Later pandemic scenario 2 e reduced radiotherapy resources Later pandemic scenario 2 e reduced radiotherapy resources All cases How highly would you prioritize this patient’s treatment compared to all other cancer patients in your centre? All cases If there was a critical shortage of RT capacity, would you recommend further hypofractionation beyond what you have described above? All cases If you answered yes to the question above, what would be the maximum degree of hypofractionation you would propose to a patient in your clinical service? Specify the total dose, number of fractions, total treatment time, and provide any pertinent references if available All cases In the setting of reduced RT capacity, if this patient was COVID-19 positive before the start of treatment, what would be the maximum duration to postpone the initiation of radiotherapy (in weeks)? All cases In the setting of reduced RT capacity, if this patient became COVID-19 positive after starting treatment, would you recommend interrupting RT until the patient becomes asymptomatic and the test for COVID-19 negative? Early phase of the COVID-19 pandemic: risk mitigation Overall Please rank the six cases in order of priority, starting with the highest-priority case, in the setting of reduced resources Overall If you were to triage patients for treatment, in the setting of reduced RT resources, please provide up to 5 factors that you would use to decide who gets treatment, in order of importance e to triage patients for treatment, in the setting of reduced RT resources, please provide up to 5 factors that uld use to decide who gets treatment, in order of importance Question: An operable patient with stage I NSCLC is referred to you by a thoracic surgeon because timely access to surgery is not available due to surgical capacity issues. Would you treat with SBRT? Table 3 Recommendations regarding postponement of treatment Would you recommend postponing the initiation of treatment by 4e6 weeks? Case Response Case 1: stage I NSCLC Yes: 43% No: 57% Case 2: stage III NSCLC Yes: 4% No: 96% (strong consensus) Case 3: PORT NSCLC Yes: 82% (strong consensus) No: 18% Case 4: LS SCLC Yes: 11% No: 89% (strong consensus) Case 5: PCI SCLC Yes: 70% (consensus) No: 30% Case 6: Palliative NSCLC Yes: 4% No: 96% (strong consensus) Surgical capacities might become especially at risk because of the strong need for intensive care and ventilators in patients with severe COVID-19 infection. We therefore addressed a situation where an operable patient with stage I NSCLC is referred to radiation oncology by a thoracic surgeon because timely access to surgery is not available due to surgical capacity issues: it was asked whether treatment with SBRT would be offered. There was a 100% consensus to offer SBRT. Question: Which multi-modality strategies would you consider as reasonable in order to address the challenges in this early phase of the COVID-19 pandemic? This question was asked for the curative stage III NSCLC (case 2) only. It was explicitly described that the patient International Journal of Radiation Oncology Biology Physics Guckenberger et al. 636 Table 4 Fractionation recommendations Would you recommend hypofractionating beyond your usual fractionation? Early phase of the COVID-19 pandemic: risk mitigation Case Standard fractionations Response Maximum degree of hypofractionation supported Case 1: stage I NSCLC SBRT: 45e54 Gy in 3 Fx, 48 Gy in 4 fractions Yes: 50% No: 50% 30e34 in 1 Fx17: 90% support if choosing hypofractionation (strong consensus) Case 2: stage III NSCLC Radiochemotherapy: 60e66 Gy in 30e33 Fx over 6e6.5 weeks Yes: 46% No: 54% Case 3: PORT NSCLC PORT: 50e60 Gy over 5e6 weeks Yes: 29% No: 71% (consensus) Case 4: LS SCLC Radiochemotherapy: 60e66 Gy in 30e33 Fx over 6e6.5 weeks, or 45 Gy in 30 Fx over 3 weeks using BID fractions of 1.5 Gy Yes: 33% No: 67% (consensus) Case 5: PCI SCLC PCI: 25 Gy in 10 Fx over 2 weeks Yes: 7% No: 93% (strong consensus) Case 6: Palliative NSCLC 30 Gy in 10 Fx over 2 weeks Yes: 89% (strong consensus) No: 11% Favored fractionations: 20 Gy in 5 Fx (30%)18 17 Gy in 2 Fx (37%)19 8e10 Gy in 1Fx (33%)20 Would you recommend hypofractionating beyond your usual fractionation? does not have any contraindications against the guideline recommended standard-of-care concomitant radio- chemotherapy. Concurrent radiochemotherapy achieved strong consensus as the preferred treatment strategy. Radiotherapy alone, chemotherapy followed by radio- therapy and chemotherapy followed by radiochemotherapy were not considered as reasonable treatment strategies by >33% of participants. We also asked if respondents would recommend against any standard concurrent chemotherapy agents (e.g. cisplatin- etoposide, cisplatin-vinka alkaloid, cisplatin-pemetrexed, carboplatin-paclitaxel, or carboplatin monotherapy), and Table 5 Recommended hypofractionation regimens based on availability/use of concurrent and sequential radiochemotherapy, or radiotherapy alone Would you consider hypofractionated radiotherapy as appropriate? Case 2 stage III NSCLC Response Maximum degree of hypofractionation supported Radiotherapy only Yes: 97% (strong consensus) No: 3% 60 Gy in 15 Fx (33%)21,22 60 Gy in 20 Fx (27%)23 60e66 Gy in 24e30 Fx (2.2e2.75 Gy/day) (23%)24 55 Gy in 20 Fx (13%)25 None (3%) Sequential radiochemotherapy Yes: 97% (strong consensus) No: 3% 60e66 Gy in 24e30 Fx (2.2e2.75 Gy/day) (27%)24 55 Gy in 20 Fx (27%)25 60 Gy in 15 Fx (23%)21,22 60 Gy in 20 Fx (20%)23 None (3%) Concomitant radiochemotherapy Yes: 27% No: 73% (consensus) See footnote* * Although there was consensus not to recommend hypofractionation, the respondents supportive of hypofractionation (n Z 11) were asked which fractionation(s) they would support, with multiple answers allowed. Early phase of the COVID-19 pandemic: risk mitigation The favored options were 60e66 Gy in 22e30 Fx, given at 2.2e2.75 Gy/day, (75%) and 55 Gy in 20 Fx (63%). Table 5 Recommended hypofractionation regimens based on availability/use of concurrent and sequential radiochemotherapy, or radiotherapy alone Would you consider hypofractionated radiotherapy as appropriate? Concomitant radiochemotherapy Concomitant radiochemotherapy Yes: 27% No: 73% (consensus) * Although there was consensus not to recommend hypofractionation, the respondents supportive of hypofractionation (n Z 11) were asked which fractionation(s) they would support, with multiple answers allowed. The favored options were 60e66 Gy in 22e30 Fx, given at 2.2e2.75 Gy/day, (75%) and 55 Gy in 20 Fx (63%). * Although there was consensus not to recommend hypofractionation, the respondents supportive of hypofractionation (n Z 11) were asked which fractionation(s) they would support, with multiple answers allowed. The favored options were 60e66 Gy in 22e30 Fx, given at 2.2e2.75 Gy/day, (75%) and 55 Gy in 20 Fx (63%). ESTRO-ASTRO practice recommendations for lung cancer radiotherapy in COVID-19 pandemic Volume 107 Number 4 2020 ESTRO-ASTRO practice recommendations for lung cancer radiotherapy in COVID-19 pandemic 637 Volume 107 Number 4 2020 there was no consensus to recommend against any of these. The carboplatin-paclitaxel regimen was chosen as a regimen of concern most often (by 37% of respondents), potentially due to risks of myelosuppression and/or pneumonitis. there was no consensus to recommend against any of these. The carboplatin-paclitaxel regimen was chosen as a regimen of concern most often (by 37% of respondents), potentially due to risks of myelosuppression and/or pneumonitis. were described as relevant in the decision-making process of whether or not to interrupt radiotherapy in patients diagnosed as COVID-19 positive: COVID-19 related symptoms, symptoms of lung cancer, and infection with COVID-19 near the end of treatment. Question: Would your multi-modality treatment strategy change if the tumor was mutation positive (EGFR or ALK) or highly PD-L1 positive (i.e. > 50%)? (Cases 1e3) There was strong consensus (96%) not to change the treatment strategy for case 1 (stage I NSCLC) and almost consensus (64%) for case 3 (PORT NSCLC). Question: Would your multi-modality treatment strategy change if the tumor was mutation positive (EGFR or ALK) or highly PD-L1 positive (i.e. > 50%)? (Cases 1e3) Question: Handling of COVID-19 positive patients? There was consensus in all cases to postpone initiation of radiotherapy until the patient becomes asymptomatic and the test for COVID-19 becomes negative (Table 6). When patients are diagnosed as COVID-19 positive during radiotherapy treatment, there was consensus to interrupt radiotherapy until the patient becomes asymptomatic and the test for COVID-19 is negative in cases 3, 5 and 6 whereas opinions were evenly split for the cases with curative radiotherapy at the time of primary diagnosis (cases 1, 2 and 4). Later phase of the COVID-19 pandemic: lack of radiotherapy resources and need for patient triage There was strong consensus (96%) not to change the treatment strategy for case 1 (stage I NSCLC) and almost consensus (64%) for case 3 (PORT NSCLC). For the later phase of the COVID-19 pandemic with poten- tially reduced radiotherapy resources, we addressed the questions whether further hypofractionation or postponement of radiotherapy for COVID-19 positive patients would be considered as reasonable and how to prioritize and triage pa- tients. Results are summarized in Table 7. For case 3 (PCI SCLC) there was strong consensus (83%) for regular contrast- enhancedcranial MRI followupinstead ofPCI.Availability of MRI may, however, be limited during a pandemic situation. For case 2 (stage III NSCLC), we asked about induction strategies to postpone the start of radiotherapy for the populations described in this question, although these are not yet evidence-based treatment options. There was limited support but no consensus to consider induction EGFR-targeting TKI for EGFR mutated NSCLC or in- duction ALK-targeting TKI for NSCLC with ALK rear- rangement (38%); induction chemo-IO for cancers highly PD-L1 positive was not supported (17%). Regarding postponement or interruption of treatment for COVID-19 positive patients during scenario 2, since the consensus in Scenario 1 was to postpone and interrupt in all situations, it was concluded that treatment would be postponed/ interrupted until the patient recovers and is COVID-19 negative in Scenario 2 as well, since it is a more extreme example. For case 3 (PORT NSCLC) we asked about EGFR/ALK targeting TKIs and about immune-checkpoint inhibition (chemotherapy) as options to postpone radiotherapy or as alternatives to radiotherapy (although these are not yet evidence-based treatment options): none of these strategies was supported by >25% of the participants. Prioritization of cases and triage of patients Table 8 shows the ranking of cases based on relative pri- ority, their perceived priority relative to all other types of cancer cases, and the top 5 factors recommended in order to triage patients in a setting where not all patients can receive radiotherapy due to capacity shortages. Prioritization of cases and triage of patients Discussion Table 7 Recommendations regarding hypofractionation of treatment in the later phase of the COVID-19 pandemic characterized by a lack of radiotherapy resources p p characterized by a lack of radiotherapy resources Case Maximum hypofractionation considered as appropriate (66% threshold) Case 1: stage I NSCLC 30e34 Gy in 1 Fx Case 2: stage III NSCLC 55e60 Gy in 20 Fx Case 3: PORT NSCLC Consensus against hypofractionation Case 4: LS SCLC 40e45 Gy in 15 Fx Case 5: PCI SCLC Consensus against hypofractionation Case 6: Palliative NSCLC 8e10 Gy in 1 Fx This ESTRO-ASTRO practice recommendation used methodologies that are established quality indicators for regular consensus and guideline processes:26 the practice recommendation was ofﬁcially endorsed by the ESTRO and ASTRO societies, a sufﬁciently large group of inter- national experts in lung cancer radiotherapy contributed to this recommendation, the modiﬁed Delphi process started with open questions aiming to comprehensively collect the knowledge and opinions of all participants and consensus was established by follow-up rounds of feedback and voting. A systematic review was not part of the practice recommendation due to time constraints, and especially due to a lack of evidence for pandemic situations. common lung cancer cases within the context of two different scenarios of the COVID-19 pandemic. Beyond detailed recommendations shown above, the following three take-home messages should be considered in lung cancer radiotherapy. py First, in a risk-mitigation pandemic scenario where radiotherapy resources remain available, efforts should be made to not compromise the prognosis of lung cancer pa- tients by departing from guideline-recommended radio- therapy practice. Second, in that same scenario, postponement or interruption of radiotherapy treatment of COVID-19 positive patients should be considered to avoid exposure of cancer patients and staff to an increased risk of COVID-19 infection. Third, in a severe pandemic scenario characterized by reduced resources, if patients must be triaged, important factors included potential for cure, relative beneﬁt of radiation, life expectancy, and perfor- mance status. All co-authors therefore encourage practitioners to consider the results of this ESTRO-ASTRO practice recommendation on whether and how to adapt radiotherapy for lung cancer to the COVID-19 pandemic. Discussion This Delphi process was able to achieve consensus in many important aspects of lung cancer radiotherapy in the current COVID-19 pandemic. A total of 32 interna- tional experts in lung cancer radiotherapy completed 3 rounds of a consensus-building process and addressed six For case 2 (stage III NSCLC) and case 4 (LS SCLC) with longer radiotherapy treatments, the following factors Table 6 Recommendations on delay or interruption of treatment in COVID-19 positive patients Patient case Time patient is diagnosed as COVID-19 positive Postpone or interrupt RT? Case 1: Stage I NSCLC Start of Tx Yes: 96% (Strong consensus) After start of Tx Yes: 54% Case 2: Stage III NSCLC Start of Tx Yes: 100% (Strong consensus) After start of Tx Yes: 57% Case 3: PORT NSCLC Start of Tx Yes: 96% (Strong consensus) After start of Tx Yes: 68% (Consensus) Case 4: LS SCLC Start of Tx Yes: 89% (Strong consensus) After start of Tx Yes: 48% Case 5: PCI SCLC Start of Tx Yes: 93% (Strong consensus) After start of Tx Yes: 67% (consensus) Case 6: Palliative NSCLC Start of Tx Yes: 74% (Consensus) After start of Tx Yes: 78% (Consensus) Table 6 Recommendations on delay or interruption of treatment in COVID-19 positive patients Patient case Time patient is diagnosed as COVID-19 positive International Journal of Radiation Oncology Biology Physics Guckenberger et al. 638 This joint ESTRO-ASTRO practice recommendation aims to provide rapid, pragmatic and balanced guidance in common clinical scenarios of radiotherapy for lung cancer. Practitioners must use their clinical judgement when considering how these consensus statements apply to their individual clinical practice. These consensus statements are not absolute clinical practice recommendations. Clinical decisions should take into account all clinical factors, and in some settings the consensus recommendations may not be appropriate. The decision-making process will be inﬂuenced by various stakeholders (governments, health care authorities, hospital and university administration), will be restricted by logistical and ﬁnancial aspects, will need to follow the appropriate legal frameworks, and will need to be put into political and cultural context. The ability to implement hypofractionation may depend on departmental resources available (e.g. physicist). ESTRO Disclaimer ESTRO cannot endorse all statements or opinions made on the guidelines. Regardless of the vast professional knowl- edge and scientiﬁc expertise in the ﬁeld of radiation oncology that ESTRO possesses, the Society cannot inspect all information to determine the truthfulness, accuracy, reliability, completeness or relevancy thereof. Under no circumstances will ESTRO be held liable for any decision taken or acted upon as a result of reliance on the content of the guidelines. The component information of the guide- lines is not intended or implied to be a substitute for pro- fessional medical advice or medical care. The advice of a medical professional should always be sought prior to commencing any form of medical treatment. To this end, all component information contained within the guidelines is done so for solely educational and scientiﬁc purposes. ESTRO and all of its staff, agents and members disclaim any and all warranties and representations with regards to the information contained on the guidelines. This includes any implied warranties and conditions that may be derived from the aforementioned guidelines. R. Dziadziuszko: AstraZeneca, Pﬁzer, Roche, Seattle Genetics, Takeda Advisory Boards. C. Faivre-Finn: research funding from AstraZeneca and Elekta and Astra Zeneca Advisory Boards. M. Guckenberger: research funding from AstraZeneca. K. Higgins: Astra Zenenca Advisory Board/Consultant, Genetech Advisory Board, Varian Consultant. Research Funding ReﬂeXion Medical. C. Le Pechoux: Participation in Astra Zeneca, Nano- biotix and Roche Advisory Boards, Participation in educational meetings Amgen, Astra Zeneca, Lilly, Med- scape, PrimeOncology. Y. Lievens: personal fees from AstraZeneca and RaySearch. Y. Lievens: personal fees from AstraZeneca and RaySearch. U. Ricardo: Astra Zeneca Advisory Board. U. Ricardo: Astra Zeneca Advisory Board. A. Rimner: Research funding from Varian Medical Systems, Boehringer Ingelheim, Pﬁzer, AstraZeneca, Merck; Personal fees from: AstraZeneca, Merck, Research to Practice, Cybrexa, More Health. A. Rimner: Research funding from Varian Medical Systems, Boehringer Ingelheim, Pﬁzer, AstraZeneca, Merck; Personal fees from: AstraZeneca, Merck, Research to Practice, Cybrexa, More Health. S. Schild: UpToDate. S. Schild: UpToDate. S. Senan: consulting/advisory fees from AstraZeneca, Merck, Celgene, and departmental research funding from ViewRay Inc., Varian Medical Systems and AstraZeneca. References 1. Li Q, et al. Early transmission dynamics in Wuhan, China, of novel coronavirus-infected pneumonia. N Engl J Med 2020. 2. Rosenbaum L. Facing Covid-19 in Italy d Ethics, Logistics, and Therapeutics on the Epidemic’s Front Line. N Engl J Med 2020. 3. Guan W, et al. Clinical Characteristics of Coronavirus Disease 2019 in China. N Engl J Med 2020. 4. Zhou F, et al. Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study. The Lancet 2020. 5. Xia Y, Jin R, Zhao J, Li W, Shen H. Risk of COVID-19 for cancer patients. Lancet Oncol 2020. 6. Ferguson NM, et al. Impact of non-pharmaceutical interventions (NPIs) to reduce COVID-19 mortality and healthcare demand 2020: 20. 7. Filippi AR, Russi E, Magrini SM, Corvo` R. COVID-19 outbreak in northern Italy: ﬁrst practical indications for radiotherapy departments. Int J Radiat Oncol Biol Phys 2020. 8. Krengli M, Ferrara E, Mastroleo F, Brambilla M, Ricardi U. Running a Radiation Oncology Department at the time of coronavirus: an Italian experience. Adv Radiat Oncol 2020. ASTRO Disclaimer Disclaimer and Adherence: American Society for Radi- ation Oncology (ASTRO) guidelines present scientiﬁc, health, and safety information and may reﬂect scientiﬁc or medical opinion. They are available to ASTRO members and the public for educational and informa- tional purposes only. Commercial use of any content in this guideline without the prior written consent of ASTRO is strictly prohibited. Adherence to this guideline does not ensure successful treatment in every situation. This guideline should not be deemed inclusive of all proper methods of care or exclusive of other methods reasonably directed to obtaining the same results. The physician must make the ultimate judgment regarding therapy considering all circumstances presented by the patient. ASTRO assumes no liability for the information, conclusions, and ﬁndings contained in its guidelines. This guideline cannot be assumed to apply to the use of these interventions performed in the context of clinical trials. This guideline is based on information available at the time the task force conducted its research and discussions on this topic. There may be new developments that are not reﬂected in this guideline and that may, over time, be a basis for ASTRO to revisit and update the guideline. M. Stuschke: research funding from AstraZeneca. S. Yom: research funding from Genentech, Merck, Bristol-Myers Squibb, BioMimetix; The other authors declare no conﬂicts of interest. The other authors declare no conﬂicts of interest. ESTRO-ASTRO practice recommendations for lung cancer radiotherapy in COVID-19 pandemic ESTRO-ASTRO practice recommendations for lung cancer radiotherapy in COVID-19 pandemic 6 639 Volume 107 Number 4 2020 cancer care in serious situations such as now e take care of yourselves as well as your patients. J. Bradley: AstraZeneca Scientiﬁc Advisory Board; Mevion Medical Systems Scientiﬁc Advisory Board; Genentech Scientiﬁc Advisory Board. M. Daly: Research funding from EMD Serono and Genentech, Boston Scientiﬁc Advisory Board, Triptych Health Partners consulting. Discussion Finally, we want to express our gratitude to all colleagues of all pro- fessions and disciplines who continue delivering optimal Table 8 Prioritization of lung cancer patients and factor for triaging of patients Prioritization of lung cancer patients Top 5 factors for triaging patients across all radiotherapy cases Rank Case* Relative Priority Compared All Other Types Cancer Cases in Department** 1. Stage III NSCLC Very high/high (71% consensus) 1. Potential for cure 2. LS-SCLC SCLC Very high/high (78% consensus) 2. Relative beneﬁt of RT vs. other treatment options 3. Stage I NSCLC High/average (near consensus: 65%) 3. Active COVID-19 infection (absence thereof) 4. Palliative NSCLC No consensus. Widely dispersed responses. 4. Life expectancy 5. PORT NSCLC Low/very low (68% consensus) 5. Performance Status 6. SCLC PCI Low/very low (81% consensus) * The six cases were ranked, with 6 points given for a #1 ranking, 5 points for #2, etc, and the average number of points was determined. The average scores, in order of ranking as listed in the table, were 5.2, 4.9, 4.1, 3.0, 2.1 and 1.7, respectively. ** Respondents were asked to prioritize each case as very high, high, average, low, or very low, corresponding to quintiles of priority (e.g. very high Z top 20%, very low Z bottom 20%), compared to all types of cancers treated in their department. Adjacent categories were combined to determine consensus. Table 8 Prioritization of lung cancer patients and factor for triaging of patients Top 5 factors for triaging patients across all radiotherapy cases Conﬂict of interest disclosures 9. Zhao Z, Bai H, Duan JC, Wang J. Individualized treatment recom- mendations for lung cancer patients at different stages of treatment A. Bezjak: AstraZeneca advisory board. Guckenberger et al. 640 International Journal of Radiation Oncology Biology Physics during the outbreak of 2019 novel coronavirus disease epidemic. Zhonghua Zhong Liu Za Zhi 2020;42:E007. during the outbreak of 2019 novel coronavirus disease epidemic. Zhonghua Zhong Liu Za Zhi 2020;42:E007. fraction radiotherapy for patients with inoperable non-small cell lung cancer and good performance status. Clin Oncol 1996;8: 167-175. fraction radiotherapy for patients with inoperable non-small cell lung cancer and good performance status. Clin Oncol 1996;8: 167-175. 10. Stroud C. Institute of Medicine (U.S.) & Forum on Medical and Public Health Preparedness for Catastrophic Events. Crisis standards of care: summary of a workshop series, National Academies Press: 2010. 20. Medical A. Research Council (MRC) randomised trial of palliative radiotherapy with two fractions or a single fraction in patients with inoperable non-small-cell lung cancer (NSCLC) and poor performance status. Medical Research Council Lung Cancer Working Party. Br J Cancer 1992;65:934-941. 11. Chinazzi M, et al. The effect of travel restrictions on the spread of the 2019 novel coronavirus (COVID-19) outbreak. Science 2020;. https: //doi.org/10.1126/science.aba9757. 12. Lancet T. COVID-19: protecting health-care workers. The Lancet 2020;395:922. 21. Westover KD, et al. Precision hypofractionated radiation therapy in poor performing patients with non-small cell lung cancer: phase 1 dose escalation trial. Int J Radiat Oncol 2015;93:72-81. 13. Francesco C, Pettke A, Michele B, Fabio P, Helleday T. Managing COVID-19 in the oncology clinic and avoiding the distraction effect. Ann Oncol 2020;S0923753420363730. https://doi.org/10.1016/j. annonc.2020.03.286. 22. Iyengar P, et al. A phase III randomized study of image guided con- ventional (60 Gy/30 fx) versus accelerated, hypofractionated (60 Gy/15 fx) radiation for poor performance status stage II and III NSCLC patientsdan interim analysis. Int J Radiat Oncol Biol Phys 2016;96:E451. 14. Christian MD. Triage Crit Care Clin 2019;35:575-589. 15. Emanuel EJ, et al. Fair allocation of scarce medical resources in the time of Covid-19. N Engl J Med 2020. 23. Osti MF, et al. Image guided hypofractionated 3-dimensional radiation therapy in patients with inoperable advanced stage non-small cell lung cancer. Int J Radiat Oncol 2013;85:e157-e163. 16. Hsu C-C, Sandford BA. The Delphi Technique: Making Sense of Consensus. 12, 8. 24. Belderbos J, et al. Conﬂict of interest disclosures Randomised trial of sequential versus concurrent chemo-radiotherapy in patients with inoperable non-small cell lung cancer (EORTC 08972e22973). Eur J Cancer 2007;43:114-121. 17. Videtic GM, et al. Long-term follow-up on NRG oncology RTOG 0915 (NCCTG N0927): A randomized phase 2 study comparing 2 stereotactic body radiation therapy schedules for medically inoperable patients with stage I peripheral non-small cell lung cancer. Int J Radiat Oncol Biol Phys 2019;103:1077-1084. 25. Maguire J, et al. SOCCAR: a randomised phase II trial comparing sequential versus concurrent chemotherapy and radical hypofractionated radiotherapy in patients with inoperable stage III Non-Small Cell Lung Cancer and good performance status. Eur J Cancer 2014;50:2939-2949. 18. Bezjak A, et al. Randomized phase III trial of single versus fraction- ated thoracic radiation in the palliation of patients with lung cancer (NCIC CTG SC.15). Int J Radiat Oncol 2002;54:719-728. 26. Ko¨tter T, Blozik E, Scherer M. Methods for the guideline-based development of quality indicatorsea systematic review. Implement Sci 2012;7:21. 19. Medical Research Council Lung Cancer Working Party, et al. Ran- domized trial of palliative two-fraction versus more intensive 13-
https://openalex.org/W4390713801	https://cimc.knu.ua/article/download/887/1419	English	null	The Role and Development of Popular Science TV Programs During Times of MilitaryActions	Aktualʹnì pitannâ masovoì̈ komunìkacìì̈	2,023	cc-by	7,085	R E S E A R C H A R T I C L E S R E S E A R C H A R T I C L E S The Role and Development of Popular Science TV Programs During Times of Military Actions Roksolana Kravchenko PhD student Department of TV and Radio, Scientific and Educational Institute of Journalism, Taras Shevchenko National University of Kyiv, Ukraine https://orcid.org/0000-0001-9063-4282 Email: lisovskalana@gmail.com PhD student Department of TV and Radio, Scientific and Educational Institute of Journalism, Taras Shevchenko National University of Kyiv, Ukraine https://orcid.org/0000-0001-9063-4282 Email: lisovskalana@gmail.com https://doi.org/10.17721/CIMC.2023.34.71-82 UDC 001.92:621.397.13:316.485.26 Citation: Kravchenko, R. (2023). The role and development of popular science TV programs during times of military actions. Current Issues of Mass Communication, 34, 71-82. https://doi.org/10.17721/CIMC.2023.34.71-82 ISSN 2312-5160 online ISSN 2786-4502 ISSN 2312-5160 online ISSN 2786-4502 online ISSN Current Issues of Mass Communication, Issue 34 (2023) Journal homepage: https://cimc.knu.ua/index © 2023 Current Issues of Mass Communication ABSTRACT ARTICLE INFO Received: 25 July 2023 Accepted: 22 November 2023 Published: 30 December 2023 The article analyzes the number of popular science TV programs on the air of national TV channels in Ukraine. Popular science content is an important aspect of public life, especially as an element of distraction from news about military operations. The goal of this study was to find out whether there are popular science programs on national TV channels, to analyze the competitive features of Ukrainian popular science TV journalism, which combines the broadcasting traditions formed by practice and modern development trends. We used the following methods: content analysis of the broadcast network of 32 national TV channels in Ukraine. We analyzed their TV programs for the presence of popular science content on the air, as well as the thematic content of TV programs in accordance with historical background, and additionally studied the websites and YouTube pages of channels that were disconnected from broadcasting in 2022; literature analysis, which made it possible to identify gaps in the research of functioning of popular science TV programs in the Ukrainian media space. Results: out of 32 TV channels, 5 broadcast popular science programs. Among them, 4 produce their own content, and one broadcasts foreign programs, mostly in the natural sciences. Of the 4 channels that have their own popular science content, 3 broadcast historical programs, and “Піксель TV” produces a project for children about chemical experi- ments. As of 2023, the share of channels with popular science content is 15.6%. Con- clusions: the level of popular science TV journalism in Ukraine is not increasing. At the same time, given the social preferences of the audience, we can say that this niche is not filled. Historical topics of video content are currently the most popular among viewers, which may be reflected in the growth in the number of viewers for channels that offer such content. KEYWORDS popular science journalism, television, societal demand, TV programs Copyright: © 2023 Roksolana Kravchenko. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. АНОТАЦІЯ У статті аналізується кількість науково-популярних телепрограм в ефірі національних телеканалів України. Науково-популярний контент є важливою стороною суспільного життя, особливо як елемент відволікання від новин про військові дії. Метою цього дослі- дження було з’ясувати, чи є на національних телеканалах науково-популярні програми, про- аналізувати конкурентоспроможність української науково-популярної тележурналістики, яка поєднує сформовані практикою традиції мовлення та сучасні тенденції розвитку. Вико- ристано такі методи: контент-аналіз мережі мовлення 32 національних телеканалів України. Ми проаналізували їхні телепрограми на предмет наявності в ефірі науково-популярного контенту, а також тематичного наповнення телепрограм відповідно до історичної довідки, а також додатково вивчили веб-сайти та YouTube-канали, які були відключені від ефірного мовлення у 2022 році; аналіз літератури, який дав змогу виявити прогалини у дослідженні функціонування науково-популярних телепрограм в українському медіапросторі. Резуль- тати: з 32 телеканалів 5 транслюють науково-популярні програми. Серед них 4 створюють власний контент, а один транслює іноземні програми, переважно природничі. З 4 каналів, які мають власний науково-популярний контент, 3 транслюють історичні програми, а «Пік- сель ТВ» випускає проєкт для дітей про хімічні досліди. Станом на 2023 рік частка каналів з науково-популярним контентом становить 15,6%. Висновки: рівень науково-популярної тележурналістики в Україні не підвищується. Водночас, враховуючи соціальні уподобання аудиторії, можна сказати, що ця ніша не заповнена. Історичні теми відеоконтенту зараз є найпопулярнішими серед глядачів, що може відобразитися на зростанні кількості глядачів каналів, які пропонують такий контент. КЛЮЧОВІ СЛОВА: науково-популярна журналістика, телебачення, соціальне замовлення, телепрограми. ABSTRACT © 2023 Current Issues of Mass Communication 71 ISSN 2312-5160 online ISSN 2786-4502 ISSN 2312-5160 online ISSN 2786-4502 © 2023 Актуальні питання масової комунікації Introduction Popular science journalism is a branch of journalism that aims to explain complex scientific concepts and discoveries to a wide audience. Its goal is to explain scientific phenomena or tech- nologies in an accessible, understandable form for the general public. Popular science journalism serves as a bridge between the complex world of scientific discovery and the society. At the same time, popular science journalism is not only about simplifying scientific content, but also about making it relevant and contextualized for everyday life. The terms "popular science journalism" and "science journalism" should not be equated, as science journalism is primarily focused on cov- ering scientific developments, research, and conclusions of the scientific community. It often tar- gets an audience that has an education in a particular scientific field or is looking for in-depth scientific information. Oksana Kyrylova, in her work "Science and Popular Science Journalism: Difficulties in Defining Concepts and Typologizing Media," draws attention to the frequent con- fusion of these definitions, so the author offers the following definitions: "science journalism pro- duces content with a focus on a trained audience without unnecessary use of infotainment methods to disseminate knowledge about events in the world of science. Popular science journalism is pri- marily concerned with popularizing scientific knowledge and science itself, using all available tools to attract the widest possible audience: infotainment, edutainment, science content, gaming content, etc." (Kyrylova, 2021, p.142). The key difference between these two concepts is the audi- ence and the depth of scientific coverage. Science journalism is aimed at an audience that is already familiar with or deeply interested in scientific details, while popular science journalism aims to attract a wider audience by simplifying and presenting scientific information in an easily digestible form. © 2023 Актуальні питання масової комунікації 72 ISSN 2312-5160 online ISSN 2786-4502 To simplify the distinction between the concepts of "popular science" and "science" journalism, we propose a visualized model of the transfer of scientific information from communicator to the audience (Figure 1). Figure 1. The difference between communicators and audience in science and popular science journalism. Source: based on the scientific publication "Scientific and Popular Science Journalism: Difficulties in Defining Concepts and Typologizing Media". Figure 1. The difference between communicators and audience in science and popular science journalism. Source: based on the scientific publication "Scientific and Popular Science Journalism: Difficulties in Defining Concepts and Typologizing Media". © 2023 Current Issues of Mass Communication Introduction In 2011, Halyna Usatenko drew attention to the broadcasting of foreign popular science programs in Russian, emphasizing that the issue of language in this context is a factor of the nation and the state as such. Educational TV channels do not use the Ukrainian lan- guage, and therefore do not contribute to the introduction of terminology that is being used in the scientific, educational, and mass use of the modern world, and therefore do not form a sense of Ukrainian presence in the development of the modern globalized world among citizens (Usatenko, 2011). One of the author's previous studies showed that even a year after Russia's full-scale invasion of Ukraine, a certain percentage of viewers (7.7%) continue to watch Russian popular science pro- grams (Flywheel of History, Notes of Cicero, Other History, Nauchpok, Topless, Utopia show, Geo, Loony, KOSMO, Nikolai Vavilov) (Kravchenko, 2023, p. 80). This prompted further re- search into the state of popular science content in Ukraine. Such trends require special attention from media professionals, because given the possible manipulation of historical facts in Russian content, the spread of Russian content to the Ukrainian audience may affect the Ukrainians’ per- ception of events such as the annexation of Crimea and the occupation of Donbas. These results also indicate a lack of scientific content in Ukraine, which is why the audience continues to turn to Russian content. On classic linear television, popular science content is developing less than, for example, en- tertainment content (Goian & Kravchenko, 2021, p. 132). However, the problem of popular science TV projects' ratings is part of a broader problem - the decline in ratings on television in general. According to the results of TV viewing monitoring over the past 5 years, the share of the audience has been falling every year. E.g., according to the TV panel, the highest share of the top 10 most rated TV channels was 14.63 (Ukraina TV channel), and the lowest was 2.46 (Plus Plus TV chan- nel). While in 2022, the highest share was 11.01 for 1+1 TV channel (Figure 2). The chart shows a consistent decline in the audience's interest in television in general. Since there are no annual public reports from a single monitoring group, we collected data from open sources: Television Industry Committee, Nielsen Ukraine, and Big Data UA. Introduction Researchers have demonstrated the interconnection between modern media and the construc- tion of national identity. Thus, TV content, with its audiovisual capabilities, has the potential to serve as a medium for representing the nation through its agents. Today, the global scientific com- munity is debating the relationship between popular science journalism and national conscious- ness. One example of popular science content’s impact on the self-identification of the population is the development of these TV programs in Israel. Merav Katz-Kimchi writes about this in detail in his article, noting that since 1968 - after the Six-Day War, from which Israel emerged victorious, having gained territories with a large Palestinian population - the state of Israel has deployed tele- vision not just to broadcast educational, entertainment, and information programs for its public, but as a propaganda tool aimed at the population of the occupied territories. And one of the ele- ments that successfully played the role of positive propaganda for Israel was popular science pro- grams. As the author notes, "thanks to the cooperation of the scientific and technological elite, the television broadcast of popular science programs in prime time made a huge contribution. With the help of these programs, the state spread a special image of the scientific and technological power of the nation". Science and technology began to play a prominent role in Zionism, which was established as a modern political movement in the late 19th century. The instrumental and practical Zionist movement, with its colonizing mission of settling Jews in Palestine and creating a state for the Jewish people, cherished science and technology as an important tool for exercising state power, examples of which include population management, army formation, mapping and engi- neering, and the provision of health care (Katz-Kimchi, 2012, p. 519). © 2023 Current Issues of Mass Communication 73 ISSN 2312-5160 online ISSN 2786-4502 ISSN 2312-5160 online ISSN 2786-4502 After the Soviet Union collapsed, Russian or foreign TV programs translated into Russian flooded the Ukrainian market, which could be one of the reasons for the stagnation of Ukrainian popular science content. In 2008, Mykhailo Kochergan warned that "television was occupied by Russian-language programs" and "the national information space was given to non-national media" (Kochergan, 2008, p. 19). © 2023 Актуальні питання масової комунікації Figure 2. Share of viewing of Ukrainian TV channels (2018-2022). Source: data from open sources: Television Industry Committee, Nielsen Ukraine, and Big Data UA. Figure 2. Share of viewing of Ukrainian TV channels (2018-2022). Source: data from open sources: Television Industry Committee, Nielsen Ukraine, and Big Data UA. The overall state of scientific programming in Ukraine was investigated by Maria Kuznetsova in her work "Thematic orientation and place in the modern Ukrainian television space of popular science programs." The author notes that as of 2016, out of 45 nationwide television channels, only 7 broadcast popular science programs series or films including the “UA: Перший” channel The overall state of scientific programming in Ukraine was investigated by Maria Kuznetsova in her work "Thematic orientation and place in the modern Ukrainian television space of popular science programs." The author notes that as of 2016, out of 45 nationwide television channels, only 7 broadcast popular science programs, series, or films, including the “UA: Перший” channel. Among the niche channels specializing in this content, she mentions “Мега”, “НЛО-TV”, “Тоніс”, and “ECO-TV”. She concludes that the fact that only niche channels broadcast popular science shows indicates that Ukrainian television does not give due attention to this content (Kuznetsova, 2016, p. 323). p g y 7 broadcast popular science programs, series, or films, including the “UA: Перший” channel. Among the niche channels specializing in this content, she mentions “Мега”, “НЛО-TV”, “Тоніс”, and “ECO-TV”. She concludes that the fact that only niche channels broadcast popular science shows indicates that Ukrainian television does not give due attention to this content (Kuznetsova, 2016, p. 323). During times of armed conflict, popular science journalism also plays a crucial role. Research by R. Kravchenko in 2022 showed that Ukrainians sometimes need to be distracted from news about the war to psychologically recover. Survey results from 460 respondents revealed that almost 65% used popular science or entertainment content to divert their attention from the war (Kravchenko, 2022, p. 138). Despite these statistics, there has been no significant leap in the de- velopment of popular science content in the Ukrainian media space. One of the reasons is the lack of interest among journalists themselves in delving into popular science journalism. Introduction The ratings may have an error depending on the quarter or high-profile events, but the chart below generally shows the situation with TV viewing. At the same time, a 2021 study by O. Goian and R. Kravchenko shows that the Ukrainian audience has a high demand for popular science TV programs. Almost three-quarters of the re- spondents (73%) said they would turn on Ukrainian TV channels more often if these channels had popular science and educational content on the air (Goian & Kravchenko, 2021, p. 133). Popular science journalism in Ukraine shows promising prospects for development, especially in social media and on the YouTube platform, where channels of this genre are growing and in- creasing their audiences. Within a year of full-scale Russian invasion, the viewership of popular science YouTube channels continued to grow, with some Ukrainian channels showing audience growth of +110% compared to 2022 (Kravchenko, 2023, p. 154). However, in contrast to YouTube, this niche remains largely undeveloped in the Ukrainian linear television landscape. Popular foreign broadcasters' programs are transmitted by the Мега channel. In the article "Is- sues of science popular programs of the Мега channel," the author, D.A. Shevchenko, points out that the majority of the channel's programming (90%) consists of foreign-produced content, with only 10% being domestic popular science television content. The popular science themes are pre- sented in four main categories: 45% of the content is historical-religious, 28% focuses on flora and fauna, 15% is related to interesting travels, adventures, and tourism, and only 12% are programs of a scientific and technical nature (Shevchenko & Kovtun, 2019, p. 352). 74 ISSN 2312-5160 online ISSN 2786-4502 Figure 2. Share of viewing of Ukrainian TV channels (2018-2022). Source: data from open sources: Television Industry Committee, Nielsen Ukraine, and Big Data UA. ISSN 2312-5160 online ISSN 2786-4502 online ISSN 2786-4502 © 2023 Current Issues of Mass Communication After the start of Russia's full-scale invasion of Ukraine, according to presidential decrees №151/2022 and №152/2022, as well as the National Council for Television and Radio Broadcast- ing's announcement, during the state of war, all nationwide TV channels were required to unite and broadcast the telethon “Unified News #UkraineTogether 24/7”. This telethon aimed to implement a unified information policy during the war. As a result, six nationwide TV channels (Suspilne (“UA:Перший”), “ICTV”, “Ми-Україна”, “1+1”, “Інтер”, “Рада”) joined forces and broadcasted the telethon, while other channels included segments of the telethon in their programming (Ukaz prezydenta Ukrainy №151/2022, Ukaz prezydenta Ukrainy №152/2022). In April 2022, it became 75 ISSN 2312-5160 online ISSN 2786-4502 ISSN 2312-5160 online ISSN 2786-4502 known that some oppositional TV channels were disconnected from digital broadcasting. The State Special Communications Service explained that the disconnection from digital broadcasting was due to decisions made by the National Security and Defense Council. However, the Secretary of the National Security and Defense Council, Oleksiy Danilov, stated that these channels allegedly refused to participate in the unified telethon, and the decision was made not due to censorship but for security reasons. The director-general of the 5th channel, Volodymyr Mzhelsky, stated that these broadcasters were asked to broadcast the telethon without participating in its production, effectively paying for the broadcasting of someone else's content (Tsyfrovyi efir “Espreso”, 2022). y p y g g ( y y p ) As of May 2023, the 5th channel, Прямий, and Еспресо are no longer available on digital television. However, their transmission continues on the internet, satellite, certain OTT platforms, and cable networks. Despite changes in the television schedules of nationwide TV channels, we have decided to examine the presence or absence of popular science content in their broadcasts. Projects were analyzed both directly on television and through the publication of content on web- sites, YouTube channels, and social media platforms of the television channels. One of the working hypotheses of this study was that popular science journalism is not suffi- ciently developed on Ukrainian linear television. This is due to several reasons, including the pri- ority given to war programs in the media and the lack of qualified personnel. The aim of this study is to investigate the state of popular science journalism in the Ukrainian television market. For further comprehensive research into the functioning of popular science jour- nalism in the Ukrainian television space, we sought to clarify the thematic orientation of popular science programs. In order to explore the potential and diversity of Ukrainian popular science con- tent, we set out to conduct a content analysis of 32 national TV channels, additionally, the publi- cation of television materials on the platforms of these TV channels was worked out. This was done because some TV channels were disconnected from broadcasting in 2022 and temporarily switched to digital platforms. Method As of 2023, there are 32 nationwide television channels in Ukraine: “UA: Перший”, “Рада”, “1+1”, “ICTV”, “СТБ”, “Інтер”, “УНІАН-TV”, “Бігуді”, “Армія TV”, “Новий Channel”, “TET”, “2+2”, “M1”, “НТН”, “Мега”, “ПлюсПлюс”, “Ми-Україна”, “1+1 Україна”, “ICTV2”, “ОЦЕ-TV”, “K1”, “K2”, “Zoom”, “Еспресо”, “XSPORT”, “Enter-фільм”, “Піксель”, “5 Chan- nel”, “TAK TV”, “Сонце”, “Суспільне Культура”, “Прямий”. Through content analysis, the programming grid of each channel was examined. As channels such as 5 Channel and Прямий were disconnected from digital broadcasting, their websites and YouTube channels were also an- alyzed among the 32 television channels. The presence or absence of popular science content in the broadcasts was studied, irrespective of the time slot. Additionally, it was determined whether the content was domestically produced or a retransmission of foreign programs. The thematic spec- ificity of the projects was described. Special attention was given to Suspilne (“UA: Перший”), as a media outlet obligated to broadcast such content. Article 4 of the Law of Ukraine "On Public Media of Ukraine" states that one of the main tasks of the National Public Broadcasting Company of Ukraine (Suspilne) is to contribute to the fullest satisfaction of the informational, cultural, and educational needs of the Ukrainian population, including the creation and dissemination of educa- tional and cognitive programs, as well as providing citizens of Ukraine with requested information products not available on the commercial market (Pro suspilni media Ukrainy : Zakon Ukrainy, 2021). Дослідження проводилося за період від 24 лютого 2022 року до 25 липня 2023 року. ) р р р у р у The dataset may be found by the link: Mendeley Data, https://doi.org/10.7910/DVN/ISNBTM The dataset may be found by the link: Mendeley Data, https://doi.org/10.7910/DVN/ISNBTM Results Through content analysis, it was found that among the 32 nationwide TV channels, popular science content is broadcasted on 5 channels: “Мега”, “Ми-Україна”, “Enter-фільм”, “5 Chan- nel”, and “Піксель”. Occasionally, interviews with scientists can also be seen on “Суспільне © 2023 Актуальні питання масової комунікації 76 ISSN 2312-5160 online ISSN 2786-4502 Культура”. Previously, popular science content was also present on Suspilne (“UA: Перший”), “1+1”, “ПлюсПлюс”, and “5 Channel”. Культура”. Previously, popular science content was also present on Suspilne (“UA: Перший”), “1+1”, “ПлюсПлюс”, and “5 Channel”. Below is a more detailed description of the qualitative and quantitative content related to pop- ular science on television channels. “UA: Перший” is the central channel of the National Public Broadcasting Company of Ukraine. In July 2020, the program "Шо? Як?" was launched on the channel. It was a popular science show for children and teenagers, where the robot Yuki and her assistant, geneticist, and gerontologist Oleksandr Kolyada, explained various phenomena from a scientific point of view. A total of 12 episodes were produced, but production was discontinued in August. “Рада” is a nationwide state information and analytical television channel that covers the ac- tivities of the Verkhovna Rada (Parliament) of Ukraine, its committees, commissions, and Ukrain- ian MPs. Its programming does not include the creation of popular science content. “Рада” is a nationwide state information and analytical television channel that covers the ac- tivities of the Verkhovna Rada (Parliament) of Ukraine, its committees, commissions, and Ukrain- ian MPs. Its programming does not include the creation of popular science content. “Рада” is a nationwide state information and analytical television channel that covers the ac- tivities of the Verkhovna Rada (Parliament) of Ukraine, its committees, commissions, and Ukrain- ian MPs. Its programming does not include the creation of popular science content. ian MPs. Its programming does not include the creation of popular science content. “1+1” is a nationwide channel that broadcasts programs of various genres. In December 2021, it produced a series of four popular science films called "Mysterious Manuscripts. Ukraine. The Return of Its History." “1+1” is a nationwide channel that broadcasts programs of various genres. In December 2021, it produced a series of four popular science films called "Mysterious Manuscripts. Ukraine. The Return of Its History." “1+1” is a nationwide channel that broadcasts programs of various genres. In December 2021, it produced a series of four popular science films called "Mysterious Manuscripts. Ukraine. Results It is part of "1+1 Media," which is owned by Ukrainian oligarch Ihor Kolomoisky. Despite this, it does not have any popular science programs. “Бігуді”. Part of "1+1 Media." The TV channel is oriented towards a female audience. Its pro- gramming includes Ukrainian and foreign TV series and cartoons. Scientific and educational pro- grams are not present on the channel. “Бігуді”. Part of "1+1 Media." The TV channel is oriented towards a female audience. Its pro- gramming includes Ukrainian and foreign TV series and cartoons. Scientific and educational pro- grams are not present on the channel. “Армія TV”. A nationwide state-owned TV channel focused on military themes, owned by the Ministry of Defence of Ukraine. The channel's specific content does not include scientific or edu- cational material. “Армія TV”. A nationwide state-owned TV channel focused on military themes, owned by the Ministry of Defence of Ukraine. The channel's specific content does not include scientific or edu- cational material. “Новий Channel”. A Ukrainian entertainment TV channel. It is part of the media holding "Starlight Media," owned by Ukrainian oligarch Viktor Pinchuk. Scientific and educational pro- grams are not broadcasted on this channel. g “TET”. A nationwide entertainment TV channel, part of the media conglomerate "1+1 Media." Scientific and educational content is not available on this channel. “TET”. A nationwide entertainment TV channel, part of the media conglomerate "1+1 Media." Scientific and educational content is not available on this channel. “2+2”. A nationwide TV channel belonging to the media conglomerate "1+1 Media," which is owned by oligarch Igor Kolomoisky. The channel is aimed at viewers aged 18 to 35. Scientific and educational programs are absent on this channel. “2+2”. A nationwide TV channel belonging to the media conglomerate "1+1 Media," which is owned by oligarch Igor Kolomoisky. The channel is aimed at viewers aged 18 to 35. Scientific and educational programs are absent on this channel. “2+2”. A nationwide TV channel belonging to the media conglomerate "1+1 Media," which is owned by oligarch Igor Kolomoisky. The channel is aimed at viewers aged 18 to 35. Scientific and educational programs are absent on this channel. “M1”. A nationwide music TV channel that broadcasts music videos and entertainment pro- grams. It is part of the media holdings "Starlight Media" and "TAVR Media." The channel's nature does not involve the creation of scientific or educational content. “НТН”. Results The Return of Its History." “Суспільне Культура” is a nationwide public television channel and online media outlet un- der the National Public Broadcasting Company of Ukraine, focused on cultural and educational topics. It creates the program "Культурний інстинкт" (Cultural Instinct), and occasionally fea- tures popularizers of history, such as Oleksandr Alfiorov and Volodymyr Viatrovych. “Суспільне Культура” is a nationwide public television channel and online media outlet un- der the National Public Broadcasting Company of Ukraine, focused on cultural and educational topics. It creates the program "Культурний інстинкт" (Cultural Instinct), and occasionally fea- tures popularizers of history, such as Oleksandr Alfiorov and Volodymyr Viatrovych. “ICTV” is a nationwide television channel belonging to the media holding company "Starlight Media." It does not have any popular science programs. “ICTV” is a nationwide television channel belonging to the media holding company "Starlight Media." It does not have any popular science programs. “ICTV” is a nationwide television channel belonging to the media holding company "Starlight Media." It does not have any popular science programs. “СТБ”, like ICTV, is a nationwide television channel owned by the media holding company "Starlight Media." It also lacks popular science programs. “СТБ”, like ICTV, is a nationwide television channel owned by the media holding company "Starlight Media." It also lacks popular science programs. “СТБ”, like ICTV, is a nationwide television channel owned by the media holding company "Starlight Media." It also lacks popular science programs. “Інтер” is an information and entertainment television channel in Ukraine, owned by oligarch Dmytro Firtash and politician Serhii Lyovochkin. It does not broadcast any popular science pro- grams. “Інтер” is an information and entertainment television channel in Ukraine, owned by oligarch Dmytro Firtash and politician Serhii Lyovochkin. It does not broadcast any popular science pro- grams. “Інтер” is an information and entertainment television channel in Ukraine, owned by oligarch Dmytro Firtash and politician Serhii Lyovochkin. It does not broadcast any popular science pro- grams. “УНІАН-TV” is positioned as a Ukrainian enlightening-informational television channel. It is part of "1+1 Media," which is owned by Ukrainian oligarch Ihor Kolomoisky. Despite this, it does not have any popular science programs. “УНІАН-TV” is positioned as a Ukrainian enlightening-informational television channel. It is part of "1+1 Media," which is owned by Ukrainian oligarch Ihor Kolomoisky. Despite this, it does not have any popular science programs. “УНІАН-TV” is positioned as a Ukrainian enlightening-informational television channel. Results The channel used to broadcast the following popular science projects: ● "The World Awaits Discovery" - an animated series about various scientific dis- coveries around the world, concluded in 2019. ● "ECO ПлюсПлюс" - short educational stories about the harm we cause to the environment every day, concluded in 2020. “Ми — Україна”. This is a nationwide information channel composed of former employees of the closed channels "Ukraine" and "Ukraine 24," previously owned by Ukrainian oligarch Rinat Akhmetov. The channel produces its own popular science program: "Paragraph" - an author's pro- ject by Ukrainian TV presenter Yulia Halushka. It reveals the historical truth behind the eternal contradictions between Ukraine and Russia. “Ми — Україна”. This is a nationwide information channel composed of former employees of the closed channels "Ukraine" and "Ukraine 24," previously owned by Ukrainian oligarch Rinat Akhmetov. The channel produces its own popular science program: "Paragraph" - an author's pro- ject by Ukrainian TV presenter Yulia Halushka. It reveals the historical truth behind the eternal contradictions between Ukraine and Russia. “1+1 Україна”. A nationwide television channel. It is part of the media conglomerate "1+1 Media." Scientific-popular content is absent on the channel. “ICTV2”. An entertainment channel belonging to the media conglomerate "Starlight Media." Scientific-popular content is absent on the channel. “ОЦЕ TV”. A nationwide entertainment channel that is part of the media conglomerate "Star- light Media." It was created as a rebranding of the "QTV" channel in 2017. Scientific-popular content is absent on the channel. “ОЦЕ TV”. A nationwide entertainment channel that is part of the media conglomerate "Star- light Media." It was created as a rebranding of the "QTV" channel in 2017. Scientific-popular content is absent on the channel. “K1”. A nationwide entertainment channel targeted towards a young audience. It belongs to media conglomerate "Inter Media Group." Scientific-popular content is absent on the channel “K2”. A Ukrainian women's channel, targeting women aged 18-50. It belongs to the media conglomerate "Inter Media Group." Scientific-popular content is absent on the channel. “K2”. A Ukrainian women's channel, targeting women aged 18-50. It belongs to the media conglomerate "Inter Media Group." Scientific-popular content is absent on the channel. “Zoom”. A Ukrainian entertainment channel of the media conglomerate "Inter Media Group." Scientific-popular content is absent on the channel. “Прямий”. A Ukrainian informational channel created on August 24, 2017, in place of the "Tonis" channel, operating in the infotainment format. Results A Ukrainian nationwide TV channel belonging to the media conglomerate Inter Medi up. Scientific and educational content is not present on this channel. “Мега”. This is a Ukrainian popular science channel under the "Inter Media Group" holding rs informative documentary series, original projects, fiction films, and historical TV show “Мега”. This is a Ukrainian popular science channel under the "Inter Media Group" holding. It airs informative documentary series, original projects, fiction films, and historical TV shows “Мега”. This is a Ukrainian popular science channel under the "Inter Media Group" holding. It airs informative documentary series, original projects, fiction films, and historical TV shows © 2023 Current Issues of Mass Communication 77 ISSN 2312-5160 online ISSN 2786-4502 from international networks like BBC, National Geographic Channel, History, Discovery C m international networks like BBC, National Geographic Channel, History, Discovery Channel from international networks like BBC, National Geographic Channel, History, Discovery Channel, Animal Planet, and Cineflix. Some of the popular science programs on this channel include: g p y y mal Planet, and Cineflix. Some of the popular science programs on this channel include: Animal Planet, and Cineflix. Some of the popular science programs on this channel incl ● "Unexpected History" - a historical documentary project produced jointly by the UK and the USA, created in 2013. ● "Unexpected History" - a historical documentary project produced jointly by the UK and the USA, created in 2013. ● "How the Universe Works" - a scientific documentary series that was originally broadcast on the Discovery Channel in 2010. ● "How the Universe Works" - a scientific documentary series that was originally broadcast on the Discovery Channel in 2010. ● "Ukraine: Forgotten History" - a Ukrainian documentary series about the history of Ukraine, presented by Andriy Romanidi, with scriptwriting by Iryna Chmola. Each episode delves into significant events in Ukrainian history using reconstructions and com- mentary from Ukrainian political scientists, historians, cultural experts, and journalists. Production took place in 2012-2013. ● "Secrets of the Deep with Jeremy Wade" - an American project exploring seas and oceans, filmed in 2020-2022. ● "Secrets of the Deep with Jeremy Wade" - an American project exploring seas and oceans, filmed in 2020-2022. “ПлюсПлюс”. This is a nationwide children's channel owned by the "1+1 Media" media con- glomerate. © 2023 Актуальні питання масової комунікації Results Scientific-popular content is absent on the channel. “Еспресо TV”. A nationwide informational channel established in November 2013. Scientific- popular content is absent on the channel. “XSPORT”. A nationwide sports and educational channel. The channel's nature does not in- volve creating scientific-popular content. “XSPORT”. A nationwide sports and educational channel. The channel's nature does not in- volve creating scientific-popular content. “Enter-фільм”. A nationwide film channel. It is part of the media conglomerate "Inter Media Group." The channel airs the series "Ukraine: The Forgotten History." It is a Ukrainian documen- tary series about the history of Ukraine, hosted by Andriy Romanidi. The series premiered in 2012 and was filmed until 2013. “Enter-фільм”. A nationwide film channel. It is part of the media conglomerate "Inter Media Group." The channel airs the series "Ukraine: The Forgotten History." It is a Ukrainian documen- tary series about the history of Ukraine, hosted by Andriy Romanidi. The series premiered in 2012 and was filmed until 2013. “Піксель TV”. A Ukrainian children's television channel. It belongs to the Inter Media Group. The channel broadcasts the Ukrainian-produced program "I Want to Know Everything" for chil- dren. “Піксель TV”. A Ukrainian children's television channel. It belongs to the Inter Media Group. The channel broadcasts the Ukrainian-produced program "I Want to Know Everything" for chil- dren. © 2023 Актуальні питання масової комунікації 78 ISSN 2312-5160 online ISSN 2786-4502 “5 Channel”. A Ukrainian informational and educational channel. From 2003 to 2021, it was owned by the 5th President of Ukraine, Petro Poroshenko. Currently, the channel airs the scientific- popular project "Time Machine." It is a historical program that talks about significant figures that influenced history, featuring unique archival documents shedding light on modern history. “5 Channel”. A Ukrainian informational and educational channel. From 2003 to 2021, it was owned by the 5th President of Ukraine, Petro Poroshenko. Currently, the channel airs the scientific- popular project "Time Machine." It is a historical program that talks about significant figures that influenced history, featuring unique archival documents shedding light on modern history. y g q g g y “TAK TV”. A Ukrainian informational channel. Scientific-popular programs are absent on the channel. “TAK TV”. A Ukrainian informational channel. Scientific-popular programs are absent on the channel. “Сонце”. A nationwide entertainment channel. Scientific-popular programs are absent on the channel. “Сонце”. A nationwide entertainment channel. Scientific-popular programs are absent on the channel. Figure 3. Results Availability of popular science TV programs on Ukrainian channels. Source: own research Figure 3. Availability of popular science TV programs on Ukrainian channels. Source: own research Figure 3. Availability of popular science TV programs on Ukrainian channels. Source: own research Thus, we found that currently on Ukrainian television, science-popular programs are only broadcasted on 5 channels (15.6%). Among them, 4 channels produce their own content (“Ми- Україна”, “Enter фільм”, “Піксель TV”, “5 Channel”), out of which 3 channels air historical programs, and “Піксель TV” creates a project for children about chemical experiments. The “Мега” channel broadcasts foreign-produced programs mainly related to natural sciences. The re- maining 27 channels produce entertainment or news content, while specialized channels like “XSPORT” or “M1” air sports and music programs, respectively (Figure 3). On the “Ми-Україна” channel, the "Paragraph" project is broadcasted on the channel's YouTube page, since the television airtime is filled with the informational marathon "Yedуni No- vyny". The “5 Channel” airs the program "Time Machine" every Sunday at 8:00 PM (prime time) and on their YouTube channel (a separate version was not created for the YouTube channel, but the entire television episode was published). The “Мега” channel is not present in the marathon, and throughout the day, the channel broad- casts scientific-popular projects, mostly of foreign production. However, their YouTube channel has been inactive for the past year. p y “Enter-фільм”, like “Мега”, is not part of the United Marathon. It broadcasts the historical series "Ukraine: The Forgotten History" every day at 8:45 AM. © 2023 Current Issues of Mass Communication 79 ISSN 2312-5160 online ISSN 2786-4502 “Піксель TV” also does not participate in the television marathon. It broadcasts the program for children of Ukrainian production "I Want to Know Everything" every day at 1:50 AM and 10:30 AM. Conclusions The research revealed that only 5 out of 32 television channels broadcast scientific-popular programs, which constitutes 15.6% of the total number of channels. This percentage matches the results obtained by Maria Kuznetsova in 2016, indicating that the increase in scientific-popular content on linear television is not happening. Additionally, 3 out of 5 (60%) of these channels air exclusively historical programs, while the Мега channel also has a significant portion of historical programs, but it also broadcasts programs related to natural and technical sciences. This imbalance with a focus on historical themes can be explained by the increased interest among Ukrainians in the context of certain events after Russia's invasion of Ukraine in 2014 and the full-scale war in 2022. The analysis of the broadcast schedule of the channels showed that only two channels broadcast science-popular programs during prime time. Such a low percentage of scientific-popular content presence on television and its almost complete absence during prime time indicates that Ukrainian television does not give due attention to such projects. It was also found that only one channel, “Мега”, broadcasts foreign-produced content, while the other four channels produce their own content. This trend shows that journalists are trying to develop scientific-popular programs, but for now, they are mainly focused on historical topics. The analysis of literature has shown that popular science television programs around the world have been used not only as a tool for educating the population, but also as a significant political influence tool to emphasize the prestige of the country. Besides, it helped to shape national identity. A strong case of this is Israel, which, after the Six Day War, deployed television in the occupied lands inhabited by Palestinians. One of the elements that served as positive propaganda for Israel was the popular science programs that were broadcast in prime time. Literature data also showed that a certain percentage of viewers (7.7%) continue to consume Russian popular science content, including historical programs, which requires special attention from media professionals. Data Availability Dataset file is available from: the Mendeley Data https://doi.org/10.7910/DVN/ISNBTM y Dataset file is available from: the Mendeley Data https://doi.org/10.7910/DVN/ISNBTM Acknowledgments g I would like to express my appreciation to Oles Goian for all his help with preparing this article at a time when the subject is of such considerable relevance and importance. Data Availability Data Availability Dataset file is available from: the Mendeley Data https://doi.org/10.7910/DVN/ISNBTM References Derzhspetszviazku: Tsyfrovyi efir «Espreso», Priamoho ta 5 kanalu vidkliuchyly na vykonannia rishennia RNBO [State Special Communication Service: Digital broadcasting of Espresso, Pryamyy and Channel 5 was cut off to implement the NSDC decision]. (2022). Retrieved from https://detector.media/infospace/article/199070/2022-05-09-derzhspetszvyazku- tsyfrovyy-efir-espreso-pryamogo-ta-5-kanalu-vidklyuchyly-na-vykonannya-rishennya- rnbo/ «My – Ukraina»: kolyshni mediinyky Akhmetova zareiestruvaly svoiu kompaniiu. detector.media. Detector media. ["We are Ukraine": Akhmetov's former media workers register their own company] (2022). Retrieved from https://detector.media/rinok/article/203254/2022-09-29- my-ukraina-kolyshni-mediynyky-akhmetova-zareiestruvaly-svoyu-kompaniyu/ «My – Ukraina»: kolyshni mediinyky Akhmetova zareiestruvaly svoiu kompaniiu. detector.media. Detector media. ["We are Ukraine": Akhmetov's former media workers register their own company] (2022). Retrieved from https://detector.media/rinok/article/203254/2022-09-29- my-ukraina-kolyshni-mediynyky-akhmetova-zareiestruvaly-svoyu-kompaniyu/ © 2023 Актуальні питання масової комунікації 80 ISSN 2312-5160 online ISSN 2786-4502 Funk, C. (2020, August 27). How Americans get science news and information. Pew Research Center’s Journalism Project. https://www.pewresearch.org/journalism/2017/09/20/science- news-and-information-today/ y Goian O. & Kravchenko R. (2021). Naukovo-populiarne telebachennia yak sotsialne zamovlennia ta osvitnia model: dosvid Ukrainy. [Popular science television as a social order and educa- tional model: the experience of Ukraine] Knowledge, Education, Law, Management, 4 (40), 130–137. https://doi.org/10.51647/ kelm.2021.4.21. Katz-Kimchi, M. (2012). Screening science, producing the nation: popular science programs on Israeli television (1968–88). Media, Culture & Society, 34(5), 519–536. https://doi.org/10.1177/0163443712442699 Kocherhan M. (2008). Movna sytuatsiia i movna polityka v Ukraini [Language situation and lan- guage policy in Ukraine]. Svitohliad. 2. 19 https://www.mao.kiev.ua/biblio/jscans/svitogliad/svit-2008-10-2/svit-2008-10-2-18- Kocherhan M. (2008). Movna sytuatsiia i movna polityka v Ukraini [Language situation and lan- guage policy in Ukraine]. Svitohliad. 2. 19 https://www.mao.kiev.ua/biblio/jscans/svitogliad/svit-2008-10-2/svit-2008-10-2-18- cochergan.pdf K h k R (2023) T d t ii tk k li kh Y T b k li Uk i i Kravchenko, R. (2023). Tendentsiia rozvytku naukovo-populiarnykh YouTube-kanaliv v Ukraini: porivniannia audytorii 2022-2023 rr [Trend of popular science YouTube channels in Ukraine: comparison of audiences in 2022-2023]. Science and technology: problems, prospects and innovations. 153-157 https://sci-conf.com.ua/x-mizhnarodna-naukovo- praktichna-konferentsiya-science-and-technology-problems-prospects-and-innovations-6- 8-07-2023-osaka-yaponiya-arhiv/ Kravchenko R. (2023). Portret audytorii naukovo-populiarnykh prohram v Ukraini [Portrait of the audience of popular science programs in Ukraine]. Rozvytok naukovoi dumky: aktualni pytannia, dosiahnennia ta innovatsii. 77-81 https://molodyivchenyi.ua/omp/index.php/con- ference/catalog/download/33/670/1377-1?inline=1 Kravchenko, R. (2022). Ukrainian scientific TV programs and YouTube channels as a distraction from war news on Russia’s invasion of Ukraine: a survey-based observational study and a content analysis. Science Editing, 9(2), 136-141. https://doi.org/10.6087/kcse.278 Kuznetsova M. M. (2016) Tematychne spriamuvannia ta mistse v suchasnomu ukrainskomu teleprostori naukovopopuliarnykh prohram [Thematic direction and place in the modern Ukrainian TV space of popular science programs]. Zbirnyk prats Naukovo-doslidnoho instytutu presoznavstva. 6. 321-332. http://nbuv.gov.ua/UJRN/ZPNDZP_2016_6_25 Kyrylova, O. (2021). © 2023 Актуальні питання масової комунікації References Naukova ta naukovo-populiarna zhurnalistyka: skladnoshchi definitsii poniat i typolohizatsii media. [Scientific and popular science journalism: the difficulties of defin- ing concepts and typologizing media] Synopsys: tekst, kontekst, media. 27(3), 141-148. https://doi.org/10.28925/2311-259x.2021.3.3 Verhovna Rada. (2021). Pro suspilni media Ukrainy : Zakon Ukrainy stanom na 23 lyp. 2021 r. [The Law of Ukraine On Public Media of Ukraine]. Retrieved from https://zakon.rada.gov.ua/laws/show/1227-18#Text Shevchenko, D. A., & Kovtun, N. O. (2017). Problematyka naukovo-populiarnykh prohram telekanalu «Meha» [Issues of popular science programs on Mega TV channel]. Aktualni Problemy Pryrodnychykh I Humanitarnykh Nauk U Doslidzhenniakh Molodykh Uchenykh «Rodzynka–2019». 352-354 https://eprints.cdu.edu.ua/3384/1/rodzinka_2_2019-352- 354.pdf Most popular contemporary TV shows in the United Kingdom (UK) 2022. Statista. (2022). Retrieved from https://www.statista.com/statistics/1025501/most-popular-contemporary- tv-shows-in-the-uk/ ident of Ukraine. (2022) Ukaz prezydenta Ukrainy №151/2022. [Presidential Decre №151/2022]. Retrieved from https://www.president.gov.ua/documents/1512022-41757 ( ) p y y [ №151/2022]. Retrieved from https://www.president.gov.ua/documents/1512022-41757 President of Ukraine. (2022) Ukaz prezydenta Ukrainy №152/2022. [Presidential Decree №152/2022]. Retrieved from https://www.president.gov.ua/documents/1522022-41761 №5 / 0 ]. e eved o ps://www.p es de .gov.ua/docu e s/ 5 0 757 President of Ukraine. (2022) Ukaz prezydenta Ukrainy №152/2022. [Presidential Decree №152/2022]. Retrieved from https://www.president.gov.ua/documents/1522022-41761 © 2023 Current Issues of Mass Communication 81 ISSN 2312-5160 online ISSN 2786-4502 ISSN 2312-5160 online ISSN 2786-4502 Usatenko H. (2011). Ukrainska/rosiiska mova televiziinykh shou. [Ukrainian/Russian language of TV shows]. Radio Liberty. Retrieved from https://www.radiosvoboda.org/a/2268195.html Wikimedia (2023). Spysok ukrainskykh telekanaliv. [List of TV channels in Ukraine] Retrieved from https://uk.wikipedia.org/wiki/%D0%A1%D0%BF%D0%B8%D1%81%D0%BE%D0%B A_%D1%83%D0%BA%D1%80%D0%B0%D1%97%D0%BD%D1%81%D1%8C%D0 %BA%D0%B8%D1%85_%D1%82%D0%B5%D0%BB%D0%B5%D0%BA%D0%B0% D0%BD%D0%B0%D0%BB%D1%96%D0%B2#cite_note-4 82
https://openalex.org/W2042555148	https://www.epj-conferences.org/articles/epjconf/pdf/2013/13/epjconf_isvh2012_04008.pdf	English	null	A Quark Matter Contribution to the Cosmic Ray Spectrum	EPJ web of conferences	2,013	cc-by	3,840	A Quark Matter Contribution to the Cosmic Ray Spectrum Kyle Lawson1,a 1Department of Physics and Astronomy, University of British Columbia, Vancouver, BC, V6T 1Z1, Canad 1Department of Physics and Astronomy, University of British Columbia, Vancouver, BC, V6T 1Z1, Canada Abstract. I will describe a possible dark matter model in which the dark matter is composed of heavy “nuggets" of standard model quarks and antiquarks bound in a high density phase of . If objects of this type are formed early in the universe’s history they may provide the observed dark matter content. In this scenario the nuggets are dark not because of their fundamentally weak interactions but because of the incredibly small number density required to explain the observed mass density of the dark matter. The correspondingly small ﬂux of these objects through the earth renders them invisible to conventional high sensitivity dark matter searches intended to detect weakly interacting particles with a ﬂux many orders of magnitude larger. Instead the greatest search potential for dark matter models of this form may come from the largest scale cosmic ray detectors. I will brieﬂy describe the properties of quark nugget dark matter and then use these properties in order to predict the signal they would produce in a variety of cosmic ray detectors. This is an Open Access article distributed under the terms of the Creative Commons Attribution License 2 0 , which . permits unrestricted use, distributi and reproduction in any medium, provided the original work is properly cited. on, EPJ Web of Conferences DOI: 10.1051/ C ⃝Owned by the authors, published by EDP Sciences, 2013 , epjconf 201 / 04008 (2013) 35204008 52 EPJ Web of Conferences DOI: 10.1051/ C ⃝Owned by the authors, published by EDP Sciences, 2013 , epjconf 201 / 04008 (2013) 35204008 52 EPJ Web of Conferences DOI: 10.1051/ C ⃝Owned by the authors published by EDP S , epjconf 201 / 04008 (2013) 35204008 52 DOI: 10.1051/ C ⃝Owned by the authors, published by EDP Sciences, 2013 , epjconf 201 / ( ) 35204008 aklawson@phas.ubc.ca 1 Quark matter as dark matter ter the phase transition the hadronic matter not bound in nuggets annihilates with the remaining antimatter. How- ever the remaining plasma has an excess of quarks corre- sponding to the excess of antiquarks locked in the nuggets. The remaining matter makes up the visible component of the universe. This mechanism can explain the observed visible matter to dark matter ratio if the matter component of the universe is composed of dark antimatter nuggets, dark matter nuggets and visible matter in a 3:2:1 ratio. More details on this model may be found in [2], [3], [4] and the references therein. Conventional dark matter models generally assume the ex- istence of a new fundamental particle which interacts only weakly with the particles of the standard model and which carries a conserved charge to prevent its decay over cos- mological timescales. The properties of this new particle are then to be extracted from the observed properties of the dark matter. As an alternative to invoking new funda- mental physics there exists a class of dark matter models in which the dark matter is a composite object consisting of standard model particles. In the dark matter model con- sidered here the dark matter takes the form of nuggets of standard model light quarks locked in a colour supercon- ducting phase (or whatever phase is actually realized in the high density limit of .) It has been speculated that such objects could be formed at phase transition of the early universe at which time the essentially “freeze out" of interactions with the surrounding plasma. This model is similar to earlier objects such as “stranglets" in which the high density phase is stabilized by the presence of strange quarks [1]. Article available at http://www.epj-conferences.org or http://dx.doi.org/10.1051/epjconf/20125204008 This is an Open Access article distributed under the terms of the Creative Commons Attribution License 2 0 , which . permits unrestricted use, distributi and reproduction in any medium, provided the original work is properly cited. on, Article available at http://www.epj-conferences.org or http://dx.doi.org/10.1051/epjconf/20125204008 4 Estimated ﬂux and energy For present purposes the most important consequence of this model is the presence of massive nuggets of nuclear density antiquarks in the solar neighbourhood. Given a lo- cal dark matter density of roughly 1GeV cm−3 and a galac- tic velocity of 200 km/s the ﬂux of quark nuggets is, dN dA dt = nv ≈ 1025 B ! km−2yr−1. (1) (1) Where the nuggets’ average baryon charge, B, should be taken to be around 1025. While this ﬂux is far below the sensitivity level of conventional dark matter searches it is accessible to large scale cosmic ray detectors being com- parable to the cosmic ray ﬂux near the limit. The annihilation of galactic protons is a more com- plicated many body problem than the case of electrons. A proton colliding with the nugget will penetrate into the nugget and annihilate to jets of light mesons. This process results in two distinct annihilation signals. If one of the resulting jets is directed back towards the nugget surface the energy will rapidly heat a region of the positron layer which will subsequently emit in the xray band [9]. Jets unable to reach the surface dissipate their energy in the nugget increasing its temperature and resulting in thermal emission. The scale of the thermal spectrum is determined by the surrounding matter density and for the galactic cen- tre will be in the microwave band [10]. In both of these bands an excess emission above expected galactic back- grounds are detected. The telescope detects an excess of emission from the galactic centre which it char- acterizes as consistent with an 8keV plasma though such a A nugget of quarks will deposit thermal energy in the atmosphere as it collides with atmospheric molecules, however, this process will produce only modest levels of ionization and will not generate any relativistic secondary particles. As such the prospects for detecting these events in a cosmic ray experiment are very limited. If however the nugget is composed antiquarks the resulting annihilation of atmospheric molecules will release considerably more energy, much of it in the form high energy secondary par- ticles. In order to estimate energy scale of these events we consider the case in which all of the atmospheric material in the path of the nugget annihilates completely. 2 Nugget structure At asymptotically large densities the ground state of  is a colour superconducting quark matter state with equal numbers of u,d and s quarks. This state is electrically and composed purely of the three light quark ﬂavours. How- ever, at the surface of a nugget composed of quark mat- ter the density must drop to zero from its bulk value. As the density drops so does the ability of the quark mat- ter to support a strange quark population of the heavier strange quarks. As such the outer shell of the quark matter contains only u and d quarks and carries a strong electri- cal ﬁeld due to the charge diﬀerence between these two species. At the nugget surface the energy density of these electric ﬁelds will be near the energy scale. In or- der to maintain overall electrical neutrality the nugget ac- quires a surrounding shell of electromagnetically bound leptons known as the electrosphere (though in the case of an antiquark nugget the electrosphere is in fact composed of positrons.) The electrosphere was modeled in detail in [5] and runs from nuclear scale densities near the surface The particular model considered here diﬀers from ear- lier composite dark matter candidates in that these objects may be composed of either matter or antimatter and was originally proposes as an explanation of the matter dom- inance of the universe. It is assumed that the very early universe contained equal amounts of matter and antimat- ter and that a baryogenesis process must occur resulting in the present day matter dominance. In this scenario the dark matter forms through a charge separation mechanism in which the formation of nuggets of antimatter proceeds slightly faster than the rate of matter nugget formation. Af- EPJ Web of Conferences EPJ Web of Conferences down to atomic densities far from the quark matter where the electrons see only a strongly screened central charge. down to atomic densities far from the quark matter where the electrons see only a strongly screened central charge. While the actual nugget structure may be rather com- plicated depending on the speciﬁc phase realized many of these details are incidental to the general eﬀects that I will consider below. Of primary importance is the pres- ence of a higher than nuclear density quark matter core in a non-baryonic phase and a surrounding electrosphere of electrons or positrons. 3 Observational constraints While, as argued above, the nuggets are eﬀectively “dark" their interaction with visible matter that they do encounter is by not weak in any sense. It is only the rarity of these in- teractions makes them diﬃcult to observe. At present the best constraints on the model are derived from astrophys- ical observations. Of particular interest are observations of regions where both the visible and dark matter densi- ties are substantially above average. Here I brieﬂy outline some of the relevant observations. While none of these sources independently provide a smoking gun the absence of any one of these sources, in roughly the observed ratio, would have immediately ruled out this dark matter model. Based on the observed scale of emission and assuming a typical dark matter proﬁle for the galactic centre we may estimate the average quark nugget mass required to produce a diﬀuse continuum at this scale (smaller nuggets have a larger number density and thus produce a stronger diﬀuse emission.) A rough ﬁt to the galactic backgrounds would suggest that the nuggets have an average baryonic charge of B ∼1025 or, equivalently, a mass of roughly a gram. A stronger than expected 511keV line, associated with low momentum electron positron annihilation, is observed from the galactic centre [6]. This indicates the presence of signiﬁcantly more antimatter in the galactic centre than is predicted by standard astrophysical models and may be explained by the annihilation of galactic electrons in the outer, low temperature, positron layers of a population of quark nuggets [7]. The satellite observes an excess of diﬀuse gamma rays from the galactic centre at the MeV energy scale. In the quark nugget model this emission is neces- sarily co-produced with the 511 keV line [8]. While the majority of galactic electrons incident on a nugget will an- nihilate almost immediately on contact with the outer lay- ers of the electrosphere a small fraction will be capable of penetrating suﬃciently deep into the nugget that they an- nihilate with positrons carrying momenta near the binding energy scale of the quark matter. Detailed calculations of the structure of the positron layer and annihilation rates produce a 511keV to MeV emission ratio consistent with that observed [5]. 2 Nugget structure plasma is too hot to remain bound to the galaxy and has no obvious heating mechanism [11]. At the microwave scale the satellite observes a “haze" associated with the galaxy. The scale of both of these diﬀuse sources is con- sistent with the predictions of the quark nugget dark matter model. While the actual nugget structure may be rather com- plicated depending on the speciﬁc phase realized many of these details are incidental to the general eﬀects that I will consider below. Of primary importance is the pres- ence of a higher than nuclear density quark matter core in a non-baryonic phase and a surrounding electrosphere of electrons or positrons. The observations outlined above all arise from the in- teractions of visible and dark matter in the relatively high density region near the galactic centre. Recent work has also analyzed the contribution of these objects the cosmo- logical background [12]. The density of the universe at very early phases of it’s evolution was comparable to that of the galactic centre today and, as such, we would ex- pect a similar diﬀuse radiation contribution to that from the galactic centre but redshifted down in energy by roughly three orders of magnitude. In fact, sky temperature mea- surements by the 2 6 Surface particle counts In a conventional air shower the production of secondary particles continues only as long as the shower contains particles of suﬃcient energy to break up atmospheric nu- clei. Beyond the point where ionization becomes the dom- inate interaction secondary particles cease to be created and only those particles with suﬃcient momenta will reach the earth’s surface where they may be detected. Con- versely, a quark nugget traveling through the atmosphere will continue to emit secondary particles across all atmo- spheric depths. These secondary particles are preferen- tially “beamed" in the forward direction of the nugget’s path as the majority of annihilations will occur on the nugget’s forward surface and only those annihilation com- ponents directed back towards the nugget surface are able to escape from the quark matter. The injection energy of these secondary particles depends only on nuclear physics processes within the nugget and consequently will be es- sentially constant across all atmospheric depths. Rather than producing the typical “pancake" of comoving sec- ondary particles observed in a conventional shower the nugget will carry with it a cone of secondary particles emitted along its line of motion. Following the same logic which lead to equation (2) the total number of nucleons striking the nugget will be, Etot = XatπR2 n mN ≈1016eV Rn 10−5cm 2 . (3) (3) Where mN is the average mass per nucleon. This number however represents only an upper limit on the number of annihilations as thermal and radiation eﬀects will likely cause some fraction of the incoming matter to be deﬂected before reaching the nugget’s surface [13]. In order to estimate the number of secondary particles which these nuclear annihilations will produce consider the fate of an atmospheric nucleus incident on the surface of an antiquark nugget. Rather than annihilating directly on contact the nucleus will penetrate some distance into the nugget due to the relatively small wave function over- lap between the nucleons and the cooper pair states of the quark matter. When the nucleon annihilates the annihila- tion products will stream oﬀas hadronic jets along with the other light modes of the superconductor. As the nu- cleus’ initial momentum is negligible in comparison to the total energy released the resulting jets will be randomly oriented relative to the nugget’s surface. 4 Estimated ﬂux and energy The at- mospheric depth is on the order of 1kg cm−2 and a quark nugget with a baryon number on the order of 1025 will 04008-p.2 ISVHECRI 2012 have a radius of 10µm so that the energy generated by an- nihilations may be estimated as by thermal photons.) The partial deﬂection of incoming at- mospheric matter combined with the relatively small num- ber of annihilation products may push the number of emit- ted secondary particles below the annihilation rate (3) by up to a few orders of magnitude however, even with these suppression eﬀects, the total number of secondaries will remain at least comparable to the number generated in a ultra high energy cosmic ray induced air shower. Shower evolution is then dependent on the thermal properties of the nugget and the surrounding atmospheric density. The evolution of these parameters with atmospheric depth has been discussed in [13] where it is argued that the shower evolution may in fact take a more complex form then sim- ply tracking the rise in atmospheric density. Etot = 2Xatc2πR2 n ≈1026eV Rn 10−5cm 2 . (2) (2) While much of this energy will be thermalized within the nugget some will be released into the atmosphere and may be observable. Some discussion of the nuggets’ thermo- dynamics and interactions with the atmosphere is required before this energy can be translated observational conse- quences. However the scale of the total energy produced as an antiquark nugget crosses the atmosphere is large enough to suggest that these events may be observable to detectors of suﬃcient area. While much of this energy will be thermalized within the nugget some will be released into the atmosphere and may be observable. Some discussion of the nuggets’ thermo- dynamics and interactions with the atmosphere is required before this energy can be translated observational conse- quences. However the scale of the total energy produced as an antiquark nugget crosses the atmosphere is large enough to suggest that these events may be observable to detectors of suﬃcient area. The most important distinguishing feature of the air shower associated with a quark nugget passing through the atmosphere is the distinctive time scale involved. While the emitted secondary particles travel at essentially the speed of light the nugget itself moves at a much slower galactic speed. This results in an air shower which may evolve over tens of milliseconds. 5 Atmospheric ﬂuorescence The production of atmospheric ﬂuorescence requires a large number of relativistic secondary particles. As these particles lose energy to the surrounding atmosphere they excite nitrogen molecules which subsequently emit in the UV band. The ﬂuorescence yield of a quark nugget in- duced air shower may thus be estimated from the number of high energy secondary particles which the nugget in- jects into the atmosphere. 6 Surface particle counts If one of these jets happens to be directed back towards the quark surface any components of the particle cascade which are weakly coupled to the quark matter may escape across the sur- face. Strongly bound hadronic particles will remain bound to the nugget while electrons and positrons are unlikely to be able to be able to penetrate out through the electro- sphere and the associated strong electric ﬁelds. Conse- quently, muons produced in either the initial annihilation or the subsequent cascade will dominate particle emission from nugget (thought the total energy release is dominated The nuggets themselves will move at a much slower velocity than the secondary particles it emits (unlike con- ventional showers where both the primary and secondary particles move at the speed of light.) This introduces a much longer timescale for the arrival of shower compo- nents. In [13] this timescale was estimated as follows. The highest energy shower components are likely to be muons with QCD scale energies (E ∼GeV) in the ref- erence frame of a surface observer these particles have a lifetime on the order of 10µs during which time they travel a couple kilometers. Thus the cone of relativistic muons will extend roughly a kilometer ahead of the nugget. Once this cone reaches the surface relativistic muons will begin 04008-p.3 EPJ Web of Conferences 7 Radio signal The secondary particles emitted by the nugget will be de- ﬂected by the earth’s magnetic ﬁeld leading to the emis- sion of geosynchrotron radiation. This eﬀect has also been considered in the case of air showers induced by a single ultrahigh energy primary and would allow for observations with a much higher duty cycle than conventional ﬂuores- cence detection. As with the case of atmospheric ﬂuorescence the total magnitude of the atmospheric radio signal will be compa- rable to that of a conventional air shower. This is for the simple reason that total number of contributing relativistic particles (which sets the magnitude of both the ﬂuores- cence and geosynchrotron signals.) As with the ﬂuores- cence signal the clearest diﬀerentiating feature between a quark nugget driven air shower and that of a ultra high en- ergy cosmic ray is in the signal timing. The fact that the nugget travels considerably slower than the speed of light causes the radio emission to occur over millisecond time scales. 8 Conclusion to be seen in surface detectors. As the nuggets travel at velocities v ∼200km/s it will take on the order of a few milliseconds for the nugget itself to reach the surface at which point the emitted muons will be entirely absorbed by the earth and will no longer be detectable. A surface detector will thus detect particles over timescales from the millisecond down depending on the distance of the detec- tor from the shower centre. If the dark matter in our galaxy consists of nuggets of quark and antiquark matter then these nuggets, while evad- ing conventional dark matter searches, may be detectable by large scale cosmic ray detectors. If a nugget of anti- matter strikes the earth’s atmosphere it will induce an air shower of secondary particles at a scale similar to that pro- duced by a single ultra high energy cosmic ray. These air showers are detectable by the same means currently used to detect air showers associated with ultrahigh energy cos- mic rays. The main distinguishing feature of this type of shower is the signiﬁcantly longer time duration associated with the relatively low velocity of the primary. If cur- rently operating or planned cosmic ray detectors are ca- pable of searching for signals on this longer (millisecond) timescale they will be able to explore most of the interest- ing parameter space for dark matter of this form. References [1] Witten, E., Phys. Rev. D 30, (1984) 272-285 [2] Zhitnitsky, A., JCAP 10, (2003) [3] Oaknin, D. and Zhitnitsky, A., Phys. Rev. D 71, (2005) 023519 [1] Witten, E., Phys. Rev. D 30, (1984) 272-285 [2] Zhitnitsky, A., JCAP 10, (2003) [3] Oaknin, D. and Zhitnitsky, A., Phys. Rev. D 71, (2005) 023519 [4] Zhitnitsky, A., Phys. Rev. D 74, (2006), 043515 [5] Forbes, M.M., Lawson, K. and Zhitnitsky, A., Phys. Rev. D 82, (2010), 083510 [6] Knodlseder, J. et al., Astron. and Astrophys. 441, (2005) 513-532 A second signiﬁcant distinguishing feature of a quark nugget air shower in the radio band arises from the fact that the nugget itself is heated as it passes through the at- mosphere reaching keV scale temperatures near the earth’s surface. Despite the small area of the emitting surface such a high temperature can result in appreciable levels of ther- mal radiation. While this radiation is generally an order of magnitude smaller than the associated geosynchrotron signal it is spread across a broader range of wavelengths and a wider range of emission directions [14]. [7] Zhitnitsky, A., Phys. Rev. D 76, (2007) 103518 [8] Lawson, K. and Zhitnisky, A., JCAP 0801, (2008), 22 [9] Forbes, M.M. and Zhitnitsky, A., JCAP 0801, (2008), 23 [10] Forbes, M.M. and Zhitnitsky, A., Phys. Rev. D 78, (2008), 083505 [11] Muno, M.P. et al, Astrophys. J. 613, (2004), 326-342 [12] Lawson, K. and Zhitnitsky, A.R., arXiv:1210.2400 [13] Lawson, K., Phys. Rev. D 83, (2011), 103520 As the nugget’s temperature is set by the density of the surrounding material the thermal emission may be boosted once the nugget enters the earth. It has been suggested [15] that the thermal emission from a nugget passing through the radio transparent antarctic ice will present a relatively strong signal across the radio band. [14] Lawson, K., arXiv:1208.0042 [15] Gorham, P., Phys. Rev. D 86, (2012), 123005 04008-p.4
https://openalex.org/W4210688548	https://www.medrxiv.org/content/medrxiv/early/2022/01/30/2022.01.29.22270064.full.pdf	English	null	The Correlation Between Brain Performance Capacity and COVID-19: A Cross-sectional Survey and Canonical Correlation Analysis	medRxiv (Cold Spring Harbor Laboratory)	2,022	cc-by	11,669	1 The Correlation Between Brain Performance Capacity and COVID-19: A Cross- 2 sectional Survey and Canonical Correlation Analysis 3 Y Liu a, X Chen b, JS Xian c, R Wang c, K Ma c, K Xu d, X Yang c, FL Wang c, N Mu c, S Wang c, 4 Lai c, T Li a, CY Yang c, YL Quan c, H Feng c, LH Wang a, TN Chen c, d* s preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice. . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint CC BY 4 0 I t ti l li It i d il bl d perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in was not certified by peer review) The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: . CC-BY 4.0 International license It is made available under a perpetuity. . CC-BY 4.0 International license It is made available under a p p y 1 The Correlation Between Brain Performance Capacity and COVID-19: A Cross- 2 sectional Survey and Canonical Correlation Analysis 3 Y Liu a, X Chen b, JS Xian c, R Wang c, K Ma c, K Xu d, X Yang c, FL Wang c, N Mu c, S Wang c, Y 4 Lai c, T Li a, CY Yang c, YL Quan c, H Feng c, LH Wang a, TN Chen c, d* 5 a Army Medical University, Southwest Hospital, Administrative Office, Chongqing, 400038, China. 6 b Chongqing Medical University, University Town Hospital, Pediatrics Department, Chongqing 7 400031, China 8 c Army Medical University, Southwest Hospital, Neurosurgery Department, Chongqing 400038, China 9 d Army Medical University, Southwest Hospital, Army Aviation Medicine Teaching and Research 10 Office, Chongqing, 400038, China. 11 1 NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice. 400031, China . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint 12 ABSTRACT 13 Objective: To generate a concept of brain performance capacity (BPC) with sleep, fatigue and mental 14 workload as evaluation indicators and to analyze the correlation between BPC and the impact of COVID- 15 19. 16 Methods: A cluster sampling method was adopted to randomly select 259 civil air crew members. The 17 measurements of sleep quality, fatigue and mental workload (MWL) were assessed using the Pittsburgh 18 Sleep Quality Index (PSQI), Multidimensional Fatigue Inventory (MFI-20) and NASA Task Load Index. 19 The impact of COVID-19 included 7 dimensions scored on a Likert scale. Canonical correlation analysis 20 (CCA) was conducted to examine the relationship between BPC and COVID-19. 21 Results: A total of 259 air crew members participated in the survey. Participants’ average PSQI score 22 was 7.826 (SD = 3.796), with 49.8% reporting incidents of insomnia, mostly of a minor degree. 23 Participants’ MFI was an average 56.112 (SD = 10.040), with 100% reporting some incidence of fatigue, 24 mainly severe. The weighted mental workload (MWL) score was an average of 43.084 (SD = 17.543), 25 with reports of mostly a mid-level degree. There was a significant relationship between BPC and 26 COVID-19, with a canonical correlation coefficient of 0.507 (P=0.000), an eigenvalue of 0.364 and a 27 contribution rate of 69.1%. All components of the BPC variable set: PSQI, MFI and MWL contributed 28 greatly to BPC, with absolute canonical loadings of 0.790, 0.606 and 0.667, respectively; the same was 29 true for the COVID-19 variable set, with absolute canonical loadings ranging from 0.608 to 0.951. 30 Conclusion: Multiple indicators to measure BPC and the interrelationship of BPC and COVID-19 should 31 be used in future research to gain a comprehensive understanding of anti-epidemic measures to ensure 32 victory in the battle against the spread of the disease. 33 Keywords: COVID-19, mental health, brain performance capacity, flight crews 12 ABSTRACT 13 Objective: To generate a concept of brain performance capacity (BPC) with sleep, fatigue and mental 14 workload as evaluation indicators and to analyze the correlation between BPC and the impact of COVID- 15 19. . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint 400031, China 16 Methods: A cluster sampling method was adopted to randomly select 259 civil air crew members. The 17 measurements of sleep quality, fatigue and mental workload (MWL) were assessed using the Pittsburgh 18 Sleep Quality Index (PSQI), Multidimensional Fatigue Inventory (MFI-20) and NASA Task Load Index. 19 The impact of COVID-19 included 7 dimensions scored on a Likert scale. Canonical correlation analysis 20 (CCA) was conducted to examine the relationship between BPC and COVID-19. 34 INTRODUCTION 34 INTRODUCTION 35 Brain performance capacity, a complex concept, is associated with memory, cognitive capacities, 36 attention, and work efficiency. This has been tested by research and resulted in much evidence. As early 37 as 1975, Kahn, R L(1) held the idea that impaired memory and depression were the products of poor 38 brain performance. L. Nyberg also argued that brain maintenance constituted the primary determinant of 39 successful memory aging (2). In modern society, it is vital for optimal performance to enhance or 40 preserve the cognitive performance of personnel working in stressful, demanding and/or high-tempo 41 environments (3). Furthermore, the ability to focus attention on tasks may contribute to peak brain 42 performance and high-level work efficiency. A. Yamashita found that individuals with attention- 43 deficit/hyperactivity disorder (ADHD) spent more time and energy maintaining optimal and effective 44 performance behavior during tasks that required sustained attention (4). 45 It is well known that in some professional contexts, such as aviation, where staying in good physical 46 condition is requisite, there are additional traits, such as BPC, that are important factors for success. 47 However, how can this aspect be valuated? We have a remarkably poor understanding of what the direct 48 indicator might be in measuring BPC. 49 However, there are some factors, such as sleep, fatigue-free state, and appropriate level of mental 50 workload, which are thought to be related to performance; vigilant attention and long memory are also 51 traits that may indirectly boost peak brain performance. Much evidence suggests that the right number 52 of hours and quality of sleep improves working memory (5), vigilant attention (6), concentration (7), but 53 fatigue and high-level mental workload impair them (8-13). Taking flight crews as an example, members 54 and particularly pilots, are submitted to high, exigent work demands and must manage multiple tasks at 55 the same time. In addition, they are continuously exposed to stimuli that compete for their attention and 35 Brain performance capacity, a complex concept, is associated with memory, cognitive capacities, 36 attention, and work efficiency. This has been tested by research and resulted in much evidence. As early 37 as 1975, Kahn, R L(1) held the idea that impaired memory and depression were the products of poor 38 brain performance. L. Nyberg also argued that brain maintenance constituted the primary determinant of 39 successful memory aging (2). 12 ABSTRACT 33 Keywords: COVID-19, mental health, brain performance capacity, flight crews 2 . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a pe petu ty . CC-BY 4.0 International license It is made available under a p p y . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint 34 INTRODUCTION In modern society, it is vital for optimal performance to enhance or 40 preserve the cognitive performance of personnel working in stressful, demanding and/or high-tempo 41 environments (3). Furthermore, the ability to focus attention on tasks may contribute to peak brain 42 performance and high-level work efficiency. A. Yamashita found that individuals with attention- 43 deficit/hyperactivity disorder (ADHD) spent more time and energy maintaining optimal and effective 44 performance behavior during tasks that required sustained attention (4). 3 45 It is well known that in some professional contexts, such as aviation, where staying in good physical 46 condition is requisite, there are additional traits, such as BPC, that are important factors for success. 47 However, how can this aspect be valuated? We have a remarkably poor understanding of what the direct 48 indicator might be in measuring BPC. 49 However, there are some factors, such as sleep, fatigue-free state, and appropriate level of mental 50 workload, which are thought to be related to performance; vigilant attention and long memory are also 51 traits that may indirectly boost peak brain performance. Much evidence suggests that the right number 52 of hours and quality of sleep improves working memory (5), vigilant attention (6), concentration (7), but 53 fatigue and high-level mental workload impair them (8-13). Taking flight crews as an example, members 54 and particularly pilots, are submitted to high, exigent work demands and must manage multiple tasks at 55 the same time. In addition, they are continuously exposed to stimuli that compete for their attention and 45 It is well known that in some professional contexts, such as aviation, where staying in good physical 46 condition is requisite, there are additional traits, such as BPC, that are important factors for success. 47 However, how can this aspect be valuated? We have a remarkably poor understanding of what the direct 48 indicator might be in measuring BPC. 49 However, there are some factors, such as sleep, fatigue-free state, and appropriate level of mental 50 workload, which are thought to be related to performance; vigilant attention and long memory are also 51 traits that may indirectly boost peak brain performance. Much evidence suggests that the right number 52 of hours and quality of sleep improves working memory (5), vigilant attention (6), concentration (7), but 53 fatigue and high-level mental workload impair them (8-13). . CC-BY 4.0 International license It is made available under a p p y . CC-BY 4.0 International license It is made available under a p p y 34 INTRODUCTION Taking flight crews as an example, members 54 and particularly pilots, are submitted to high, exigent work demands and must manage multiple tasks at 55 the same time. In addition, they are continuously exposed to stimuli that compete for their attention and 3 . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint CC BY 4 0 I i l li I i d il bl d perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in was not certified by peer review) The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: . CC-BY 4.0 International license It is made available under a pe petu ty . CC-BY 4.0 International license It is made available under a pe petu ty . CC-BY 4.0 International license It is made available under a pe petu ty 4 56 ability to manage their resources to make the right decisions. These task-oriented responsibilities are 57 further complicated by complex flying factors, such as long trips and varying shift flights. 58 It was reported that the majority of accidents, 60%~90%, are attributed to “human error” (14-16) 59 and occur when flight crews are subjected to a high and intensive mental workload level (17). Thus, BPC 60 is a prerequisite to ensure flight safety. The COVID-19 pandemic has had a huge impact on people’s 61 lives, the economy, physical-mental health, and BPC, since the outbreak began in 2020. Because of the 62 instability and resurgence of the disease, it will continue to have a negative impact for some time even 63 though we have entered a new period known as the normalized era of COVID-19. Like the general 64 population, aviation crews have been exposed to high-risk environments COVID-19 environments for a 65 long time. 34 INTRODUCTION have physical and mental illness. The eligible participants were asked to answer all the 81 questions about sleep, fatigue and mental workload in the previous month. Subjects logged on to 82 Questionnaire Star via WeChat with a smartphone or computer. Individuals completing the survey 83 received a red envelope with a random amount of money as compensation from Questionnaire Star. 84 Demographics including gender, age, BMI, position, and social jet lag, were collected. Respondents 85 were also asked about the severity of the impact of COVID-19 on them and scored their responses on a 86 Likert scale (1= never, 5 = extremely serious). This scale included seven dimensions: fear of self- 87 infection (C1); fear of ineffective prevention and control (C2); physical unwellness caused by regular 88 nucleic acid testing (C3); increased workload of personal protective equipment (C4); sleep quality (C5); 89 mental fatigue (C6); and physical fatigue (C7). 90 2.2 Scales to measure indicators of BPC 91 2.2.1 Fatigue Assessment 92 A multidimensional fatigue inventory (MFI-20) was conducted to evaluate the fatigue of 93 commercial pilots. The scale was created by Ellen Smets and colleagues (19) in 1995 and has been 94 translated into various languages by other researchers worldwide. The 20-item scale was designed to 95 evaluate five dimensions of fatigue: general fatigue, physical fatigue, reduced activity, reduced 96 motivation and mental fatigue. Twenty questions were rated on a 5-point scale (“yes, that is true” to “no, 97 that is not true”). The higher the total score was, the greater the fatigue. The total MFI scores range from 98 0 to 100 and are interpreted as no fatigue (0-20), minor fatigue (21-39), serious fatigue (40-59), severe 99 fatigue (60-79) and extremely severe fatigue (80-100). This scale has been verified by Chinese scholars 78 in this survey, b. have the ability to use basic modern information technology, such as WeChat or 79 computers. Exclusion criteria: a. not full-attended during the investigation request period due to any 80 reasons, b. have physical and mental illness. The eligible participants were asked to answer all the 81 questions about sleep, fatigue and mental workload in the previous month. Subjects logged on to 82 Questionnaire Star via WeChat with a smartphone or computer. Individuals completing the survey 83 received a red envelope with a random amount of money as compensation from Questionnaire Star. 34 INTRODUCTION Thus, this study aimed to assess the impact in the following ways: 1) update information about 66 the sleep quality, fatigue and mental workload of commercial flight crews in the normalized era of 67 COVID-19 in China, and 2) generate a dependent variable cluster named BPC with indicators such as 68 sleep quality, fatigue and mental workload and to explore the relationship between BPC and the impact 69 of COVID-19 in this new era. 70 METHODS 71 2.1 Participants and study design 72 This was a cluster sampling and cross-sectional study. The recommended minimum sample size 73 was 200, which was 10 times the number of MFI-20 for a maximum number of entries in this study (18). 74 Estimating a 20% invalid survey response rate, the expected sample was at least 240. The final sample 75 was 259. 76 Questionnaire Star was used to collect sociodemographic data and key targets (sleep, fatigue and 77 mental workload) from 25 November to 2 December 2021. Inclusion criteria: a. volunteer to participate . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint CC BY 4 0 I i l li I i d il bl d perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in was not certified by peer review) The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: . CC-BY 4.0 International license It is made available under a pe petu ty . CC-BY 4.0 International license It is made available under a p p y . CC-BY 4.0 International license It is made available under a p p y 5 78 in this survey, b. have the ability to use basic modern information technology, such as WeChat or 79 computers. Exclusion criteria: a. not full-attended during the investigation request period due to any 80 reasons, b. 34 INTRODUCTION 84 Demographics including gender, age, BMI, position, and social jet lag, were collected. Respondents 85 were also asked about the severity of the impact of COVID-19 on them and scored their responses on a 86 Likert scale (1= never, 5 = extremely serious). This scale included seven dimensions: fear of self- 87 infection (C1); fear of ineffective prevention and control (C2); physical unwellness caused by regular 88 nucleic acid testing (C3); increased workload of personal protective equipment (C4); sleep quality (C5); 89 mental fatigue (C6); and physical fatigue (C7). 90 2.2 Scales to measure indicators of BPC 91 2.2.1 Fatigue Assessment 92 A multidimensional fatigue inventory (MFI-20) was conducted to evaluate the fatigue of 93 commercial pilots. The scale was created by Ellen Smets and colleagues (19) in 1995 and has been 94 translated into various languages by other researchers worldwide. The 20-item scale was designed to 78 in this survey, b. have the ability to use basic modern information technology, such as WeChat or 79 computers. Exclusion criteria: a. not full-attended during the investigation request period due to any 80 reasons, b. have physical and mental illness. The eligible participants were asked to answer all the 81 questions about sleep, fatigue and mental workload in the previous month. Subjects logged on to 82 Questionnaire Star via WeChat with a smartphone or computer. Individuals completing the survey 83 received a red envelope with a random amount of money as compensation from Questionnaire Star. 84 Demographics including gender, age, BMI, position, and social jet lag, were collected. Respondents 85 were also asked about the severity of the impact of COVID-19 on them and scored their responses on a 86 Likert scale (1= never, 5 = extremely serious). This scale included seven dimensions: fear of self- 87 infection (C1); fear of ineffective prevention and control (C2); physical unwellness caused by regular 88 nucleic acid testing (C3); increased workload of personal protective equipment (C4); sleep quality (C5); 89 mental fatigue (C6); and physical fatigue (C7). 92 A multidimensional fatigue inventory (MFI-20) was conducted to evaluate the fatigue of 93 commercial pilots. The scale was created by Ellen Smets and colleagues (19) in 1995 and has been 94 translated into various languages by other researchers worldwide. The 20-item scale was designed to 95 evaluate five dimensions of fatigue: general fatigue, physical fatigue, reduced activity, reduced 96 motivation and mental fatigue. . CC-BY 4.0 International license It is made available under a p p y 34 INTRODUCTION Twenty questions were rated on a 5-point scale (“yes, that is true” to “no, 97 that is not true”). The higher the total score was, the greater the fatigue. The total MFI scores range from 98 0 to 100 and are interpreted as no fatigue (0-20), minor fatigue (21-39), serious fatigue (40-59), severe 99 fatigue (60-79) and extremely severe fatigue (80-100). This scale has been verified by Chinese scholars 5 5 . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a p p y 100 and can be used to test the fatigue of Chinese pilots (20). 101 2.2.2 Sleep Assessment 102 The Pittsburgh Sleep Quality Index (PSQI) was used to assess the sleep quality of the participants. 103 This mature scale, widely used to evaluate sleep quality, has 7 items: sleep latency, sleep duration, sleep 104 quality, habitual efficiency, sleep disturbance, use of sleeping medications and daytime dysfunction21. 105 Total PSQI scores range from 0 to 21, and 8 points or more is regarded as a sleep disorder, 8-10 points 106 as a minor sleep disorder, 11-15 points as a serious sleep disorder, and 16 points or more as a severe 107 sleep disorder (22). The higher the score was, the worse the sleep quality. 108 2.2.3 Mental Workload Assessment 109 NASA Task Load Index (NASA TLX) (23) was utilized to measure the mental workload of the 110 participants. It is a tool for measuring and conducting a subjective mental workload (MWL) assessment. 111 It allows the subject to determine the MWL of a participant while they are performing a task. It rates 112 performance across six dimensions to determine an overall workload rating. The six dimensions are as 113 follows: 114 (1). Mental demand: How much thinking, deciding, or calculating was required to perform the task? 115 (2). Physical demand: How much physical activity and intensity was required to complete the task? 116 (3). Temporal demand: What kind of time pressure was involved in completing the task? 117 (4). Effort: How hard does the participant have to work to maintain his or her level of performance? . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint 34 INTRODUCTION is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint CC BY 4 0 I i l li I i d il bl d perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in was not certified by peer review) The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: . CC-BY 4.0 International license It is made available under a p p y 122 asked to rate their score on an interval scale ranging from low (1) to high (100). The TLX also employs 123 a paired comparisons procedure. This involves presenting 15 pairwise combinations to the participants 124 and asking them to select the scale from each pair that has the most effect on the workload during the 125 task under analysis. This procedure accounts for two potential sources of between-rater variability: 126 differences in workload definition between the raters and differences in the sources or workload between 127 the tasks. The total MWL score is 0-100 points. The higher the scores, the higher the mental workload. 128 In this study, we used the weighted workload score. This scale has been verified by Chinese scholars and 129 can be used to test the mental workload of Chinese pilots. 130 2.3 Statistical analyses 131 The statistical analyses were carried out by IBM SPSS Statistics 26.0. All answer sheets were error- 132 checked by the study team. The QQ chart and Bonferroni test were used to check the former's normality 133 and homogeneity of the variance. Canonical Correlation Analysis (CCA), a multivariate analysis of the 134 association between sets of multiple independent and dependent variables, was administered to explore 135 the relationship between BPC and the impact of COVID-19. Each set can be named. The canonical 136 correlation coefficient (Rc) maximizes the correlation between the canonical variate obtained by 137 synthesizing multiple independent and dependent variables. 34 INTRODUCTION 118 (5). Performance: How successful was the individual in completing the task? 119 (6). Frustration level: How insecure and discouraged or secure and content did the participant feel 120 during the task? 121 Each subscale is presented to the participants either during or after the experimental trial. They are 100 and can be used to test the fatigue of Chinese pilots (20). 101 2.2.2 Sleep Assessment 102 The Pittsburgh Sleep Quality Index (PSQI) was used to assess the sleep quality of the participants. 103 This mature scale, widely used to evaluate sleep quality, has 7 items: sleep latency, sleep duration, sleep 104 quality, habitual efficiency, sleep disturbance, use of sleeping medications and daytime dysfunction21. 105 Total PSQI scores range from 0 to 21, and 8 points or more is regarded as a sleep disorder, 8-10 points 106 as a minor sleep disorder, 11-15 points as a serious sleep disorder, and 16 points or more as a severe 107 sleep disorder (22). The higher the score was, the worse the sleep quality. 6 108 2.2.3 Mental Workload Assessment 109 NASA Task Load Index (NASA TLX) (23) was utilized to measure the mental workload of the 110 participants. It is a tool for measuring and conducting a subjective mental workload (MWL) assessment. 111 It allows the subject to determine the MWL of a participant while they are performing a task. It rates 112 performance across six dimensions to determine an overall workload rating. The six dimensions are as 113 follows: 114 (1). Mental demand: How much thinking, deciding, or calculating was required to perform the task? 115 (2). Physical demand: How much physical activity and intensity was required to complete the task? 116 (3). Temporal demand: What kind of time pressure was involved in completing the task? 117 (4). Effort: How hard does the participant have to work to maintain his or her level of performance? 118 (5). Performance: How successful was the individual in completing the task? 119 (6). Frustration level: How insecure and discouraged or secure and content did the participant feel 120 during the task? 121 Each subscale is presented to the participants either during or after the experimental trial. They are 6 . CC-BY 4.0 International license It is made available under a perpetuity. loading >0.30 was regarded as an important contribution. . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint 34 INTRODUCTION In this study, we constructed an independent 138 variable set named COVID-19 that affected the dependent variable set: sleep quality, fatigue state and 139 mental workload, which was named brain performance capacity (BPC). The relationships between the 140 2 sets of variables were interpreted using canonical loadings to determine how much each variable 141 contributed to its own set and using canonical cross-loadings to determine how much each variable 142 contributed to the other set. The significance level (p) of the correlation was set at 0.05, and a 143 loading >0.30 was regarded as an important contribution. 131 The statistical analyses were carried out by IBM SPSS Statistics 26.0. All answer sheets were error- 132 checked by the study team. The QQ chart and Bonferroni test were used to check the former's normality 133 and homogeneity of the variance. Canonical Correlation Analysis (CCA), a multivariate analysis of the 134 association between sets of multiple independent and dependent variables, was administered to explore 135 the relationship between BPC and the impact of COVID-19. Each set can be named. The canonical 136 correlation coefficient (Rc) maximizes the correlation between the canonical variate obtained by 137 synthesizing multiple independent and dependent variables. In this study, we constructed an independent 138 variable set named COVID-19 that affected the dependent variable set: sleep quality, fatigue state and 139 mental workload, which was named brain performance capacity (BPC). The relationships between the 140 2 sets of variables were interpreted using canonical loadings to determine how much each variable 141 contributed to its own set and using canonical cross-loadings to determine how much each variable 142 contributed to the other set. The significance level (p) of the correlation was set at 0.05, and a 143 loading >0.30 was regarded as an important contribution. 7 7 . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a p p y 144 RESULTS 145 3.1 Participant 146 In total, 259 flight crew members participated, and the number of valid responses was 100% without any 147 missing data. Overall, 51.7% of the flight crew members (n = 134) were male, and 48.3% were female 148 (n=125). The age of the participants ranged from 17 to 55 years, with an average of 28.204 ± 6.728. The 149 weighted NASA-TLX scores ranged from 0 to 100, and the mean score was 43.084 ± 17.543. The PSQI 150 scores ranged from 0~22 with a mean score of 7.826 ± 3.796, and the MFI score ranged from 22 to 77 151 with a mean score of 56.112 ± 10.040. The insomnia incidence of the participants was 49.8%, mainly 152 minor insomnia (30.2%), followed by serious (15.8%) and severe insomnia (1.9%), with a significant 153 difference among the insomnia degrees (X2=134.745, P=0.000). The fatigue incidence was 100%, mainly 154 severe fatigue (47.9%), followed by serious (42.1%) and minor fatigue (10%), with significant difference 155 among them (X2=64.548, P=0.000). The distribution of civil aviation crew BPCs is detailed in Table 1. 156 3.2 CCA of BPC and COVID-19 157 3.2.1 Spearman’s Correlation 158 Table 2 shows the correlation coefficient matrix among the three components of BPC scales and the 159 seven domains of COVID-19 impact on the civil aviation crews. Except for C4 and MFI, the correlation 160 coefficients among the components were significant, from the lowest absolute value of 0.140 (P<0.05) 161 to the largest absolute value of 0.896 (P < 0.01). 162 3.2.2 Canonical Correlation 163 The CCA was performed between BPC variables and COVID-19 variables, and the outcome 164 contained 3 pairs of canonical variables. The correlation coefficient of the first pair of canonical variates 165 was statistically significant (P=0.000 < 0.05), with the correlation coefficient having a value of 0.517, 162 3.2.2 Canonical Correlation 163 The CCA was performed between BPC variables and COVID-19 variables, and the outcome 164 contained 3 pairs of canonical variables. The correlation coefficient of the first pair of canonical variates 165 was statistically significant (P=0.000 < 0.05), with the correlation coefficient having a value of 0.517, 8 8 CC-BY 4 0 International license It is made available under a perpetuity. 144 RESULTS is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a p p y 166 an eigenvalue of 0.364, and a contribution rate of 69.1%. Thus, only the first pair of canonical variables 167 was analyzed further (Table 3). 168 According to Table 4, which shows the standardized canonical correlation coefficients, a set of 169 linear combinations within a group of typical variables can be obtained: 170 BPC=-0.485MWL-0.610PSQI-0.312MFI 171 COVID-19=0.088C1-0.114C2-0.511C3+0.464C4-0.872C5+0.086C6-0.132C7 172 The proportion variance explained by the BPC set was 48.3%, and that explained by the COVID- 173 19 set was 58%. 174 Fig 1 shows the canonical loadings of the BPC set and the COVID-19 set. All MWL, PSQI and MFI 175 contribute greatly to BPC, with an absolute value of the lowest canonical loading coefficient of 0.606. 176 All components of COVID-19 contribute greatly to COVID-19, with absolute values of canonical 177 loading coefficients from 0.608 to 0.951. This indicates that the indicators MWL, PSQI and MFI could 178 well inflect the BPC. Similarly, C1, C2, C3, C4, C5, C6 and C7 present the impacts of COVID-19 on 179 civil aviation crews. 180 DISCUSSION 181 This study updated the data on civil aviation crews regarding sleep quality, fatigue and mental 182 workload in the new stage of normalized epidemic era of COVID-19 in China. The insomnia incidence 183 was 49.8%, and gender, position and driving experience had no significant impact. There was a high 180 DISCUSSION 181 This study updated the data on civil aviation crews regarding sleep quality, fatigue and mental 182 workload in the new stage of normalized epidemic era of COVID-19 in China. The insomnia incidence 183 was 49.8%, and gender, position and driving experience had no significant impact. There was a high 184 level of fatigue, with a 100% incidence reported by participants, which was much higher than that 185 reported by US army aviators (24). 144 RESULTS Men were more tired than women. The MWL was reported to be 186 mid-level, with a weighted score of 43.084 ± 17.543. Furthermore, the older the age and the higher the 187 BMI were, the worse the sleep, the more fatigued and the higher the mental workload level. 9 9 . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint CC perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in was not certified by peer review) The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: . CC-BY 4.0 International license It is made available under a p p y . CC-BY 4.0 International license It is made available under a p p y . CC-BY 4.0 International license It is made available under a p p y 188 It also first reported here that a comprehensive evaluation of BPC can be made through the 189 assessment index cluster we built. Many studies have reported the effects of sleep quality, fatigue and 190 mental workload on individuals’ memory, attention, reaction, and mental health, which could directly or 191 indirectly respond to brain performance. Yamashita, A hold the idea (25) that individuals with attention- 192 deficit/hyperactivity disorder (ADHD) have unstable and unreliable behaviors related to the state of their 193 brains, whereas those without ADHD have an optimal brain state that guides their more stable and 194 reliable behavior. The brain has plasticity. L. L. Beason-Held (26) tested the hypotheses that stable task 195 performance could be supported by stable brain performance or that stable performance could be 196 maintained by changing brain functions if the change was a compensatory reorganization of function. 197 We also found that sleep quality, fatigue and mental workload responded well to BPC. 144 RESULTS Each indicator 198 contributed greatly to BPC, as there were strong loadings of MWL (loading=-0.677), PSQI (loading=- 199 0.790), and MFI (loading=-0.606). As concluded by Walker, Matthew P, et al., the brain under sleep 200 deprivation exhibits inattention, decreased working memory, and insufficient motivation to learn (27). 201 We also found that the higher the MWL level was, the lower the BPC; the worse the sleep quality was, 202 the lower the BPC; and the more fatigued an individual was, the lower his or her BPC. These findings 203 may provide new perspectives on how to evaluate BPC. 204 Additionally, this study highlighted the impact of COVID-19 on civil flight crews. The COVID-19 205 pandemic has had a profound impact on the aviation industry, leading to job loss, furlough, disruption of 206 careers and training trajectories and uncertainties, even leading to self-harm among airline employees 207 (28). In addition, there is evidence that COVID-19 induces fears, including fear of infection and fear of 208 passing the infection to others due to ineffective prevention and control measures from the government. 209 Due to the vagaries of the epidemic, people receive nucleic acid tests (NATs) on a fixed schedule or and 10 . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint CC BY 4 0 I i l li I i d il bl d perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in was not certified by peer review) The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: . CC-BY 4.0 International license It is made available under a perpetuity. . CC-BY 4.0 International license It is made available under a pe petu ty 210 at any time during the period of normalized control of the epidemic in China. 144 RESULTS Special occupations in the 211 airlines and health care require not only masks but also personal protective equipment such as gloves, 212 gowns and hats, which is bound to increase the workload of these employees. COVID-19 causes sleep 213 problems such as difficulties initiating and maintaining sleep (29) and fatigue, including mental fatigue 214 and physical fatigue (30). With the statistical analysis of canonical loadings of the COVID-19 set, we 215 also found that the fear of infection(C1), fear of ineffective prevention and control(C2), physical 216 unwellness caused by NAT(C3), increased workload of personal protective equipment(C4), sleep 217 quality(C5), mental fatigue(C6) and physical fatigue(C7) all contributed greatly to the impact of COVID- 218 19, with a loading of -0.608, -0.689, -0.774, -0.617, -0.951, -0.818 and -0.813, respectively. The more 219 severe the impact of COVID-19 was, the worse the BPC. 211 airlines and health care require not only masks but also personal protective equipment such as gloves, 212 gowns and hats, which is bound to increase the workload of these employees. COVID-19 causes sleep 213 problems such as difficulties initiating and maintaining sleep (29) and fatigue, including mental fatigue 214 and physical fatigue (30). With the statistical analysis of canonical loadings of the COVID-19 set, we 215 also found that the fear of infection(C1), fear of ineffective prevention and control(C2), physical 216 unwellness caused by NAT(C3), increased workload of personal protective equipment(C4), sleep 217 quality(C5), mental fatigue(C6) and physical fatigue(C7) all contributed greatly to the impact of COVID- 218 19, with a loading of -0.608, -0.689, -0.774, -0.617, -0.951, -0.818 and -0.813, respectively. The more 219 severe the impact of COVID-19 was, the worse the BPC. 220 We also explored the relationship between BPC and the impact of COVID-19 with CCA. We found 221 that a moderate positive correlation existed between BPC and the impact of COVID-19, with a 222 correlation of 0.517. The correlation indicated that the more severe the impact of COVID-19 was on civil 223 aviation crews, the stronger the BPC. However, the pandemic has been in a normalized control period in 224 China. 144 RESULTS is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint CC BY I i l li I i d il bl d perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in was not certified by peer review) The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: . CC-BY 4.0 International license It is made available under a pe petu ty . CC-BY 4.0 International license It is made available under a p p y . CC-BY 4.0 International license It is made available under a p p y 232 cannot provide a cause-and-effect relationship. Because BPC is dynamic, more rigorous study designs, 233 such as cohort studies, should be conducted in the future. Second, although sleep, fatigue and mental 234 workload were successfully used as evaluation indicators of BPC, the assessment tools in this study were 235 mainly scales that are subjective. Therefore, more evaluation indicators of BPC should be explored and 236 combined with objective evaluation indicators. 237 CONCLUSION 238 BPC could be assessed by sleep quality, fatigue and mental workload. In the normalized epidemic 239 era of COVID-19 in China, the pandemic can positively affect the level of brain performance of 240 commercial flight crews. Considerations including multiple indicators to measure BPC and the 241 relationship between BPC and COVID-19 should be considered in future research to gain a 242 comprehensive view of anti-epidemic measures as soon as possible. 243 Acknowledgements 244 Ethical approval and consent to participate 245 Guided by the 2000 Declaration of Helsinki for ethical standards, the protocol was approved by the 246 Ethics Committee of the First Affiliated Hospital of Army Medical University, PLA (Ref: KY2021047). 247 Informed consent was obtained from the participants prior to their participation. There, they were able 248 to read a 2-page informed consent form. After reading the consent, they could decide whether to take 249 part in this survey. Those who agreed to participate in the study were required to electronically sign the 250 consent form and then to complete the questionnaire; otherwise, the Questionnaire Star would end the 251 interview directly. 252 Funding 253 This study was supported by Chongqing Municipal Education Commission titled Research on the . 144 RESULTS As of 24:00 on November 25, 2021, there had been 860 confirmed cases (including 8 severe 225 cases), 93,087 cured and discharged cases, 4,636 deaths, 98,583 confirmed cases and 4 suspected cases; 226 a total of 1,315,485 close contacts had been traced, and 28,342 close contacts were still under medical 227 observation, according to a report from the National Health Commission of the People’s Republic of 228 China (31). COVID-19 was still a predictor of BPC. This finding may suggest that there is a long way to 229 go in the fight against COVID-19, and the negative influences should not be underestimated. There are 230 some valid concerns about how to boost the BPC to respond to the impact of COVID-19 on individuals. 11 220 We also explored the relationship between BPC and the impact of COVID-19 with CCA. We found 221 that a moderate positive correlation existed between BPC and the impact of COVID-19, with a 222 correlation of 0.517. The correlation indicated that the more severe the impact of COVID-19 was on civil 223 aviation crews, the stronger the BPC. However, the pandemic has been in a normalized control period in 224 China. As of 24:00 on November 25, 2021, there had been 860 confirmed cases (including 8 severe 225 cases), 93,087 cured and discharged cases, 4,636 deaths, 98,583 confirmed cases and 4 suspected cases; 226 a total of 1,315,485 close contacts had been traced, and 28,342 close contacts were still under medical 227 observation, according to a report from the National Health Commission of the People’s Republic of 228 China (31). COVID-19 was still a predictor of BPC. This finding may suggest that there is a long way to 229 go in the fight against COVID-19, and the negative influences should not be underestimated. There are 230 some valid concerns about how to boost the BPC to respond to the impact of COVID-19 on individuals. 231 This study has limitations that need to be considered. First, this was a cross-sectional study, so it 11 . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a perpetuity. This study was supported by Chongqing Municipal Education Commission titled Research on the 144 RESULTS CC-BY 4.0 International license It is made available under a 232 cannot provide a cause-and-effect relationship. Because BPC is dynamic, more rigorous study designs, 233 such as cohort studies, should be conducted in the future. Second, although sleep, fatigue and mental 234 workload were successfully used as evaluation indicators of BPC, the assessment tools in this study were 235 mainly scales that are subjective. Therefore, more evaluation indicators of BPC should be explored and 236 combined with objective evaluation indicators. 237 CONCLUSION 238 BPC could be assessed by sleep quality, fatigue and mental workload. In the normalized epidemic 239 era of COVID-19 in China, the pandemic can positively affect the level of brain performance of 240 commercial flight crews. Considerations including multiple indicators to measure BPC and the 241 relationship between BPC and COVID-19 should be considered in future research to gain a 242 comprehensive view of anti-epidemic measures as soon as possible. 243 Acknowledgements 244 Ethical approval and consent to participate 245 Guided by the 2000 Declaration of Helsinki for ethical standards, the protocol was approved by the 246 Ethics Committee of the First Affiliated Hospital of Army Medical University, PLA (Ref: KY2021047). 247 Informed consent was obtained from the participants prior to their participation. There, they were able 237 CONCLUSION 238 BPC could be assessed by sleep quality, fatigue and mental workload. In the normalized epidemic 239 era of COVID-19 in China, the pandemic can positively affect the level of brain performance of 240 commercial flight crews. Considerations including multiple indicators to measure BPC and the 241 relationship between BPC and COVID-19 should be considered in future research to gain a 242 comprehensive view of anti-epidemic measures as soon as possible. 12 CC BY 4 0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint CC-BY 4 0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint perpetuity. 144 RESULTS is the author/funder, who has granted medRxiv a license to display the preprint in d by peer review) The copyright holder for this this version posted January 30, 2022. ; g/10.1101/2022.01.29.22270064 . CC-BY 4.0 International license It is made available under a p p y . CC-BY 4.0 International license It is made available under a p p y 254 Construction of Intelligent Supervision System of Mental Workload and Its Key Technologies f g p y f y g 255 (CYS21523). 256 Competing interesting 257 The authors declare that they have no competing interests. All authors have read the manuscript and 258 agreed to publish. 259 Clinical Trial Registry 260 This study is registered in the Chinese Clinical Trial Registry, number ChiCTR2100053133. 261 Authors’ contributions 262 HF, TNC and LHW conceived, designed, and supervised the study; YL, XC, YL, NM, KX collected 263 the data; YL, JSX, FLW and TL finalized the analysis, designed the study and interpreted the findings; 264 YL and XC drafted the manuscript; YLQ, SW, RW and CYY interpreted the findings and helped revise 265 drafts of the manuscript. all authors have read and approved the final version of the manuscript, and agree 266 with the order of presentation of the authors. 267 References 268 1. Kahn RL, Zarit SH, Hilbert NM, Niederehe G. Memory complaint and impairment in the 269 aged. The effect of depression and altered brain function. Archives of general psychiatry. 270 1975;32(12):1569-73. 271 2. Nyberg L, Lovden M, Riklund K, Lindenberger U, Backman L. Memory aging and brain 272 maintenance. Trends Cogn Sci. 2012;16(5):292-305. 273 3. Tooley KL. Effects of the Human Gut Microbiota on Cognitive Performance, Brain 274 Structure and Function: A Narrative Review. Nutrients. 2020;12(10):17. Tooley KL. Effects of the Human Gut Microbiota on Cognitive Performance, Brain Structure and Function: A Narrative Review. Nutrients. 2020;12(10):17. Yamashita A, Rothlein D, Kucyi A, Valera EM, Esterman M. Brain state-based detection of 13 . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint 276 attentional fluctuations and their modulation. Neuroimage. 2021;236. 277 5. Goldi M, Rasch B. Effects of targeted memory reactivation during sleep at home depend on 278 sleep disturbances and habituation. 144 RESULTS Npj Science of Learning. 2019;4(1). 279 6. Mao TX, Dinges D, Deng Y, Zhao K, Yang ZJ, Lei H, et al. Impaired Vigilant Attention 280 Partly Accounts for Inhibition Control Deficits After Total Sleep Deprivation and Partial 281 Sleep Restriction. Nature and Science of Sleep. 2021;13:1545-60. 282 7. Hudson AN, Van Dongen HPA, Honn KA. Sleep deprivation, vigilant attention, and brain 283 function: a review. Neuropsychopharmacology. 2020;45(1):21-30. 284 8. Hoskens MCJ, Uiga L, Cooke A, Capio CM, Masters RSW. The effects of fatigued working 285 memory functions on hypothesis testing during acquisition of a motor skill. Journal of 286 experimental psychology General. 2021. 287 9. Jain S, Nataraja NP. The Effect of Fatigue on Working Memory and Auditory Perceptual 288 Abilities in Trained Musicians. American Journal of Audiology. 2019;28(2):483-94. 289 10. Wang H, Liu X, Li J, Xu T, Bezerianos A, Sun Y, et al. Driving Fatigue Recognition With 290 Functional Connectivity Based on Phase Synchronization. IEEE Transactions on Cognitive 291 and Developmental Systems. 2021;13(3):668-78. 292 11. Riley EA, Owora A, McCleary J, Anderson A. Sleepiness, Exertion Fatigue, Arousal, and 293 Vigilant Attention in Persons With Chronic Aphasia. American Journal of Speech-Language 294 Pathology. 2019;28(4):1491-508. 295 12. Xing X, Zhong B, Luo H, Rose T, Li J, Antwi-Afari MF. Effects of physical fatigue on the attentional fluctuations and their modulation. Neuroimage. 2021;236. Goldi M, Rasch B. Effects of targeted memory reactivation during sleep at home depend on sleep disturbances and habituation. Npj Science of Learning. 2019;4(1). Sleep Restriction. Nature and Science of Sleep. 2021;13:1545-60. Hudson AN, Van Dongen HPA, Honn KA. Sleep deprivation, vigilant attention, and brain function: a review. Neuropsychopharmacology. 2020;45(1):21-30. 283 function: a review. Neuropsychopharmacology. 2020;45(1):21-30. 284 8. Hoskens MCJ, Uiga L, Cooke A, Capio CM, Masters RSW. The effects of fatigued working 285 memory functions on hypothesis testing during acquisition of a motor skill. Journal of 286 experimental psychology General. 2021. 287 9. Jain S, Nataraja NP. The Effect of Fatigue on Working Memory and Auditory Perceptual 288 Abilities in Trained Musicians. American Journal of Audiology. 2019;28(2):483-94. 289 10. Wang H, Liu X, Li J, Xu T, Bezerianos A, Sun Y, et al. Driving Fatigue Recognition With 290 Functional Connectivity Based on Phase Synchronization. IEEE Transactions on Cognitive 291 and Developmental Systems. 2021;13(3):668-78. 292 11. Riley EA, Owora A, McCleary J, Anderson A. Sleepiness, Exertion Fatigue, Arousal, and 293 Vigilant Attention in Persons With Chronic Aphasia. 144 RESULTS American Journal of Speech-Language 294 Pathology. 2019;28(4):1491-508. 14 . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a p p y 298 13. Wang H, Liu X, Hu H, Wan F, Li T, Gao L, et al. Dynamic Reorganization of Functional 299 Connectivity Unmasks Fatigue Related Performance Declines in Simulated Driving. Ieee 300 Transactions on Neural Systems and Rehabilitation Engineering. 2020;28(8):1790-9. 301 14. Park S-J, Jeon E-C. A Study on the Safety and Management Plan of Aviation Maintenance 302 due to Human Error. Journal of the Korean Society of Manufacturing Process Engineers. 303 2020;19(10):68-73. 304 15. Lyu T, Song W, Du K. Human Factors Analysis of Air Traffic Safety Based on HFACS-BN 305 Model. Applied Sciences-Basel. 2019;9(23). 306 16. Budde D, Hinkelbein J, Boyd DD. Analysis of Air Taxi Accidents (20042018) and 307 Associated Human Factors by Aircraft Performance Class. Aerospace medicine and human 308 performance. 2021;92(5):294-302. 309 17. Christopher D. Wickens WSH, Justin G. Hollands, Simon Banbury. Engineering Psychology 310 and Human Performance. 5th Edition ed. New York: Routledge; 2021. 311 18. Chenfeng. Medical Multivariate Statistics: China Statistics Press; 2019. 312 19. Smets EM, Garssen B, Bonke B, De Haes JC. The Multidimensional Fatigue Inventory (MFI) 313 psychometric qualities of an instrument to assess fatigue. J Psychosom Res. 1995;39(3):315- 314 25. 315 20. Lin R, Yan C, Rui J. The reliability and validity of the Chinese version of the 316 Multidimensional Fatigue Scale for pilots. Chin J Convalescent Med 2016;25:332-4.(In 317 Chinese) 318 21. Buysse DJ, Reynolds CF, 3rd, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh Sleep 319 Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res. Connectivity Unmasks Fatigue Related Performance Declines in Simulated Driving. Ieee and Human Performance. 5th Edition ed. New York: Routledge; 2021. Chenfeng. Medical Multivariate Statistics: China Statistics Press; 2019. 15 . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint 144 RESULTS ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint 16 320 1989;28(2):193-213. 321 22. Jixian L, Maoqin T, Lei H, et al. Reliability and validity of the Pittsburgh sleep quality index. 322 Shandong Mental Health Center, Jinan 1996:103-7.(In Chinese) 323 23. Hart SG, Staveland LE. Development of NASA-TLX (Task Load Index): Results of 324 Empirical and Theoretical Research. In: Hancock PA, Meshkati N, editors. Advances in 325 Psychology. 52: North-Holland; 1988. p. 139-83. 326 24. Kelley AM, Feltman KA, Curry IP. A Survey of Fatigue in Army Aviators. Aerosp Med 327 Hum Perform. 2018;89(5):464-8. 328 25. Yamashita A, Rothlein D, Kucyi A, Valera EM, Esterman M. Brain state-based detection of 329 attentional fluctuations and their modulation. Neuroimage. 2021;236:118072. 330 26. Beason-Held LL, Kraut MA, Resnick SM. Longitudinal changes in aging brain function. 331 Neurobiol Aging. 2008;29(4):483-96. 332 27. Krause AJ, Simon EB, Mander BA, Greer SM, Saletin JM, Goldstein-Piekarski AN, et al. 333 The sleep-deprived human brain. Nat Rev Neurosci. 2017;18(7):404-18. 334 28. Vuorio A, Bor R. Self-Harm in Aviation Medicine-A Complex Challenge During a 335 Pandemic. Front Public Health. 2021;9:5. 336 29. Becker SP, Dvorsky MR, Breaux R, Cusick CN, Taylor KP, Langberg JM. Prospective 337 examination of adolescent sleep patterns and behaviors before and during COVID-19. Sleep. 338 2021;44(8):11. 339 30. Elanwar R, Hussein M, Magdy R, Eid RA, Yassien A, Abdelsattar AS, et al. Physical and 340 Mental Fatigue in Subjects Recovered from COVID-19 Infection: A Case-Control Study. 341 Neuropsychiatric disease and treatment. 2021;17:2063-71. 320 1989;28(2):193-213. 321 22. Jixian L, Maoqin T, Lei H, et al. Reliability and validity of the Pittsburgh sleep quality index. 322 Shandong Mental Health Center, Jinan 1996:103-7.(In Chinese) 323 23. Hart SG, Staveland LE. Development of NASA-TLX (Task Load Index): Results of 324 Empirical and Theoretical Research. In: Hancock PA, Meshkati N, editors. Advances in 325 Psychology. 52: North-Holland; 1988. p. 139-83. 326 24. Kelley AM, Feltman KA, Curry IP. A Survey of Fatigue in Army Aviators. Aerosp Med 327 Hum Perform. 2018;89(5):464-8. 328 25. Yamashita A, Rothlein D, Kucyi A, Valera EM, Esterman M. Brain state-based detection of 329 attentional fluctuations and their modulation. Neuroimage. . CC-BY 4.0 International license It is made available under a pe petu ty perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in was not certified by peer review) The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: 144 RESULTS 2021;236:118072. 330 26. Beason-Held LL, Kraut MA, Resnick SM. Longitudinal changes in aging brain function. 331 Neurobiol Aging. 2008;29(4):483-96. 332 27. Krause AJ, Simon EB, Mander BA, Greer SM, Saletin JM, Goldstein-Piekarski AN, et al. 333 The sleep-deprived human brain. Nat Rev Neurosci. 2017;18(7):404-18. 334 28. Vuorio A, Bor R. Self-Harm in Aviation Medicine-A Complex Challenge During a 335 Pandemic. Front Public Health. 2021;9:5. 336 29. Becker SP, Dvorsky MR, Breaux R, Cusick CN, Taylor KP, Langberg JM. Prospective 337 examination of adolescent sleep patterns and behaviors before and during COVID-19. Sleep. 338 2021;44(8):11. 339 30. Elanwar R, Hussein M, Magdy R, Eid RA, Yassien A, Abdelsattar AS, et al. Physical and 328 25. Yamashita A, Rothlein D, Kucyi A, Valera EM, Esterman M. Brain state-based detection of 329 attentional fluctuations and their modulation. Neuroimage. 2021;236:118072. 330 26. Beason-Held LL, Kraut MA, Resnick SM. Longitudinal changes in aging brain function. Neurobiol Aging. 2008;29(4):483-96. 331 Neurobiol Aging. 2008;29(4):483-96. 332 27. Krause AJ, Simon EB, Mander BA, Greer SM, Saletin JM, Goldstein-Piekarski AN, et al. 332 27. Krause AJ, Simon EB, Mander BA, Greer SM, Saletin JM, Goldstein-Piekarski AN, et al. 333 Th l d i d h b i N R N i 2017 18(7) 404 18 333 The sleep-deprived human brain. Nat Rev Neurosci. 2017;18(7):404-18. 333 The sleep-deprived human brain. Nat Rev Neurosci. 2017;18(7):404-18. 334 28. Vuorio A, Bor R. Self-Harm in Aviation Medicine-A Complex Challenge During a Pandemic. Front Public Health. 2021;9:5. 336 29. Becker SP, Dvorsky MR, Breaux R, Cusick CN, Taylor KP, Langberg JM. Prospective 337 examination of adolescent sleep patterns and behaviors before and during COVID-19. Sleep. 338 2021;44(8):11. Neuropsychiatric disease and treatment. 2021;17:2063-71. 16 CC-BY 4 0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for th this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a p p y . 2021. National Health Commission of the People's Republic of China2021. China NHCotPsRo. The Latest Situation of the COVID-19 as of 24:00 on November 25， 342 31. China NHCotPsRo. The Latest Situation of the COVID-19 as of 24:00 on November 25， 144 RESULTS CC-BY 4.0 International license It is made available under a p p y 17 17 . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a perpetuity. 346 Tables Table 1 Distribution of Civil Aviation Crews’ BPC (n=259) 347 Item PSQI (Mean ± SD) MFI (Mean ± SD) MWL (Mean ± SD) gender Male(n=134) Female(n=125) t value P value 8.08±4.04 7.55±3.50 1.13 0.26 53.99±11.04 58.39±8.30 -3.65 0.00 44.06±18.65 42.04±16.28 0.93 0.35 age/y 28.20±2.73 t value P value 20.38±7.35 44.63 0.00 -27.91±12.3 36.52 0.00 -14.8±18.75 -12.77 0.00 BMI 21.11±3.89 t value P value 13.28±5.42 39.42 0.00 -35±11.17 -50.43 0.00 -21.9±17.93 -19.73 0.00 position pilots(n=78) flight attendant(n=138) security(n=43) F value P value 7.60±4.08 7.96±3.87 7.79±3.00 0.23 0.80 54.06±11.7 56.96±9.1 57.09±9.31 2.35 0.10 44.18±18.18 42.88±16.73 41.75±19.17 0.28 0.75 driving experience/y 5(n=187) >5(n=72) t value P value 7.58±3.41 8.47±4.60 -1.50 0.14 56.33±10.22 55.56±9.61 0.56 0.58 42.29±17.32 45.14±18.06 -1.17 0.24 Table 2 Spearman's Correlation between BPC and COVID-19 348 348 Table 2 Spearman's Correlation between BPC and COVID-19 MWL PSQI MFI C1 C2 C3 C4 C5 C6 C7 MWL 1 .171 .203 .167 .236 .229 .210 .345 .327 .285 PSQI .171 1 .190 .255 .320 .355 .296 .393 .329 .354 MFI .203 .190 1 .149* .140* .140* 0.078 .272 .187 .198 C1 .167 .255** .149* 1 .797 .668 .749 .674 .692 .662 C2 .236 .320 .140* .797 1 .660 .752 .737 .743 .716 C3 .229 .355 .140* .668 .660 1 .817 .692 .648 .644 C4 .210 .296 0.078 .749 .752 .817 1 .698 .721 .703 C5 .345 .393 .272 .674 .737 .692 .698 1 .896 .850 C6 .327 .329 .187 .692 .743 .648 .721 .896 1 .867 C7 .285 .354 .198 .662 .716 .644 .703 .850 .867 1 349 Note: . Correlation is significant at the 0.01 level (2-tailed); . level (2-tailed). . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint 349 Note: . Correlation is significant at the 0.01 level (2-tailed); . Correlation is significant at the 0.05 350 level (2-tailed). 362 C4=increased workload of personal protective equipment; C5=sleep quality; C6=mental fatigue; 359 Abbreviations: BPC=brain performance capacity; MWL=mental workload; PSQI= Pittsburgh Sleep 360 Quality Index; MFI= Multidimensional Fatigue Inventory; C1=self-fear of infection; C2=fear of 361 ineffective prevention and control; C3=physical unwellness caused by regular nucleic and testing; . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint 144 RESULTS Correlation is significant at the 0.05 350 level (2 tailed) 18 . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint CC BY 4 0 I i l li I i d il bl d perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in was not certified by peer review) The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: . CC-BY 4.0 International license It is made available under a pe petu ty . CC-BY 4.0 International license It is made available under a p p y . CC-BY 4.0 International license It is made available under a p p y 351 Abbreviations: MWL=mental workload; PSQI= Pittsburgh Sleep Quality Index; MFI= Multidimensional 352 Fatigue Inventory ; C1=self-fear of infection; C2=fear of ineffective prevention and control; C3=physical 353 unwellness caused by regular nucleic and testing; C4=increased workload of personal protective 354 equipment; C5=sleep quality; C6=mental fatigue; C7=physical fatigue. 355 356 Table 3 Results of CCA between BPC and COVID-19 　 Correlation Eigenvalue Wilks Statistic F u P 1 0.517 0.364 0.691 4.681 21.000 0.000 2 0.206 0.044 0.943 1.247 12.000 0.247 3 0.124 0.016 0.985 0.789 5.000 0.558 357 358 Table 4 Standardized Canonical Correlation Coefficients of BPC set and COVID-19 set 　Variable set Variable Standardized Canonical Correlation Coefficients Proportion Variance of Explained MWL -0.485 PSQI -0.610 BPC MFI -0.312 0.483 C1 0.088 C2 -0.114 C3 -0.511 C4 0.464 C5 -0.872 C6 0.086 COVID-19 C7 -0.132 0.580 359 Abbreviations: BPC=brain performance capacity; MWL=mental workload; PSQI= Pittsburgh Sleep 360 Quality Index; MFI= Multidimensional Fatigue Inventory; C1=self-fear of infection; C2=fear of Table 3 Results of CCA between BPC and COVID-19 19 . CC-BY 4.0 International license It is made available under a perpetuity. . CC-BY 4.0 International license It is made available under a p p y 144 RESULTS is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 30, 2022. ; https://doi.org/10.1101/2022.01.29.22270064 doi: medRxiv preprint 363 C7=physical fatigue. 364 365 Figure Caption 366 Figure 1: CCA of BPC and COVID-19 367 Abbreviations: CCA=canonical correlation analysis; BPC=brain performance capacity; MWL=mental 368 workload; PSQI= Pittsburgh Sleep Quality Index; MFI= Multidimensional Fatigue Inventory; C1=self- 369 fear of infection; C2=fear of ineffective prevention and control; C3=physical unwellness caused by 370 regular nucleic acid testing; C4=increased workload of personal protective equipment; C5=sleep quality; 371 C6=mental fatigue; C7=physical fatigue. 367 Abbreviations: CCA=canonical correlation analysis; BPC=brain performance capacity; MWL=mental 368 workload; PSQI= Pittsburgh Sleep Quality Index; MFI= Multidimensional Fatigue Inventory; C1=self- 369 fear of infection; C2=fear of ineffective prevention and control; C3=physical unwellness caused by 370 regular nucleic acid testing; C4=increased workload of personal protective equipment; C5=sleep quality; 371 C6=mental fatigue; C7=physical fatigue 20
https://openalex.org/W4310495751	https://www.businessperspectives.org/images/pdf/applications/publishing/templates/article/assets/17403/PPM_2022_04_Strilets.pdf	English	null	State support for the digitalization of SMEs in European countries	Problems and perspectives in management/Problems & perspectives in management	2,022	cc-by	11,185	“State support for the digitalization of SMEs in European countries AUTHORS Viktoriia Strilets Serhiy Frolov Viktoriia Datsenko Oksana Tymoshenko Maksym Yatsko ARTICLE INFO Viktoriia Strilets, Serhiy Frolov, Viktoriia Datsenko, Oksana Tymoshenko and Maksym Yatsko (2022). State support for the digitalization of SMEs in European countries. Problems and Perspectives in Management, 20(4), 290-305. doi:10.21511/ppm.20(4).2022.22 DOI http://dx.doi.org/10.21511/ppm.20(4).2022.22 RELEASED ON Thursday, 01 December 2022 RECEIVED ON Monday, 22 August 2022 ACCEPTED ON Friday, 18 November 2022 LICENSE This work is licensed under a Creative Commons Attribution 4.0 International License JOURNAL "Problems and Perspectives in Management" ISSN PRINT 1727-7051 ISSN ONLINE 1810-5467 PUBLISHER LLC “Consulting Publishing Company “Business Perspectives” FOUNDER LLC “Consulting Publishing Company “Business Perspectives” NUMBER OF REFERENCES 49 NUMBER OF FIGURES 4 NUMBER OF TABLES 10 © The author(s) 2022. This publication is an open access article. © The author(s) 2022. This publication is an open access article. © The author(s) 2022. This publication is an open access article. INTRODUCTION Serhiy Frolov, Doctor of Economics, Professor, Department of Finance, Banking and Insurance, Sumy National Agrarian University, Ukraine. (Corresponding author) SMEs are an essential element of countries’ national economies in the world, contributing significantly to their development (49.7% – in the total number of employees, 60% – in the created added value, and 50% Viktoriia Datsenko, Ph.D. in Economics, Associate Professor, Dean of the Faculty of Economics, Business and International Relations, University of Customs and Finance, Ukraine. – in the volume of exports). However, the recent changes in the busi- ness environment (e.g., the transition to Industry 4.0, the economic downturn due to COVID-19, and the Russian war against Ukraine) caused many challenges for the business models of SMEs. Thus, since the beginning of the COVID-19 pandemic, every fourth SME entity in EU countries is on the verge of bankruptcy; 26% of young and 18% of experienced SMEs went bankrupt during the year. The situation is worsening due to the shock from inflation and rising energy prices caused by Russian aggression. The need to adapt to changing condi- tions and maintain competitiveness stimulates SMEs to develop alter- native business forms and diversify their offers and markets. – in the volume of exports). However, the recent changes in the busi- ness environment (e.g., the transition to Industry 4.0, the economic downturn due to COVID-19, and the Russian war against Ukraine) caused many challenges for the business models of SMEs. Thus, since the beginning of the COVID-19 pandemic, every fourth SME entity in EU countries is on the verge of bankruptcy; 26% of young and 18% of experienced SMEs went bankrupt during the year. The situation is worsening due to the shock from inflation and rising energy prices caused by Russian aggression. The need to adapt to changing condi- tions and maintain competitiveness stimulates SMEs to develop alter- native business forms and diversify their offers and markets. Oksana Tymoshenko, Ph.D. in Economics, Associate Professor, Department of Economics, Poltava University of Economics and Trade, Ukraine. Oksana Tymoshenko, Ph.D. in Economics, Associate Professor, Department of Economics, Poltava University of Economics and Trade, Ukraine. Maksym Yatsko, Ph.D. in Economics, Associate Professor, Department of Accounting and Auditing, State Higher Educational Institution “Uzhgorod National University”, Ukraine. Business digitalization is considered a key to achieving success in the long term. businessperspectives.org businessperspectives.org businessperspectives.org Problems and Perspectives in Management, Volume 20, Issue 4, 2022 Viktoriia Strilets (Ukraine), Serhiy Frolov (Ukraine), Viktoriia Datsenko (Ukraine), Oksana Tymoshenko (Ukraine), Maksym Yatsko (Ukraine) Keywords JEL Classification М13, М15, О33 JEL Classification М13, М15, О33 Abstract COVID-19 and Russia’s war against Ukraine have created many challenges for the business models of SMEs, stimulating them to develop alternative forms of business. Among the key alternatives, one should single out digitization. However, imperfections of Ukraine’s state support system for digital transformation necessitate adapting the positive practice of leading countries with excellent business digitalization. Therefore, this paper structures the practical experience of state support of European countries in the digitalization of SMEs. The study conducts a cluster analysis to identify countries whose experience in state support for business digitalization is the best for adoption. Clustering employed the criterion of using key digitization instruments: internet of things, integration of business processes into electronic processes, digital integration with customers/suppliers, cloud services, big data analysis, and artificial intelligence. Thus, 4 clusters were identified (the first cluster comprises 16 countries, the second – 2, the third – 5, and the fourth – 7). These findings contributed to the selection of some key measures. They included: expanding the financial capabilities of SMEs to adopt digital technologies in their processes; formation of digital hard skills among entrepreneurs and increasing the personnel potential of SME digitization; creation of a favorable business environment to stimulate the digital activity of SMEs; and popu- larization of business digitalization strategy as a component of digital transformation strategy in society. Received on: 22nd of August, 2022 Accepted on: 18th of November, 2022 Published on: 1st of December, 2022 Received on: 22nd of August, 2022 Accepted on: 18th of November, 2022 Published on: 1st of December, 2022 Received on: 22nd of August, 2022 Accepted on: 18th of November, 2022 Published on: 1st of December, 2022 © Viktoriia Strilets, Serhiy Frolov, Viktoriia Datsenko, Oksana Tymoshenko, Maksym Yatsko, 2022 Viktoriia Strilets, Doctor of Economics, Associate Professor, Department of International Economics and International Economic Relations, Poltava University of Economics and Trade, Ukraine. Keywords digital transformation, state support, small and medium enterprises, artificial intelligence digital transformation, state support, small and medium enterprises, artificial intelligence BUSINESS PERSPECTIVES www.businessperspectives.org LLC “СPС “Business Perspectives” Hryhorii Skovoroda lane, 10, Sumy, 40022, Ukraine This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International license, which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited. Conflict of interest statement: Author(s) reported no conflict of interest digital transformation, state support, small and medium enterprises, artificial intelligence Viktoriia Strilets (Ukraine), Serhiy Frolov (Ukraine), Viktoriia Datsenko (Ukraine), Oksana Tymoshenko (Ukraine), Maksym Yatsko (Ukraine) Received on: 22nd of August, 2022 Accepted on: 18th of November, 2022 Published on: 1st of December, 2022 This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International license, which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited. 1. LITERATURE REVIEW al., 2022; Kuznyetsova et al., 2022), the insurance market (Polinkevych et al., 2021), and the chatbots (Jansom et al., 2022) became significantly more developed during the COVID-19 pandemic. Digitalization is fundamental for economic growth and job creation in modern conditions. Therefore, its development occupies a leading place in the Strategy “Europe 2021–2027”. Furthermore, digi- talization is a consistent unification of all sectors of the economy and adaptation of subjects to the new conditions of the digital economy (Parviainen et al., 2022). Generally, business digitalization describes how firms can improve their operations and gener- ate value by transforming their business models (Samper et al., 2022). Parida et al. (2019) showed that digital technologies help innovate organiza- tional models and offer opportunities for value creation in industrial ecosystems. Yu et al. (2022) noted that business digitalization models re- fer to continuous improvement. Here, informa- tion, computing, communication, and connec- tion technologies are vital for an organization’s restructuring. Liu et al. (2011) described digitalization as integra- tion of digital technologies and business process- es into the digital economy. Bethlendi and Szőcs (2022) considered it a transformation of products, processes, and organizational aspects. Matt et al. (2015) concluded that it involves transformations in organizational methods, roles, and offers due to the implementation of digital technologies. Finally, Vial (2019) noted that it is improving processes by transforming its properties using information, computing, and communication technologies. As a result, digitalization increases countries’ eco- nomic security and ensures financial systems’ sta- bility (Shkolnyk et al., 2022). Business digitalization is multifaceted, combin- ing various tools with different goals and forcing the reconfiguration of strategic assets. Only some SMEs can adopt such transformations. The small- er the company is, the less likely it will adopt digi- tal tools and the more likely it will limit the use of elementary services. Business digitalization is a critical component of digitalization in society. Currently, digital technol- ogies have penetrated almost all aspects of human life and opened new opportunities for businesses (Simatupang & Widjaja, 2012). Digitalization not only helps to overcome the information asymme- try between businesses and investors (Pedchenko et al., 2018) but also stimulates the development of SMEs in other business aspects. For example, the digitalization of banking services (Pham et Business digitalization involves transformations of business models, services, goods, and struc- tures (Hess et al., 2016; Heavin & Power, 2018). INTRODUCTION It involves implementing digital technologies that can im- prove the ability of SMEs to collect, combine, process, and use busi- ness data. The digitization of SME business processes can help retain and increase the number of consumers, increase competitiveness, and save on operating and transaction costs. Conflict of interest statement: Author(s) reported no conflict of interest 290 http://dx.doi.org/10.21511/ppm.20(4).2022.22 http://dx.doi.org/10.21511/ppm.20(4).2022.22 Problems and Perspectives in Management, Volume 20, Issue 4, 2022 However, the existence of both internal problems in the development of digital processes of SMEs and those related to the imperfection of the system of state support for the digital transformation forces to adopt the best positive practice of EU countries. SMEs face significant digital innovation barriers be- cause of their limited resources and opportunities compared to big enterprises. Working with digital technologies is significantly different from traditional ways of doing business. This requires moderniza- tion of the business model and additional state assistance. Teteriatnyk (2018), Strilets (2019), and Okoi et al. (2022) investigated the role of government in improving the productivity and growth of small businesses. Moreover, it was found that the government’s involvement in supporting the digitalization of SMEs is necessary. Therefore, the issue of structuring the practical experience of state support for business digitalization to identify those leading countries whose experience in forming a clear legal framework and adequate state support is possible for adopting in outsider countries is relevant and necessary. Therefore, the issue of structuring the practical experience of state support for business digitalization to identify those leading countries whose experience in forming a clear legal framework and adequate state support is possible for adopting in outsider countries is relevant and necessary. http://dx.doi.org/10.21511/ppm.20(4).2022.22 1. LITERATURE REVIEW As a result, it accelerates the company’s economic de- velopment, increases its productivity and perfor- mance, minimizes transaction costs, and opens ways to new markets (Williams, 2021). According to Natorina (2017), digital transformations are crucial to rational management decisions, includ- ing digitalizing operational processes, virtualiz- 291 Problems and Perspectives in Management, Volume 20, Issue 4, 2022 volume of sales through digital channels. It is ar- gued that business activities have become more favorable during the pandemic through digital channels. Therefore, governments should provide support programs (tax incentives and subsidies) for SMEs. ing employee effectiveness, and transactional pro- ductivity systems. As for the impact of digitaliza- tion of business on productivity, dynamic manag- ers can harness the potential of new methods and encourage their adoption (Li et al., 2018). Digitalization of business is also a driving force that contributes to its advancement. Its main ad- vantages include saving time, increasing business productivity, automation of production and other internal company processes, and improving com- munications. Moreover, it offers cross-selling/up- selling opportunities, reaching a new level of cus- tomer service and encouraging them to purchase more products, introducing competitive opportu- nities due to improved customer experience and overall optimization. Finally, it contributes to in- creasing GDP per capita and reducing CO2 emis- sions (Melnyk et al., 2021). According to Reim et al. (2022), SMEs may overcome obstacles and obtain various benefits (e.g., hiring highly-qualified staff) via digital technologies adoption. Sestino et al. (2020) explored big data and the in- ternet of things as tools for business digitalization strategies. The study asserted that these technolo- gies can increase the competitiveness of compa- nies. Furthermore, IoT offers possibilities to ana- lyze consumer attitudes, behavior, and intentions. Lestari et al. (2021) highlighted the role of e-com- merce for small businesses in improving their operations. The results showed that non-imple- mentation of e-commerce resulted in lower reve- nues for small businesses than those that adopted e-commerce. They concluded that e-commerce could benefit both governments and small busi- nesses in times of crisis, e.g., COVID-19. Business digitalization across industries has en- abled the internet of things (IoT), cloud comput- ing, big data analytics, and artificial intelligence. By employing these tools, firms can enhance their performance (Lia et al., 2022; Polyakov et al., 2021). Thus, business digitalization involves cyber- physical systems, 3D printing, robotics, big data, and the internet of things. http://dx.doi.org/10.21511/ppm.20(4).2022.22 1. LITERATURE REVIEW Moreover, to achieve success, mobile tools and applications, social me- dia, analytical tools and applications, platforms for communication and data sharing, as well as collaboration applications, are necessary. Another essential tool of business digitalization – digital platforms – was thoroughly researched by Oro (2018), Li et al. (2021), and Akhter et al. (2022). They found that digital platforms positive- ly impact SMEs’ business innovations and recon- figuration capabilities. An intelligent digital plat- form that brings together innovative stakehold- ers (e.g., government, academia, companies, and community) enhances cooperation between vari- ous actors interested in sharing information and promoting and selling the results of SME activity (Maquera et al., 2022). According to Cenamor et al. (2019), a digital platform supports information sharing and enables companies to unify and up- date their data. To digitize their offers, modern companies should adopt digital tools. Ting et al. (2020) noted that artificial intelligence (AI), 5G wireless networks, machine learning (ML), and deep learning (DL), as well as blockchain systems, contributed to a unique and beneficial environment in all sectors of the economy. However, more than applying digital technologies is needed to stimulate business digitalization. The development of all business digitalization tools deserves the government’s close attention and determines the need for financial, informational, advisory, and other types of support. Small and medium-sized businesses should possess the nec- essary digital strategies, capabilities, and culture and talent growth initiatives to achieve digital transformation successfully. Many studies are fo- cused on the elements of a successful digital busi- ness strategy and the role of government support Almost a third of companies invest up to 50 mil- lion US dollars in the automation and digitization of business processes (Pereira et al., 2022). Thus, Maglakelidze and Erkomaishvili (2021) found that one of the alternative tools to overcome the chal- lenges of the pandemic, which has given impetus to changes in business behavior, is to increase the 292 http://dx.doi.org/10.21511/ppm.20(4).2022.22 Problems and Perspectives in Management, Volume 20, Issue 4, 2022 Problems and Perspectives in Management, Volume 20, Issue 4, 2022 2. METHODOLOGY • Stage 6: Building a dendogram to form the number of clusters; • Stage 7: Determining the fullness of clusters; • Stage 8: Providing key characteristics of clusters; • Stage 9: Generalizing conclusions. Descriptive characteristics of cluster analysis are given in Appendix B. Descriptive characteristics of cluster analysis are given in Appendix B. http://dx.doi.org/10.21511/ppm.20(4).2022.22 2. METHODOLOGY (Matt et al., 2015; Fischer et al., 2020). SMEs need a stakeholder collaboration strategy to adopt and develop digitization practices (Nudurupati et al., 2022). Due to the significant investments needed, it could be challenging for SMEs to adopt digital technologies. Thus, governmental support is cru- cial here (Yuan et al., 2021). The paper uses cluster analysis (for grouping coun- tries into clusters according to the level of appli- cation of business digitization tools by SMEs). Clustering SMEs by the level of business digital- ization is needed to identify countries with a fa- vorable environment for the digital transforma- tion of enterprises since these countries are char- acterized by adequate state support. The proposed sequence of cluster analysis involves the following stages: Governments support SMEs in adopting various digital platforms. However, the spread of such platforms depends on the structural and politi- cal aspects of countries and regions. For example, OECD (2021a, 2021b), EC (2022), and GEM (2022) showed that the country’s socio-economic and de- mographic features, population awareness of digi- tal tools, concentration of platforms across sectors, and its structural aspects (freedom and the rule of law, product market regulation and digital ser- vices) affect the use of platforms in a country. • Stage 1: Analyzing analytical and statistical data and forming of a database for analysis; • Stage 2: Forming a sample of the leading indi- cators for evaluating the digitization of SMEs; • Stage 3: Studying modern trends of changes in the leading indicators (Appendix A); In response to COVID-19, countries introduced various initiatives to promote SMEs’ digitiza- tion, including online platforms. However, no conceptual approach to studying state support for the digitalization of SMEs has been elaborat- ed. In particular, there is a research gap in terms of the experience of using essential tools: online platforms, the internet of things, the integration of business processes into electronic processes; artificial intelligence; digital integration with customers/suppliers; 3D printing and robotics; big data analysis; cloud services, and others. In addition, there are scarce studies on how the improvement of sustainable competitiveness can describe the development of digitalization (Petkovski et al., 2022). Although the literature highlights numerous tools of digital innova- tions and their positive impact on the develop- ment of SMEs, the interdependence between the competencies of governments and digital inno- vations and the positive practices of using such dependence have been studied in a rather frag- mentary manner. 3. RESULTS Digitalization plays a significant role in increas- ing the sustainability of SMEs. It offers opportu- nities to restructure industries and enhance busi- ness competition. Companies adopt new business models and create value propositions via digital technologies. Therefore, SMEs enhance their per- formance, i.e., increase effectiveness and reduce operational costs, increase revenues, obtain more practical knowledge and innovations, and im- The level of business digitization in countries is di- rectly dependent on the favorable business envi- ronment created by the government for the digital transformation of society. Therefore, the purpose of this study is to structure the practical experi- ence of state support in European countries in the field of digitalization of SMEs. 293 http://dx.doi.org/10.21511/ppm.20(4).2022.22 Problems and Perspectives in Management, Volume 20, Issue 4, 2022 Source: EC (2022), OECD (2021a), GEM (2022). Source: EC (2022), OECD (2021a), GEM (2022). Figure 1. The use of digital technologies by business entities according to size in EU countries in 2020, % 0 20 40 60 80 100 B2G Interactions Enterprise Resource Planning Social Media Customer Relations Management Electronic Voting Cloud Technologies E-Commerce High Speed Broadband Connection Supply Management Big data Electronic Booking and Ordering Small Business Medium Business Big Business Customer Relations Management . The use of digital technologies by business entities according to size in EU countries in 2020, % prove customer engagement. Nevertheless, SMEs’ digitization level is still deficient (Figure 1). prove customer engagement. Nevertheless, SMEs’ digitization level is still deficient (Figure 1). Finland, Denmark, and Belgium. In these coun- tries, the level of integration of SMEs in electron- ic business is more than 45%. On the other hand, the level of integration in Latvia, Malta, Austria, Luxembourg, Lithuania, and France is somewhat lower but still relatively high. According to OECP calculations, other things be- ing equal, the probability of closing an SME was approximately 10% higher than that of enterprises with a number of employees from 50 to 249. At the same time, SMEs in the field of wholesale and retail trade and hotel and catering establishments are particularly vulnerable to the consequences of the pandemic (OECD, 2021a). This evidence is con- sistent with the statement about the importance and necessity of government support for SMEs, especially regarding the digitization of their busi- ness processes. Therefore, the government should promote the competitiveness and productivity of SMEs, in particular, by encouraging the digitiza- tion of SME business processes. 3. RESULTS The European Union pays considerable attention to implementing digital technologies in busi- ness processes, defined primarily in the strategy “Europe 2021-2027”. However, at the moment, the EU member states have an unequal use of digital tools. Therefore, in order to determine the most successful comprehensive approaches of the gov- ernments to stimulate business digitalization, it is proposed to cluster countries according to the level of implementation of all digitalization com- ponents. Moreover, it is vital to identify those “star countries” whose level of SME digitalization is the highest, and, therefore, their practices are the most successful (Table 1). In recent years, the digitalization of business has continued at a rapid pace. Across industries, busi- ness entities rapidly equip their employees with digital tools. However, smaller companies could be more successful in this sphere. There are also differences in implementing electronic business among countries (Figure 2). The descriptive characteristics of the cluster anal- ysis (Appendix B) made it possible to determine a sufficient validity of clusters: there are 16 coun- tries in cluster 1, 2 – in cluster 2, 5 – in cluster 3, and 7 – in cluster 4. Thus, studies on the integration of SMEs in digi- tal technologies in foreign countries have shown that these processes take place most effectively in Therefore, the first cluster has relatively low (com- pared to other clusters) indicators of the use of 294 http://dx.doi.org/10.21511/ppm.20(4).2022.22 http://dx.doi.org/10.21511/ppm.20(4).2022.22 Problems and Perspectives in Management, Volume 20, Issue 4, 2022 Source: EC (2022), OECD (2021a), GEM (2022). 3. RESULTS 20 26 36 35 36 35 31 20 27 35 42 34 33 33 37 37 39 36 36 38 34 22 21 25 28 26 36 35 23 28 33 55 36 28 36 20 25 30 35 40 45 50 55 2012 2013 2014 2015 2017 2019 2021 Italy Cyprus Greece Czech Republic Germany Group with a sufficient level of integration in 2021 (31-36%) 295 8 12 14 14 13 13 19 19 19 26 24 22 22 21 18 26 26 26 24 24 22 8 15 16 21 27 24 22 17 20 22 24 27 27 23 0 20 40 60 80 100 120 2012 2013 2014 2015 2017 2019 2021 Hungary Bulgaria Croatia Estonia Ireland 20 26 36 35 36 35 31 20 27 35 42 34 33 33 37 37 39 36 36 38 34 22 21 25 28 26 36 35 23 28 33 55 36 28 36 20 25 30 35 40 45 50 55 2012 2013 2014 2015 2017 2019 2021 Italy Cyprus Greece Czech Republic Germany 10 8 9 15 24 31 38 21 35 37 37 39 39 38 23 24 29 28 27 31 38 22 39 33 39 46 47 43 24 30 44 40 39 42 44 32 32 34 38 37 47 44 6 11 16 21 26 31 36 41 46 2012 2013 2014 2015 2017 2019 2021 Latvia Luxembourg Malta Lithuania Austria France 31 36 37 35 37 42 47 32 32 41 45 39 48 49 32 40 46 49 53 51 56 30 35 40 45 50 55 60 2012 2013 2014 2015 2017 2019 2021 Finland Denmark Belgium Group with a low level of integration in 2021 (18-23%) Group with a sufficient level of integration in 2021 (31-36%) Group with an average level of integration in 2021 (38-44%) Group with a high level of integration in 2021 (>45%) Figure 2. 3. RESULTS Trends in the integration of SMEs into e-business in some countries in 2012-2021, % in the structure of SMEs 8 12 14 14 13 13 19 19 19 26 24 22 22 21 18 26 26 26 24 24 22 8 15 16 21 27 24 22 17 20 22 24 27 27 23 0 20 40 60 80 100 120 2012 2013 2014 2015 2017 2019 2021 Hungary Bulgaria Croatia Estonia Ireland Group with a low level of integration in 2021 (18-23%) Group with a sufficient level of integration in 2021 (31-36%) 10 8 9 15 24 31 38 21 35 37 37 39 39 38 23 24 29 28 27 31 38 22 39 33 39 46 47 43 24 30 44 40 39 42 44 32 32 34 38 37 47 44 6 11 16 21 26 31 36 41 46 2012 2013 2014 2015 2017 2019 2021 Latvia Luxembourg Malta Lithuania Austria France Group with an average level of integration in 2021 (38-44% 31 36 37 35 37 42 47 32 32 41 45 39 48 49 32 40 46 49 53 51 56 30 35 40 45 50 55 60 2012 2013 2014 2015 2017 2019 2021 Finland Denmark Belgium %) Group with a high level of integration in 2021 (>45%) Group with an average level of integration in 2021 (38-44%) Figure 2. Trends in the integration of SMEs into e-business in some countries in 2012-2021, % in the structure of SMEs Figure 2. Trends in the integration of SMEs into e-business in some countries in 2012-2021, % in the structure of SMEs 295 http://dx.doi.org/10.21511/ppm.20(4).2022.22 Problems and Perspectives in Management, Volume 20, Issue 4, 2022 Table 1. Stages of cluster analysis Stage 1: Analyzing analytical and statistical data and forming a database for analysisi Database of the European Commission; Organization for economic development and cooperation (reports on the act digitalization of business; Global monitoring of entrepreneurship (GEM reports).iii g g p gi g gii atistical data on the use of digital technologies and opportunities (internet of things D1, integration of business processes into ectronic processes D2; artificial intelligence D3; digital integration with customers/suppliers D4; 3D printing and robotics D5; big d alysis D6; cloud services D7). Stage 3: Studying current trends in changes of the main indicators Formation of a database by country. Stage 3: Studying current trends in changes of the main indicators Formation of a database by country. digital technologies, particularly in artificial in- telligence, digital integration with customers and suppliers, and cloud services. However, it has an adequate level of big data analysis and the use of 3D i i d b i Thi l i l d Germany, Portugal, Slovakia, France, Croatia, and the Czech Republic. The second cluster showed high use of internet of hi d i i f b i i Stage 3: Studying current trends in changes of the main indicators Formation of a database by country. Stage 4: Selecting software for cluster analysis IPSS Statistics, Deductor Studio. Stage 5: Selecting cluster analysis criteria Number of observations 31, K-means analysis, intergroup connection method, Euclidean distance, error 0.05 Stage 6: Building a dendogram to form the number of clusters Dendogram of countries’ distribution according to the level of digitalization Result: allocation of 4 clusters; cluster 1 – 16 countries; cluster 2 – 3; cluster 3 – 5; cluster 4 – 7 Stage 7: Determining the clusters Cluster 1 Austria, Belgium, Ireland, Spain, Cyprus, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Germany, Portugal, Slovakia, France, Croatia, Czech Republic Cluster 2 Greece, Italy, Finland Cluster 3 Denmark, Estonia, Norway, Slovenia, Sweden Cluster 4 Bulgaria, Bosnia and Herzegovina, Poland, Romania, Serbia, Turkey, Hungary Stage 8: Providing key characteristics of clusters Stage 9: Generalizing conclusions Stage 4: Selecting software for cluster analysis IPSS Statistics, Deductor Studio. Stage 1: Analyzing analytical and statistical data and forming a database for analysisi In addition, various support initiatives stimulate SMEs to adopt on- line platforms, although the prevalence of online platforms differs due to structural and procedure causes (OECD, 2021a, 2019; EC, 2022). Among the latest trends, it is also worth highlighting fi- nancial support programs for the digitalization of SMEs (Norway, Switzerland, Greece, etc.) and ad- visory and informational support provided by the governments. The third cluster is a leader in cloud services, big data analysis, and artificial intelligence. It suffi- ciently uses the internet of things and has an av- erage level of digital integration with customers/ suppliers. This cluster includes Denmark, Estonia, Norway, Slovenia, and Sweden. The last (fourth) cluster is an outsider and has the lowest indicators for all the studied indicators, which indicates the lowest level of digitalization of SMEs in the countries of this group. This clus- ter includes Bulgaria, Bosnia and Herzegovina, Poland, Romania, Serbia, Turkey, and Hungary. Table 2 shows that SMEs need appropriate policies and support programs from the government to successfully implement digital transformations. It is crucial to analyze the role of governmental sup- port in increasing the harmonization of govern- ment policies and programs with SMEs’ demands. Therefore, the development of a conceptual mod- el of state support for the digitization of SMEs is necessary. This model should include main barri- ers and challenges of SMEs digitization, govern- The results showed that the countries of the sec- ond and third clusters are examples to follow. In addition, it is appropriate to study the example of the first cluster regarding the use of big data anal- ysis, 3D printing, and robotics. Thus, Table 2 sum- marizes the main measures implemented by the state bodies of the leading countries to stimulate the digitization of business. Table 2. State support measures for business digitalization Table 2. State support measures for business digitalization Source: OECD (2021a, 2019), EC (2022), GEM (2022). Country State support measures for business digitalization Estonia Implementation of the “Digitalization Grant” program. Providing financial support for digital technologies and robots and automation in the manufacturing and mining industries. France A public online platform for the cooperation of SMEs wishing to digitize their business with public and private consultants throughout the country. The ability to implement functions: creating a digital strategy, increasing the presence on the Internet, developing the clientele, selling on the internet, etc. Stage 1: Analyzing analytical and statistical data and forming a database for analysisi Stage 5: Selecting cluster analysis criteria Number of observations 31, K-means analysis, intergroup connection method, Euclidean distance, error 0.05 Stage 6: Building a dendogram to form the number of clusters Dendogram of countries’ distribution according to the level of digitalization Result: allocation of 4 clusters; cluster 1 – 16 countries; cluster 2 – 3; cluster 3 – 5; cluster 4 – 7 Result: allocation of 4 clusters; cluster 1 – 16 countries; cluster 2 – 3; cluster 3 – 5; cluster 4 – 7 Stage 6: Building a dendogram to form the number of clusters Dendogram of countries’ distribution according to the level of digitalization Dendogram of countries’ distribution according to the level of digitalization Stage 7: Determining the clusters Cluster 1 Austria, Belgium, Ireland, Spain, Cyprus, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Germany, Portugal, Slovakia, France, Croatia, Czech Republic Cluster 2 Greece, Italy, Finland Cluster 3 Denmark, Estonia, Norway, Slovenia, Sweden Cluster 4 Bulgaria, Bosnia and Herzegovina, Poland, Romania, Serbia, Turkey, Hungary Stage 8: Providing key characteristics of clusters Stage 9: Generalizing conclusions Stage 7: Determining the clusters Cluster 1 Austria, Belgium, Ireland, Spain, Cyprus, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Germany, Portugal, Slovakia, France, Croatia, Czech Republic Cluster 2 Greece, Italy, Finland Cluster 3 Denmark, Estonia, Norway, Slovenia, Sweden Cluster 4 Bulgaria, Bosnia and Herzegovina, Poland, Romania, Serbia, Turkey, Hungary Stage 8: Providing key characteristics of clusters Stage 9: Generalizing conclusions Germany, Portugal, Slovakia, France, Croatia, and the Czech Republic. digital technologies, particularly in artificial in- telligence, digital integration with customers and suppliers, and cloud services. However, it has an adequate level of big data analysis and the use of 3D printing and robotics. This cluster includes Austria, Belgium, Ireland, Spain, Cyprus, Latvia, Lithuania, Luxembourg, Malta, Netherlands, The second cluster showed high use of internet of things and integration of business processes into electronic business; the highest level of digital in- tegration with customers/suppliers; and the popu- 296 http://dx.doi.org/10.21511/ppm.20(4).2022.22 http://dx.doi.org/10.21511/ppm.20(4).2022.22 Problems and Perspectives in Management, Volume 20, Issue 4, 2022 larization of cloud services. This cluster includes Greece, Italy, and Finland. Analysis of state support for digitization found that authorities have started to implement aware- ness-raising activities and policy actions regard- ing online platforms (Australia, Denmark, France, and the United Kingdom). Stage 1: Analyzing analytical and statistical data and forming a database for analysisi Consultants provide information on available financing options; information channels to offer data on support initiatives from the government and private sector actors; a daily radio show that discussed future digital trends. Denmark The government’s program “SMEs: Digital” is part of the National Government’s Strategy for Digital Growth; the initiative “SME: Digital Features the E-Commerce Centre” is designed to help small and medium-sized enterprises with online sales, includes grants for private consultations to clarify and develop the potential of the company in the field of e-commerce; E-Commerce Center workshops where online selling challenges are discussed with industry experts. Sweden The Digi-Lift initiative promoted digital transformations with an emphasis on the industrial sector. It is part of the smart industry strategy. Italy Due to the pandemic, Italian regions offer financial support to companies and self-employed workers to work from home. Funding is provided to cover the cost of training and purchasing digital tools/subscriptions to digital platforms for remote work. Slovenia “Digital vouchers” program support micro-SMEs with up to 10000 euros to improve digital competences, prepare a digital strategy, ensure cyber security, and implement digital marketing. Ireland Implementation of Digital Trading Online Voucher. Micro-enterprises can receive a voucher for 2500 euros to develop online sales and access to free online training. Luxembourg A digital technology fund with co-financing from public and private actors and focusing on venture investments in startups working in the field of ICT and related industries, including cyber security, big data, digital health, next- generation media and communication networks, digital training, satellite telecommunications, and services. Portugal Opendays i4.0 is an information campaign to popularize the need for digital transformation among SMEs. Germany Go-digital is a program to support SMEs (up to 100 employees and a total balance sheet of up to 20 million euros) in IT security, digitization of business processes, and digital market development, thanks to the expertise provided by consulting firms. State support measures for business digitalization tonia Implementation of the “Digitalization Grant” program. Providing financial support for digital technologies and robo and automation in the manufacturing and mining industries. A public online platform for the cooperation of SMEs wishing to digitize their business with public and private consultants throughout the country. The ability to implement functions: creating a digital strategy, increasing the presence on the Internet, developing the clientele, selling on the internet, etc. http://dx.doi.org/10.21511/ppm.20(4).2022.22 Stage 1: Analyzing analytical and statistical data and forming a database for analysisi Support drivers • development of digital technologies; digital strategies; • culture and development of talented digital specialists; • digital shifts in the industry; change in the competitive environment; • changes in normative legal acts; • opportunities to increase • business efficiency. Financial support Consulting, information and non- financial support A set of support measures (financial and non- financial) with an industry approach or technology focus. With an industry approach or focus on digital technologies. State support for digitization of SMEs • subsidies; • provision of grants; • reduction of taxation. Forms of support Types of support • digital training and seminars to improve knowledge in digital technologies; • consultations on legislation in digital transformation. • providing access to online platforms and electronic programs; • introducing electronic governance; • strengthening G2B interaction. Barriers • lack of funding; • lack of qualified specialists; • lack of digital/technological capabilities, skills, and knowledge; • high cost of digital technologies. • creation of regulations and standards of digital technologies, etc. State support for digitization of SMEs Consulting, information and non- financial support A set of support measures (financial and non- financial) with an industry approach or technology focus. Figure 3. Conceptual model of state support for digitization of SMEs nancing programs for the digitalization of SMEs following the example of Estonia, Slovenia, or Luxembourg); advisory and informational sup- port (information campaigns for the promotion of digital technologies; seminars and trainings on digital competencies similar to Portugal and Germany), state online platforms (as practiced in France), and ensuring the operation of digi- tal payment systems and technologies, which is envisaged by the strategy of the national govern- ment of Denmark. Accordingly, crucial steps on the way to stimulating business digitalization are: expanding the financial capabilities of SMEs to adopt digital technologies in their processes; formation of digital hard skills among entrepre- neurs and, on this basis, increasing the personnel potential of digitization of SMEs; creation of a favorable business environment to stimulate the digital activity of SMEs; popularization of the strategy of digitalization of business as a compo- nent of the strategy of digital transformation of society (Figure 4). mental functions in supporting the digitization of SMEs, drivers for stimulating the digitization of SMEs, and classification of existing types and forms of state support (Figure 3). Stage 1: Analyzing analytical and statistical data and forming a database for analysisi Consultants provide information on available financing options; information channels to offer data on support initiatives from the government and private sector actors; a daily radio show that discussed future digital trends. The government’s program “SMEs: Digital” is part of the National Government’s Strategy for Digital Growth; the initiative “SME: Digital Features the E-Commerce Centre” is designed to help small and medium-sized enterprises with online sales, includes grants for private consultations to clarify and develop the potential of the company in the field of e-commerce; E-Commerce Center workshops where online selling challenges are discussed with industry experts. The Digi-Lift initiative promoted digital transformations with an emphasis on the industrial sector. It is part of the smart industry strategy. Due to the pandemic, Italian regions offer financial support to companies and self-employed workers to work from home. Funding is provided to cover the cost of training and purchasing digital tools/subscriptions to digital platforms for remote work. “Digital vouchers” program support micro-SMEs with up to 10000 euros to improve digital competences, prepare a digital strategy, ensure cyber security, and implement digital marketing. Implementation of Digital Trading Online Voucher. Micro-enterprises can receive a voucher for 2500 euros to develop online sales and access to free online training. Opendays i4.0 is an information campaign to popularize the need for digital transformation among SMEs. Go-digital is a program to support SMEs (up to 100 employees and a total balance sheet of up to 20 million euros) in IT security, digitization of business processes, and digital market development, thanks to the expertise provided by consulting firms. 297 http://dx.doi.org/10.21511/ppm.20(4).2022.22 Problems and Perspectives in Management, Volume 20, Issue 4, 2022 Problems and Perspectives in Management, Volume 20, Issue 4, 2022 Figure 3. Conceptual model of state support for digitization of SMEs Implementation of digital payment systems and technologies. • creation of a mobile/digital payment system, and development of the business ecosystem. Digitization of SMEs Goal: Accelerating the digital transformation of SMEs by: • training and a course on digitization; provision of digital services; • promotion of digital technologies and tools; • provision of funding/subsidy for digital transformation; • creation of regulations and standards of digital technologies, etc. Barriers • lack of funding; • lack of qualified specialists; • lack of digital/technological capabilities, skills, and knowledge; • high cost of digital technologies. http://dx.doi.org/10.21511/ppm.20(4).2022.22 4. DISCUSSION el of digital transformation. However, SMEs face the same problems when changing their business model in favor of digitalization. Therefore, it is im- portant to provide small and medium-sized enter- prises with government support to adopt digital technologies with limited resources (information- al, financial, or personnel). Supporting Fishchuk et al. (2018), the state should assume several roles si- multaneously for digital modernization and trans- formation. They include: leader and experimenter (attraction of innovative SMEs and IT companies to carry out digital transformation of state bodies and enterprises; integration of services of private companies into their processes through online platforms, etc.); regulator and defender (norma- tive and legal regulation (in particular, standard- ization), training and retraining of personnel, as well as ensuring access to capital, etc.); and pop- ularizer (increasing competence in data science, artificial intelligence, and digital transformation). The previous literature noted the existing challeng- es of business digitalization (Lestari et al., 2021; Liu et al., 2011; Maglakelidze & Erkomaishvili, 2021), but the importance of government support for business digitalization, especially in countries with a low digital transformation rating, was not discussed in details. Scientists have identified business digitalization as a key tool for the devel- opment of SMEs in modern conditions (Cenamor et al., 2019; Guseva & Legominova, 2018; Li et al., 2018) and the importance of state support for busi- ness digitalization (Holotiuk & Beimborn, 2017; Hess et al., 2016; Lia et al., 2022). However, this was done at a very general level without specifying what the government support measures should be and how they can solve the existing digitization problems. There must be a variety of solutions for all countries as each country has a different lev- Stage 1: Analyzing analytical and statistical data and forming a database for analysisi Key measures of state support to stimulate the digitalization of SMEs in the global econom Key measures of state support that make it possible to stimulate the digitization of SMEs in the global economy Increasing the personnel potential of digitization of SMEs Providing SMEs with advisory and informational support and assistance; creation of consultative bodies and consultative groups at the national and subnational levels Encouragement of training and improvement of qualifications of SMEs, formation of data culture in SMEs, increasing the digital security profile of SMEs Expanding the financial capabilities of SMEs Involvement of financial technologies and alternative sources of financing, encouragement of business innovations Eimplification of SME access to financial resources through venture capital funds and grant programs Adapting the business environment to stimulate the digital activity of SMEs Establishment of a favorable regulatory framework, promotion of e-government and e-services for SMEs, deployment of high-quality digital infrastructure, creation of state online platforms Popularizing the strategy of digitalization of business as a component of the strategy of digital transformation of society Development of long-term strategic goals, creation of management mechanisms in new policy areas Providing SMEs with advisory and informational support and assistance; creation of consultative bodies and consultative groups at the national and subnational levels Encouragement of training and improvement of qualifications of SMEs, formation of data culture in SMEs, increasing the digital security profile of SMEs Key measures of state support that make it possible to stimulate the digitization of SMEs in the global economy Involvement of financial technologies and alternative sources of financing, encouragement of business innovations Involvement of financial technologies and alternative sources of financing, encouragement of business innovations Eimplification of SME access to financial resources through venture capital funds and grant programs Establishment of a favorable regulatory framework, promotion of e-government and e-services for SMEs, deployment of high-quality digital infrastructure, creation of state online platforms Establishment of a favorable regulatory framework, promotion of e-government and e-services for SMEs, deployment of high-quality digital infrastructure, creation of state online platforms Adapting the business environment to stimulate the digital activity of SMEs Popularizing the strategy of digitalization of business as a component of the strategy of digital transformation of society Popularizing the strategy of digitalization of business as a component of the strategy of digital transformation of society Popularizing the strategy of digitalization of business as a component of the strategy of digital transformation of society Development of long-term strategic goals, creation of management mechanisms in new policy areas Development of long-term strategic goals, creation of management mechanisms in new policy areas Stage 1: Analyzing analytical and statistical data and forming a database for analysisi The heterogeneity of SME actions, the multiplici- ty of their business environments, and the urgent challenges facing them require a fundamental re- thinking of state policy regarding SMEs and digi- talization. A general government approach will be a key success factor, including effective multi-lev- el governance mechanisms at the national and sub-national, regional, and urban levels, interna- tional learning, and enhanced monitoring and evaluation capacity. The study of the positive practice of state sup- port for the digitalization of SMEs in these coun- tries contributed to the selection of a number of key forms of state support that will stimulate the digitalization of SMEs in the global econo- my (Figure 4). They are: financial support (fi- 298 http://dx.doi.org/10.21511/ppm.20(4).2022.22 Problems and Perspectives in Management, Volume 20, Issue 4, 2022 Figure 4. transformation: Defining meta objectives using business process management. Information & Man- agement, 57(5), 1-13. https://doi. org/10.1016/j.im.2019.103262 CONCLUSION A study of the process of digital transformation in many countries showed that the pace of digitization of small and medium-sized businesses is lower than that of big ones. The literature review proved their 299 http://dx.doi.org/10.21511/ppm.20(4).2022.22 Problems and Perspectives in Management, Volume 20, Issue 4, 2022 greatest vulnerability to the challenges of the pandemic, the Ukrainian-Russian war, and shocks from rising energy prices. The paper substantiates the necessity of using digital technologies as an alternative to business development in overcoming negative consequences. The results found that the level of dig- itization of business in countries is directly dependent on the favorable business environment created by the government for the digital transformation of society. In order to determine the countries whose experience in state support of business digitalization is the best for adoption, clustering was carried out according to the criterion of using crucial digitalization tools by SMEs in these countries. The results made it possible to highlight a number of countries whose experience is sufficiently practical and, therefore, worthy of attention in terms of adoption in countries striving for digital development. The key positions of the state support strategies for SME digitalization in these countries became the basis for the formation of a conceptual model of state support for SME digitalization. It includes the following key components: the goal of SME digitalization; digitization drivers; barriers and challenges of digitization; types and forms of state support for digitization. In addition, the study of the experience of digitally active countries made it possible to outline the strategic prospects of digital business trans- formation for countries with a low level of digitization. They are: expanding the financial capabilities of SMEs to adopt digital technologies; formation of digital hard skills among entrepreneurs and, on this basis, increasing the personnel potential of digitization; creation of a favorable business environment to stimulate the digital activity of SMEs; and popularization of the strategy of business digitalization as a component of the digital transformation strategy of society. AUTHOR CONTRIBUTIONS Conceptualization: Viktoriia Strilets, Serhiy Frolov, Oksana Tymoshenko. Conceptualization: Viktoriia Strilets, Serhiy Frolov, Oksana Tymoshenko. Data curation: Viktoriia Datsenko, Maksym Yatsko. Data curation: Viktoriia Datsenko, Maksym Yatsko. Formal analysis: Maksym Yatsko, Oksana Tymoshenko. y y y Investigation: Serhiy Frolov, Viktoriia Datsenko, Oksana Tymoshenko. Investigation: Serhiy Frolov, Viktoriia Datsenko, Oksana Tymoshenko. g y Methodology: Viktoriia Strilets, Viktoriia Datsenko. g y Methodology: Viktoriia Strilets, Viktoriia Datsenko. Software: Oksana Tymoshenko, Maksym Yatsko. Software: Oksana Tymoshenko, Maksym Yatsko. Validation: Serhiy Frolov, Viktoriia Datsenko. Validation: Serhiy Frolov, Viktoriia Datsenko. Visualization: Viktoriia Strilets, Maksym Yatsko. Writing – original draft: Viktoriia Strilets, Serhiy Frolov, Oksana Tymoshenko, Viktoriia Datsen Maksym Yatsko. Writing original draft: Viktoriia Strilets, Serhiy Frolov, Oksana Tymoshenko, Viktoriia Datsenko, Maksym Yatsko. Writing – review & editing: Viktoriia Strilets, Serhiy Frolov. Fishchuk, V., Matyushko, V., Chernev, E., Yurchak, O., & Lavryk, Ya. (2018). Ukraina 2030E – kraina z rozvynutoiu tsyfrovoiu ekonomikoiu [Ukraine 2030E – a country with a developed digital economy]. Ukrainian Institute of the Future. (In Ukrainian). Re- trieved from https://strategy.uifu- ture.org/kraina-z-rozvinutoyu- cifrovoyu-ekonomikoyu.html 2. Bethlendi, A., & Szőcs, Á. (2022). How the Fintech ecosystem changes with the entry of Big Tech companies. Investment Management and Financial Inno- vations, 19(3), 38-48. https://doi. org/10.21511/imfi.19(3).2022.04 transformation: Defining meta objectives using business process management. Information & Man- agement, 57(5), 1-13. https://doi. org/10.1016/j.im.2019.103262 Fishchuk, V., Matyushko, V., Chernev, E., Yurchak, O., & Lavryk, Ya. (2018). Ukraina 2030E – kraina z rozvynutoiu tsyfrovoiu ekonomikoiu [Ukraine 2030E – a country with a developed digital economy]. Ukrainian Institute of the Future. (In Ukrainian). Re- trieved from https://strategy.uifu- ture.org/kraina-z-rozvinutoyu- cifrovoyu-ekonomikoyu.html 5. Fischer, M., Imgrund, F., Janiescha, C., & Winkelman, A. (2020). Strategy archetypes for digital 4. EC. (2022). Statistical information of the European Commission. Re- trieved from https://ec.europa.eu/ eurostat/web/main/data/database 3. Cenamor, J., Parida, V., & Wincent, J. (2019). How entrepreneurial SMEs compete through digital platforms: The roles of digital platform capability, network capability and ambidexterity. Journal of Business Research, 100, 196-206. https://doi.org/10.1016/j. jbusres.2019.03.035 REFERENCES 3. Cenamor, J., Parida, V., & Wincent, J. (2019). How entrepreneurial SMEs compete through digital platforms: The roles of digital platform capability, network capability and ambidexterity. Journal of Business Research, 100, 196-206. https://doi.org/10.1016/j. jbusres.2019.03.035 4. EC. (2022). Statistical information of the European Commission. Re- trieved from https://ec.europa.eu/ eurostat/web/main/data/database 5. Fischer, M., Imgrund, F., Janiescha, C., & Winkelman, A. (2020). Strategy archetypes for digital 1. Akhter A., Anwarul Islam K. M., Karim M., & Latif, W. B. (2022). Examining determinants of digital entrepreneurial inten- tion: A case of graduate students. Problems and Perspectives in Management, 20(3), 153-163. http://dx.doi.org/10.21511/ ppm.20(3).2022.13 3. Cenamor, J., Parida, V., & Wincent, J. (2019). How entrepreneurial SMEs compete through digital platforms: The roles of digital platform capability, network capability and ambidexterity. Journal of Business Research, 100, 196-206. https://doi.org/10.1016/j. jbusres.2019.03.035 3. Cenamor, J., Parida, V., & Wincent, J. (2019). How entrepreneurial SMEs compete through digital platforms: The roles of digital platform capability, network capability and ambidexterity. Journal of Business Research, 100, 196-206. https://doi.org/10.1016/j. jbusres.2019.03.035 6. 4. 4. EC. (2022). Statistical information of the European Commission. Re- trieved from https://ec.europa.eu/ eurostat/web/main/data/database 2. Bethlendi, A., & Szőcs, Á. (2022). How the Fintech ecosystem changes with the entry of Big Tech companies. Investment Management and Financial Inno- vations, 19(3), 38-48. https://doi. org/10.21511/imfi.19(3).2022.04 2. Bethlendi, A., & Szőcs, Á. (2022). How the Fintech ecosystem changes with the entry of Big Tech companies. Investment Management and Financial Inno- vations, 19(3), 38-48. https://doi. org/10.21511/imfi.19(3).2022.04 5. Fischer, M., Imgrund, F., Janiescha, C., & Winkelman, A. (2020). Strategy archetypes for digital 300 http://dx.doi.org/10.21511/ppm.20(4).2022.22 Problems and Perspectives in Management, Volume 20, Issue 4, 2022 Platform for Community-Based Rural Tourism – A Novel Concept Development in Peru. Sustain- ability, 14(13), 7907. https://doi. org/10.3390/su14137907 7. Global Entrepreneurship Monitor (GEM). (2022). Global Entre- preneurship Monitor 2021/2022 Global report: Opportunity Amid Disruption. Retrieved from https:// gemconsortium.org/report/gem- 20212022-global-report-opportu- nity-amid-disruption 14. Lestari, D., Zainurossalamia, S., Siti, M., Wardhani, W., & Yuda- ruddin, R. (2021). The impact of COVID-19 pandemic on performance of small enterprises that are e-commerce adopters and non-adopters. Problems and Perspectives in Management, 19(3), 467-477. https://doi.org/10.21511/ ppm.19(3).2021.38 21. Melnyk, L., Kubatko, O., Piven, V., Klymenko, K., & Rybina, L. (2021). Digital and economic transforma- tions for sustainable development promotion: A case of OECD countries. Environmental Econom- ics, 12(1), 140-148. https://doi. org/10.21511/ee.12(1).2021.12 8. Guseva, O. Yu., & Legominova, S. V. (2018). REFERENCES Dydzhytalizatsiia – yak instrument udoskonalennia biznes-protsesiv, yikh optymizat- siia [Digitization as a tool for improving business processes, their optimization]. Ekonomika. Menedzhment. Biznes – Economy. Management. Business, 1(23), 33-39. (In Ukrainian). Retrieved from http://nbuv.gov.ua/UJRN/ ecmebi_2018_1_7 15. Li, J. O., Liu, H., Ting, D. S. J., Jeon, S., Chan, R. V. P., Kim, J. E., Sim, D. A., Thomas, P. B. M., Lin, H., Chen, Y., Sakomoto, T., Loewen- stein, A., Lam, D. S. C., Pasquale, L. R., Wong, T. Y., Lam, L. A., & Ting, D. S. W. (2021). Digital tech- nology, tele-medicine and artificial intelligence in ophthalmology: A global perspective. Progress in reti- nal and eye research, 82, 100900. https://doi.org/10.1016/j.pretey- eres.2020.100900 22. Matt, G., Hess, T., & Benlian, A. (2015). Digital transformation strategies. Business & Informa- tion Systems Engineering, 57(5), 339-343. https://doi.org/10.1007/ s12599-015-0401-5 23. Natorina, A. O. (2017). Didzhytal- transformatsiia diialnosti pid- pryiemstv [Digital-transformation of enterprises activities]. Ekonomi- ka ta derzhava – Economy and the State, 4, 90-93. (In Ukrainian). 9. Heavin, C., & Power, D. J. (2018). Challenges for digital transforma- tion-towards a conceptual decision support guide for managers. Jour- nal of Decision Systems, 27(sup1), 38-45. https://doi.org/10.1080/124 60125.2018.1468697 16. Li, L., Su, F., Zhang, W., & Mao, J. Y. (2018). Digital transforma- tion by SME entrepreneurs: a capability perspective. Information Systems Journal, 28(6), 1129-1157. https://doi.org/10.1111/isj.12153 24. Nudurupati, S. S., Budhwar, P., Pappu, R. P., Chowdhury, S., Kondala, M., Chakraborty, A., & Ghosh, S. K. (2022). Transforming sustainability of Indian small and medium-sized enterprises through circular economy adoption. Journal of Business Research, 149, 250-269. https://doi.org/10.1016/j. jbusres.2022.05.036 10. Hess, T., Benlian, A., Matt, C., & Wiesböck, F. (2016). Options for formulating a digital transforma- tion strategy. MIS Quarterly Ex- ecutive, 25(2), 123-139. Retrieved from https://boris.unibe.ch/id/ eprint/105447 17. Lia, C., Khan, A., Ahmad, H., & Shahzad, M. (2022). Business ana- lytics competencies in stabilizing firms’ agility and digital innova- tion amid COVID-19. Journal of Innovation & Knowledge, 7(4), 100246. Retrieved from https:// pesquisa.bvsalud.org/global- literature-on-novel-coronavirus- 2019-ncov/resource/pt/covid- who-2031460 11. Holotiuk, F., & Beimborn, D. (2017). Critical success factors of digital business strategy. 13th Inter- national Conference on Wirtschafts- informatik (pp. 991-1005). St. Gallen, Switzerland. Retrieved from https://wi2017.ch/images/ wi2017-0244.pdf 25. OECD. (2019). Financing SMEs and Entrepreneurs 2019: An OECD Scoreboard. Paris: OECD Publishing. Retrieved from https:// www.oecd.org/cfe/smes/Policy- Highlights-Scoreboard2019-Final- opt.pdf 18. Liu, D.-Y., Chen, S.-W., & Chou, T.-C. (2011). Resource fit in digital transformation: Lessons learned from the CBC Bank global e-banking project. REFERENCES Management Decision, 49(10), 1728-1742. https://doi. org/10.1108/00251741111183852 12. Jansom, A., Srisangkhajorn, T., & Limarunothai, W. (2022). How chatbot e-services motivate com- munication credibility and lead to customer satisfaction: The perspective of Thai consumers in the apparel retailing context. Innovative Marketing, 18(3), 15-27. https://doi.org/10.21511/ im.18(3).2022.02 26. OECD. (2021a). The Digital Transformation of SMEs. Paris: OECD Publishing. https://doi. org/10.1787/bdb9256a-en 27. OECD. (2021b). SME and Entre- preneurship Outlook 2021. Paris: OECD Publishing. https://doi. org/10.1787/97a5bbfe-en 19. Maglakelidze, A., & Erkomaishvili, G. (2021). Challenges of small and medium enterprises dur- ing the COVID-19 pandemic: Case of Georgia. Problems and Perspectives in Management, 19(4), 20-28. http://dx.doi.org/10.21511/ ppm.19(4).2021.02 28. Okoi, I., Ubi, I., Iheanacho, M., Emori, E., & Sunday, E. (2022). Strategic entrepreneurship prac- tices and performance of small and medium-sized enterprises in Nigeria. Problems and Perspectives in Management, 20(1), 108-116. http://dx.doi.org/10.21511/ ppm.20(1).2022.10 13. Kuznyetsova, А., Boiarko, І., Khutorna, М., & Zhezherun, Y. (2022). Development of financial inclusion from the standpoint of ensuring financial stability. Public and Municipal Finance, 11(1), 20-36. https://doi.org/10.21511/ pmf.11(1).2022.03 20. Maquera, G., da Costa, B. F., Men- doza, Ó., Salinas, R. A., & Haddad, N. (2022). Intelligent Digital 301 http://dx.doi.org/10.21511/ppm.20(4).2022.22 Problems and Perspectives in Management, Volume 20, Issue 4, 2022 Problems and Perspectives in Management, Volume 20, Issue 4, 2022 29. Oro, D. (2018, November 13). Gartner Identifies Top 10 Strategic IoT Technologies and Trends from 2018–2023. IoT Central. Retrieved from https://www.iotcentral. io/blog/gartner-identifies-top- 10-strategic-iot-technologies-and- trends-f ology, practice] (457 p.). Poltava: PUET. (In Ukrainian). Retrieved from http://dspace.puet.edu.ua/ bitstream/123456789/8544/1/ Strilec%20Zabezpechennia%20 rozvytku%20malyh%20pidpryem- stv.pdf Yermoshenko, A. (2021). Change of business models of Ukrainian insurance companies in the condi- tions of COVID-19. Insurance Markets and Companies, 12(1), 83-98. https://doi.org/10.21511/ ins.12(1).2021.08 ology, practice] (457 p.). Poltava: PUET. (In Ukrainian). Retrieved from http://dspace.puet.edu.ua/ bitstream/123456789/8544/1/ Strilec%20Zabezpechennia%20 rozvytku%20malyh%20pidpryem- stv.pdf 4. Teteriatnyk, B. S. (2018). Did- zhytyzatsiia ta didzhytalizatsiia v konteksti virtualizatsii hospodar- skoi diialnosti [Digitization and digitalization in the context of virtual economic activity. Vseu- krainska naukovo-praktychna konferentsiia “Innovatsiina systema ta informatsiini tekhnolohii v suchasnii nautsi” – All-Ukrainian scientific and practical conference “Innovative system and information technologies in modern science”. Kharkiv, Ukraine. (In Ukrainian). Retrieved from https://ndipzir.org. ua/14-06-2017 44. 37. Polyakov, M., Khanin, I., Shevchenko, G., & Bilozubenko, V. (2021). Data mining as a cogni- tive tool: Capabilities and limits. Knowledge and Performance Man- agement, 5(1), 1-13. https://doi. org/10.21511/kpm.05(1).2021.01 30. Parida, V., Sjodin, D., & Reim, W. (2019). REFERENCES Reviewing literature on digitalization, business model in- novation, and sustainable industry: Past achievements and future promises. Sustainability, 11(2), 391. https://doi.org/10.3390/ su11020391 38. Reim, W., Yli-Viitala, P., & Arrasvuori, J. (2022). Tackling business model challenges in SME internationalization through digi- talization. Journal of Innovation & Knowledge, 7(3), 100199. https:// doi.org/10.1016/j.jik.2022.100199 31. Parviainen, P., Tihinen, M., Kaa- riainen, J., & Teppola, S. (2022). Tackling the digitalization chal- lenge: How to benefit from digita- lization in practice. International Journal of Information Systems and Project Management, 5(1), 63-77. https://doi.org/10.12821/ ijispm050104 39. Samper, M. G., Sukier, H. B., Palencia, D. B., Molina, R. I. R., Alfaro, K. B., Sánchez, Y. S., & Sarmiento, A. C. F. (2022). Digital transformation of business models: influence of operation and trade variables. Procedia Computer Science, 203, 565-569. https://doi. org/10.1016/j.procs.2022.07.125 45. Ting, D. S. W., Carin, L., Dzau, V., & Wong, T. Y. (2020). Digi- tal technology and COVID-19. Nature Medicine, 26(4), 459-461. https://doi.org/10.1038/s41591- 020-0824-5 32. Pedchenko, N., Strilec, V., & Rudenko, N. (2018). Method of Delpni as an innovative tool for managing asymmetric informa- tion in financial relationships of potential investors and small business entities. Marketing and Management of Innovations, 3, 68-80. https://doi.org/10.21272/ mmi.2018.3-06 46. Vial, G. (2019). Understanding digital transformation: A review and a research agenda. Journal of Strategic Information Sys- tems, 28(2), 118-144. https://doi. org/10.1016/j.jsis.2019.01.003 40. Sestino, A., Prete, M., Piper, L., & Guido, G. (2020). Internet of Things and Big Data as enablers for business digitalization strategies. Technovation, 98, 102173. https:// doi.org/10.1016/j.technova- tion.2020.102173 33. Pereira, C. S., Veloso, B., Durão, N., & Moreira, F. (2022). The influence of technological innovations on international business strategy before and during COVID-19 pandemic. Procedia Computer Science, 196, 44-51. https://doi. org/10.1016/j.procs.2021.11.071 47. Williams, L. (2021). Concepts of Digital Economy and Industry 4.0 in Intelligent and information sys- tems. International Journal of Intel- ligent Networks, 2, 122-129. https:// doi.org/10.1016/j.ijin.2021.09.002 41. Shkolnyk, I., Frolov, S., Orlov, V., Datsenko, V., & Kozmenko, Y. (2022). The impact of financial digitalization on ensuring the economic security of a country at war: New measurement vec- tors. Investment Management and Financial Innovations, 19(3), 119- 138. http://dx.doi.org/10.21511/ imfi.19(3).2022.11 48. Yuan, S., Musibau, H. O., Genç, S. Y., Shaheen, R., Ameen, A., & Tan, Z. (2021). Digitalization of economy is the key factor behind fourth industrial revolution: How G7 countries are overcoming with the financing issues? Technological Forecasting and Social Change, 165. https://doi.org/10.1016/j.tech- fore.2020.120533 34. REFERENCES Petkovski, I., Fedajev, A., & Bazen, J. (2022). Modelling Complex Relationships Between Sustainable Competitiveness and Digitali- zation. Journal of Competitive- ness, 14(2), 79-96. http://dx.doi. org/10.7441/joc.2022.02.05 42. Simatupang, T. M., & Widjaja, F. B. (2012). Benchmarking of In- novation Capability in the Digital Industry. Procedia – Social and Behavioral Sciences, 65, 948-954. http://dx.doi.org/10.1016/j.sb- spro.2012.11.225 35. Pham, T. T., Hoang, H. T. T., & Do, H. T. T. (2022). What is the future of the bank branch in the midst of the 4.0 industrial revolution? Answers from Vietnam. Banks and Bank Systems, 17(2), 199- 208. https://doi.org/10.21511/ bbs.17(2).2022.17 36. Polinkevych, O., Glonti, V., Baranova, V., Levchenko, V., & 49. Yu, H., Fletcher, M., & Buck, T. (2022). Managing digital transfor- mation during re-international- ization: Trajectories and implica- tions for performance. Journal of International Management, 28(4), 100947. https://doi.org/10.1016/j. intman.2022.100947 43. Strilets, V. Yu. (2019). Zabezpech- ennia rozvytku maloho pidpryiem- nytstva: teoriia, metodolohiia, prak- tyka [Ensuring the development of small enterprises: theory, method- 302 http://dx.doi.org/10.21511/ppm.20(4).2022.22 Problems and Perspectives in Management, Volume 20, Issue 4, 2022 APPENDIX A APPENDIX A Table A1. Initial data for the cluster analysis of countries by the level of SME digitization Source: Authors’ elaboration. REFERENCES Country Internet of things Integration of business processes into electronic processes Artificial intelligence Digital integration with customers/ suppliers, supply chain management 3D printers and robotics Big data analysis Cloud services Austria 48 44 7 19 3 5 37 Belgium 25 56 8 21 5 19 49 Bulgaria 14 21 3 9 1 5 10 Bosnia and Herzegovina 13 25 2 11 3 3 6 Greece 21 34 4 8 2 20 Denmark 17 49 20 55 8 21 62 Estonia 16 22 2 61 2 7 55 Ireland 31 23 6 16 2 20 56 Spain 26 48 6 32 5 5 27 Italy 31 31 5 95 4 6 59 Cyprus 32 33 2 12 5 2 48 Latvia 25 38 5 12 2 5 25 Lithuania 24 43 3 25 3 8 29 Luxembourg 20 38 12 12 3 15 29 Malta 24 38 8 21 8 25 53 Netherlands 18 41 10 22 5 23 62 Germany 34 36 9 15 6 15 38 Norway 22 38 9 63 3 16 61 Poland 15 30 2 11 2 6 24 Portugal 20 51 16 16 4 9 31 Romania 9 16 1 15 1 4 13 Serbia 15 21 0 19 1 1 25 Slovakia 24 30 4 14 3 4 33 Slovenia 46 34 9 57 3 4 39 Turkey 20 27 2 19 3 5 8 Hungary 20 19 3 12 3 5 23 Finland 39 47 12 81 6 17 71 France 20 44 5 21 3 18 26 Croatia 21 22 8 39 3 11 36 Czech Republic 28 35 3 9 4 7 42 Sweden 38 33 8 43 5 11 73 Table A1. Initial data for the cluster analysis of countries by the level of SME digitization http://dx.doi.org/10.21511/ppm.20(4).2022.22 Note: Created in IBM SPSS Statistics. APPENDIX B Table B1. Initial centers of clusters Indicators Cluster 1 2 3 4 D1 48 31 17 9 D2 44 31 49 16 D3 7 5 20 1 D4 19 95 55 15 D5 3 4 8 1 D6 5 6 21 4 D7 37 59 62 13 Note: Created in IBM SPSS Statistics. Table B1. Initial centers of clusters 303 http://dx.doi.org/10.21511/ppm.20(4).2022.22 Problems and Perspectives in Management, Volume 20, Issue 4, 2022 Table B2. Chronology of iterations Iteration Change of cluster centers 1 2 3 4 1 22.134 9.626 19.202 8.431 2 1.302 3.209 3.200 1.054 3 .077 1.070 .533 .132 4 .005 .357 .089 .016 5 .000 .119 .015 .002 6 1.559E-5 .040 .002 .000 7 9.170E-7 .013 .000 3.216E-5 8 5.394E-8 .004 6.859E-5 4.020E-6 9 3.173E-9 .001 1.143E-5 5.025E-7 10 1.866E-10 .000 1.905E-6 6.281E-8 Note: Created in IBM SPSS Statistics. a: Iterations are terminated because the maximum number of iterations has been completed. Iterations did not lead to convergence. The maximum change of the absolute coordinate for any center: 000. The current iteration is 10. The minimum distance between the initial centers: 50.941. Note: Created in IBM SPSS Statistics. a: Iterations are terminated because the maximum number of iterations has been completed. Iterations did not lead to convergence. The maximum change of the absolute coordinate for any center: 000. The current iteration is 10. The minimum distance between the initial centers: 50.941. Note: Created in IBM SPSS Statistics. a: Iterations are terminated because the maximum number of iterations has been completed. Iterations did not lead to convergence. The maximum change of the absolute coordinate for any center: 000. The current iteration is 10. The minimum distance between the initial centers: 50.941. Table B3. APPENDIX B Belonging of countries to clusters Observation number Country Cluster Distance 1 Austria 1 23,517 2 Belgium 1 21,408 3 Bulgaria 4 7,781 4 Bosnia and Herzegovina 4 10,541 5 Greece 2 15,85 6 Denmark 3 23,043 7 Estonia 3 20,860 8 Ireland 1 25,413 9 Spain 1 21,001 10 Italy 2 14,439 11 Cyprus 1 18,057 12 Latvia 1 17,315 13 Lithuania 1 13,654 14 Luxembourg 1 14,901 15 Malta 1 19,951 16 Netherlands 1 27,412 17 Germany 1 10,133 18 Norway 3 11,034 19 Poland 4 11,618 20 Portugal 1 18,696 21 Romania 4 9,635 22 Serbia 4 11,581 23 Slovakia 1 14,652 24 Slovenia 3 27,527 25 Turkey 4 11,357 26 Hungary 4 9,927 27 Finland 2 14,440 28 France 1 16,615 29 Croatia 1 26,719 30 Czech Republic 1 13,045 31 Sweden 3 22,396 Table B3. Belonging of countries to clusters Table B3. Belonging of countries to clusters Note: Created in IBM SPSS Statistics. Note: Created in IBM SPSS Statistics. Note: Created in IBM SPSS Statistics. F-criterion should be used only for description purposes, since the clusters are selected so that the difference between observations in different clusters is maximal. The observed significance levels are not adjusted for this and, therefore, cannot be used to test the hypothesis of equality of cluster means. Table B3. Belonging of countries to clusters Note: Created in IBM SPSS Statistics. 304 http://dx.doi.org/10.21511/ppm.20(4).2022.22 Problems and Perspectives in Management, Volume 20, Issue 4, 2022 Table B4. End centers of clusters Table B4. End centers of clusters Table B4. End centers of clusters Factor Cluster 1 2 3 4 D1 Internet of things 26 35 28 15 D2 Integration of business processes into electronic processes 39 39 35 23 D3 Artificial intelligence 7 9 10 2 D4 Digital integration with customers/suppliers 19 88 56 14 D5 3D printing and robotics 4 5 4 2 D6 Big data analysis 12 12 12 4 D7 Cloud services 39 65 58 16 Note: Created in IBM SPSS Statistics. Table B5. Distance between the end centers of clusters Clusters 1 2 3 4 1 74,227 41,654 32,267 2 74,227 33,971 93,426 3 41,654 33,971 63,332 4 32,267 93,426 63,332 Note: Created in IBM SPSS Statistics. Table B6. Number of observations in each cluster Cluster 1 16,000 2 2,000 3 5,000 4 7,000 Valid 30,000 Missed 1,000 Note: Created in IBM SPSS Statistics. Table B7. Distance of ANOVA clusters Indicators Cluster error F Significance Average square Degrees of freedom Average square Degrees of freedom D1 312.281 3 64.256 26 4.860 .008 D2 434.713 3 73.893 26 5.883 .003 D3 70.037 3 15.329 26 4.569 .011 D4 4581.129 3 52.768 26 86.816 .000 D5 8.722 3 2.646 26 3.296 .036 D6 106.924 3 41.657 26 2.567 .076 D7 2328.838 3 120.006 26 19.406 .000 Note: Created in IBM SPSS Statistics. F-criterion should be used only for description purposes, since the clusters are selected so that the difference between observations in different clusters is maximal. The observed significance levels are not adjusted for this and, therefore, cannot be used to test the hypothesis of equality of cluster means. Factor Cluster 1 2 3 4 D1 Internet of things 26 35 28 15 D2 Integration of business processes into electronic processes 39 39 35 23 D3 Artificial intelligence 7 9 10 2 D4 Digital integration with customers/suppliers 19 88 56 14 D5 3D printing and robotics 4 5 4 2 D6 Big data analysis 12 12 12 4 D7 Cloud services 39 65 58 16 Note: Created in IBM SPSS Statistics. Table B5. Distance between the end centers of clusters Table B5. Distance between the end centers of clusters Table B5. Distance between the end centers of clusters Clusters 1 2 3 4 1 74,227 41,654 32,267 2 74,227 33,971 93,426 3 41,654 33,971 63,332 4 32,267 93,426 63,332 Note: Created in IBM SPSS Statistics. Table B6. Table B4. End centers of clusters Number of observations in each cluster Cluster 1 16,000 2 2,000 3 5,000 4 7,000 Valid 30,000 Missed 1,000 Note: Created in IBM SPSS Statistics. Table B7. Distance of ANOVA clusters Table B7. Distance of ANOVA clusters Indicators Cluster error F Significance Average square Degrees of freedom Average square Degrees of freedom D1 312.281 3 64.256 26 4.860 .008 D2 434.713 3 73.893 26 5.883 .003 D3 70.037 3 15.329 26 4.569 .011 D4 4581.129 3 52.768 26 86.816 .000 D5 8.722 3 2.646 26 3.296 .036 D6 106.924 3 41.657 26 2.567 .076 D7 2328.838 3 120.006 26 19.406 .000 Note: Created in IBM SPSS Statistics. F-criterion should be used only for description purposes, since the clusters are selected so that the difference between observations in different clusters is maximal. The observed significance levels are not adjusted for this and, therefore, cannot be used to test the hypothesis of equality of cluster means. 305 http://dx.doi.org/10.21511/ppm.20(4).2022.22
https://openalex.org/W3120953645	https://europepmc.org/articles/pmc7981333?pdf=render	English	null	Functional and directed connectivity of the cortico-limbic network in mice in vivo	Brain structure & function	2,021	cc-by	13,002	* Heiko J. Luhmann luhmann@uni‑mainz.de Abstract Higher cognitive processes and emotional regulation depend on densely interconnected telencephalic and limbic areas. Central structures of this cortico-limbic network are ventral hippocampus (vHC), medial prefrontal cortex (PFC), basolateral amygdala (BLA) and nucleus accumbens (NAC). Human and animal studies have revealed both anatomical and functional alterations in specific connections of this network in several psychiatric disorders. However, it is often not clear whether functional alterations within these densely interconnected brain areas are caused by modifications in the direct pathways, or alternatively through indirect interactions. We performed multi-site extracellular recordings of spontaneous activity in three different brain regions to study the functional connectivity in the BLA–NAC–PFC–vHC network of the lightly anesthetized mouse in vivo. We show that BLA, NAC, PFC and vHC are functionally connected in distinct frequency bands and deter- mined the influence of a third brain region on this connectivity. In addition to describing mutual synchronicity, we determined the strength of functional connectivity for each region in the BLA–NAC–PFC–vHC network. We find a region-specificity in the strength of feedforward and feedback connections for each region in its interaction with other areas in the network. Our results provide insights into functional and directed connectivity in the cortico-limbic network of adult wild-type mice, which may be helpful to further elucidate the pathophysiological changes of this network in psychiatric disorders and to develop target-specific therapeutic interventions. Keywords Multi-electrode recordings · Hippocampus · Medial prefrontal cortex · Basolateral amygdala · Nucleus accumbens · Temporal partial directed coherence Functional and directed connectivity of the cortico‑limbic network in mice in vivo Zeinab Khastkhodaei1 · Muthuraman Muthuraman2 · Jenq‑Wei Yang1 · Sergiu Groppa2 · He Received: 5 February 2020 / Accepted: 16 December 2020 / Published online: 13 January 2021 © The Author(s) 2021 Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/s0042 9-020-02202-7. 2 Section of Movement Disorders and Neurostimulation, Biomedical Statistics and MULTIMODAL Signal Processing Unit, Department of Neurology, University Medical Center of the Johannes Gutenberg University Mainz, Duesbergweg 6, 55128 Mainz, Germany Brain Structure and Function (2021) 226:685–700 https://doi.org/10.1007/s00429-020-02202-7 Brain Structure and Function (2021) 226:685–700 https://doi.org/10.1007/s00429-020-02202-7 ORIGINAL ARTICLE Introduction visual cortex). Therefore, we have chosen the in vivo sys- tem to analyze functional interactions in the cortico-limbic network consisting of BLA, PFC, NAC and vHC. Given the anatomical intermingled topographies of functionally distinct neurons in BLA (McGarry and Carter 2017; Beyeler et al. 2018) and NAC (Gangarossa et al. 2013), it is advantageous to record network activity and multi-unit activity at multiple sites. Multi-electrode array (MEA) technology offers the advantage to record the activity (i) from different and even deep brain areas simultaneously, (ii) at multiple sites in the target brain region, and (iii) from the local network (local field poten- tial, LFP) and multi- or single-unit activity (MUA, SUA, respectively). However, depending on the brain regions of interest, such multi-site MEA recordings are difficult to perform in awake or even freely moving animals (but see Hultman et al. 2018). Here we studied in adult wild-type male C57BL/6 N mice the functional and directed connectivity of four brain structures within the cortico-limbic network (vHC–BLA–NAC–PFC) using extracellular MEA record- ings. We measured in three areas simultaneously sponta- neous LFPs and MUA in a controlled lightly anesthetized state. For each brain region, we determined the dominant frequency and the functional interactions between two con- nected brain regions. The influence of a third brain region on the interactions between two connected brain areas was determined by quantification of the partial coherence between these three signals. To detect and quantify the func- tional interactions within and between these networks we used temporal partial directed coherence (tPDC), a novel method previously applied in the analysis of EEG, MEG and fMRI signals (Leistritz et al. 2013; Anwar et al. 2016; Vergotte et al. 2017; Muthuraman et al. 2018). Rather than merely describing mutual synchronicity, as in cross-cor- relation and coherence analysis, tPDC gives information whether and how two brain structures interact over time and in specific frequency bands via direct feedforward or feed- back connections (Baccala and Sameshima 2001; Muthura- man et al. 2018; Pagnotta and Plomp 2018). i Although imaging techniques are powerful methods to monitor the activity from the whole brain (e.g. fMRI) or from hundreds of neurons (e.g. calcium imaging), the temporal resolution of these techniques are inferior as compared to electrophysiological recordings. Further- more, EEG and MEG have a poor spatial resolution and do not allow recordings from deep brain structures. Introduction on densely interconnected telencephalic and limbic areas (Simons and Spiers 2003; Sesack and Grace 2010; Pres- ton and Eichenbaum 2013). The ventral hippocampus (vHC), medial prefrontal cortex (PFC), basolateral amyg- dala (BLA) and nucleus accumbens (NAC) are central structures to compose these cortico-limbic network inter- actions. Recently, experimental studies in animals and clinical data obtained in humans provided evidence that structural anomalies and aberrant functional coupling in specific connections of this cortico-limbic network may be related to several psychiatric diseases, such as schizophre- nia, depression, and post-traumatic stress disorder (Fornito and Bullmore 2015; Hillary and Grafman 2017). It has been recently demonstrated that mice exposed to stress in a model of chronic social defeat display reduced myelin protein content in the NAC and a reduced myelin thickness only in the PFC of stress susceptible mice (Bonnefil et al. 2019). Dysfunction in the vHC to PFC pathway and its reciprocal connectivity to the BLA have been suggested Higher cognitive processes and emotional regulation, such as associative learning, episodic memory formation, decision making, threat processing, and anxiety, depend Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/s0042 9-020-02202-7. * Heiko J. Luhmann luhmann@uni‑mainz.de 1 Institute of Physiology, University Medical Center of the Johannes Gutenberg University Mainz, Duesbergweg 6, 55128 Mainz, Germany 2 Section of Movement Disorders and Neurostimulation, Biomedical Statistics and MULTIMODAL Signal Processing Unit, Department of Neurology, University Medical Center of the Johannes Gutenberg University Mainz, Duesbergweg 6, 55128 Mainz, Germany (0123 1 Brain Structure and Function (2021) 226:685–700 686 as key mechanisms underlying not only some psychiat- ric disorders, but also stress regulation and resilience to emotional stress (Godsil et al. 2013; Padilla-Coreano et al. 2019). For example, hippocampus and prefrontal cortex show reduced connectivity in depression (Genzel et al. 2015), abnormal synchrony in schizophrenia (Dick- erson et al. 2012) and disruption of plasticity following an acute stress exposure (Rocher et al. 2004). A recent study reported the role of hippocampal-prefrontal theta communication in regulating avoidance behavior (Padilla- Coreano et al. 2019). Furthermore, BLA hyperactivity and NAC hypoactivity have been observed in several mood disorders (for review see Russo et al. 2012). However, it is often not clear whether functional alterations within these densely interconnected brain areas are caused by modifi- cations in the direct pathways, or alternatively through indirect connections. Therefore, the identification of the pathway(s) undergoing pathophysiological changes is essential to develop a specific therapeutic intervention. Materials and methods BLA–NAC and BLA–PFC. Using this approach, we were not able to study the vHC–BLA connectivity. Recordings from vHC were obtained through a craniotomy located 3 mm lateral to the midline and 3 mm posterior to Bregma. Recordings from NAC were obtained from 0.8 mm lateral to the midline and 1.1 mm anterior to the Bregma. Due to space limitations, we reached PFC via a contralateral cra- niotomy located 0.5 mm lateral to the midline and 1.5 mm anterior to the Bregma in a 60° angle. Recordings from BLA were obtained through a craniotomy located 2.8 mm lateral to the midline and 1.5 mm posterior to Bregma. For vHC, BLA and NAC, we used a 32-channel MEA in a 4-shank configuration (A4 × 8-A32, Neuronexus; 200 µm inter-shank spacing), and for PFC we used a one shank 32-channel MEA (A1 × 32-Poly2, Neuronexus). Extracel- lular signals were recorded at 20 kHz with a ME256-FIA- MPA system or ME2100 system (Multi Channel Systems, Reutlingen, Germany). All procedures related to the care and treatments of ani- mals were approved by a local ethics committee (#23 177- 07/G 14-1-080) and followed the German and European national regulations (European Communities Council Directive, 86/609/EEC). In total 16 C57BL6/N male mice were used, which ranged in age from 2 to 3 months. Nine of these mice were used for simultaneous BLA–NAC–PFC recordings and 7 were used for vHC-NAC-PFC record- ings. All recordings have been performed in the right hemisphere. Introduction Thus, intracranial recordings are currently the most powerful method to perform recordings with high spatial and tem- poral resolution at the network and single cell level even in deep brain structures. For obvious reasons intracranial recordings can be temporally performed in only a rather limited group of patients, cases of Parkinson´s disease during deep brain stimulation (Cagnan et al. 2019) or in pharmaco-resistant epilepsy (Sinha et al. 2017), and in these patients the intracranial electrodes are located in regions relevant to these diseases. Therefore, multi-site recordings simultaneously from many brain regions can be performed only in non-human models. In recent years, the number of in vitro and in vivo animal studies that have performed simultaneous recordings from multiple regions has been increasing. Despite the well-known advantages of in vitro approaches such as cellular and subcellular reso- lution (e.g. patch-clamp recordings from defined cellular compartments, such as dendrite or presynaptic terminal), experimental control of the extracellular milieu (e.g. to apply drugs in specific concentrations reaching the target structure without restraints by the blood–brain barrier), or well-defined control of presynaptic stimulation (e.g. mon- osynaptic single-fiber activation), in vitro systems do not fully recapitulate network connectivity (especially of long- range projections), the influence of neuromodulatory sys- tems (e.g. cholinergic or serotonergic modulatory action) and spontaneous network activity (e.g. alpha rhythm in Using this approach, we demonstrate that vHC–BLA–NAC–PFC are highly interconnected in their LFP and spike activity. We show the influence of PFC in synchronizing NAC with BLA and vHC, albeit in differ- ent extent. We find a region-specificity in the strength of feedforward and feedback connections for each region in its interaction with other areas in the network. To our knowl- edge this is the first report demonstrating not only the tech- nical feasibility of simultaneous MEA recordings in three brain regions of interest in the lightly anesthetized mouse in vivo, but also the applicability of the directed connectiv- ity (tPDC) method in multi-site brain region recordings and analyses using small animal models. 1 3 3 3 Brain Structure and Function (2021) 226:685–700 687 Surgical preparation and electrophysiological recordings Anesthesia was induced by 5% isoflurane and maintained during surgery by a combination of urethane (375 mg/kg, ip), chlorprothixene hydrochloride (2 mg/kg, ip) and iso- flurane (1–2%). Lidocain hydrochloride gel 2% was used for local analgesia and atropine (0.3 mg/kg, sc) to reduce bronchial secretions. Animal temperature was kept at 37 °C. A custom-designed head post was mounted to the skull using cyanoacrylate adhesive (Permabond Engineer- ing Adhesives, Eastleigh, UK). As reference we used a wire placed into the cerebellum. During recordings, iso- flurane was reduced to 0% to minimize anesthesia-induced unwanted side effects on neural activity and functional connectivity (Williams et al. 2010; Bukhari et al. 2018; Paasonen et al. 2018). Previous fMRI experiments com- paring six different anesthesia protocols have demonstrated that the pattern of functional connectivity during urethane anesthesia, even with three times higher urethane concen- trations (1250 mg/kg, ip) than used in our experiments, is similar to that observed in awake rodents. Functional con- nectivity of the default mode network, interhemispheric and complex-network parameters (e.g. correlation coeffi- cients, modularity, clustering coefficients) and thalamocor- tical connectivity is better preserved in urethane anesthesia as compared to the other anesthetics tested (Paasonen et al. 2018). Depending on the level of anesthesia as assessed by the breathing rate and presence or absence of the pinch toe reflex, additional doses of chlorprothixene hydrochlo- ride (2 mg/kg, ip) were injected (after approximately 3 h). Recordings typically lasted for 5 h. Functional connectivity analyses a Schematic illustration of MEAs in BLA, NAC, vHC and PFC. Note that a one-shank MEA for recording in PFC was inserted from the contralateral hemisphere. Sec- tions from Paxinos and Franklin (2001). b Photographs of coro- nal sections showing the track and localization of the MEAs, covered with DiI for the BLA– NAC–PFC configuration (top) and for the vHC–NAC–PFC configuration (bottom). The distance from Bregma: − 1.46, 0.86, 1.98 mm for BLA, NAC and PFC, respectively (top) and − 2.80, 1.10 and 1.70 mm for vHC, NAC and PFC respec- tively (bottom). In this and fol- lowing figures, BLA is shown in blue color, NAC in red, vHC in black and PFC in green te of coherence value is indicated by over- et al. 1995). To test the significance of the a particular frequency we used confidence % 𝜔)\|\|\| 2 ̂Syy(𝜔) . where the magnitude of CYxy ( 휔 ) is called coherency between the two signals x and y. Again, the overcap indicates the estimate of that value (Halliday et al. 1995). Cxy\|z(𝜔) = \|\|\| CYxy(𝜔) − CYxz(𝜔) CYzy(𝜔)\|\|\| 2 (1 −̂Cxz(𝜔) )(1 −̂Czy(𝜔) ) , on and configu- or simultane- the cortico- f the adult Schematic EAs in BLA, PFC. Note that A for recording ted from the misphere. Sec- os and Franklin raphs of coro- wing the track of the MEAs, for the BLA–i guration (top) –NAC–PFC ttom). The egma: − 1.46, r BLA, NAC ively (top) and 1.70 mm for PFC respec- n this and fol- LA is shown in in red, vHC in green Cxy\|z(𝜔) = \|\|\| CYxy(𝜔) − CYxz(𝜔) CYzy(𝜔)\|\|\| 2 (1 −̂Cxz(𝜔) )(1 −̂Czy(𝜔) ) , Cxy\|z(𝜔) = \|\|\| CYxy(𝜔) − CYxz(𝜔) CYzy(𝜔)\|\|\| 2 (1 −̂Cxz(𝜔) )(1 −̂Czy(𝜔) ) , ̂C(𝜔) = \|\|\|̂Sxy(𝜔)\|\|\| 2 ̂Sxx(𝜔)̂Syy(𝜔) . (1 −̂Cxz(𝜔) )(1 −̂Czy(𝜔) ) where the magnitude of CYxy ( 휔 ) is called coherency between the two signals x and y. Again, the overcap indicates the estimate of that value (Halliday et al. 1995). The estimate of coherence value is indicated by over- cap (Halliday et al. 1995). To test the significance of the coherence at a particular frequency we used confidence limit as an indicator, which was calculated at the 100% 훼 as given by 1 −(1 −훼)1∕(M−1)(Halliday et al. 1995; Govindan et al. 2006) where 훼 is set to 0.90, so the confidence limit is 1 −0.11∕(M−1) . Functional connectivity analyses To calculate the spike count correlation during spontaneous activity we followed our previous paper (Yang et al. 2013) with modified parameters for surrogate, which includes 100 trials and 100 ms jittering time. The correlation coefficient was considered significant if the values differed from the mean of the surrogate by > 2 SD. To test the neuronal syn- chrony across multiple areas in the cortico-limbic network as described earlier (Govindan et al. 2006) we measured the LFP coherence between each pair of recording sites. The LFP coherence analysis was computed by band-pass filtering at 0.5–250 Hz. To reduce the adverse effects of volume con- duction, we subtracted signals from two neighboring elec- trodes and obtained bipolar LFPs. Next, for each mouse we selected one representative channel in each area and com- puted the coherence for each representative channel pair. Finally, we averaged the coherences across all mice. Having x(t) and y(t) as the two simultaneously measured signals, we divided the length of each signal (N) into multiple segments (M) in a length of (L), so that N = LM. To estimate both ordinary and partial coherence it is important to select an optimal length of L to maintain both sensitivity and reliabil- ity of the estimation. Here, we always used N = 6,000,000, L = 40,000 and M = 150 for coherence analysis. To measure the power spectra Sxx ( 휔 ) and Syy ( 휔 ) of the signals x and y, we performed Fourier transformation of the autocorrelation function in each window. The Fourier transformation of the cross-correlation function of the signal x and y in each win- dow was considered as cross spectrum of each pair Sxy ( 휔 ). Then we averaged the power spectra and the cross spectrum across all the segments for each pair to get the estimate of the same. The coherence was calculated as follows: We performed extracellular MEA recordings in the vHC–BLA–NAC–PFC network, simultaneously with two technically feasible combinations of MEA localizations in vHC–NAC–PFC or BLA–NAC–PFC (Fig. 1, Suppl. Fig. 1). In each animal we recorded either vHC–NAC–PFC or BLA–NAC–PFC and measured functional connectivity in five distinct pairs: vHC–NAC, vHC–PFC, NAC–PFC, 1 1 3 3 3 Brain Structure and Function (2021) 226:685–700 688 Fig. 1 Localization and configu- ration of MEAs for simultane- ous recordings in the cortico- limbic network of the adult mouse in vivo. Functional connectivity analyses Two-time series were considered as “cor- related” if the coherence value was above this confidence limit and “uncorrelated” if the coherence value was below this confidence limit. 3 Spontaneous activity in cortico‑limbic network Spontaneous activity in the cortico-limbic network was char- acterized by stable neural activity in all recorded regions over a long time period such that the average firing rates at the beginning and at the end of the experiments were compa- rable (6.36 ± 3.32 vs. 7.63 ± 5.97 Hz in BLA, 21.78 ± 19.06 vs. 26.23 ± 18.61 in NAC, 4.85 ± 3.27 vs. 5.21 ± 3.92 in PFC, 7.75 ± 7.9 vs. 8.03 ± 8.19 in vHC, n = 992 channels from 16 animals, mean ± SD; Fig. 2a, b). Given the previous data on cortico-limbic network synchrony in lower frequen- cies (Hultman et al. 2016), we concentrated our further anal- ysis on frequencies below gamma (all frequencies less than 32 Hz). Defining the frequency bands as delta = 0.5–3 Hz, theta = 4–7 Hz, alpha = 8–15 Hz, beta = 16–31 Hz and gamma = 32–100 Hz, LFPs in all four brain regions showed a high power in lower frequencies. The strongest power of LFP in BLA, NAC and PFC were observed in the delta band and in vHC in both delta and theta bands. In the BLA–NAC–PFC configuration (Fig. 2c) the median power in the delta band was 2.04 (BLA), 4.58 (NAC), 1.92 µV2/ Hz (PFC), and in theta band 0.52 (BLA), 0.66 (NAC), 0.33 µV2/Hz (PFC). Higher frequencies showed lower power with alpha 0.09 (BLA), 0.19 (NAC), 0.09 µV2/Hz (PFC), and beta 0.02 (BLA), 0.02 (NAC), 0.02 µV2/Hz (PFC). In the vHC-NAC-PFC configuration (Fig. 2d) the median power of delta was 2.37 (vHC), 3.76 (NAC), and 2.77 µV2/Hz (PFC), and theta 1.61 (vHC), 1.86 (NAC), 1.8 µV2/Hz (PFC). The median power was lower for alpha 0.52 (vHC), 0.45 (NAC), 0.48 µV2/Hz (PFC), and for beta 0.2 (vHC), 0.11 (NAC), 0.12 µV2/Hz (PFC). These findings are not only in agree- ment with previous reports (Dzirasa et al. 2011a; Hultman et al. 2016; Padilla-Coreano et al. 2019), but also indicate the stability of all recording sites over experimental sessions lasting up to 5 h, as the average firing rates are relatively constant during the entire recording session. Time‑resolved partial directed coherence (tPDC) By concurrently using two Kalman fil- ters working in parallel with one another, it is possible to estimate both states and model parameters of the system at each time instant. After estimating the time-varying multi- variate (MVAR) coefficients, the next step is to use those coefficients for the calculation of causality between the time series. Since DEKF yields the time varying MVAR coefficients at each time point, we can calculate the par- tial directed coherence (PDC) at each time point as well. Afterwards, a time–frequency plot using all PDCs can be concatenated to produce a time–frequency plot. The pre- cise distribution of the MVAR coefficients is not known; we used the surrogate method called bootstrapping (Kaminski et al. 2001) to check for the significance of the results. This method is based on the random shuffling of the subjected time series and hence it is data driven. In short, it divides the original time series into smaller non-overlapping windows and randomly shuffles the order of these windows to create a new time series. The MVAR model was fitted to this shuffled time series. This process was repeated 1000 times and their average are calculated. The resulting 95th percentile value of the null distribution is the significance threshold value for all connections. Additionally, we applied time reversal technique (Haufe et al. 2013) as a second significance test on the connections already identified by tPDC using data driven bootstrapping surrogate significance test. region (xy vs. xz; xy vs. yz; xz vs. yz) using MATLAB. All ANOVA analysis were followed by post hoc using mult- compare function in MATLAB. For the ANOVA results that showed interaction between factors, a paired t test was performed following ANOVA. To test the difference in cross-correlations we performed a Mann–Whitney U test in GraphPad Prism. region (xy vs. xz; xy vs. yz; xz vs. yz) using MATLAB. All ANOVA analysis were followed by post hoc using mult- compare function in MATLAB. For the ANOVA results that showed interaction between factors, a paired t test was performed following ANOVA. To test the difference in cross-correlations we performed a Mann–Whitney U test in GraphPad Prism. Histology For post hoc identification of the recording sites, we coated each electrode with fluorescent lipophilic dye (DiI, D282, Invitrogen). After recordings, mice were transcardially per- fused under xylazine hydrochloride anesthesia (30 mg/kg) with 0.1 M phosphate buffer (PB), followed by 4% para- formaldehyde (PFA). Brains were removed and post-fixed in 4% PFA at 4 °C overnight and after rinsing 3 times with 0.1 M PB transferred to 30% sucrose in 0.1 M PB solu- tion. Using a freezing microtome (Leica, Wetzlar, Germany) brains were sliced at 200 µm thickness and were inspected for the presence of the tracers using a fluorescent microscope (Fig. 1b). Time‑resolved partial directed coherence (tPDC) Using time–frequency causality we cannot only focus on a particular frequency itself but can also analyze the time- dynamics of the causality at that frequency. Based on state-space modelling, the time–frequency causality esti- mation method of tPDC relies on dual-extended Kalman filtering (DEKF) (Wan and Nelson 2001). These results are i Similarly, we measured the partial coherence between the three signals x(t), y(t) and z(t) as follows: 1 3 3 Brain Structure and Function (2021) 226:685–700 689 an estimate of the time-varying dependent autoregressive (AR) coefficients: one EKF estimates the states and feeds this information to the second DEKF, which estimates the model parameters and back propagates this information to the initial EKF. By concurrently using two Kalman fil- ters working in parallel with one another, it is possible to estimate both states and model parameters of the system at each time instant. After estimating the time-varying multi- variate (MVAR) coefficients, the next step is to use those coefficients for the calculation of causality between the time series. Since DEKF yields the time varying MVAR coefficients at each time point, we can calculate the par- tial directed coherence (PDC) at each time point as well. Afterwards, a time–frequency plot using all PDCs can be concatenated to produce a time–frequency plot. The pre- cise distribution of the MVAR coefficients is not known; we used the surrogate method called bootstrapping (Kaminski et al. 2001) to check for the significance of the results. This method is based on the random shuffling of the subjected time series and hence it is data driven. In short, it divides the original time series into smaller non-overlapping windows and randomly shuffles the order of these windows to create a new time series. The MVAR model was fitted to this shuffled time series. This process was repeated 1000 times and their average are calculated. The resulting 95th percentile value of the null distribution is the significance threshold value for all connections. Additionally, we applied time reversal technique (Haufe et al. 2013) as a second significance test on the connections already identified by tPDC using data driven bootstrapping surrogate significance test. an estimate of the time-varying dependent autoregressive (AR) coefficients: one EKF estimates the states and feeds this information to the second DEKF, which estimates the model parameters and back propagates this information to the initial EKF. Functional connectivity of MUA within the cortico‑limbic network Statistical analyses were performed in MATLAB and Graph- Pad Prism. To investigate the direction- and region-specific- ity of the connections we performed an ANOVA with the within-subject factor direction of connectivity (x–y vs. y–x; x–z vs. z–x; y–z vs. z–y), and the between-subject factors To measure the functional connectivity within the cortico- limbic network, we computed cross-correlation analyses 1 Brain Structure and Function (2021) 226:685–700 690 690 Brain Structure and Function (2021) 226:685 700 (spike count correlation: rSC) of MUA during spontaneous activity for the five paired brain regions (Fig. 3). Figure 3a shows examples of cross-correlations for five paired regions. The cross-correlation functions of BLA–PFC interactions showed the largest peak at a positive lag of 10–20 ms, which means BLA (reference channel) is driving PFC (target chan- nel) (Fig. 3a, top left panel, median across 9 mice 1 ms, Fig. 3b). We also found a correlation between the number of spikes in BLA and NAC, which occurred at a lag of 0–10 ms, indicating BLA is leading NAC (Fig. 3a, median across 9 Fig. 2 Simultaneous recordings of spontaneous activity in three brain regions. a Local field potentials (LFPs) and multi-unit activ- ity (MUA) recorded simultaneously in BLA, NAC and PFC. Lower traces show recordings at higher temporal resolution. b Same as in (a) for simultaneous recordings in vHC, NAC and PFC. c Power spectra of spontaneous LFP activity in BLA, NAC and PFC averaged across all mice (n = 9). Thick lines represent mean and thin lines rep- resent ± SD. In boxplots, frequency bands are definded as delta from 0 to 3 Hz, theta from 4 to 7 Hz, alpha from 8 to 15 Hz, and beta from 16 to 31 Hz. d Same as in (c), but for simultaneous recordings in vHC, NAC and PFC in 7 mice. Please note the different scale of NAC boxplot in (c) as compared to other boxplots. In this and following figures, thin red lines in boxplots indicate the medians and red crosses show outliers resent ± SD. In boxplots, frequency bands are definded as delta from 0 to 3 Hz, theta from 4 to 7 Hz, alpha from 8 to 15 Hz, and beta from 16 to 31 Hz. d Same as in (c), but for simultaneous recordings in vHC, NAC and PFC in 7 mice. Functional connectivity of MUA within the cortico‑limbic network Please note the different scale of NAC boxplot in (c) as compared to other boxplots. In this and following figures, thin red lines in boxplots indicate the medians and red crosses show outliers Fig. 2 Simultaneous recordings of spontaneous activity in three brain regions. a Local field potentials (LFPs) and multi-unit activ- ity (MUA) recorded simultaneously in BLA, NAC and PFC. Lower traces show recordings at higher temporal resolution. b Same as in (a) for simultaneous recordings in vHC, NAC and PFC. c Power spectra of spontaneous LFP activity in BLA, NAC and PFC averaged across all mice (n = 9). Thick lines represent mean and thin lines rep- means BLA (reference channel) is driving PFC (target chan- nel) (Fig. 3a, top left panel, median across 9 mice 1 ms, Fig. 3b). We also found a correlation between the number of spikes in BLA and NAC, which occurred at a lag of 0–10 ms, indicating BLA is leading NAC (Fig. 3a, median across 9 (spike count correlation: rSC) of MUA during spontaneous activity for the five paired brain regions (Fig. 3). Figure 3a shows examples of cross-correlations for five paired regions. The cross-correlation functions of BLA–PFC interactions showed the largest peak at a positive lag of 10–20 ms, which 1 3 3 Brain Structure and Function (2021) 226:685–700 691 Fig. 3 Spike count cross-correlations within the cortico-limbic net- work. a Representative cross-correlations between two brain regions. In all graphs black curves correspond to mean of surrogate, red curves correspond to mean standard deviation of surrogate and red crosses correspond to the maximum strength of correlation and “ref” stands for reference channel. b Comparison of the lag time of each region with two other connected regions averaged across all mice (n = 9 for BLA-NAC-PFC, n = 7 for vHC-NAC-PFC recording condi- tion). In the boxplot * represents p < 0.05 and *p < 0.002. c Com- parison of the maximum correlation strength of each region with two other connected regions averaged across all mice (n = 9 for BLA- NAC-PFC, n = 7 for vHC-NAC-PFC recording condition). In the box- plot represents p < 0.05, p < 0.001, p < 10–4 Fig. 3 Spike count cross-correlations within the cortico-limbic net- work. a Representative cross-correlations between two brain regions. Functional connectivity of MUA within the cortico‑limbic network In all graphs black curves correspond to mean of surrogate, red curves correspond to mean standard deviation of surrogate and red crosses correspond to the maximum strength of correlation and “ref” stands for reference channel. b Comparison of the lag time of each region with two other connected regions averaged across all mice (n = 9 for BLA-NAC-PFC, n = 7 for vHC-NAC-PFC recording condi- tion). In the boxplot represents p < 0.05 and *p < 0.002. c Com- parison of the maximum correlation strength of each region with two other connected regions averaged across all mice (n = 9 for BLA- NAC-PFC, n = 7 for vHC-NAC-PFC recording condition). In the box- plot represents p < 0.05, p < 0.001, p < 10–4 mice 9 ms, Fig. 3b). NAC and PFC showed a clear peak of correlation occurring at a positive lag of 0–10 ms, demon- strating that NAC drives PFC (Fig. 3a, lower panel, median across 9 mice 1 ms, Fig. 3b). In vHC–NAC–PFC recordings, the cross-correlation analysis revealed that vHC is correlated with PFC and NAC. The vHC–NAC and vHC–PFC cross- correlations illustrate that vHC drives both NAC and PFC (Fig. 3a, top, median across 7 mice 11.5 for vHC–NAC and 23 ms for vHC–PFC, Fig. 3b). It should also be noted that for each pair of the five tested connections, the time lag of cross-correlation peak was not always consistent across different channel combinations, such that in a given paired region we observed negative, positive and/or zero time lags in different channel combinations (Fig. 3b). Comparing the lag time of correlation between different pairs, we found that the BLA–NAC time lag of correlation is significantly longer than NAC-PFC (p = 0.04, Mann–Whitney U test; Fig. 3b). In addition, the lag time of correlation between vHC and PFC is significantly longer than BLA–PFC correlation’s lag time (p = 0.01, Fig. 3b) and also NAC–PFC correlation’s lag time (p < 0.002, Fig. 3b). Although the medians of lag time of cross-correlation in all five pairs are in positive range the 1 3 Brain Structure and Function (2021) 226:685–700 692 areas, we noted that the correlation in BLA–PFC is stronger than the vHC–PFC correlation (p = 0.02, Mann–Whitney U test; Fig. 3c) and the correlation in BLA–NAC is stronger than the vHC–NAC correlation (p = 0.01, Mann–Whitney U test; Fig. 3c). Functional connectivity of MUA within the cortico‑limbic network Altogether, the spike count cross-correlation results demonstrate that within the cortico-limbic network the BLA–NAC correlation is the strongest among all four connections studied. Furthermore, BLA makes a stronger connection with NAC and PFC than vHC does with these two regions. variability of the results was rather high as evident in large error bars. Under this condition, resolving the directional- ity based on merely cross-correlation analysis is difficult (Bastos et al. 2018) and therefore, we used another method to determine the direction of functional connectivity (see below). In summary, a substantial number of channels in all five pairs showed significant correlations during spontane- ous activity. Next, we asked whether the correlation strength of a given region with the second area is different than that of the third region. To address this question, we compared the rSC values of each pair (e.g. rSC of BLA–PFC) with that of its corresponding third region (i.e. rSC of BLA–NAC) and averaged across all mice. We found that for BLA, the corre- lation in the BLA–NAC pair is significantly stronger than the correlation in the BLA–PFC pair (p < 0.001, Mann–Whitney U test; Fig. 3c). For NAC, we found that the BLA–NAC correlation is significantly stronger than the NAC–PFC correlation (p < 10–4, Mann–Whitney U test; Fig. 3c). In vHC–NAC–PFC recordings, we found that the strength of correlation in NAC–PFC is stronger than the vHC-PFC cor- relation (p < 0.001, Mann–Whitney U test; Fig. 3c). Compar- ing the correlation between BLA and vHC with the other Ordinary and partial coherence of LFPs during spontaneous activity As expected, vHC and PFC also were sig- nificantly coherent and showed a maximum peak occurring at 1 Hz (0.17 ± 0.13, Fig. 4a, right). These results demon- strate coherence between all five pairs, although the coher- ence strength is different across pairs such that BLA–PFC coherence is the strongest and vHC–NAC coherence is the weakest. found that the maximum coherence of vHC–NAC decreased by approximately 31% from 0.16 ± 0.13 to 0.11 ± 0.11 after removing the PFC influence (Fig. 4b, second plot from right). Despite of this reduction the vHC–NAC coherence remained significant and therefore, PFC influence can only partly describe the vHC–NAC coherence. When comparing ordinary coherence and partial coherence for vHC–PFC pair, we noticed that the peak of coherence between vHC and PFC declined by 30% from 0.17 ± 0.13 to 0.12 ± 0.11 after removing the NAC influence. This reduction demonstrates the NAC contribution in synchronizing vHC and PFC, how- ever, although the majority of vHC and PFC coherency is preserved without NAC involvement. Our results in partial coherence analysis suggest that within the cortico-limbic network an individual region plays an important role in syn- chronizing other regions although the contribution on this synchronization is different and varies from 30% (influence of NAC on vHC–PFC synchrony) to 60% (influence of PFC on BLA–NAC synchrony). Next, we tested whether the observed coherence in each pair is due to common influence of a third area. We meas- ured the partial coherence for each pair after removing the influence of the third area and averaged it across all mice (Fig. 4b). Left panel of Fig. 4b shows partial coherence between BLA and NAC when the contribution of PFC as a potential common influence has been removed (BLA–NAC/ PFC). When comparing ordinary coherence and partial coherence for the BLA–NAC pair, we found that the peak of coherence between BLA and NAC decreased by 60% from 0.35 ± 0.17 to 0.14 ± 0.15 after removing the contribution of PFC (Fig. 4b, left). This result demonstrates the powerful impact of the PFC on the BLA-NAC coherency, although it cannot fully account for the BLA–NAC connection, as their partial coherence still remains significant even after removing the PFC common influence. Similarly, we com- puted partial coherence for the BLA–PFC pair by excluding the NAC influence (Fig. 4b, second panel from left). Ordinary and partial coherence of LFPs during spontaneous activity We found that the maximum coherence of BLA-PFC decreased by 42.5% from 0.4 ± 0.23 to 0.23 ± 0.16 after removing the NAC influence. Since BLA-PFC coherence after removing NAC contribution still remains significant, it cannot fully be explained by the NAC influence. This means that despite the remarkable contribution of NAC in BLA-PFC coher- ence, the BLA–PFC coherence is mainly a result of either direct communication between BLA and PFC or indirect via other brain regions, which were not recorded in our experiments. Finally, we removed the BLA to measure its influence on the coherency between NAC and PFC (Fig. 4b, middle panel). We found that the peak of partial coherence in the NAC–PFC pair decreased by approximately 50% from 0.35 ± 0.21 to 0.17 ± 0.13. This finding demonstrates the important implication of BLA in synchronizing the NAC and PFC areas. However, the NAC–PFC coherency cannot fully be explained by BLA influence, as it remained sig- nificant even after removing the BLA influence. Similar to BLA–NAC–PFC recordings, we computed partial coherence for the vHC–NAC and vHC–PFC pairs by excluding the PFC and NAC influence, respectively (Fig. 4b, right panels). We Ordinary and partial coherence of LFPs during spontaneous activity Next, we asked whether and how areas of the cortico-limbic network are also correlated in their LFP activity (Fig. 4). To address this question, we measured the coherency for BLA–NAC, BLA–PFC, NAC–PFC (n = 9 animals), and vHC–NAC and vHC–PFC (n = 7) for each mouse and then averaged it across all animals (Fig. 4a). We found that BLA and NAC are significantly coherent at ~ 0.5 Hz (0.35 ± 0.17, mean ± SD; Fig. 4a, left). Similarly, BLA and PFC showed a significant coherence with similar pattern as that observed Fig. 4 Ordinary coherence and partial coherence in the BLA-NAC- PFC and the vHC–NAC–PFC network. a Ordinary coherence plots for five tested pairs. Average of ordinary coherence in BLA-NAC (n = 9), BLA–PFC (n = 9), NAC–PFC (n = 9), vHC-NAC (n = 7) and vHC–PFC (n = 7). b Same as in a, but for partial coherence after removing the influence of another brain region. Note changes in peak coherence as compared to plots in (a). In all ordinary and par- tial coherence plots solid lines show the means and shaded areas show the standard deviations of coherence averaged across all mice. Dashed lines show the confidence limit of significance for coherence removing the influence of another brain region. Note changes in peak coherence as compared to plots in (a). In all ordinary and par- tial coherence plots solid lines show the means and shaded areas show the standard deviations of coherence averaged across all mice. Dashed lines show the confidence limit of significance for coherence Fig. 4 Ordinary coherence and partial coherence in the BLA-NAC- PFC and the vHC–NAC–PFC network. a Ordinary coherence plots for five tested pairs. Average of ordinary coherence in BLA-NAC (n = 9), BLA–PFC (n = 9), NAC–PFC (n = 9), vHC-NAC (n = 7) and vHC–PFC (n = 7). b Same as in a, but for partial coherence after 1 3 Brain Structure and Function (2021) 226:685–700 693 in BLA–NAC. The maximum coherence of BLA–PFC occurred at 0.5 Hz (0.4 ± 0.23, Fig. 4a, second panel from left). We also found a significant coherency in NAC–PFC, which tended to be strongest at lower frequencies with a peak occurring at 0.5 Hz (0.35 ± 0.21; Fig. 4a, middle). In vHC–NAC–PFC recordings, vHC and NAC showed the maximum coherency at 1 Hz (0.16 ± 0.13, Fig. 4a, second plot from right). Time resolved partial directed coherence during spontaneous activity b Comparison between the total tPDC values for vHC-NAC (left), vHC-PFC (middle) and NAC-PFC (right) in theta band averaged across 7 mice aged across all mice (n = 9). b Comparison between the total tPDC values for vHC-NAC (left), vHC-PFC (middle) and NAC-PFC (right) in theta band averaged across 7 mice Fig. 5 Comparison of areas’ interactions in the BLA-NAC-PFC (a) and vHC-NAC-PFC (b) recording configuration for theta activity only. a Comparison between the total tPDC values for BLA-NAC (left), BLA-PFC (middle) and NAC-PFC (right) in theta band aver- aged across all mice (n = 9). b Comparison between the total tPDC values for vHC-NAC (left), vHC-PFC (middle) and NAC-PFC (right) in theta band averaged across 7 mice with vHC and NAC, such that in all frequency bands except for theta the tPDC values for total PFC inputs was signifi- cantly lower than total PFC output (ANOVA, main effect, p = 0.01 [delta], p < 10–3 [alpha], p < 10–3 [beta], Fig. 6e, g, h). In the theta band the tPDC values for vHC to PFC were significantly higher than the tPDC values for PFC to vHC (ANOVA, direct extension of main effect, p = 0.01, Fig. 7f). In contrast, in the other frequency bands, there was no sig- nificant difference between the strength of information flow for vHC to PFC and for PFC to vHC (Fig. 7e, g, h). Finally, in all frequencies the tPDC values for PFC to NAC were significantly higher than NAC to PFC (ANOVA, direct extension of main effect, p < 10–4 [delta], p < 10–4 [theta], p < 10–4 [alpha], p < 10–3 [beta], Fig. 7e–h). In summary, these results indicate direct connectivity between vHC, NAC and PFC with a clear frequency-specific role for vHC. In the vHC–NAC–PFC network, vHC act as network driver only in theta band, whereas in the other frequency bands more an interplay between these nodes was found. total input to the PFC (p > 0.05, Fig. 6a–d). However, the PFC showed a region-specificity in the direction of infor- mation flow for interacting with NAC and BLA (Fig. 7a–d). Indeed, the tPDC values for PFC to BLA were significantly lower than tPDC values for BLA to PFC (ANOVA, direct extension of main effect, p < 10–3 [delta], p < 10–4 [theta], p < 10–5 [alpha] and p < 10–5 [beta], Fig. 7a–d). Time resolved partial directed coherence during spontaneous activity In contrast, the tPDC values for PFC to NAC were significantly higher than the tPDC values for NAC to PFC (ANOVA, direct extension of main effect, p = 0.01 [delta] and p < 10–3[alpha], Fig. 7a, c). These results not only demonstrate the direct connectivity between BLA, NAC and PFC in distinctive frequency bands, but also suggest a target-specificity in the strength of information flow between these regions. Indeed, in the BLA–NAC–PFC network, BLA strongly transmits information to the other regions, while NAC predominantly acts as a target region and PFC regulates BLA–NAC–PFC network. In the vHC–NAC–PFC network every single area inter- acted with two others without any significant region-prefer- ences, as shown exemplary in Fig. 5b for the theta frequency. We found that the tPDC values for vHC outputs in theta band were significantly higher than the tPDC values for vHC inputs (ANOVA, main effect, p < 10–3; Fig. 6f). However, there was no significant difference for vHC interactions with NAC or with PFC in other frequency bands (Fig. 7e–h). In all frequencies we found a significantly higher informa- tion flow in the inputs to NAC as compared to the outputs from NAC (ANOVA, main effect, p < 10–4 [delta], p < 10–3 [theta], p < 10–4 [alpha], p < 10–3 [beta], Fig. 6e–h). This directionality in NAC interaction with two other regions was region-specific only in delta band where the tPDC values for PFC to NAC were significantly higher than the tPDC values for vHC to NAC (ANOVA, direct extension of main effect, p = 0.02; Fig. 7e). Similar to vHC and NAC, we found a directionality in information flow in interaction of PFC Time resolved partial directed coherence during spontaneous activity We used tPDC to quantify the directed connectivity in five different frequency bands for both the BLA-NAC-PFC and the vHC-NAC-PFC network (Figs. 5, 6 and 7). Since the LFP power in the gamma range was rather small, we focused on delta to beta frequencies. In the BLA-NAC-PFC network, every single area interacted with the two others without any significant region-preferences. This is illustrated in Fig. 5a exemplary for the activity in the theta band. However, when we took the direction of information flow into account, we observed significant direction preferences in informa- tion flow for all three regions in multiple frequencies. We found that in all frequencies bands the tPDC values for BLA outputs were significantly higher than the tPDC values for BLA inputs (ANOVA, main effect, p < 10–7 [delta], p < 10–6 [theta], p < 10–8 [alpha] and p < 10–9 [beta], Fig. 6a–d). This directionality of the BLA in interaction with others, how- ever, was not region specific, as the tPDC values for BLA to PFC were comparable to tPDC values for BLA to NAC and tPDC values for NAC to BLA were similar to tPDC values for PFC to BLA (p > 0.05, Fig. 7a–d). The directionality pat- tern of NAC interactions with others was opposite to what we observed in the BLA, as the strength of information flow from NAC to BLA and PFC was significantly weaker than the strength of information flow from BLA and PFC to NAC (ANOVA, main effect, p < 10–6 [delta], p < 10–4 [theta], p < 10–6 [alpha] and p < 10–5 [beta], Fig. 6a–d). Similar to the BLA, we did not find a significant region specificity in direc- tionality neither in inputs to NAC from BLA and PFC, nor in the outputs from NAC to BLA and PFC (p > 0.05, Fig. 7a-d). In the PFC, the total PFC output was comparable with the 1 1 3 3 Brain Structure and Function (2021) 226:685–700 694 Fig. 5 Comparison of areas’ interactions in the BLA-NAC-PFC (a) and vHC-NAC-PFC (b) recording configuration for theta activity only. a Comparison between the total tPDC values for BLA-NAC (left), BLA-PFC (middle) and NAC-PFC (right) in theta band aver- aged across all mice (n = 9). Discussion Information flow for NAC (middle) and for PFC (right). In all panels represents p < 0.05, p < 0.001, p < 10–4, *p < 10–5, * p < 10–6, **p < 10–7, ***p < 10–8 and *****p < 10–9 Fig. 7 Region-specific directionality of information flow in different frequency bands in the BLA-NAC-PFC (left) and vHC-NAC-PFC network (right). For further details, see Fig. 6 and text Fig. 7 Region-specific directionality of information flow in different frequency bands in the BLA-NAC-PFC (left) and vHC-NAC-PFC network (right). For further details, see Fig. 6 and text Fig. 6 Comparison of inward and outward information flow in BLA, vHC, NAC and PFC in all frequency bands in the BLA-NAC-PFC (left, n = 9 mice) and vHC-NAC-PFC recording configuration (right, n = 7). a–d Information flow for BLA (left) and NAC (middle) in delta (a), theta (b), alpha (c) and beta band (d). Information flow for PFC in (right, all p > 0.05). e–h, inward and outward information flow for vHC in all frequency bands in the vHC-NAC-PFC recording configuration (left). Information flow for NAC (middle) and for PFC (right). In all panels represents p < 0.05, p < 0.001, p < 10–4, *p < 10–5, * p < 10–6, **p < 10–7, ***p < 10–8 and *****p < 10–9 Fig. 7 Region-specific directionality of information flow in different frequency bands in the BLA-NAC-PFC (left) and vHC-NAC-PFC network (right). For further details, see Fig. 6 and text Discussion Using MEAs, we recorded in lightly anesthetized adult mice the spontaneous activity simultaneously at multiple sites of the cortico-limbic network and analyzed the direction and the strength of the functional interactions by spike count cross-correlations and by time-resolved partial directed coherence (tPDC) calculations. Given our light anesthesia protocol for which we needed to use only one-third of the common dosage of urethane without repetition, we assume that the adverse effects of anesthesia on our results are rather small. This assumption is supported by the similarity of our results with previous observations obtained in awake rodents (see below). Our results demonstrate that BLA, NAC, PFC and vHC, as the key structures of the cortico-limbic network, 1 3 3 Brain Structure and Function (2021) 226:685–700 695 B a St uctu e a d u ct o ( 0 ) 6:685 00 Fig. 6 Comparison of inward and outward information flow in BLA, vHC, NAC and PFC in all frequency bands in the BLA-NAC-PFC (left, n = 9 mice) and vHC-NAC-PFC recording configuration (right, n = 7). a–d Information flow for BLA (left) and NAC (middle) in delta (a), theta (b), alpha (c) and beta band (d). Information flow for PFC in (right, all p > 0.05). e–h, inward and outward information flow for vHC in all frequency bands in the vHC-NAC-PFC recording configuration (left). Information flow for NAC (middle) and for PFC (right). In all panels represents p < 0.05, p < 0.001, p < 10–4, *p < 10–5, * p < 10–6, **p < 10–7, ***p < 10–8 and ******p < 10–9 are highly interconnected and synchronized during their l Region‑specific spontaneous activity All four brain regions revealed the strongest spontaneous activity in the delta and theta frequency range and in the Fourier spectrum a gradual power decrease from alpha over beta to gamma (Fig. 2). Although the presence of delta f Fig. 6 Comparison of inward and outward information flow in BLA, vHC, NAC and PFC in all frequency bands in the BLA-NAC-PFC (left, n = 9 mice) and vHC-NAC-PFC recording configuration (right, n = 7). a–d Information flow for BLA (left) and NAC (middle) in delta (a), theta (b), alpha (c) and beta band (d). Information flow for PFC in (right, all p > 0.05). e–h, inward and outward information flow for vHC in all frequency bands in the vHC-NAC-PFC recording configuration (left). Functional connectivity in cortico‑limbic network In agreement with previous observations, all five pairs that we examined in the present study showed synchrony in multiple frequencies [for review (Bocchio et al. 2017)]. As shown previously (Dzirasa et al. 2011a, b; Hultman et al. 2016), we identified the delta and theta band as the domi- nant synchrony rhythms in the BLA–NAC–PFC network. Moreover, our results provide evidence for the contribution of a third brain region on the pair synchrony of two con- nected areas. Specifically, more than half of the observed coherence in the BLA–NAC is mediated by the common PFC input, indicating a powerful impact of PFC in enhanc- ing the BLA–NAC synchrony. A similar influence of PFC in synchronizing BLA and NAC was observed for the vHC- NAC pair in vHC–NAC–PFC recordings, albeit at a smaller intensity (31% vs. 60%), such that more than one-third of the vHC–NAC synchrony magnitude could be explained by the PFC influence. Our results also revealed the implication of NAC in cortico-limbic synchronization, although its magni- tude differs for different paired areas. While NAC accounts for a remarkable amount of the BLA–PFC coherence, its influence is less pronounced in vHC-PFC synchrony such that the majority of vHC–PFC synchrony (~ 70%) is pre- served after removing the NAC contribution. Given this finding and previous observations (Siapas et al. 2005; Wier- zynski et al. 2009; Godsil et al. 2013; Padilla-Coreano et al. 2016; Liu and Carter, 2018), it is most likely that vHC and PFC synchrony is mediated through a direct connectivity, although the role of an unrecorded area cannot be excluded. In contrast to vHC-PFC, NAC-PFC synchrony is remarkably dependent on the input from a third connected area. Indeed, nearly half of the observed coherence in NAC-PFC could be explained by BLA influence. Altogether, our results indi- cate that all four regions contribute in vHC-BLA-NAC-PFC Region‑specific spontaneous activity All four brain regions revealed the strongest spontaneous activity in the delta and theta frequency range and in the Fourier spectrum a gradual power decrease from alpha over beta to gamma (Fig. 2). Although the presence of delta activity is a strong indicator for the effects of anesthesia, delta and theta activity are also prominent rhythms of the hippocampus and prefrontal cortex of awake rats during a working-memory task (Fujisawa and Buzsáki 2011). Hip- pocampal respiration rhythms between 2 and 12 Hz and driven by the olfactory bulb have been demonstrated in awake mice (Nguyen et al. 2016). So-called type 2 theta activity (5–9 Hz), also known as "immobility theta", is resistant to urethane and is a dominant hippocampal rhythm during aroused immobility (Sainsbury and Montoya 1984). are highly interconnected and synchronized during their spontaneous activity. Determining the influence of each region in network synchrony and measuring the strength of feedforward and feedback connections, we identified the PFC as a powerful regulator for network synchrony, the BLA as a major source of information flow, NAC as a hub for converging information flow from other network struc- tures and vHC as a main controller for PFC in theta band. These findings may contribute to a better understanding of the physiological and pathophysiological interactions in the cortico-limbic network. 1 Brain Structure and Function (2021) 226:685–700 696 presentation or during performing a task (Cohen and Kohn 2011). In our data we observed a high variability in the lag time of cross-correlations between different pairs. Therefore, we decided to obtain the directionality information of con- nections using tPDC analysis. Although spike cross-corre- lation has been used extensively for functional connectivity analysis and in some cases for inferring the directionality of connections, it might not be an informative method of analy- sis to reveal the direction of connection across areas with bidirectional connectivity, which is the case for the brain regions that we examined in our study (Russo and Nestler 2013). As Bastos and Schoffelen (2016) discuss in their review article, the spike cross-correlation analysis could be informative for those regions with a dominant unidirectional connectivity (e.g. the retino-geniculate feedforward path- way; see Usrey et al. 1998), while this might not be the case for regions with a bidirectional connectivity as in our study. Region‑specific spontaneous activity Spontaneous oscillations in the delta and theta range have been previously recorded in the amygdala of urethane anes- thetized and behaving mice, respectively (Seidenbecher et al. 2003; Bazelot et al. 2015) [for review (Bocchio et al. 2017)]. Furthermore, single units in BLA, vHC and PFC of behaving rats fire at frequencies below 4 Hz (Sotres-Bayon et al. 2012). In the NAC of awake freely-moving mice spon- taneous activity is in the theta-delta range with a peak at 1–2 Hz (Luo et al. 2018). We noticed a difference in the PFC LFP power spectrum between the BLA–NAC–PFC and vHC–NAC–PFC experimental condition. We can rule out an anesthesia effect, because in both conditions animals were anesthetized in the same way and showed comparable breathing rates (data not shown). Although, we cannot pro- vide a full explanation for this difference, we speculate that it might be due to layer-specific differences in PFC. A layer- specific LFP power has been reported by Bastos et al. (2018) in frontal cortex recordings of monkeys. Their results show that the LFP power in frequencies below 20 Hz is stronger in deep layers than in superficial layers. Given this find- ing, we speculate that the LFPs in BLA–NAC–PFC versus vHC–NAC–PFC condition may have been recorded in dif- ferent layers of PFC. We cannot fully exclude that LFPs in BLA–NAC–PFC were mostly recorded in superficial layers whereas LFPs in the vHC–NAC–PFC condition mostly in deep layers of PFC. However, our data on the spontane- ous LFP activity in BLA, NAC, vHC and PFC are in good agreement with these previous observations, supporting the assumption that the light urethane anesthesia used in our experiments has only mild effects on the spontaneous net- work activity. Directed connectivity in cortico‑limbic network In addition to describing mutual synchronicity, we deter- mined the direction and strength of information flow for each pair in the cortico-limbic network (Fig. 8). For all five pairs, we found reciprocal connections, which is consistent with previous studies (Little and Carter 2013; Senn et al. 2014; Hultman et al. 2016; Reppucci and Petrovich 2016; Klavir et al. 2017; McGarry and Carter 2017). However, the strength of the information flow, depending on target area, was weighted toward either directions. In particular, BLA appeared to be as the principle source of information flow in the BLA-NAC-PFC network, as the connectivity from BLA to NAC and PFC was remarkably stronger than that from NAC and PFC to BLA. These results might be sound in contrast to our results in coherence, where PFC has been suggested as a key structure in BLA-NAC-PFC synchrony. However, it should be considered that our results in mutual synchrony can only describe the existence of synchrony for each pair and determine the contribution of a third recorded area (Muthuraman et al. 2008) and the direction of informa- tion flow cannot be resolved by relying merely on coherence analysis. In contrast, tPDC analysis provides the comple- mentary information and particularly the direction of con- nection (Muthuraman et al. 2018). Unlike vHC-PFC, we did not find any frequency-specific difference directionality for vHC-NAC communication, as the strength of connection from vHC to NAC was compara- ble to the strength of connection from NAC to vHC. In the delta band, NAC was influenced mainly by PFC rather than vHC, which is in line with the known role of the PFC slow wave in regulating limbic areas (Karalis et al. 2016; Hultman et al. 2016). In conclusion, our results provide insights to the functional connectivity in the cortico-limbic network by determining the strength and direction of information flow within this network and implicating PFC as a key structure. Unlike BLA, the NAC implication in this network seems to be more as a “receiver”, as the strength of NAC output to BLA and PFC was much weaker than the inputs from BLA and PFC to NAC. This finding is consistent with previous observations on NAC innervation (Phillipson and Griffiths 1985). Unlike NAC and BLA, PFC behaves differently in its interaction with NAC and BLA, functioning as a “donor” in communicating with NAC and as a “recipient” in commu- nicating with BLA. Correlation of spontaneous activity between two brain regions A number of our observations are in good agreement with previous reports confirming strong interactions between specific brain regions within the cortico-limbic network. Furthermore, we provide novel insights into the functional connectivity in different frequencies ranging from delta to beta. Using spike count cross-correlation, we determined the coupling strengths between two connected brain regions. Consistent with previous studies (Pitkänen et al. 2000; Sei- denbecher et al. 2003; Hoover and Vertes 2007; Stujenske et al. 2014; Hultman et al. 2016; McGarry and Carter 2017; Burgos-Robles et al. 2017), all five pairs that we tested were closely connected and revealed spike-count correlations. The rsc in our results compared to previous studies that measured spike count correlations (Cohen and Newsome 2008; Leavitt et al. 2013; Taub et al. 2018) were smaller. However, these previous studies measured the correlation between neurons located in the same brain region, close to each other, and having similar stimulus preferences in response to stimulus 1 3 3 Brain Structure and Function (2021) 226:685–700 697 regulating limbic dynamics (Hultman et al. 2016; McGarry and Carter 2017), although our study cannot answer this question directly. network synchrony and PFC slow waves are the key player in this synchronization (Karalis et al. 2016; Hultman et al. 2016). Our results in cortico-limbic network show that vHC is highly interconnected with PFC and that this connection is bidirectional. Although this is in contrast with the thoughts of unidirectional connectivity from vHC-PFC (Xu et al. 2019), the interplaying of hippocampus and PFC in memory has been also reported by other studies (Preston and Eichen- baum 2013), supporting bidirectional connectivity in vHC- PFC. Although the strength of information flow from vHC to PFC was stronger than PFC to vHC in theta rhythm, for other frequencies (delta, alpha and beta) there was no signifi- cant difference between information flow from vHC to PFC and from PFC to vHC, indicating bidirectional connectivity for vHC and PFC, at least in LFP activity. This is consistent with the previous report on interplaying hippocampus and PFC in memory (Preston and Eichenbaum, 2013).ii Directed connectivity in cortico‑limbic network This finding is in line with our partial coherence results where the 60% of BLA–NAC synchrony is mediated by PFC. Our findings, therefore, might be support- ing evidence for the established regulatory role of PFC in Conflicts of interest/competing interest The authors declare no con- flict of interest. Conflicts of interest/competing interest The authors declare no con- flict of interest. Ethical approval All procedures related to the care and treatments of animals were approved by a local ethics committee (#23 177-07/G 14-1-080) and followed the German and European national regulations (European Communities Council Directive, 86/609/EEC). Open Access This article is licensed under a Creative Commons Attri- bution 4.0 International License, which permits use, sharing, adapta- tion, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Conclusion In conclusion, we have characterized the functional connec- tivity in the cortico-limbic network of lightly anesthetized wild-type mice. Performing simultaneous recordings in mul- tiple brain regions by using multi-channels probes, we found that BLA, NAC, PFC and vHC are highly interconnected and synchronized in their spontaneous activity. We also determined the implication of each region in synchroniz- ing the other areas of the network. Our results revealed the powerful role of the PFC in synchronizing the cortico-limbic network, which is consistent with previous findings. Given the observed bidirectional connectivity in all five tested pairs, we measured the strength of both feedforward and feedback connections by applying a time-resolved partial directed coherence analyses, as a powerful computational tool, to our data. We found that in the BLA–NAC–PFC network, BLA serves as a principal source of information flow, NAC functions more as a “recipient” and PFC acts as a “donor” for NAC and as a “recipient” for BLA. We also found that in the theta band the strength of information flow from vHC to PFC is stronger than in the opposite direc- tion. Our results obtained in lightly anesthetized wild-type mice provide further information on the physiological con- nectivity of the cortico-limbic network and represent the basis to study alterations in this network in animal models of psychiatric disorders. We hypothesize that the connectiv- ity between these regions may be altered in mouse models of psychiatric disorders (Hultman et al. 2016; Karalis et al. 2016; Padilla-Coreano et al. 2019). Limitations Third, we did not include gamma band activity in our tPDC analysis, as activities in the gamma range were rather small. However, previous studies in mice demonstrated gamma band alterations and impli- cations in anxiety-related disorders (Popescu et al. 2009; Dzirasa et al. 2011b). Using our experimental approach, the role of gamma rhythms in the cortico-limbic network may be elucidated in behaviorally challenged animals. Fourth, we did not investigate the architecture of func- tional connections, which would have provided comple- mentary insights into specific interaction between areas. Indeed, using a correlation matrix the topology of func- tional connections can be examined and a graph theory (Rubinov and Sporns 2010) can be employed to describe the connection organizations in the network. Acknowledgments We thank Beate Krumm for excellent help with his- tology. This work was supported by grants of the Deutsche Forschun- gsgemeinschaft to S.G. (CRC 1193, B05) and to H.J.L. (CRC 1193, B04). Acknowledgments We thank Beate Krumm for excellent help with his- tology. This work was supported by grants of the Deutsche Forschun- gsgemeinschaft to S.G. (CRC 1193, B05) and to H.J.L. (CRC 1193, B04). Author contributions Conceptualization: MM, SG, HJL; methodology: ZK, MM, J-WY, SG, HJL; performing experiments: ZK, J-WY; data analysis: ZK, MM; writing manuscript: ZK, MM, SG, HJL; supervi- sion: MM, SG, HJL. Funding Open Access funding enabled and organized by Projekt DEAL. This work was supported by grants of the Deutsche Forschun- gsgemeinschaft to S.G. (CRC 1193, B05) and to H.J.L. (CRC 1193, B04). Limitations Our results in cortico-limbic functional connectivity in lightly anesthetized mice have a few limitations. First, although our anesthesia protocol consisting of a combina- tion of a very low dose of urethane and adding a sedative was designed to reduce the adverse effects of urethane on brain state as much as possible, the influence of even this light anesthesia on our results should not be neglected and therefore a direct translation from our results to awake state is difficult. Second, we did not study the BLA-vHC Fig. 8 Summary diagram of directed functional connectivity in the cortico-limbic network for spontaneous activity in the delta (a), theta (b), alpha (c) and beta (d) frequency band. Thickness of arrows indicates normalized values of tPDC strength (see scale right) Fig. 8 Summary diagram of directed functional connectivity in the cortico-limbic network for spontaneous activity in the delta (a), theta (b), alpha (c) and beta (d) frequency band. Thickness of arrows indicates normalized values of tPDC strength (see scale right) 1 3 3 698 Brain Structure and Function (2021) 226:685–700 communication, due to technical difficulties for record- ing these two areas simultaneously using our multi-site MEA approach. However, the interaction of BLA and vHC and its relevant role in emotion, cognition and psychiatric disorders has been probed in previous studies [for review (Bocchio et al. 2017)]. Third, we did not include gamma band activity in our tPDC analysis, as activities in the gamma range were rather small. However, previous studies in mice demonstrated gamma band alterations and impli- cations in anxiety-related disorders (Popescu et al. 2009; Dzirasa et al. 2011b). Using our experimental approach, the role of gamma rhythms in the cortico-limbic network may be elucidated in behaviorally challenged animals. Fourth, we did not investigate the architecture of func- tional connections, which would have provided comple- mentary insights into specific interaction between areas. Indeed, using a correlation matrix the topology of func- tional connections can be examined and a graph theory (Rubinov and Sporns 2010) can be employed to describe the connection organizations in the network. communication, due to technical difficulties for record- ing these two areas simultaneously using our multi-site MEA approach. However, the interaction of BLA and vHC and its relevant role in emotion, cognition and psychiatric disorders has been probed in previous studies [for review (Bocchio et al. 2017)]. References Anwar AR, Muthalib M, Perrey S, Galka A, Granert O, Wolff S, Heute U, Deuschl G, Raethjen J, Muthuraman M (2016) Effec- tive connectivity of cortical sensorimotor networks during finger movement tasks: a simultaneous fNIRS, fMRI, EEG Study. Brain Topogr 29:645–660 Baccala LA, Sameshima K (2001) Partial directed coherence: a new concept in neural structure determination. Biol Cybern 84:463–474f Bastos AM, Schoffelen JM (2016) A tutorial review of functional con- nectivity analysis methods and their interpretational pitfalls. Front Syst Neurosci 9:175 Bastos AM, Loonis R, Kornblith S, Lundqvist M, Miller EK (2018) Laminar recordings in frontal cortex suggest distinct lay- ers for maintenance and control of working memory. PNAS 115(5):1117–1122 Bazelot M, Bocchio M, Kasugai Y, Fischer D, Dodson PD, Ferraguti F, Capogna M (2015) Hippocampal theta input to the amygdala shapes feedforward inhibition to gate heterosynaptic plasticity. Neuron 87:1290–1303 Beyeler A, Chang CJ, Silvestre M, Leveque C, Namburi P, Wildes CP, Tye KM (2018) Organization of valence-encoding and 1 3 Brain Structure and Function (2021) 226:685–700 699 projection-defined neurons in the basolateral amygdala. Cell Rep 22:905–918 Haufe S, Nikulin VV, Müller KR, Nolte G (2013) A critical assess- ment of connectivity measures for EEG data: a simulation study. Neuroimage 64:120–133 Bocchio M, Nabavi S, Capogna M (2017) Synaptic plasticity, engrams, and network oscillations in amygdala circuits for storage and retrieval of emotional memories. Neuron 94:731–743i Hillary FG, Grafman JH (2017) Injured brains and adaptive networks: the benefits and costs of hyperconnectivity. Trends Cogn Sci 21:385–401f Bonnefil V, Dietz K, Amatruda M, Wentling M, Aubry AV, Dupree JL, Temple G, Park HJ, Burghardt NS, Casaccia P, Liu J (2019) Region-specific myelin differences define behavioral conse- quences of chronic social defeat stress in mice. eLife 8:e40855l Hoover WB, Vertes RP (2007) Anatomical analysis of afferent projec- tions to the medial prefrontal cortex in the rat. Brain Struct Funct 212:149–179 Hultman R, Mague SD, Li Q, Katz BM, Michel N, Lin L, Wang J, David LK, Blount C, Chandy R, Carlson D, Ulrich K, Carin L, Dunson D, Kumar S, Deisseroth K, Moore SD, Dzirasa K (2016) Dysregulation of prefrontal cortex-mediated slow-evolv- ing limbic dynamics drives stress-induced emotional pathology. Neuron 91:439–452 Bukhari Q, Schroeter A, Rudin M (2018) Increasing isoflurane dose reduces homotopic correlation and functional segregation of brain networks in mice as revealed by resting-state fMRI. References Sci Rep 8:10591 p Burgos-Robles A, Kimchi EY, Izadmehr EM, Porzenheim MJ, Ramos- Guasp WA, Nieh EH, Felix-Ortiz AC, Namburi P, Leppla CA, Presbrey KN, Anandalingam KK, Pagan-Rivera PA, Anahtar M, Beyeler A, Tye KM (2017) Amygdala inputs to prefrontal cortex guide behavior amid conflicting cues of reward and punishment. Nat Neurosci 20:824–835 Hultman R, Ulrich K, Sachs BD, Blount C, Carlson DE, Ndubuizu N, Bagot RC, Parise EM, Vu MT, Gallagher NM, Wang J, Silva AJ, Deisseroth K, Mague SD, Caron MG, Nestler EJ, Carin L, Dzirasa K (2018) Brain-wide electrical spatiotemporal dynam- ics encode depression vulnerability. Cell 173:166–180 Cagnan H, Mallet N, Moll CKE, Gulberti A, Holt AB, Westphal M, Gerloff C, Engel AK, Hamel W, Magill PJ, Brown P, Sharott A (2019) Temporal evolution of beta bursts in the parkinsonian cortical and basal ganglia network. Proc Natl Acad Sci USA 116:16095–16104 Kaminski M, Ding M, Truccolo WA, Bressler SL (2001) Evaluating causal relations in neural systems: granger causality, directed transfer function and statistical assessment of significance. Biol Cybern 85:145–157 Cohen MR, Newsome WT (2008) Context-dependent changes in func- tional circuitry in visual area MT. Neuron 60:162–173 Karalis N, Dejean C, Chaudun F, Khoder S, Rozeske RR, Wurtz H, Bagur S, Benchenane K, Sirota A, Courtin J, Herry C (2016) 4-Hz oscillations synchronize prefrontal-amygdala circuits dur- ing fear behavior. Nat Neurosci 19:605 Cohen MR, Kohn A (2011) Measuring and interpreting neuronal cor- relations. Nat Neurosci 14:811–819 Dickerson DD, Restieaux AM, Bilkey DK (2012) Clozapine admin- istration ameliorates disrupted long-range synchrony in a neu- rodevelopmental animal model of schizophrenia. Schizophr Res 135:112–115 Klavir O, Prigge M, Sarel A, Paz R, Yizhar O (2017) Manipulating fear associations via optogenetic modulation of amygdala inputs to prefrontal cortex. Nat Neurosci 20:836–844 Leavitt ML, Pieper F, Sachs A, Joober R, Martinez-Trujillo JC (2013) Structure of spike count correlations reveals functional interactions between neurons in dorsolateral prefrontal cortex area 8a of behaving primates. PLoS ONE 8:e61503i Dzirasa K, Fuentes R, Kumar S, Potes JM, Nicolelis MAL (2011a) Chronic in vivo multi-circuit neurophysiological recordings in mice. J Neurosci Methods 195:36–46 Dzirasa K, McGarity DL, Bhattacharya A, Kumar S, Takahashi JS, Dunson D, McClung CA, Nicolelis MA (2011b) Impaired limbic gamma oscillatory synchrony during anxiety-related behavior in a genetic mouse model of bipolar mania. References J Neurosci 31:6449–6456 Leistritz L, Pester B, Doering A, Schiecke K, Babiloni F, Astolfi L, Witte H (2013) Time-variant partial directed coherence for analysing connectivity: a methodological study. Philos Trans A Math Phys Eng Sci 371:20110616 Little JP, Carter AG (2013) Synaptic mechanisms underlying strong reciprocal connectivity between the medial prefrontal cortex and basolateral amygdala. J Neurosci 33:15333–15342 Fornito A, Bullmore ET (2015) Connectomics: a new paradigm for understanding brain disease. Eur Neuropsychopharmacol 25:733–748 Fujisawa S, Buzsáki G (2011) A 4 Hz oscillation adaptively syn- chronizes prefrontal, VTA, and hippocampal activities. Neuron 72:153–165 Liu X, Carter AG (2018) Ventral hippocampal inputs preferentially drive corticocortical neurons in the infralimbic prefrontal cor- tex. J Neurosci 38:7351–7363 Luo YJ, Li YD, Wang L, Yang SR, Yuan XS, Wang J, Cherasse Y, Lazarus M, Chen JF, Qu WM, Huang ZL (2018) Nucleus accumbens controls wakefulness by a subpopulation of neurons expressing dopamine D1 receptors. Nat Commun 9:1576 Gangarossa G, Espallergues J, de Kerchove DA, El Mestikawy S, Ger- fen CR, Herve D, Girault JA, Valjent E (2013) Distribution and compartmental organization of GABAergic medium-sized spiny neurons in the mouse nucleus accumbens. Front Neural Circuits 7:22 McGarry LM, Carter AG (2017) Prefrontal cortex drives distinct projection neurons in the basolateral amygdala. Cell Rep 21:1426–1433 Genzel L, Dresler M, Cornu M, Jager E, Konrad B, Adamczyk M, Friess E, Steiger A, Czisch M, Goya-Maldonado R (2015) Medial prefrontal-hippocampal connectivity and motor memory consolidation in depression and schizophrenia. Biol Psychiatry 77:177–186 Muthuraman M, Govindan RB, Deuschl G, Heute U, Raethjen J (2008) Differentiating phase shift and delay in narrow band coherent signals. Clin Neurophysiol 119:1062–1070 Godsil BP, Kiss JP, Spedding M, Jay TM (2013) The hippocampal- prefrontal pathway: the weak link in psychiatric disorders? Eur Neuropsychopharmacol 23:1165–1181 Muthuraman M, Raethjen J, Koirala N, Anwar AR, Mideksa KG, Elble R, Groppa S, Deuschl G (2018) Cerebello-cortical net- work fingerprints differ between essential, Parkinson’s and mim- icked tremors. Brain 141:1770–1781 Govindan RB, Raethjen J, Arning K, Kopper F, Deuschl G (2006) Time delay and partial coherence analyses to identify cortical connec- tivities. Biol Cybern 94:262–275 Nguyen C, Muller VC, Wolfenstetter T, Yanovsky Y, Draguhn A, Tort AB, Brankack J (2016) Hippocampal respiration-driven rhythm distinct from theta oscillations in awake mice. References J Neu- rosci 36:162–177 Halliday DM, Rosenberg JR, Amjad AM, Breeze P, Conway BA, Farmer SF (1995) A framework for the analysis of mixed time series/point process data–theory and application to the study of physiological tremor, single motor unit discharges and electro- myograms. Prog Biophys Mol Biol 64:237–278 Paasonen J, Stenroos P, Salo RA, Kiviniemi V, Grohn O (2018) Functional connectivity under six anesthesia protocols and the awake condition in rat brain. Neuroimage 172:9–20 1 3 700 Brain Structure and Function (2021) 226:685–700 Sesack SR, Grace AA (2010) Cortico-Basal Ganglia reward network: microcircuitry. Neuropsychopharm 35:27–47 Padilla-Coreano N, Bolkan SS, Pierce GM, Blackman DR, Hardin WD, Garcia-Garcia AL, Spellman TJ, Gordon JA (2016) Direct ventral hippocampal-prefrontal input is required for anxiety- related neural activity and behavior. Neuron 89:857–866 y p y p Siapas AG, Lubenov EV, Wilson MA (2005) Prefrontal phase locking to hippocampal theta oscillations. Neuron 46:141–151 Simons JS, Spiers HJ (2003) Prefrontal and medial temporal lobe inter- actions in long-term memory. Nat Rev Neurosci 4:637–648 Padilla-Coreano N, Canetta S, Mikofsky RM, Alway E, Passecker J, Myroshnychenko MV, Garcia-Garcia AL, Warren R, Teboul E, Blackman DR, Morton MP, Hupalo S, Tye KM, Kellendonk C, Kupferschmidt DA, Gordon JA (2019) Hippocampal-pre- frontal theta transmission regulates avoidance behavior. Neuron 104:601-610.e604 Sinha N, Dauwels J, Kaiser M, Cash SS, Brandon WM, Wang Y, Taylor PN (2017) Predicting neurosurgical outcomes in focal epilepsy patients using computational modelling. Brain 140:319–332 Sotres-Bayon F, Sierra-Mercado D, Pardilla-Delgado E, Quirk GJ (2012) Gating of fear in prelimbic cortex by hippocampal and amygdala inputs. Neuron 76:804–812 Pagnotta MF, Plomp G (2018) Time-varying MVAR algorithms for directed connectivity analysis: critical comparison in simulations and benchmark EEG data. PLoS ONE 13:e0198846 Stujenske JM, Likhtik E, Topiwala MA, Gordon JA (2014) Fear and safety engage competing patterns of theta-gamma coupling in the basolateral amygdala. Neuron 83:919–933 and benchmark EEG data. PLoS ONE 13:e0198846 Paxinos G, Franklin KBJ (2001) The mouse brain in stereotaxic coor- dinates. Academic Press, San Diegofi Phillipson OT, Griffiths AC (1985) The topographic order of inputs to nucleus accumbens in the rat. Neuroscience 16:275–296 Taub AH, Perets R, Kahana E, Paz R (2018) Oscillations synchronize amygdala-to-prefrontal primate circuits during aversive learning. Neuron 97:291–298 Pitkänen A, Pikkarainen M, Nurminen N, Ylinen A (2000) Reciprocal connections between the amygdala and the hippocampal forma- tion, perirhinal cortex, and postrhinal cortex in rat - A review. References Parahippocampal Reg 911:369–391 Usrey WM, Reppas JB, Reid RC (1998) Paired-spike interactions and synaptic efficacy of retinal inputs to the thalamus. Nature 395:384–387 Popescu AT, Popa D, Pare D (2009) Coherent gamma oscillations couple the amygdala and striatum during learning. Nat Neurosci 12:801–807 Vergotte G, Torre K, Chirumamilla VC, Anwar AR, Groppa S, Per- rey S, Muthuraman M (2017) Dynamics of the human brain net- work revealed by time-frequency effective connectivity in fNIRS. Biomed Opt Express 8:5326–5341i Preston AR, Eichenbaum H (2013) Interplay of hippocampus and pre- frontal cortex in memory. Curr Biol 23:R764–R773 Wan EA, Nelson AT (2001) Dual extended Kalman filter methods. In: Haykin S (ed) Kalman filtering and neural networks. Wiley, New York, pp 123–173 Reppucci CJ, Petrovich GD (2016) Organization of connections between the amygdala, medial prefrontal cortex, and lateral hypo- thalamus: a single and double retrograde tracing study in rats. Brain Struct Funct 221:2937–2962 Wierzynski CM, Lubenov EV, Gu M, Siapas AG (2009) State-depend- ent spike-timing relationships between hippocampal and prefron- tal circuits during sleep. Neuron 61:587–596 Rocher C, Spedding M, Munoz C, Jay TM (2004) Acute stress-induced changes in hippocampal/prefrontal circuits in rats: effects of anti- depressants. Cereb Cortex 14:224–229 Williams KA, Magnuson M, Majeed W, LaConte SM, Peltier SJ, Hu X, Keilholz SD (2010) Comparison of alpha-chloralose, medetomi- dine and isoflurane anesthesia for functional connectivity mapping in the rat. Magn Reson Imaging 28:995–1003 Rubinov M, Sporns O (2010) Complex network measures of brain connectivity: uses and interpretations. Neuroimage 52:1059–1069 Russo SJ, Nestler EJ (2013) The brain reward circuitry in mood disor- ders. Nat Rev Neurosci 14:609–625 Xu H, Liu L, Tian Y, Wang J, Li J, Zheng J, Zhao H, He M, Xu TL, Duan S, Xu H (2019) A disinhibitory microcircuit mediates con- ditioned social fear in the prefrontal cortex. Neuron 102:668–682 Russo SJ, Murrough JW, Han MH, Charney DS, Nestler EJ (2012) Neurobiology of resilience. Nat Neurosci 15:1475–1484 Yang JW, An S, Sun JJ, Reyes-Puerta V, Kindler J, Berger T, Kilb W, Luhmann HJ (2013) Thalamic network oscillations synchronize ontogenetic columns in the newborn rat barrel cortex. Cereb Cor- tex 23:1299–1316 Sainsbury RS, Montoya CP (1984) The relationship between type 2 theta and behavior. Physiol Behav 33:621–626 Seidenbecher T, Laxmi TR, Stork O, Pape HC (2003) Amygdalar and hippocampal theta rhythm synchronization during fear memory retrieval. References Science 301:846–850f Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Senn V, Wolff SBE, Herry C, Grenier F, Ehrlich I, Grundemann J, Fadok JP, Muller C, Letzkus JJ, Lüthi A (2014) Long-range con- nectivity defines behavioral specificity of amygdala neurons. Neuron 81:428–437 1 3
https://openalex.org/W4213078611	https://bmcpublichealth.biomedcentral.com/track/pdf/10.1186/s12889-022-12705-z	English	null	Outcomes of a community-led online-based HIV self-testing demonstration among cisgender men who have sex with men and transgender women in the Philippines during the COVID-19 pandemic: a retrospective cohort study	BMC public health	2,022	cc-by	9,433	Outcomes of a community-led online- based HIV self-testing demonstration among cisgender men who have sex with men and transgender women in the Philippines during the COVID-19 pandemic: a retrospective cohort study Patrick C. Eustaquio1†, Roberto Figuracion Jr2†, Kiyohiko Izumi3, Mary Joy Morin4, Kenneth Samaco3, Sarah May Flores4, Anne Brink5 and Mona Liza Diones2 Eustaquio et al. BMC Public Health (2022) 22:366 https://doi.org/10.1186/s12889-022-12705-z Eustaquio et al. BMC Public Health (2022) 22:366 https://doi.org/10.1186/s12889-022-12705-z © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativeco mmons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Abstract Introduction: The Philippines, which has the fastest rising HIV epidemic globally, has limited options for HIV testing and its uptake remains low among cisgender men who have sex with men (cis-MSM) and transgender women (TGW), especially amid the COVID-19 pandemic. As HIV self-testing (HIVST) and technology-based approaches could syner- gize to expand uptake of HIV testing, we aimed to evaluate the outcomes of a community-led online-based HIVST demonstration and to explore factors associated with HIVST-related behaviours and outcomes. Methods: We did a secondary data analysis among cis-MSM and TGW who participated in the HIVST demonstration, who were recruited online and tested out-of-facility, in Western Visayas, Philippines, from March to November 2020. We reviewed data on demographics, sexuality-, and context-related variables. Using multivariable logistic regression, we tested for associations between the aforementioned covariates and two primary outcomes, opting for directly- assisted HIVST (DAH) and willingness to secondarily distribute kits. Results: HIVST kits were distributed to 647 individuals (590 cis-MSM, 57 TGW), 54.6% were first-time testers, 10.4% opted DAH, and 46.1% were willing to distribute to peers. Reporting rate was high (99.3%) with 7.6% reactivity rate. While linkage to prevention (100%) and care (85.7%) were high, pre-exposure prophylaxis (PrEP) (0.3%) and antiretrovi- ral therapy (ART) (51.0%) initiation were limited. There were no reports of adverse events. Those who were employed, had recent anal intercourse, opted for DAH, not willing to secondarily distribute, and accessed HIVST during minimal to no quarantine restriction had significantly higher reactivity rates. Likelihood of opting for DAH was higher among those who had three or more partners in the past year (aOR = 2.01 [CI = 1.01–4.35]) and those who accessed during Philippines Full list of author information is available at the end of the article Introductionh The limited demand for HIV testing among the key pop- ulations (KP) has challenged the Philippines to reverse its HIV epidemic, where annual incidence of new infec- tions and AIDS-related death increased by 237 and 315%, respectively, over the past decade [1]. Although esti- mated national prevalence is at 0.2%, the epidemic is con- centrated among KP with prevalence disproportionately higher among people who inject drugs (PWID) (29.0%), cisgender men who have sex with men (cis-MSM) (5.0%), transgender women (TGW) (4.9%), and female sex workers (0.6%) [2]. Improvements in the first 95% of the UNAIDS 95–95-95 targets were noted in the recent years until the COVID-19 pandemic has decreased HIV tests done by 61% in 2020, ultimately leading to 68% of estimated people living with HIV (PLHIV) knowing their status in 2021 [a], similar to the proportion estimated 5 years ago [2]. An equally important approach is the use of technol- ogy-based interventions—this is the concurrent use of technology to expand reach, accelerate scale-up, and facilitate cost-efficient and instantaneous service delivery, responding to the inherent restrictions in face-to-face services [16]. Examples of these are online-based inter- ventions, which if used with HIVST seem to synergisti- cally remove barriers to HIV testing among cis-MSM especially among first-time testers [17, 18]. Even though the proportion of Filipinos accessing the internet (67.0%) and using social media (80.7%) are higher than the global average [19], this approach has not been maximized, yet has been increasingly used during the COVID-19 pan- demic [20]. The diagnosis gap is a known driver of the HIV epi- demic [3]. The low uptake of HIV testing among cis- MSM and TGW has been attributed to meager options for testing in the Philippines [4, 5], limited currently to facility-based and community-based testing. The former is the more prevalent model [6] and involves using rapid diagnostic test (RDT) kits, available only in Department of Health (DOH)-accredited stand-alone laboratories, hospitals, and clinics, and is only facilitated by medical technologists specifically trained for HIV [7]. Whereas community-based testing is carried out by trained lay providers during community visits and outreach pro- grams using RDT kits. To address the low uptake amid the limited choices, expanding options may be key to upscaling access and uptake of HIV testing. Keywords: HIV self-testing, Cisgender men who have sex with men, Transgender women, Community-based interventions, Digital health, Differentiated service delivery, Philippines, Low- and middle-income countries maximal quarantine restrictions (aOR = 4.25 [CI = 2.46–7.43]). Odds of willingness to share were higher among those in urban areas (aOR = 1.64 [CI = 1.15–2.36]) but lower among first-time testers (aOR = 0.45 [CI = 0.32–0.62]). C l i HIVST ld ff ti l h h d t h l ti Whil th d d i i li maximal quarantine restrictions (aOR = 4.25 [CI = 2.46–7.43]). Odds of willingness to share were hi in urban areas (aOR = 1.64 [CI = 1.15–2.36]) but lower among first-time testers (aOR = 0.45 [CI = 0.3 Conclusions: HIVST could effectively reach hard-to-reach populations. While there was demand in accessing online- based unassisted approaches, DAH should still be offered. Uptake of PrEP and same-day ART should be upscaled by decentralizing these services to community-based organizations. Differentiated service delivery is key to respond to preferences and values of key populations amid the dynamic geographical and sociocultural contexts they are in. Keywords: HIV self-testing, Cisgender men who have sex with men, Transgender women, Community-based interventions, Digital health, Differentiated service delivery, Philippines, Low- and middle-income countries enrollment to treatment [13]. In the Philippines, limited evidence shows acceptability and preference of blood- based over fluid-based tests among cis-MSM and TGW [14, 15]. Introductionh The World Health Organization (WHO) has recommended HIV self-testing (HIVST), which involves the use of RDT kits for individuals to perform and interpret on their own [8]. This may remove barriers in the current HIV testing in the Philippines, including geographical distance, lack of confidentiality or privacy, conflicting schedules, and stigma [4, 5, 9–12]. HIVST has been shown to be safe and effective at increasing uptake and frequency of HIV test- ing without compromising condom use, social safety, and We aimed to describe the outcomes of a community- led online-based HIVST demonstration project done in Western Visayas, Philippines, particularly, in terms of reach, reporting and reactivity rates, and successful linkage to services. Furthermore, we aimed to explore the demographic, sexuality-, and context-related factors associated with HIVST-related behavior and preferences, particularly opting for DAH and willingness to share HIVST kits to their partners and peers. © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativeco mmons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Eustaquio et al. BMC Public Health (2022) 22:366 Page 2 of 11 maximal quarantine restrictions (aOR = 4.25 [CI = 2.46–7.43]). Odds of willingness to share were higher among those in urban areas (aOR = 1.64 [CI = 1.15–2.36]) but lower among first-time testers (aOR = 0.45 [CI = 0.32–0.62]). Conclusions: HIVST could effectively reach hard-to-reach populations. While there was demand in accessing online- based unassisted approaches, DAH should still be offered. Uptake of PrEP and same-day ART should be upscaled by decentralizing these services to community-based organizations. Differentiated service delivery is key to respond to f d l f k l ti id th d i hi l d i lt l t t th i maximal quarantine restrictions (aOR = 4.25 [CI = 2.46–7.43]). Odds of willingness to share were higher among those in urban areas (aOR = 1.64 [CI = 1.15–2.36]) but lower among first-time testers (aOR = 0.45 [CI = 0.32–0.62]). Conclusions: HIVST could effectively reach hard-to-reach populations. While there was demand in accessing online- based unassisted approaches, DAH should still be offered. Uptake of PrEP and same-day ART should be upscaled by decentralizing these services to community-based organizations. Differentiated service delivery is key to respond to preferences and values of key populations amid the dynamic geographical and sociocultural contexts they are in. Study design, setting, and participants We did a multiple-center, retrospective cohort analysis of participants recruited in a community-led online-based HIVST demonstration in Western Visayas, Philippines, implemented from March to November 2020. The STROBE statement checklist of items was used to guide the development of this research [21]. Western Visayas is in the center of the Philippines and is composed of six provinces separated in three different islands. Its two highly urbanized cities (Bacolod City and Iloilo City) are HIV high burden areas [1]. HIVST was demonstrated in the region in 2020 by the DOH West- ern Visayas because, firstly, two thirds of new HIV cases Eustaquio et al. BMC Public Health (2022) 22:366 Page 3 of 11 Page 3 of 11 in the Philippines are detected outside Metro Manila and the Western Visayas is among the areas with highest HIV incidence, contributing 6.2% of newly diagnosed cases in 2019 nationally [22], and secondly, almost one fourth of PLHIV in the region has not been diagnosed in 2019 [22].h and support on linking them to appropriate services. For validation purposes, the participants were asked to show the outcome of the HIVST kit by sending a photo or through a video call. In rare cases when the result was invalid (n = 4), they were offered retesting using their preferred strategy (DAH or unassisted) but with a differ- ent HIVST kit. g g The demonstration project was implemented by dif- ferent CBOs led by the main study site, Family Planning Organization of the Philippines-Iloilo (FPOP-Iloilo)- Rajah Community Center. Both online and offline recruitment campaigns were conducted, using social media platforms and face-to-face invites in social and sexual networks, respectively. As the HIVST demonstra- tion was only limited among individuals within the West- ern Visayas region, the campaigns were targeted among cis-MSM and TGW in the said region. These campaigns led interested individuals to an online sign-up sheet. All adults residing in the six provinces in Western Visayas who signed-up were eligible to receive the HIVST ser- vices from implementing CBOs which started to distrib- ute INSTI® HIV Self-Test kits (BioLytical Laboratories, Richmond, British Columbia, Canada) in March 2020. While extreme lockdowns were implemented due to the unprecedented COVID-19 pandemic, the online-based nature of the program allowed continuity of provision of HIVST. Study design, setting, and participants Those who tested reactive were referred directly to HIV treatment facilities and follow-up calls were conducted at two weeks, four weeks, and then every four weeks until ART initiation or twelve weeks, whichever came first, to determine self-reported linkage to the cascade of HIV care services. Those who neither responded to follow- ups nor reported their cascade outcomes within twelve weeks were tagged as lost to follow-up. Verification of the self-reported cascade outcomes were legislatively pos- sible only if they were eventually enrolled in the main study site, FPOP-Iloilo. Meanwhile, those who tested non-reactive were routinely provided with risk reduc- tion counseling and were offered to be enrolled in the HIV preexposure prophylaxis (PrEP) program of FPOP- Iloilo (the only provider in the region during the span of the study) as part of the post-test counseling. The project officers in each CBO were designated to collect all the data using a standardized data collection sheet. Using convenience sampling, we performed a second- ary data analysis among people who opted and consented for the HIVST demonstration who fit the following inclu- sion criteria as follows: (1) self-identified as cis-MSM or TGW, (2) 18 years old and above, and (3) opted for online-based services. The following were excluded: (1) assigned female at birth, (2) assigned male at birth and identified as heterosexual, (3) opted for offline services, and (4) those who eventually disclosed that they were known PLHIV. We created a research dataset for the purposes of the secondary analysis from the deidentified dataset from the implementers, which included participants who fit the study inclusion and exclusion criteria. We cleaned the dataset and ensured that recoding would preserve the original data as much as possible. The following outcomes were included: (1) HIVST result (reactive or non-reac- tive), (2) whether they opted for directly assisted (DAH) (i.e., in-person demonstration and / or supervision by a provider) or unassisted HIVST [8], (3) whether they were willing to distribute the kits to their partners or peers (i.e., secondary distribution of HIVST kits) or not [8], (4) linkage to appropriate HIV services, i.e., enrollment to care (confirmatory testing and treatment) among those reactive and prevention services (risk reduction coun- seling, condoms and lubricants, and/or PrEP) among those non-reactive, and (5) reports of adverse events such as suicidal attempts, coercion, and social harm [8, 23, 24]. Study design, setting, and participants Included covariates were (1) demographics (age, gender identity, and employment), (2) sexuality-related variables including (a) anal sex within the past 3 months, (b) num- ber of male partners for the past 12 months, (c) history of HIV testing, i.e., first-time tester or not [25], and (d) source of information regarding the HIVST program, and (3) context-related variables such as (a) time, measured in the date of access of the HIVST service, and (b) place of residence. These variables were determined a priori [25, 26]. Some quantitative variables were transformed into Ethical approvalh The study adhered to the principles of the Declaration of Helsinki and the ethical approval (NEC Code: 2021–004) was provided by the National Ethics Commission of the Philippine Council on Health Research and Develop- ment, Department of Science and Technology, Republic of the Philippines. Procedures When signing up online, the participants were provided with pre-test and programmatic information. Upon pro- viding their electronically recorded consent, data on demographics, sexual risk and behavior, and HIV testing related behavior and preference were collected through self-reporting. Thereafter, the participants were reached by the implementers through phone calls to verify the intent and data they provided. Participants accessed the HIVST package, either through pick-up or courier, con- taining the HIVST kit itself, instructional materials (con- taining information on HIV, on how to use, interpret, and dispose of the kit, on accessing the support hotline, and linkage to appropriate HIV-related services), and condoms and lubricants. Participants were followed- up through phone calls within two days upon access to determine the outcomes, to provide post-test counseling Eustaquio et al. BMC Public Health (2022) 22:366 Page 4 of 11 categories, particularly, (1) age, grouped into less than or equal to 24 or 25 and over, signifying the young KP group [8], (2) number of male sexual partners in the past 12 months, grouped based on the median number based on national biobehavioral surveillance [4]. Some qualita- tive variables were recoded: (1) the extent of quarantine restrictions into “None to minimal” or “Maximal”, based on the date and location of the individual participation, and (2) the place of residence classified into either urban or rural. categories, particularly, (1) age, grouped into less than or equal to 24 or 25 and over, signifying the young KP group [8], (2) number of male sexual partners in the past 12 months, grouped based on the median number based on national biobehavioral surveillance [4]. Some qualita- tive variables were recoded: (1) the extent of quarantine restrictions into “None to minimal” or “Maximal”, based on the date and location of the individual participation, and (2) the place of residence classified into either urban or rural. final multivariate analyses. Chi-square tests were used to assess collinearity of potential predictors. We used c statistics and Hosmer-Lemeshow statistics to assess pre- dictive power and model fit. We used p < 0.05 to deter- mine significant outputs in Chi-square and Fisher exact tests and crude (cOR) and adjusted odds ratio (aOR). All analyses were performed using R version 4.0.3. Statistical analysis Descriptive statistics were done to summarize the predic- tors. We performed Chi-square and Fisher exact tests to compare baseline characteristics, stratified by reported HIV test result. To describe the outcomes of the HIVST demonstration, we determined the prevalence at each component of the testing cascade. Moreover, we per- formed multivariate logistic regression using complete case analyses and backward elimination to determine predictors associated with our outcomes of interest: (1) opting for DAH and (2) willingness to distribute. Pre- dictors found to be statistically associated in the initial bivariate analyses using p < 0.25 were included in the Results From March to November 2020, 768 HIVST kits were distributed (Fig. 1). Due to missing documentation, 33 participants were not assessed for eligibility. Among those assessed, 88 were excluded based on the inclusion and exclusion criteria. Eventually, 647 participants were included in the analysis. Fig. 1 Flow diagram of the retrospective cohort study. HIVST – HIV self-testing, PLHIV – people living with HIV Eustaquio et al. BMC Public Health (2022) 22:366 Page 5 of 11 Table 1 Sociodemographic factors, sexual risk and behavior, and HIV testing-related behavior and preferences of the HIVST demonstration project participants, disaggregated into reported HIV testing result Table 1 Sociodemographic factors, sexual risk and behavior, and HIV testing-related behavior and preferences of the HIVST demonstration project participants, disaggregated into reported HIV testing result Table 1 Sociodemographic factors, sexual risk and behavior, and HIV testing-related behavior and preferences of the HIVST demonstration project participants, disaggregated into reported HIV testing result a Denominator is number of individuals reported HIVST result disaggregated based on baseline characteristic b Fisher exact test. a Denominator is number of individuals reported HIVST result disaggregated based on baseline characteristic b Fisher exact test. All other comparison of proportions were done using Chi-square test b Fisher exact test. All other comparison of proportions were done using Chi-square test significant at p < 0.05 a Denominator is number of individuals reported HIVST result disaggregated based on baseline characteristic Results BMC Public Health (2022) 22:366 Page 6 of 11 Table 2 HIV self-testing demonstration study outcomes a Defined as being enrolled into a treatment facility b Includes condoms and lubricants and behavioral risk reduction counseling n / N (%) Distributed 647 Reported results 643 / 647 (99.3%) Reactive 49 / 643 (7.6%) Linked to carea 42 / 49 (85.7%) Initiated on antiretroviral therapy 25 / 49 (51.0%) Lost to follow-up 7 / 49 (14.3%) Non-reactive 594 / 643 (92.4%) Linked to prevention servicesb 594 / 594 (100%) Initiated on pre-exposure prophylaxis 2 /594 (0.3%) Adverse events reports 0 / 647 (0.0%) Table 2 HIV self-testing demonstration study outcomes Median age of the participants was 26 (interquartile range 23–30) years. Majority self-identified as cis-MSM (91.2%), and most were employed (77.6%) and residing in an urban area (69.1%) (Table 1). Many (70.2%) had three or more sexual partners in the past 12 months and 51.8% had anal sexual intercourse in the past 3 months. Among those distributed with HIVST kits, more than half (54.6%) had never tested previously for HIV, most (89.6%) preferred unassisted HIVST, and almost half (46.1%) were willing to distribute kits to their sexual part- ners and peers. Furthermore, reporting rate of HIVST result was high at 99.3%. g Of the 643 who reported their HIVST outcomes, 49 (7.6%) tested reactive. The proportions of testing reac- tive were significantly higher among those employed (p = 0.023), who had anal intercourse in the past 3 months (p = 0.021), who opted for DAH (p = 0.018), not willing to distribute the HIVST kits (p < 0.000), and who accessed HIVST during none to minimal quarantine restrictions (p = 0.017) compared to their corresponding counterparts. There was no significant difference in the proportion of those tested reactive between first-time testers and those with a history of HIV testing (p = 0.743). Moreover, among those who tested reactive, 42 (85.7%) were eventually linked to care and 25 (51.0%) were initiated on ART during the study period (Table 2). Among those non-reactive, all 594 participants (100%) were provided prevention services through routine provision of risk reduction counseling and condoms and lubricants. Only 2 (0.3%) were successfully linked to PrEP services. Lastly, there were no reports of adverse events in the program. Results It is striking that there seemed to be no difference in reac- tivity rate between first-time and ever testers, especially considering that in the Philippines all of those who come for HIV testing are routinely provided with risk reduction counselling [7, 32] which would be expected to decrease their risk for HIV. Our finding suggests that the aforemen- tioned may have had marginal impact, as noted in other studies [33]. Nonetheless, HIV testing is a good avenue to educate KP regarding HIV and their risks. Hence, the DOH should not only consider reviewing its risk reduction coun- seling strategy but also advocate for and upscale all aspects of combination prevention [34], particularly pre- and post-exposure prophylaxis, condom use, and addressing stigma and discrimination, which have been determined as national priority interventions for HIV prevention [35]. Only a few (10.4%) opted for DAH (Table 3). The likeli- hood of opting for DAH was higher among those who had three or more partners in the past year (aOR = 2.01 [CI = 1.01–4.35], p = 0.049) and among those who accessed HIVST during maximal quarantine restrictions (aOR = 4.25 [CI = 2.46–7.43], p < 0.00). y Although only a minority in our cohort (10.4%) opted for DAH, the following findings have important implications for policy. Firstly, reactivity among those who opted for DAH was significantly higher compared to unassisted, sim- ilar with another study [36]. Moreover, we found that those with three or more sexual partners in the past year had twice higher odds of opting for DAH. There is evidence on the presence of anxiety related to the HIVST process, par- ticularly linkage to care, and this translates to a desire for assistance among cis-MSM [14], TGW [14, 31], and other KP [37, 38]. Secondly, while it may be intuitive that testing for the first-time is associated with higher odds of DAH as seen in previous studies [14, 25, 39], it was the opposite in our bivariate model. Participants may have been enticed by the privacy, convenience, and independence that HIVST offers. Lastly, we found that stricter COVID-19-related quarantine restrictions were associated with higher likeli- hood of DAH. Results All other comparison of proportions were done using Chi-square test Distributed Reported results (N = 643) Non-Reactive Reactive p-value n (% of total of 647) n (%)a n (%)a Age group 0.441 18–24 243 (37.6%) 227 (93.4%) 16 (6.6%) 25 and over 404 (62.4%) 367 (91.8%) 33 (8.3%) Missing 0 (0.0%) Gender identity 0.234b Cis-MSM 590 (91.2%) 539 (92.0%) 47 (8.0%) Transgender woman 57 (8.8%) 55 (96.5%) 2 (3.5%) Missing 0 (0.0%) Employment 0.023 * Unemployed 143 (22.1%) 139 (97.2%) 4 (2.8%) Employed 502 (77.6%) 455 (91.2%) 44 (8.8%) Missing 2 (0.3%) Location 0.556 Urban 447 (69.1%) 182 (91.5%) 17 (8.5%) Rural 200 (30.9%) 412 (92.8%) 32 (7.2%) Missing 0 (0.0%) Number of male sex partners in the past 12 months 0.757 Less than 3 190 (29.4%) 177 (93.2%) 13 (6.8%) 3 or more 454 (70.2%) 417 (92.5%) 34 (7.5%) Missing 3 (0.5%) Anal course in the past 3 months 0.018 * No 309 (47.8%) 293 (95.1%) 15 (4.9%) Yes 335 (51.8%) 300 (90.1%) 33 (9.9%) Missing 0 (0.5%) First-time tester 0.743 No 294 (45.4%) 269 (92.8%) 21 (7.2%) Yes 353 (54.6%) 325 (92.1%) 28 (7.9%) Missing 0 (0.0%) Preference for assistance < 0.00 * Unassisted 580 (89.6%) 539 (93.6%) 37 (6.4%) Directly-assisted 67 (10.4%) 55 (82.1%) 12 (17.9%) Missing 0 (0.0%) Willingness to distribute HIVST kits to sexual and social network < 0.00 * No 348 (53.8%) 308 (88.5%) 40 (11.5%) Yes 298 (46.1%) 285 (96.9%) 9 (3.1%) Missing 1 (0.2%) Source of information about HIVST 0.518b Social network 616 (95.2%) 567 (92.6%) 45 (7.4%) Partner notification 24 (3.7%) 21 (87.5%) 3 (12.5%) Provider-initiated 7 (1.1%) 6 (85.7%) 1 (14.3%) Missing 0 (0.0%) Community quarantine restrictions 0.017 * None to minimal 574 (88.7%) 526 (92.0%) 46 (8.0%) Maximum 73 (11.3%) 68 (93.2%) 3 (4.1%) Missing 0 (0.0%) Eustaquio et al. BMC Public Health (2022) 22:366 Eustaquio et al. Results We could only speculate that the perceived limited access to healthcare services amid a time of public health crisis and uncertainty may have reinforced depend- ence on health providers and peers especially given that Almost half (46.1%) were willing to distribute the HIVST kits to their partners and peers (Table 4). The likelihood of willingness to share was higher among those residing in urban (aOR = 1.64 [CI = 1.15–2.36], p = 0.007), whereas it was lower among first-time testers (aOR = 0.45 [CI = 0.32– 0.62], p < 0.00). Discussion We found that online-based HIVST reached many first- time testers among cis-MSM and TGW, similar with previous studies [27–31]. Reporting and linkage to care and prevention rates were high but ART and PrEP initia- tion were sub-optimal. Reactivity rate and HIVST prefer- ences were associated with participants’ vulnerabilities and context. Eustaquio et al. BMC Public Health (2022) 22:366 Page 7 of 11 Table 3 Predictors of opting directly assisted HIVST c statistic = 0.70; Hosmer-Lemeshow χ2 = 0.17433, df = 8, p-value = 1 * significant at < 0.25 for crude odds ratio (cOR); ** significant at < 0.05 for adjusted odds ratio (aOR) Directly assisted HIVST Crude OR p-value Adjusted OR p-value n / N (% among distributed) OR (95% CI) aOR (95% CI) Age group 18–24 21 / 243 (8.6%) 1.00 25 and over 46 / 404 (11.4%) 1.36 (0.80–2.38) 0.269 Missing 0 / 647 (0.0%) Gender identity Cis-MSM 64 / 590 (10.8%) 1.00 1.00 Transgender woman 3 / 57 (5.3%) 0.46 (0.11–1.29) 0.197* 0.69 (0.16–2.08) 0.562 Missing 0 / 647 (0.0%) Employment Unemployed 11 / 143 (7.7%) 1.00 Employed 55 / 502 (11.0%) 1.48 (0.78–3.05) 0.258 Missing 2 / 647 (0.3%) Location Rural 42 / 447 (9.4%) 1.00 – Urban 25 / 200 (12.5%) 0.73 (0.43–1.24) 0.233* – – – Missing 0 / 647 (0.0%) Number of male sex partners in the past 12 months Less than 3 10 / 190 (5.3%) 1.00 1.00 3 or more 56 / 454 (12.3%) 2.53 (1.32–5.38) 0.009* 2.01 (1.01–4.35) 0.049** Missing 3 / 647 (0.5%) First-time tester No 40 / 294 (13.6%) 1.00 – Yes 27 / 353 (7.6%) 0.53 (0.31–0.88) 0.015* – – – Missing 0 / 647 (0.0%) Source of information about HIVST Social network 61 / 616 (9.9%) 1.00 Partner notification 5 / 24 (20.8%) 2.39 (0.77–6.20) 0.093 Provider-initiated 1 / 7 (14.3%) 1.52 (0.09–9.08) 0.702 Missing 0 / 647 (0.0%) Community quarantine restrictions None to minimal 115 / 580 (19.8%) 1.00 1.00 Maximum 36 / 67 (53.7%) 4.70 (2.79–7.95) < 0.00* 4.25 (2.46–7.43) < 0.00** Missing 0 / 647 (0.0%) * significant at < 0.25 for crude odds ratio (cOR); ** significant at < 0.05 for adjusted odds ratio (aOR) 47], respectively, and kits being delivered by trained provid- ers themselves. Discussion Moreover, ensuring DAH may be crucial if the Philippines introduces oral fluid-based test, to address lack familiarity as Filipino KPs are more accustomed to blood-based tests. PLHIV and KP are at increased risk of vulnerability to both HIV and COVID and its physical, mental, and social comorbidities [40, 41]. Therefore, as DAH was associated with better retention [42] and higher ART initiation [43], even during the COVID-19 pandemic [44], implemen- tation of HIVST in the Philippines should provide and expand options for direct assistance that go beyond in- person demonstration and also include emotional support [45]. This could involve capacitating community-based testing providers and “seeds” to provide demonstrations and peer support to their communities and networks [46, Secondary distribution has been shown to increase the reach, positivity yield, and cost-efficiency of HIV testing among cis-MSM [26, 46, 48]. Like other studies which showed increased distribution [26, 49], we found that willingness to distribute was higher among those with prior HIV testing. This is reassuring as we also found that Eustaquio et al. Discussion BMC Public Health (2022) 22:366 Page 8 of 11 Table 4 Predictors of willingness to distribute HIVST to sexual partners and peers c statistic = 0.70; Hosmer-Lemeshow χ2 = 1.1911, df = 8, p-value = 0.9967 * significant at < 0.25 for crude odds ratio (cOR); ** significant at < 0.05 for adjusted odds ratio (aOR) Willingness to distribute Crude OR p-value Adjusted OR p-value n / N (% among distributed) OR (95% CI) aOR (95% CI) Age group 18–24 100 / 242 (41.3%) 1.00 – 25 and over 198 / 404 (49.0%) 1.36 (0.99–1.89) 0.058* – – – Missing 1 / 647 (0.2%) Gender identity Cis-MSM 280 / 589 (47.5%) 1.00 – Transgender woman 18 / 57 (31.6%) 0.51 (0.28–0.90) 0.023* – – – Missing 1 / 647 (0.2%) Employment Unemployed 61 / 143 (42.7%) 1.00 Employed 236 / 501 (47.1%) 1.20 (0.82–1.75) 0.347 Missing 2 / 647 (0.3%) Location Rural 223 / 447 (49.9%) 1.00 1.00 Urban 75 / 199 (37.7%) 1.65 (1.17–2.32) 0.004* 1.64 (1.15–2.36) 0.007** Missing 1 / 647 (0.2%) Number of male sex partners in the past 12 months Less than 3 83 / 189 (43.9%) 1.00 3 or more 214 / 454 (47.1%) 1.14 (0.81–1.60) 0.456 Missing 4 / 647 (0.6%) First-time tester No 172 / 294 (58.5%) 1.00 1.00 Yes 126 / 352 (35.8%) 0.40 (0.29–0.54) < 0.00* 0.45 (0.32–0.62) < 0.00** Missing 1 / 647 (0.2%) Source of information about HIVST Social network 281 / 616 (45.6%) 1.00 Partner notification 13 / 23 (56.5%) 1.55 (0.67–3.68) 0.306 Provider-initiated 4 / 7 (57.1%) 1.59 (0.35–8.13) 0.546 Missing 1 / 647 (0.2%) Community quarantine restrictions None to minimal 43 / 348 (12.4%) 1.00 1.00 Maximum 108 / 298 (36.2%) 4.03 (2.73–6.05) < 0.00* 3.60 (2.41–5.45) < 0.00** Missing 1 / 647 (0.2%) Table 4 Predictors of willingness to distribute HIVST to sexual partners and peers * significant at < 0.25 for crude odds ratio (cOR); ** significant at < 0.05 for adjusted odds ratio (aOR) and HIV education [53], and higher acceptability of HIV interventions [51]. This is opportune as urban areas are priority sites for sustainable and effective HIV response [54]; as willingness was high, secondary distribution of HIVST kits could augment current HIV testing practices through approaches like index testing and sexual and social network testing [46, 47]. Discussion There is plenty of evidence that secondary distribution [55] and technology-assisted models [18, 55] play a role in increasing testing uptake among cis-MSM and TGW, whereas community-based models were found to be more effective among young online-based HIVST can effectively reach to first-time testers, consistent with other studies [17, 18]. Hence, in the Philippines, where less than half (43%) of cis-MSM and TGW were ever tested for HIV [4], technology-based HIVST has the potential to increase the proportion of ever tested for HIV [17, 18] and, consequently, facilitate initial and repeat testing among their networks though secondary distribution [30, 50]. We also found that resid- ing in urban areas was associated with increased odds of willingness to distribute. This may be due to the dense clustering of KP [51], higher access to queer culture [52] Page 9 of 11 Eustaquio et al. BMC Public Health (2022) 22:366 self-selection bias. Generalizing the findings of our study must be done with caution. Lastly, there are limitations of in the use of stepwise backward elimination. Although it prevents overfitting and allow different combinations of variables [59–62], there is considerable variance when different samples are used [62] and there is potential for inappropriate variables to be included in the model [59, 60, 62]. We did, however, ensure that there were suf- ficient events per variable [60, 63] and that we explored a priori predictors, respectively. Thus, we are confident that the models predict the outcomes within the context of the study. people and male partners of females in antenatal clinics [55]. The knowledge gap on effective distribution mod- els among other KPs, like PWID, people in prisons, and female sex workers, may be attributed to the dispropor- tionately limited studies among these vulnerable groups. Hence, further studies are required to fully respond to their values and preferences on HIV testing. Despite high rates of uptake, reporting, and referral to services, we found suboptimal initiation of antiretrovi- ral interventions. Apart from the limitations brought by COVID-19, suboptimal initiation may be explained by the fact that only one in eleven CBOs in the demonstra- tion was capable of prescribing ART or PrEP, similar to the experience in Thailand [43]. However, when treat- ment was also CBO-led, as in the HIVST demonstra- tion in Vietnam, higher initiation rates was noted [56]. Discussion Furthermore, despite that rapid ART initiation has been recommended by the WHO since 2017, the current HIV treatment guidelines in the Philippines in 2018 did not mention this [b] and may explain the low ART initiation rate. Meanwhile, poor PrEP initiation may be explained by cost [57], especially that, unlike ART, PrEP is neither state-sponsored nor covered by health insurance in the Philippines. Overall, the benefits of online-based HIVST could not be maximized without concurrent innovations in treatment and prevention. Although a few treatment and PrEP facilities are CBOs or have partner CBOs in the Philippines, continuing the endeavor by the DOH to further decentralize HIV-related services to CBOs should be prioritized. Likewise, technology-supported inter- ventions or seamless online-to-offline transition during ART or PrEP prescribing, linkage to care, and reten- tion, should be considered and further studied. Lastly, local treatment guidelines should be revised to allow rapid ART initiation. Acknowledgments Data used were collected from participants in the HIVST demonstration project (iScreen) in Western Visayas, Philippines. We would like to acknowl- edge the vital role of different CBOs who made the HIVST demonstration project possible: Rajah Community Center, Montecarlo Clan, Cabatuan LGBT Association, La Villa Pride, GBT Janiuay, LGBTQ PAVIA Chapter, Red Seahorse, Bagani Community Center, Ogtonganon Mask Stewards, Kamini Community Center, and Red Lace Boracay. The investigating team would also like to acknowledge the following: Adrian Hort Ramos, DOH-CHD Region 6 HIV program coordinator, for providing his technical inputs in the demonstra- tion project, Vincent Misterio, FPOP-Iloilo iScreen project coordinator, and other members of the iScreen implementation team for organizing the implementation of the HIVST demonstration project. We would also like to acknowledge Danvic Rosadiño for his technical review of the research proto- col and to our peer reviewers for providing us crucial feedback on improving this manuscript. Abbreviations AIDS: Acquired immune deficiency syndrome; aOR: Adjusted odds ratio; ART: Antiretroviral therapy; CBO: Community-based organization; CI: Confidence interval; cis-MSM: Cisgender men who have sex with men; cOR: Crude odds ratio; COVID-19: Coronavirus disease in 2019; DAH: Directly- assisted HIV self-testing; DOH: Department of Health; FPOP: Family Planning Organization of the Philippines; HIV: Human immunodeficiency virus; HIVST: HIV self-testing; KP: Key population; PLHIV: People living with human immunodeficiency virus.; PrEP: Pre-exposure prophylaxis; STROBE: Strength- ening the reporting of observational studies; TGW: Transgender women; UNAIDS: Joint United Nations Programme on HIV/AIDS; WHO: World Health Organization. The primary strength of this study was the technol- ogy-based delivery of the demonstration project; this allowed numerous and precise data points to be used to explore associations. Furthermore, to our knowledge, this is the first association study to consider the poten- tial influence of quarantine restrictions on HIV service delivery in the Philippines. Meanwhile, it is important to acknowledge some study limitations. Firstly, as this study is a secondary data analysis, we were bound to the limitations of the primary data collection such as high potential for information bias, as much of the data was collected through self-reporting which is particularly vul- nerable to social desirability bias. However, verification was done whenever possible. Moreover, likewise with a previous study [58], the willingness to distribute HIVST kits were collected at baseline and, hence, may be influ- enced by the uncertainty of their HIV status. Secondly, the online-based convenience sampling may have led to Conclusions We have shown that a community-based online-based HIVST intervention is safe and has the potential to increase uptake of HIV testing and linkage to appropriate service among cis-MSM and TGW, yet initiation of ART and PrEP were low. The study emphasized the impor- tance of providing different options for HIVST which suite their values and preferences of KP. Geographi- cal, temporal, and sociocultural contexts are important considerations in ensuring differentiated services are provided. References 1. Department of Health Epidemiology Bureau. A briefer on the Philippine HIV estimates 2020. Manila: Department of Health; 2021. 1. Department of Health Epidemiology Bureau. A briefer on the Philippine HIV estimates 2020. Manila: Department of Health; 2021. 22. Department of Health Epidemiology Bureau. HIV/AIDS and ART registry of the Philippines - December 2019. Manila: Department of Health; 2019. 2. Joint United Nations Programme on HIV/AIDS. UNAIDS Data 2020. Geneva: UNAIDS; 2020. 2. Joint United Nations Programme on HIV/AIDS. UNAIDS Data 2020. Geneva: UNAIDS; 2020. 23. Mulubwa C, Hensen B, Phiri MM, Shanaube K, Schaap AJ, Floyd S, et al. Community based distribution of oral HIV self-testing kits in Zambia: a cluster-randomised trial nested in four HPTN 071 (PopART) intervention communities. Lancet HIV. 2019;6(2):e81–92. 3. Fisher M, Pao D, Brown AE, Sudarshi D, Gill ON, Cane P, et al. Deter- minants of HIV-1 transmission in men who have sex with men: a combined clinical, epidemiological and phylogenetic approach. AIDS. 2010;24(11):1739–47. 3. Fisher M, Pao D, Brown AE, Sudarshi D, Gill ON, Cane P, et al. Deter- minants of HIV-1 transmission in men who have sex with men: a combined clinical, epidemiological and phylogenetic approach. AIDS. 2010;24(11):1739–47. 24. Brown AN, Djimeu EW, Cameron DB. A review of the evidence of harm from self-tests. AIDS Behav. 2014;18(Suppl 4):S445–9. 4. Department of Health Epidemiology Bureau. 2018 integrated HIV Behav- ioral and serologic surveillance (IHBSS). Manila: Department of Health; 2020. 4. Department of Health Epidemiology Bureau. 2018 integrated HIV Behav- ioral and serologic surveillance (IHBSS). Manila: Department of Health; 2020. 25. Lyons CE, Coly K, Bowring AL, Liestman B, Diouf D, Wong VJ, et al. Use and acceptability of HIV self-testing among first-time testers at risk for HIV in Senegal. AIDS Behav. 2019;23(Suppl 2):130–41. 5. United Nations Development Programme. Missing in action - loss of cli- ents from HIV testing, treat- ment, care and support services: case studies of gay men and other men who have sex with men in Manila missing in action. Bangkok: United Nations Development Programme; 2017. 5. United Nations Development Programme. Missing in action - loss of cli- ents from HIV testing, treat- ment, care and support services: case studies of gay men and other men who have sex with men in Manila missing in action. Bangkok: United Nations Development Programme; 2017. 26. Xiao W, Yan L, Chen L, Fu G, Yang H, Yang C, et al. Author details 1 1 LoveYourself, Inc, 715 Anglo Bldg., Shaw Blvd, 1550 Mandaluyong City, Philip- pines. 2 Family Planning Organization of the Philippines, Inc, Iloilo Chapter – Rajah Community Center, 2F Dulalia Building, Rizal St, 5000 Iloilo City, Iloilo City Proper, Philippines. 3 World Health Organization, Philippines, Ground Floor Building 3 San Lazaro Compound, C. S. Gatmaitan Ave, Santa Cruz, 1000 Manila, Metro Manila, Philippines. 4 Department of Health, Philippines, Compound San Lazaro St, Santa Cruz, 1000 Manila, Metro Manila, Philippines. 5 Independent consultant, London, UK. 17. LeGrand S, Muessig KE, Horvath KJ, Rosengren AL, Hightow-Weidman LB. Using technology to support HIV self-testing among MSM. Curr Opin HIV AIDS. 2017;12(5):425–31. 18. McGuire M, de Waal A, Karellis A, Janssen R, Engel N, Sampath R, et al. HIV self-testing with digital supports as the new paradigm: A systematic review of global evidence (2010-2021). EClinicalMedicine. 2021;39:101059. 19. DataReportal. Digital 2021 The Philippines 2021 [Available from: https:// datareportal.com/reports/digital-2021-philippines. Received: 1 October 2021 Accepted: 28 January 2022 Received: 1 October 2021 Accepted: 28 January 2022 20. Quilantang MIN, Bermudez ANC, Operario D. Reimagining the future of HIV service implementation in the Philippines based on lessons from COVID-19. AIDS Behav. 2020;24(11):3003–5. 21. von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP, et al. Strengthening the reporting of observational studies in epidemi- ology (STROBE) statement: guidelines for reporting observational studies. BMJ. 2007;335(7624):806–8. Availability of data and materials 10. Rutstein SE, Ananworanich J, Fidler S, Johnson C, Sanders EJ, Sued O, et al. Clinical and public health implications of acute and early HIV detection and treatment: a scoping review. J Int AIDS Soc. 2017;20(1):21579. 10. Rutstein SE, Ananworanich J, Fidler S, Johnson C, Sanders EJ, Sued O, et al. Clinical and public health implications of acute and early HIV detection and treatment: a scoping review. J Int AIDS Soc. 2017;20(1):21579. Due to ethical reasons, the dataset created and analyzed is not publicly avail- able as it contains potentially sensitive information. For further inquiries, email may be sent to the corresponding author, Dr. Patrick C. Eustaquio via patrick@ loveyourself.ph. 11. Hatzold K, Gudukeya S, Mutseta MN, Chilongosi R, Nalubamba M, Nkhoma C, et al. HIV self-testing: breaking the barriers to uptake of test- ing among men and adolescents in sub-Saharan Africa, experiences from STAR demonstration projects in Malawi, Zambia and Zimbabwe. J Int AIDS Soc. 2019;22(Suppl 1):e25244. 11. Hatzold K, Gudukeya S, Mutseta MN, Chilongosi R, Nalubamba M, Nkhoma C, et al. HIV self-testing: breaking the barriers to uptake of test- ing among men and adolescents in sub-Saharan Africa, experiences from STAR demonstration projects in Malawi, Zambia and Zimbabwe. J Int AIDS Soc. 2019;22(Suppl 1):e25244. Funding 8. World Health Organization. Consolidated guidelines on HIV prevention, testing, treatment, service delivery and monitoring: recommendations for a public health approach. Geneva: World Health Organization; 2021. 8. World Health Organization. Consolidated guidelines on HIV prevention, testing, treatment, service delivery and monitoring: recommendations for a public health approach. Geneva: World Health Organization; 2021. The funding for ethics application and publication were provided by the Joint United Nations Programme on HIV/AIDS Unified Budget, Results and Account- ability Framework 2020. The funders of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. 9. Golub SA, Gamarel KE. The impact of anticipated HIV stigma on delays in HIV testing behaviors: findings from a community-based sample of men who have sex with men and transgender women in new York City. AIDS Patient Care STDS. 2013;27(11):621–7. 9. Golub SA, Gamarel KE. The impact of anticipated HIV stigma on delays in HIV testing behaviors: findings from a community-based sample of men who have sex with men and transgender women in new York City. AIDS Patient Care STDS. 2013;27(11):621–7. Competing interests The authors declare that they have no competing interests. 16. Simoni JM, Kutner BA, Horvath KJ. Opportunities and challenges of digital technology for HIV treatment and prevention. Curr HIV/AIDS Rep. 2015;12(4):437–40. Authors’ contributions All authors conceived and designed the research question and analysis and provided technical inputs in the draft. RFJ and MD collected the data. PCE performed the statistical analysis while all other others contributed to the data analysis. PCE and RFJ wrote the first draft of the manuscript and all Page 10 of 11 Eustaquio et al. BMC Public Health (2022) 22:366 Eustaquio et al. BMC Public Health (2022) 22:366 Eustaquio et al. BMC Public Health (2022) 22:366 other authors provided technical inputs and contributed for the revisions. All authors have agreed on the final version of the manuscript. 7. Department of Health. Administrative order 2017–0019 policies and guidelines in the conduct of HIV testing Services in Health Facilities. Philippines: DOH; 2017. 7. Department of Health. Administrative order 2017–0019 policies and guidelines in the conduct of HIV testing Services in Health Facilities. Philippines: DOH; 2017. Ethics approval and consent to participate pp 12. Piamonte S. Behavioral, Normative, and Control Beliefs of Filipino Men Who Have Sex with Men on Repeat HIV Testing and Counseling. Acta Med Philipp. 2021. 12. Piamonte S. Behavioral, Normative, and Control Beliefs of Filipino Men Who Have Sex with Men on Repeat HIV Testing and Counseling. Acta Med Philipp. 2021. In accordance with the international standards and national guidelines, the study adhered to the principles of the Declaration of Helsinki and the research protocol received ethics approval from the National Ethics Commission of the Philippine Council on Health Research and Development, Department of Sci- ence and Technology, Republic of the Philippines (NEC Code: 2021–004). Data collection, processing, and management strictly adhered to the Republic Act 10173 Philippines Data Privacy Act of 2012 and the Republic Act 11166 HIV/ AIDS Control Act. All participants voluntarily gave their electronically recorded informed consent to participate in the study. 13. Witzel TC, Eshun-Wilson I, Jamil MS, Tilouche N, Figueroa C, Johnson CC, et al. Comparing the effects of HIV self-testing to standard HIV testing for key populations: a systematic review and meta-analysis. BMC Med. 2020;18(1):381. 13. Witzel TC, Eshun-Wilson I, Jamil MS, Tilouche N, Figueroa C, Johnson CC, et al. Comparing the effects of HIV self-testing to standard HIV testing for key populations: a systematic review and meta-analysis. BMC Med. 2020;18(1):381. 14. Gohil J, Baja ES, Sy TR, Guevara EG, Hemingway C, Medina PMB, et al. Is the Philippines ready for HIV self-testing? BMC Public Health. 2020;20(1):34. 14. Gohil J, Baja ES, Sy TR, Guevara EG, Hemingway C, Medina PMB, et al. Is the Philippines ready for HIV self-testing? BMC Public Health. 2020;20(1):34. 15. Pagtakhan R, Rosadiño J, Briñes M, Baja E. Community-led conversations on awareness and acceptability of HIV self-screening among men who have sex with men and transgender women in metropolitan Manila, Philippines. International AIDS Society Conference on HIV Science; 2019; Mexico City: International AIDS Society; 2019. 6. Ross AG, Ditangco RA, Belimac JG, Olveda RM, Mercado ES, Chau TN, et al. The dire sexual health crisis among MSM in the Philippines: an exploding HIV epidemic in the absence of essential health services. Int J Infect Dis. 2015;37:6–8. References Can self-testing increase HIV testing among men who have sex with men: A systematic review and meta-analysis. PLoS One. 2017;12(11):e0188890. 31. Witzel CTWT, McCabe L, Gabriel M, WOlton A, Gafos M, Ward D, et al. Impact and acceptability of HIV self-testing for trans men and trans women: A mixed-methods subgroup analysis of the SELPHI randomised controlled trial and process evaluation in England and Wales. EClinicalMedicine. 2021;32(100700). 50. Li S, Zhang J, Mao X, Lu T, Gao Y, Zhang W, et al. Feasibility of indirect sec- ondary distribution of HIV self-test kits via WeChat among men who have sex with men: National Cross-sectional Study in China. J Med Internet Res. 2021;23(10):e28508. 51. He L, Pan X, Yang J, Ma Q, Jiang J, Wang W, et al. HIV risk behavior and HIV testing among rural and urban men who have sex with men in Zheji- ang Province, China: A respondent-driven sampling study. PLoS One. 2020;15(4):e0231026. 32. Department of Health. Department Memorandum No. 2020–0276: Interim Guidelines on Community-based HIV Screening. Philippines: DOH; 2020. 33. Zajac K, Kennedy CE, Fonner VA, Armstrong KS, O’Reilly KR, Sweat MD. A systematic review of the effects of Behavioral counseling on sexual risk behaviors and HIV/STI prevalence in low- and middle-income countries. AIDS Behav. 2015;19(7):1178–202. 52. Preston DB, D’Augelli AR, Kassab CD, Cain RE, Schulze FW, Starks MT. The influence of stigma on the sexual risk behavior of rural men who have sex with men. AIDS Educ Prev. 2004;16(4):291–303. 34. Brody C, Sok S, Tuot S, Pantelic M, Restoy E, Yi S. Do combination HIV prevention programmes result in increased empowerment, inclusion and agency to demand equal rights for marginalised populations in low-income and middle-income countries? A systematic review. BMJ Glob Health. 2019;4(5):e001560. 53. Colby D, Minh TT, Toan TT. Down on the farm: homosexual behaviour, HIV risk and HIV prevalence in rural communities in Khanh Hoa province. Vietnam Sex Transm Infect. 2008;84(6):439–43. 54. Joint United Nations Programme on HIV/AIDS. Global AIDS Update 2021 - Confronting Inequalities: Lessons for Pandemic Responses from 40 Years of AIDS. 2021. 35. Department of Health, editor. Philippine Health Sector HIV Strategic Plan. In: Department of Health, editor. Manila: Department of Health; 2020. 55. Eshun-Wilson I, Jamil MS, Witzel TC, Glidded DV, Johnson C, Le Trouneau N, et al. A systematic review and network Meta-analyses to assess the effec- tiveness of human immunodeficiency virus (HIV) self-testing distribution strategies. Clin Infect Dis. 2021;73(4):e1018–e28. 36. References Sexual network distribu- tion of HIV self-testing kits: findings from the process evaluation of an intervention for men who have sex with men in China. PLoS One. 2020;15(4):e0232094. 6. Ross AG, Ditangco RA, Belimac JG, Olveda RM, Mercado ES, Chau TN, et al. The dire sexual health crisis among MSM in the Philippines: an exploding HIV epidemic in the absence of essential health services. Int J Infect Dis. 2015;37:6–8. 6. Ross AG, Ditangco RA, Belimac JG, Olveda RM, Mercado ES, Chau TN, et al. The dire sexual health crisis among MSM in the Philippines: an exploding HIV epidemic in the absence of essential health services. Int J Infect Dis. 2015;37:6–8. 27. Wang Z, Lau JTF, Ip M, Ho SPY, Mo PKH, Latkin C, et al. A randomized con- trolled trial evaluating efficacy of promoting a home-based HIV self-testing with online counseling on increasing HIV testing among men who have sex with men. AIDS Behav. 2018;22(1):190–201. Page 11 of 11 Eustaquio et al. BMC Public Health (2022) 22:366 Page 11 of 11 Eustaquio et al. BMC Public Health (2022) 22:366 28. Tang W, Wei C, Cao B, Wu D, Li KT, Lu H, et al. Crowdsourcing to expand HIV testing among men who have sex with men in China: A closed cohort stepped wedge cluster randomized controlled trial. PLoS Med. 2018;15(8):e1002645. 47. Lillie T, Boyee D, Kamariza G, Nkunzimana A, Gashobotse D, Persaud N. Increasing testing options for key populations in Burundi through peer- assisted HIV self-testing: descriptive analysis of routine programmatic data. JMIR Public Health Surveill. 2021;7(9):e24272. 48. Okoboi S, Castelnuovo B, Van Geertruyden JP, Lazarus O, Vu L, Kalibala S, et al. Cost-effectiveness of peer-delivered HIV self-tests for MSM in Uganda. Front Public Health. 2021;9:651325. 29. Jamil MS, Prestage G, Fairley CK, Grulich AE, Smith KS, Chen M, et al. Effect of availability of HIV self-testing on HIV testing frequency in gay and bisexual men at high risk of infection (FORTH): a waiting-list randomised controlled trial. Lancet HIV. 2017;4(6):e241–e50. 49. Sha YX, Y; Wang Y; Ong, J; Ni, Y; Lu, Y; Cheng, M. Comparing the effective- ness of secondary distribution of HIV self-testing to testing card referral in promoting HIV testing among men who have sex with men in China: A quasi-experimental study. medRxiv. 2021. 30. Zhang C, Li X, Brecht ML, Koniak-Griffin D. References Girault P, Misa Wong C, Jittjang S, Fongkaew K, Cassell MM, Lertpiriyasu- wat C, et al. Uptake of oral fluid-based HIV self-testing among men who have sex with men and transgender women in Thailand. PLoS One. 2021;16(8):e0256094. 56. Green KE, Vu BN, Phan HT, Tran MH, Ngo HV, Vo SH, et al. From conventional to disruptive: upturning the HIV testing status quo among men who have sex with men in Vietnam. J Int AIDS Soc. 2018;21(Suppl 5):e25127. 37. Burke VM, Nakyanjo N, Ddaaki W, Payne C, Hutchinson N, Wawer MJ, et al. HIV self-testing values and preferences among sex workers, fishermen, and mainland community members in Rakai, Uganda: A qualitative study. PLoS One. 2017;12(8):e0183280. 57. van Dijk M, de Wit JBF, Guadamuz TE, Martinez JE, Jonas KJ. Slow uptake of PrEP: Behavioral predictors and the influence of Price on PrEP uptake among MSM with a high interest in PrEP. AIDS Behav. 2021;25(8):2382–90. 38. Wulandari LPL, Kaldor J, Guy R. Uptake and acceptability of assisted and unassisted HIV self-testing among men who purchase sex in brothels in Indonesia: a pilot intervention study. BMC Public Health. 2020;20(1):730. 58. Sharma A, Chavez PR, MacGowan RJ, McNaghten AD, Mustanski B, Gravens L, et al. Willingness to distribute free rapid home HIV test kits and to test with social or sexual network associates among men who have sex with men in the United States. AIDS Care. 2017;29(12):1499–503. 39. Phanuphak N, Anand T, Jantarapakde J, Nitpolprasert C, Himmad K, Sungsing T, et al. What would you choose: online or offline or mixed services? Feasibility of online HIV counselling and testing among Thai men who have sex with men and transgender women and factors associated with service uptake. J Int AIDS Soc. 2018;21(Suppl 5):e25118.i 59. Ratner B. Variable selection methods in regression: ignorable problem, out- ing notable solution. J Target Meas Anal Mark. 2010;18:65–75. 60. Hosmer D, Lemeshow S, Sturdivant RX. Applied logistic regression. New York: John Wiley & Sons, Inc.; 2013. 40. Waterfield KC, Shah GH, Etheredge GD, Ikhile O. Consequences of COVID-19 crisis for persons with HIV: the impact of social determinants of health. BMC Public Health. 2021;21(1):299. y 61. Steyerberg E. Clinical prediction models: a practical approach to develop- ment, validation, and updating: Springer Science & Business Media; 2008. 41. Iversen J, Sabin K, Chang J, Morgan Thomas R, Prestage G, Strathdee SA, et al. References COVID-19, HIV and key populations: cross-cutting issues and the need for population-specific responses. J Int AIDS Soc. 2020;23(10):e25632. 62. Chowdhury MZI, Turin TC. Variable selection strategies and its impor- tance in clinical prediction modelling. Fam Med Community Health. 2020;8(1):e000262. 63. Austin PC, Steyerberg EW. Events per variable (EPV) and the relative perfor- mance of different strategies for estimating the out-of-sample validity of logistic regression models. Stat Methods Med Res. 2017;26(2):796–808. 42. Tonen-Wolyec S, Kayembe Tshilumba C, Batina-Agasa S, Marini Djang’eing’a R, Hayette MP, Belec L. Comparison of practicability and effectiveness between unassisted HIV self-testing and directly assisted HIV self-testing in the Democratic Republic of the Congo: a randomized feasibility trial. BMC Infect Dis. 2020;20(1):830. 63. Austin PC, Steyerberg EW. Events per variable (EPV) and the relative perfor- mance of different strategies for estimating the out-of-sample validity of logistic regression models. Stat Methods Med Res. 2017;26(2):796–808. Publisher’s Note 43. Phanuphak N, Jantarapakde J, Himmad L, Sungsing T, Meksena R, Pho- mthong S, et al. Linkages to HIV confirmatory testing and antiretroviral therapy after online, supervised, HIV self-testing among Thai men who have sex with men and transgender women. J Int AIDS Soc. 2020;23(1):e25448. Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. 44. Maatouk I, Nakib ME, Assi M, Farah P, Makso B, Nakib CE, et al. Community- led HIV self-testing for men who have sex with men in Lebanon: lessons learned and impact of COVID-19. Health Res Policy Syst. 2021;19(Suppl 1):50. 45. Janssen R, Engel N, Esmail A, Oelofse S, Krumeich A, Dheda K, et al. Alone but supported: A qualitative study of an HIV self-testing app in an observa- tional cohort study in South Africa. AIDS Behav. 2020;24(2):467–74. 46. Okoboi S, Lazarus O, Castelnuovo B, Nanfuka M, Kambugu A, Mujugira A, et al. Peer distribution of HIV self-test kits to men who have sex with men to identify undiagnosed HIV infection in Uganda: A pilot study. PLoS One. 2020;15(1):e0227741.
https://openalex.org/W4244447299	https://www.researchsquare.com/article/rs-148386/latest.pdf	English	null	Association between Liver Enzymes and Dyslipidemia in Yemeni Patients with Type Two Diabetes Mellitus	Research Square (Research Square)	2,021	cc-by	6,033	Association between Liver Enzymes and Dyslipidemia in Yemeni Patients with Type Two Diabetes Mellitus Lotfi S. Bin Dahman (  lotfydahman@hu.edu.ye ) College of Medicine and Health Sciences, Hadhramout University https://orcid.org/0000-0002-1526- 4955 Mariam A. Humam College of Medicine and Health Sciences, Hadhramout University Omar H. Barahim College of Medicine and Health Sciences, Hadhramout University Mohammed A. Balfas College of Medicine and Health Sciences, Hadhramout University Research Article Keywords: liver enzymes, dyslipidemia, type 2 diabetes mellitus, Yemeni Patients Posted Date: March 15th, 2021 DOI: https://doi.org/10.21203/rs.3.rs-148386/v4 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Lotfi S. Bin Dahman (  lotfydahman@hu.edu.ye ) Research Article Page 1/11 Page 1/11 Abstract The current study was aimed to assess the association between liver enzymes and blood lipid profile in a sample of Yemeni patients with T2D. This is a case-control study comprising 142 T2D patients and 142 healthy control subjects at the Ibn-Sina hospital's outpatient clinics in Mukalla, Yemen. Fasting blood glucose (FBG), total cholesterol, triglyceride, high-density lipoprotein cholesterol (HDL-cholesterol), alanine aminotransferase (ALT), aspartate aminotransferase (AST), and gamma-glutamyltransferase (GGT) were determined using Cobas Integra Plus 400 autoanalyzer. Also, anthropometric and blood pressure measurements were taken from each participant. Independent sample T-test and Pearson correlation coefficient were used. T2D patients had significantly higher FBG (P= <0.0001), total cholesterol (P= <0.0001), LDL-cholesterol (P= <0.0001), and GGT (P= <0.0001) while, HDL-cholesterol was significantly lower in T2D patients (P= 0.021). In correlation analysis, serum GGT was positively associated with FBG (r= 0.216; P= <0.0001), total cholesterol (r= 0.196; P= 0.0001), triglyceride (r= 0.123; P= 0.038), and LDL-cholesterol (r= 0.209; P= <0.0001). Also, serum ALT was positively associated with FBG (r= 0.145, P= 0.014) and triglyceride (r= 0.172, P= 0.004). We conclude that higher ALT and GGT are used as the predictive biomarkers for NAFLD in T2D patients with hyperlipidemia. Therefore, routine screening of liver enzymes and lipid profile in T2D patients is recommended to detect liver abnormalities and diminish diabetes complications. 2.1. Study Design and Subjects Selection This is a case-control study carried out at the College of Medicine and Health Sciences, Hadhramout University. The subjects were selected from the diabetic outpatient clinic of Ibn-Sina hospital, Mukalla, during the period from January to May 2020. A total of 284 Yemeni adult subjects, randomly selected and recruited into this study. At recruitment, an in-person interview was conducted using a structured questionnaire to collect health-related information. The study group was subdivided into two groups: 142 healthy control subjects composed of 51 males and 91 females (age: 46.0±7.94 yr.), and 142 T2D patients consisting of 64 males and 78 females (age: 54.0±8.29 yr.). T2D patients were those who reported being diagnosed with T2D. Healthy control subjects were selected from the remaining participants free of T2D and matched for age, sex, and dialect group with cases on a 1:1 ratio. Moreover, the chosen healthy control subjects were screened for the presence of undiagnosed T2D at the time of blood donation by measuring fasting blood glucose (FBG). Written consent was obtained from each participant entered into the study. The current study was approved by the Ethics Committee of the Medicine and Health Sciences College, Hadhramout University, Mukalla, Yemen. Patients with co- morbidities such as chronic liver disease, chronic renal disease, cardiovascular disease, and malignancy were excluded. 2.2. Data Collection A gave questionnaire form focusing on demographic information, and diabetes history was given to all subjects. The patient's demographic data, clinical presentation, medical history, and physical findings were taken from each participant. These data are included: The patient's age, sex, smoking status (never, current or past), hypertension status (yes or no), diabetes status (yes or no), diabetes duration (years), diabetes medication, and diabetes complications. Participants were diagnosed with diabetes according to medical history, present intake of diabetes medications, or the American Diabetes Association (ADA) criteria [14]. Patients with T2DM were defined as fasting blood glucose level ≥126 mg/dl (≥7.0 mmol/L), 2-hour postprandial plasma glucose level ≥200 mg/dl (≥11.1 mmol/L), or HbA1c ≥6.5% [14]. Classification of Body Mass Index (BMI) was based on WHO [15]. 1. Introduction Diabetes mellitus is defined as a metabolic disorder characterized by chronic hyperglycemia, which results from defective insulin action and secretion or both [1]. World Health Organization (WHO) projects that the number of diabetic patients will exceed 350 million by 2030 [1]. Previous data have documented liver disease is a significant cause of morbidity and mortality of type 2 diabetes patients [2] [3]. It is known that the liver is a vital organ in the metabolism of carbohydrates and in maintaining glucose homeostasis during fasting and postprandial period [2] [4]. Non-alcoholic fatty liver disease (NAFLD) is the scope of chronic liver disease in T2D patients [5], which is characterized by excess deposition of fat in the liver and associated with hepatic insulin resistance [3] and T2D risk [5]. Serum alanine aminotransferase (ALT) and gamma-glutamyltransferase (GGT) are promising biomarkers of NAFLD. ALT has been considered the specific liver injury marker, as found in high concentrations in hepatocytes [6]. At the same time, GGT is present on the surface of most cell types and highly active in the liver, kidneys, and pancreas [7]. Also, GGT is responsible for extracellular glutathione catabolism and may be linked to oxidative stress [8] and chronic inflammation [9]; both oxidative stress and chronic inflammation are essential pathways for hepatic insulin resistance (IR) and subsequently T2D development [10]. Accordingly, hyperinsulinemia and IR play an essential role in lipid abnormalities for T2D patients [2] [11]. Also, altered lipoprotein patterns and liver enzymes have been identified as independent risk factors for the development of cardiovascular disease (CVD) [10] [11] [12]. Moreover, higher levels of triglycerides, Page 2/11 Page 2/11 LDL-cholesterol, total cholesterol, and lower HDL-cholesterol levels were reported in T2D patients than normal healthy subjects [13]. However, few studies reported the correlation between liver enzymes and lipid profiles in T2D patients; hence, this case-control study was conducted to assess the association between liver enzymes and blood lipid profiles in a sample of Yemeni patients with T2D. LDL-cholesterol, total cholesterol, and lower HDL-cholesterol levels were reported in T2D patients than normal healthy subjects [13]. However, few studies reported the correlation between liver enzymes and lipid profiles in T2D patients; hence, this case-control study was conducted to assess the association between liver enzymes and blood lipid profiles in a sample of Yemeni patients with T2D. 2.5. Statistical Analysis Data were analyzed by using the Statistical Package for the Social Sciences for Windows (version 24) and are expressed by mean ± standard deviation (SD) for continuous variables (normally distributed). Non-continuous variables are represented by median (inter-quartile range) and n (percentage) for categorical variables. Independent Student's t-test used for normally distributed continuous variables and Wilcoxon signed-rank test for skewed continuous variables. The Pearson correlation test was performed with ALT, AST, and GGT as the dependent variables. The statistical analysis was conducted at a 95% confidence level, and a P-value <0.05 was considered statistically significant. 2.3. Anthropometric and Blood Pressure Measurements Weight and height were measured following measured according to WHO guidelines [15]. Body mass index (BMI) was calculated as weight/height2 (Kg/m2). Obese subjects were defined as BMI ≥30 kg/m2 and normal-weight subjects having a BMI of 18–25 according to WHO guidelines [15]. Patients who had Page 3/11 Page 3/11 a blood pressure of ≥140/90 mmHg or were taking antihypertensive medications were diagnosed with hypertension [16]. A true healthy normal ALT level ranges from 29 to 33 IU/l for males and 19 to 25 IU/l for females, and levels above this should be assessed as described by the American College of Gastroenterology (ACG) [17]. a blood pressure of ≥140/90 mmHg or were taking antihypertensive medications were diagnosed with hypertension [16]. A true healthy normal ALT level ranges from 29 to 33 IU/l for males and 19 to 25 IU/l for females, and levels above this should be assessed as described by the American College of Gastroenterology (ACG) [17]. 2.4. Biochemical Investigations Ten milliliters of the venous blood sample was obtained from consenting subjects. The blood samples were collected by vein puncture in tubes without anticoagulant. The blood samples were then transported to the laboratory immediately. The serum was then separated and stored at −20°C freezers till analyses. The serum samples of matched case–control pairs were randomly placed next to each other with the case/control status blinded to the laboratory personnel and were processed and tested in the same batch. All laboratory equipment was calibrated. Thawing freezing was avoided by dividing the samples into aliquots. Plasma fasting blood glucose (FBG), total cholesterol, triglycerides, and HDL-cholesterol were determined enzymatically using a chemical autoanalyzer (Cobas Integra 400 Plus, Roche diagnostic GmbH, Mannheim, Switzerland), following the standard procedures as described by the manufacturer. Concentrations of LDL-cholesterol were calculated using Friedwald's formula [18]. All biochemical investigations were analyzed in the National Center for Public Health Labs–Mukalla, Yemen. 3. Results And Discussion Descriptive statistics of anthropometric and biochemical data of the study population are presented in table 1. T2D patients had significantly increased BMI (P= 0.008), systolic BP (P= <0.0001), diastolic BP (P= <0.0001), FBG (P= <0.0001), total cholesterol (P= <0.0001), LDL-cholesterol (P= <0.0001), and GGT (P= 0.016) as compared to healthy control subjects. We found no significant difference in serum triglyceride (P= 0.097) and ALT (P= 0.07). Healthy control subjects had significantly increased HDL- cholesterol (P= 0.021) and AST (P= 0.001) compared to T2D patients. On the other hand, 31.7% of T2D patients had hypertension, whereas 6.3% of healthy control subjects had hypertension. Besides, in T2D patients, the current smokers were 4.2%, and the former smokers were 3.5%. According to BMI criteria, 38.7% of T2D patients had overweight, and 24.6% with obese compared to healthy control subjects (40.1%, 14.1%). Page 4/11 Pearson correlation using ALT, AST, and GGT as dependent variables is presented in table 2. ALT was positively associated with FBG (r= 0.145, P= 0.014), triglyceride (r= 0.172, P= 0.004), AST (r= 590, P= <0.001), and GGT (r= 0.507, P= <0.001) respectively. Serum GGT was positively associated with systolic BP (r= 0.134, P= 0.024), diastolic BP (r= 0.218, P= <0.001), FBG (r= 0.216, P= <0.0001), total cholesterol (r= 0.196, P= 0.0001), triglyceride (r= 0.123, P= 0.038), LDL-cholesterol (r= 0.209, P= <0.0001), and AST (r= 0.366, P= <0.0001) across the combined group. Pearson correlation using ALT, AST, and GGT as dependent variables is presented in table 2. ALT was positively associated with FBG (r= 0.145, P= 0.014), triglyceride (r= 0.172, P= 0.004), AST (r= 590, P= <0.001), and GGT (r= 0.507, P= <0.001) respectively. Serum GGT was positively associated with systolic BP (r= 0.134, P= 0.024), diastolic BP (r= 0.218, P= <0.001), FBG (r= 0.216, P= <0.0001), total cholesterol (r= 0.196, P= 0.0001), triglyceride (r= 0.123, P= 0.038), LDL-cholesterol (r= 0.209, P= <0.0001), and AST (r= 0.366, P= <0.0001) across the combined group. Using partial correlation analysis (table 3), controlling for age and BMI, significant positive association between ALT with AST (r= 0.589, P= <0.0001) and ALT (r= 0.514, P= <0.0001) remained significant across the combined group, whilst, the association between ALT with FBG and triglyceride was no longer significant. 3. Results And Discussion Using the same analysis, the association between GGT with systolic BP (r= 0.124, P= 0.038), diastolic BP (r= 0.213, P= <0.0001), FBG (r= 0.213, P= <0.0001), total cholesterol (r= 0.199, P= 0.001), triglyceride (r= 0.127, P= 0.033), and LDL-cholesterol (r= 0.208, P= <0.0001) remained significant before and after age and BMI as adjustment. Table 1. Anthropometric and biochemical data of healthy controls and T2D patients Variables Healthy controls T2D patients P-value N 142 142 Age (years) 46.0±7.94 54.0±8.29 <0.0001 Sex: male/female 51(35.9)/91(64.1) 63(44.4)/78(54.9) Weight (kg) 71.12±10.67 69.61±13.83 <0.0001 Height (cm) 164.57±8.47 159.97±10.04 <0.0001 BMI (kg/m2) 26.31±3.95 27.21±4.94 0.008 Systolic BP (mmHg) 115.28±13.11 128.80±20.92 <0.0001 Diastolic BP (mmHg) 70.45±9.02 79.47±9.90 <0.0001 BMI classification: Normal weight Overweight Obese 65(45.8) 57(40.1) 20(14.1) 52(36.6) 55(38.7) 35(24.6) History of hypertension: Yes/no 9(6.3)/133(93.7) 45(31.7)/97(68.3) Smoking status: Never Smoker Current Smoker Former Smoker 142(100) 0(0) 0(0) 131(92.3) 6(4.2) 5(3.5) FBG (mmol/L) 5.18±0.91 8.91±2.89 <0.0001 Total cholesterol (mmol/L) 4.70±0.77 5.16±1.20 <0.0001 Triglyceride (mmol/L) 1.24±0.37 1.16±0.42 0.097 HDL-cholesterol (mmol/L) 1.67±0.42 1.57±0.34 0.021 LDL-cholesterol (mmol/L) 2.77±0.80 3.35±1.17 0.001 ALT (IU/L) 13.1(8.37-19.3) 11.6(7.3-16.8) 0.07 AST (IU/L) 21.2(17.8-28.7) 16.4(13.3-21.7) 0.001 GGT (IU/L) 25.1(16.8-34.7) 29.2(18.4-49.7) <0.0001 Data were presented as mean±SD for normal continuous variables and median (interquartile range) for continuous non-norma variables. Independent sample T-test for normally distributed continuous variables and Mann-Whitney U test for skewed continuous variables. P-value <0.05 was considered statistically significant. BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; FBG, fasting blood glucose; HDL-C, high- density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, gamma-glutamyltransferase. Table 1. Anthropometric and biochemical data of healthy controls and T2D patients Data were presented as mean±SD for normal continuous variables and median (interquartile range) for continuous non-normal variables. Independent sample T-test for normally distributed continuous variables and Mann-Whitney U test for skewed continuous variables. P-value <0.05 was considered statistically significant. BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; FBG, fasting blood glucose; HDL-C, high- density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, gamma-glutamyltransferase. ALT (IU/L) AST (IU/L) GGT (IU/L) Pearson correlation using ALT, AST and GGT as dependent variables in the combined groups studied after Age and BMI adjustment as a covariance g , p g p g adjustment as a covariance N= 284 ALT AST GGT r P-value r P-value r P-value Sex (M/F) 0.116 0.051 0.114 0.055 0.017 0.772 Systolic BP (mmHg) -0.018 0.764 -0.053 0.377 0.124* 0.038 Diastolic BP (mmHg) 0.024 0.686 -0.078 0.194 0.213 0<.0001 FBG (mmol/L) 0.161 0.007 -0.074 0.213 0.213 <0.0001 Total cholesterol (mmol/L) 0.027 0.652 0.085 0.155 0.199 0.001 Triglyceride (mmol/L) 0.171 0.004 0.090 0.130 0.127 0.033 HDL-cholesterol (mmol/L) -0.104 0.081 -0.026 0.668 -0.056 0.351 LDL-cholesterol (mmol/L) 0.052 0.388 0.087 0.147 0.208 <0.0001 ALT (IU/L) 0.589 <0.0001 0.514 <0.0001 AST (IU/L) 0.589 <0.0001 0.368 <0.0001 GGT (IU/L) 0.514 <0.0001 0.368 <0.0001 Pearson correlation coefficient with corresponding p-value (p<0.05 is significant). ** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed). BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; FBG, fasting blood glucose; HDL-C, high- density lipoprotein cholesterol; LDL-cholesterol, low-density lipoprotein cholesterol; ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, gamma-glutamyltransferase. Correlation is significant at the 0.05 level (2 tailed). BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; FBG, fasting blood glucose; HDL-C, high- density lipoprotein cholesterol; LDL-cholesterol, low-density lipoprotein cholesterol; ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, gamma-glutamyltransferase. ALT (IU/L) AST (IU/L) GGT (IU/L) Data were presented as mean±SD for normal continuous variables and median (interquartile range) for continuous non-normal variables. Independent sample T-test for normally distributed continuous variables and Mann-Whitney U test for skewed continuous variables. P-value <0.05 was considered statistically significant. BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; FBG, fasting blood glucose; HDL-C, high- density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, gamma-glutamyltransferase. Page 5/11 Page 5/11 Table 2. Pearson correlation using ALT, AST and GGT as dependent variables in the combined study group N= 284 ALT AST GGT r P-value r P-value r P-value Age (years) -0.046 0.443 0.010 0.860 0.047 0.433 Sex (M/F) 0.119* 0.046 0.116 0.050 0.016 0.789 Weight (kg) -0.001 0.989 -0.008 0.895 -0.004 0.945 Height (cm) 0.098 0.101 0.071 0.235 -0.058 0.334 BMI (kg/m2) -0.070 0.241 -0.059 0.326 0.033 0.538 Systolic BP (mmHg) -0.037 0.533 -0.058 0.334 0.134* 0.024 Diastolic BP (mmHg) 0.013 0.830 -0.080 0.178 0.218** <0.001 FBG (mmol/L) 0.145* 0.014 -0.067 0.260 0.216** <0.0001 Total cholesterol (mmol/L) 0.027 0.653 0.081 0.176 0.196* 0.0001 Triglyceride (mmol/L) 0.172** 0.004 0.087 0.141 0.123* 0.038 HDL-cholesterol (mmol/L) -0.091 0.124 -0.023 0.699 -0.064 0.285 LDL-cholesterol (mmol/L) 0.047 0.429 0.082 0.170 0.209 <0.0001 ALT (IU/L) 0.590 <0.0001 0.507 <0.0001 AST (IU/L) 0.590 <0.0001 0.366 <0.0001 GGT (IU/L) 0.507 <0.0001 0.366 <0.0001 Pearson correlation coefficient with corresponding p-value (p<0.05 is considered a significant). Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed). BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; FBG, fasting blood glucose; TC, total cholesterol; TG, triglyceride; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, gamma-glutamyltransferase. correlation using ALT, AST and GGT as dependent variables in the combined study group Table 2. Pearson correlation using ALT, AST and GGT as dependent variables in the combined study group Correlation is significant at the 0.05 level (2 tailed). BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; FBG, fasting blood glucose; TC, total cholesterol; TG, triglyceride; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, gamma-glutamyltransferase. Table 3. Pearson correlation using ALT, AST and GGT as dependent variables in the combined groups studied after Age and BMI adjustment as a covariance Table 3. Correlation is significant at the 0.05 level (2 tailed). BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; FBG, fasting blood glucose; TC, total cholesterol; TG, triglyceride; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, gamma-glutamyltransferase. 4. Discussion Although the incidence of diabetes is increasing worldwide and its prevalence is higher in developing countries, no studies have examined the relationship between elevated liver enzymes and T2D risk in Yemeni patients. Our research, therefore, was focused on the liver as the vital organ contributing to glucose homeostasis during the fasting and postprandial stage. Also, most people aged ≥45 years in Page 6/11 Page 6/11 developing countries have diabetes [19]. These findings were convenient with our study showed that T2D patients had significantly higher mean age compared to healthy control subjects (Table 1). developing countries have diabetes [19]. These findings were convenient with our study showed that T2D patients had significantly higher mean age compared to healthy control subjects (Table 1). Our present findings also observed significantly increased BMI, systolic BP, and diastolic BP in T2D patients than healthy control subjects. The present study also showed that serum FBG, total cholesterol, and LDL-cholesterol were significantly higher in T2D patients than healthy control subjects. At the same time, there is no significant difference found among both groups for serum triglyceride. In contrast, HDL- cholesterol was significantly lower in T2D patients. Our study further revealed higher levels of GGT in T2D patients. In contrast, AST was significantly lower in T2D patients. Besides, no significant difference was found among both groups for ALT. Such a positive relationship between liver enzymes and blood lipids profile in T2D patients has been observed in previous studies [2] 20] [21] [22] [23]. This finding supports the role of hepatic insulin resistance in NAFLD's pathogenesis in patients with T2D [24] [25]. Moreover, Cho et al. reported a correlation between ALT activity and increased fatty liver [26]. The impairment of the normal process of synthesis and elimination of triglycerides may progress to fibrosis, cirrhosis, and hepatocellular carcinoma [27] [28]. In addition to its effect on lipid metabolism, insulin also contributes a pro-inflammatory effect to liver abrasion [29]. Thus, inflammation contributes to IR. Moreover, Pro-inflammatory cytokines and transcription factors are highly expressed in white adipose tissue and the liver. In contrast, obesity, a state of chronic low-grade inflammation and a risk factor for IR and NAFLD, is induced by overnutrition. It is a primary cause of decreased insulin sensitivity. al. also suggested that these aminotransferases' mild or chronic elevations may be due to NAFLD [40] [41] al. also suggested that these aminotransferases' mild or chronic elevations may be due to NAFLD [40] [41]. The present study's strength included adjustment for well-established diabetes risk factors, including BMI, blood lipids, and hypertension. However, there are some limitations: Our sample size may be small and thus underpowered to detect the interaction with ALT and GGT. We measured liver enzymes only once and may not represent a long-term profile. We did not measure hepatitis B and C infection, resulting in elevated liver enzymes. We did not measure insulin, CRP, leptin, and adiponectin as the predictive biomarkers links between obesity, hepatic IR, and related inflammatory disorders in T2D patients. Thus, a further large sample size with measurement of insulin, CRP, leptin, adiponectin, and interleukins is required to confirm these correlations. We conclude that higher ALT and GGT are used as the predictive biomarkers for NAFLD in T2D patients with hyperlipidemia. Conflicts of Interest The authors declare no conflicts of interest. Acknowledgments The authors are grateful to Al-Huda Medical Agency, Mukalla, Yemen, for funding the study and the Ibn- Sina Hospital, Mukalla, Yemen for technical support. Also, we are thankful to the physicians and nurses who recruited and collected the data of the participants. Also, thanks to the National Center for Public Health Laboratories – Mukalla for the biochemical investigations performance. Special thanks to Students of Medical Laboratory Sciences Department (Ebrahim Al-Muhamedi, Ali Alqaaiti, Saleh Daiban, Sabri Barafah, Afaf Aldibani, Safa Basawaid, and Noor Zahfan) for the performance of data and sample collection, data entry, and biochemical investigations. Special thanks to Ms. Nasiba Al-Aidros for the statistical analysis. 5. Conclusion Higher levels of ALT and GGT may be used as the predictive markers for NAFLD in T2D patients with hyperlipidemia. Thus, routine screening of liver enzymes and lipid profile in T2D patients is recommended for the early detection of liver abnormalities and diminish diabetes complications. 4. Discussion Obesity leads to lipid accumulation and activates the c- Jun N-terminal kinase (JNK) and nuclear factor-kappa B (NF-κB) signaling pathways, which consequently increase the production of pro-inflammatory cytokines, such as tumor necrosis factor-alpha (TNF-α) and interleukin-6 (IL-6) [30]. Besides, various adipose tissue-derived proteins, such as adiponectin and leptin, are considered significant links between obesity, IR, and related inflammatory disorders [31]. GGT is known as a marker of hepatobiliary disorders and is associated with other pathological conditions like diabetes. Free radicals generated by diabetes consume glutathione which induces the increased expression of GGT in hepatocytes. Various studies have suggested the association of GGT concentra- tions with T2D [32] [33] [34] [35], and hyperlipidemia [36]. These findings agree with our study; GGT was significantly associated with the hyperglycemic and hyperlipidemia profile. We observed ALT and GGT together were positively correlated. Moreover, some data also reported elevated GGT levels with ALT in T2D patients with dyslipidemia [33] [34] [37]. Even though we did not confirm the presence of fatty liver by ultrasound techniques, we showed the relationship of ALT, AST, and GGT with the predictors of diabetes and lipid profile parameters, presenting hepatocellular injury. GGT is known as a marker of hepatobiliary disorders and is associated with other pathological conditions like diabetes. Free radicals generated by diabetes consume glutathione which induces the increased expression of GGT in hepatocytes. Various studies have suggested the association of GGT concentra- tions with T2D [32] [33] [34] [35], and hyperlipidemia [36]. These findings agree with our study; GGT was significantly associated with the hyperglycemic and hyperlipidemia profile. We observed ALT and GGT together were positively correlated. Moreover, some data also reported elevated GGT levels with ALT in T2D patients with dyslipidemia [33] [34] [37]. Even though we did not confirm the presence of fatty liver by ultrasound techniques, we showed the relationship of ALT, AST, and GGT with the predictors of diabetes and lipid profile parameters, presenting hepatocellular injury. A study of male Korean workers found that AST was independently associated with diabetes [38], while in a study of male Japanese office workers, AST was not associated with T2D risk [34]. Some studies also reported that ALT is a significant predictor of diabetes while AST is not [39]. Our data agree with our findings as AST does not show considerable relationship with the studied parameters. Besides, Clark et Page 7/11 Page 7/11 References 1. World Health Organization (1999) Definition, diagnosis and classification of diabetes mellitus and its complications: Report of a WHO consultation. Geneva: World Health Organization, 1999. https://apps.who.int/iris/handle/10665/66040. Page 8/11 Page 8/11 Page 8/11 2. Al-Jameil, N., Khan, F. A., Arjumand, S., et al. (2014) Associated liver enzymes with hyperlipidemic profile in type 2 diabetes patients. Int J Clin Exp Pathol, 7, 4345-4349. [PubMed]. 3. Hanley, A.J., Williams K., Festa, A., et al. (2004) Elevations in marker of liver injury and risk of type 2 diabetes-The insulin resistance atherosclerosis study. Diabetes, 53, 2623-2632. https://doi.org/10.2337/diabetes.53.10.2623. 4. Gavin, N. and Levinthal, A.S.T. (1999) Liver disease and diabetes mellitus. Clin Diabetes, 17. 4. Gavin, N. and Levinthal, A.S.T. (1999) Liver disease and diabetes mellitus. Clin Diabetes, 17. 5. Ballestri, S., Zona, S., Targher, , et al. (2016) Nonalcoholic fatty liver disease is associated with an almost twofold increased risk of incident type 2 diabetes and metabolic syndrome. Evidence from a systematic review and meta-analysis. J Gastroenterol Hepatol, 31, 936–44. https://doi.org/10.1111/jgh.13264. 6. Giannini, E.G., Testa, R. and Savarino, V. (2005) Liver enzyme alteration: a guide for clinicians. CMAJ, 172, 367–379. https://doi.org/10.1503/cmaj.1040752. 6. Giannini, E.G., Testa, R. and Savarino, V. (2005) Liver enzyme alteration: a guide for clinicians. CMAJ, 172, 367–379. https://doi.org/10.1503/cmaj.1040752. 7. Hanigan, M.H. and Frierson, H.F., Jr. (1996) Immunohistochemical detection of gamma-glutamyl transpeptidase in normal human tissue. J Histochem Cytochem, 44, 1101–11108. https://doi.org/10.1177/44.10.8813074. 7. Hanigan, M.H. and Frierson, H.F., Jr. (1996) Immunohistochemical detection of gamma-glutamyl transpeptidase in normal human tissue. J Histochem Cytochem, 44, 1101–11108. https://doi.org/10.1177/44.10.8813074. 8. Turgut, O. and Tandogan, I. (2011) Gamma-glutamyltransferase to determine cardiovascular risk: shifting the paradigm forward. J Atheroscler Thromb, 18, 177–181. https://doi.org/10.5551/jat.6189. 8. Turgut, O. and Tandogan, I. (2011) Gamma-glutamyltransferase to determine cardiovascular risk: shifting the paradigm forward. J Atheroscler Thromb, 18, 177–181. https://doi.org/10.5551/jat.6189. 9. Lee, D.H. and Jacobs, D.R., Jr. (2005) Association between serum gamma-glutamyltransferase and C- reactive protein. Atherosclerosis, 178, 327–330. https://doi org/10 1016/j atherosclerosis 2004 08 027 9. Lee, D.H. and Jacobs, D.R., Jr. (2005) Association between serum gamma-glutamyltransferase and C- reactive protein. Atherosclerosis, 178, 327–330. https://doi.org/10.1016/j.atherosclerosis.2004.08.027. 10. Ye-Li, W., Woon-Puay, K., Jian-Min, Y. and An, P. (2016) Association between liver enzymes and incident type 2 diabetes in Singapore Chinese men and women. BMJ Open Diabetes Research and Care, 4, e000296. https://doi.org/10.1136/bmjdrc-2016-000296. 10. Ye-Li, W., Woon-Puay, K., Jian-Min, Y. and An, P. References (1972) Estimation of the concentration of low- density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem, 18, 499–502. https://doi.org/10.1093/clinchem/36.1.15. 18. Friedewald, W.T., Levy, R.I. and Fredrickson, D.S. (1972) Estimation of the concentration of low- density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem, 18, 499–502. https://doi.org/10.1093/clinchem/36.1.15. 19. Nwarfor, A. and Owhoji, A. (2001) The prevalence of diabetes mellitus in port-Harcourt correlates with the socio-economic status. J Appl Sci Environ Mgt, 5, 75-77. [Google Scholar]. 19. Nwarfor, A. and Owhoji, A. (2001) The prevalence of diabetes mellitus in port-Harcourt correlates with the socio-economic status. J Appl Sci Environ Mgt, 5, 75-77. [Google Scholar]. 20. Jain, H.R., Shetty, V., Singh, G.S. and Shetty, S. (2016) A Study of Lipid Profile in Diabetes Mellitus. Int J Sci Stud, 4, 56-61. [Google Scholar]. 20. Jain, H.R., Shetty, V., Singh, G.S. and Shetty, S. (2016) A Study of Lipid Profile in Diabetes Mellitus. Int J Sci Stud, 4, 56-61. [Google Scholar]. 21. Han, N., Soe, H.K. and Htet, A. (2012) Determinants of Abnormal Liver Function Tests in Diabetes Patients in Myanmar. Int J Diab Res, 1, 36-41. https://doi.org/10.5923/j.diabetes.20120103.02. 21. Han, N., Soe, H.K. and Htet, A. (2012) Determinants of Abnormal Liver Function Tests in Diabetes Patients in Myanmar. Int J Diab Res, 1, 36-41. https://doi.org/10.5923/j.diabetes.20120103.02. 22. Belay, Z., Daniel, S., Tedla, K. and Gnanasekaran, N. (2014) Impairment of liver function tests and lipid profiles in type 2 diabetic patients treated at the diabetic center in Tikur Anbessa specialized teaching hospital (Tasth), Addis Ababa, Ethiopia. J Diabetes Metab, 5, 454. https://doi.org/10.4172/2155-6156.1000454. 22. Belay, Z., Daniel, S., Tedla, K. and Gnanasekaran, N. (2014) Impairment of liver function tests and lipid profiles in type 2 diabetic patients treated at the diabetic center in Tikur Anbessa specialized teaching hospital (Tasth), Addis Ababa, Ethiopia. J Diabetes Metab, 5, 454. https://doi.org/10.4172/2155-6156.1000454. 23. Adeniran, S.A., Dolapo, P.O., Oluwole, A.B., et al. (2013) Liver Enzymes and Lipid Profile Among Type 2 Diabetic Patients in Osogbo, Nigeria. Greener J Med Sci, 3, 174-178. https://doi.org/10.15580/GJMS.2013.5.011313373. 24. Tolman, K.G., Fonseca, V., Dalpiaz, A., et al. (2007) Spectrum of liver disease in type 2 diabetes and management of patients with diabetes and liver disease. Diabetes Care, 30, 734-743. https://doi.org/10.2337/dc06-1539. 25. Nannipieri, M., Gonzales, C., Baldi, S., et al. (2005) Liver enzymes, the metabolic syndrome, and incident diabetes: The Mexico City diabetes study. Diabetes Care, 28, 1757-1762. https://doi.org/10.2337/diacare.28.7.1757. 26. References (2016) Association between liver enzymes and incident type 2 diabetes in Singapore Chinese men and women. BMJ Open Diabetes Research and Care, 4, e000296. https://doi.org/10.1136/bmjdrc-2016-000296. 11. Ginsberg, H.N., Zhang, Y.L. and Hernandez-Ono, A. (2006) Metabolic syndrome: focus on dyslipidemia. Obesity (Silver Spring), 14 Suppl 1:41S-49S. https://doi.org/10.1038/oby.2006.281. 11. Ginsberg, H.N., Zhang, Y.L. and Hernandez-Ono, A. (2006) Metabolic syndrome: focus on dyslipidemia. Obesity (Silver Spring), 14 Suppl 1:41S-49S. https://doi.org/10.1038/oby.2006.281. 12. Balaji, A.S. (2014) Serum alanine transaminases and lipid profile in type 2 diabetes mellitus Indian patient. J Diab Res, [Epub ahead of print]. https://doi.org/10.5171/2013.613176. 12. Balaji, A.S. (2014) Serum alanine transaminases and lipid profile in type 2 diabetes mellitus Indian patient. J Diab Res, [Epub ahead of print]. https://doi.org/10.5171/2013.613176. 13. Rajeswari, S., Kumar, A., Gandhi, M. and Swaminathan, S. (2014) Association between Lipid Profile and Liver Function Tests in Diabetic Patients. Int J Pure App Biosci, 2(4):26-31. 13. Rajeswari, S., Kumar, A., Gandhi, M. and Swaminathan, S. (2014) Association between Lipid Profile and Liver Function Tests in Diabetic Patients. Int J Pure App Biosci, 2(4):26-31. 14. American Diabetes Association (2016) Standards of medical care in diabetes-2016. Diabetes Care, 39, 101–106. https://doi.org/10.2337/dc16-S003. 14. American Diabetes Association (2016) Standards of medical care in diabetes-2016. Diabetes Care, 39, 101–106. https://doi.org/10.2337/dc16-S003. Organization (1995) Physical status: the use and interpretation of anthropom 15. World Health Organization (1995) Physical status: the use and interpretation of anthropometry (1995) Report of WHO expert committee. WHO Technical Report Series, no. 854. Geneva: WHO, 321– 344. https://apps.who.int/iris/handle/10665/37003. 15. World Health Organization (1995) Physical status: the use and interpretation of anthropometry (1995) Report of WHO expert committee. WHO Technical Report Series, no. 854. Geneva: WHO, 321– 344. https://apps.who.int/iris/handle/10665/37003. 16. Chobanian, A.V., Bakris, G.L., Black, H.R., et al. (2003) Seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension, 42, 1206– 1252. https://doi.org/10.1161/01.HYP.0000107251.49515.c2. 16. Chobanian, A.V., Bakris, G.L., Black, H.R., et al. (2003) Seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension, 42, 1206– 1252. https://doi.org/10.1161/01.HYP.0000107251.49515.c2. Page 9/11 Page 9/11 17. Kwo, P.Y., Cohen, S.M. and Lim, J.K. (2017) ACG clinical guideline: evaluation of abnormal liver chemistries. Am J Gastroenterol, 112, 18–35. https://doi.org/10.1038/ajg.2016.517. 17. Kwo, P.Y., Cohen, S.M. and Lim, J.K. (2017) ACG clinical guideline: evaluation of abnormal liver chemistries. Am J Gastroenterol, 112, 18–35. https://doi.org/10.1038/ajg.2016.517. 18. Friedewald, W.T., Levy, R.I. and Fredrickson, D.S. References Cho, N.H., Jang, H.C., Choi, S.H., et al. (2007) Abnormal liver function test predicts type 2 diabetes: a community-based prospective study. Diabetes Care, 30, 2566–2568. https://doi.org/10.2337/dc07- 0106. 27. Tolman, K.G., Fonseca, V., Tan, M.H. and Dalpiaz, A. (2004) Hepatobiliary disease in type 2 diabetes mellitus. Ann Intern Med, 141, 946-956. https://doi.org/10.7326/0003-4819-141-12-200412210- 00011. 27. Tolman, K.G., Fonseca, V., Tan, M.H. and Dalpiaz, A. (2004) Hepatobiliary disease in type 2 diabetes mellitus. Ann Intern Med, 141, 946-956. https://doi.org/10.7326/0003-4819-141-12-200412210- 00011. 28. Chatila, R. and West, A.B. (1996) Hepatomegaly and abnormal liver tests due to glycogenesis in adults with diabetes. Med Balt, 75, 327-333. https://doi.org/10.1097/00005792-199611000-00003. 28. Chatila, R. and West, A.B. (1996) Hepatomegaly and abnormal liver tests due to glycogenesis in adults with diabetes. Med Balt, 75, 327-333. https://doi.org/10.1097/00005792-199611000-00003. 29. Balaji, A.S. (2014) Serum alanine transaminases and lipid profile in type 2 diabetes mellitus Indian patient. J Diab Res, [Epub ahead of print]. https://doi.org/10.5171/2013.613176. 29. Balaji, A.S. (2014) Serum alanine transaminases and lipid profile in type 2 diabetes mellitus Indian patient. J Diab Res, [Epub ahead of print]. https://doi.org/10.5171/2013.613176. 30. Sharma, M., Vikram, N.K., Misra, A., et al. (2013) Assessment of 11-beta hydroxysteroid dehydrogenase (11-betaHSD1) 4478T> G and tumor necrosis factor-alpha (TNF-alpha)-308G>A polymorphisms with obesity and insulin resistance in Asian Indians in North India, Mol Biol Rep, 40, 6261–70. [PubMed]. 30. Sharma, M., Vikram, N.K., Misra, A., et al. (2013) Assessment of 11-beta hydroxysteroid dehydrogenase (11-betaHSD1) 4478T> G and tumor necrosis factor-alpha (TNF-alpha)-308G>A polymorphisms with obesity and insulin resistance in Asian Indians in North India, Mol Biol Rep, 40, 6261–70. [PubMed]. Page 10/11 Page 10/11 31. Chen, Z., Yu, R., Xiong, Y., et al. (2017) A vicious circle between insulin resistance and inflammation in nonalcoholic fatty liver disease, Lipids in Health and Disease, 16, 203-211. [PubMed]. 32. Lee, D.H., Silventonein, K., Jacobs, D.R., et al. (2014) Gamma glutamyltransferase, obesity and the risk of type 2 diabetes observational cohort study among 20,158 middle aged men and women. J Clin Endocrinol Metab, 89, 5410-5414. https://doi.org/10.1210/jc.2004-0505. 32. Lee, D.H., Silventonein, K., Jacobs, D.R., et al. (2014) Gamma glutamyltransferase, obesity and the risk of type 2 diabetes observational cohort study among 20,158 middle aged men and women. J Clin Endocrinol Metab, 89, 5410-5414. https://doi.org/10.1210/jc.2004-0505. 33. Lee, D.H., Ha, M.H., Kim, J.H., et al. (2003) Gamma glutamyltransferase and diabetes-a four year follow up study. Diabetologia, 46, 359-364. https://doi.org/10.1007/s00125-003-1036-5. 33. Lee, D.H., Ha, M.H., Kim, J.H., et al. References (2003) Gamma glutamyltransferase and diabetes-a four year follow up study. Diabetologia, 46, 359-364. https://doi.org/10.1007/s00125-003-1036-5. 34. Nakanishi, N., Suzuki, K. and Tatara, K. (2004) Serum gamma glutamyltransferase and risk of metabolic syndrome and type 2 diabetes in middle aged Japanese men. Diabetes Care, 27, 1427- 1432. https://doi.org/10.2337/diacare.27.6.1427. 34. Nakanishi, N., Suzuki, K. and Tatara, K. (2004) Serum gamma glutamyltransferase and risk of metabolic syndrome and type 2 diabetes in middle aged Japanese men. Diabetes Care, 27, 1427- 1432. https://doi.org/10.2337/diacare.27.6.1427. 35. Lee, D.H., Jacobs, D.R., Gross, M., Kiefe CI, et al. (2003) Gamma glutamyltransferase is a predictor of incident diabetes and hypertension: the Coronary Artery Risk Development in Young Adults (CARDIA) Study. Clin Chem, 49, 1358-1366. https://doi.org/10.1373/49.8.1358. 35. Lee, D.H., Jacobs, D.R., Gross, M., Kiefe CI, et al. (2003) Gamma glutamyltransferase is a predictor of incident diabetes and hypertension: the Coronary Artery Risk Development in Young Adults (CARDIA) Study. Clin Chem, 49, 1358-1366. https://doi.org/10.1373/49.8.1358. 36. Sakuta, H., Suzuki, T., Yasuda, H. and Ito, T. (2005) Gamma glutamyltransferase and airflow obstruction in middle-aged men. Eur J Intern Med, 16, 348-351. https://doi.org/10.1016/j.ejim.2005.06.005. 36. Sakuta, H., Suzuki, T., Yasuda, H. and Ito, T. (2005) Gamma glutamyltransferase and airflow obstruction in middle-aged men. Eur J Intern Med, 16, 348-351. https://doi.org/10.1016/j.ejim.2005.06.005. 37. Marchesini, G., Brizi, M., Bianchi, G., et al. (2001) Nonalcoholic fatty liver disease: a feature of the metabolic syndrome. Diabetes, 50, 1844-1850. https://doi.org/10.2337/diabetes.50.8.1844. 37. Marchesini, G., Brizi, M., Bianchi, G., et al. (2001) Nonalcoholic fatty liver disease: a feature of the metabolic syndrome. Diabetes, 50, 1844-1850. https://doi.org/10.2337/diabetes.50.8.1844. 38. Ahn, H.R., Nam, H.S., Park, K.S., et al. (2014) The association between liver enzymes and risk of type 2 diabetes: the Namwon study. Diabetol Metab Syndr, 6, 14. https://doi.org/10.1186/1758-5996-6-14. 38. Ahn, H.R., Nam, H.S., Park, K.S., et al. (2014) The association between liver enzymes and risk of type 2 diabetes: the Namwon study. Diabetol Metab Syndr, 6, 14. https://doi.org/10.1186/1758-5996-6-14. 39. Vozarova, B., Stefan, N., Lindsay, R.S., et al. (2002) High alanine aminotransferase is associated wit decreased hepatic insulin sensitivity and predicts the development of type 2 diabetes. Diabetes, 51, 1889–1895. https://doi.org/10.2337/diabetes.51.6.1889. 39. Vozarova, B., Stefan, N., Lindsay, R.S., et al. (2002) High alanine aminotransferase is associated with decreased hepatic insulin sensitivity and predicts the development of type 2 diabetes. Diabetes, 51, 1889–1895. https://doi.org/10.2337/diabetes.51.6.1889. 40. Clark, J.M. and Diehl, A.M. (2003) Nonalcoholic fatty liver disease: an underrecognized cause of cryptogenic cirrhosis. JAMA, 290, 3000-3004. https://doi.org/10.1001/jama.289.22.3000. 40. Clark, J.M. References and Diehl, A.M. (2003) Nonalcoholic fatty liver disease: an underrecognized cause of cryptogenic cirrhosis. JAMA, 290, 3000-3004. https://doi.org/10.1001/jama.289.22.3000. 41. Clark, J.M., Brancati, F.L. and Diehl, A.M. (2003) The prevalence and etiology of elevated aminotransferase levels in the United States. Am J Gastroenterol, 98, 960-967. https://doi.org/10.1111/j.1572-0241.2003.07486.x. 41. Clark, J.M., Brancati, F.L. and Diehl, A.M. (2003) The prevalence and etiology of elevated aminotransferase levels in the United States. Am J Gastroenterol, 98, 960-967. https://doi.org/10.1111/j.1572-0241.2003.07486.x. Page 11/11 Page 11/11
https://openalex.org/W3009032614	https://hal.archives-ouvertes.fr/hal-02486559/file/2020-Cuello_BioForA-FiPS_AtrFtirMicroscpectroscopy_WoodCWatCellLevel.pdf	English	null	ATR-FTIR Microspectroscopy Brings a Novel Insight Into the Study of Cell Wall Chemistry at the Cellular Level	Frontiers in plant science	2,020	cc-by	11,660	To cite this version: Clément Cuello, Paul Marchand, Françoise F. Laurans, Camille Grand-Perret, Veronique Laine-Prade, et al.. ATR-FTIR Microspectroscopy Brings a Novel Insight Into the Study of Cell Wall Chemistry at the Cellular Level. Frontiers in Plant Science, 2020, 11, pp.105. 10.3389/fpls.2020.00105. hal- 02486559 Distributed under a Creative Commons Attribution 4.0 International License HAL Id: hal-02486559 https://hal.science/hal-02486559v1 Submitted on 21 Feb 2020 L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. Distributed under a Creative Commons Attribution 4.0 International License ORIGINAL RESEARCH published: 21 February 2020 doi: 10.3389/fpls.2020.00105 In conclusion, ATR-FTIR microspectroscopy offers new possibilities for studying cell wall composition at the cell level Reviewed by: Lennart Salmén, Research Institutes of Sweden (RISE), Sweden Barbara Hinterstoisser, University of Natural Resources and Life Sciences Vienna, Austria Frédéric Jamme, Soleil Synchrotron, France Reviewed by: Lennart Salmén, Research Institutes of Sweden (RISE), Sweden Barbara Hinterstoisser, University of Natural Resources and Life Sciences Vienna, Austria Frédéric Jamme, Soleil Synchrotron, France *Correspondence: Annabelle Déjardin annabelle.dejardin@inrae.fr †Present address: Paul Marchand, INRAE, AgroParisTech, Université Paris-Saclay, ECOSYS, Thiverval-Grignon, France Specialty section: This article was submitted to Technical Advances in Plant Science, a section of the journal Frontiers in Plant Science Received: 19 November 2019 Accepted: 23 January 2020 Published: 21 February 2020 Specialty section: This article was submitted to Technical Advances in Plant Science, a section of the journal Frontiers in Plant Science Specialty section: This article was submitted to Technical Advances in Plant Science, a section of the journal Frontiers in Plant Science Received: 19 November 2019 Accepted: 23 January 2020 Published: 21 February 2020 Edited by: Andras Gorzsas, Umeå University, Sweden Edited by: Andras Gorzsas, Umeå University, Sweden Edited by: Andras Gorzsas, Umeå University, Sweden Wood is a complex tissue that fulﬁlls three major functions in trees: water conduction, mechanical support and nutrient storage. In Angiosperm trees, vessels, ﬁbers and parenchyma rays are respectively assigned to these functions. Cell wall composition and structure strongly varies according to cell type, developmental stages and environmental conditions. This complexity can therefore hinder the study of the molecular mechanisms of wood formation, underlying the construction of its properties. However, this can be circumvented thanks to the development of cell-speciﬁc approaches and microphenotyping. Here, we present a non-destructive microphenotyping method based on attenuated total reﬂectance–Fourier transformed infrared (ATR-FTIR) microspectroscopy. We applied this technique to three types of poplar wood: normal wood of staked trees (NW), tension and opposite wood of artiﬁcially tilted trees (TW, OW). TW is produced by angiosperm trees in response to mechanical strains and is characterized by the presence of G ﬁbers, exhibiting a thick gelatinous extra- layer, named G-layer, located in place of the usual S2 and/or S3 layers. By contrast, OW located on the opposite side of the trunk is totally deprived of ﬁbers with G-layers. We developed a workﬂow for hyperspectral image analysis with both automatic pixel clustering according to cell wall types and identiﬁcation of differentially absorbed wavenumbers (DAWNs). As pixel clustering failed to assign pixels to ray S-layers with sufﬁcient efﬁciency, the IR proﬁling and identiﬁcation of DAWNs were restricted to ﬁber and vessel cell walls. As reported elsewhere, this workﬂow identiﬁed cellulose as the main component of the G-layers, while the amount in acetylated xylans and lignins were shown to be reduced. These results validate ATR-FTIR technique for in situ characterization of G layers. In addition, this study brought new information about IR proﬁling of S-layers in TW, OW and NW. While OW and NW exhibited similar proﬁles, TW ﬁbers S-layers combined characteristics of TW G-layers and of regular ﬁber S-layers. Unexpectedly, vessel S-layers of the three kinds of wood showed signiﬁcant differences in IR proﬁling. Keywords: ATR-FTIR microspectroscopy, phenotyping, poplar, cell wall, G-layer, wood INTRODUCTION Indeed, the wavenumbers emitted by infrared beams are differentially absorbed according to the chemical bonds, therefore the functional groups, present in the sample. Cell wall composition may be determined by prediction equations based on near-infrared absorbance spectra of ground plant materials. This technique has been successfully used to predict cell wall composition in Arabidopsis thaliana (Jasinski et al., 2016), forage crops (Fairbrother and Brink, 1990; Molano et al., 2016; Baldy et al., 2017; Li et al., 2017), rice (Huang et al., 2017) and poplar (Gebreselassie et al., 2017). Mid-infrared (MIR) spectroscopy is complementary to predictive near-infrared (NIR) spectroscopy. Indeed, MIR spectroscopy targets the fundamental vibrations of the molecules, which gives a good visualization of the molecular bonds present in the sample (Bertrand and Dufour, 2006; Allison, 2011) and also to assign variations in the absorbance at a given wavenumber to a class of compounds (McCann et al., 1992; Kacurakova and Wilson, 2001). For example, FTIR spectroscopy was used to discriminate between different Arabidopsis cell wall mutants (Chen et al., 1998). FTIR spectroscopy coupled with microscopy imaging makes possible in situ analysis of cell wall assembly (Mouille et al., 2003; Gierlinger et al., 2008; Gorzsás et al., 2011; Chazal et al., 2014; Gierlinger, 2018). Wood cell walls exhibit a multi-layered structure, determinant for wood mechanical properties: middle lamella and primary cell wall are overlaid by three layers of secondary cell walls (SCW), named S1, S2, and S3, S2 being the thickest. Each layer results from the assembly of semi-crystalline cellulose microﬁbrils embedded in an amorphous matrix of polysaccharides and lignins. They differ according to cellulose microﬁbril orientation and the nature and proportion of the matrix components (pectins, hemicelluloses and lignins) (Mellerowicz and Sundberg, 2008). The middle lamella and primary cell wall are known to be rich in lignins, pectins and xyloglucans, with apparently cellulose microﬁbrils oriented at random. By contrast, the S-layers are rich in xylans with oriented cellulose microﬁbrils: the angle formed between the cell axis and cellulose microﬁbrils, named microﬁbril angle (MFA) is wide in the S1- and S3-layers, while it is fairly low in the S2-layer (Barnett and Bonham, 2004). At the end of cell wall deposition, ligniﬁcation occurs, leading to the impermeabilization of the cell walls. INTRODUCTION Angiosperm trees produce tension wood on the upper side of inclined trunks or branches in order to reorient them or at least to reach a new equilibrium. In poplar, tension wood ﬁbers exhibit a thick gelatinous extra-layer, named G-layer (Jourez, 1997). Located in place of the usual S2 and/or S3 layers, G-layer is rich in cellulose, rhamnogalacturonan type I pectins (RG-I), and arabinogalactan proteins (AGP) (Gorshkova et al., 2015; Guedes et al., 2017), with nearly no lignin (Joseleau et al., 2004; Pilate et al., 2004; Gierlinger and Schwanninger, 2006). In G- layer, cellulose is more crystalline and MFA is close to 0° (Norberg and Meier, 1966; Gorshkova et al., 2015). Trees are able to live long and reach a considerable height, partly thanks to the remarkable properties of their wood. Indeed, wood —or secondary xylem—fulﬁlls three major functions in trees: (i) water conduction from the roots to the crown, (ii) mechanical support of the ever-increasing mass of the growing tree submitted to different environmental cues (wind, slope, light, …), and (iii) storage of temporary reserves, important for tree perennial growth (Plomion et al., 2001; Déjardin et al., 2010). In Angiosperm trees, vessels, ﬁbers and parenchyma rays are respectively assigned to each of these functions. These different xylem cell types originate from the differentiation of cambial cells. When reaching their ﬁnal dimensions, xylem cells build a secondary cell wall that will be deposited onto the primary wall and middle lamella. Vessels are rapidly submitted to programmed cell death after ligniﬁcation of their cell walls, then ﬁbers have the same destiny (Courtois‐Moreau et al., 2009), while ray cells remain alive for several years (Nakaba et al., 2012). Therefore, wood results in the complex assembly of the cell walls of dead ﬁbers and vessels, connected to living parenchyma rays (Déjardin et al., 2010). Thus, wood is both a highly complex and variable tissue, which does not facilitate the study of the molecular mechanisms underlying its formation and the building of its properties. However, this hindrance may be alleviated thanks to the development of cell-speciﬁc approaches and microphenotyping. Particularly, the development of microphenotyping methods may provide extensive biochemical information at the cell wall level. In the last decade, Fourier Transform InfraRed (FTIR) spectroscopy made possible high throughput biochemical analyses of lignocellulosic biomass. Citation: Cuello C, Marchand P, Laurans F, Grand-Perret C, Lainé-Prade V, Pilate G and Déjardin A (2020) ATR- FTIR Microspectroscopy Brings a Novel Insight Into the Study of Cell Wall Chemistry at the Cellular Level. Front. Plant Sci. 11:105. doi: 10.3389/fpls.2020.00105 Keywords: ATR-FTIR microspectroscopy, phenotyping, poplar, cell wall, G-layer, wood February 2020 \| Volume 11 \| Article 105 1 Frontiers in Plant Science \| www.frontiersin.org Microphenotyping Using ATR-FTIR Cuello et al. Frontiers in Plant Science \| www.frontiersin.org INTRODUCTION Lignins are phenolic polymers merely resulting, in hardwood species, from the polymerization of two classes of monolignols (coniferyl alcohol or G unit and synapyl alcohol or S unit) that differ in their degree of methoxylation (Freudenberg and Neish, 1968). They are synthesized in the cytoplasm, then transported to the cell wall, turned into chemical radicals by the activity of laccases and peroxidases and probably get polymerized in a random manner. Lignin structure is therefore highly diverse and difﬁcult to predict (van Parijs et al., 2010). In this methodological study, we present a non-destructive high-throughput microphenotyping method based on ATR-FTIR microspectroscopy, one of the most resolving methods currently available. The objective was to develop a method to obtain IR spectra speciﬁc to wood cell types, ﬁbers, vessels and rays. Therefore, we have developed a high-throughput pipeline for hyperspectral image analysis with both automatic pixel clustering and identiﬁcation of differentially absorbed wavenumbers between samples. We have implemented this pipeline on different types of wood, including tension wood, a model that has been extensively characterized from a biochemical point of view. We conﬁrmed previously known results, which validates this in situ approach, and we generated for the ﬁrst time an IR proﬁle for vessel walls in three different types of wood. Wood cell walls are highly variable according to developmental processes or various environmental stresses. First, it is known that secondary cell wall composition differs between vessels, ﬁbers and ray cells (e.g. Donaldson et al., 2001). The cell wall of vessel elements is enriched in G-units while ﬁber cell wall is richer in S-units (Terashima et al., 1993). Wood cell wall composition can also be affected by various environmental cues. For example, in response to mechanical strains, February 2020 \| Volume 11 \| Article 105 2 Microphenotyping Using ATR-FTIR Cuello et al. for each type of wood using R (v. 3.4.4) (Figure 1). Principal component analysis (PCA) was applied on raw spectra using PCA function implemented in FactoMineR package (v. 1.40). False RGB images were then obtained using quilt.plot (ﬁelds, v.9.6). Red, blue and green intensities were based on pixel contribution to ﬁrst, second and third principal component, respectively. Plant Material and Sample Preparation S p p The study was carried out on six three-month-old ramets, originating from in vitro micropropagated shoots of the INRA 717-1B4 clone (Populus tremula L. x Populus alba L.). Two ramets were grown in a greenhouse in spring 2011 and the four others in the autumn and winter 2012. In the latter case, the plants were supplemented with sufﬁcient amount of artiﬁcial light to keep the cambium active. In order to induce the production of TW, three of these trees (TT_11-A in 2011, TT_12-A and TT_12-B in 2012) were tilted at 45° two months before sampling. The three other poplar plants (UT_11-C, UT_12-C and UT_12-D) were grown upright in the same conditions and produce no or very few TW. Three-centimeter-long stem fragments were sampled at the base of each tree. NW was collected on upright trees, TW and OW on the upper and lower side of artiﬁcially tilted trees, respectively. The samples were stored at –20°C until use. For each tree, 20 μm-thick cross-sections were cut from the frozen wood samples using a RM2155 microtome (LEICA, Wetzlar, Germany). The sections were ethanol-dried between two glass slides to ensure ﬂatness and stored at 23°C and 45% RH until ATR-FTIR analysis. INTRODUCTION Based on these false RGB images, speciﬁc mean representative spectrum (MRS) were built for each cell wall class, from the average of 90 selected spectra (that is 30 spectra manually selected in one of the three images of each biological replicate of each wood type). Both MRS and pixel spectra were corrected using the detrend function implemented in the prospectr package (v. 0.1.3). Correlations between each pixel spectrum and MRS of its type of wood were determined using the Spearman correlation test implemented in cor.test (stats, v.3.5.2). The correlation score was used to assign pixels to a cell wall class or to lumen. A pixel was considered as lumen when its correlation scores with all MRS were below the threshold value of 0.90. Pixels with very similar correlation scores with all MRS (less than 1% difference) were classiﬁed as “not assigned” (NA). Finally, the pixel was assigned to the cell wall class whose MRS has the highest correlation score with the pixel. Statistical Analysis For each cell type, a PCA was carried out on the whole spectra to discriminate between the different types of wood. Loadings were calculated as follow: Loadings = Eigenvector ﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ Eigenvalues p . Due to the huge size of the data set, pixels corresponding to the cell wall of each cell type in each image were averaged to produce heatmaps using heatmap.2 function (gplots package, v. 3.0.1.1), scaling on columns. Heatmaps dendrograms were calculated considering Pearson’s correlation between wavenumbers as a distance. Multivariate Image Analysis We analyzed ten classes of cell wall corresponding to the existing associations between cell wall layers (G- or S-layers), cell types (ﬁber, ray or vessel) and wood types (TW, OW or NW): TW ﬁber G-layer, TW ﬁber S-layer, OW ﬁber S-layer, NW ﬁber S- layer, TW ray S-layer, OW ray S-layer, NW ray S-layer, TW vessel S-layer, OW vessel S-layer and NW vessel S-layer. Multivariate image analysis has been performed independently Hyperspectral Imaging ATR-FTIR images were produced from stem cross-sections by a mid-IR light (1,800–850 cm–1) at a 4 cm–1 spectral resolution using a Spotlight 400 FTIR imaging system coupled to a Spectrum 400 FTIR spectrophotometer (PERKIN ELMER, Wellesley, USA). According to the manufacturer’s procedure, the wood section was pushed by pressure into direct contact with the tip of the 600-μm diameter plane Germanium crystal. Supplementary Figure 1 shows the contact area after pressure release showing the intimate contact of the sample with the crystal. Sixteen scans per pixel were taken in order to enhance signal-to-noise ratio. For each type of wood (TW, OW and NW), three (100 x 100) μm² images from each cross section were taken at a (1.56 x 1.56) μm² pixel size. The acquisition time of an image of this size was 40 minutes. This maximal pixel resolution corresponded to an oversampling factor of two, compared with the diffraction-limited spatial resolution of 3.1 μm. At this pixel resolution, the image size was 4,096 pixels. Therefore, for each type of wood, the data set was composed of 36,864 spectra (3 different trees x 3 images x 4,096 pixels). The spectra were then corrected using different functions of the SpectrumImage software (v.1.6.4): background correction during acquisition, noise reduction and atmospheric correction, using default parameters. No ATR correction was performed. DAWNs were determined using the pairwise Wilcoxon test from the R stats package (v.3.5.2). A Bonferroni adjusted P-value of 0.001 was used as cut-off criterion. Only DAWNs located at a local maximum (± 3 cm–1) were considered for further analysis. Local maxima were identiﬁed on the average spectra resulting from all pixels assigned to a class of cell wall. A single wavenumber (s) was considered as a local maximum when its absorbance was higher compared to s-2 and s+2 cm–1. The assignation of DAWNs to cell wall components was based on an in-lab database (Table S1) compiling data from the literature and from the Histochem database (Durand et al., 2019; https://pﬂ. grignon.inra.fr/shistochem/). RESULTS (D) Ninety pixels were manually attributed to each of ray, ﬁber and vessel cell walls and to G-layers. (E) Mean representative spectra (MRS), which is the pixel average of each of these classes, and pixel spectra were detrend. Correlations between pixel spectra and MRS were determined using Spearman’s correlation test. Pixel were assigned to a given class using a 0.9 correlation score threshold. Green: vessel, tan: ray, red: ﬁber, black: lumen, white: NA. (F) Differentially absorbed wavenumbers (DAWNs) were identiﬁed using pairwise Wilcoxon test. A Bonferroni adjusted P-value of 0.001 was used as cut-off criterion. types from the different types of wood. Classic clustering methods, such as hierarchical clustering based on principal components, Ascending Hierarchical Classiﬁcation, k- nearest neighbor and k-means were not conclusive. The most effective clustering method was the assignation based on Spearman’s correlation score between pixel spectra and a mean representative spectrum (MRS) for each class of cell wall. categorized as ray S-layers, three as ﬁber S-layers, four as lumen and seven were not assigned. Conversely, the wrong assignment of ray spectra was rather high reaching 52% in OW, 28% in NW and 38% in TW. , Subsequently, we used this workﬂow to assign the 36,864 spectra from the pixels of each type of wood. In NW, 50% of the spectra were categorized as ﬁber S-layers, 14% as ray S- layers, 9% as vessel S-layers and 23% as lumen, while 4% of the spectra could not be assigned. In OW, 41% of the spectra corresponded to ﬁber S-layers, 16% to ray S-layers and 11% to vessel S-layers, while 26% of the spectra, with a correlation score below 0.90, were considered as lumen. Finally, 6% of the spectra with similar correlation levels to MRS from all classes, were not assigned and therefore excluded from the analysis. In TW, the ﬁber G-layer was the largest class with 44% of the spectra. The ﬁber S-layers, ray S-layers, vessel S-layers and lumen categories were composed of 34%, 8%, 2% and 12% of the spectra, respectively. Interestingly, unlike OW and NW, all TW spectra have been assigned (Figure 2B). The comparison between the raw IR images and the corresponding cluster maps (Figure 3) revealed a correct adjustment especially for ﬁbers and vessels, indicative of the efﬁciency of the clustering method used to predict these two cell types. RESULTS From Raw Infra-Red Images to Pixel Clusters of the Different Cell Wall Types One of the most important steps of the workﬂow was to produce a good assignation of the pixels to the different cell From Raw Infra-Red Images to Pixel Clusters of the Different Cell Wall Types One of the most important steps of the workﬂow was to produce a good assignation of the pixels to the different cell February 2020 \| Volume 11 \| Article 105 Frontiers in Plant Science \| www.frontiersin.org 3 Microphenotyping Using ATR-FTIR Cuello et al. FIGURE 1 \| Workﬂow. Sample preparation. Poplar trees were grown in a greenhouse for three months. (A) 20 μm-thick ethanol dried cross section were produced. Image acquisition. (B) (100 x 100) μm² images were acquired from 20 μm thick cross sections using an infrared (IR) microscope. Background subtraction, noise reduction and atmospheric correction were applied on raw IR images using the SpectrumImage software (v.1.6.4). Multivariate image analyses were carried out using homemade scripts encoded in R (v. 3.4.4). (C) Principal component analyses (PCA) were performed on raw spectra. RGB intensities were derived from pixel contribution to PC1 for Red, PC2 for Green and PC3 for Blue. (D) Ninety pixels were manually attributed to each of ray, ﬁber and vessel cell walls and to G-layers. (E) Mean representative spectra (MRS), which is the pixel average of each of these classes, and pixel spectra were detrend. Correlations between pixel spectra and MRS were determined using Spearman’s correlation test. Pixel were assigned to a given class using a 0.9 correlation score threshold. Green: vessel, tan: ray, red: ﬁber, black: lumen, white: NA. (F) Differentially absorbed wavenumbers (DAWNs) were identiﬁed using pairwise Wilcoxon test. A Bonferroni adjusted P-value of 0.001 was used as cut-off criterion. FIGURE 1 \| Workﬂow. Sample preparation. Poplar trees were grown in a greenhouse for three months. (A) 20 μm-thick ethanol dried cross section were produced. Image acquisition. (B) (100 x 100) μm² images were acquired from 20 μm thick cross sections using an infrared (IR) microscope. Background subtraction, noise reduction and atmospheric correction were applied on raw IR images using the SpectrumImage software (v.1.6.4). Multivariate image analyses were carried out using homemade scripts encoded in R (v. 3.4.4). (C) Principal component analyses (PCA) were performed on raw spectra. RGB intensities were derived from pixel contribution to PC1 for Red, PC2 for Green and PC3 for Blue. Frontiers in Plant Science \| www.frontiersin.org RESULTS The pixels that could not be assigned mainly corresponded to pixels at the interface between cell types. As the assignation for ray S-layers was not very efﬁcient, IR proﬁling was restricted to ﬁber and vessel SCW. We applied this method to the spectra used to calculate the MRS of the different classes of cell wall (Table S2). When applied to pixels with a known assignment, our method correctly assigned 48% to 100% of the pixels depending on the cell wall category (Figure 2A). All the spectra from TW ﬁber G-layer were properly assigned, whereas, no spectrum from any TW S-layer class was wrongly assigned to the G-layer class. Likewise, all spectra of OW ﬁber S-layers were correctly assigned, while only three of them were identiﬁed as ray S- layers in NW and two of them were also identiﬁed as ray S- layers in TW. Our method correctly assigned a large part of vessel spectra: 78% in OW, 82% in NW and 73% in TW. In TW, nearly all wrongly assigned vessel spectra (23 out of 24) were classiﬁed as ray S-layers. In NW, only six vessel spectra were classiﬁed as ray S-layers, while four others had a correlation score below 0.90 with all MRS and were therefore classiﬁed as lumen. Finally, six spectra that presented similar correlation values to all MRS were in consequence not assigned. In OW, six vessel spectra were February 2020 \| Volume 11 \| Article 105 Frontiers in Plant Science \| www.frontiersin.org Microphenotyping Using ATR-FTIR Cuello et al. FIGURE 2 \| Assignation of pixel to cell-wall categories. (A) Clustering efﬁciency evaluated on 90 manually assigned pixels per cell wall type. Red: Normal wood (NW), Black: Opposite wood (OW), Pigeon blue: Tension wood (TW), FibS: ﬁber S-layers, FibG: TW ﬁber G-layers, Ves: vessel S-layers, Ray: ray S-layers. (B) Distribution of the 36,864 pixels in cell wall classes after assignation using Spearman’s correlation score between pixel spectra and mean representative spectra. NW, normal wood; OW, opposite wood; TW, tension wood; Red, ﬁber S-layers; Blue, TW ﬁber G-layer; Green, vessel S-layers; Tan, ray S-layers; Black, lumen; Grey, not assigned. FIGURE 2 \| Assignation of pixel to cell-wall categories. (A) Clustering efﬁciency evaluated on 90 manually assigned pixels per cell wall type. Red: Normal wood (NW), Black: Opposite wood (OW), Pigeon blue: Tension wood (TW), FibS: ﬁber S-layers, FibG: TW ﬁber G-layers, Ves: vessel S-layers, Ray: ray S-layers. IR Proﬁling of Fiber and Vessel SCW Fibers as a shoulder on the spectra of NW and OW ﬁber S-layers. The spectrum of TW ﬁber S-layers is intermediate, in particular with weak bands at 1,506 cm–1, resembling the spectra of NW and OW; it also presents the G-layer characteristic bands at 1,136, 1,316, and 1,052 cm–1. In terms of absorbance differences, the bands at 1,736 and 1,236 cm–1 mainly differentiate TW and OW/ NW ﬁbers, with a higher absorbance in the latter (Figure 4B): they can be mainly attributed to acetylated xylans, and to a lesser extent to lignins (Table S1). The PCA loading plots underline the spectral features associated with the ﬁrst and second dimensions, which are responsible for most of the variance (Supplementary Figure S3A). The samples were discriminated only by the ﬁrst dimension (Figure 4A). The loading for PC1 is quite complex and cannot be easily interpreted by simple changes in few compounds. The loading is positive mostly for six ranges of absorbance bands (1,576–14, 1,492–74, 1,446–28, 1,412–1,384, 1,360–1,280, 1,014–992 cm–1). The ﬁrst four ranges of bands do not correspond to known IR bands, while the two last groups may correspond to cellulose (crystalline cellulose at 1,335 cm–1 and 1,318–12 cm–1, C–O valence vibration at 996–85 cm–1) and lignins (1,330–24 cm–1, S ring plus G ring condensed). The loading is negative for four ranges of bands (1,758–08, 1,268– 06, 1,136–16, 1,098–68 cm–1). The ﬁrst three groups contain bands attributed to hemicelluloses, cellulose and lignins (Table S1). In order to identify more precisely differences between cell wall, we have identiﬁed DAWNs in the vicinity of a local maximum (± 3 cm–1), that were further used to generate a heatmap (Figure 4C, Table S3). The analysis was limited to these DAWNs because the most relevant biological information in an IR spectrum is carried by the wavenumbers at the level of the most intense bands. The heatmap shows 2 clusters of DAWNs that have a very contrasted absorbance proﬁle between TW G-layer and S-layers of OW or NW ﬁbers, while Fibers The ﬁber spectra, assigned by automatic clustering, were ﬁrst analysed by PCA. When performed on all the images, PCA clearly discriminated three groups of spectra: TW ﬁber G-layer, TW ﬁber S-layers and a group combining spectra from both NW and OW ﬁber S-layers. RESULTS (B) Distribution of the 36,864 pixels in cell wall classes after assignation using Spearman’s correlation score between pixel spectra and mean representative spectra. NW, normal wood; OW, opposite wood; TW, tension wood; Red, ﬁber S-layers; Blue, TW ﬁber G-layer; Green, vessel S-layers; Tan, ray S-layers; Black, lumen; Grey, not assigned. Frontiers in Plant Science \| www.frontiersin.org IR Proﬁling of Fiber and Vessel SCW Fibers However, in the latter group, two populations of spectra were detected, with the minor one gathering spectra all coming from the same 4 images (3 from NW and 1 from OW) (Supplementary Figure S2). These 4 images were therefore removed from the subsequent analysis as they may induce unwanted variability for technical or biological reasons. A new PCA was performed (Figure 4A), whose ﬁrst dimension explained 42.5% of the variance and discriminated rather well TW ﬁber G-layer from the ﬁber S-layers of both OW and NW, whereas TW ﬁber S-layers stood in between. Figure 4B shows the average infrared spectrum of pixels assigned to the four types of ﬁber layers. The TW ﬁber G-layer spectrum shows signiﬁcant changes compared to the OW and NW ﬁber S-layers spectra, including the disappearance of two bands at 1,594 and 1,506 cm–1, corresponding to aromatic skeletal vibration of lignins (Table S1) and the appearance of a well-resolved double band at 1,336 and 1,316 cm–1, that can be attributed to cellulose (Table S1, O-H in-plane bending and CH2 rocking vibration, respectively). It is also characterized by the appearance of a band at 1200 cm–1, that can be attributed to OH in-plane deformation in cellulose. This band is masked in the other S- layers because of the overlap of a more intense band at 1236 cm–1, corresponding to acetyl and carboxyl vibration in xylan and to C-C, C-O and C=O stretch in lignins. Finally, a last band is clearly visible in TW ﬁber G and S-layers at 1,052 cm–1, corresponding to C-O valence vibration, mainly from C3-O3H, most probably from cellulose. This latter band is detectable only February 2020 \| Volume 11 \| Article 105 Frontiers in Plant Science \| www.frontiersin.org Frontiers in Plant Science \| www.frontiersin.org 5 Microphenotyping Using ATR-FTIR Cuello et al. TW ﬁber S-layer has an intermediate proﬁle. Cluster B gathers DAWNs with a high absorbance in NW and OW S-layers and a low absorbance in TW S- and G-layers. On the reverse, Cluster E groups DAWNs with a high absorbance in TW G-layer and to a lesser extent in TW S-layers and a low absorbance in NW and OW S-layers. Cluster B contains DAWNs in the range of 1,742–28 and 1,238–32 cm–1 related mainly to acetylated xylans, and in the range of 1,596–90 and 1,508–04 cm–1 related to lignins. IR Proﬁling of Fiber and Vessel SCW Fibers Surprisingly, this latter group was weakly absorbed in NW ﬁber S-layers, compared to OW ﬁber S-layers. Cluster B also contains FIGURE 3 \| Comparison of raw images (A–C) and cluster map (D–F) of tension wood (A, D), opposite wood (B, E) and normal wood (C, F) using one representative image per type of wood. Note that these images were not the ones used to determine MRS. (A–C) In raw images, the intensity in red reﬂects the mean absorbance of the pixel. (D–F) Red, ﬁber S-layers; Blue, ﬁber G-layer; Green, vessel S-layers; Tan, ray S-layers; White, NA; Black, lumen. FIGURE 3 \| Comparison of raw images (A–C) and cluster map (D–F) of tension wood (A, D), opposite wood (B, E) and normal wood (C, F) using one representative image per type of wood. Note that these images were not the ones used to determine MRS. (A–C) In raw images, the intensity in red reﬂects the mean absorbance of the pixel. (D–F) Red, ﬁber S-layers; Blue, ﬁber G-layer; Green, vessel S-layers; Tan, ray S-layers; White, NA; Black, lumen. FIGURE 3 \| Comparison of raw images (A–C) and cluster map (D–F) of tension wood (A, D), opposite wood (B, E) and normal wood (C, F) using one representative image per type of wood. Note that these images were not the ones used to determine MRS. (A–C) In raw images, the intensity in red reﬂects the mean absorbance of the pixel. (D–F) Red, ﬁber S-layers; Blue, ﬁber G-layer; Green, vessel S-layers; Tan, ray S-layers; White, NA; Black, lumen. S-layers. Cluster B contains DAWNs in the range of 1,742–28 and 1,238–32 cm–1 related mainly to acetylated xylans, and in the range of 1,596–90 and 1,508–04 cm–1 related to lignins. Surprisingly, this latter group was weakly absorbed in NW ﬁber S-layers, compared to OW ﬁber S-layers. Cluster B also contains TW ﬁber S-layer has an intermediate proﬁle. Cluster B gathers DAWNs with a high absorbance in NW and OW S-layers and a low absorbance in TW S- and G-layers. On the reverse, Cluster E groups DAWNs with a high absorbance in TW G-layer and to a lesser extent in TW S-layers and a low absorbance in NW and OW February 2020 \| Volume 11 \| Article 105 Frontiers in Plant Science \| www.frontiersin.org 6 Microphenotyping Using ATR-FTIR Cuello et al. IR Proﬁling of Fiber and Vessel SCW Fibers IGURE 4 \| Chemical differences between the G-layer from tension wood ﬁbers and the S-layers from ﬁbers of tension, opposite and normal wood. (A) PCA score ot. Blue, TW ﬁber G-layer; Turquoise, TW ﬁber S-layers; Purple, OW ﬁber S-layers; Red, NW ﬁber S-layers. Ellipses encompass 95% of the data in a normal stribution. Numbers in brackets refer to the number of pixels assigned to the different cell wall categories. Color intensity reﬂects the density of individuals. (B) verage ATR-FTIR spectra. Blue, TW ﬁber G-layer; Turquoise, TW ﬁber S-layers; Purple, OW ﬁber S-layers; Red, NW ﬁber S-layers. The green dotted line insertion epresents a zoom of the [1,510–1,300] cm–1 region. (C) Heatmap of differentially absorbed wavenumbers. TW_FibG, TW ﬁber G-layer, TW_FibS; TW ﬁber S-layers; W_FibS, OW ﬁber S-layers; NW_FibS, NW ﬁber S-layers. FIGURE 4 \| Chemical differences between the G-layer from tension wood ﬁbers and the S-layers from ﬁbers of tension, opposite and normal wood. (A) PCA score plot. Blue, TW ﬁber G-layer; Turquoise, TW ﬁber S-layers; Purple, OW ﬁber S-layers; Red, NW ﬁber S-layers. Ellipses encompass 95% of the data in a normal distribution. Numbers in brackets refer to the number of pixels assigned to the different cell wall categories. Color intensity reﬂects the density of individuals. (B) Average ATR-FTIR spectra. Blue, TW ﬁber G-layer; Turquoise, TW ﬁber S-layers; Purple, OW ﬁber S-layers; Red, NW ﬁber S-layers. The green dotted line insertion represents a zoom of the [1,510–1,300] cm–1 region. (C) Heatmap of differentially absorbed wavenumbers. TW_FibG, TW ﬁber G-layer, TW_FibS; TW ﬁber S-layers; OW_FibS, OW ﬁber S-layers; NW_FibS, NW ﬁber S-layers. from C-N stretching, Table S1). Finally, some groups of DAWNs cannot be easily related to known IR assignment (1,790–74, 1,696– 92, 1,562, 1,540–30, 1,478–74 cm–1). from C-N stretching, Table S1). Finally, some groups of DAWNs cannot be easily related to known IR assignment (1,790–74, 1,696– 92, 1,562, 1,540–30, 1,478–74 cm–1). DAWNs in the range of 1,248–1,244 cm–1, that may be attributed to lignins (stretching of phenolics), and in the range of 1,464–58 cm–1, where both cellulose, lignins and xylans may contribute (CH2 of pyran ring symmetric scissoring; OH- and CH- deformation). Cluster E contains DAWNs that may correspond to crystalline cellulose (1,428–26, 1,336–1,314 cm–1, which appear as a double band in the TW G-layer spectra). Likewise, it contains DAWNs in the range of 1,282–78 and 1,162–56 cm–1 that may be also attributed to cellulose. IR Proﬁling of Fiber and Vessel SCW Fibers Cluster E contains DAWNs in the range of 1,056–50 and 1,036–30 cm–1, that can be attributed to cellulose, hemicelluloses (and also to lignins for the latter group). Interestingly, it contains a group of DAWNs at 1,518–12 cm–1 that can be attributed to lignins (aromatic skeletal vibration, higher absorbance in G lignin in comparison to S lignin, Table S1) and one in the range of 1,556–48 cm–1 than can be attributed to proteins (Amide II: N-H deformation + stretching contribution DAWNs in the range of 1,248–1,244 cm–1, that may be attributed to lignins (stretching of phenolics), and in the range of 1,464–58 cm–1, where both cellulose, lignins and xylans may contribute (CH2 of pyran ring symmetric scissoring; OH- and CH- deformation). Cluster E contains DAWNs that may correspond to crystalline cellulose (1,428–26, 1,336–1,314 cm–1, which appear as a double band in the TW G-layer spectra). Likewise, it contains DAWNs in the range of 1,282–78 and 1,162–56 cm–1 that may be also attributed to cellulose. Cluster E contains DAWNs in the range of 1,056–50 and 1,036–30 cm–1, that can be attributed to cellulose, hemicelluloses (and also to lignins for the latter group). Interestingly, it contains a group of DAWNs at 1,518–12 cm–1 that can be attributed to lignins (aromatic skeletal vibration, higher absorbance in G lignin in comparison to S lignin, Table S1) and one in the range of 1,556–48 cm–1 than can be attributed to proteins (Amide II: N-H deformation + stretching contribution Figure 5 shows images reconstructed on the basis of the absorbance level for each pixel at a given wavenumber. Three wavenumbers were chosen: 1,736 cm–1 (acetylated xylans), 1,316 cm–1 (crystalline cellulose), and 1,236 cm–1 (acetylated xylans and lignins). In particular, the images clearly show the absorbance gradient between ﬁber G-layer and S-layer in TW. from C-N stretching, Table S1). Finally, some groups of DAWNs cannot be easily related to known IR assignment (1,790–74, 1,696– 92, 1,562, 1,540–30, 1,478–74 cm–1). Vessels PCA analysis on vessel S-layers spectra slightly discriminates the spectra from TW, OW and NW, according to the ﬁrst two dimensions, explaining 36 and 17.7% of the variance (Figure 6A). When the ﬁrst dimension partially separates TW vessel S- layers from both OW and NW vessel S-layers, the second one February 2020 \| Volume 11 \| Article 105 Frontiers in Plant Science \| www.frontiersin.org 7 Microphenotyping Using ATR-FTIR Cuello et al. FIGURE 5 \| Infrared images of selected DAWNs of representative images of NW (A, D, G), OW (B, E, H) and TW (C, F, I). Selected DAWNs: 1,236 cm–1 corresponding to acetylated xylans and lignins (A–C), 1,316 cm–1 corresponding to crystalline cellulose (D–F) and 1,736 cm–1 corresponding to acetylated xylans (G–I). Red, ﬁber S-/G-layers; Green, vessel S-layers; Tan, ray S-layers; White, NA; Black, lumen. Red, and green intensities are proportionate to the absorbance of pixels assigned to ﬁber S/G-layers and vessel S-layers, respectively. FIGURE 5 \| Infrared images of selected DAWNs of representative images of NW (A, D, G), OW (B, E, H) and TW (C, F, I). Selected DAWNs: 1,236 cm–1 corresponding to acetylated xylans and lignins (A–C), 1,316 cm–1 corresponding to crystalline cellulose (D–F) and 1,736 cm–1 corresponding to acetylated xylans (G–I). Red, ﬁber S-/G-layers; Green, vessel S-layers; Tan, ray S-layers; White, NA; Black, lumen. Red, and green intensities are proportionate to the absorbance of pixels assigned to ﬁber S/G-layers and vessel S-layers, respectively. slightly distinguishes OW and NW vessel S-layers. The average infrared spectra of the vessel S-layers from the three types of wood appear rather similar (Figure 6B). The average spectra of OW and TW vessel S-layers exhibit an important noise between approximately 1,800 and 1,450 cm–1. Nevertheless, we can clearly distinguish in these two spectra, three double bands at 1,736, 1,460, and 1,374 cm–1, while there is only one band in the NW spectrum at these wavenumbers. Acetylated xylans mainly contribute to the ﬁrst band and cellulose, hemicelluloses and lignins to the two others. The bands at 1,736 (acetylated xylans), 1,332–26 (lignins, S ring plus G ring condensed), 1,236 (acetylated xylans and lignins) and 1,034 cm–1 (cellulose and lignins) makes possible to differentiate TW, NW and OW vessel S-layers (Figure 6B). Vessels DAWNs of 1,374–72 cm–1 may be attributed both to cellulose and hemicelluloses (CH deformation vibration and CH bending, Table S1). The isolated DAWN of 1,506 cm–1 can be assigned to lignins (aromatic skeletal vibrations; G > S). Besides, a large number of DAWNs cannot be easily related to known IR assignment (1,686–82, 1,572–68, 1,560–58, 1,538–28, 918, 914– 10, 936–26 cm–1). Cluster F gathers DAWNs that are less absorbed in TW vessel S-layers, when compared to OW and NW vessel S-layers. DAWNs in the range of 1,236–32 cm–1 correspond to acetylated xylans. The 3 groups of DAWNs in the same region of the IR spectra (1,744–36, 1,732–26, 1,722–10 cm–1) are due to C=O stretching, that are mainly due to acetylated xylans too, but the involvement of esteriﬁed lignins cannot be excluded Vessels The PCA loading plots (Supplementary Figure S3B) show that on PCA ﬁrst dimension, the loading is positive for four spectral zones (1,580–26, 1,494–74, 1,450–1,380, and 976–904 cm–1), that cannot be easily assigned to reported IR bands except the region of 1,430–21 cm–1 that can be assigned to cellulose, hemicelluloses and lignins (Table S1), 996–85 cm–1 to cellulose and 925–15 cm–1 to lignins. The loading is negative for three ranges of bands (1,760–22 cm–1, 1,256–36 cm–1 and 1,138– 1,034 cm–1), corresponding respectively to lignins and acetylated xylans, lignins, cellulose and lignins (Table S1). On the second dimension, two main ranges of bands (1,800–1,756 cm–1 and 1,206–1,164 cm–1) presented positive loadings, the second one being potentially attributed to cellulose, hemicelluloses and lignins (Table S1). The loading is negative mainly for two groups of bands (1,676–56 and 1,650–1,590 cm–1). They both correspond to lignins (C=O stretch in conjugated p-substituted aryl ketones/aromatic skeletal vibrations, higher absorbance in G lignin in comparison to S lignin, Table S1). As done for ﬁber analysis, the DAWNs in the vicinity of a local maximum (± 3 cm–1) were used to generate a heatmap (Figure 6C, Table S4). The heatmap shows more variability between samples, compared to the heatmap for ﬁber, most likely because vessel/ﬁber area ratio is small in wood and in consequence, there were fewer vessel pixels available. Cluster A corresponds to DAWNs that have a low absorbance in NW vessel S-layers, and a high absorbance in TW vessel S-layers, while the absorbance is generally contrasted between the two sections of OW analyzed. February 2020 \| Volume 11 \| Article 105 Frontiers in Plant Science \| www.frontiersin.org 8 Microphenotyping Using ATR-FTIR Cuello et al. FIGURE 6 \| Chemical differences between vessel S-layers of tension, opposite and normal wood. (A) PCA score plot. Turquoise, TW vessel S-layers; Purple, OW vessel S-layers; Red, NW vessel S-layers. Ellipses encompass 95% of the data in a normal distribution. Numbers in brackets refer to the number of pixels assigned to the different cell wall categories. Color intensity reﬂects the density of individuals. (B) Average ATR-FTIR spectra. Turquoise, TW vessel S-layers; Purple, OW vessel S-layers; Red, NW vessel S-layers. The green dotted line insertion represents a zoom of the [1,500–1,300] cm–1 region. (C) Heatmap of differentially absorbed wavenumbers. TW_Ves, TW vessel S-layers; OW_Ves, OW vessel S-layers; NW_Ves, NW vessel S-layers. Vessels FIGURE 6 \| Chemical differences between vessel S-layers of tension, opposite and normal wood. (A) PCA score plot. Turquoise, TW vessel S-layers; Purple, OW vessel S-layers; Red, NW vessel S-layers. Ellipses encompass 95% of the data in a normal distribution. Numbers in brackets refer to the number of pixels assigned to the different cell wall categories. Color intensity reﬂects the density of individuals. (B) Average ATR-FTIR spectra. Turquoise, TW vessel S-layers; Purple, OW vessel S-layers; Red, NW vessel S-layers. The green dotted line insertion represents a zoom of the [1,500–1,300] cm–1 region. (C) Heatmap of differentially absorbed wavenumbers. TW_Ves, TW vessel S-layers; OW_Ves, OW vessel S-layers; NW_Ves, NW vessel S-layers. (Table S1). On Figure 6B, the spectrum for NW shows a unique band around 1,734, while several bands and shoulders can be seen in TW and OW, probably related to a decreased amount of xylans, making visible other individual bands. Finally, both cellulose and lignins may contribute to DAWNs in the range of 1,040–32 cm–1 (C–O valence vibration, mainly from C3–O3H; aromatic C-H in- plane deformation, G > S). Cluster A contains DAWNs in the range of 1,364–60 and 1,346– 42 cm–1, that may be mostly attributed to cellulose and in the range of 1,336–22 cm–1, that may be attributed to crystalline cellulose and also to lignins (S ring plus G ring condensed). DAWNs of 1,374–72 cm–1 may be attributed both to cellulose and hemicelluloses (CH deformation vibration and CH bending, Table S1). The isolated DAWN of 1,506 cm–1 can be assigned to lignins (aromatic skeletal vibrations; G > S). Besides, a large number of DAWNs cannot be easily related to known IR assignment (1,686–82, 1,572–68, 1,560–58, 1,538–28, 918, 914– 10, 936–26 cm–1). Cluster F gathers DAWNs that are less absorbed in TW vessel S-layers, when compared to OW and NW vessel S-layers. DAWNs in the range of 1,236–32 cm–1 correspond to acetylated xylans. The 3 groups of DAWNs in the same region of the IR spectra (1,744–36, 1,732–26, 1,722–10 cm–1) are due to C=O stretching, that are mainly due to acetylated xylans too, but the involvement of esteriﬁed lignins cannot be excluded Cluster A contains DAWNs in the range of 1,364–60 and 1,346– 42 cm–1, that may be mostly attributed to cellulose and in the range of 1,336–22 cm–1, that may be attributed to crystalline cellulose and also to lignins (S ring plus G ring condensed). DISCUSSION In this study, we set up a non-destructive method for the microphenotyping of wood cell walls, based on ATR-FTIR microspectroscopy, that makes possible to obtain, in a rather short time, IR spectra from the different wood cell types, thus overcoming the complexity of wood tissue. We have developed a high-throughput pipeline for hyperspectral image analysis February 2020 \| Volume 11 \| Article 105 Frontiers in Plant Science \| www.frontiersin.org 9 Microphenotyping Using ATR-FTIR Cuello et al. (Figure 1) with both automatic pixel clustering according to cell wall types and identiﬁcation of DAWNs between samples. This pipeline was used to compare the biochemical composition of the cell wall from both ﬁbers and vessels originating from three kinds of wood in poplar: normal wood of upright trees, tension and opposite wood of artiﬁcially tilted trees. As the preparation of the samples is very simple—microtome sectioning at 20 μm, followed by ethanol drying—the acquisition of the images is fast—40 min for a (100 x 100) μm2 hyperspectral image—and the analysis is mostly an automated process, this method is suitable to screen for differences in IR spectra, a large number of samples, at the ﬁber or vessel cell walls. Thanks to the high refractive index of the Germanium crystal (4.01), the spatial resolution of ATR-FTIR microspectrocopy is 3.1 μm, and based on an oversampling factor of two, it is possible to reach a pixel resolution of (1.56 x 1.56) μm2. This is sufﬁcient to get IR spectra at the level of secondary cell walls, and with respects to the possibility of signal contamination, we obtained the best results for ﬁber cell walls, due to their higher thickness compared to vessel and ray cell walls. However, this method is not resolutive enough to reach the level of the cell wall layer. For such studies, some other methods of IR nanospectroscopy, such as scattering scanning near-ﬁeld optical microscopy (s-SNOM) or atomic force microscopy infrared nanospectroscopy (AFM-IR), would be more appropriate. Indeed, the resolution of such methods is beyond the classical diffraction limit of light. These techniques were recently used to investigate the chemical composition of primary cell walls in Populus vascular cambium (Pereira et al., 2018). The drawback of these techniques is that ultrathin sections need to be prepared, with potential chemical treatments like DMSO that may alter cell wall composition, and therefore IR spectra. DISCUSSION In this study, the IR spectra of TW ﬁber G-layer is indeed very similar to what was obtained by Olsson et al. (2011) on TW and isolated G-layers, with a large decrease of the bands at 1,736 and 1,236 cm–1, no bands at 1,594 and 1,506 cm–1, and the appearance of 3 speciﬁc bands at 1,136, 1,316, and 1,200 cm–1. Basically, it has also the same proﬁle as the one presented in Gierlinger et al. (2008), except that we did not observe, the large band at 1,645 cm–1 attributed to the deformation of vibration of adsorbed and free water: this probably reﬂects a difference in the water status of the samples analyzed. Our results indicate that TW ﬁber G-layers contained higher amounts of crystalline cellulose and lower amounts of acetylated xylans and lignins than ﬁber S-layers, although it is not so striking for TW S-layers (see supra for discussion on this). This is consistent with a number of studies demonstrating that the G-layer was mainly composed of cellulose (e.g. Norberg and Meier, 1966; Nishikubo et al., 2007; Guedes et al., 2017). This is also in accordance with the absence or the low levels of xylans and lignins reported A critical step to identify differences in cell wall composition using ATR-FTIR microspectroscopy is to unequivocally assign pixels to the right type of cell wall. The clustering method used here makes possible such an assignment with a high efﬁciency for both ﬁber S- and G-layers and with an acceptable level of success for vessels. The low score for the assignation of ray pixels probably results from the high diversity in ray cell walls. Indeed, as rays are uniseriate in Populus (IAWA Committee, 1989), there are very few ray-to-ray cell walls on transverse sections and most ray cell walls are either neighbored by ﬁber or vessel cell walls. Therefore, in this study, the pixels classiﬁed as ray cell walls correspond in fact to a mixture of ray/vessel and ray/ﬁber cell walls, whose composition may differ accordingly. Ray cells can be classiﬁed into three different cell types, named contact cells, intermediate cells and isolated cells (Braun, 1967; Nakaba et al., 2012) that may have cell walls with rather different features. DISCUSSION addition, parenchyma ray cells stay alive longer than vessels and ﬁbers and their cell wall composition may also be subjected to modiﬁcations according to their age. These aspects may be at least partly circumvented by applying ATR-FTIR technique on longitudinal sections in place of transverse sections. The score for the assignation of vessel pixels is not as high as for ﬁber pixels: as for rays, vessel pixels may in fact correspond to a mixture of vessel/vessel, vessel/ray and vessel/ﬁber cell walls. The fact that vessels have thin cell walls may also contribute to some incorrect assignments. Finally, the lower score for the assignment of pixels in OW compared to NW and TW (Figure 2B) may be linked to the reduced growth of xylem cells resulting from the position of OW on the lower side of the tilted stem. The smaller number of pixels assigned to vessel cell wall in TW compared to OW and NW is in accordance with the observations from a number of studies using other approaches, that the number of vessels is decreased in TW in comparison to OW (Jourez et al., 2001; Ruelle et al., 2006; Tarmian et al., 2009) or to NW (Chow, 1947). We also observed in TW a decreased number of lumen- associated pixels, which certainly result in the reduced lumen size due to the presence of the G-layer (Clair et al., 2006; Ruelle et al., 2006; Déjardin et al., 2010; Gorshkova et al., 2010). Finally, the smaller number of pixels assigned to ﬁber S-layers in TW supports the fact that S-layers are reduced in G ﬁbers, in relation to the deposition of the G-layer. ATR-FTIR microscopy provided IR proﬁling of ﬁbers and vessel cell wall layers and DAWNs were identiﬁed and used to characterize differences in biochemical composition between layers. Another critical step was to correctly assign these DAWNs to wood polymers as the absorption band from a given functional group may originate from several wood polymers and bands are overlapping in the spectrum. This is the reason why we focused on the DAWNs in the vicinity of visible peaks on the spectra to get information easier to interpret on a biological level. TW and G ﬁbers have been extensively studied using various complementary methods (FTIR, Raman imaging, biochemical analyses, immunolocalization). These are valuable data to validate our results using in situ IR proﬁling. Frontiers in Plant Science \| www.frontiersin.org CONCLUSION Thanks to the high pixel resolution of ATR-FTIR hyperspectral images—(1.56 x 1.56) μm² —we were able to get IR proﬁling of ﬁbers and vessels at the cell layer level. We have developed a high-throughput pipeline for hyperspectral image analysis with both automatic pixel clustering and identiﬁcation of differentially absorbed wavenumbers between samples. We have implemented this pipeline on different types of wood, including tension wood, a model that has been extensively characterized from a biochemical point of view. We conﬁrmed previously known results on G layers, which validates this in situ approach, and we generated an IR proﬁle for vessel S-layers in 3 different types of wood, as well as for TW ﬁber S-layers, providing information on the composition of cell layers rarely described in the literature. This methodological study paves the way for the microphenotyping of cell walls to rapidly and ﬁnely characterize various genetic resources or trees submitted to various stresses, and thus to overcome the complexity of wood tissue for all studies aimed at understanding wood formation. We consider that this technique is useful to give some clues about the cell wall compounds underlying cell wall differences. However, with regards to the inherent limitations of this technique (spatial resolution, difﬁculties to unambiguously assign bands to speciﬁc cell wall compounds), the hypotheses raised have to be validated by a complementary low-throughput technique, like immunocytochemistry. Regarding ﬁber S-layers, many DAWNs exhibited important differences in absorbance between TW ﬁber S-layers and ﬁber S- layers from OW or NW. From our observations, the former seemed to have a biochemical composition rather similar to what is described in the G-layer, while some features remained common to the S-layer from OW and NW. For example, the absorption bands for lignins and acetylated xylans were present in TW ﬁber S-layers but lower than in OW and NW. These observations may reﬂect biological differences in ﬁber cell wall assembly between the different wood types. This may be related to the fact that the S-layers in G ﬁbers are not fully developed, in comparison to the S-layers of OW and NW ﬁbers. Alternatively, it remains possible that the induction of tension wood formation, beside inducing G-layer formation, may also act on S-layer differentiation. However, we cannot rule out that these differences result from signal contaminations by the adjacent G-layer. CONCLUSION Therefore, it remains to verify with a larger number of observations that our results reﬂect the biological reality or if the close proximity of the G-layer interferes with the measurements. AUTHOR CONTRIBUTIONS CC, FL, GP, and AD designed the research. CC, PM, FL, CG-P, and VL-P acquired the data. CC, PM, and AD analyzed the data. CC, GP and AD wrote the article. CC, FL, GP, and AD designed the research. CC, PM, FL, CG-P, and VL-P acquired the data. CC, PM, and AD analyzed the data. CC, GP and AD wrote the article. DATA AVAILABILITY STATEMENT p y y IR proﬁling was also obtained for vessel S-layers. The two ﬁrst dimensions of PCA were able to discriminate between the three types of wood (Figure 6A) showing that there are some chemical differences in the cell wall of vessels, even between NW and OW, while ﬁber S-layers were very similar for both types of wood. The analysis of DAWNs between the different kinds of wood was not easy. Indeed, there was an important variability between the different sections analyzed [see for example differences between TT_11A and TT_12B, the 2 replicates for OW vessel cell wall (Figure 6C)]. This certainly reﬂects the low number of pixels analyzed (in comparison to ﬁbers) that mainly results from the fact that vessels have thin cell wall and are not numerous (still in comparison to ﬁbers). As for ﬁber cell walls, there are obviously important differences between vessel cell wall from NW and TW, with more cellulose and less acetylated xylans in the latter, two features also observed in TW ﬁber cell wall. Although, G-layer is absent from vessel cell wall, we cannot rule out a response at the vessel cell wall to stem tilting. The differences between NW and OW vessel S-layers, based on PC2 loadings, are mainly explained The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation, to any qualiﬁed researcher. DISCUSSION While contact cells are located predominantly within the upper and lower lines of individual ray lines and are connected to adjacent vessel elements through pits, intermediate cells are located within the same radial lines but are not adjacent to vessel elements. Isolated cells are located within the radial cell lines with no connections with vessels (Braun, 1967). In Populus sieboldii x P. grandidentata, it has been shown that secondary cell wall thickening is initiated earlier in the upper and lower cell lines of a ray (Nakaba et al., 2012). In February 2020 \| Volume 11 \| Article 105 10 Microphenotyping Using ATR-FTIR Cuello et al. elsewhere. The higher absorbance of a DAWN attributed to proteins is relevant since arabinogalactan proteins were reported as very abundant in TW in a number of studies (e.g. Lafarguette et al., 2004; Andersson-Gunnerås et al., 2006). Interestingly, while most lignin speciﬁc DAWNs exhibited lower absorbance in the G- layer, one of them presented a higher absorbance (1,518–12 cm–1, aromatic skeletal vibration, higher absorbance in G lignin in comparison to S lignin). In addition, a positive loading on the ﬁrst dimension of PCA was found for a group of bands corresponding also to lignins (1,330–24 cm–1, S ring plus G ring condensed). This suggests that some lignins, likely different from the ﬁber S-layer lignins, are actually present in the G-layer, as previously reported in poplar (Joseleau et al., 2004; Gierlinger and Schwanninger, 2006) and in other species (Ghislain et al., 2016; Higaki et al., 2017). Thus, the in situ technique described in this paper is validated on TW ﬁber G-layers, since the results we obtained on individual G-layers were in accordance with other studies made at the tissue level or on isolated G-layers. by differences in lignins. As we did not ﬁnd any information on vessel cell wall composition in the bibliography, it is difﬁcult to go further on these observations without performing measurements on a larger number of samples or using complementary techniques like immunocytochemistry. Frontiers in Plant Science \| www.frontiersin.org REFERENCES Fairbrother, T. E., and Brink, G. E. (1990). Determination of cell wall carbohydrates in forages by near infrared reﬂectance spectroscopy. Anim. Feed Sci. Technol. 28, 293–302. doi: 10.1016/0377-8401(90)90160-A Allison, G. G. (2011). “Application of Fourier Transform Mid-Infrared Spectroscopy (FTIR) for Research into Biomass Feed-Stocks” in Fourier Transforms - New Analytical Approaches and FTIR Strategies. Ed. G. Nikolic (London: IntechOpen). doi: 10.5772/15785 Freudenberg, K., and Neish, A. C. (1968). Constitution and biosynthesis of lignin (Berlin Heidelberg: Springer-Verlag), 129 pp. Gebreselassie, M. N., Ader, K., Boizot, N., Millier, F., Charpentier, J.-P., Alves, A., et al. (2017). Near-infrared spectroscopy enables the genetic analysis of chemical properties in a large set of wood samples from Populus nigra (L.) natural populations. Ind. Crops Prod. 107, 159–171. doi: 10.1016/ j.indcrop.2017.05.013 Andersson-Gunnerås, S., Mellerowicz, E. J., Love, J., Segerman, B., Ohmiya, Y., Coutinho, P. M., et al. (2006). Biosynthesis of cellulose-enriched tension wood in Populus: global analysis of transcripts and metabolites identiﬁes biochemical and developmental regulators in secondary wall biosynthesis. Plant J. Cell Mol. Biol. 45, 144–165. doi: 10.1111/j.1365-313X.2005.02584.x Biol. 45, 144–165. doi: 10.1111/j.1365-313X.2005.02584.x Ghislain, B., Nicolini, E.-A., Romain, R., Ruelle, J., Yoshinaga, A., Alford, M. H., et al. (2016). Multilayered structure of tension wood cell walls in Salicaceae sensu lato and its taxonomic signiﬁcance. Bot. J. Linn. Soc 182, 744–756. doi: 10.1111/boj.12471 Baldy, A., Jacquemot, M.-P., Griveau, Y., Bauland, C., Reymond, M., and Mechin, V. (2017). Energy values of registered corn forage hybrids in france over the last 20 years rose in a context of maintained yield increase. Am. J. Plant Sci. 08, 1449–1461. doi: 10.4236/ajps.2017.86099 Gierlinger, N., and Schwanninger, M. (2006). Chemical imaging of poplar wood cell walls by confocal raman microscopy. Plant Physiol. 140, 1246–1254. doi: 10.1104/pp.105.066993 Barnett, J. R., and Bonham, V. A. (2004). Cellulose microﬁbril angle in the cell wall of wood ﬁbers. Biol. Rev. 79, 461–472. doi: 10.1017/S1464793103006377 Bertrand, D., and Dufour, E. (2006). La spectroscopie infrarouge et ses applications analytiques (Tec & Doc. Paris: Lavoisier). Gierlinger, N., Goswami, L., Schmidt, M., Burgert, I., Coutand, C., Rogge, T., et al. (2008). In situ FT-IR microscopic study on enzymatic treatment of poplar wood cross-sections. Biomacromolecules 9, 2194–2201. doi: 10.1021/ bm800300b Braun, H. J. (1967). Entwicklung und Bau der Holzstrahlen unter dem Aspekt der kontakt-isolations-differenzierung gegenüber dem hydrosystem. i. das prinzip der kontakt-isolations-differenzierung. Holzforschung 21, 33–37. doi: 10.1515/ hfsg.1967.21.2.33 Gierlinger, N. (2018). New insights into plant cell walls by vibrational microspectroscopy. Appl. Spectrosc. REFERENCES Rev. 53, 517–551. doi: 10.1080/ 05704928.2017.1363052 Chazal, R., Robert, P., Durand, S., Devaux, M.-F., Saulnier, L., Lapierre, C., et al. (2014). Investigating lignin key features in maize lignocelluloses using infrared spectroscopy. Appl. Spectrosc. 68, 1342–1347. doi: 10.1366/14-07472 Gorshkova, T. A., Gurjanov, O. P., Mikshina, P. V., Ibragimova, N. N., Mokshina, N. E., Salnikov, V. V., et al. (2010). Speciﬁc type of secondary cell wall formed by plant ﬁbers. Russ. J. Plant Physiol. 57, 328–341. doi: 10.1134/ S1021443710030040 Chen, L., Carpita, N. C., Reiter, W.-D., Wilson, R. H., Jeffries, C., and McCann, M. C. (1998). A rapid method to screen for cell-wall mutants using discriminant analysis of Fourier transform infrared spectra. Plant J. 16, 385–392. doi: 10.1046/j.1365- 313x.1998.00301.x Gorshkova, T., Mokshina, N., Chernova, T., Ibragimova, N., Salnikov, V., Mikshina, P., et al. (2015). Aspen tension wood ﬁbers contain b-(1!4)- galactans and acidic arabinogalactans retained by cellulose microﬁbrils in gelatinous walls. Plant Physiol. 169, 2048–2063. doi: 10.1104/pp.15.00690 Chow, K. Y. (1947). A comparative study of the structure and chemical composition of tension wood and normal wood in beech (Fagus sylvatica L.). Forestry 20, 62–77. doi: 10.1093/forestry/20.1.62 Gorzsás, A., Stenlund, H., Persson, P., Trygg, J., and Sundberg, B. (2011). Cell- speciﬁc chemotyping and multivariate imaging by combined FT-IR microspectroscopy and orthogonal projections to latent structures (OPLS) analysis reveals the chemical landscape of secondary xylem: The chemical landscape of secondary xylem. Plant J. 66, 903–914. doi: 10.1111/j.1365- 313X.2011.04542.x Clair, B., Ruelle, J., Beauchêne, J., Prévost, M. F., and Fournier, M. (2006). Tension wood and opposite wood in 21 tropical rain forest species. IAWA J. 27, 329– 338. doi: 10.1163/22941932-90000158 Courtois‐Moreau, C. L., Pesquet, E., Sjödin, A., Muñiz, L., Bollhöner, B., Kaneda, M., et al. (2009). A unique program for cell death in xylem ﬁbers of Populus stem. Plant J. 58, 260–274. doi: 10.1111/j.1365-313X.2008.03777.x Guedes, F. T. P., Laurans, F., Quemener, B., Assor, C., Lainé-Prade, V., Boizot, N., et al. (2017). Non-cellulosic polysaccharide distribution during G-layer formation in poplar tension wood ﬁbers: abundance of rhamnogalacturonan I and arabinogalactan proteins but no evidence of xyloglucan. Planta 246, 857– 878. doi: 10.1007/s00425-017-2737-1 j Déjardin, A., Laurans, F., Arnaud, D., Breton, C., Pilate, G., and Leplé, J.-C. (2010). Déjardin, A., Laurans, F., Arnaud, D., Breton, C., Pilate, G., and Leplé, J.-C. (2010). Wood formation in Angiosperms. C. R. Biol. 333, 325–334. doi: 10.1016/ j.crvi.2010.01.010 Wood formation in Angiosperms. C. R. Biol. 333, 325–334. FUNDING This research was funded by Centre Val de Loire Region, APR-IR #2016-00108472 (OPeNSPeNU). This research was funded by Centre Val de Loire Region, APR-IR #2016-00108472 (OPeNSPeNU). Frontiers in Plant Science \| www.frontiersin.org February 2020 \| Volume 11 \| Article 105 Microphenotyping Using ATR-FTIR Cuello et al. SUPPLEMENTARY MATERIAL The authors gratefully acknowledge the LICA (Laboratoire d'Ingéniérie Cellulaire de l'Arbre) and Phenobois (Wood and Tree Physicochemical Phenotyping Facility for Genetic Resources) for the provision of infra-red instruments. Phenobois is supported by the programme “Investments for the Future” (ANR-10-EQPX-16, XYLOFOREST) from the French National Agency for Research. The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpls.2020.00105/ full#supplementary-material g ) ( Tree Physicochemical Phenotyping Facility for Genetic Resources) for the provision of infra-red instruments. Phenobois is supported by the programme “Investments for the Future” (ANR-10-EQPX-16, XYLOFOREST) from the French National Agency for Research. SUPPLEMENTARY FIGURE 1 \| Crystal impacts on UT_12-C cross-section. Green circles pinpoint the three impacts. SUPPLEMENTARY FIGURE 1 \| Crystal impacts on UT_12-C cross-section. Green circles pinpoint the three impacts. SUPPLEMENTARY FIGURE 2 \| PCA score plot of ﬁbres on the whole data set. Blue: TW ﬁbre G-layer, Turquoise: TW ﬁbre S-layers, Purple: OW ﬁbre S-layers, Red: NW ﬁbre S-layers. Ellipses encompass 95% of the data in a normal distribution. Color intensity reﬂects the density of individuals. The authors would also like to thank the staff of the INRAE experimental unit GBFOR (UE 911) for the establishment and management of the experimental plantation in Orléans, France. CC has been funded by a University fellowship. distribution. Color intensity reﬂects the density of individuals. SUPPLEMENTARY FIGURE 3 \| PCA loading plots of ﬁbre S- and G-layers (A) and vessel S-layers (B). Red: PC1, Green: PC2. SUPPLEMENTARY FIGURE 3 \| PCA loading plots of ﬁbre S- and G-layers (A) and vessel S-layers (B). Red: PC1, Green: PC2. REFERENCES doi: 10.1016/ j.crvi.2010.01.010 Donaldson, L., Hague, J., and Snell, R. (2001). Lignin distribution in coppice poplar, linseed and wheat straw. Holzforschung 55, 379–385. doi: 10.1515/ HF.2001.063 Higaki, A., Yoshinaga, A., and Takabe, K. (2017). Heterogeneous distribution of xylan and lignin in tension wood G-layers of the S1+G type in several Japanese hardwoods. Tree Physiol. 37, 1767–1775. doi: 10.1093/treephys/tpx144 Durand, S., Papineau, P., Robert, P., Devaux, M.-F., and Guillon, F. (2019). HISTOCHEM: a database of reference spectra for plant cell wall polymers. Berichte Aus Dem. Julius Kühn-Inst. 204, 91–92. doi: 10.5073/berjki.2019.204.000 Huang, J., Li, Y., Wang, Y., Chen, Y., Liu, M., Wang, Y., et al. (2017). A precise and consistent assay for major wall polymer features that distinctively determine February 2020 \| Volume 11 \| Article 105 Frontiers in Plant Science \| www.frontiersin.org 12 Microphenotyping Using ATR-FTIR Cuello et al. biomass sacchariﬁcation in transgenic rice by near-infrared spectroscopy. Biotechnol. Biofuels 10, 294. doi: 10.1186/s13068-017-0983-x Nishikubo, N., Awano, T., Banasiak, A., Bourquin, V., Ibatullin, F., Funada, R., et al. (2007). Xyloglucan Endo-transglycosylase (XET) Functions in gelatinous layers of tension wood ﬁbers in poplar—a glimpse into the mechanism of the balancing act of trees. Plant Cell Physiol. 48, 843–855. doi: 10.1093/pcp/ pcm055 IAWA Committee (1989). IAWA List of Microscopic Features for Hardwood Identiﬁcation IAWA Bulletin Vol. 10. Eds. E. A. Wheeler, P. Baas and P. Gasson (Leiden: Brill publishers), 219–332. Norberg, P. H., and Meier, H. (1966). Physical and chemical properties of the gelatinous layer in tension wood ﬁbers of aspen (Populus tremula L.). Holzforschung 20, 174–178. doi: 10.1515/hfsg.1966.20.6.174 Jasinski, S., Lécureuil, A., Durandet, M., Bernard-Moulin, P., and Guerche, P. (2016). Arabidopsis Seed Content QTL Mapping Using High-Throughput Phenotyping: The Assets of Near Infrared Spectroscopy. Front. Plant Sci. 7, 1682. doi: 10.3389/fpls.2016.01682 Olsson, A.-M., Bjurhager, I., Gerber, L., Sundberg, B., and Salmén, L. (2011). Ultra-structural organisation of cell wall polymers in normal and tension wood of aspen revealed by polarisation FTIR microspectroscopy. Planta 233, 1277– 1286. doi: 10.1007/s00425-011-1384-1 Joseleau, J.-P., Imai, T., Kuroda, K., and Ruel, K. (2004). Detection in situ and characterization of lignin in the G -layer of tension wood ﬁbers of Populus deltoides. Planta 219, 338–345. doi: 10.1007/s00425-004-1226-5 Pereira, L., Flores-Borges, D. N. A., Bittencourt, P. R. L., Mayer, J. L. S., Kiyota, E., Araujo, P., et al. (2018). Infrared nanospectroscopy reveals the chemical nature of pit membranes in water-conducting cells of plant xylem. Plant Physiol. 177, 1629–1638. REFERENCES doi: 10.1016/j.crvi.2004.07.006 Jourez, B., Riboux, A., and Leclercq, A. (2001). Anatomical characteristics of tension wood and opposite wood in young inclined stems of poplar (Populus euramericana cv ‘GHOY’). IAWA J. 22, 133–157. doi: 10.1163/22941932- 90000274 Jourez, B. (1997). Le bois de tension. 1 Déﬁnition et distribution dans l’arbre. Biotechnol. Agron. Société Environ. 1, 100–112. http://hdl.handle.net/2268/ 95404 Pilate, G., Déjardin, A., Laurans, F., and Leplé, J.-C. (2004). Tension wood as a model for functional genomics of wood formation. New Phytol. 164, 63–72. doi: 10.1111/j.1469-8137.2004.01176.x Kacurakova, M., and Wilson, R. (2001). Developments in mid-infrared FT-IR spectroscopy of selected carbohydrates. Carbohydr. Polym. 44, 291–303. doi: 10.1016/S0144-8617(00)00245-9 Plomion, C., Leprovost, G., and Stokes, A. (2001). Wood formation in trees. Plant Physiol. 127, 1513–1523. doi: 10.1104/pp.010816 Ruelle, J., Clair, B., Beauchêne, J., Prévost, M. F., and Fournier, M. (2006). Tension wood and opposite wood in 21 tropical rain forest species. IAWA J. 27, 341– 376. doi: 10.1163/22941932-90000159 Lafarguette, F., Leplé, J.-C., Déjardin, A., Laurans, F., Costa, G., Lesage-Descauses, M.-C., et al. (2004). Poplar genes encoding fasciclin-like arabinogalactan proteins are highly expressed in tension wood. New Phytol. 164, 107–121. doi: 10.1111/j.1469-8137.2004.01175.x Tarmian, A., Remond, R., Faezipour, M., Karimi, A., and Perré, P. (2009). Reaction wood drying kinetics: tension wood in Fagus sylvatica and compression wood in Picea abies. Wood Sci. Technol. 43, 113–130. doi: 10.1007/s00226-008-0230-5 Li, K., Wang, H., Hu, X., Ma, F., Wu, Y., Wang, Q., et al. (2017). Genetic and quantitative trait locus analysis of cell wall components and forage digestibility in the Zheng58 × HD568 Maize RIL population at anthesis stage. Front. Plant Sci. 8, 1472. doi: 10.3389/fpls.2017.01472 Terashima, N., Fukushima, K., He, L.-F., and Takabe, K. (1993). “Comprehensive Model of the Ligniﬁed Plant Cell Wall,” in Forage Cell Wall Structure and Digestibility (Madison, WI: American Society of Agronomy, Crop Science Society of America, Soil Science Society of America), 247–270. doi: 10.2134/ 1993.foragecellwall.c10 McCann, M. C., Hammouri, M., Wilson, R., Belton, P., and Roberts, K. (1992). Fourier transform infrared microspectroscopy is a new way to look at plant cell walls. Plant Physiol. 100, 1940–1947. doi: 10.1104/pp.100.4.1940 Mellerowicz, E. J., and Sundberg, B. (2008). Wood cell walls: biosynthesis, developmental dynamics and their implications for wood properties. Curr. Opin. Plant Biol. 11, 293–300. doi: 10.1016/j.pbi.2008.03.003 van Parijs, F. R. D., Morreel, K., Ralph, J., Boerjan, W., and Merks, R. M. H. (2010). Modeling lignin polymerization. I. simulation model of dehydrogenation polymers. Plant Physiol. 153, 1332–1344. Frontiers in Plant Science \| www.frontiersin.org February 2020 \| Volume 11 \| Article 105 REFERENCES doi: 10.1104/pp.110.154468 Molano, M. L., Cortés, M. L., Ávila, P., Martens, S. D., and Muñoz, L. S. (2016). Near infrared spectroscopy (NIRS) calibration equations to predict nutritional quality parameters of tropical forages. Trop. Grassl.-Forrajes Trop. 4, 139–145. doi: 10.17138/tgft(4)139-145 Conﬂict of Interest: The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Mouille, G., Robin, S., Lecomte, M., Pagant, S., and Höfte, H. (2003). Classiﬁcation and identiﬁcation of Arabidopsis cell wall mutants using Fourier-Transform InfraRed (FT-IR) microspectroscopy. Plant J. 35, 393–404. doi: 10.1046/j.1365- 313X.2003.01807.x Copyright © 2020 Cuello, Marchand, Laurans, Grand-Perret, Lainé-Prade, Pilate and Déjardin. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Copyright © 2020 Cuello, Marchand, Laurans, Grand-Perret, Lainé-Prade, Pilate and Déjardin. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Nakaba, S., Begum, S., Yamagishi, Y., Jin, H.-O., Kubo, T., and Funada, R. (2012). Differences in the timing of cell death, differentiation and function among three different types of ray parenchyma cells in the hardwood Populus sieboldii × P. grandidentata. Trees 26, 743–750. doi: 10.1007/s00468-011-0640-0 February 2020 \| Volume 11 \| Article 105 Frontiers in Plant Science \| www.frontiersin.org 13
https://openalex.org/W1991430024	https://hal.science/hal-02372142/document	English	null	Protein-tyrosine Phosphatase PTPL1/FAP-1 Triggers Apoptosis in Human Breast Cancer Cells	Journal of biological chemistry/The Journal of biological chemistry	2,002	cc-by	15,154	1 The abbreviations used are: IGF, insulin-like growth factor; DCC/ FCS, dextran-coated charcoal-treated fetal calf serum; IRS-1, insulin receptor substrate-1; OH-Tam, 4-hydroxytamoxifen; PI3-K, phosphati- dylinositol (PI) 3-kinase; PTP, protein-tyrosine phosphatase; TUNEL, terminal deoxynucleotidyltransferase-mediated dUTP-biotin nick end labeling; wt, wild type. Protein-tyrosine Phosphatase PTPL1/FAP-1 Triggers Apoptosis in Human Breast Cancer Cells* Received for publication September 3 2002 and in revised form September 26 2002 Received for publication, September 3, 2002, and in revised form, September 26, 2002 Published, JBC Papers in Press, September 26, 2002, DOI 10.1074/jbc.M208950200 Received for publication, September 3, 2002, and in revised form, September 26, 2002 Published, JBC Papers in Press, September 26, 2002, DOI 10.1074/jbc.M208950200 Guillaume Bompard‡, Carole Puech, Christine Pre´bois, Franc¸oise Vignon§, and Gilles Freiss From the Inserm, Unit 540, Molecular and Cellular Endocrinology of Cancers, 60 rue de Navacelles, 34090 Montpellier, France Guillaume Bompard‡, Carole Puech, Christine Pre´bois, Franc¸oise Vignon§, and Gilles Freiss From the Inserm, Unit 540, Molecular and Cellular Endocrinology of Cancers, 60 rue de Navacelles, 34090 Montpellier, France Studies in Jurkat leukemia cells have suggested that protein-tyrosine phosphatase PTPL1/FAP-1 rescues Fas- induced cell death. However, we have previously shown that this enzyme triggers 4-hydroxytamoxifen-induced growth inhibition in human breast cancer cells. The present study addresses the role of PTPL1/FAP-1 in an- tiestrogen-regulated apoptotic effect and insulin-like growth factor-I survival action in MCF7 cells and fur- ther identifies the impacted signaling pathway. By ter- minal deoxynucleotidyltransferase-mediated dUTP-bio- tin nick end labeling and cytoplasmic nucleosome enzyme-linked immunosorbent assay, we demonstrated that 4-hydroxytamoxifen-induced apoptosis was totally lost in PTPL1/FAP-1 antisense transfectants in which enzyme expression was abrogated, revealing the crucial role of this phosphatase in the apoptotic process in hu- man breast cancer cells. Time-dependent expression of PTPL1/FAP-1 in MCF7 cells completely abolished the survival action of insulin-like growth factor-I. This ef- fect occurred through a highly significant reduction in phosphatidylinositol 3-kinase/Akt pathway activation (80% reduction in phosphatidylinositol 3-kinase activ- ity, 55% inhibition of Akt activation) accompanied by a 65% decrease in insulin receptor substrate-1 growth fac- tor-induced tyrosine phosphorylation. These results provide the first evidence that PTPL1/FAP-1 has a key role in the apoptotic process in human breast cancer cells independent of Fas but associated with an early inhibition of the insulin receptor substrate-1/phosphati- dylinositol 3-kinase pathway. Our data therefore sug- gest new therapeutic routes and strengthen the impor- tance of identifying endogenous regulators and substrates of this phosphatase in breast tumors. vitro in human breast cancer cell models that steroid hormones (mostly estrogens) and growth factors (epidermal growth fac- tor, transforming growth factor , IGF-I1 and II, etc.) are the major signals affecting proliferation (2, 3). Most of these factors stimulate cell division and/or promote cell survival by mecha- nisms yet poorly understood, thus conferring growth advantage to tumor-responsive cells. Protein-tyrosine Phosphatase PTPL1/FAP-1 Triggers Apoptosis in Human Breast Cancer Cells* Received for publication September 3 2002 and in revised form September 26 2002 On the other side, growth inhibitors and various antagonists have been shown to increase breast cancer cell apoptosis (4, 5). Estrogens and their antagonists mediate their action through their nuclear receptors (estrogen receptors and ), acting as transcriptional factors in coordination with numerous additional transcriptional cofactors. Peptide hormones and growth factor-signaling pathways involve the activation of sev- eral kinase cascades (mitogen-activated protein kinases, phos- phatidylinositol (PI) 3-kinase, etc.). We and others provided evidence for the existence of multiple interactions between these seemingly distinct pathways (6–10). We previously dem- onstrated that steroidal and non-steroidal antiestrogens such as tamoxifen, which is currently used in breast cancer adjuvant therapy, inhibit the growth of estrogen receptor-positive cells not only by acting as competitors of hormone agonists on the nuclear receptors but also by preventing growth factor mito- genic action in the total absence of estrogens (6). Cell growth inhibition, evidenced by the drastic decrease in the overall cell DNA measurement, correlated with the expression of some protein-tyrosine phosphatases (PTP) (11), among which is PTPL1/FAP-1 (12–15), whose mRNA accumulation is time- and dose-dependently regulated by 4-hydroxytamoxifen (OH-Tam) and ICI 182,780 (16). Using stable transfectants expressing PTPL1/FAP-1 RNA antisense, we have shown that this partic- ular enzyme was a key factor in the OH-Tam-induced anti- growth factor action. The role of PTP in breast cancer cell apoptosis is poorly understood, and SHP-1 was the only enzyme yet clearly impli- cated in somatostatin-induced apoptosis (17). Besides, a study in Jurkat cells proposed an antiapoptotic role of PTPL1/FAP-1 by inhibition of the Fas-signaling pathway (15), whereas MCF7 cells like the majority of the breast cancer cell lines are resist- ant to Fas-induced apoptosis (18). Breast cancer is one of the most common malignancies af- fecting women in the Western countries. Although mortality and survival rates decreased in the UK (1), breast cancer inci- dence is still increasing. It is therefore crucial in the coming years to design new therapeutic strategies based on the up- dated knowledge of the mechanisms by which tumor growth is sustained. Our present study was designed to evaluate the role of PTPL1/FAP-1 in the apoptotic process and to determine PTPL1/FAP-1 action on the survival signaling pathway stimu- lated by IGF-I. Using PTPL1/FAP-1 antisense transfectants in which enzyme expression is abrogated, we here demonstrate Breast cancer proliferation is the result of the balance be- tween cell division and cell apoptosis. * This work was supported by Inserm, by the Association pour la Recherche sur le Cancer, and by the Ligue Re´gionale Contre le Cancer. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ‡ Recipient of a Ph.D. grant from the Ministe`re de la Recherche et de l’Enseignement Supe´rieur and an ARC fellowship. § To whom correspondence should be addressed. Tel.: 33-4-67-04-37- 63; Fax: 33-4-67-54-05-98; E-mail: vignon@u540.montp.inserm.fr. Vol. 277, No. 49, Issue of December 6, pp. 47861–47869, 2002 Printed in U.S.A. Vol. 277, No. 49, Issue of December 6, pp. 47861–47869, 2002 Printed in U.S.A. THE JOURNAL OF BIOLOGICAL CHEMISTRY © 2002 by The American Society for Biochemistry and Molecular Biology, Inc. This is an Open Access article under the CC BY license. * This work was supported by Inserm, by the Association pour la Recherche sur le Cancer, and by the Ligue Re´gionale Contre le Cancer. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Protein-tyrosine Phosphatase PTPL1/FAP-1 Triggers Apoptosis in Human Breast Cancer Cells* Received for publication September 3 2002 and in revised form September 26 2002 It has been shown in 47862 47862 ated by Bradford assay (19), was immunoprecipitated with antiserum against IRS-1 (Upstate Biotechnology) and retained on protein G-Sepharose. The immunoprecipitates were washed twice with phos- phate-buffered saline, 0.5% Nonidet P-40, twice with 100 mM Tris-HCl (pH 7.4), 0.5 M LiCl, 1 mM dithiothreitol, and 0.2 mM Na3VO4, and twice with 10 mM Tris-HCl (pH 7.4), 100 mM NaCl, 1 mM dithiothreitol, and 0.2 mM Na3VO4. All washes were performed at 4 °C. The reaction mixtures (50 l), containing the immunoprecipitates in 20 mM Tris-HCl (pH 7.4), 80 mM NaCl, 25 mM MgCl2, 0.5 mM EGTA, 10 M ATP, 1 Ci of [-32P]ATP, and 2 g of PI/phosphatidylserine were incubated at 25 °C for 30 min. The reactions were terminated, and the lipids were extracted by the addition of 10 volumes of chloroform/methanol/HCl (11.6 N) (50:100:1). The mixture was then vortexed, and the organic phase was extracted with 2 volumes of methanol, 4 N HCl (1:1). The extracted products were desiccated, dissolved in 10 l of chloroform, and separated by thin-layer chromatography in a developing solution composed of chloroform, methanol, 4 M NH4OH (9:7:2). The production of phosphatidylinositol 3-phosphate was detected and quantified by counting with a Fujix-Bas 1000. ated by Bradford assay (19), was immunoprecipitated with antiserum against IRS-1 (Upstate Biotechnology) and retained on protein G-Sepharose. The immunoprecipitates were washed twice with phos- phate-buffered saline, 0.5% Nonidet P-40, twice with 100 mM Tris-HCl (pH 7.4), 0.5 M LiCl, 1 mM dithiothreitol, and 0.2 mM Na3VO4, and twice with 10 mM Tris-HCl (pH 7.4), 100 mM NaCl, 1 mM dithiothreitol, and 0.2 mM Na3VO4. All washes were performed at 4 °C. The reaction mixtures (50 l), containing the immunoprecipitates in 20 mM Tris-HCl (pH 7.4), 80 mM NaCl, 25 mM MgCl2, 0.5 mM EGTA, 10 M ATP, 1 Ci of [-32P]ATP, and 2 g of PI/phosphatidylserine were incubated at 25 °C for 30 min. The reactions were terminated, and the lipids were extracted by the addition of 10 volumes of chloroform/methanol/HCl (11.6 N) (50:100:1). The mixture was then vortexed, and the organic phase was extracted with 2 volumes of methanol, 4 N HCl (1:1). The extracted products were desiccated, dissolved in 10 l of chloroform, and separated by thin-layer chromatography in a developing solution composed of chloroform, methanol, 4 M NH4OH (9:7:2). The production of phosphatidylinositol 3-phosphate was detected and quantified by counting with a Fujix-Bas 1000. 47862 that protein-tyrosine phosphatase PTPL1/FAP-1 abolishes the survival action of IGF-I by a drastic inhibition of the IRS-1/ PI3-kinase/Akt pathway. Altogether, these data bring the first evidence for a new biological role of PTPL1/FAP-1 independent of Fas, which might offer new breast cancer therapeutic strategies. EXPERIMENTAL PROCEDURES Western Blot Analysis of p85/IRS-1 Association—Cells were lysed in lysis buffer, and IRS-1 from extracts containing equal amounts of pro- tein was immunoprecipitated with antiserum against IRS-1 and re- tained on protein G-Sepharose (see PI3-K assay). After washes, the immunoprecipitates were separated on an 10% gel by SDS-PAGE. Pro- teins were electrotransferred to Hybond-P membrane (Amersham Bio- sciences) and incubated with antiserum against p85 (anti PI3-K p85, Upstate Biotechnology). Proteins were finally visualized as previously described. Western Blot Analysis of Akt Activation—Cells were lysed in lysis buffer (see PI3-K assay) and separated on a 10% gel by SDS-PAGE. Proteins were electrotransferred to Hybond-P membrane (Amersham Biosciences) and incubated with anti-Akt or anti-phospho-Akt (New England Biolabs, Ser-473). Proteins were visualized as previously described. Western Blot Analysis of Akt Activation—Cells were lysed in lysis buffer (see PI3-K assay) and separated on a 10% gel by SDS-PAGE. Proteins were electrotransferred to Hybond-P membrane (Amersham Biosciences) and incubated with anti-Akt or anti-phospho-Akt (New England Biolabs, Ser-473). Proteins were visualized as previously described. Terminal Deoxynucleotidyltransferase-mediated dUTP-biotin Nick End Labeling (TUNEL) Assay—Apoptotic cells were detected by TUNEL assay by using the ApopTag direct in situ apoptosis detection kit with peroxidase (Oncor). Attached cells were harvested by trypsinization, combined with unattached cells, fixed for 20 min in 4% paraformaldehyde, permeabilized for 5 min in 0.1% Triton X-100, and resuspended in phosphate-buffered saline. Next 2.5 105 cells were spread on glass slides by use of a Cytospin. Detection of apoptotic cells and counterstaining with hematoxylin were performed according to the instructions provided by the manufacturer. The slides were viewed and photographed, and the apoptotic index was calculated as the percentage of nuclei labeled by peroxidase using a Samba analyzer (Alcatel). At least 103 cells were examined and evaluated in each sample. Western Blot Analysis of IRS-1 Tyrosine Phosphorylation—Cells were lysed in lysis buffer, and IRS-1 from extracts containing equal amounts of protein was immunoprecipitated with antiserum against IRS-1 and retained on protein G-Sepharose (see PI3-K assay). After washes, the immunoprecipitates were separated on an 8% gel by SDS- PAGE. Proteins were electrotransferred to Hybond-P membrane (Am- ersham Biosciences) and incubated with anti-Tyr(P) (4G10) or anti- IRS-1 (Upstate Biotechnology). Proteins were visualized as mentioned above. EXPERIMENTAL PROCEDURES Materials—All biochemicals were from Sigma unless otherwise specified. Cell Culture—MCF7 human breast cancer cells were obtained from the Michigan Cancer Foundation (Detroit, MI). Establishment and characterization of stable PTPL1 antisense-expressing clones were pre- viously described (16), B3 and B10, respectively, corresponding to clones 3 and 10. MCF7 cells and derived clones were maintained in Ham’s F-12/Dulbecco’s modified Eagle’s medium (1:1) supplemented with 10% fetal calf serum (FCS) (Invitrogen). HEK 293 cells were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% FCS (Invitrogen). Before all hormonal treatments, the cells were stripped of endogenous steroids by successive passages in medium without phenol red containing 10% (2 days), then 3% (5 days) charcoal- stripped (DCC) FCS. They were finally treated with E2, IGF-I, OH- Tam, IGF-I plus OH-Tam, or ethanol alone (control) in the presence of 1% DCC/FCS. Western Blot Analysis of PTPL1 Expression—Cells were lysed in radioimmune precipitation assay buffer (50 mM Tris (pH 7.5), 150 mM NaCl, 0.5% Nonidet P-40, 0.1% SDS, 0.5% deoxycholate, 10 mM NaF, 1 mM Na3VO4, 1 mM phenylmethylsulfonyl fluoride, and 100 kallikrein units/ml Trasylol) and separated on a 10% gel by SDS-PAGE. Proteins were electrotransferred to Hybond-P membrane (Amersham Bio- sciences) and incubated with antiserum against PTPL1/FAP-1 (H 300, Santa Cruz Biotechnology) or against actin (Sigma). Proteins were later visualized with horseradish peroxidase-conjugated goat anti-rabbit im- munoglobulin (Sigma) followed by Renaissance chemiluminescence sys- tem (PerkinElmer Life Sciences). Western Blot Analysis of PTPL1 Expression—Cells were lysed in radioimmune precipitation assay buffer (50 mM Tris (pH 7.5), 150 mM NaCl, 0.5% Nonidet P-40, 0.1% SDS, 0.5% deoxycholate, 10 mM NaF, 1 mM Na3VO4, 1 mM phenylmethylsulfonyl fluoride, and 100 kallikrein units/ml Trasylol) and separated on a 10% gel by SDS-PAGE. Proteins were electrotransferred to Hybond-P membrane (Amersham Bio- sciences) and incubated with antiserum against PTPL1/FAP-1 (H 300, Santa Cruz Biotechnology) or against actin (Sigma). Proteins were later visualized with horseradish peroxidase-conjugated goat anti-rabbit im- munoglobulin (Sigma) followed by Renaissance chemiluminescence sys- tem (PerkinElmer Life Sciences). DNA Fragmentation—Attached cells were collected in a buffer con- taining 5 mM Tris (pH 8.0), 10 mM EDTA, and 0.5% Triton X-100 combined with unattached cells and incubated for 20 min at 4 °C. The lysate was centrifuged for 30 min at 13,000 g (4 °C). EXPERIMENTAL PROCEDURES The pellet containing unfragmented DNA was assayed for DNA content by the diaminobenzoic acid fluorometric method (LS-5 spectrometer, 405 nm excitation, 495 nm emission, PerkinElmer Life Sciences) after 2 washes with methanol as in Vignon et al. (6). The supernatant was treated with 300 g of proteinase K/ml for 30 min at 60 °C. DNA was extracted twice with equal volumes of phenol, chloroform, isoamyl alcohol (25:24:1) and precipitated with 1/10 volume of sodium acetate (3 N) and 2.5 volumes of 100% ethanol. The pellets were washed once with 70% ethanol and resuspended in buffer containing 10 mM Tris (pH 8.0), 1 mM EDTA, and 50 mM NaCl in the proportions of 1 l of buffer for 2 g of unfragmented DNA assayed in the first pellet. Cytoplasmic DNA was then treated with 60 g of RNase A/ml for 60 min at 37 °C. 15 l of cytoplasmic DNA was run on a 1.5% agarose gel for 4 h at 50 V. The gel was stained with ethidium bromide and photographed under UV light. DNA Fragmentation—Attached cells were collected in a buffer con- taining 5 mM Tris (pH 8.0), 10 mM EDTA, and 0.5% Triton X-100 combined with unattached cells and incubated for 20 min at 4 °C. The lysate was centrifuged for 30 min at 13,000 g (4 °C). The pellet containing unfragmented DNA was assayed for DNA content by the diaminobenzoic acid fluorometric method (LS-5 spectrometer, 405 nm excitation, 495 nm emission, PerkinElmer Life Sciences) after 2 washes with methanol as in Vignon et al. (6). The supernatant was treated with 300 g of proteinase K/ml for 30 min at 60 °C. DNA was extracted twice with equal volumes of phenol, chloroform, isoamyl alcohol (25:24:1) and precipitated with 1/10 volume of sodium acetate (3 N) and 2.5 volumes of 100% ethanol. The pellets were washed once with 70% ethanol and resuspended in buffer containing 10 mM Tris (pH 8.0), 1 mM EDTA, and 50 mM NaCl in the proportions of 1 l of buffer for 2 g of unfragmented DNA assayed in the first pellet. Cytoplasmic DNA was then treated with 60 g of RNase A/ml for 60 min at 37 °C. 15 l of cytoplasmic DNA was run on a 1.5% agarose gel for 4 h at 50 V. The gel was stained with ethidium bromide and photographed under UV light. RESULTS Biochemical Evaluations of OH-Tam-induced Apoptosis in Human Breast Cancer Cells—Our previous studies about the respective roles of OH-Tam, IGF-I, and the protein-tyrosine phosphatase PTPL1 on MCF7 growth regulation were based upon the evaluation of total cell numbers assessed by DNA cell measurement after diverse treatments (11). However, this glo- bal estimation did not discriminate between the relative con- tributions of the mitotic and/or the apoptotic indexes to the net growth rate. The presence in MCF7 cells of widely accepted morphological modifications induced by the apoptotic process have been long recognized both in vitro (4, 20) and in animal xenografts after steroid antagonist treatments or estrogen withdrawal (21). However, the presence of apoptosis-associated biochemical evidence such as DNA fragmentation remained a matter of debate in the literature (22, 23). Therefore, we have first evaluated whether IGF-I and OH-Tam modulations of apoptosis were associated with DNA strand breaks in our pres- ent culture conditions. We have indeed shown that OH-Tam treatment induced a marked DNA fragmentation in MCF7 Quantitation of Apoptosis by the Cell Death ELISA—2 105 cells were plated on 30-mm duplicate dishes. At the indicated time points, one dish was assayed for total DNA contents by the diaminobenzoic acid method (6), and cells from the other dish were washed once with phosphate-buffered saline and assayed for the presence of nucleosomal fragments in the cytoplasm by a cell death detection ELISA kit (Roche Molecular Biochemicals) according to the manufacturer’s instructions. The ELISA results were normalized to the total DNA value from the duplicate dish. Quantitation of Apoptosis by the Cell Death ELISA—2 105 cells were plated on 30-mm duplicate dishes. At the indicated time points, one dish was assayed for total DNA contents by the diaminobenzoic acid method (6), and cells from the other dish were washed once with phosphate-buffered saline and assayed for the presence of nucleosomal fragments in the cytoplasm by a cell death detection ELISA kit (Roche Molecular Biochemicals) according to the manufacturer’s instructions. The ELISA results were normalized to the total DNA value from the duplicate dish. p PI3-K Assay—Phospholipid mixtures (2 mg/ml) containing PI and phosphatidylserine were dried under a stream of nitrogen and soni- cated twice at 100 watts for 15 s in 10 mM Tris-HCl (pH 7.4). The resulting vesicles (PI/phosphatidylserine) were used as a substrate for PI3-K. The preparation of cell extracts and immunoprecipitation for PI3-K were performed at 4 °C. This paper is available on line at http://www.jbc.org 47861 PTPL1 and Apoptosis in Breast Cancer RESULTS Low molecular mass DNA was extracted and analyzed on a 1.5% agarose gel stained with ethidium bromide. B, after stripping from endogenous steroids, cells were cultured in the presence of the indicated compounds (same concentrations as above) for 7 days before collection of cells and spreading on glass slides. Detection of apoptotic cells was performed by TUNEL assay, and cells with typical nuclear staining were quantified. Bars, 10 m. cells (Fig. 1A, fourth lane). With this technique, we are, thus, able to confirm the apoptotic effect of OH-Tam (4) and observe the survival action of IGF-I in MCF7 cells (Fig. 1A, third lane). Furthermore, by examining DNA laddering, we have shown that the antiestrogen was able to counteract the survival action of IGF-I. DNA degradation was indeed equivalent in control untreated cells and in cells submitted to a 4-day simultaneous treatment with IGF-I and OH-Tam (Fig. 1A, compare lanes two and five). We have quantified apoptosis with two alternative biochemical techniques, the TUNEL assay, which allows the determination of the percentage of labeled apoptotic nuclei after cytocentrifugation of the pooled cell fractions (attached, loosely attached, and floating; Fig. 1B, typical nuclear staining) or a quantitative ELISA of cytoplasmic histone-associated DNA fragments (mono- and oligonucleosomes). Using these two assays, we have first applied a long term treatment (7 days) to quantify apoptosis in wild-type (wt) MCF7 cells in conditions in which we previously demonstrated a drastic inhibition of IGF- I-stimulated overall growth by OH-Tam (11). The two methods gave very similar results and confirmed OH-Tam pro-apoptotic effect as well as its capacity to inhibit IGF-I survival activity in wt MCF7 cells (Fig. 2, A and B, black bars). In the TUNEL assay (Fig. 2A), although less than 10% of the cells were apo- ptotic in control culture conditions, we observed that OH-Tam treatment raised the percentage of apoptosis to greater than 40%, whereas as high as 30% apoptosis was counted in the presence of IGF-I and OH-Tam. Similarly, ELISA provided evidence of a 3.9-fold increase of apoptosis with OH-Tam alone and a 2.6-fold augmentation in the presence of IGF-I and OH- Tam (Fig. 2B). The results obtained with both biochemical methods strongly provided evidence that the drastic increase in OH-Tam-induced apoptosis was responsible for its anti-growth factor activity. RESULTS The cultured cells were washed 3 times with ice-cold phosphate-buffered saline and then lysed in 500 l of lysis buffer (20 mM Tris-HCl (pH 7.4), 137 mM NaCl, 1 mM MgCl2, 1 mM CaCl2, 0.5% Nonidet P-40, 1 mM Na3VO4, 1 mM phenylmethylsulfonyl fluoride, 1 mM dithiothreitol, 10% glycerol, and 100 kallikrein units/ml Trasylol). After gentle shaking for 30 min, the cell extracts were ob- tained by centrifugation in a microcentrifuge at 13,000 g for 5 min. The PI3-K from extracts containing equal amounts of protein, evalu- PI3-K Assay—Phospholipid mixtures (2 mg/ml) containing PI and phosphatidylserine were dried under a stream of nitrogen and soni- cated twice at 100 watts for 15 s in 10 mM Tris-HCl (pH 7.4). The resulting vesicles (PI/phosphatidylserine) were used as a substrate for PI3-K. The preparation of cell extracts and immunoprecipitation for PI3-K were performed at 4 °C. The cultured cells were washed 3 times with ice-cold phosphate-buffered saline and then lysed in 500 l of lysis buffer (20 mM Tris-HCl (pH 7.4), 137 mM NaCl, 1 mM MgCl2, 1 mM CaCl2, 0.5% Nonidet P-40, 1 mM Na3VO4, 1 mM phenylmethylsulfonyl fluoride, 1 mM dithiothreitol, 10% glycerol, and 100 kallikrein units/ml Trasylol). After gentle shaking for 30 min, the cell extracts were ob- tained by centrifugation in a microcentrifuge at 13,000 g for 5 min. The PI3-K from extracts containing equal amounts of protein, evalu- PTPL1 and Apoptosis in Breast Cancer 47863 FIG. 1. OH-Tam induces DNA damages and inhibits IGF-I survival effect in hormone-responsive MCF7 cells. A, DNA fragmentation. Cells were stripped from endogenous steroids and maintained for 4 days in 1% DCC/FCS (Control), supplemented with 10 nM IGF-I, 50 nM OH-Tam, or both simultaneously. Low molecular mass DNA was extracted and analyzed on a 1.5% agarose gel stained with ethidium bromide. B, after stripping from endogenous steroids, cells were cultured in the presence of the indicated compounds (same concentrations as above) for 7 days before collection of cells and spreading on glass slides. Detection of apoptotic cells was performed by TUNEL assay, and cells with typical nuclear staining were quantified. Bars, 10 m. FIG. 1. OH-Tam induces DNA damages and inhibits IGF-I survival effect in hormone-responsive MCF7 cells. A, DNA fragmentation. Cells were stripped from endogenous steroids and maintained for 4 days in 1% DCC/FCS (Control), supplemented with 10 nM IGF-I, 50 nM OH-Tam, or both simultaneously. RESULTS activity required PTPL1 expression and correlated with phos- phatase up-regulation (16) led us to propose that PTPL1 was a key factor in OH-Tam-induced cell death. To demonstrate the role of PTPL1/FAP-1 in apoptosis, we have compared by TUNEL assay or ELISA the effect of various combinations of 7-day treatments on the degree of apoptosis in wt MCF7 cells, the void plasmid mock-transfectant clone (P4), and two MCF7- derived stable clones transfected with PTPL1 cDNA antisense plasmids in which OH-Tam was no longer able to block IGF-I action on growth (16). We showed an overall lower sensitivity of antisense clones to OH-Tam treatment, with a positive corre- lation between resistance to apoptosis and antisense expres- sion (Fig. 2, A and B). After treatment with antiestrogen alone there was a gradual significant decrease in the level of apo- ptosis, from 3.5 to 4 times the control level in wt MCF7 and mock-transfected P4 to 2–3-fold control in low or high antisense expresser B10 or B3 clones (Fig. 2, A and B). After simultane- ous treatment with IGF-I and OH-Tam, the apoptosis level reached 2–3 times the control level (60–70% of OH-Tam alone level) in PTPL1-expressing cells and went down to 1.3–1.5 times the control level in the B10 clone to finally attain the control level or even slightly below the control level in the B3 antisense clone (Fig. 2, A and B). These data clearly point to the essential role of PTPL1 in the apoptotic process. cells (Fig. 1A, fourth lane). With this technique, we are, thus, able to confirm the apoptotic effect of OH-Tam (4) and observe the survival action of IGF-I in MCF7 cells (Fig. 1A, third lane). Furthermore, by examining DNA laddering, we have shown that the antiestrogen was able to counteract the survival action of IGF-I. DNA degradation was indeed equivalent in control untreated cells and in cells submitted to a 4-day simultaneous treatment with IGF-I and OH-Tam (Fig. 1A, compare lanes two and five). We have quantified apoptosis with two alternative biochemical techniques, the TUNEL assay, which allows the determination of the percentage of labeled apoptotic nuclei after cytocentrifugation of the pooled cell fractions (attached, loosely attached, and floating; Fig. 1B, typical nuclear staining) or a quantitative ELISA of cytoplasmic histone-associated DNA fragments (mono- and oligonucleosomes). RESULTS of two or three exper- iments as indicated; * indicates p 0.05, and ** indicates p 0.01 (B3 or B10 versus MCF7 and P4) according to Student’s t test. PTPL1 and Apoptosis in Breast Cancer 47864 FIG. 3. Expression of PTPL1 and quantification of apoptosis by cytoplasmic nucleosomes ELISA after 5 days of antiestrogen treatment. After stripping from endogenous steroids, wild-type MCF7 cells (MCF7-wt), mock-transfected clone (P4), and PTPL1 antisense- transfected clones (B3 and B10) were cultured in the presence of the indicated compounds for 5 days before collection of cells and evaluation of PTPL1 expression in A or quantification of apoptosis by ELISA of cytoplasmic histone-associated DNA fragments in B. A, cells were lysed in radioimmune precipitation assay buffer, and lysates were analyzed for enzyme expression in control (C) and OH-Tam-treated cells by direct immunoblotting with antisera specific for N-terminal PTPL1 (H300, Santa Cruz, top panel) and for actin (bottom panel). PTPL1 expression in HEK 293 cells (HEK) is shown for comparison. B, results obtained with ELISA are expressed as the means S.D. of two or three exper- iments as indicated; * indicates p 0.05, and ** indicates p 0.01 (B3 or B10 versus MCF7 and P4) according to Student’s t test. FIG. 3. Expression of PTPL1 and quantification of apoptosis b t l i l ELISA ft 5 d f ti t FIG. 3. Expression of PTPL1 and quantification of apoptosis by cytoplasmic nucleosomes ELISA after 5 days of antiestrogen treatment. After stripping from endogenous steroids, wild-type MCF7 cells (MCF7-wt), mock-transfected clone (P4), and PTPL1 antisense- transfected clones (B3 and B10) were cultured in the presence of the indicated compounds for 5 days before collection of cells and evaluation of PTPL1 expression in A or quantification of apoptosis by ELISA of cytoplasmic histone-associated DNA fragments in B. A, cells were lysed in radioimmune precipitation assay buffer, and lysates were analyzed for enzyme expression in control (C) and OH-Tam-treated cells by direct immunoblotting with antisera specific for N-terminal PTPL1 (H300, Santa Cruz, top panel) and for actin (bottom panel). PTPL1 expression in HEK 293 cells (HEK) is shown for comparison. B, results obtained with ELISA are expressed as the means S.D. of two or three exper- iments as indicated; * indicates p 0.05, and ** indicates p 0.01 (B3 or B10 versus MCF7 and P4) according to Student’s t test. FIG. RESULTS 2. Quantification of apoptosis by TUNEL assay or cyto- plasmic nucleosomes ELISA in PTPL1-positive or -defective cells. After stripping from endogenous steroids, wt MCF7 cells (MCF7- wt), mock-transfected clone (P4), and PTPL1 antisense transfected clones (B3 and B10) were cultured in the presence of the indicated compounds for 7 days (A and B). After collection of cells, apoptosis was quantified by TUNEL assay (A) or ELISA of cytoplasmic histone-asso- ciated DNA fragments (B). Results are expressed as the means S.D. of two or three experiments as indicated; in A, * indicates p 0.05, and ** indicates p 0.01 (B3 versus MCF7 and P4); in B, * indicates p 0.05, and ** indicates p 0.01 (B3 or B10 versus MCF7 and P4) according to Student’s t test. (C, control.) Ref. 24), but the survival pathway used by IGF-I against ap- optotic injury is uncertain. In many cell lines, the importance of the PI3-K/Akt pathway has been demonstrated (25), but it is also clear that the IGF-IR is connected to several alternative pathways, which include mitogen-activated protein kinase ac- tivation (26). We therefore tested in MCF7 cells the impact of specific inhibitors of these two major pathways on apoptosis and IGF-I survival effect. Representative data of Fig. 4 clearly demonstrated that the MEK-1 (mitogen-activated protein ki- nase/extracellular signal-regulated kinase kinase) inhibitor PD98059 was without effect, whereas the PI3-K inhibitor LY294002 induced apoptosis and partially inhibited IGF-I pro- tective effect in this breast cancer cell line. This partial remain- ing of IGF-I survival signaling indicated that, as in other cell lines (27), additional connecting pathways contributed to IGF-I antiapoptotic activity in MCF7 cells. However, the results ob- tained with LY294002 pointed to the fundamental function of the PI3-K/Akt pathway in the control of cell death in MCF7 cells, and furthermore, the similarity between LY294002 and OH-Tam action suggested an inhibitory action of the antiestro- gen, and most probably of PTPL1, on the PI3-K/Akt pathway (Fig. 4). verely reduced in B10 and completely abolished in B3 as compared with wt MCF7 or P4 mock transfectant (Fig. 3A). We then quantified apoptosis in wt MCF7 and on the three stable derived transfectants after a 5-day treatment with the various combinations of compounds. In these conditions, OH-Tam dis- played the same apoptotic activity and capacity to reverse IGF-I survival effect in wt or mock-transfected MCF7 cells (Fig. 3B). RESULTS Using these two assays, we have first applied a long term treatment (7 days) to quantify apoptosis in wild-type (wt) MCF7 cells in conditions in which we previously demonstrated a drastic inhibition of IGF- I-stimulated overall growth by OH-Tam (11). The two methods gave very similar results and confirmed OH-Tam pro-apoptotic effect as well as its capacity to inhibit IGF-I survival activity in wt MCF7 cells (Fig. 2, A and B, black bars). In the TUNEL assay (Fig. 2A), although less than 10% of the cells were apo- ptotic in control culture conditions, we observed that OH-Tam treatment raised the percentage of apoptosis to greater than 40%, whereas as high as 30% apoptosis was counted in the presence of IGF-I and OH-Tam. Similarly, ELISA provided evidence of a 3.9-fold increase of apoptosis with OH-Tam alone and a 2.6-fold augmentation in the presence of IGF-I and OH- Tam (Fig. 2B). The results obtained with both biochemical methods strongly provided evidence that the drastic increase in OH-Tam-induced apoptosis was responsible for its anti-growth factor activity. In the B3 clone, after 7 days of OH-Tam treatment we ob- served a residual induction of apoptosis and a remaining inhi- bition of IGF-I survival effect (Fig. 2, A and B). These effects could result either from an accumulation of PTPL1 mRNA bypassing antisense inhibition or from distinct mechanisms of apoptosis induction by OH-Tam that do not require PTPL1 expression. We had shown that a 5-day treatment with OH- Tam was sufficient to increase PTP activity in wt MCF7, whereas it remained inefficient in B3 and B10 clones (16). By direct Western blotting with an N-terminal PTPL1/FAP-1 an- tibody, we have now demonstrated that enzyme protein expres- sion and antiestrogen regulation (5-day treatment) were se- The Protein-tyrosine Phosphatase PTPL1 Plays a Key Role in Apoptosis Induction—The present results associated with our previous data demonstrating that OH-Tam anti-growth factor PTPL1 and Apoptosis in Breast Cancer 47864 verely reduced in B10 and completely abolished in B3 as compared with wt MCF7 or P4 mock transfectant (Fig. 3A). We th tifi d t i i t MCF7 d th th t bl Ref. 24), but the survival pathway used by IGF-I against ap- optotic injury is uncertain. RESULTS In many cell lines, the importance of the PI3-K/Akt pathway has been demonstrated (25), but it is also clear that the IGF-IR is connected to several alternative pathways, which include mitogen-activated protein kinase ac- tivation (26). We therefore tested in MCF7 cells the impact of specific inhibitors of these two major pathways on apoptosis and IGF-I survival effect. Representative data of Fig. 4 clearly demonstrated that the MEK-1 (mitogen-activated protein ki- nase/extracellular signal-regulated kinase kinase) inhibitor PD98059 was without effect, whereas the PI3-K inhibitor LY294002 induced apoptosis and partially inhibited IGF-I pro- t ti ff t i thi b t ll li Thi ti l i FIG. 2. Quantification of apoptosis by TUNEL assay or cyto- plasmic nucleosomes ELISA in PTPL1-positive or -defective cells. After stripping from endogenous steroids, wt MCF7 cells (MCF7- wt), mock-transfected clone (P4), and PTPL1 antisense transfected clones (B3 and B10) were cultured in the presence of the indicated compounds for 7 days (A and B). After collection of cells, apoptosis was quantified by TUNEL assay (A) or ELISA of cytoplasmic histone-asso- ciated DNA fragments (B). Results are expressed as the means S.D. of two or three experiments as indicated; in A, * indicates p 0.05, and ** indicates p 0.01 (B3 versus MCF7 and P4); in B, * indicates p 0.05, and ** indicates p 0.01 (B3 or B10 versus MCF7 and P4) according to Student’s t test. (C, control.) FIG. 3. Expression of PTPL1 and quantification of apoptosis by cytoplasmic nucleosomes ELISA after 5 days of antiestrogen treatment. After stripping from endogenous steroids, wild-type MCF7 cells (MCF7-wt), mock-transfected clone (P4), and PTPL1 antisense- transfected clones (B3 and B10) were cultured in the presence of the indicated compounds for 5 days before collection of cells and evaluation of PTPL1 expression in A or quantification of apoptosis by ELISA of cytoplasmic histone-associated DNA fragments in B. A, cells were lysed in radioimmune precipitation assay buffer, and lysates were analyzed for enzyme expression in control (C) and OH-Tam-treated cells by direct immunoblotting with antisera specific for N-terminal PTPL1 (H300, Santa Cruz, top panel) and for actin (bottom panel). PTPL1 expression in HEK 293 cells (HEK) is shown for comparison. B, results obtained with ELISA are expressed as the means S.D. RESULTS We have therefore estimated Akt activation by probing Western blots with specific antibody to phosphorylated Akt and evi- denced a drastic inhibition (55%) of its activation in wild-type and mock-transfected MCF7 cells (see Fig. 8, A and B). This indicates that the reduction in PI3-K activation by OH-Tam was sufficient for severely impeding this survival pathway. We have shown (Fig. 5A) that IRS-1 expression was not affected by OH-Tam treatment. Therefore, inhibition of PI3-K activity ob- served after 4 days of OH-Tam treatment in MCF7 cells could be due to a direct PI3-K inhibition or IRS-1 dephosphorylation. We have then evaluated IRS-1 tyrosine phosphorylation by IRS-1 immunoprecipitation and anti-phosphotyrosine Western blots. In mock-transfected MCF7 cells (P4) we observed a dras- tic inhibition of IRS-1 phosphorylation (65%) without any loss of IRS-1 expression (see Fig. 9, A and B). This indicates that OH-Tam is acting on IRS-1/PI3-K/Akt pathway. TPL1 expression is necessary for PI3-K inhibitio FIG. 5. PTPL1 expression is necessary for PI3-K inhibition. Wild-type MCF7 (MCF7-wt), mock-transfected clone (P4), and PTPL1 antisense-transfected clones (B3 and B10) cultured in 1% DCC/FCS were treated for 4 days with 50 nM OH-Tam or vehicle alone (ethanol) before 10 min of IGF-I (20 nM) stimulation. PI3-K activity was assayed in the presence of [-32P]ATP on IRS-1 immunoprecipitate from cell lysate using phosphatidylinositol as substrate. A, immunoblotting of immunoprecipitated IRS-1. B, a representative autoradiography of phosphatidylinositol 3-phosphate separated by thin-layer chromatogra- phy. C, quantitative analysis of PI3-K activity in cells pretreated with OH-Tam before IGF-I stimulation. Results are expressed as a percent- age of the activity quantified in cells pretreated with vehicle alone. Results are the mean S.D. of two to four experiments; * indicates p 0.02, and ** indicates p 0.01 (B10 or B3 versus MCF7 and P4) according to Student’s t test. tained in the two MCF7-derived PTPL1-defective clones (B3 and B10) (Fig. 3B). As expected, OH-Tam no longer blocked IRS-1 phosphorylation, p85 association with IRS-1, PI3-K ac- tivity, and Akt activation by IGF-I in these clones (Figs. 5–9), thus confirming the inhibitory role of PTPL1 on PI3-K/Akt pathway. In previous studies (11, 16), we indeed showed that significant induction of PTP activity and PTPL1 expression required 2–3 days of treatment with antiestrogen. In a time- course experiment (Fig. RESULTS 6A, bottom panel). These data indicate, as suggested previously when com- paring LY294002 and OH-Tam effects on apoptosis (Fig. 4), that the antiestrogen has a strong inhibitory action on the PI3-K/Akt pathway. We did not know yet if the residual PI3-K activity observed after 4 days of OH-Tam treatment in MCF7 cells could be sufficient for full activation of the downstream effector in this survival pathway, the activation of Akt. We have therefore estimated Akt activation by probing Western blots with specific antibody to phosphorylated Akt and evi- denced a drastic inhibition (55%) of its activation in wild-type and mock-transfected MCF7 cells (see Fig. 8, A and B). This indicates that the reduction in PI3-K activation by OH-Tam was sufficient for severely impeding this survival pathway. We have shown (Fig. 5A) that IRS-1 expression was not affected by OH-Tam treatment. Therefore, inhibition of PI3-K activity ob- served after 4 days of OH-Tam treatment in MCF7 cells could be due to a direct PI3-K inhibition or IRS-1 dephosphorylation. We have then evaluated IRS-1 tyrosine phosphorylation by IRS-1 immunoprecipitation and anti-phosphotyrosine Western blots. In mock-transfected MCF7 cells (P4) we observed a dras- tic inhibition of IRS-1 phosphorylation (65%) without any loss of IRS-1 expression (see Fig. 9, A and B). This indicates that OH-Tam is acting on IRS-1/PI3-K/Akt pathway. pathway in PTPL1 action, we have tested the effect of OH-Tam pretreatment, on the association of PI-3K p85 regulatory sub- unit with IRS-1, and on the PI-3K activity after IGF-I activa- tion in wt MCF7 cells and transfectants. OH-Tam significantly reduced PI3-K activation by IGF-I in wild-type and P4 mock- transfected MCF7 cells (respectively, 75 and 85% inhibition) (Fig. 5, B and C; see Fig. 7) without any significant change in IRS-1 expression (Fig. 5A). Moreover, OH-Tam selectively in- hibited to the same extent the p85/IRS-1 association in the P4 mock-transfected clone (90% inhibition) (Fig. 6A, top panel, and B) without any change in p85 expression (Fig. 6A, bottom panel). These data indicate, as suggested previously when com- paring LY294002 and OH-Tam effects on apoptosis (Fig. 4), that the antiestrogen has a strong inhibitory action on the PI3-K/Akt pathway. We did not know yet if the residual PI3-K activity observed after 4 days of OH-Tam treatment in MCF7 cells could be sufficient for full activation of the downstream effector in this survival pathway, the activation of Akt. RESULTS On the opposite, the antiestrogen could not significantly induce apoptosis and led to partial (B10) or no (B3) inhibition of IGF-I survival effect in PTPL1-defective clones (Fig. 3B). After such a shorter period of treatment, we obtained a perfect cor- relation between PTPL1 expression and apoptosis induction or IGF-I survival effect inhibition. The next step was therefore to determine which signaling pathway was affected in PTPL1 apoptosis induction. verely reduced in B10 and completely abolished in B3 as compared with wt MCF7 or P4 mock transfectant (Fig. 3A). We then quantified apoptosis in wt MCF7 and on the three stable derived transfectants after a 5-day treatment with the various combinations of compounds. In these conditions, OH-Tam dis- played the same apoptotic activity and capacity to reverse IGF-I survival effect in wt or mock-transfected MCF7 cells (Fig. 3B). On the opposite, the antiestrogen could not significantly induce apoptosis and led to partial (B10) or no (B3) inhibition of IGF-I survival effect in PTPL1-defective clones (Fig. 3B). After such a shorter period of treatment, we obtained a perfect cor- relation between PTPL1 expression and apoptosis induction or IGF-I survival effect inhibition. The next step was therefore to determine which signaling pathway was affected in PTPL1 apoptosis induction. This pathway originates with the phosphorylation of the major IGF-IR substrate, IRS-1 (28), which in turn activates PI3-K and downstream Akt (25). Among the concluding steps are the inhibitory phosphorylations by Akt of the caspase 9 (29) and of Bad (30), one of the pro-apoptotic members of the Bcl-2 family of proteins. To further document the involvement of this PTPL1-induced Apoptosis Is Mediated by Inhibition of the PI3-K/Akt Pathway—It is well established that IGF-I has a strong antiapoptotic function in many cell lines (for review, see FIG. 4. PI3-kinase activity is crucial for MCF7 cell survival. After steroid withdrawal cells were treated for 3 days with PD 98059 (MEK-1 (mitogen-activated protein kinase/extracellular signal-regu- lated kinase kinase) inhibitor; 75 M) and LY 294002 (PI3-K inhibitor 25 M) in combination or not with IGF-I (10 nM) or for 7 days with OH-Tam (50 nM) in combination or not with IGF-I. Apoptosis was quantified by ELISA of cytoplasmic histone-associated DNA fragments. PTPL1 and Apopto PTPL1 and Apoptosis in Breast Cancer d h C d d d f l ( FIG. 5. PTPL1 expression is necessary for PI3-K inhibition. RESULTS PI3-kinase activity is crucial for MCF7 cell survival. After steroid withdrawal cells were treated for 3 days with PD 98059 (MEK-1 (mitogen-activated protein kinase/extracellular signal-regu- lated kinase kinase) inhibitor; 75 M) and LY 294002 (PI3-K inhibitor 25 M) in combination or not with IGF-I (10 nM) or for 7 days with OH-Tam (50 nM) in combination or not with IGF-I. Apoptosis was quantified by ELISA of cytoplasmic histone-associated DNA fragments. FIG. 5. PTPL1 expression is necessary for PI3-K inhibition. Wild-type MCF7 (MCF7-wt), mock-transfected clone (P4), and PTPL1 antisense-transfected clones (B3 and B10) cultured in 1% DCC/FCS were treated for 4 days with 50 nM OH-Tam or vehicle alone (ethanol) before 10 min of IGF-I (20 nM) stimulation. PI3-K activity was assayed in the presence of [-32P]ATP on IRS-1 immunoprecipitate from cell lysate using phosphatidylinositol as substrate. A, immunoblotting of immunoprecipitated IRS-1. B, a representative autoradiography of phosphatidylinositol 3-phosphate separated by thin-layer chromatogra- phy. C, quantitative analysis of PI3-K activity in cells pretreated with OH-Tam before IGF-I stimulation. Results are expressed as a percent- age of the activity quantified in cells pretreated with vehicle alone. Results are the mean S.D. of two to four experiments; * indicates p 0.02, and indicates p 0.01 (B10 or B3 versus MCF7 and P4) according to Student’s t test. p p FIG. 4. PI3-kinase activity is crucial for MCF7 cell survival. After steroid withdrawal cells were treated for 3 days with PD 98059 (MEK-1 (mitogen-activated protein kinase/extracellular signal-regu- lated kinase kinase) inhibitor; 75 M) and LY 294002 (PI3-K inhibitor 25 M) in combination or not with IGF-I (10 nM) or for 7 days with OH-Tam (50 nM) in combination or not with IGF-I. Apoptosis was quantified by ELISA of cytoplasmic histone-associated DNA fragments. pathway in PTPL1 action, we have tested the effect of OH-Tam pretreatment, on the association of PI-3K p85 regulatory sub- unit with IRS-1, and on the PI-3K activity after IGF-I activa- tion in wt MCF7 cells and transfectants. OH-Tam significantly reduced PI3-K activation by IGF-I in wild-type and P4 mock- transfected MCF7 cells (respectively, 75 and 85% inhibition) (Fig. 5, B and C; see Fig. 7) without any significant change in IRS-1 expression (Fig. 5A). Moreover, OH-Tam selectively in- hibited to the same extent the p85/IRS-1 association in the P4 mock-transfected clone (90% inhibition) (Fig. 6A, top panel, and B) without any change in p85 expression (Fig. PTPL1/FAP-1 Is Not Associated to Fas Pathway in Breast RESULTS of two to four experiment 0.02, and indicates p 0.01 (B10 or B3 versus according to Student’s t test. PTPL1 and Apoptosis in Breast Cancer pathway in PTPL1 action, we have tested the effect of OH-Tam pretreatment, on the association of PI-3K p85 regulatory sub- unit with IRS-1, and on the PI-3K activity after IGF-I activa- tion in wt MCF7 cells and transfectants. OH-Tam significantly reduced PI3-K activation by IGF-I in wild-type and P4 mock- transfected MCF7 cells (respectively, 75 and 85% inhibition) (Fig. 5, B and C; see Fig. 7) without any significant change in IRS-1 expression (Fig. 5A). Moreover, OH-Tam selectively in- hibited to the same extent the p85/IRS-1 association in the P4 mock-transfected clone (90% inhibition) (Fig. 6A, top panel, and B) without any change in p85 expression (Fig. 6A, bottom panel). These data indicate, as suggested previously when com- paring LY294002 and OH-Tam effects on apoptosis (Fig. 4), that the antiestrogen has a strong inhibitory action on the PI3-K/Akt pathway. We did not know yet if the residual PI3-K activity observed after 4 days of OH-Tam treatment in MCF7 cells could be sufficient for full activation of the downstream effector in this survival pathway, the activation of Akt. We have therefore estimated Akt activation by probing Western blots with specific antibody to phosphorylated Akt and evi- denced a drastic inhibition (55%) of its activation in wild-type and mock-transfected MCF7 cells (see Fig. 8, A and B). This indicates that the reduction in PI3-K activation by OH-Tam was sufficient for severely impeding this survival pathway. We have shown (Fig. 5A) that IRS-1 expression was not affected by OH-Tam treatment. Therefore, inhibition of PI3-K activity ob- served after 4 days of OH-Tam treatment in MCF7 cells could be due to a direct PI3-K inhibition or IRS-1 dephosphorylation. We have then evaluated IRS-1 tyrosine phosphorylation by IRS-1 immunoprecipitation and anti-phosphotyrosine Western blots. In mock-transfected MCF7 cells (P4) we observed a dras- tic inhibition of IRS-1 phosphorylation (65%) without any loss tained in the two MCF7-derived PTPL1-defective clones (B3 and B10) (Fig. 3B). As expected, OH-Tam no longer blocked IRS-1 phosphorylation, p85 association with IRS-1, PI3-K ac- tivity, and Akt activation by IGF-I in these clones (Figs. 5–9), thus confirming the inhibitory role of PTPL1 on PI3-K/Akt pathway. In previous studies (11, 16), we indeed showed that significant induction of PTP activity and PTPL1 expression FIG. 4. RESULTS Wild-type MCF7 (MCF7-wt), mock-transfected clone (P4), and PTPL1 antisense-transfected clones (B3 and B10) cultured in 1% DCC/FCS were treated for 4 days with 50 nM OH-Tam or vehicle alone (ethanol) before 10 min of IGF-I (20 nM) stimulation. PI3-K activity was assayed in the presence of [-32P]ATP on IRS-1 immunoprecipitate from cell lysate using phosphatidylinositol as substrate. A, immunoblotting of immunoprecipitated IRS-1. B, a representative autoradiography of phosphatidylinositol 3-phosphate separated by thin-layer chromatogra- phy. C, quantitative analysis of PI3-K activity in cells pretreated with OH-Tam before IGF-I stimulation. Results are expressed as a percent- age of the activity quantified in cells pretreated with vehicle alone. Results are the mean S.D. of two to four experiments; * indicates p 0.02, and ** indicates p 0.01 (B10 or B3 versus MCF7 and P4) according to Student’s t test. sis in Breast Cancer 47865 47865 way in PTPL1 action, we have tested the effect of OH-Tam reatment, on the association of PI-3K p85 regulatory sub- with IRS-1, and on the PI-3K activity after IGF-I activa- in wt MCF7 cells and transfectants. OH-Tam significantly ced PI3-K activation by IGF-I in wild-type and P4 mock- sfected MCF7 cells (respectively, 75 and 85% inhibition) 5, B and C; see Fig. 7) without any significant change in 1 expression (Fig. 5A). Moreover, OH-Tam selectively in- ed to the same extent the p85/IRS-1 association in the P4 k-transfected clone (90% inhibition) (Fig. 6A, top panel, and without any change in p85 expression (Fig. 6A, bottom l). These data indicate, as suggested previously when com- ng LY294002 and OH-Tam effects on apoptosis (Fig. 4), the antiestrogen has a strong inhibitory action on the K/Akt pathway. We did not know yet if the residual PI3-K ity observed after 4 days of OH-Tam treatment in MCF7 could be sufficient for full activation of the downstream tor in this survival pathway, the activation of Akt. We therefore estimated Akt activation by probing Western with specific antibody to phosphorylated Akt and evi- ed a drastic inhibition (55%) of its activation in wild-type mock-transfected MCF7 cells (see Fig. 8, A and B). This ates that the reduction in PI3-K activation by OH-Tam sufficient for severely impeding this survival pathway. We shown (Fig. 5A) that IRS-1 expression was not affected by Tam treatment. RESULTS Therefore, inhibition of PI3-K activity ob- ed after 4 days of OH-Tam treatment in MCF7 cells could ue to a direct PI3-K inhibition or IRS-1 dephosphorylation. have then evaluated IRS-1 tyrosine phosphorylation by 1 immunoprecipitation and anti-phosphotyrosine Western . In mock-transfected MCF7 cells (P4) we observed a dras- nhibition of IRS-1 phosphorylation (65%) without any loss RS-1 expression (see Fig. 9, A and B). This indicates that Tam is acting on IRS-1/PI3-K/Akt pathway. ur previous experiments indicated that PTPL1 was a key ator of OH-Tam-induced apoptosis (Figs. 2 and 3), and our experiments demonstrated that these apoptotic effects are ated by inhibition of the Akt/PI3-K pathway (Fig. 4), thus ting to a role of PTPL1 on this pathway. To evaluate tained in the two MCF7-derived PTPL1-defect and B10) (Fig. 3B). As expected, OH-Tam no IRS-1 phosphorylation, p85 association with IR tivity, and Akt activation by IGF-I in these clon thus confirming the inhibitory role of PTPL1 pathway. In previous studies (11, 16), we indee significant induction of PTP activity and PTP required 2–3 days of treatment with antiestrog course experiment (Fig. 7), we confirmed in the tant P4 the perfect correlation between the tim PTPL1 induction of expression and the neutraliz PI3-K/Akt pathway, whereas no effect was seen defective B3 clone. Altogether our data on apo type and PTPL1 antisense transfectants are the G. 4. PI3-kinase activity is crucial for MCF7 cell survival. steroid withdrawal cells were treated for 3 days with PD 98059 K-1 (mitogen-activated protein kinase/extracellular signal-regu- kinase kinase) inhibitor; 75 M) and LY 294002 (PI3-K inhibitor M) in combination or not with IGF-I (10 nM) or for 7 days with am (50 nM) in combination or not with IGF-I. Apoptosis was tified by ELISA of cytoplasmic histone-associated DNA fragments. FIG. 5. PTPL1 expression is necessary for PI Wild-type MCF7 (MCF7-wt), mock-transfected clone antisense-transfected clones (B3 and B10) cultured were treated for 4 days with 50 nM OH-Tam or vehicl before 10 min of IGF-I (20 nM) stimulation. PI3-K acti in the presence of [-32P]ATP on IRS-1 immunoprec lysate using phosphatidylinositol as substrate. A, im immunoprecipitated IRS-1. B, a representative aut phosphatidylinositol 3-phosphate separated by thin-lay phy. C, quantitative analysis of PI3-K activity in cells OH-Tam before IGF-I stimulation. Results are expres age of the activity quantified in cells pretreated wit Results are the mean S.D. RESULTS Results were expressed as a percentage of the activity quantified in cells pretreated with vehicle alone. Results are the means S.D. of two experiments; * indicates p 0.05, and ** indicates p 0.01 (B3 versus P4) according to Student’s t test. FIG. 6. Inhibition of p85 association with IRS-1 in PTPL1-pos- itive or -defective cells. Mock-transfected clone (P4) and PTPL1 antisense transfected clone (B3) cultured in 1% DCC/FCS were treated for 4 days with 50 nM OH-Tam or vehicle alone (ethanol) before a 10-min IGF-I (20 nM) stimulation. A, IRS-1 immunoprecipitates were analyzed by immunoblotting (WB) with an antibody specific for p85 (IP IRS-1; top panel). The same cell lysates were analyzed by direct immu- noblotting with the antiserum specific for p85 (Direct; bottom panel). B, the signals from the experiments were quantified by scanning densi- tometry, and the level of p85 associated with IRS-1 in each sample was normalized to the level of total p85. (C, control.) Cancer Cells—On the basis of data obtained in Jurkat leuke- mia cells, Sato et al. (15) proposes a model for an anti-apoptotic role of PTPL1/FAP-1 by inhibition of Fas death receptor trans- duction through interactions of PTPL1/FAP-1 PDZ domains with Fas cytoplasmic terminal domain. They further suggest that the uncontrolled growth of cells in some pathologies in- cluding cancer could result from a correlation between PTPL1/ FAP-1 expression and resistance to Fas-induced cell death. A series of investigations has confirmed this assumption (31–33), whereas other groups failed to find such a correlation (34–36), which altogether seemed to highly depend on the cell line examined and therefore strongly suggested that it was a tissue- or a cell-specific event. We have evaluated the ability of an anti-Fas antibody to induce apoptosis in three estrogen-sensi- tive breast cancer cell lines, which expressed varying levels of PTPL1/FAP-1, wild-type MCF7 cells (low expression), T47D cells (3.5-fold more expression than in MCF7 cells) (Fig. 10B), and the B3 clone (no expression). Fig. 10A displayed the ab- sence of Fas-mediated apoptosis in MCF7 cells and a 20-fold induction of apoptosis by Fas antibody in the T47D cell line, clearly providing evidence for the absence of a positive corre- lation between PTPL1/FAP-1 expression and Fas resistance in these breast cancer cell lines. RESULTS Furthermore, the absence of sensitivity to Fas antibody in B3 clone (which did not express PTPL1) demonstrated that the PTPL1/FAP-1 expression was not responsible for Fas resistance in human breast cancer cells. FIG. 7. Time-course inhibition of PI3-K activation in PTPL1- positive or -defective cells. Mock-transfected clone (P4) and PTPL1 antisense transfected clone (B3) cultured in 1% DCC/FCS were treated for 1–4 days (d) with 50 nM OH-Tam or vehicle alone (ethanol) before 10 min of IGF-I (20 nM) stimulation. PI3-K activity associated to IRS-1 was assayed as in Fig. 4. A, a representative autoradiography of phosphati- dylinositol 3-phosphate separated by thin-layer chromatography. (C, control.) B, quantitative analysis of PI3-K activity in cells pretreated with OH-Tam before IGF-I stimulation. Results were expressed as a percentage of the activity quantified in cells pretreated with vehicle alone. Results are the means S.D. of two experiments; * indicates p 0.05, and ** indicates p 0.01 (B3 versus P4) according to Student’s t test. cer. The role of PTP in this process in breast cancer is poorly documented, and the relation between PTPL1 and apoptosis is still a matter of debate in the literature and seems to be highly dependent on cell type or species. In this study, we have ad- dressed the role of PTPL1 in the model of antiestrogen-induced apoptosis in hormone-responsive human breast cancer cells positive or defective for this particular tyrosine phosphatase. We have shown that PTPL1 is necessary for the early inhibi- tion of IRS-1/PI3-K/Akt pathway, which leads to increased apoptosis and inhibition of IGF-I survival effect. Therefore these results first demonstrate that PTPL1 has a pro-apoptotic effect in human breast cancer cells. PTPL1 Is a Pro-apoptotic Enzyme Acting on the PI3-kinase/ Akt Pathway—Several studies have shown that antiestrogen inhibits growth factor effects on proliferation (6, 7) as well as on activation of AP-1-controlled genes (8) and IRS-1 phosphoryl- ation (37). We have presently demonstrated that OH-Tam not only emphasizes the apoptotic effect of steroid withdrawal but RESULTS 7), we confirmed in the mock transfec- tant P4 the perfect correlation between the time necessary for PTPL1 induction of expression and the neutralizing activity on PI3-K/Akt pathway, whereas no effect was seen in the PTPL1- defective B3 clone. Altogether our data on apoptosis in wild- type and PTPL1 antisense transfectants are therefore concord- ant with the concept that PTPL1 is a pro-apoptotic enzyme in estrogen receptor-positive breast cancer cells by early inhibi- tion of the PI3-K/Akt pathway. tained in the two MCF7-derived PTPL1-defective clones (B3 and B10) (Fig. 3B). As expected, OH-Tam no longer blocked IRS-1 phosphorylation, p85 association with IRS-1, PI3-K ac- tivity, and Akt activation by IGF-I in these clones (Figs. 5–9), thus confirming the inhibitory role of PTPL1 on PI3-K/Akt pathway. In previous studies (11, 16), we indeed showed that significant induction of PTP activity and PTPL1 expression required 2–3 days of treatment with antiestrogen. In a time- course experiment (Fig. 7), we confirmed in the mock transfec- tant P4 the perfect correlation between the time necessary for PTPL1 induction of expression and the neutralizing activity on PI3-K/Akt pathway, whereas no effect was seen in the PTPL1- defective B3 clone. Altogether our data on apoptosis in wild- type and PTPL1 antisense transfectants are therefore concord- ant with the concept that PTPL1 is a pro-apoptotic enzyme in estrogen receptor-positive breast cancer cells by early inhibi- tion of the PI3-K/Akt pathway. Our previous experiments indicated that PTPL1 was a key mediator of OH-Tam-induced apoptosis (Figs. 2 and 3), and our last experiments demonstrated that these apoptotic effects are mediated by inhibition of the Akt/PI3-K pathway (Fig. 4), thus pointing to a role of PTPL1 on this pathway. To evaluate PTPL1 contribution, we have thus evaluated the effect of a 4-day OH-Tam treatment on IRS-1 phosphorylation, p85 asso- ciation with IRS-1, PI3-K activity, and Akt activation in wt MCF7 or P4 mock transfectant by comparison with that ob- PTPL1/FAP-1 Is Not Associated to Fas Pathway in Breast PTPL1 and Apoptosis in Breast Cancer Cancer Cells—On the basis of data obtained in Jurkat leuke- mia cells, Sato et al. (15) proposes a model for an anti-apoptotic role of PTPL1/FAP-1 by inhibition of Fas death receptor trans- duction through interactions of PTPL1/FAP-1 PDZ domains with Fas cytoplasmic terminal domain. RESULTS They further suggest that the uncontrolled growth of cells in some pathologies in- cluding cancer could result from a correlation between PTPL1/ FAP-1 expression and resistance to Fas-induced cell death. A series of investigations has confirmed this assumption (31–33), whereas other groups failed to find such a correlation (34–36), which altogether seemed to highly depend on the cell line examined and therefore strongly suggested that it was a tissue- or a cell-specific event. We have evaluated the ability of an anti-Fas antibody to induce apoptosis in three estrogen-sensi- tive breast cancer cell lines, which expressed varying levels of PTPL1/FAP-1, wild-type MCF7 cells (low expression), T47D cells (3.5-fold more expression than in MCF7 cells) (Fig. 10B), and the B3 clone (no expression). Fig. 10A displayed the ab- sence of Fas-mediated apoptosis in MCF7 cells and a 20-fold induction of apoptosis by Fas antibody in the T47D cell line, clearly providing evidence for the absence of a positive corre- l ti b t PTPL1/FAP 1 i d F i t i cer. The role of PTP in this process in breast cancer is poorly documented, and the relation between PTPL1 and apoptosis is still a matter of debate in the literature and seems to be highly dependent on cell type or species. In this study, we have ad- dressed the role of PTPL1 in the model of antiestrogen-induced apoptosis in hormone-responsive human breast cancer cells positive or defective for this particular tyrosine phosphatase. We have shown that PTPL1 is necessary for the early inhibi- tion of IRS-1/PI3-K/Akt pathway, which leads to increased apoptosis and inhibition of IGF I survival effect Therefore FIG. 6. Inhibition of p85 association with IRS-1 in PTPL1-pos- itive or -defective cells. Mock-transfected clone (P4) and PTPL1 antisense transfected clone (B3) cultured in 1% DCC/FCS were treated for 4 days with 50 nM OH-Tam or vehicle alone (ethanol) before a 10-min IGF-I (20 nM) stimulation. A, IRS-1 immunoprecipitates were analyzed by immunoblotting (WB) with an antibody specific for p85 (IP IRS-1; top panel). The same cell lysates were analyzed by direct immu- noblotting with the antiserum specific for p85 (Direct; bottom panel). B, the signals from the experiments were quantified by scanning densi- tometry, and the level of p85 associated with IRS-1 in each sample was normalized to the level of total p85. (C, control.) FIG. 7. Time-course inhibition of PI3-K activation in PTPL1- positive or -defective cells. RESULTS Mock-transfected clone (P4) and PTPL1 antisense transfected clone (B3) cultured in 1% DCC/FCS were treated for 1–4 days (d) with 50 nM OH-Tam or vehicle alone (ethanol) before 10 min of IGF-I (20 nM) stimulation. PI3-K activity associated to IRS-1 was assayed as in Fig. 4. A, a representative autoradiography of phosphati- dylinositol 3-phosphate separated by thin-layer chromatography. (C, control.) B, quantitative analysis of PI3-K activity in cells pretreated with OH-Tam before IGF-I stimulation. Results were expressed as a percentage of the activity quantified in cells pretreated with vehicle alone. Results are the means S.D. of two experiments; * indicates p 0.05, and ** indicates p 0.01 (B3 versus P4) according to Student’s t test. PTPL1 and Apoptosis in Breast Cancer 47866 ke- otic ns- ins est in- L1/ . A 33), 36), ine ue- an nsi- s of 7D B), ab- old ne, rre- e in of ess was lls. cer. The role of PTP in this process in breast cancer is poorly documented, and the relation between PTPL1 and apoptosis is still a matter of debate in the literature and seems to be highly dependent on cell type or species. In this study, we have ad- dressed the role of PTPL1 in the model of antiestrogen-induced apoptosis in hormone-responsive human breast cancer cells positive or defective for this particular tyrosine phosphatase. We have shown that PTPL1 is necessary for the early inhibi- tion of IRS-1/PI3-K/Akt pathway, which leads to increased apoptosis and inhibition of IGF-I survival effect. Therefore these results first demonstrate that PTPL1 has a pro-apoptotic effect in human breast cancer cells. PTPL1 Is a Pro-apoptotic Enzyme Acting on the PI3-kinase/ Akt Pathway—Several studies have shown that antiestrogen inhibits growth factor effects on proliferation (6, 7) as well as on pos- PL1 ated e a were (IP mu- . B, nsi- was FIG. 7. Time-course inhibition of PI3-K activation in PTPL1- positive or -defective cells. Mock-transfected clone (P4) and PTPL1 antisense transfected clone (B3) cultured in 1% DCC/FCS were treated for 1–4 days (d) with 50 nM OH-Tam or vehicle alone (ethanol) before 10 min of IGF-I (20 nM) stimulation. PI3-K activity associated to IRS-1 was assayed as in Fig. 4. A, a representative autoradiography of phosphati- dylinositol 3-phosphate separated by thin-layer chromatography. (C, control.) B, quantitative analysis of PI3-K activity in cells pretreated with OH-Tam before IGF-I stimulation. 2 G. Bompard, C. Puech, C. Pre´bois, F. Vignon, and G. Freiss, unpub- lished observations. DISCUSSION We showed that there was no blockade of the IRS-1/PI3-K/Akt pathway in PTPL1-defective clones, indicating the predomi- nant role of PTPL1 in this inhibition. It has previously been shown by other groups (39, 40) that OH-Tam does not directly affect the expression of IRS-1/2 and IGF-IR, although it inhib- its estrogen-induced expression of IRS-1 and IGF-IR. These results, associated with the present data on IRS-1 phosphoryl- ation, point at an action of the tyrosine phosphatase at the level of some sites of the major tyrosine-phosphorylated proteins implicated in this signal transduction pathway. To identify the target(s) of PTPL1/FAP-1 in this transduction pathway we are now developing a panel of PTPL1 mutants able to specifically trap its endogenous substrates (41). effect (7-day treatment) or no effect at all (5-day treatment) on IGF-I survival effect in cells defective in PTPL1. This is clear cut evidence for the role of PTPL1 in the anti-survival and pro-apoptotic effects of OH-Tam. To further understand PTPL1 implication and mechanism of action in apoptosis, we have tested how the absence of PTPL1 expression affected the ability of OH-Tam to inhibit IGF-I-induced PI3-K and Akt activities. We showed that there was no blockade of the IRS-1/PI3-K/Akt pathway in PTPL1-defective clones, indicating the predomi- nant role of PTPL1 in this inhibition. It has previously been shown by other groups (39, 40) that OH-Tam does not directly affect the expression of IRS-1/2 and IGF-IR, although it inhib- its estrogen-induced expression of IRS-1 and IGF-IR. These results, associated with the present data on IRS-1 phosphoryl- ation, point at an action of the tyrosine phosphatase at the level of some sites of the major tyrosine-phosphorylated proteins implicated in this signal transduction pathway. To identify the target(s) of PTPL1/FAP-1 in this transduction pathway we are now developing a panel of PTPL1 mutants able to specifically trap its endogenous substrates (41). also inhibits the survival effect of IGF-I. In many cell lines, the importance of the IRS-1/PI3-K/Akt pathway in survival has been demonstrated (25), but alternative pathways implying mitogen-activated protein kinase have been evidenced (38). In the 32D cell line devoid of IRS-1 two alternative pathways were demonstrated; one involves mitogen-activated protein kinase activation via Shc phosphorylation, and a second one, inde- pendent of mitogen-activated protein kinase and PI3-K, im- plies a mitochondrial translocation of Raf (26). DISCUSSION To further understand PTPL1 implication and mechanism of action in apoptosis, we have tested how the absence of PTPL1 expression affected the ability of OH-Tam to inhibit IGF-I-induced PI3-K and Akt activities. We showed that there was no blockade of the IRS-1/PI3-K/Akt th i PTPL1 d f ti l i di ti th d i ospho- ession. -trans- isense- ultured 4 days ne (eth- M) stim- yzed by specific panel). e con- h anti- he sig- quan- nd the sample . PTPL1 and Apoptosis in Breast Cancer 47867 PTPL1 and Apoptosis in Breast Cancer 47867 FIG. 8. Inhibition of Akt phospho- rylation requires PTPL1 expression. Wild-type MCF7 (MCF7-wt), mock-trans- fected clone (P4), and PTPL1 antisense- transfected clones (B3 and B10) cultured in 1% DCC/FCS were treated for 4 days with 50 nM OH-Tam or vehicle alone (eth- anol) before 10 min of IGF-I (20 nM) stim- ulation. A, cell lysates were analyzed by immunoblotting with an antibody specific for phospho-Akt (Akt-P; bottom panel). Equivalent amounts of Akt were con- firmed by reprobing the blots with anti- Akt antibody (Akt; top panel). B, the sig- nals from the experiments were quan- tified by scanning densitometry, and the level of phosphorylation in each sample was normalized to the level of Akt. PTPL1 and Apoptosis in Breast Cancer 47867 FIG. 8. Inhibition of Akt phospho- rylation requires PTPL1 expression. Wild-type MCF7 (MCF7-wt), mock-trans- fected clone (P4), and PTPL1 antisense- transfected clones (B3 and B10) cultured in 1% DCC/FCS were treated for 4 days with 50 nM OH-Tam or vehicle alone (eth- anol) before 10 min of IGF-I (20 nM) stim- ulation. A, cell lysates were analyzed by immunoblotting with an antibody specific for phospho-Akt (Akt-P; bottom panel). Equivalent amounts of Akt were con- firmed by reprobing the blots with anti- Akt antibody (Akt; top panel). B, the sig- nals from the experiments were quan- tified by scanning densitometry, and the level of phosphorylation in each sample was normalized to the level of Akt. effect (7-day treatment) or no effect at all (5-day treatment) on IGF-I survival effect in cells defective in PTPL1. This is clear cut evidence for the role of PTPL1 in the anti-survival and pro-apoptotic effects of OH-Tam. To further understand PTPL1 implication and mechanism of action in apoptosis, we have tested how the absence of PTPL1 expression affected the ability of OH-Tam to inhibit IGF-I-induced PI3-K and Akt activities. DISCUSSION In MCF7 cells, the use of specific inhibitors showed the predominant role of PI3-K pathway for the protective action of IGF-I. Whereas PD98059 was without effect, the results obtained with OH- Tam and LY294002 were analogous, suggesting that the an- tiestrogen can act as an inhibitor of the PI3-K pathway. Studies from our laboratory show that some of the IGF-I effects such as immediately early gene regulation or stimulation of glyceral- dehyde-3-phosphate dehydrogenase expression were not inhib- ited by short term (8) or long term tamoxifen treatment,2 indi- cating that some but not all the transducing pathways of IGF-I receptor were discontinued after antiestrogen treatment. We have now shown that pre-treatment with OH-Tam induced a reduction of IRS-1 phosphorylation and PI3-K activation as well as an inhibition of IGF-I-induced Akt activation. Our results clearly demonstrate that OH-Tam inhibits IGF-I sur- vival effect through a blockade of the IRS-1/PI3-K/Akt pathway. PTPL1/FAP-1 Apoptotic Effect Is Independent of the Fas Pathway—This study is one of the first evidences of a pro- apoptotic effect of a tyrosine phosphatase counteracting the adverse protective effect of a tyrosine kinase in breast cancer. Effectively, the role of these enzymes in cancer is poorly un- derstood. Keane et al. (18) demonstrate by inducible transfec- tion that DEP-1 had an antiproliferative effect in MCF7 cells, and Ardini et al. (42) show a positive correlation between high expression of PTP and delayed tumor growth and metastasis. Only SHP-1 was clearly implicated in the apoptotic process in breast cancer; the membrane recruitment of this enzyme by the somatostatin receptor or Fas has been shown to be an early event in their signaling that induced intracellular acidification- dependent apoptosis (17). PTPL1/FAP-1 Apoptotic Effect Is Independent of the Fas Pathway—This study is one of the first evidences of a pro- apoptotic effect of a tyrosine phosphatase counteracting the adverse protective effect of a tyrosine kinase in breast cancer. Effectively, the role of these enzymes in cancer is poorly un- derstood. Keane et al. (18) demonstrate by inducible transfec- tion that DEP-1 had an antiproliferative effect in MCF7 cells, and Ardini et al. (42) show a positive correlation between high expression of PTP and delayed tumor growth and metastasis. DISCUSSION Regulation of apoptosis is a fundamental feature in the con- trol of normal development but also in the progression of can- y cell lines, the survival has ways implying denced (38). In pathways were protein kinase nd one, inde- nd PI3-K, im- n MCF7 cells, minant role of GF-I. Whereas ned with OH- g that the an- hway. Studies effects such as on of glyceral- were not inhib- atment,2 indi- hways of IGF-I reatment. We am induced a effect (7-day treatment) or no effect at all (5-day treatment) on IGF-I survival effect in cells defective in PTPL1. This is clear cut evidence for the role of PTPL1 in the anti-survival and pro-apoptotic effects of OH-Tam. To further understand PTPL1 implication and mechanism of action in apoptosis, we have tested how the absence of PTPL1 expression affected the ability of OH-Tam to inhibit IGF-I-induced PI3-K and Akt activities. We showed that there was no blockade of the IRS-1/PI3-K/Akt pathway in PTPL1-defective clones, indicating the predomi- nant role of PTPL1 in this inhibition. It has previously been shown by other groups (39, 40) that OH-Tam does not directly affect the expression of IRS-1/2 and IGF-IR, although it inhib- its estrogen-induced expression of IRS-1 and IGF-IR. These results, associated with the present data on IRS-1 phosphoryl- ation, point at an action of the tyrosine phosphatase at the level of some sites of the major tyrosine-phosphorylated proteins implicated in this signal transduction pathway. To identify the target(s) of PTPL1/FAP-1 in this transduction pathway we are now developing a panel of PTPL1 mutants able to specifically trap its endogenous substrates (41). PTPL1/FAP 1 A i Eff I I d d f h F 1 and Apoptosis in Breast Cancer 47867 of IGF-I. In many cell lines, the K/Akt pathway in survival has alternative pathways implying ase have been evidenced (38). In 1 two alternative pathways were itogen-activated protein kinase lation, and a second one, inde- protein kinase and PI3-K, im- ti f R f (26) I MCF7 ll effect (7-day treatment) or no effect at all (5-day treatment) on IGF-I survival effect in cells defective in PTPL1. This is clear cut evidence for the role of PTPL1 in the anti-survival and pro-apoptotic effects of OH-Tam. DISCUSSION Only SHP-1 was clearly implicated in the apoptotic process in breast cancer; the membrane recruitment of this enzyme by the somatostatin receptor or Fas has been shown to be an early event in their signaling that induced intracellular acidification- dependent apoptosis (17). We have previously shown that the up-regulation of PTPL1 is crucial for the overall effect of tamoxifen on growth factor activity (16). Here, using stable antisense PTPL1 clones, we have evaluated the implication of this enzyme in the apoptotic process. We now demonstrate that OH-Tam had a reduced PTPL1/FAP 1 d t i h b thi t i tid lt d i th i d ti f F di t d S-1 phospho- ve or -defec- ed clone (P4) sfected clone 1% DCC/FCS h 50 nM OH- anol) before a mulation. A, were analyzed antibody spe- Tyr; top pan- IRS-1 immu- ed by immu- ibody specific anel). B, the ts were quan- etry, and the each sample s of IRS-1. FIG. 10. PTPL1/FAP-1 expression is not responsible for Fas resistance in breast cancer cells. A, wild-type MCF7 cells, B3 antisense clone, and T47D cells cultured in 1% DCC/FCS were treated for 18 h with 1 g/ml cycloheximide in the absence of antibody (control) or with 100 ng/ml anti-Fas-activating antibody (DX2;1 R&D Systems). Apoptosis was quantified by ELISA of cytoplasmic histone-associ- ated DNA fragments. Results were ex- pressed as a percentage of apoptosis quan- tified in control cells. Results are the means S.D. of three experiments. B, ex- pression of PTPL1/FAP-1 was assayed by reverse transcription-PCR as in Freiss et al. (16) on total RNA from MCF7 and T47D cells cultured in 1% DCC/FCS. Results are the means S.D. of three experiments us- ing, respectively, glyceraldehyde-3-phos- phate dehydrogenase, hypoxanthine-gua- nine phosphoribosyltransferase, and 2- macroglobulin as the reference genes. PTPL1 and Apoptosis in Breast Cancer 47868 PTPL1 and Apoptosis in Breast Cancer The relation between PTPL1/FAP-1 and apoptosis has been addressed by several groups and remains controversial. The first evidence of such a relation was brought up by Sato et al. (15) with the demonstration of the interaction of FAP-1 with the pro-apoptotic receptor Fas. In this initial study and a recent one from the same group (43) it was shown that in transfectant clones of Jurkat or TMK cells overexpressing FAP-1 the apo- ptotic effect of a Fas agonist antibody was inhibited by 50%. DISCUSSION The three C-terminal amino acids (SLV) of human Fas were necessary and sufficient for its interaction with the second PDZ this tripeptide resulted in the induction of Fas-mediated apo- ptosis in a colon cancer cell line expressing Fas and FAP-1. This observation indicated that an interaction between Fas C-ter- minal domain and the PDZ domain of a protein modulates Fas signal transduction inhibition, and the authors propose that FAP-1 could be this inhibitory protein (45). On the other hand, Cuppen et al. (46) show that there was no interaction between murine Fas, which lacked the SLV terminal sequence, and PTP-BL (mouse homologue of PTPL1/FAP-1). These data strongly suggest that this interaction, which is not conserved FIG. 9. Inhibition of IRS-1 phospho- rylation in PTPL1-positive or -defec- tive cells. Mock-transfected clone (P4) and PTPL1 antisense-transfected clone (B3 and B10) cultured in 1% DCC/FCS were treated for 4 days with 50 nM OH- Tam or vehicle alone (ethanol) before a 10-min IGF-I (20 nM) stimulation. A, IRS-1 immunoprecipitates were analyzed by immunoblotting with an antibody spe- cific for phosphotyrosine (P-Tyr; top pan- el). Equivalent amounts of IRS-1 immu- noprecipitates were analyzed by immu- noblotting (WB) with an antibody specific for IRS-1 (IRS-1; bottom panel). B, the signals from the experiments were quan- tified by scanning densitometry, and the level of phosphorylation in each sample was normalized to the levels of IRS-1. FIG. 10. PTPL1/FAP-1 expression is not responsible for Fas resistance in breast cancer cells. A, wild-type MCF7 cells, B3 antisense clone, and T47D cells cultured in 1% DCC/FCS were treated for 18 h with 1 g/ml cycloheximide in the absence of antibody (control) or with 100 ng/ml anti-Fas-activating antibody (DX2;1 R&D Systems). Apoptosis was quantified by ELISA of cytoplasmic histone-associ- ated DNA fragments. Results were ex- pressed as a percentage of apoptosis quan- tified in control cells. Results are the means S.D. of three experiments. B, ex- pression of PTPL1/FAP-1 was assayed by reverse transcription-PCR as in Freiss et al. (16) on total RNA from MCF7 and T47D cells cultured in 1% DCC/FCS. Results are the means S.D. of three experiments us- ing, respectively, glyceraldehyde-3-phos- phate dehydrogenase, hypoxanthine-gua- nine phosphoribosyltransferase, and 2- macroglobulin as the reference genes. p p FIG. 9. Inhibition of IRS-1 phospho- rylation in PTPL1-positive or -defec- tive cells. DISCUSSION Mock-transfected clone (P4) and PTPL1 antisense-transfected clone (B3 and B10) cultured in 1% DCC/FCS were treated for 4 days with 50 nM OH- Tam or vehicle alone (ethanol) before a 10-min IGF-I (20 nM) stimulation. A, IRS-1 immunoprecipitates were analyzed by immunoblotting with an antibody spe- cific for phosphotyrosine (P-Tyr; top pan- el). Equivalent amounts of IRS-1 immu- noprecipitates were analyzed by immu- noblotting (WB) with an antibody specific for IRS-1 (IRS-1; bottom panel). B, the signals from the experiments were quan- tified by scanning densitometry, and the level of phosphorylation in each sample was normalized to the levels of IRS-1. FIG. 10. PTPL1/FAP-1 expression is not responsible for Fas resistance in breast cancer cells. A, wild-type MCF7 cells, B3 antisense clone, and T47D cells cultured in 1% DCC/FCS were treated for 18 h with 1 g/ml cycloheximide in the absence of antibody (control) or with 100 ng/ml anti-Fas-activating antibody (DX2;1 R&D Systems). Apoptosis was quantified by ELISA of cytoplasmic histone-associ- ated DNA fragments. Results were ex- pressed as a percentage of apoptosis quan- tified in control cells. Results are the means S.D. of three experiments. B, ex- pression of PTPL1/FAP-1 was assayed by reverse transcription-PCR as in Freiss et al. (16) on total RNA from MCF7 and T47D cells cultured in 1% DCC/FCS. Results are the means S.D. of three experiments us- ing, respectively, glyceraldehyde-3-phos- phate dehydrogenase, hypoxanthine-gua- nine phosphoribosyltransferase, and 2- macroglobulin as the reference genes. FIG. 10. PTPL1/FAP-1 expression is not responsible for Fas resistance in breast cancer cells. A, wild-type MCF7 cells, B3 antisense clone, and T47D cells cultured in 1% DCC/FCS were treated for 18 h with 1 g/ml cycloheximide in the absence of antibody (control) or with 100 ng/ml anti-Fas-activating antibody (DX2;1 R&D Systems). Apoptosis was quantified by ELISA of cytoplasmic histone-associ- ated DNA fragments. Results were ex- pressed as a percentage of apoptosis quan- tified in control cells. Results are the means S.D. of three experiments. B, ex- pression of PTPL1/FAP-1 was assayed by reverse transcription-PCR as in Freiss et al. (16) on total RNA from MCF7 and T47D cells cultured in 1% DCC/FCS. Results are the means S.D. of three experiments us- ing, respectively, glyceraldehyde-3-phos- phate dehydrogenase, hypoxanthine-gua- nine phosphoribosyltransferase, and 2- macroglobulin as the reference genes. The relation between PTPL1/FAP-1 and apoptosis has been addressed by several groups and remains controversial. PTPL1 and Apoptosis in Breast Cancer 47869 A., A’Hern, R., Salter, J., Detre, S., Nicholson, R., Robertson, J., Smith, I. E., A., A’Hern, R., Salter, J., Detre, S., Nicholson, R., Robertson, J., Smith, I. E., and Dowsett, M. (1997) Int. J. Cancer 72, 608–613 function. In the same study, Cuppen et al. (46) could not gen- erate an inhibition of Fas-induced apoptosis by overexpressing PTP-BL in mouse T cell lymphoma transfected with human Fas, which interacted with PTP-BL, indicating that this inter- action is not sufficient for inhibiting Fas transduction in this cell type. A., A Hern, R., Salter, J., Detre, S., Nicholson, R., Robertson, J., Smith, I. E., and Dowsett, M. (1997) Int. J. Cancer 72, 608–613 and Dowsett, M. (1997) Int. J. Cancer 72, 608–613 6. Vignon, F., Bouton, M. M., and Rochefort, H. (1987) Biochem. Biophys. Res. Commun. 146, 1502–1508 7. Ignar-Trowbridge, D. M., Nelson, K. G., Bidwell, M. C., Curtis, S. W., Washburn, T. F., McLachlan, J. A., and Korach, K. S. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 4658–4662 8. Philips, A., Chalbos, D., and Rochefort, H. (1993) J. Biol. Chem. 268, 14103–14108 yp Several laboratories have looked for a correlation between PTPL1/FAP-1 expression and resistance to Fas-induced apo- ptosis in different cell lines or tissues. Some studies have provided evidence for a high level of FAP-1 mRNA expression in Kaposi’s sarcoma (31) and, in pancreatic adenocarcinomas (33), a higher expression in T helper cells type 1 (which are resistant to apoptosis) than in T helper cells type 2 (which are sensitive to Fas ligand) (32). Other groups failed to show any correlation between Fas resistance and FAP-1 expression in colon cancer (34), in subclones of S49.1 murine T lymphocytes (35), in adult T cell leukemia cell lines (36), and in testicular germ cells where exposure to mono-(2-ethylhexyl) phthalate induces massive cell death and increases FAP-1 expression (47). We have presently evaluated the sensitivity to Fas-medi- ated apoptosis of three breast cancer cell lines expressing var- ious levels of PTPL1/FAP-1 mRNA; they are T47D cells, which expressed high levels of enzyme, wild-type MCF7 cells, ex- pressing moderate levels of PTPL1, and the B3 transfectant MCF7 subclone, in which enzyme expression was switched off. T47D cells were sensitive, whereas MCF7 cells were resistant to Fas-mediated apoptosis as previously described (48), indi- cating the absence of correlation between PTPL1/FAP-1 ex- pression and Fas sensitivity. PTPL1 and Apoptosis in Breast Cancer Kyprianou, N., English, H. F., Davidson, N. E., and Isa Res. 51, 162–166 21. Huovinen, R., Warri, A., and Collan, Y. (1993) Int. J. Can 22. Gooch, J. L., and Yee, D. (1999) Cancer Lett. 144, 31–37 23. Gooch, J. L., Van Den Berg, C. L., and Yee, D. (1999) Breast Cancer Res. Treat. 56, 1–10 24. Baserga, R., Prisco, M. and Hongo, A. (1999) in The IGF System (Rosenfeld, R. G., and Roberts, C. T., Jr., eds) pp. 329–353, Humana Press Inc., Totowa, NJ 5. Kulik, G., Klippel, A., and Weber, M. J. (1997) Mol. Cell. Biol. 25. Kulik, G., Klippel, A., and Weber, M. J. (1997) Mol. Cell. Biol. 17, 1595–1606 26. Peruzzi, F., Prisco, M., Dews, M., Salomoni, P., Grassilli, E., Romano, G., Calabretta, B., and Baserga, R. (1999) Mol. Cell. Biol. 19, 7203–7215 26. Peruzzi, F., Prisco, M., Dews, M., Salomoni, P., Grassilli, E., Romano, G., Calabretta, B., and Baserga, R. (1999) Mol. Cell. Biol. 19, 7203–7215 g 27. Navarro, M., and Baserga, R. (2001) Endocrinology 142, 1073–1081 28. Myers, M. G., Jr., Grammer, T. C., Wang, L. M., Sun, X. J., Pierce, J. Blenis, J., and White, M. F. (1994) J. Biol. Chem. 269, 28783–28789 29. Cardone, M. H., Roy, N., Stennicke, H. R., Salvesen, G. S., Franke, T. F., Stanbridge, E., Frisch, S., and Reed, J. C. (1998) Science 282, 1318–1321 30. Datta, S. R., Dudek, H., Tao, X., Masters, S., Fu, H., Gotoh, Y., and Greenberg, M. E. (1997) Cell 91, 231–241 31. Mori, S., Murakami-Mori, K., Jewett, A., Nakamura, S., and Bonavida, B. (1996) Cancer Res. 56, 1874–1879 32. Zhang, X., Brunner, T., Carter, L., Dutton, R. W., Rogers, P., Bradley, L., Sato, T., Reed, J. C., Green, D., and Swain, S. L. (1997) J. Exp. Med. 185, 1837–1849 33. Ungefroren, H., Voss, M., Jansen, M., Roeder, C., Henne-Bruns, D., Kremer, B., and Kalthoff, H. (1998) Cancer Res. 58, 1741–1749 Conclusions—We have presently demonstrated that the pro- tein-tyrosine phosphatase PTPL1/FAP-1, whose expression is stimulated by antiestrogens in hormone-sensitive breast can- cer cells, plays a major role in the inhibition of IGF-I survival pathway in these tumor cells by interfering on the IRS-1/PI3- K/Akt pathway. The pro-apoptotic effect of PTPL1 in breast cancer could suggest new therapeutic routes and points at the necessity of further studies on the mechanisms by which such enzymes are regulated by yet unknown endogenous factors in breast tumors. 2. Freiss, G., Prebois, C., and Vignon, F. (1993) Breast Cancer Res. Treat. 27, 57–68 1. Peto, R., Boreham, J., Clarke, M., Davies, C., and Beral, V. (2000) Lancet 355, 1822 4. Bardon, S., Vignon, F., Montcourrier, P., and Rochefort, H. (1987) Cancer Res. 47, 1441–1448 5. Ellis, P. A., Saccani-Jotti, G., Clarke, R., Johnston, S. R., Anderson, E., Howell, 3. Dickson, R. B., and Lippman, M. E. (1995) Endocr. Rev. 16, 559–589 PTPL1 and Apoptosis in Breast Cancer 34. Tillman, D. M., Harwood, F. G., Gibson, A. A., and Houghton, J. A. (1998) Cell Death. Differ. 5, 450–457 35. Broome, H. E., Dargan, C. M., Brunner, T., and Green, D. R. (1998) Cell Death. Differ. 5, 200–205 ff , 36. Yamada, Y., Sugahara, K., Tsuruda, K., Nohda, K., Hata, T., Maeda, T., H d M T M H hib T J h T T M Mi ki 36. Yamada, Y., Sugahara, K., Tsuruda, K., Nohda, K., Hata, T., Maeda, T., Honda, M., Tawara, M., Hayashibara, T., Joh, T., Tomonaga, M., Miyazaki, Y., and Kamihira, S. (1999) Cancer Lett. 147, 215–219 kova, M. A., and Surmacz, E. (1997) Cancer Res. 57, 2606–2610 37. Guvakova, M. A., and Surmacz, E. (1997) Cancer Res. 57, 2 38. Parrizas, M., Saltiel, A. R., and LeRoith, D. (1997) J. Biol. Chem. 272, 154–161 39. Lee, A. V., Jackson, J. G., Gooch, J. L., Hilsenbeck, S. G., Coronado-Heinsohn, E., Osborne, C. K., and Yee, D. (1999) Mol. Endocrinol. 13, 787–796 40. Molloy, C. A., May, F. E., and Westley, B. R. (2000) J. Biol. Chem. 275, 12565–12571 41. Flint, A. J., Tiganis, T., Barford, D., and Tonks, N. K. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 1680–1685 Acknowledgments—We thank Dr. E. Salin-Drouin (Besins-Iscovesco, Paris, France) for providing OH-Tam and acknowledge the participa- tion of M. Errabih and N. Wildbaum in some experiments during their graduate courses. We are grateful to J. Y. Cance for final preparation of the drawings and figures. 42. Ardini, E., Agresti, R., Tagliabue, E., Greco, M., Aiello, P., Yang, L. T., Menard, S., and Sap, J. (2000) Oncogene 19, 4979–4987 43. Li, Y., Kanki, H., Hachiya, T., Ohyama, T., Irie, S., Tang, G., Mukai, J., and Sato, T. (2000) Int. J. Cancer 87, 473–479 43. Li, Y., Kanki, H., Hachiya, T., Ohyama, T., Irie, Sato, T. (2000) Int. J. Cancer 87, 473–479 Sato, T. (2000) Int. J. Cancer 87, 473–479 44. Saras, J., Engstrom, U., Gonez, L. J., and Heldin, C. H. (1997) J. Biol. Chem. 272, 20979–20981 DISCUSSION The first evidence of such a relation was brought up by Sato et al. (15) with the demonstration of the interaction of FAP-1 with the pro-apoptotic receptor Fas. In this initial study and a recent one from the same group (43) it was shown that in transfectant clones of Jurkat or TMK cells overexpressing FAP-1 the apo- ptotic effect of a Fas agonist antibody was inhibited by 50%. The three C-terminal amino acids (SLV) of human Fas were necessary and sufficient for its interaction with the second PDZ domain of PTPL1 (44). The direct cytoplasmic microinjection of this tripeptide resulted in the induction of Fas-mediated apo- ptosis in a colon cancer cell line expressing Fas and FAP-1. This observation indicated that an interaction between Fas C-ter- minal domain and the PDZ domain of a protein modulates Fas signal transduction inhibition, and the authors propose that FAP-1 could be this inhibitory protein (45). On the other hand, Cuppen et al. (46) show that there was no interaction between murine Fas, which lacked the SLV terminal sequence, and PTP-BL (mouse homologue of PTPL1/FAP-1). These data strongly suggest that this interaction, which is not conserved along the evolution, does not play a crucial role in FAP-1 PTPL1 and Apoptosis in Breast Cancer PTPL1 and Apoptosis in Breast Cancer Moreover, B3 and MCF7 cells were equally resistant, indicating that in MCF7 cells, PTPL1 expression is not responsible for the resistance to Fas-mediated apoptosis. Altogether, these studies suggested a dual effect of PTPL1 on apoptosis. On one side, in human cells sensitive to Fas-mediated apoptosis, overexpression of PTPL1/FAP-1 could inhibit Fas signaling in a cell- and species-specific manner. On the other hand, in cells not expressing Fas or in which Fas is not efficient as is the case in most breast cancer cell lines, PTPL1/FAP-1 could have a pro-apoptotic effect by inhibition of the IRS-1/PI3-K/Akt signaling pathway. 9. Kato, S., Endoh, H., Masuhiro, Y., Kitamoto, T., Uchiyama, S., Sasaki, H., Masushige S Gotoh Y Nishida E Kawashima H Metzger D and 9. Kato, S., Endoh, H., Masuhiro, Y., Kitamoto, T., Uchiyama, S., Sasaki, H., Masushige, S., Gotoh, Y., Nishida, E., Kawashima, H., Metzger, D., and Chambon, P. (1995) Science 270, 1491–1494 10. Migliaccio, A., Di Domenico, M., Castoria, G., de Falco, A., Bontempo, P., Nola, E d A i hi F (1996) EMBO J 15 1292 1300 10. Migliaccio, A., Di Domenico, M., Castoria, G., de Falco, A., Bontempo, P., Nola, E., and Auricchio, F. (1996) EMBO J. 15, 1292–1300 11. Freiss, G., and Vignon, F. (1994) Mol. Endocrinol. 8, 1389–1396 12. Banville, D., Ahmad, S., Stocco, R., and Shen, S. H. (1994) J. Biol. Chem. 269, 22320–22327 13. Maekawa, K., Imagawa, N., Nagamatsu, M., and Harada, S. (1994) FEBS Lett. 337, 200–206 14. Saras, J., Claesson-Welsh, L., Heldin, C. H., and Gonez, L. J. (1994) J. Biol. Chem. 269, 24082–24089 15. Sato, T., Irie, S., Kitada, S., and Reed, J. C. (1995) Science 268, 411–415 16. Freiss, G., Puech, C., and Vignon, F. (1998) Mol. Endocrinol. 12, 568–579 17. Thangaraju, M., Sharma, K., Leber, B., Andrews, D. W., S ik C B ( ) J B l Ch 17. Thangaraju, M., Sharma, K., Leber, B., Andrews, D. W., Shen, S. H., and Srikant, C. B. (1999) J. Biol. Chem. 274, 29549–29557 g j , , , , , , , , Srikant, C. B. (1999) J. Biol. Chem. 274, 29549–29557 18. Keane, M. M., Ettenberg, S. A., Lowrey, G. A., Russell, E. K., and Lipkowitz, S. (1996) Cancer Res. 56, 4791–4798 19. Bradford, M. M. (1976) Anal. Biochem. 72, 248–254 20. Kyprianou, N., English, H. F., Davidson, N. E., and Isaacs, J. T. (1991) Cancer Res. 51, 162–166 20. REFERENCES 45. Yanagisawa, J., Takahashi, M., Kanki, H., Yano-Yanagisawa, H., Tazunoki, T., Sawa, E., Nishitoba, T., Kamishohara, M., Kobayashi, E., Kataoka, S., and Sato, T. (1997) J. Biol. Chem. 272, 8539–8545 1. Peto, R., Boreham, J., Clarke, M., Davies, C., and Beral, V. (2000) Lancet 355, 1822 46. Cuppen, E., Nagata, S., Wieringa, B., and Hendriks, W. (1997) J. Biol. Chem. 272, 30215–30220 2. Freiss, G., Prebois, C., and Vignon, F. (1993) Breast Cancer Res. Treat. 27, 57–68 47. Boekelheide, K., Lee, J., Shipp, E. B., Richburg, J. H., and Li, G. (1998) Toxicol. Lett. 102–103, 503–508 3. Dickson, R. B., and Lippman, M. E. (1995) Endocr. Rev. 16, 559–589 4. Bardon, S., Vignon, F., Montcourrier, P., and Rochefort, H. (1987) Cancer Res. 47, 1441–1448 48. Keane, M. M., Lowrey, G. A., Ettenberg, S. A., Dayton, M. A., and Lipkowitz, S. (1996) Cancer Res. 56, 4236–4243
W4310675600.txt	https://eprints.bournemouth.ac.uk/37920/1/ijerph-19-16150-v4.pdf	en		Open Questions on the Electromagnetic Field Contribution to the Risk of Neurodegenerative Diseases	International journal of environmental research and public health/International journal of environmental research and public health	2,022	cc-by	9,493	International Journal of Environmental Research and Public Health Review Open Questions on the Electromagnetic Field Contribution to the Risk of Neurodegenerative Diseases Joanna Wyszkowska 1, * and Colin Pritchard 2 1 2 * Department of Animal Physiology and Neurobiology, Faculty of Biological and Veterinary Sciences, Nicolaus Copernicus University in Toruń, Lwowska 1, 87-100 Toruń, Poland Faculty of Health and Social Sciences, Bournemouth University, Bournemouth Gateway, 106 St. Pauls Rd, Bournemouth BH88AJ, UK Correspondence: joanna.wyszkowska@umk.pl Abstract: This work presents the current state of knowledge about the possible contributory influence of the electromagnetic field on the occurrence of neurodegenerative diseases such as Alzheimer’s and Parkinson’s disease, amyotrophic lateral sclerosis, and multiple sclerosis. Up-to-date literature indicates both favourable and adverse effects of electromagnetic exposure on human health, making it difficult to come to valid and unambiguous conclusions. The epidemiological data analysis from the World Health Organization statistics shows a substantial rise in neurological mortality compared with rises in total populations in developed countries over a mere 15-year period. The largest of the analysed countries produced odds ratios of >100%. The contribution of electromagnetic exposure to the incidence of neurodegenerative diseases is still undoubtedly open to discussion, and it requires further in-depth research to assess the action mechanism of electromagnetic fields in neurodegenerative diseases. The limitations of research published hitherto and the problem of drawing unequivocal conclusions are also in focus. Citation: Wyszkowska, J.; Pritchard, Keywords: EMF; neurological mortality; early onset dementias; nervous system; environmental pollutants; occupational exposure; Alzheimer’s disease; oxidative stress C. Open Questions on the Electromagnetic Field Contribution to the Risk of Neurodegenerative Diseases. Int. J. Environ. Res. Public Health 2022, 19, 16150. https:// doi.org/10.3390/ijerph192316150 Academic Editor: Paul B. Tchounwou Received: 13 October 2022 Accepted: 29 November 2022 Published: 2 December 2022 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Copyright: © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). 1. Introduction The increasing number of man-made sources of electromagnetic field (EMF) raises interest in occupational groups about its impact on human health, especially concerning the high level of exposure. While there are some beneficial and therapeutic applications of EMF, there are more and more publications devoted to the unfavourable effects of EMF exposure on humans, mostly pointing to the deterioration of their well-being, disruptions to the functions of the nervous system, or linking it to the occurrence of cancer [1–8]. Many functions of the human body are controlled by electric potentials and currents: the transmission of electric signals in the neuromuscular system, the blood flow associated with the movement of charged particles, and membrane transport phenomena all depend on electric charges and potentials [9]. Thus, EMF has many potential targets for action. However, to fully understand the characteristics of EMF influence, it is necessary to know the exact underlying mechanisms. Various research is being conducted to determine the effectiveness and safety of the application of EMF in medical treatment. It is crucial to know how powerful EMF can be without disturbing homeostasis, whether compensatory mechanisms appear and whether the effect of EMF is cumulative [10]. It is also necessary to pay attention to the International Agency for Research on Cancer (IARC) 2B classification for possibly human carcinogenic outcomes from chronic EMF exposure [11]. The classification was based on studies demonstrating an association between two types of brain tumours, glioma, and acoustic neuroma, with exposure to radiofrequency EMF from wireless phones. Although the report found that evidence from the occupational and environmental radiofrequency Int. J. Environ. Res. Public Health 2022, 19, 16150. https://doi.org/10.3390/ijerph192316150 https://www.mdpi.com/journal/ijerph Int. J. Environ. Res. Public Health 2022, 19, 16150 2 of 15 EMF exposure was inadequate, the conclusion is that there could be some risk. Therefore, it is necessary to keep a close watch for a link between cell phones and cancer risk, especially in the category of heavy users. In the case of power-frequency EMF, only the magnetic component was classified as possibly carcinogenic to humans. In the report, studies conducted on occupational exposure pointed to a possible increased risk of leukaemia, brain tumours, and male breast cancer. However, their interpretation was difficult mainly due to methodological limitations and a lack of appropriate exposure measurements [12]. EMF treatment should be conducted under well-controlled conditions, supported by evidence of biological activity for EMF, which may lead to a positive or negative outcome depending on the exposure parameters. This is to ensure the safety of therapeutic treatments involving EMF, even more so when a direct action of an EMF occurring at one body location may have an indirect effect in another location [13–16]. According to published results, the most beneficial therapeutic cycle should include 10 to 14 daily treatments. In selected cases, it is advisable to repeat the cycle after four weeks [10]. Reflecting on human biology, we need to bear in mind that the brain is essentially an electro-biochemical organ, so potentially, as EMF passes through the human body, it might be expected that the brain and neurological system will most likely be affected by EMF exposure [3,17]. Special attention is needed for cases of chronic exposure, where conclusions from research into the beneficial effects during short and controlled medical EMF treatments may not apply. The biological effect of exposure to EMF generally depends primarily on its frequency. This is due to the fact that different frequencies interact with the body in different ways: low-frequency EMF may cause the stimulation of nerves and muscles by induced electric potentials, while a high-frequency field induces thermal effects that may lead to a rise in body temperature. EMF exposure at a frequency exceeding 10 GHz causes EMF energy absorption mainly near the surface of the body [9,18]. There have been articles indicating a relationship between the higher incidence of neurodegenerative diseases (NDD) and increased exposure to EMF [1,2,19–21]. Researchers are trying to define a mechanism that could explain this correlation by considering the contribution of oxidative stress, which is closely related to the occurrence of neurological diseases and may be developed under the influence of EMF, among other factors [16,22,23]. However, the key point is that it is not possible to simply split the impact of EMF exposure and multiple other interactive environmental pollutants (e.g., increases in background hormone-disruptive chemicals, air pollution, food additives, and petrochemicals from the motor and air transport), which all overlap and contribute to organism response. Each of these factors may play an important role in people’s lives. Epidemiologic evidence appears to suggest that workers, especially in electrical occupations, may be at an increased risk caused not only by higher levels of EMF exposure but also by exposure to other factors (e.g., metals, chemicals) that may interact with EMF [24]. It should be taken into account that the rise in incidents of neurological diseases is a reflection of the Gompertzian hypothesis, which states that people’s life span is becoming longer, so they are developing more age-related diseases [25]. In other words, it is suggested that the apparent increases in the incidence of NDD are mainly due to demographics, i.e., there are more neurological diseases because there are more elderly people. However, we will present and use the latest epidemiological data, based on World Health Organization (WHO) statistics, updated as of December 2018 [26], which contains evidence that seriously challenges such simplistic reassurances and provides cause for concern. With regard to EMF, there has been a degree of uncertainty as studies have not found any statistically significant negative effects related to EMF exposure [21,27,28], and other research results have even indicated that short-term exposure to low-level EMF can help improve memory in Alzheimer patients [29], and functional and mental status of poststroke patients [30]. This paper is an overview of the results arising from the epidemiological, in vitro, and in vivo studies that investigated whether EMF exposure has an influence on the occurrence of neurodegenerative diseases. A literature search was conducted on the online databases Int. J. Environ. Res. Public Health 2022, 19, 16150 3 of 15 of PubMed and Google Scholar and the official reports of the Scientific Committee on Emerging and Newly Identified Health Risks working for the European Council of the European Union [18] and WHO [31]. The following terms were searched in the online databases individually or in combination: “neurodegenerative disease”, “Alzheimer”, “amyotrophic lateral sclerosis”, “ALS”, “Parkinson’s disease”, “multiple sclerosis”, “MS”, “exposure”, “magnetic field”, “electromagnetic field”. The comprehensiveness of the literature search was verified using reviews and reference lists of other publications. 2. Neurodegenerative Diseases (NDD) Neurodegeneration is the progressive loss of the structure or function of neurons, including their death. There are hundreds of disorders that could be described as neurodegenerative diseases (NDD). They are often associated with deficits in brain function (e.g., memory and cognition, or movement–dependent on the predominant neuronal population impacted). Many of these diseases are rare, but a few are common and include Alzheimer’s disease (AD), Parkinson’s disease (PD), amyotrophic lateral sclerosis (ALS), and multiple sclerosis (MS) (see Table 1 for the summary). They represent one of the gravest health concerns currently affecting developed countries. Specific environmental factors and lifestyle, alone or in combination with genetic susceptibility factors, are considered to play a key role in the pathogenesis of NDD [23,32]. The risk of being affected by an NDD increases dramatically with age. More people living longer means that more individuals are affected by NDD, which is why it is so important to improve our understanding of what causes NDD and to develop new approaches for treatment and prevention [33]. Table 1. Overview of common neurodegenerative diseases. Based on data from the public domains [34,35]. Disease Main Neuropathology Symptoms - Alzheimer’s disease (AD) Beta-amyloid deposits and neurofibrillary tangles in the cerebral cortex and subcortical grey matter - Parkinson’s disease (PD) Loss of neurons that produce dopamine–a chemical messenger in the brain - - Amyotrophic lateral sclerosis (ALS) Multiple sclerosis (MS) Loss of neurons in the motor cortex (upper motor neurons) and motor neurons in the brain stem and central spinal cord (lower motor neurons) Loss of memory, an inability to learn new things, loss of language function, a deranged perception of space, an inability to perform calculations, indifference, depression, delusions, and other manifestations These deficits affect patients’ social functioning and make it difficult or impossible for them to carry on with their daily lives AD is inexorably progressive and fatal within 5 to 10 years Motor symptoms: rigidity, tremor at rest, slowness of voluntary movement, stooped posture, a shuffling, small-step gait, difficulty with balance Non-motor symptoms: expressionless face, soft voice, olfactory loss, mood disturbances, dementia, sleep disorders, and autonomic dysfunction, including constipation, cardiac arrhythmias, and hypotension The most common cause of death is pneumonia - Trouble walking or running, writing, speech problems The majority of patients die, usually from respiratory paralysis, within 2–3 years from the onset of symptoms - Numbness or weakness in one or more limbs, electric-shock sensations that occur with certain neck movements, tremor, lack of coordination or unsteady gait, partial or complete loss of vision, often with pain during eye movement, prolonged double vision, blurry vision Slightly more than two in every five people with MS died from the disease or from complications Inflammatory demyelinating processes in the brain and spinal cord (CNS) - Int. J. Environ. Res. Public Health 2022, 19, 16150 4 of 15 3. Electromagnetic Field and Neurodegenerative Diseases Researchers have been looking for environmental factors responsible for the development of NDD. Several reports indicate that exposure to electric and magnetic fields may be associated with an increased risk of NDD. The focus of attention is occupational exposure with a relatively high level of EMF exposure, which may be associated with a significant duration of exposure. Several studies [20,21,36–38] have addressed this issue. 3.1. Earlier Epidemiological Studies Based on a thorough analysis of death certificates, it was observed that there is a higher fatality ratio from NDD among people professionally exposed to EMF (e.g., electric power line/cable workers, welders, electricians) than in other professional groups [39]. However, the occurrence of AD and ALS was more strongly associated with EMF exposure than PD [40]. In a similar study [41], a higher mortality rate because of AD in men exposed to the magnetic field was stated; in contrast, in another study, ALS deaths had no connection to magnetic field exposure [39]. However, no clear correlation between the results with the actual level of EMF exposure was revealed. Additionally, researchers’ attention was drawn to the death rate of people inhabiting areas adjacent to high-voltage lines. The authors of this article observed an increased mortality rate due to NDD (in particular AD) in residents living near (<50 m) 220–380 kV power lines [42]. A Swedish study seems to reinforce the evidence for a relationship between occupational EMF exposure and AD, however, it showed elevated risks only for a subgroup of manual workers before the age of 75 [43]. Davanipour et al. [44] studied the possible relationship between EMF exposure and severe cognitive dysfunction. The results indicate that working with EMF exposure (10−4 –10−2 mT) may increase the risk of severe cognitive dysfunction. Smoking and older age (75+) may increase the deleterious effect of EMF exposure [44]. The elevated risks of dementia, motor neuron disease, MS, and epilepsy and lower risks of PD in relation to exposure to EMF (10−4 –10−3 mT) were observed in a large cohort of Danish utility sector employees [45]. On the other hand, a study [46] involving an extensive analysis based on a sample of 30,631 people employed in Danish utility companies did not observe the correlation between PD, AD, or any other diseases of the central nervous system and occupational exposure to EMF (10−4 –10−3 mT). Parlett et al. [47], likewise, indicated no increased rate of mortality from motor neuron disease related to people employed in the electronics sector (~3 × 10−6 mT). The conducted cohort study showed only 40 (out of 3,000,000 examined people) deaths from a motor neuron disease during an average of 8.8 years of observation [47]. In the majority of the available data, no association between PD and EMF exposure has been observed [37,39,46]. The reviewed papers indicate a possible relationship between NDD and EMF, though they also emphasise the methodological limitations, and so no consistent results and unambiguous conclusions have been reached. For instance, Ahlbom [19] concluded that there is relatively compelling evidence indicating that electric utility work may be associated with an increased risk of ALS. However, EMF exposure is only one of several possible contributing factors. For AD, the evidence for an association with EMF is relatively flimsy [19]. In 2006, in a meta-analysis including eight studies published between January 2000 and July 2005, the increased risk of AD was confirmed [37]. Zhou et al. [38] conducted a metaanalysis of seventeen epidemiological studies. Although the findings were not consistent, the authors indicated a slight but significant ALS risk increase among those performing jobs related to EMF exposure (50/60 Hz, 3 × 10−4 –10−3 mT) [38]. Gunnarsson and Bodin [48] in their meta-analysis of sixteen studies (1998–2017) showed recently that occupational exposure to EMF (50/60 Hz, 10−4 –10−3 mT) seemed to involve a 10% increase in risk for ALS and AD, though no such indication of risk was found for PD [48,49]. The majority of studies based on death certificate examination indicated no association between EMF exposure and the risk of NDD [41,43]. An elevated risk of AD and ALS was shown in small-scale studies or only for subgroups of 65–75-year-old people or manual work- Int. J. Environ. Res. Public Health 2022, 19, 16150 5 of 15 ers [38,43,44,47]. Studies of motor neuron disease occurrence based on EMF-level exposure assessment showed less evidence of the EMF effect than those relying on job titles alone [18]. Epidemiological studies on neurological diseases in relation to radiofrequency (100 kHz < f ≤ 300 GHz) EMF exposure show no clear effect, though the evidence is limited. Studies focused on an association between mobile phone use and migraine, vertigo, and the risk of PD and MS [50]. The main problems are conflicting results and methodological limitations [18]. Although the publications regarding the association between EMF and NDD are quite numerous, it is important to note that all those analyses are based solely on death certificates and medical documentation, and therefore demonstrate a certain degree of methodological weaknesses (EMF and other environmental factors lack sufficient characterisation). Numerous external factors, such as the severity of work, physical or mental work, and lifestyle can be determinants with regard to the risk of NDD among different professional groups. Moreover, the studied data from death certificates mostly concern people who lived and worked in the 1970s, 1980s, and 1990s (people subject to occupational exposure to EMF were predominantly physical workers). Such data may lead to conclusions that are inadequate and inapplicable these days. Epidemiological studies have the advantage of long-term observations, though they still contain limitations, such as the relatively crude EMF exposure assessment and the coexistence of many other factors determining neurophysiological pathologies. On the other hand, the small studies have a low statistical power due to a small number of events. Additionally, as EMF exposure is ubiquitous, it is difficult to find an unexposed reference group, and instead, a quantitative contrast is chosen by comparing low versus high exposure levels. Other limitations of this type of study include the fact that most review articles are based on publications stored in electronic databases, usually PubMed, and only English-language publications are included there. These limitations of epidemiological studies are insufficient to conclude that EMF exposure increases the risk of NDD [18]. 3.2. Laboratory Experiments Several studies have shown that EMF exposure modifies physiological and biochemical processes leading to immune cell activation [15], resulting in increased reactive oxygen species (ROS) formation, enhanced phagocytic activity, and increased cytokine release [23,51]. It was shown that EMF can cause mild oxidative stress (increase in ROS and changes in antioxidant levels) in many tissues of the body [16,52,53]. The increase in plasma concentration of pro-inflammatory cytokines and an elevation in blood parameters, such as white blood cells, lymphocytes, hemoglobin, and hematocrit levels in rats exposed to EMF (50 Hz, 7 mT, 24 h) were also demonstrated [15,54]. Inflammation in the central nervous system often occurs in the case of AD, PD, or in the case of chronic neurological disorders. In the review [53], the authors indicate that exposure to EMF can cause redox reactions (50 Hz, 0.1–1.0 mT, 7 days) and the induction of oxidative stress (50 Hz, 2 mT, 3 h) in the rodent brain. This increases the level of free radicals after exposure to EMF (50 Hz, 7 mT, 30 min/day, 10 days), which, in turn, leads to oxidative damage to the lipids in the brain of mice and rats. In the experimental model of a rat exposed to 50 Hz EMF (0.1 and 0.5 mT, 7 days), there was a strong toxic effect disturbing the antioxidant effect. It was shown that exposure to 50 Hz frequency EMF (0.1–1.0 mT, 10 days) affects the antioxidant capacity of enzymes in the brain of both young and old rats. However, in older rats, a large decrease in all major anti-oxidative enzymes was observed, thus indicating an age-dependent greater susceptibility to the induction of oxidative stress as a result of exposure to the EMF [53,55]. The age-related differences in the influence of the EMF appear also in the paper of Ivancsits et al. [56], where the authors established an age-related decrease in DNA repair efficiency of EMF (50 Hz, 1 mT, 15 h)-induced DNA strand breaks [56]. Analysis of the body of experimental evidence reveals that it is still unclear whether or not exposure to microwaves (EMF of frequency > 300 MHz) affects the nervous system, Int. J. Environ. Res. Public Health 2022, 19, 16150 6 of 15 including neurobehavioural disorders, although some of the studies suggest a non-thermal level effect on learning, memory or behaviour. Some studies on NDD have shown evidence of a potential correlation between EMF and the mechanism of neurodegeneration. However, the correlation is not clearly defined and studies cannot explain the precise mechanisms. Further studies of these effects are needed [57,58]. The inconsistencies in neuronal parameters in response to microwaves reveal an indeterminacy in identifying the molecular impacts of EMF, and in discriminating thermal from non-thermal effects [59]. The lack of conclusive evidence stems from the ambiguity regarding exposure, proper protocols, control groups, and dosimetry in many studies. Additional experiments are required to assess whether longer-term exposure could be associated with symptoms [18]. Co-exposures of several factors may have a significant influence on the development of NDD. Deng et al. [60] investigated whether memory impairments produced in mice by chronic aluminium (Al) treatment (200 mg/kg) could be modulated by magnetic field exposure (50 Hz, 2 mT for 4 h/day, 6 days/week). It was found that both aluminium and EMF could have an impact on learning memory and pro-oxidative function in mice by neuronal cell loss and overexpression of phosphorylated tau protein in the hippocampus and cerebral cortex. However, there was no evidence of any association between EMF exposure and aluminium loading [60]. Zhang et al. [61] also used chronic Al treatment as a contributing factor to cognitive function impairment in AD to examine whether or not EMF (50 Hz, 0.1 mT) and Al have synergistic effects on AD pathogenesis. The results showed learning and memory impairment, neuronal cell loss, and high density of amyloidβ (Aβ) in the hippocampus and cerebral cortex in Al treatment rats. EMF exposure had no effect on the pathogenesis of AD induced by Al overload [61]. Contrary results were obtained by Liu et al. [62]. This group investigated EMF exposure (50 Hz, 0.4 mT, 60 d) combined with intraperitoneal D-galactose (50 mg/kg, 42 d) which can cause premature aging and organ decline, and Aβ25–35 hippocampal (5 µL) injection inducing AD-like clinicopathological features. All these factors were implemented to establish a complex rat model and relationship between EMF exposure and AD development. The results showed that EMF partially improved the cognitive and clinicopathologic symptoms of AD rats, which indicates that certain conditions of EMF exposure could delay the development of AD in rats [62]. It should be emphasised that the result offers only the possibility of using EMF exposure with specific parameters in medical treatment, and does not exclude possible contraindications to such therapy or its side effects. Overall, extremely low-frequency magnetic field and radiofrequency EMF were evaluated as possibly carcinogenic to humans (Group 2B). On the other hand, the power-frequency electric field, as well as static electric and magnetic field, was judged “not classifiable” based on “inadequate” evidence from both humans and animals (Group 3) [11,12]. Maaroufi et al. [58] tested the hypothesis of a possible link between an iron overload in the brain and neurodegenerative disorders. They studied whether combined radiofrequency EMF exposure (900 MHz, 0.05–0.18 W/kg, 1 h/day, 21 d) and iron overload (which is neurotoxic and can contribute to learning deficits, etc.) influenced the outcome of spatial cognitive tasks, neurochemistry, and oxidative stress in rats. The results show that rats exposed to EMF displayed impaired exploratory activity, but not in the navigation and working memory tasks. Some changes in dopamine levels in certain brain regions were noted, but not in all parts of the brain. There were no consistent effects on parameters related to oxidative balance in the brain. The iron overload did not exacerbate the effects of radiofrequency EMF exposure [58]. Exposure to EMF may interact with chemical agents by exhibiting an increase or a decrease in the effects of the latter. Nevertheless, due to the small number of investigations available and the large variety of protocols adopted (different chemical treatments and different EMF exposure conditions), it is not possible to draw valid conclusions [18]. The question of whether or not EMF increases the incidence of morbidity in people with a genetic predisposition is still open. Experimental results are inconsistent, e.g., some indicate that EMF induces brain DNA damage [63], while others show that EMF exposure Int. J. Environ. Res. Public Health 2022, 19, 16150 7 of 15 does not result in a significant effect on inflammation-related genes or protein expression in the immune cells [64]. Animal studies on changes relevant to human NDD in the context of EMF exposure are scarce [2]. Both in vivo and in vitro experiments used various models and EMF exposure conditions, mostly acute or short-term (with exposures ranging from a few minutes to several days). 4. Evidence for Accelerating Neurological Mortality (NM) 2000–2015 The problem with exploring an environmental impact on human health is that it ignores “individual” epigenetic variations. Discoveries reveal that a wide array of environmental, dietary, behavioural, and medical experiences can significantly affect the future development and health of an individual [65]. Based upon earlier research, the question was asked regarding whether there has been an increase in neurological mortality (NM), outstripping changes in population between 1989 and 2015, and the answer was strongly in the affirmative [66]. In the analysed period, the proportion of the over 75-year-olds in the general population rose substantially, e.g., more than doubling in Japan and Spain, and increasing by more than 50% in another 12 analysed countries [24,26]. This problem also occurs in countries that were not described in the WHO report e.g., in Poland where the proportion of over 75-year-olds in the general population increased by 43% between 2002 and 2015 [67]. Likewise, substantial rises in NM were noted in 11 of the 21 analysed developed countries (DC). One remarkable change was that the USA initially had the 15th highest neurological mortality rate out of 21 countries (1979–1997), but by 2015 it had risen to be the second highest [24,26,66]. The initial explanation for these changes was demographics, namely the Gompertzian hypothesis that more people were now living longer and developing age-related diseases [25]. However, the new phenomena of rises in early-onset forms of dementia, occurring in the last 10 years, were ignored [68,69]. Indeed, supportive charities have been developed to help with this growing problem [70], all of which points strongly towards environmental factors. This does not exclude underlying genetic factors, indeed with a greater understanding of epigenetics, it is now appreciated that environmental changes can trigger underlying genetic predispositions [24,66]. Bearing in mind how long genetic and environmental changes need to make a visible impact upon patterns of human health, this work led to a realisation about the relatively short time in which neurological death rates have been accelerating amongst DC. Here we present the results of the analysis of the most recent WHO mortality data. available, updated December 2018 [26], which demonstrates the extent of the acceleration of NM in two categories: ‘Nervous Disease Deaths’, which includes the major conditions such as Motor Neurone Disease, PD, MS, etc., and, ‘AD and Other Dementia Deaths’, in 21 analysed DC, over a mere 15 year period, 2000 to 2015 [26,71]. The data show that in every DC the rates of death from general neurological diseases such as Motor Neurone Disease, PD and MS for 55–74-year-olds had a greater increase, rather than from dementias in the same age range. While NM rates fell between 2000 and 2015 in Belgium −3%, Canada −11% and France −28%, in twelve countries a substantial increase was observed (>20%). There were notable rises in the larger countries, such as Germany by 52%, Japan 60%, Sweden 52%, USA 50% and the UK 51%, for the combined sexes, in just 15 years. The changes in total NM rates in 21 selected DC divided by gender are shown in Table 2. Numbers of total NM are represented via the WHO Age-Standardised-Death-Rates (ASDR) per million (pm). The age-standardized mortality rate is a weighted average of the age-specific mortality rates per 100,000 persons, where the weights are the proportions of persons in the corresponding age groups of the WHO standard population. Standardization by age is important for most health problems. Int. J. Environ. Res. Public Health 2022, 19, 16150 8 of 15 Table 2. Male and female combined neurological mortality (NM) in Age-Standardised Death Rates (ASDR), split by sex rates per million (pm) in selected developed countries. Based upon WHO [26] data. Country Total Male NM ASDR NM/pm Change Total Female NM ASDR NM/pm [%] 1. Australia 2000 2015 246 383 2. Austria 2000 2015 3. Change [%] 56 231 380 65 129 226 75 91 202 122 Belgium 2000 2015 238 405 70 274 367 34 4. Canada 2000 2015 358 393 10 345 399 16 5. Denmark 2000 2015 246 405 65 206 419 103 6. France 2000 2014 331 334 1 280 322 15 7. Finland 2000 2015 481 999 108 462 938 103 8. Germany 2000 2015 169 302 79 117 262 124 9. Greece2000 2015 151 228 51 77 101 31 10. Ireland 2000 2014 217 408 88 194 405 109 11. Italy 2000 2015 231 288 25 200 270 35 12. Japan 2000 2015 71 125 76 49 100 104 13. Netherland 2000 2015 260 477 83 272 482 92 14. New Zealand 2000 2013 291 344 18 238 342 44 15. Norway 2000 2015 262 368 40 204 309 51 16. Portugal 2000 2014 162 292 80 121 228 88 17. Spain 2000 2015 298 394 32 291 401 38 18. Sweden 2000 2015 260 398 53 251 436 74 19. Switzerland 2000 2015 312 346 11 274 400 46 20. UK 2000 2015 217 531 145 192 558 191 21. USA 2000 2015 330 557 69 325 606 86 Int. J. Environ. Res. Public Health 2022, 19, 16150 9 of 15 The ASDR in all age groups rose by more than 50% in 15 DC in this century, reaching the highest value in Germany 125%, Japan 151% and UK 202%. The smallest increase was observed in France and Canada, 8% and 14%, respectively, with an overall average increase of 59% over the period (Table 2). However, rates per million perhaps do not reflect the practical situation faced by families and public health services regarding conditions that usually are present for more than a decade before death. In Table 3. we compare the population of people aged 55–74 and the total populations of the eight largest analysed DC, which gives a more realistic indication of the increasing rate of NM. Table 3. Comparison the population of people aged 55–74 and the total populations of the eight largest analysed developed countries (actual numbers, both sexes, neurological mortality, population (in millions)). Odds ratios (OR) of total population to neurological mortality (NM). Based upon WHO [26] data. Population of 55–74 Years Old Country 2000–2015 Total NM Total Population Change 2000–2015 [%] Change Odds Ratio [%] Canada Neurological mortality: Population (in millions): 2649–3652 0.496–0.761 +38 +53 19,293–35,091 30.791–35.255 +82 +14 59% France Neurological mortality: Population (in millions): 6236–5997 10.628–13.956 −4 +31 40,594–71,543 58.898–64.129 +76 +9 62% Germany Neurological mortality: Population (in millions:) 5790–9332 18.424–19.491 +61 +6 22,543–73,310 82.188–81.687 +225 −1 227% Italy Neurological mortality: Population (in millions): 5693–6542 12.598–14.231 +68 +13 27,554–61,678 56.924–60.731 +124 +7 110% Japan Neurological mortality: Population (in millions): 4438–8099 29.392–33.471 +82 +14 14,023–56,027 125.612–125.319 +299 −1 300% Spain Neurological mortality: Population (in millions): 3892–5007 7.888–9.876 +29 +25 26,679–62,871 40.174–46.410 +135 +16 104% UK Neurological mortality: Population (in millions): 4650–9019 11.065–13.792 +94 +25 24,601–103,550 59.704–65.110 +321 +9 286% USA Neurological mortality: Population (in millions:) 21,818–48,047 42.666–67.380 +120 +58 174,708–436,438 281.421–319.929 +150 +14 120% All the countries, including the outliers Canada and France, showed a considerable increase, even in the group of 55–74-year-olds. In terms of total NM, compared with rises in total populations, France and Canada produced an odds ratio of 62% and 59%, respectively, and the other six countries’ odds ratios were >100% (Table 3). Let us look into the changes in the populations using data from the UK and the USA as an example. In Britain, the 55–74 population rose 25%, but NM cases went from 4650 to 9019 (up 94%), while total NM in the entire population went from 24,601 to 103,550, more than trebling, while the total population rose by just 9% in this century. In the USA, the older age band population (55–74 year-olds) rose by an impressive 58%, but their NM went Int. J. Environ. Res. Public Health 2022, 19, 16150 10 of 15 from 21,818 to 48,047, up 120%. With regard to total NM, they rose from 174,708 to 436,438, up 150%, while the population rose by just 14% [26]. Surely the only word to describe these changes is acceleration and in just 15 years. Clearly, having a greater number of older people increases the risk of more age-related deaths, but for that to happen at such speed there seem to be major environmental influences. Whether EMF plays a considerable role is still an open question, but less so when it has come to be seen as one factor amongst various other interactive environmental pollutants, and possibly a triggering factor. Whatever the likely multiple interactive causes of these changes may be, numbers like these are a matter of great concern, and the authorities need to respond by determining exactly the reasons underlying these rises, and how to make the environment safer. 5. Discussion This work aims to present current knowledge about the influence of EMF on the incidence of NDD, which is becoming progressively common in today’s world. Currently, in the era of rapid technical progress, people are surrounded by devices emitting an EMF, and the number of NDD occurrences is rising. It is perhaps unsurprising then, given this apparent correlation, that researchers are trying to understand whether there is any causation between the two issues. The thesis that EMF increases the risk of NDD deserves thorough and comprehensive research, research that crosses disciplinary boundaries due to the interaction of many environmental factors upon human health. Studies investigating the possible effects of EMF exposure on NDD are too diverse with regard to applied EMF, the duration of exposure, and the statistical methods to draw any reasonable and satisfactory conclusion [18]. In the case of PD and MS, there is not enough research to determine whether EMF affects their development. However, some scientists cast a shadow of uncertainty claiming that EMF contributes to the formation of oxidative stress in the body, and therefore leads to the incidence of these diseases. However, many studies are indicating the participation of EMF in the development of AD and ALS. Although the results are not consistent, there is an increased risk of AD observed across populations. Undoubtedly, further intensive research is needed to assess the mechanism of EMF acting on NDD. The effects on ROS, lipid peroxidation, and antioxidant defence are among the proposed mechanisms, though none of them has been finally defined. The difficulties with the identification and experimental validation of the EMF influence mechanism are due to the variability of biological responses and a lack of consistency in the findings. To summarise, the published results are not unequivocal and are often contradictory (Figure 1), so further research is needed to thoroughly explain the mechanism of action of EMF on the central nervous system, and to explain its potential relationship with NDD. Another important factor that needs to be considered is that the development of a disease does not solely result from environmental factors. As illnesses usually depend on potential genetic predispositions, two individuals exposed to the same noxious pollutant may develop various medical conditions. In the course of seeking explanations for the impact of EMF on human health, we have concluded that the previous view, whereby most apparent increases in incidences of disease were due to the demographics (having more elderly people in the population), might be flawed. This is clear, not only from the remarkable increase in early-onset-dementias (EOD), but from how, starting in the late 1990s, disproportionate rises in NM have been reported by more and more DC, indicating that the disease process was initiated at an earlier life stage. Of course, having more over 75-year-olds in the overall population involves more age-related diseases, but such NM were often treble the rate of population rises in this age group. So, the question arises–if death is inevitable, why is it due to neurological causes? Perhaps the most thought-provoking finding is the acceleration of NM in a relatively short period of just 15 years. The numbers are alarming, and it would be negligent of us if we failed to emphasise the extent of the problem and the possible contributory causes. Int. J. Environ. Res. Public Health 2022, 19, 16150 11 of 15 Current scientific data are not sufficient to determine the dependence of the particular effects on the EMF exposure parameters, and thus determine the numerical value of the exposure threshold at which the defence mechanisms of each human body are insufficient (or sufficient) to protect health. As mentioned, co-exposures of several factors may have a significant influence on the development of NDD. Exposure to EMF may act as an age-dependent risk factor. With age, more and more damaged or misfolded defective molecules are stored in inclusion bodies (“garbage bags”) in and between neurons, thus enhancing the degeneration of cells. The defective molecules also disturb the function of the neurons, leading to cell death [48]. It has been demonstrated that EMF can activate the cellular stress response through increased levels of stress proteins, such as HSP70 [72]. Some authors highlighted that the onset of stress response through EMF exposure should be considered as a defence reaction of the cell to damaging agents [73]. Other authors have suggested the beneficial effects of EMF acting as a mild stressor and inducing protection against various stressors [74,75]. Cellular homeostatic mechanisms may quickly compensate for the physiological disturbances [72]. However, EMF may also decrease the tolerance threshold towards additional oxidativebased challenges. Co-exposure to EMF and other stress factors could trigger the failure of the antioxidant cell response leading to oxidative damage and functional impairment. This, in turn, may significantly increase the risk of the development of NDD [76]. This paper presents the available data on the influence of EMF on the incidence of neurodegenerative EMF Int. J. Environ. Res. Public Health 2022, 19, x FOR PEER REVIEW diseases and the mechanisms of this impact. The role of 12 of 16 as a factor in increased mortality in populations as a result of neurodegenerative diseases is also considered. The work aimed to show how complex this problem is, how difficult it is to compare data with each other, and, therefore, despite the existence of a lot of difficult to define conclusions. There is no doubt that the impact of EMF on the incidence data, it is difficult to define conclusions. There is no doubt that the impact of EMF on of neurodegenerative diseases cannot be overlooked and more systematic, standardized the incidence of neurodegenerative diseases cannot be overlooked and more systematic, research, e.g., using animal models, a well-designed EMF exposure system, and wellstandardized research, e.g., using animal models, a well-designed EMF exposure system, defined dosimetry should be conducted. and well-defined dosimetry should be conducted. Figure 1. Distribution of EMF effects obtained in epidemiological studies (triangles), Figure 1. Distribution of EMF effects obtained in epidemiological studies (triangles), animalanimal studies studies (squares), andand in experiments on cells in publications discusseddiscussed in this work (squares), in experiments on(circles)—reported cells (circles)—reported in publications in(for this work 50/60 Hz): no effects—white, positive—grey, negative—black; reference level for occupational (for 50/60 Hz): no effects—white, positive—grey, negative—black; reference level for occupational exposure is marked [77]. exposure is marked [77]. 6. Conclusions Genes, environment, and behaviour significantly determine life expectancy and the types of diseases a person may be plagued with. Environmental and behaviour stressors decide whether a genetic disposition will manifest. To what extent EMF must be counted among those stressors, science cannot say with certainty at this time: • Studies investigating the possible effects of EMF exposure on NDD are too diverse Int. J. Environ. Res. Public Health 2022, 19, 16150 12 of 15 6. Conclusions Genes, environment, and behaviour significantly determine life expectancy and the types of diseases a person may be plagued with. Environmental and behaviour stressors decide whether a genetic disposition will manifest. To what extent EMF must be counted among those stressors, science cannot say with certainty at this time: • • • • • • Studies investigating the possible effects of EMF exposure on NDD are too diverse with regard to applied EMF, the duration of exposure, and statistical methods to draw any reasonable and satisfactory conclusion. The difficulties with the identification and experimental validation of the EMF influence mechanism are due to the variability of biological responses and a lack of consistency in the findings. There are a number of significant factors besides EMF influencing NDD, such as age, a low level or lack of education, or serious or repeated minor head injuries, and various toxic environmental and occupational agents (including such things as solvents, pesticides, and toxic metals). EMF may interact with other multiple environmental pollutants and/or occupational factors. EMF may have a beneficial impact as a mild stress factor inducing protection against various stressors or, on the contrary, may disturb the stress response of cells, leading to oxidative damage and functional impairment There is no concrete evidence of the positive or negative effects of EMF, however, research should still be carried out in this field, so as not to overlook such a risk factor. Author Contributions: Conceptualization, J.W. and C.P.; methodology, J.W.; investigation, J.W. and C.P.; writing—original draft preparation, J.W. and C.P.; writing—review and editing, J.W.; visualization, J.W.; supervision, J.W. and C.P. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Ethical approval was not required as the manuscript was based on published material in the public sphere. Acknowledgments: We would like to thank Jolanta Karpowicz (Central Institute for Labour Protection– National Research Institute (CIOP-PIB), Poland) for her valuable comments on the manuscript. Conflicts of Interest: The authors declare no conflict of interest. References 1. 2. 3. 4. 5. 6. 7. Terzi, M.; Ozberk, B.; Deniz, O.G.; Kaplan, S. The Role of Electromagnetic Fields in Neurological Disorders. J. Chem. Neuroanat. 2016, 75, 77–84. [CrossRef] Mattsson, M.-O.; Simkó, M. Is There a Relation between Extremely Low Frequency Magnetic Field Exposure, Inflammation and Neurodegenerative Diseases? A Review of in Vivo and in Vitro Experimental Evidence. Toxicology 2012, 301, 1–12. [CrossRef] [PubMed] Lai, H. Neurological Effects of Non-Ionizing Electromagnetic Fields. The Bioinitiative Report 2012. 2012. Available online: http://www.bioinitiative.org (accessed on 3 October 2022). Saliev, T.; Begimbetova, D.; Masoud, A.-R.; Matkarimov, B. Biological Effects of Non-Ionizing Electromagnetic Fields: Two Sides of a Coin. Prog. Biophys. Mol. Biol. 2019, 141, 25–36. [CrossRef] [PubMed] Wyszkowska, J.; Jankowska, M.; Gas, P. Electromagnetic Fields and Neurodegenerative Diseases. Przeglad ˛ Elektrotechniczny 2019, 1, 129–133. [CrossRef] Hosseinabadi, M.B.; Khanjani, N.; Ebrahimi, M.H.; Haji, B.; Abdolahfard, M. The Effect of Chronic Exposure to Extremely Low-Frequency Electromagnetic Fields on Sleep Quality, Stress, Depression and Anxiety. Electromagn. Biol. Med. 2019, 38, 96–101. [CrossRef] Touitou, Y.; Selmaoui, B.; Lambrozo, J. Assessment of Cortisol Secretory Pattern in Workers Chronically Exposed to ELF-EMF Generated by High Voltage Transmission Lines and Substations. Environ. Int. 2022, 161, 107103. [CrossRef] [PubMed] Int. J. Environ. Res. Public Health 2022, 19, 16150 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 13 of 15 Touitou, Y.; Lambrozo, J.; Mauvieux, B.; Riedel, M. Evaluation in Humans of ELF-EMF Exposure on Chromogranin A, a Marker of Neuroendocrine Tumors and Stress. Chronobiol. Int. 2020, 37, 60–67. [CrossRef] Reilly, J.P. Peripheral Nerve Stimulation by Induced Electric Currents: Exposure to Time-Varying Magnetic Fields. Med. Biol. Eng. Comput. 1989, 27, 101. [CrossRef] Madkan, A.; Blank, M.; Elson, E.; Chou, K.-C.; Geddis, M.S.; Goodman, R. Steps to the Clinic with ELF EMF. Nat. Sci. 2009, 1, 157–165. [CrossRef] IARC Working Group on the Evaluation of Carcinogenic Risks to Humans. Non-Ionizing Radiation, Part 2: Radiofrequency Electromagnetic Fields. In IARC Monographs on the Evaluation of Carcinogenic Risks to Humans; International Agency for Research on Cancer (IARC): Lyon, France, 2013; Volume 102. Available online: http://monographs.iarc.fr/ENG/Monographs/vol102/ (accessed on 3 October 2022). IARC Working Group on the Evaluation of Carcinogenic Risks to Humans. Non-Ionizing Radiation, Part 1: Static and Extremely Low-Frequency (ELF) Electric and Magnetic Fields. In IARC Monographs on the Evaluation of Carcinogenic Risks to Humans; International Agency for Research on Cancer (IARC): Lyon, France, 2002; Volume 80, Available online: http://monographs.iarc. fr/ENG/Monographs/vol80/ (accessed on 3 October 2022). Hazlewood, C.F.; Markov, M. Trigger Points and Systemic Effect for EMF Therapy. Environmentalist 2009, 29, 232–239. [CrossRef] Jankowska, M.; Pawlowska-Mainville, A.; Stankiewicz, M.; Rogalska, J.; Wyszkowska, J. Exposure to 50 Hz Electromagnetic Field Changes the Efficiency of the Scorpion Alpha Toxin. J. Venom. Anim. Toxins Trop. Dis. 2015, 21, 38. [CrossRef] Wyszkowska, J.; J˛edrzejewski, T.; Piotrowski, J.; Wojciechowska, A.; Stankiewicz, M.; Kozak, W. Evaluation of the Influence of in Vivo Exposure to Extremely Low-Frequency Magnetic Fields on the Plasma Levels of pro-Inflammatory Cytokines in Rats. Int. J. Radiat. Biol. 2018, 94, 909–917. [CrossRef] [PubMed] Klimek, A.; Nowakowska, A.; Kletkiewicz, H.; Wyszkowska, J.; Maliszewska, J.; Jankowska, M.; Peplowski, L.; Rogalska, J. Bidirectional Effect of Repeated Exposure to Extremely Low-Frequency Electromagnetic Field (50 Hz) of 1 and 7 MT on Oxidative/Antioxidative Status in Rat’s Brain: The Prediction for the Vulnerability to Diseases. Oxidative Med. Cell. Longev. 2022, 2022, e1031211. [CrossRef] [PubMed] D’Angelo, C.; Costantini, E.; Kamal, M.A.; Reale, M. Experimental Model for ELF-EMF Exposure: Concern for Human Health. Saudi J. Biol. Sci. 2015, 22, 75–84. [CrossRef] [PubMed] Scientific Committee on Emerging and Newly Identified Health Risks (SCENIHR) Potential Health Effects of Exposure to Electromagnetic Fields (EMF). Available online: https://health.ec.europa.eu/publications/potential-health-effects-exposureelectromagnetic-fields-emf_en (accessed on 26 September 2022). Ahlbom, A. Neurodegenerative Diseases, Suicide and Depressive Symptoms in Relation to EMF. Bioelectromagn. J. Bioelectromagn. Soc. Soc. Phys. Regul. Biol. Med. Eur. Bioelectromagn. Assoc. 2001, 22, S132–S143. [CrossRef] Riancho, J.; Sanchez de la Torre, J.R.; Paz-Fajardo, L.; Limia, C.; Santurtun, A.; Cifra, M.; Kourtidis, K.; Fdez-Arroyabe, P. The Role of Magnetic Fields in Neurodegenerative Diseases. Int. J. Biometeorol. 2021, 65, 107–117. [CrossRef] Filippini, T.; Hatch, E.E.; Vinceti, M. Residential Exposure to Electromagnetic Fields and Risk of Amyotrophic Lateral Sclerosis: A Dose–Response Meta-Analysis. Sci. Rep. 2021, 11, 11939. [CrossRef] Migliore, L.; Coppedè, F. Environmental-Induced Oxidative Stress in Neurodegenerative Disorders and Aging. Mutat. Res. Toxicol. Environ. Mutagen. 2009, 674, 73–84. [CrossRef] Reale, M.; Kamal, M.A.; Patruno, A.; Costantini, E.; D’Angelo, C.; Pesce, M.; Greig, N.H. Neuronal Cellular Responses to Extremely Low Frequency Electromagnetic Field Exposure: Implications Regarding Oxidative Stress and Neurodegeneration. PLoS ONE 2014, 9, e104973. [CrossRef] Pritchard, C.; Silk, A.; Hansen, L. Are Rises in Electro-Magnetic Field in the Human Environment, Interacting with Multiple Environmental Pollutions, the Tripping Point for Increases in Neurological Deaths in the Western World? Med. Hypotheses 2019, 127, 76–83. [CrossRef] Easton, D.M. Gompertzian Growth and Decay: A Powerful Descriptive Tool for Neuroscience. Physiol. Behav. 2005, 86, 407–414. [CrossRef] [PubMed] WHO. Mortality Database—WHO. Available online: https://www.who.int/data/data-collection-tools/who-mortality-database (accessed on 26 September 2022). Villarini, M.; Gambelunghe, A.; Giustarini, D.; Ambrosini, M.V.; Fatigoni, C.; Rossi, R.; Dominici, L.; Levorato, S.; Muzi, G.; Piobbico, D. No Evidence of DNA Damage by Co-Exposure to Extremely Low Frequency Magnetic Fields and Aluminum on Neuroblastoma Cell Lines. Mutat. Res. Toxicol. Environ. Mutagen. 2017, 823, 11–21. [CrossRef] [PubMed] Zubko, O.; Gould, R.L.; Gay, H.C.; Cox, H.J.; Coulson, M.C.; Howard, R.J. Effects of Electromagnetic Fields Emitted by GSM Phones on Working Memory: A Meta-analysis. Int. J. Geriatr. Psychiatry 2017, 32, 125–135. [CrossRef] [PubMed] Guerriero, F.; Botarelli, E.; Mele, G.; Polo, L.; Zoncu, D.; Renati, P.; Sgarlata, C.; Rollone, M.; Ricevuti, G.; Maurizi, N.; et al. An Innovative Intervention for the Treatment of Cognitive Impairment–Emisymmetric Bilateral Stimulation Improves Cognitive Functions in Alzheimer’s Disease and Mild Cognitive Impairment: An Open-Label Study. Neuropsychiatr. Dis. Treat. 2015, 11, 2391–2404. [CrossRef] Cichoń, N.; Rzeźnicka, P.; Bijak, M.; Miller, E.; Miller, S.; Saluk, J. Extremely Low Frequency Electromagnetic Field Reduces Oxidative Stress during the Rehabilitation of Post-Acute Stroke Patients. Adv. Clin. Exp. Med. 2018, 27, 1285–1293. [CrossRef] Int. J. Environ. Res. Public Health 2022, 19, 16150 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 14 of 15 WHO. Extremely Low Frequency Fields. Environmental Health Criteria Monograph No. 238. Available online: https://www.who.int/ publications-detail-redirect/9789241572385 (accessed on 26 September 2022). Cannon, J.R.; Greenamyre, J.T. The Role of Environmental Exposures in Neurodegeneration and Neurodegenerative Diseases. Toxicol. Sci. 2011, 124, 225–250. [CrossRef] Dugger, B.N.; Dickson, D.W. Pathology of Neurodegenerative Diseases. Cold Spring Harb. Perspect. Biol. 2017, 9, a028035. [CrossRef] Alzheimer’s Disease Fact Sheet. Available online: https://www.nia.nih.gov/health/alzheimers-disease-fact-sheet (accessed on 3 October 2018). Neuropathology. Available online: https://neuropathology-web.org/ (accessed on 27 September 2022). García, A.M.; Sisternas, A.; Hoyos, S.P. Occupational Exposure to Extremely Low Frequency Electric and Magnetic Fields and Alzheimer Disease: A Meta-Analysis. Int. J. Epidemiol. 2008, 37, 329–340. [CrossRef] Hug, K.; Röösli, M.; Rapp, R. Magnetic Field Exposure and Neurodegenerative Diseases–Recent Epidemiological Studies. Soz.-Präventivmedizin 2006, 51, 210–220. [CrossRef] Zhou, H.; Chen, G.; Chen, C.; Yu, Y.; Xu, Z. Association between Extremely Low-Frequency Electromagnetic Fields Occupations and Amyotrophic Lateral Sclerosis: A Meta-Analysis. PLoS ONE 2012, 7, e48354. [CrossRef] Savitz, D.A.; Checkoway, H.; Loomis, D.P. Magnetic Field Exposure and Neurodegenerative Disease Mortality among Electric Utility Workers. Epidemiol. Camb. Mass. 1998, 9, 398–404. [CrossRef] Haakansson, N.; Gustavsson, P.; Johansen, C.; Floderus, B. Neurodegenerative Diseases in Welders and Other Workers Exposed to High Levels of Magnetic Fields. Epidemiology 2003, 14, 420–426. [CrossRef] Feychting, M.; Jonsson, F.; Pedersen, N.L.; Ahlbom, A. Occupational Magnetic Field Exposure and Neurodegenerative Disease. Epidemiology 2003, 14, 413–419. [CrossRef] [PubMed] Huss, A.; Spoerri, A.; Egger, M.; Röösli, M.; Study, S.N.C. Residence near Power Lines and Mortality from Neurodegenerative Diseases: Longitudinal Study of the Swiss Population. Am. J. Epidemiol. 2009, 169, 167–175. [CrossRef] Andel, R.; Crowe, M.; Feychting, M.; Pedersen, N.L.; Fratiglioni, L.; Johansson, B.; Gatz, M. Work-Related Exposure to Extremely Low-Frequency Magnetic Fields and Dementia: Results from the Population-Based Study of Dementia in Swedish Twins. J. Gerontol. Ser. Biomed. Sci. Med. Sci. 2010, 65, 1220–1227. [CrossRef] [PubMed] Davanipour, Z.; Tseng, C.-C.; Lee, P.-J.; Markides, K.S.; Sobel, E. Severe Cognitive Dysfunction and Occupational Extremely Low Frequency Magnetic Field Exposure among Elderly Mexican Americans. Br. J. Med. Med. Res. 2014, 4, 1641–1662. [CrossRef] [PubMed] Pedersen, C.; Poulsen, A.H.; Rod, N.H.; Frei, P.; Hansen, J.; Grell, K.; Raaschou-Nielsen, O.; Schüz, J.; Johansen, C. Occupational Exposure to Extremely Low-Frequency Magnetic Fields and Risk for Central Nervous System Disease: An Update of a Danish Cohort Study among Utility Workers. Int. Arch. Occup. Environ. Health 2017, 90, 619–628. [CrossRef] [PubMed] Johansen, C. Exposure to Electromagnetic Fields and Risk of Central Nervous System Disease in Utility Workers. Epidemiology 2000, 11, 539–543. [CrossRef] Parlett, L.E.; Bowman, J.D.; van Wijngaarden, E. Evaluation of Occupational Exposure to Magnetic Fields and Motor Neuron Disease Mortality in a Population-Based Cohort. J. Occup. Environ. Med. Coll. Occup. Environ. Med. 2011, 53, 1447. [CrossRef] Gunnarsson, L.-G.; Bodin, L. Occupational Exposures and Neurodegenerative Diseases—A Systematic Literature Review and Meta-Analyses. Int. J. Environ. Res. Public. Health 2019, 16, 337. [CrossRef] Frei, P.; Poulsen, A.H.; Mezei, G.; Pedersen, C.; Cronberg Salem, L.; Johansen, C.; Roosli, M.; Schuz, J. Residential Distance to High-Voltage Power Lines and Risk of Neurodegenerative Diseases: A Danish Population-Based Case-Control Study. Am. J. Epidemiol. 2013, 177, 970–978. [CrossRef] [PubMed] Poulsen, A.H.; Stenager, E.; Johansen, C.; Bentzen, J.; Friis, S.; Schüz, J. Mobile Phones and Multiple Sclerosis—A Nationwide Cohort Study in Denmark. PLoS ONE 2012, 7, e34453. Rollwitz, J.; Lupke, M.; Simkó, M. Fifty-Hertz Magnetic Fields Induce Free Radical Formation in Mouse Bone Marrow-Derived Promonocytes and Macrophages. Biochim. Biophys. Acta BBA-Gen. Subj. 2004, 1674, 231–238. [CrossRef] [PubMed] Kıvrak, E.G.; Yurt, K.K.; Kaplan, A.A.; Alkan, I.; Altun, G. Effects of Electromagnetic Fields Exposure on the Antioxidant Defense System. J. Microsc. Ultrastruct. 2017, 5, 167–176. [CrossRef] Consales, C.; Merla, C.; Marino, C.; Benassi, B. Electromagnetic Fields, Oxidative Stress, and Neurodegeneration. Int. J. Cell Biol. 2012, 2012, 683897. [CrossRef] Mahaki, H.; Jabarivasal, N.; Sardarian, K.; Zamani, A. Effects of Various Densities of 50 Hz Electromagnetic Field on Serum IL-9, IL-10, and TNF-α Levels. Int. J. Occup. Environ. Med. 2020, 11, 24–32. [CrossRef] Falone, S.; Mirabilio, A.; Carbone, M.C.; Zimmitti, V.; Di Loreto, S.; Mariggiò, M.A.; Mancinelli, R.; Di Ilio, C.; Amicarelli, F. Chronic Exposure to 50 Hz Magnetic Fields Causes a Significant Weakening of Antioxidant Defence Systems in Aged Rat Brain. Int. J. Biochem. Cell Biol. 2008, 40, 2762–2770. [CrossRef] Ivancsits, S.; Diem, E.; Jahn, O.; Rüdiger, H.W. Age-Related Effects on Induction of DNA Strand Breaks by Intermittent Exposure to Electromagnetic Fields. Mech. Ageing Dev. 2003, 124, 847–850. [CrossRef] Arendash, G.W.; Mori, T.; Dorsey, M.; Gonzalez, R.; Tajiri, N.; Borlongan, C. Electromagnetic Treatment to Old Alzheimer’s Mice Reverses β-Amyloid Deposition, Modifies Cerebral Blood Flow, and Provides Selected Cognitive Benefit. PLoS ONE 2012, 7, e35751. [CrossRef] Int. J. Environ. Res. Public Health 2022, 19, 16150 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 15 of 15 Maaroufi, K.; Had-Aissouni, L.; Melon, C.; Sakly, M.; Abdelmelek, H.; Poucet, B.; Save, E. Spatial Learning, Monoamines and Oxidative Stress in Rats Exposed to 900MHz Electromagnetic Field in Combination with Iron Overload. Behav. Brain Res. 2014, 258, 80–89. [CrossRef] Teimori, F.; Khaki, A.A.; Hemmati, R.; Rajabzadeh, A. Probably Role of Antioxidants Against EMFs-Induced Effects on Central Nervous System Structures: A Mini Review. Crescent J. Med. Biol. Sci. 2017, 4, 92–98. Deng, Y.; Zhang, Y.; Jia, S.; Liu, J.; Liu, Y.; Xu, W.; Liu, L. Effects of Aluminum and Extremely Low Frequency Electromagnetic Radiation on Oxidative Stress and Memory in Brain of Mice. Biol. Trace Elem. Res. 2013, 156, 243–252. [CrossRef] [PubMed] Zhang, C.; Li, Y.; Wang, C.; Lv, R.; Song, T. Extremely Low-Frequency Magnetic Exposure Appears to Have No Effect on Pathogenesis of Alzheimer’s Disease in Aluminum-Overloaded Rat. PLoS ONE 2013, 8, e71087. [CrossRef] [PubMed] Liu, X.; Zuo, H.; Wang, D.; Peng, R.; Song, T.; Wang, S.; Xu, X.; Gao, Y.; Li, Y.; Wang, S. Improvement of Spatial Memory Disorder and Hippocampal Damage by Exposure to Electromagnetic Fields in an Alzheimer’s Disease Rat Model. PLoS ONE 2015, 10, e0126963. [CrossRef] [PubMed] Mariucci, G.; Villarini, M.; Moretti, M.; Taha, E.; Conte, C.; Minelli, A.; Aristei, C.; Ambrosini, M.V. Brain DNA Damage and 70-KDa Heat Shock Protein Expression in CD1 Mice Exposed to Extremely Low Frequency Magnetic Fields. Int. J. Radiat. Biol. 2010, 86, 701–710. [CrossRef] Bouwens, M.; de Kleijn, S.; Ferwerda, G.; Cuppen, J.J.; Savelkoul, H.F.J.; Kemenade, B.M.L.V. Low-Frequency Electromagnetic Fields Do Not Alter Responses of Inflammatory Genes and Proteins in Human Monocytes and Immune Cell Lines. Bioelectromagnetics 2012, 33, 226–237. [CrossRef] Rothstein, M.A.; Cai, Y.; Marchant, G.E. The Ghost in Our Genes: Legal and Ethical Implications of Epigenetics. Health Matrix 2009, 19, 1–62. Pritchard, C.; Rosenorn-Lanng, E.; Silk, A.; Hansen, L. Controlled Population-Based Comparative Study of USA and International Adult [55–74] Neurological Deaths 1989–2014. Acta Neurol. Scand. 2017, 136, 698–707. [CrossRef] demografia.stat.gov.pl Baza Demografia—Główny Urzad ˛ Statystyczny. Available online: https://demografia.stat.gov.pl/ BazaDemografia/StartIntro.aspx (accessed on 26 September 2022). Bonvicini, C.; Scassellati, C.; Benussi, L.; Di Maria, E.; Maj, C.; Ciani, M.; Fostinelli, S.; Mega, A.; Bocchetta, M.; Lanzi, G. Next Generation Sequencing Analysis in Early Onset Dementia Patients. J. Alzheimers Dis. 2019, 67, 243–256. [CrossRef] Strand, B.H.; Knapskog, A.-B.; Persson, K.; Edwin, T.H.; Bjertness, E.; Engedal, K.; Selbaek, G. The Loss in Expectation of Life Due to Early-Onset Mild Cognitive Impairment and Early-Onset Dementia in Norway. Dement. Geriatr. Cogn. Disord. 2019, 47, 355–365. [CrossRef] Meet the Team. Available online: https://www.youngdementiauk.org/meet-team (accessed on 27 December 2019). Pritchard, C.; Silk, A.; Hansen, L. Accelerating Rises of Neurological Deaths in the UK and the Other Western Countries 20002015: Urgent Need for Policy Response 2019. Available online: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3494401 (accessed on 3 October 2022). Zeni, O.; Simkó, M.; Scarfi, M.R.; Mattsson, M.-O. Cellular Response to ELF-MF and Heat: Evidence for a Common Involvement of Heat Shock Proteins? Front. Public Health 2017, 5, 280. [CrossRef] [PubMed] Blank, M.; Goodman, R. Electromagnetic Fields Stress Living Cells. Pathophysiology 2009, 16, 71–78. [CrossRef] [PubMed] Guerriero, F.; Ricevuti, G. Extremely Low Frequency Electromagnetic Fields Stimulation Modulates Autoimmunity and Immune Responses: A Possible Immuno-Modulatory Therapeutic Effect in Neurodegenerative Diseases. Neural Regen. Res. 2016, 11, 1888–1895. [CrossRef] [PubMed] Carmody, S.; Wu, X.L.; Lin, H.; Blank, M.; Skopicki, H.; Goodman, R. Cytoprotection by Electromagnetic Field-Induced Hsp70: A Model for Clinical Application. J. Cell. Biochem. 2000, 79, 453–459. [CrossRef] [PubMed] Falone, S.; Grossi, M.R.; Cinque, B.; D’Angelo, B.; Tettamanti, E.; Cimini, A.; Di Ilio, C.; Amicarelli, F. Fifty Hertz Extremely Low-Frequency Electromagnetic Field Causes Changes in Redox and Differentiative Status in Neuroblastoma Cells. Int. J. Biochem. Cell Biol. 2007, 39, 2093–2106. [CrossRef] International Commission on Non-Ionizing Radiation Protection. Guidelines for Limiting Exposure to Time-Varying Electric and Magnetic Fields (1 Hz to 100 KHz). Health Phys. 2010, 99, 818–836. [CrossRef]
https://openalex.org/W2184154077	https://storage.googleapis.com/jnl-vt-j-jcte-files/journals/1/articles/121/submission/proof/121-1-211-1-10-20171115.pdf	English	null	A Survey of Vocational Training Needs of 15-25 Years Old Out-of-School Youths in Bauchi Metropolis	Journal of career and technical education	2,007	cc-by	8,362	A Survey of Vocational Training Needs of 15 – 25 Years Old Out-of-School Youths in Bauchi Metropolis M. I. BELLO Abubakar Tafawa Balewa University I. M. DANJUMA Abubakar Tafawa Balewa University A. Y. ADAMU Abubakar Tafawa Balewa University ABSTRACT The main purpose of the study was to explore and describe the vocational training needs of 15 - 25 years old out-of-school youths in Bauchi Metropolis of Bauchi State, Nigeria. Specifically, the researchers sought to; describe their demographic characteristics, examine their vocational training needs, the influence of gender on the vocational needs, and also identify problems associated in their vocational training needs. The design for the study was survey that involved 128, 15 - 25 years old out-of-school youths. Since the study population frame did not exist, a cluster sampling technique was employed in the study. The cluster sample was drawn from eight wards that constitute Bauchi Metropolis. Each ward taken as a cluster. The instrument used for the study was a 16-item questionnaire, which was constructed by the researchers and validated by experts from Vocational and Technology Education Programme of Abubakar Tafawa Balewa University. Data collected were analyzed using frequency and percentage statistical tools. Finally conclusions were reached and recommendations made. ©2007 - Journal of Career and Technical Education, Vol. 23, No. 1, Fall, 2007 – Page 55 A Survey of Vocational Training Needs of 15 – 25 Years Old Out-of-School Youths in Bauchi Metropolis A Survey of Vocational Training Needs of 15 – 25 Years Old Out-of-School Youths in Bauchi Metropolis Background of the Study Youths represent the future and hope of every country. The high returns on resources invested in youths today have both immediate and long term benefits. Where they existed and are functioning well, youth programmes play an important role in building life skills of individual young people, strengthening families and communities, and working towards sustainable development as a major contributor to the overall progress of a country (Seiders, 1985). Many countries have realized that the development and harnessing the potential of the youths can best be achieved through a sound educational system. In Nigeria for instance, the training given to youths at the secondary school level is both pre-vocational and academic. According to the National Policy on Education (NPE, 2004), the aim of secondary school is to make a person to be productive to himself and the society. Academic and vocational courses are offered at the Junior Secondary School (JSS) level. These courses are suppose to provide definite purpose and meaning to education by relating to occupational goals, provides technical knowledge and work skills necessary for employment, and develop abilities, attitudes, work habits and appreciation which contributes to a satisfying and productive life. Abdullahi (1993) states that vocational education is expected among other things to assist students to acquire relevant occupational and technical skills, prepare for future occupations, and make successful transition from school to world of work. However, it is very disturbing to observe that most of our graduates at both the JSS and SSS levels that have not further their studies stay without job and further education. Despite the fact that the training at JSS level was target at preparing this group of youths with both academic and vocational skills that would make them become self- reliant and also, prepare them for further studies. What then was the problem? Was it that these youths have not been given the adequate training that would enable them identifying their needs, so that they can initiate and pursue available vocation? What are the problems that are militating against the efforts being made to help the youths became self-reliant and useful to the society? This study attempts to find out answers to some of these questions. ©2007 - Journal of Career and Technical Education, Vol. 23, No. 1, Fall, 2007 – Page 56 Literature Review United Nations (1985) described youth as those persons falling between the ages of 15 - 24 years. By this definition, therefore children are those persons under the age of 14. According to World Bank (2006), the term "youth" in general refers to those who are between the ages of 15 - 25. However, it is worth noting that Article 1 of the United Nations defines children as persons up to the age of 18. Though by that explanation many countries also draw a line on youth at the age at which a person is given equal treatment under the law - often referred to as the "age of maturity". This age is often 18 in many countries, and once a person passes this age, he is considered to be an adult. As enshrined in Nigeria, it is expected that when a child attained the ages of 18, he/she must have acquired a skill that will make a person to be productive to himself and the society (NPE, 2004), which by implication can fend for himself with minimum supervision from elders. However, the operational definition and nuances of the term 'youth' often vary from country to country, depending on the specific socio-cultural, institutional, economic and political factors. For example, in Botswana, the national youth policy adopted in 1996, defined youth as a proportion of the population falling within the age range of 12 to 29 years (Squire, 2000). Education the common property of everyone, the prime creator and conveyor of knowledge in any facet, is the most complex human endeavour (Galadima, 2003). In other words, it is the only means by which a society can become for better or for worse. That is to say strengthening general education at the primary and secondary school levels are the first priority for public policies to improve productivity and flexibility of the work force. In addition to generating broad benefits to society, general education directly increases worker mobility and productivity and increases the access (chance) of the poor and socially disadvantaged groups to training and wage employment (World Bank, 1991). For example, in Peru, workers with complete secondary school education have a 50% greater chance of receiving training. That is to say, training in specific skills is more effective when trainees have strong literacy, numeracy and problem solving skills, which by implication, it is expected that lower secondary education can provide this foundation. Literature Review General education provides the society with values, communicative and manipulative skills on which effective vocational education is based. Without knowledge of reading, writing and arithmetic, good vocational education cannot be imparted. Vocational education builds on a foundation of good general education (Okoro, 1993). World Bank (1991) reports that, 70% of the proprietors in Ibadan, Nigeria who engaged in radio and television repairs have secondary education. That is to say general education teaches those values, skills and knowledge which each citizen should have in order to understand the society in which he lives and play an intelligent part in its affairs. For example electricity can be taught as general education and as vocational education. As general education, the emphasis is on helping students understand the principles of electricity and to become intelligent users of electrical appliances. Okoro (1993) defines vocational education as "any form of education whose primary purpose is to prepare persons for employment in recognized occupations". That is to say vocational education provides skills, knowledge, and attitudes necessary for effective employment in specific occupations. To further buttress the definition given above, vocational education can be conceived as a comprehensive term referring to those aspects of educational process involving, in addition to general, the acquisition of practical skills, attitudes, understanding and knowledge relating to occupations in various sectors of economic and social life. The Nigerian National Policy on Education (NPE, 2004) highlighted aspects of vocational education as: (a) an integral part of general education (b) a means of preparing for occupational fields and effective participation in the world of work. (c) an aspect of lifelong learning and a preparation for responsible citizenship (d) an instrument for promoting environmentally sound sustainable development (e) a method of facilitating poverty alleviation. (a) an integral part of general education (b) a means of preparing for occupational fields and effective participation in the world of work. (c) an aspect of lifelong learning and a preparation for responsible citizenship (d) an instrument for promoting environmentally sound sustainable development (e) a method of facilitating poverty alleviation. These are consistent with those found in UNESCO and ILO (2002) reports. The provision of vocational education starts at the Junior Secondary level in Nigeria, where students/youths are exposed to vocational aspects such as Business Studies and Introductory Technology. Introductory Technology comprises of some theoretical knowledge and practical skills in electricity and electronic, woodwork, metal work, and building trades. ©2007 - Journal of Career and Technical Education, Vol. 23, No. 1, Fall, 2007 – Page 57 Literature Review It is expected that at the end of the three-year Junior Secondary School the youths must have been equipped to either pursue higher education or opt out into the world of work. Abdullahi (1993) states that, vocational education is expected among other things to assist students to acquire relevant occupational and technical skills, prepare for future occupations, and make successful transition from school to world of work. However, it has been observed that the present practice in vocational education curriculum development in Nigeria encourages lack of harmony between the educational practice and the world of work (Oranu, 1990). To buttress the above point UNESCO ©2007 - Journal of Career and Technical Education, Vol. 23, No. 1, Fall, 2007 – Page 57 (1979) had earlier reported that the institutions responsible for manpower training are rarely in close contact with industries (1979) had earlier reported that the institutions responsible for manpower training are rarely in close contact with industries Okoro (2003) categorizes into three the vocational needs that Nigeria in particular may dwelt into so as to reduce the rate of unemployment taking into cognizance the cost implication as follows: (1) Production of Artisans - This includes vocations like mechanics, furniture making and electricians. (1) Production of Artisans - This includes vocations like mechanics, furniture making and electricians. (2) Production of Agriculturalist - This includes vocations like poultry farming, horticulture, animal husbandry. Production of Agriculturalist - This includes vocations like poultry farmin horticulture, animal husbandry. , y (3) Production of Fashion Designers - This includes vocations like tailoring, dyers, barbers. (3) Production of Fashion Designers - This includes vocations like tailoring, dyers, barbers. He also interviewed teachers, young school leavers and students of pre-vocational subjects in Nigeria on the problems militating against proper imparting of vocational skills in schools and revealed that lack of funds, lack of facilities, mismanagement of resources, lack of qualified personnel, and lack of cooperation from principals, among others. Akubudike (2003) also reports other problems such as; Inadequate Staffing, Poor Attitude of Students, Lack of Proper Guidance and Counseling, Un-coordination of Career-Oriented School Clubs or Association, Lack of Exposure to Public Lectures. Suffice it to say that these problems might have contributed to the increasing high rates of school drop-outs in most cities and villages in Nigeria, which became the problem for their immediate society and the government. ©2007 - Journal of Career and Technical Education, Vol. 23, No. 1, Fall, 2007 – Page 58 Literature Review (5) In order to develop work maturity skills, more opportunities for supervised work experience that provide close articulation between in-school educational experience and on-the-job experience. (6) Emphasis on applied basic skills and employability skills in secondary programs (2) In order to improve its image, more and better publicity concerning vocational education should be put in place. (3) The teaching of basic academic skills such as reading, mathematics, and writing should be integrated into vocational instructions. (4) Identification of and instruction in a common core of employability skills that are transferable across occupations including problem-solving and decision making skills and skills necessary for getting and keeping a job should be introduced. (5) In order to develop work maturity skills, more opportunities for supervised work experience that provide close articulation between in school educational (4) Identification of and instruction in a common core of employability skills that are transferable across occupations including problem-solving and decision making skills and skills necessary for getting and keeping a job should be introduced. (5) In order to develop work maturity skills, more opportunities for supervised work experience that provide close articulation between in-school educational experience and on-the-job experience. (6) p j p (6) Emphasis on applied basic skills and employability skills in secondary programs and technical skills in post secondary programmes should be inculcated. Stanwick (2005, 2006) and Sherman (2006) investigated outcomes from various levels of vocational education and training and found out that the employment outcomes six months after training depend on the level of vocational education and training undertaken. They also considered initial outcomes by gender of vocational education and training (VET) students. They found that males had a smoother transition to employment, obtaining better employment outcomes six months after training when compared with females. Smith and Green (2005) conducted a follow-up survey of school students one to three years after they had left school. They found that participating in a school-based apprenticeship provides a clear pathway into apprenticeship or traineeships in a similar industry area. Using the Longitudinal Survey of Australian Youth Data, Anlezark, Karmel and Ong (2006) found that school VET programs provide a clear pathway for some students, particularly for boys studying in the areas of building and engineering. However, few girls appear to keep on with the VET subjects taken at school. ©2007 - Journal of Career and Technical Education, Vol. 23, No. 1, Fall, 2007 – Page 59 Literature Review Even though, successive governments in Nigeria have brought in youth empowerment programmes. Of recent the National Economic Empowerment and Development Strategy (NEEDS), which was aimed at creating a new Nigeria citizen who values hard work and who realizes that one cannot have something for nothing (NEEDS, 2005). Obiefuna (2003) reports that most of the Nigerian youths want to have good things in life but are not ready to be subjected to the world of work. As the world of work is further changing, most of the employers require certain skills from school leavers before being employed. This change in employers' expectation has resulted from the changing nature of workers roles in the work place. Although employers are still interested in hiring individuals who have received occupationally specific training, they also want individuals with a solid grounding in basic academic skills. The types of basic skills employers expect their employees to possess have been listed in a number of publications (e.g. College Entrance Examination, 1984; Carnevale, Gainer and Meltzer, 1988). For example, Hollenbeck (1987) in South Dakota and What Employers Say about Vocational Education in South Carolina (1986) in their researches highlighted some gray areas which need to be mended in order to achieve a purposeful objective in order to deliver the expectations of employers. The areas highlighted includes; Basic Skills, Communication, Image, and Employability Skills. They also made ©2007 - Journal of Career and Technical Education, Vol. 23, No. 1, Fall, 2007 – Page 58 a number of suggestions, which if looked at seriously can really contribute to vocational education programmes in Nigeria. The suggestions include: a number of suggestions, which if looked at seriously can really contribute to vocational education programmes in Nigeria. The suggestions include: (1) More communication and closer collaboration between business/industry and vocational education institutions should be enhanced. (1) More communication and closer collaboration between business/industry and vocational education institutions should be enhanced. vocational education institutions should be enhanced. (2) In order to improve its image, more and better publicity concerning vocational education should be put in place. (3) The teaching of basic academic skills such as reading, mathematics, and writing should be integrated into vocational instructions. (4) Identification of and instruction in a common core of employability skills that are transferable across occupations including problem-solving and decision making skills and skills necessary for getting and keeping a job should be introduced. Literature Review They also noted a mismatch between the VET courses and programs which students undertook at school, and those which they studied after school. This leads to the question of whether school VET programs need to be better linked with the world of work and post-school study or, alternatively, should concentrate on providing a broad pre-vocational skills. Lamb and Vickers (2006) further examined the issue of whether VET programs within schools need to be aligned with those outside schools, and found that linked programs typically result in smoother transitions to work, particularly for students who do not further their education. By tracking young people, researchers are now able to conclude that vocational education and training assists the transition to work. However, the smoothness of this transition does vary depending on student demographics and the nature of the VET program undertaken. Students undertaking JSS 1 and II may not, in the short term, attain the jobs they hoped for without further training. This is in agreement with Harris, Rainey and Sumner (2006) who stated that students who experience multiple pathways are generally not aware of the career development services (which are usually carried out at senior level of secondary school) available or do not think they need them. ©2007 - Journal of Career and Technical Education, Vol. 23, No. 1, Fall, 2007 – Page 60 Statement of the Problem The total youth population in the world today is estimated to be about 1,500 million and is projected to grow to 2 billion by the year 2050 (Seiders, 1985). He also states that the total number of youths will remain greater in rural areas over the next ten to fifteen years; all future growth in number of youth will be in the cities. Going by what is happening and considering the rate at which the youths in Nigeria are graduating at both JSS and SSS levels and a high percentage of them are doing nothing in terms of being engaged in vocations or trades that could earn them a living and preventing them from becoming a menace in the society, one will start to wonder what is going to happen in the future. Experience has shown that some of the graduates of both the JSS and SSS are jobless. Therefore, they become liability to both their parents and the society. Some engage in vices that are detrimental to themselves and the society. However, efforts have been made by the government and some NGOs to address some of these problems through setting up training centres and programmes which some youths undergo in order to become self-reliant. Despite these efforts the problem is becoming very persistent among our youths as their population is increasing. Probably most of these programmes that were targeted at our youths of the secondary school level age group have not been based on sufficient data from empirical studies on their specific needs and aspirations. As such there is the tendency that such laudable programmes may not be able to adequately address the problems of this group of youths in the society. It is against this background that this study examines the vocational training needs of out-of-school youths in Bauchi metropolis. Population The population of the study comprises of all 15-25 years old out-of-school youths in Bauchi Metropolis of Bauchi State. Research Design g The study is a survey research. A survey research is a research that employs the study of large and small population by selecting and studying sample chosen from the population to discover the relative incidence, which by impact can easily serve as a forecaster and predictor (Olaitan, Ali, Eyoh and Sowande, 2000). In this study, the vocational training needs of 15-25 years old out-of-school youths were examined and described. Objectives of the Study The study sought to achieve the following objectives: (1) Describe the demographic characteristics of 15 - 25 years old out-of-school youths in Bauchi metropolis. (2) Examine the vocational training needs of 15 - 25 years old out-of-school youths in Bauchi metropolis. (3) Examine the influence of gender on the vocational training needs of 15 - 25 years old out-of-school youths in Bauchi metropolis. (4) Identify the problems associated in their vocational training needs. Describe the demographic characteristics of 15 - 25 years old out-of-school youths in Bauchi metropolis. Examine the vocational training needs of 15 - 25 years old out-of-school youths in Bauchi metropolis. p Examine the influence of gender on the vocational training needs of 15 - 25 years old out-of-school youths in Bauchi metropolis. (4) y Identify the problems associated in their vocational training needs. (4) Identify the problems associated in their vocational training needs. Research Questions The following research questions were investigated: (1) What are the demographic characteristics of the 15 - 25 years old out-of-school youths in Bauchi Metropolis? (2) What are the vocational training needs of 15 - 25 years old out-of-school youths in BauchiMetropolis? (3) How does gender influences the vocational training needs of 15 - 25 years old out-of-school youths in Bauchi Metropolis? (4) What are the problems associated with the vocational training needs of 15 - 25 years old out-of-school youths in Bauchi Metropolis? How does gender influences the vocational training needs of 15 - 25 years old out-of-school youths in Bauchi Metropolis? What are the problems associated with the vocational training needs of 15 - 25 years old out-of-school youths in Bauchi Metropolis? Area of the Study The area of the study was Bauchi Metropolis in Bauchi State of Nigeria. Bauchi Metropolis been the capital of Bauchi State Nigeria is located at 10o19’N, 9o50’E with a population of 316,173 (2004). The city lies on the Port Harcourt – Maiduguri railway line. Bauchi metropolis consists of eight administrative wards (units). These are: Hardo Ward, Dan'iya Ward, Makama Sarkin Baki 1 Ward, Makama Sarkin Baki 2 Ward, Majidadi A Ward, Majidadi B Ward, Dawaki Ward, and Dankade Ward respectively. ©2007 - Journal of Career and Technical Education, Vol. 23, No. 1, Fall, 2007 – Page 61 Sample and Sampling Technique A total of 128, 15-25 years old out-of-school youths in Bauchi Metropolis participated in the study. The study was aimed at describing the vocational training needs of 15 - 25 years old out-of-school youths in Bauchi Metropolis. Hence the need to employ a sampling method that would ensure a proper representation of the larger population of the out-of-school youths was necessary. In addition the study population frame did not exist; therefore, a cluster sampling method was employed. Cluster sampling involves first selecting large grouping (e.g. wards or units) and then selecting the sample of the study (i.e. elements) from each grouping (Olaitan et.al., 2000). The eight administrative wards of the metropolis constitute the clusters. The parameters that were considered for the selection of the study sample were a youth must be within the age range of 15 - 25 years old and out-of-school. He or she must have possessed at least a ©2007 - Journal of Career and Technical Education, Vol. 23, No. 1, Fall, 2007 – Page 61 minimum qualification of Junior Secondary School or Senior Secondary School Certificate. Using simple random sampling technique, a total of one hundred and twenty eight (128) 15 - 25 years old out-of-school youths were sampled from the eight clusters (wards). Research Instrument For the purpose of this research, one instrument was developed and used by the researchers. That is a 16-item Vocational Training Needs Questionnaire (VNTQ). The VNTQ sought demographic information on the participants such as sex, age, address, marital status, educational status, employment status and agency where training was acquired, which describes the characteristics of the out-of-school youths. It also sought other information such as preferred vocational training needs and the problems usually faced when receiving training. The VNTQ was validated by experts from Vocational and Technology Education Programme at Abubakar Tafawa Balewa University, Bauchi, Nigeria. ©2007 - Journal of Career and Technical Education, Vol. 23, No. 1, Fall, 2007 – Page 62 Method of Data Analysis The data collected was analyzed using frequency and simple percentage to answer the research questions raised. RESULTS AND INTERPRETATION RESULTS AND INTERPRETATION Research Question One What are the Demographic Characteristics of the Out-of-School Youths in Bauchi Metropolis? Table 1: Frequencies and Percentages of Demographic Characteristics of 15 - 25 Years Old Out of-School Youths in Bauchi Metropolis (n = 128) DEMOGRAPHIC CHARACTERISTICS FREQUENCY PERCENTAGE A. Gender (i) Male (ii) Female 79 49 61.79 38.21 B. Age Range (i) 15 – 20 Years (ii) 20 – 25 Years 80 48 62.5 37.5 C. Marital Status (i) Single (ii) Married (iii) Divorced (iv) Widowed 97 22 6 3 75.78 17.19 4.69 2.34 D. Educational Status (i) Attended JSS (ii) Attended SSS 41 87 32.03 67.97 E. Employment Status (i) Trained and Employed (ii) Trained and Unemployed (iii) Untrained and Unemployed 20 27 81 15.63 21.09 63.28 ©2007 J l f C d T h i l Ed i V l 23 N 1 F ll 2007 P 62 Research Question One What are the Demographic Characteristics of the Out-of-School Youths in Bauchi Metropolis? Table 1: Frequencies and Percentages of Demographic Characteristics of 15 - 25 Years Old Out of-School Youths in Bauchi Metropolis (n = 128) DEMOGRAPHIC CHARACTERISTICS FREQUENCY PERCENTAGE A. Gender (i) Male (ii) Female 79 49 61.79 38.21 B. Age Range (i) 15 – 20 Years (ii) 20 – 25 Years 80 48 62.5 37.5 C. Marital Status (i) Single (ii) Married (iii) Divorced (iv) Widowed 97 22 6 3 75.78 17.19 4.69 2.34 D. Educational Status (i) Attended JSS (ii) Attended SSS 41 87 32.03 67.97 E. Employment Status (i) Trained and Employed (ii) Trained and Unemployed (iii) Untrained and Unemployed 20 27 81 15.63 21.09 63.28 Table 1 shows that 61.79% and 38.21 % of the respondents were male and female respectively, 62.5% and 37.5% of the respondents were in the age group of 15 - 20 and 20 - 25 years old. The percentages of single, married, divorced and widowed respondents were 75.78%, 17.19%, 4.69% and 2.34% respectively. 32.03% of the respondents attended Junior Secondary School, while 67.97% of the respondents attended Senior Secondary School. 15.63% of the respondents were trained and employed, 21.09% were trained and unemployed and 63.28% were untrained and unemployed. ©2007 - Journal of Career and Technical Education Vol 23 No 1 Fall 2007 – Page 63 Research Question Two What are the Preferred Vocational Training Needs of 15 - 25 Years old Out-of-School Youths in Bauchi Metropolis? ©2007 - Journal of Career and Technical Education, Vol. 23, No. 1, Fall, 2007 – Page 63 RESULTS AND INTERPRETATION Table 2: Frequencies, Percentages and Ranking of Preferred Vocational Training Needs of 15 - 25 Years old Out-of-School Youths in Bauchi Metropolis (n = 128) S/NO. VOCATIONAL TRAINING AREA FREQUENCY PERCENTAGE RANK 1. Computer Maintenance and Operation Work 24 18.75 1st 2. Tailoring 12 9.38 2nd 3. Electrical Installation and Maintenance work 11 8.59 3rd 4. Furniture Making 8 6.25 4th 5. Further Education 7 5.47 5th 6. Hair Dresser/Barbing 6 4.69 6th 7. Carpentry and Joinery 5 3.91 7th 8. Bricklaying and Concrete work (Mason work) 4 3.13 8th 9. Wood Machining Operation 4 3.13 “ 10. Graphics Art (Artist) 4 3.13 “ 11. Football Playing 4 3.13 “ 12. Catering Services 3 2.34 9th 13. Photograph and Video Coverage 3 2.34 “ 14. Dyeing and Bleaching 3 2.34 “ 15. Office Assistant 3 2.34 “ 16. Knitting 3 2.34 “ 17. Radio, Television and Appliances Repairs 2 1.56 10th 18. Plumbing and Pipe fitting 2 1.56 “ 19. Ceramic making 2 1.56 “ 20. Book keeping 2 1.56 “ 21. Achaba (Commercial Motorcycling Services) 2 1.56 “ 22. VCD Dubbing and Rental Services 2 1.56 “ 23. Poultry farming 2 1.56 “ 24. Automotive Trade 2 1.56 “ 25. Trading 2 1.56 “ 26. Textile trade 1 0.78 11th 27. Typing or Data Processing 1 0.78 “ 28. Store keeping 1 0.78 “ 29. Yoghurt and Ice Cream Making 1 0.78 “ 30. Bakery making 1 0.78 “ 31. Animal Rearing 1 0.78 “ 32. Black marketing 0 0 12th 33. GSM Telephone Services 0 0 “ 34. Shoe making 0 0 “ 35. Blacksmithing 0 0 “ 36. Horticulture and Gardening Services 0 0 “ 37. Soap making 0 0 “ 38. Lotion and Cream making 0 0 “ Table 2, presents the frequencies, percentages and ranking of the responses of the respondents when requested to identify a vocational training area in which they are interested in pursuing from a list of options provided. The youths chose Computer Maintenance and Operation Works (18.75%), followed by Tailoring (9.38%), then Electrical Installation and Maintenance work (8.59%), Furniture Making (6.25%), Further Education (5.47%), Hair Dresser/Barbing (4.69%), and Carpentry and Joinery (3.91 %). Other training needs options attracted percentages from 3.13% and below, indicating that they were not very popular among the youth. ©2007 - Journal of Career and Technical Education, Vol. 23, No. ©2007 - Journal of Career and Technical Education, Vol. 23, No. 1, Fall, 2007 – Page 64 RESULTS AND INTERPRETATION 1, Fall, 2007 – Page 64 Research Question Three How does Gender Influences the Vocational Training Needs of 15 - 25 Years Old Out-of-School Youths in Bauchi Metropolis? Table 3: Frequencies and Percentages of Influence of Gender on Vocational Training Needs of 15 - 25 years Old Out-of-School Youths in Bauchi Metropolis (n = 128) MALE FEMALE S/NO. VOCATIONAL TRAINING AREA FREQ % FREQ % 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. Vocations Dominated by Male Computer Maintenance and Operation work Tailoring Electrical Installation and Maintenance Work Furniture Making Further Education Carpentry and Joinery Wood Machining Operator Graphics Art (Artist) Football Playing Catering Services Photograph and Video Coverage Office Assistant Plumbing and Pipe Fitting Achaba (Commercial Motorcycling Services) 14 8 7 5 5 4 4 4 3 2 2 2 2 2 2 2 10.94 6.25 5.47 3.91 3.91 3.13 3.13 3.13 2.34 1.56 1.56 1.56 1.56 1.56 1.56 1.56 10 4 4 3 2 1 0 0 1 1 1 1 1 0 0 0 7.81 3.13 3.13 2.34 1.56 0.78 0 0 0.78 0.78 0.78 0.78 0 0 0 0 Research Question Three How does Gender Influences the Vocational Training Needs of 15 - 25 Years Old Out-of-School Youths in Bauchi Metropolis? 17. 18. 19. Automotive Trade Trading Textile Trade Store Keeping Yoghurt and Ice Cream Making 1 1 1 0.78 0.78 0.78 0 0 0 0 0 0 SUB-TOTAL 71 55.47 28 21.88 1. 2. 3. 4. 5. 6. Vocations Dominated by Females Hair Dresser/barbing Dyeing and Bleaching Knitting Typing or Data Processing Bakery Making Animal Rearing 0 0 0 0 0 0 0 0 0 0 0 0 6 3 3 1 1 1 4.69 2.34 2.34 0.78 0.78 0.78 SUB-TOTAL 0 0 15 11.72 1. 2. 3. 4. 5. 6. Equal Domination by Sexes Bricklaying and Concrete Work (Mason) Radio, Television and Appliances Repairs Ceramic Making Book Keeping VCD Dubbing and Rental Services Poultry Farming 2 1 1 1 1 1 1.56 0.78 0.78 0.78 0.78 0.78 2 1 1 1 1 1 1.56 0.78 0.78 0.78 0.78 0.78 SUB-TOTAL 7 5.47 7 5.47 1. 2. 3. 4. 5. 6. 7. Research Question Four What are the Problems Associated with the Vocational Training Needs of 15 - 25 Years old Out-of-School Youths in Bauchi Metropolis? ©2007 - Journal of Career and Technical Education, Vol. 23, No. 1, Fall, 2007 – Page 65 RESULTS AND INTERPRETATION Vocations not Chosen by both Sexes Black marketing GSM Telephone Services Shoe Making Blacksmithing Horticulture and Gardening Services Soap Making Lotion and Cream making 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 SUB-TOTAL 0 0 0 0 TOTAL 78 60.94 50 39.06 In Table 3, 50% of the vocations were preferred by males only, while 15.79% of the vocations were preferred by females only. Also 15.79% of the vocations were preferred by both sexes and 18.42% of the vocations were not chosen by both sexes. Research Question Four What are the Problems Associated with the Vocational Training Needs f 15 25 Y ld O t f S h l Y th i B hi M t li ? In Table 3, 50% of the vocations were preferred by males only, while 15.79% of the vocations were preferred by females only. Also 15.79% of the vocations were preferred by both sexes and 18.42% of the vocations were not chosen by both sexes. Table 4: Frequencies and Percentages of Problems Associated with Vocational Training Needs of 15 - 25 Years old Out-of-School Youths in Bauchi Metropolis (n = 128) S/NO. PROBLEMS FACED FREQUENCY PERCENTAGE 1. Lack of Career Guidance 53 41.41 2. Lack of Moral Support 45 35.16 3. Abuse by the Adults 33 25.78 4. Low Income Earning 27 21.09 5. Lack of Financial Support 24 18.75 6. Lack of Involvement in Decision Making 22 17.19 7. Hunger 21 16.41 8. Lack of Employment Security 20 15.63 9. Lack of Recreational Facilities 13 10.16 10. Lack of Health Education 8 6.25 11. Hazard 7 5.47 12. Lack of Legal Support 6 4.69 Table 4: Frequencies and Percentages of Problems Associated with Vocational Training Needs of 15 - 25 Years old Out-of-School Youths in Bauchi Metropolis (n = According to Table 4, the youths had agreed that the major problems associated with their vocational training needs as implied by their percentage response were Lack of Career Guidance (41.41 %), Lack of Moral Support (35.16%), Abuse by the Adults (25.78%), Low Income Earning (21.09%). Others may not be considered as major problems because their percentage responses were below 20%. ©2007 - Journal of Career and Technical Education, Vol. 23, No. 1, Fall, 2007 – Page 66 DISCUSSION AND CONCLUSION Table 1 presents the results of demographic characteristics of 15 - 25 years old out-of-school youths in Bauchi Metropolis. The results indicate higher percentage of male (61.79%) respondent than female (38.21 %) respondents. The low percentage of female respondents could partly be attributed to the low enrolment of females in school in the area of the study. Also the percentage of respondents with age range of 15 - 20 years old was greater than those of 20-25 years old. This may be due to the range of year a student was expected to graduate from either junior or senior secondary school. Because according to NPE revised (2004), it is expected that a child is to be enrolled into primary school when he attained 6 years, that is to say it is expected that by 15 or 18 years, a child must have graduated from either junior or senior secondary school. The results also shows that marital status of most (75.78%) of the respondents are single. This might be due to the nature of environment of the study. Because nowadays parents only give out their son/daughter to marriage when the bride is engage vocationally, that is having a reliable job. Also the educational status from the results shows that most (67.97%) of the respondents attended senior secondary school. This might be attributed to the number of secondary schools that exists in the area of the study, because out of the 22 secondary schools existing in the area, only two terminate at junior secondary school level. The results also shows that most (63.28%) of the respondents are untrained and unemployed. This might be as a result of lack of knowledge of reading, writing and arithmetic ability due to lack of foundation from the scratch. Because Okoro (1993) attributed that vocational education builds on a foundation of good general education. Table 2 present the results of Preferred Vocational Training Needs of 15 - 25 years old out-of- school youths in Bauchi Metropolis. The results indicate a relatively high percentage of the participants selecting Computer Maintenance and Operation Work (18.75%). The high percentage observed with this vocation can be attributed to the obvious fact that computer operation and manipulation covers all facets of life, be it social, political, economical, religious, etc. DISCUSSION AND CONCLUSION That is to say for any organization or individual that wants to achieve maximum output; it must have to incorporate the use of computer. This is further supported by Chimezie (2003) who argues that "if we are to reap the benefits of science, driven industries, we must develop a computer literate society". Considering this statement, it has become very imperative for all concern to foster computer literacy in all facets of our life today. And this will best be achieved through a sound educational system that takes care of the needs of all citizens. Galadima (2003) acknowledged that "education is the prime creator and conveyor of knowledge in any facet. Tailoring (9.38%) was observed to be the next preferred vocational training needs after Computer Maintenance and Operation Work. This preference could be attributed to the fact that tailoring is a vocation that has been in existence for long and will continue to exist because of the continuous demand for services been rendered by this vocation. That is to say any human being who is sane, he/she must wear dresses, which are usually sown by tailors. Okoro (2003) report that our people have acquired invaluable entrepreneurial skills in tailoring and have in fact taken over this business with resounding success and remarkable ease due to its importance. He further reiterated that nowadays some of the garments sewn locally in Nigeria competes favourably with those sown in Europe and Asia. However, the preferred vocational training need next to tailoring was Electrical Installation and Maintenance Work (8.59%). This observed relatively high percentage could partly be attributed to the technological development we have been experiencing because of the way local craft and manual operation workable materials are always turned to be used electronically. For example, grinding stone, boiling pot, manual typewriter, etc are now been converted to be used electronically. And again there has been an influx of many different types of electronic appliances because of the high demand. Furniture Making (6.25%) was the next preferred vocation that followed Electrical Installation and Maintenance. The observed percentage could partly be attributed to the change in life styles and the relatively improved economic standard of the society. The olden ways of furniture decorations in our midst have changed drastically, both in the rural or urban settings. ©2007 - Journal of Career and Technical Education, Vol. 23, No. 1, Fall, 2007 – Page 67 DISCUSSION AND CONCLUSION There is also an increase in the demand for the modern furniture as a result of many marriages that are being contracted almost on weekly basis in the study area and also the raw materials for furniture making are readily available and affordable. Furniture Making was followed by Further Education (5.47%); this was the fifth in the hierarchy of the preferred vocations. This observed percentage might be unconnected with the problem of persistent poor performances of the youths in public examinations such as NECO and W AEC SSCE. Consistently over years less than 10% of the students who finished secondary school in the State proceed to higher institutions. The observed percentage could also be due to the menace of unemployment of skilled manpower and also the poor guidance and counseling services in the secondary schools. Even though, Akubudike (2003) had emphasized that guidance and counseling should be left for the school, because the schools are in better position to spot out students' talents and potentials skills that can be properly tapped. That is to say the youth are aware that opportunities for productive work are growing at a rate far beyond the capabilities of somebody that is educated. Hair Dresser/Barbing (4.69%) was next after Further Education. The observed percentage response could be attributed to the demand for such services as a result of the change in the life style of members of the society. And also the profit derived from such business. Carpentry and Joinery (3.91%) was next after Hair Dresser/Barbing. The observed percentage responses could be attributed to the growing demand for the services rendered by such vocation resulting from the sprang in building projects around. Table 3 present the results of Influence of Gender on the Vocational Training Needs of 15 – 25 years old out-of-school youths in Bauchi Metropolis. The results showed that about half (50%) of the vocations were indicated by males only as need areas. These includes Computer Maintenance and Operation Works; Tailoring; Electrical Installation and Maintenance Work; Furniture Making; Further Education; Carpentry and Joinery; Wood Machining Operation; Graphic Art (Artist); Football Playing; Catering Services; Photograph and Video Coverage; Office Assistant; Plumbing and Pipe Fitting; Achaba; Automotive Trade; Trading; Textile Trade; Store Keeping; and Yoghurt and Ice Cream Making. This observed high percentage response could be attributed to the nature of vocations which are mostly male-oriented in the environment of the study. ©2007 - Journal of Career and Technical Education, Vol. 23, No. 1, Fall, 2007 – Page 68 DISCUSSION AND CONCLUSION These are then followed by vocations indicated by females only as need areas. They include Hair Dresser/Barbing, Dyeing and Bleaching, Knitting, Typing or Data Processing, Bakery Making and Animal Rearing. This observed percentage can be attributed to the fact that these are female-oriented vocations in environment of the study. While the vocational need areas that attract equal response proportion by both sexes, include, Bricklaying and Concrete Work, Radio, Television and Appliances Repairs, Ceramic Making, Book Keeping, VCD Dubbing and Rental Services, and Poultry Farming. One interesting thing about this result was the equal response of both sexes in Bricklaying and Concrete Work, Radio, Television and Appliances Repairs which are usually regarded as male-oriented vocations in the area of the study. Vocational need areas that were not preferred by both sexes includes; Black Marketing; GSM Telephone Services; Shoe Making; Blacksmithing; Horticulture and Gardening Services; Soap Making; and Lotion and Cream Making. Table 4 presents the results of Problems Associated with the Vocational Training Needs of 15 - 25 Years old Out-of-School Youths in Bauchi Metropolis. The results indicate a high percentage in Lack of Career Guidance (41.41 %). The high percentage observed with this problem could be attributed to the experience of most of the respondents during their secondary schools' where the guidance and counseling services are very inadequate. At the age between 15 and 25 years, it is widely believed that for someone to be self-reliant, he/she must be guided appropriately so that at the end, he/she will turn out to be productive to himself and the society at large. Akubudike (2003) has argued that proper guidance and counseling is needed to channel students' interest appropriately. Lack of Career Guidance was followed by Lack of Moral Support ©2007 - Journal of Career and Technical Education, Vol. 23, No. 1, Fall, 2007 – Page 68 (35.16%). This observed percentage response could be attributed to the non-challant attitude shown by some parents towards their children education, especially parents who lack formal education. For example, most of the parents do not care to check on their wards academic performance nor check on what they are really doing after graduation so that they can boost their moral or correct them where necessary. Lack of Moral Support was further followed by Abuse by Adults (25.78%). DISCUSSION AND CONCLUSION This observed percentage response can be attributed to the societal believe that for the younger ones to be of good character, they must show obedience to adults, especially their close relations, even when such obedience would be at their own disadvantage. Considering this situation critically, instead of this believe to yield any positive result, it usually turned out to be a measure problem that makes the younger ones sometimes to be running away from places where they are receiving training. At the end the situation turns out to be a problem to both the youth themselves and the society at large. Abuse by Adults was followed by Low Income Earning (21.09%). This observed percentage can be attributed to the lack of adequate intervention from the government in supplementing the trainees with an allowances to cushion their sufferings because for a trainer to train a trainee and also forfeit part of his profit to the trainee as income is always not possible in our society. Based on the findings of the study, the following conclusions were made: (a) The youth were generally between the ages of 15 - 20 years, (b) Majority of the youths in the metropolis want to pursue careers in Computer Maintenance and Operation Works, (c) About half (19 out of 38) of the vocations considered were preferred by males only, (d) 15.79% (6 out of 38) of the vocations were preferred by females only, and (e) The youths face many problems with their vocational training needs. The major ones were Lack of Career Guidance, Lack of Moral Support and Abuse by Adults. Based on this therefore, it was recommended that Government, especially Bauchi State Government/Individuals (1) should provide vocational training programmes in Computer Maintenance and Operation Works for the 15 - 25 years old out-of-school youths in Bauchi Metropolis, (2) should encourage the establishment of youth clubs (e.g. Farming Club, Computer Literacy Club, etc.) and youth movements in the metropolis in order to serve as an eye opener for their future career opportunities, (3) should provide the necessary leadership, financial and other logistical support to the youths so that they can develop and manage self-sustaining enterprises that can create employment in the metropolis, (4) should provide a means of bridging the gap that exists between schools and industry in the State. ©2007 - Journal of Career and Technical Education, Vol. 23, No. 1, Fall, 2007 – Page 69 Akubudike, J. A. (2003). Achieving the Objectives of Entrepreneurship Education at Secondary School Level: The Problems and Prospects. Nigerian Journal of Curriculum Studies, 10(2), 331 - 334. REFERENCES Abdullahi, J. M. (1993). Preservice Skill Needs for Increasing Employability of Technical College Graduates. Unpublished M.Sc Thesis, University of Nigeria, Nsukka. Abdullahi, J. M. (1993). Preservice Skill Needs for Increasing Employability of Technical College Graduates. Unpublished M.Sc Thesis, University of Nigeria, Nsukka. Akubudike, J. A. (2003). Achieving the Objectives of Entrepreneurship Education at Secondary School Level: The Problems and Prospects. Nigerian Journal of Curriculum Studies, 10(2), 331 - 334. Akubudike, J. A. (2003). Achieving the Objectives of Entrepreneurship Education at Secondary School Level: The Problems and Prospects. Nigerian Journal of Curriculum Studies, 10(2), 331 - 334. Anlezark, A., Karmel, T. and Ong, K. (2006). Have school vocational education and training programs been successful? Adelaide, NCVER. Carnevale, A. P.; Gainer, L. J.; and Meltzer, A. S. (1988). "Workplace Basics: The Skills Employers Want." ERIC Document Reproduction Service, No. ED 299462. Chimezie, O. S. (2003). Computerphobia in Higher Education: How to go About it. Nigerian Journal of Curriculum Studies, 10(2), 293 - 299. Galadima, 1. (2003). Disparity between Expected and Actual Outcomes in the Nigerian Educational System. Nigerian Journal of Curriculum Studies, 10(2), 457 - 460. Harris, R., Rainey, L. and Sumner, R. (2006). Crazy paving or stepping stones? Learning pathways within and between vocational education and training and higher education. Adelaide: NCVER. Hollenbeck, K. (1987). Employer Perception of Vocational Education in South Dakota: Findings and Recommendations. Columbus: The Ohio State University, ERIC Document Reproduction Service, No. ED 304547. Lamb, S. and Vickers, M. (2006). Variations in VET provisions across Australian schools and their effects on student outcomes, Research report 48, Australian Council for Educational Research, Melbourne. National Economic Empowerment and Development Strategy (2005). Abuja: National Planning Commission. National Policy on Education 4th Edition (2004). Lagos: Nigerian Educational Research and Development Council (NERDC). "" Obiefuna, C. A. (2003). The School-Industry Linkage: An Insight into the Technology Curriculum of the Nigerian Universities. Nigerian Journal of Curriculum Studies, 10(2), 255 - 258. Okoro, 1. F. (2003). Curriculum Diversification as a Basis for Entrepreneurship: For Economic Empowerment of the Young School Leavers. Nigerian Journal of Curriculum Studies, 10(2), 312 - 317. Okoro, O. M. (1993). Principles and Methods in Vocational and Technical Education. Nsukka: University Trust Publishers. Olaitan, S. O.; Ali, A.; Eyoh, E. O. and Sowande, K. G. (2000). Research Skills in Education and Social Sciences. Owerri: Cape Publishers International Limited. Oranu, R. N. (1990). ©2007 - Journal of Career and Technical Education, Vol. 23, No. 1, Fall, 2007 – Page 70 ©2007 - Journal of Career and Technical Education, Vol. 23, No. 1, Fall, 2007 – Page 71 REFERENCES The Appraisal of Youth Employment in a Depressed Economy: Emerging Policies and Programmes. Nigerian Vocation Journal, 3, 35 - 62. Seiders, R. W. (1985). Background Paper: FAO's role in support of rural youth programmes and possibilities for the future. Rome: FAO Development Educational Exchange papers. Sherman, R. (2006). Down the track: TAFE outcomes for young people two years on. Adelaide: NCVER. Smith, E. and Green, A. (2005). How workplace experiences while at school affect career pathways. Adelaide, NCVER. Squire, P. J. (2000). A Survey of Vocational Training Needs of 10 -18 Years Old Out-of- School Rural Youths in the North-East District of Botswana. Journal of Vocational Education Research, 25(2). Stanwick, J. (2005). Australian Qualifications Framework lower-level qualifications: Pathways to where for young people? Adelaide: NCVER. Stanwick, J. (2006). Outcomes from higher-level vocational education and training qualifications, Adelaide, NCVER. The College Entrance Examination Board (1984). Academic Preparation for the World of Work. New York: CEEB. UNESCO & ILO (2002). Technical and Vocational Education and Training for the Twenty-first Century. UNESCO and ILO Recommendations http://www.unesco.org/education/ or http://www.ilo.org/skills/ (Retrieved 13th September, 2004). UNESCO (1979). Development of Technical and Vocational Education. A comparative study. Paris UNESCO. United Nations (1985). Youth at the United Nations. Retrieved on 11th April, 2007 from www.un.org/youth. What Employers' Say About Vocational Education in South Carolina (1986). A Study of Vocational Education. "Report #5". Columbia: South Carolina State Council on Vocational and Technical Education. In: I. Susan (Ed). Employers Expectations of Vocational Education. ERIC Digest, No. 90. World Bank (2006). You Think!. But do you know? Retrieved on 11th April, 2007 from http://youthink.worldbank.org/glossary.php. World Bank Policy Paper (1991). Vocational and Technical Education and Training. Washington: International Bank for Reconstruction and Development/World Bank.
https://openalex.org/W2945824577	https://europepmc.org/articles/pmc6529533?pdf=render	English	null	Sensitive Skin in the Genital Area	Frontiers in medicine	2,019	cc-by	13,256	Sensitive Skin in the Genital Area Miranda A. Farage* The Procter & Gamble Company, Mason Business Center, Mason, OH, United States Miranda A. Farage* The Procter & Gamble Company, Mason Business Center, Mason, OH, United States When evaluating sensitive skin, it is common to focus on the face. However, it is becoming increasingly clear that individuals can have different perceptions about the degree of skin sensitivity at distinct anatomic sites. Structural variations in the skin of different body sites can contribute to differences in barrier function, which may contribute to differences in skin sensitivity. Potential triggering factors for skin sensitivity would be expected to vary by body site. For example, the face is exposed to all ambient environmental conditions in the course of daily life, and to several products (e.g., cosmetics for women) and practices (e.g., shaving for men) that may lead to adverse sensations associated with sensitive skin. In contrast, the skin of the genital area is protected from ambient environmental conditions. However, the genital area can be exposed to conditions of higher temperature, different habits and practices and moisture due to the semi-occlusive environment. For some individuals, additional challenges such as incontinence can provide triggering factors for skin sensitivity that effect only the genital area. This article reviews data on the perception of sensitive skin of the genital area, differences based on gender, age, racial differences, and the effects of incontinence on skin sensitivity. The effects of menopause are also considered with regards to sensitive skin perceptions and to emerging differences in biomolecular and physical measures of the urogenital skin. Keywords: self-perceived sensitive skin, genital skin, gender differences, age differences, urogenital skin Edited by: Laurent Misery, Université de Bretagne Occidentale, France INTRODUCTION TO “SENSITIVE SKIN” Reviewed by: Unni Samavedam, University of Cincinnati, United States Adele Zingoni, University of Turin, Italy Individuals with sensitive skin report a variety of unpleasant sensory reactions in response to common external factors and intrinsic stressors (1, 2). Often, the sensory eﬀects that are the hallmark of sensitive skin (such as prickling, burning, tingling, or pain) are not accompanied by erythema or other objective signs of irritation or immunological responses (1). In fact, little correlation exists between individuals’ perceptions of the sensitivity of their skin and demonstrable signs of skin reactivity to irritants (3). A consensus deﬁnition of this condition was published in 2017 as follows. Reviewed by: Unni Samavedam, University of Cincinnati, United States Adele Zingoni, University of Turin, Italy Reviewed by: Unni Samavedam, University of Cincinnati, United States Adele Zingoni, University of Turin, Italy Correspondence: Miranda A. Farage farage.m@pg.com Correspondence: Miranda A. Farage farage.m@pg.com “A syndrome deﬁned by the occurrence of unpleasant sensations (stinging, burning, pain, pruritus, and tingling sensations) in response to stimuli that normally should not provoke such sensations. These unpleasant sensations cannot be explained by lesions attributable to any skin disease. The skin can appear normal or be accompanied by erythema. Sensitive skin can aﬀect all body locations, especially the face” (4). Specialty section: This article was submitted to Dermatology, a section of the journal Frontiers in Medicine Received: 26 February 2019 Accepted: 17 April 2019 Published: 15 May 2019 Citation: Farage MA (2019) Sensitive Skin in the Genital Area. Front. Med. 6:96. doi: 10.3389/fmed.2019.00096 Specialty section: This article was submitted to Dermatology, a section of the journal Frontiers in Medicine Received: 26 February 2019 Accepted: 17 April 2019 Published: 15 May 2019 The pathogenesis of sensitive skin is unknown, but believed to be the product of multiple etiologies, including; deﬁciencies in barrier function, neurosensory dysfunction, compound-speciﬁc irritancy, and cultural inﬂuences (5, 6). REVIEW REVIEW published: 15 May 2019 doi: 10.3389/fmed.2019.00096 Citation: A sizeable proportion of people in the general population in many geographies claim sensitive skin (6, 7). For example, in Europe some degree of skin sensitivity was claimed by 50–90% of responders in several studies in France, (8–10) 75% of responders in Germany, (11) over 50% in Farage MA (2019) Sensitive Skin in the Genital Area. Front. Med. 6:96. doi: 10.3389/fmed.2019.00096 May 2019 \| Volume 6 \| Article 96 1 Frontiers in Medicine \| www.frontiersin.org Sensitive Skin of the Genitalia Farage Farage Italy, (12) and 64% in Greece (13). In the UK, 38% of the men and 51% of the women claimed to have sensitive skin (14). In the US, the prevalence of self-declared sensitive skin has been reported at 44–83% (15–19). In Japan, “very” or “rather” sensitive skin is claimed by 53% of men and 56% of women (20). Sensitive skin was claimed by 57% of the subjects in Korea (21). the skin of diﬀerent body sites can contribute to diﬀerences in barrier function, which may contribute to diﬀerences in skin sensitivity. Also, potential triggering factors for skin sensitivity would be expected to vary by body site. For example, the face is exposed to all ambient environmental conditions in the course of daily life, and to several products (e.g., cosmetics for women) and practices (e.g., shaving for men) that may lead to adverse sensations associated with sensitive skin. In contrast, the skin of the genital area is more protected from ambient environmental conditions, but this anatomic site is almost constantly semi- occluded throughout the day. y j In other geographies, the proportion of the population who perceive they have sensitive skin is lower than in Europe and the United States. In a study conducted in Mexico using 246 subjects self-diagnosed sensitive skin was found in 36% subjects, with a higher prevalence of sensitive skin among subjects with lighter skin phototypes (Type II and III) compared to darker ones (type IV and V) (22). Two survey studies have been reported from China. In a study of 9,154 individuals the prevalence of self- proclaimed sensitive or very sensitive skin has been reported as 9% among men and 16% among women (23). In a study among 408 women in China, 2% claimed they had very sensitive skin, 5% claimed they had moderately sensitive skin, and 16% claimed they had slightly sensitive skin (24). Citation: The structure of the stratum corneum (SC) varies depending on the anatomic site. The skin of the genital area is the only anatomic site other than the face where mucous membranes are exposed to the outside (27). It is highly innervated, vascularized, and has numerous active skin appendages (27). Ya-Xian et al. investigated the thickness of the SC from various body sites (results are illustrated in Figure 1) (28). These investigators found that the smallest number of cell layers was found in genital skin (6 ± 2), followed by skin of the face (9 ± 2), neck (10 ± 2), scalp (12 ± 2), trunk (13 ± 4), and extremities (15 ± 4). The palms and soles showed 47 (±24) cell layers, and the heel showed 86 (±36). These same investigators demonstrated that transepidermal water loss (TEWL), decreased and the thickness of the SC increased (28). Tagami reported that TEWL on facial skin was measured at 10 g/m2/h compared to 7 g/m2/h on the scalp and axilla, and 5 g/m2/h on the trunk and extremities (27). On the thin skin of the female vulva TEWL was measured at 25 g/m2/h (27). The explanation for diﬀerences in prevalence between countries regarding the perception of sensitive skin may be related to some of the underlying physiological causes and environmental triggers for sensitive skin, such as prevailing weather conditions, and fairer vs. darker skin type. Also, it is likely cultural inﬂuences account for some of this diﬀerence. In the study conducted in urban areas in China, Xu et al. hypothesized that some of the participants, especially older individuals, were not familiar with the concept of “sensitive skin” and, therefore, the condition may have been under-reported (23). This hypothesis was supported by the observations that some individuals who did not claim sensitive skin responded that they experienced adverse sensory eﬀects after using cosmetic products. Further, the reported prevalence was inversely proportional to the age group of the responders. The vulvar epithelium exhibits marked regional diﬀerences in structure (29). The cutaneous epithelium of the mons pubis and labia majora exhibit a keratinized, stratiﬁed, squamous structure similar to skin at other sites. However, skin in this area is more hydrated than skin at other body sites and, therefore, more permeable to some materials and more susceptible to friction eﬀects (30). Moving toward the labia minora, the degree of keratinization, and thickness of the epidermis decreases. Citation: The inner third of the labia minora is non-keratinized mucosal tissue (7). The non-keratinized vulvar skin of the labia minora exhibits increased permeability related to the absence of keratin and a loosely packed, less structured lipid barrier (29, 31). In addition, the thinner, inner epithelium represents a shorter distance for penetration of substances (29). Diﬀerences in susceptibility to irritant materials seem to be dependent on the relative permeability of the skin of the vulva to the irritant. In addition, vulvar tissue is highly innervated (32). This would be expected to increase the sensations associated with sensitive skin. The expectations of the general public may also play a role. Manufacturers of consumer products have increasingly marketed products targeted for sensitive skin. Consequently, the public has likely become more aware of this condition. This may partially explain why the proportion of the population that claims sensitive skin appears to be increasing (24). Results of a study conducted in eight European countries are consistent with a cultural component (25, 26). In Portugal, Italy, and Spain, 80–90% of the subjects in the survey population reported at least some skin sensitivity, while in Germany, Belgium, and Switzerland the proportion was just a little more than half. Since the European population is considered to be highly mobile and crossbred, the authors attributed this unexpected ﬁnding to substantially more fashion and beauty-related advertising in speciﬁc European countries (26). UNIQUE FEATURES OF GENITAL SKIN Few studies have probed perceived sensitivity at multiple anatomic sites among the same group of individuals. Saint- Martory et al. reported on a survey questionnaire study conducted in 2004–2005 among 400 individuals in France (1). The face was most often reported as the site of sensitivity (85% When evaluating sensitive skin, it is common to focus on the face. However, it is becoming increasingly clear that individuals can have diﬀerent perceptions about the degree of skin sensitivity at distinct anatomic sites (1, 7, 16). Structural variations in May 2019 \| Volume 6 \| Article 96 Frontiers in Medicine \| www.frontiersin.org 2 Farage Sensitive Skin of the Genitalia Farage FIGURE 1 \| Illustration of stratum corneum thickness at various anatomic sites. This illustration of the relative thickness of stratum corneum at various anatomic sites was adapted from data in Ya-Xian et al. (28). FIGURE 1 \| Illustration of stratum corneum thickness at various anatomic sites. This illustration of the relative thickness of stratum corneum at various anatomic sites was adapted from data in Ya-Xian et al. (28). FIGURE 2 \| Prevalence of perceived sensitive genital skin among ethnic groups. Subjects in the metropolitan area of Cincinnati, Ohio were asked to complete a sensitive skin questionnaire. Participants were asked to describe their skin sensitivity as very, moderately, slightly or not sensitive. Subsequently, participants were asked to describe the skin of speciﬁc anatomic sites including the genital area using the same four-point scale. The percentage of participants claiming any degree of skin sensitivity, and the percentage giving each degree of response (i.e., very, moderately, or slightly) are plotted. Results for ethic groups were compared using a Mantel-Haenszel Chi-Square analysis. [Data adapted from Farage (16)]. FIGURE 1 \| Illustration of stratum corneum thickness at various anatomic sites. This illustration of the relative thickness of stratum corneum at various anatomic sites was adapted from data in Ya-Xian et al. (28). FIGURE 2 \| Prevalence of perceived sensitive genital skin among ethnic groups. Subjects in the metropolitan area of Cincinnati, Ohio were asked to complete a sensitive skin questionnaire. Participants were asked to describe their skin sensitivity as very, moderately, slightly or not sensitive. Subsequently, participants were asked to describe the skin of speciﬁc anatomic sites including the genital area using the same four-point scale. UNIQUE FEATURES OF GENITAL SKIN In a study conducted in 2006 in the metropolitan Cincinnati, Ohio area, 1,039 men and women completed a questionnaire related to their perceptions of sensitive skin (16). Within this group, 77% reported some degree of perceived sensitivity of the face, compared with 61% for the body, and 56% for genital skin (16). Results from these three studies on women subjects on perceptions of sensitive skin in the genital area and sensitive skin in general are presented in Figures 3A,B, respectively. When the results from the Cincinnati and Mississippi studies were compared, the proportion of subjects who claimed some degree of genital skin sensitivity was not signiﬁcantly diﬀerent between the two studies (58.1 and 57.3%, respectively, p = 0.16) (19). A higher proportion of subjects from the Cincinnati study claimed their genital skin was very sensitive (9.4 and 3.4%, respectively, p = 0.05). When asked about sensitive skin in general (Figure 3B), a slightly higher proportion of subjects in the Mississippi study claimed some degree of sensitivity (77.5% compared to 69.1% from the Cincinnati study, p = 0.01) or that the genital skin was very sensitive (11.2% compared to 5.1% from the Cincinnati study, p = 0.02). The prevalence of perceived sensitive skin at the speciﬁc anatomic sites of the face and body were also slightly higher for the Mississippi study (data not shown). We have reported previously that the proportion of subjects who perceive their genital skin as sensitive increases with age (38). The Mississippi population was older than that surveyed in Ohio (mean ages of 45.5 and 35 years, respectively), so the slight but signiﬁcant diﬀerence in perception of sensitive genital skin cannot be explained by an age diﬀerence. Differences Based on Gender and Ethnicity Ethnicity appears to play a role in the perception of sensitive genital skin. A signiﬁcant relationship was found between ethnicity and a perception of sensitive skin in the genital area (Figure 2, p = 0.012). In contrast, in the same study, no signiﬁcant relationships were found between ethnicity and sensitive skin in general, or sensitive skin of the face or body (p = 0.15, p = 0.24, p = 0.13, respectively, data not shown) (16). This is consistent with the ﬁndings of Misery et al. in a study conducted in the US (18). UNIQUE FEATURES OF GENITAL SKIN These investigators noted that the prevalence of sensitive skin in general was similar among ethnic groups varying slightly from 43% for Caucasians to 52% for African-Americans, with no statistically signiﬁcant diﬀerence (p = 0.35). Jourdain et al. conducted a study of perceived sensitive facial skin among a population in San Francisco speciﬁcally selected to include approximately equal numbers of 4 ethnicities (15). These authors found no diﬀerences between the proportions of women in the 4 ethnic groups who perceived they had some degree of sensitive facial skin (Afro-Americans, 52%; Asians, 51%; Euro-Americans, 50%; and Hispanics, 54%). Gender diﬀerences are also found in the perception of sensitive genital skin. In the study presented in Table 1 (metropolitan Cincinnati, Ohio area), a signiﬁcantly higher proportion of women (all ethnicities combined) perceived some degree of sensitive genital skin compared to men (58.1% compared to 44.2%, p = 0.0009) (16). The gender diﬀerence seemed to be driven by the Caucasian subjects who composed the higher proportion of the test population. Among Caucasian subjects 57.0% of women perceived some degree of sensitive genital skin compared to 37.3% of men (p < 0.0001) (35). In contrast, among African-Americans there was no diﬀerence between genders regarding perceived sensitive genital skin (66.7% of women and 65.0% of men, p = 0.84) (35). There is no obvious explanation for why African-American men apparently perceive their genital skin as more sensitive compared to Caucasian men (35). In the study conducted in China (24), the proportion of women who claimed sensitive genital skin was much lower than in both studies in the US at only 6% (Figure 3A). For sensitive skin in general, the prevalence was 23% (Figure 3B). These results are consistent with the observations Xu et al. who reported a lower prevalence of perceived sensitive skin in China compared to Europe and the US (23). In two other studies conducted in Asia the prevalence of sensitive skin was found to be more consistent with western cultures. Kamide et al. conducted a survey of the general population in Japan. Among the 1,500 of responders, “very” or “rather” sensitive skin was claimed by 54.5% of the entire group (56% of the women and 52.8% of the men) (20). In a study conducted in Korea by Kim et al. UNIQUE FEATURES OF GENITAL SKIN The percentage of participants claiming any degree of skin sensitivity, and the percentage giving each degree of response (i.e., very, moderately, or slightly) are plotted. Results for ethic groups were compared using a Mantel-Haenszel Chi-Square analysis. [Data adapted from Farage (16)]. FIGURE 2 \| Prevalence of perceived sensitive genital skin among ethnic groups. Subjects in the metropolitan area of Cincinnati, Ohio were asked to complete a sensitive skin questionnaire. Participants were asked to describe their skin sensitivity as very, moderately, slightly or not sensitive. Subsequently, participants were asked to describe the skin of speciﬁc anatomic sites including the genital area using the same four-point scale. The percentage of participants claiming any degree of skin sensitivity, and the percentage giving each degree of response (i.e., very, moderately, or slightly) are plotted. Results for ethic groups were compared using a Mantel-Haenszel Chi-Square analysis. [Data adapted from Farage (16)]. May 2019 \| Volume 6 \| Article 96 Frontiers in Medicine \| www.frontiersin.org 3 Sensitive Skin of the Genitalia Farage Farage products but were participating in unrelated consumer product studies. A second study was conducted in Mississippi using the same written questionnaire (19). In this study, participants were recruited from local organizations with no selection based on any dermatologic or other criteria and were from a predominantly rural environment. The study population consisted of 89 women with a mean age of 45.5 years. In these two studies, the proportions of African-American and Caucasian subjects were similar. A study using a similar protocol and a translation of the same written questionnaire was conducted in China and included 408 women with a mean age of 39 years (24). of responders). However, other anatomic sites were also reported as sensitive: the hands (58%), scalp (36%), feet (34%), neck (27%), torso (23%), and back (21%), in order of frequency. The prevalence of some degree of perceived sensitive skin of the scalp has been reported as 24% in the UK, (14) and 32–70% in France (33, 34). of responders). However, other anatomic sites were also reported as sensitive: the hands (58%), scalp (36%), feet (34%), neck (27%), torso (23%), and back (21%), in order of frequency. The prevalence of some degree of perceived sensitive skin of the scalp has been reported as 24% in the UK, (14) and 32–70% in France (33, 34). Frontiers in Medicine \| www.frontiersin.org UNIQUE FEATURES OF GENITAL SKIN among a nationwide sample of 1,000 Koreans, “sensitive” of “very sensitive” facial skin was claimed by 56.8% of the responders (59.2% of women and 54.4% of men) (21). The overall prevalence of perceived sensitive skin among women was evaluated in three separate studies using the same survey instrument. A ﬁrst study was conducted in a metropolitan area of the central US (Cincinnati, Ohio), and included 869 women with a mean age of 35.0 years (16, 35–37). Subjects were asked to complete a written questionnaire probing perceptions of sensitive skin. Study participants were not selected based on any criteria related to sensitive skin or hyper-reactivity to consumer The Cincinnati and Mississippi studies provide an opportunity to compare diﬀerent geographic locations within the same country, and to compare a metropolitan vs. rural environment. Neither of these factors appeared to be related to the prevalence of the perception of sensitive genital skin. The two geographic regions have diﬀering climates. Mississippi May 2019 \| Volume 6 \| Article 96 4 Farage Sensitive Skin of the Genitalia Farage FIGURE 3 \| Prevalence of perceived sensitive skin among women. Subjects in three different geographies (urban Ohio, rural Mississippi, and C complete the sensitive skin questionnaire described for Figure 2. Results from the Cincinnati and Mississippi studies were compared using a C Sensitive genital skin, (B) sensitive skin in general. [Data adapted from Farage (16) and Farage et al. (19, 24)]. FIGURE 3 \| Prevalence of perceived sensitive skin among women. Subjects in three different geographies (urban Ohio, rural Mississippi, and China) were asked to complete the sensitive skin questionnaire described for Figure 2. Results from the Cincinnati and Mississippi studies were compared using a Chi-square analysis. (A) Sensitive genital skin, (B) sensitive skin in general. [Data adapted from Farage (16) and Farage et al. (19, 24)]. valence of perceived sensitive skin among women. Subjects in three different geographies (urban Ohio, rural Mississippi, and China) were asked to nsitive skin questionnaire described for Figure 2. Results from the Cincinnati and Mississippi studies were compared using a Chi-square analysis. (A) ki (B) iti ki i l [D t d t d f F (16) d F t l (19 24)] FIGURE 3 \| Prevalence of perceived sensitive skin among women. Subjects in three different geographies (urban Ohio, rural Mississippi, and China) were asked to complete the sensitive skin questionnaire described for Figure 2. UNIQUE FEATURES OF GENITAL SKIN Results from the Cincinnati and Mississippi studies were compared using a Chi-square analysis. (A) Sensitive genital skin, (B) sensitive skin in general. [Data adapted from Farage (16) and Farage et al. (19, 24)]. May 2019 \| Volume 6 \| Article 96 Frontiers in Medicine \| www.frontiersin.org 5 Sensitive Skin of the Genitalia Farage Farage TABLE 1 \| Perceptions of self-declared sensitive genital skin by gender and ethnicity. Women Men Comparison of women vs. men Number Percent Number Percent p-value All ethnicities (N = 1, 032) 869 163 Very sensitive 82 9.4% 6 3.7% Moderately 126 14.5% 13 8.0% Slightly sensitive 297 34.2% 53 32.5% Sensitive (any degree) 505 58.1% 72 44.2% 0.0009 Not sensitive 364 41.9% 91 55.8% Caucasians (N = 802) 684 118 Very sensitive 51 7.5% 3 2.5% Moderately… 101 14.8% 9 7.6% Slightly sensitive 238 34.8% 32 27.1% Sensitive (any degree) 390 57.0% 44 37.3% <0.0001 Not sensitive 294 43.0% 74 62.7% Afro-Americans (N = 128) 108 20 Very sensitive 26 24.1% 1 5.0% Moderately… 20 18.5% 4 20.0% Slightly sensitive 26 24.1% 8 40.0% Sensitive (any degree) 72 66.7% 13 65.0% 0.84 Not sensitive 36 33.3% 7 35.0% Statistical comparisons were conducted for sensitivity (any degree) vs. not sensitive (shown in bold type) using a Mantel-Haenszel chi-square test. Other ethnicities (not shown) were not compared statistically due to the low number of participants. [Data adapted from Farage (35, 36)]. TABLE 1 \| Perceptions of self-declared sensitive genital skin by gender and ethnicity. Statistical comparisons were conducted for sensitivity (any degree) vs. not sensitive (shown in bold type) using a Mantel-Haenszel chi-square test. Other ethnicities (not shown) were not compared statistically due to the low number of participants. [Data adapted from Farage (35, 36)]. increased permeability, and slower wound healing (39–41). Such changes would lead to the conclusion that skin becomes more susceptible to irritation with aging. However, clinical assessments of responses to irritants indicate that older people tend to be less susceptibility to skin irritation compared to younger individuals (42–46). experiences mild winters, but long summers characterized by high temperatures and high levels of humidity. Cincinnati, Ohio, has mild summers but cold winters. However, when asked about perceived environmental conditions that trigger skin reactions, there were no diﬀerences between the two regions with regards to those conditions relevant to sensitive genital skin. UNIQUE FEATURES OF GENITAL SKIN In our survey studies, the questionnaire included lists of external factors (environmental and physiologic conditions) and certain consumer products and asked the responders to indicate if these items ever triggered a skin reaction. In the Cincinnati study (Table 2), a large proportion of the entire study population perceived each of the triggering factors as causing skin reactions on some occasions. However, for the group of individuals claiming some degree of sensitive genital skin the proportion was consistently higher compared those individuals who claimed their skin was not sensitive. Most of the environmental and physiologic conditions (Table 2A) and personal care items (Table 2B) were considered triggering factors by over 50% of the sensitive group, and less than half of the non- sensitive group (36). Dry and cold weather were identiﬁed by the majority of both the sensitive and non-sensitive individuals. For the women in the test group, certain feminine products were considered factors that sometimes trigger skin reactions (Table 2C). Comparison of the sensitive to non-sensitive groups indicated that the diﬀerences were consistently signiﬁcant across all potential triggering factors. In contrast to any changes in the physiological response to irritants with age, the perceptions of general skin sensitivity in western countries do not appear to change with aging. In a phone survey conducted in the US among a nationally representative sample of 994 subjects 45% declared having “sensitive” or “very sensitive” skin (18). There were no signiﬁcant diﬀerences in the prevalence when the data were considered based on age subgroups of 18–24 years, 25–34 years, 35–44 years, 45–54 years, 55–64 years, and ≥65 years. In a survey conducted in the Midwest in a major metropolitan area of the US (Cincinnati, OH), among the 1, 039 subjects, 68% claimed some degree of overall skin sensitivity (16). When subgroups of the responding population were considered, the proportion claiming sensitive skin was 67% for those 30 years and under, 69% for those 31–39 years, 61% for those 40–49 years, and 74% for those 50 years and older. There was no correlation between age and the perception of sensitive skin (p = 0.65). UNIQUE FEATURES OF GENITAL SKIN As mentioned earlier, the survey conducted in China indicated there was a statistically signiﬁcant inverse relationship (p < 0.001) between age and the prevalence of reported sensitive or very sensitive skin: 16% in the youngest group (<25 years), 14% in the middle group (25–49 years), and 10% in the oldest group (≥50 years) (23). Younger age groups may be more Effects of Aging Statistics compared responses of the group claiming some degree of genital sensitivity to those whose genital skin was not sensitive (bold type). Statistical analysis was via a Mantel-Haenszel chi-square test using all levels of perceived irritation frequency (never, sometimes, frequently, or always). These were grouped for presentation in the above table. ND; Not done. [Data on feminine products adapted from Farage (36)]. aware of the concept of sensitive skin, partially due to beauty- related advertising. skin responses in a signiﬁcantly higher proportion of the subjects claiming sensitive genital skin (p ≤0.02). The small number of responses regarding tampons likely reﬂects a much smaller proportion of women who use this product in this age group where a substantial portion of the women would be expected to be post-menopausal. In the Cincinnati study, the prevalence of sensitive skin of the genital area diﬀered signiﬁcantly based on age, increasing from 53.3% in the <30 age group to 66.3% in the >50 age group (p = 0.012) (Figure 4A) (16, 36). Among women, sensitive skin of the genital area was more likely to be declared by women age 50 and older (i.e., 70.2% of the age group) than by women in the other age groups (55.2% among women ≤30, 57.2% among women 31–39, and 61.4% among women 40–49). The association between age and prevalence was signiﬁcant among women (p = 0.012). Among men there was no apparent association between age and perceived sensitive genital skin (p = 0.17) (35). In contrast to the perception of sensitive skin of the genital area, sensitive skin in general does not appear to change with age for either gender (Figure 4B) (16, 36). The diﬀering perceptions among age groups with regards to skin sensitivity of the genital area may be related to speciﬁc changes that may occur as a woman ages, such as the onset of menopause. Findings regarding perceived sensitive skin and aging have been mixed. For example, in a study conducted in the US, Misery et al. concluded that overall sensitivity does not vary with age (18). However, these investigators also reported that perceived sensitivity of the scalp increased with age (34). In a study in France, (47) reported that the prevalence of perceived sensitive skin of the face decreased with age for both women (67% in 35–39 year-old, to 55% in 55–60 year-old) and men (35% in 45–49 year-old, to 29% in 55–60 year-old). Frontiers in Medicine \| www.frontiersin.org Effects of Aging As skin ages, certain physiological changes occur including: reduced epidermal and dermal thickness, reduced hydration, May 2019 \| Volume 6 \| Article 96 Frontiers in Medicine \| www.frontiersin.org 6 Sensitive Skin of the Genitalia Farage Farage TABLE 2 \| Perceptions about factors triggering skin responses among individuals claiming sensitive genital skin. Sensitive genital skin Genital skin not sensitive Difference between groups Comparison of groups Potential triggering factors Total responders % Sensitive to factor Total responders % Sensitive to factor p-value Spearman coefﬁcient (A) ENVIRONMENTAL AND PHYSIOLOGIC CONDITIONS (AMONG MEN AND WOMEN) Rough fabrics 551 74% 408 46% 28% <0.00001 0.32 Hot weather 546 64% 408 43% 22% <0.00001 0.22 Menstrual cycle (women only) 441 61% 305 40% 21% <0.00001 0.22 Stress 548 61% 406 38% 23% <0.00001 0.25 Humid weather 530 45% 406 26% 19% <0.00001 0.20 Dry weather 548 76% 413 63% 13% <0.00001 0.19 Cold weather 553 84% 425 78% 6% 0.0059 0.099 (B) PERSONAL CARE ITEMS (AMONG MEN AND WOMEN) Soaps (bar or liquid) 549 57% 413 11% 46% <0.00001 0.47 Undergarments/clothing 551 68% 421 23% 45% <0.00001 0.46 Perfumes/colognes 351 50% 406 36% 14% <0.00001 0.39 Deodorants/antiperspirants 391 51% 345 21% 30% <0.00001 0.31 Toilet paper 533 31% 409 7% 24% <0.00001 0.30 (C) FEMININE PRODUCTS (AMONG WOMEN ONLY) Menstrual pads 451 59% 315 12% 46% <0.00001 0.45 Feminine wipes 282 43% 225 7% 36% <0.00001 0.41 Douching products 194 35% 179 3% 32% <0.00001 0.40 Panty liners 455 45% 325 8% 37% <0.00001 0.39 Tampons 388 38% 296 4% 34% <0.00001 0.39 In the Cincinnati study responders were given lists of environmental factors and products and asked to indicate if these items ever caused skin irritation (never, sometimes, frequently, or always). Statistics compared responses of the group claiming some degree of genital sensitivity to those whose genital skin was not sensitive (bold type). Statistical analysis was via a Mantel-Haenszel chi-square test using all levels of perceived irritation frequency (never, sometimes, frequently, or always). These were grouped for presentation in the above table. ND; Not done. [Data on feminine products adapted from Farage (36)]. erceptions about factors triggering skin responses among individuals claiming sensitive genital skin. PERSONAL CARE ITEMS (AMONG MEN AND WOMEN) In the Cincinnati study responders were given lists of environmental factors and products and asked to indicate if these items ever caused skin irritation (never, sometimes, frequently, or always). Effects of Aging In a study conducted in Mexico, Hernández-Blanco et al. did not see a trend with regards to the incidence of self-diagnosed sensitive skin and age (22). We have found that sensitive skin of the face and body does not appear to change with age (38). Table 3 presents the perceptions of the age groups regarding feminine products and the association with skin responses. For all feminine products, individuals with sensitive genital skin in the ≤30, 31–39, and 40–49 age groups are more likely to experience skin eﬀects compared to those who do not have sensitive skin (all p < 0.005). For the women in the ≥50 age group, all feminine products except tampons are identiﬁed as triggering Taken as a whole, these results serve as a reminder that sensitive skin continues to be a complex problem involving a complicated interplay of physiological, psychological and cultural factors. As we unravel this phenomenon the importance of understanding diﬀerences between ethnicities, gender, age group and anatomic site is becoming increasingly clear. May 2019 \| Volume 6 \| Article 96 7 Sensitive Skin of the Genitalia Farage Farage TABLE 3 \| Perceptions about feminine products perceived to trigger skin responses among women of different age groups. Effects of Aging All women in age subgroup Sensitive genital skin Genital skin not sensitive Difference between sensitive and non-sensitive Comparison of sensitive and non-sensitive Total responders % Sensitive to factor Total responders % Sensitive to factor Total responses % Sensitive to factor p-value Spearman coefﬁcient ≤30 Menstrual pads 246 36% 137 54% 109 13% 41.2% <0.00001 0.42 Panty liners 240 25% 133 39% 107 7% 32.6% <0.00001 0.42 Tampons 229 24% 127 42% 102 3% 38.8% <0.00001 0.42 Feminine wipes 172 22% 90 39% 82 4% 35.2% <0.00001 0.42 Douching products 120 17% 59 32% 61 2% 30.6% 0.00003 0.41 31–39 Menstrual pads 394 39% 228 58% 166 13% 45.6% <0.00001 0.45 Panty liners 390 31% 223 47% 167 9% 37.6% <0.00001 0.40 Tampons 349 21% 194 35% 155 5% 29.3% <0.00001 0.35 Feminine wipes 220 27% 118 44% 102 8% 36.3% <0.00001 0.40 Douching products 158 20% 74 37% 84 5% 31.7% <0.00001 0.41 40–49 Menstrual pads 72 42% 43 65% 29 7% 58.2% 0.00001 0.56 Panty liners 76 37% 46 52% 30 13% 38.9% 0.0005 0.41 Tampons 64 31% 39 49% 25 4% 44.7% 0.001 0.44 Feminine wipes 54 24% 30 33% 24 13% 20.8% 0.001 0.44 Douching products 43 14% 26 23% 17 0% 23.1% 0.005 0.29 ≥50 Menstrual pads 46 54% 34 68% 12 17% 50.9% 0.001 0.45 Panty liners 59 37% 44 48% 15 7% 41.0% 0.001 0.45 Tampons 27 22% 19 32% 8 0% 31.6% 0.12 0.34 Feminine wipes 53 47% 41 56% 12 17% 39.4% 0.02 0.34 Douching products 43 30% 31 42% 12 0% 41.9% 0.02 0.40 Responders were given lists of personal items and products and asked to indicate if these items ever caused skin irritation. A Mantel-Haenszel chi-square test was used to compare Statistics compared responses of the group claiming some degree of genital sensitivity to those whose genital skin was not sensitive (bold type). Responders were given lists of personal items and products and asked to indicate if these items ever caused skin irritation. A Mantel-Haenszel chi-square test was used to compare Statistics compared responses of the group claiming some degree of genital sensitivity to those whose genital skin was not sensitive (bold type). Effects of Incontinence the subjects was 65.9 (±9.2) and 66.3 (±9.2), respectively. The prevalence of urinary incontinence was 48.4% among women and 25.6% among men. In a study conducted in the UK among 314 randomly selected female patients at a health promotion clinic, the prevalence of incontinence was 53.2% for the entire test population, with an incidence of 34.7% among women 20–29 years of age, and over 50% in women in age groups spanning 40– 79 years of age (56). In contrast, 21,590 male heads of household in the US participated in a survey to determine whether the respondent had symptoms of urinary incontinence (57). Overall, 12.7% reported symptoms during the previous 30 days. The association between urinary incontinence and age was signiﬁcant, with prevalence among the youngest age group (men 18–34 years) at 7.2%, and among the oldest age group (men ≥75 years) at 30.2%. Urinary incontinence is extremely common among women. Reports vary with regards to the precise percentage of the female population who suﬀer from incontinence. In a study conducted in Sweden among 3,071 women, Hagglund et al. reported an overall prevalence of 26%, with a prevalence of 12% among women under 30 years of age (48). In 1988 Jolleys reported an overall prevalence of urinary incontinence of 41% in a survey among 833 women in the UK, (49) whereas Thomas et al. reported that 16.6% of women reported occasional incontinence, (50) and 8.5% reported regular incontinence in a survey of 9,323 women in the London area. The risk of incontinence increases with age (51, 52). In a review of relevant literature, Botlero et al. reported that the prevalence among younger women who had at least one episode of urinary incontinence within the previous year was about 13%, compared to 46% among women in their 50’s and 60’s (53). Brown et al. reported a prevalence of 28% among a cohort of 2,763 participants in survey of post-menopausal women (54). Roberts et al. reported on the results of a community-based study involving 762 women and 778 men (55). The mean age (±SD) of We conducted a study to evaluate perceptions of sensitive skin in women with urinary incontinence compared to a group of age-matched controls (17). The participants included women who suﬀered from light urinary incontinence age 50 and above who participated in focus groups as part of development eﬀorts aimed toward incontinence products. Frontiers in Medicine \| www.frontiersin.org Frontiers in Medicine \| www.frontiersin.org Effects of Incontinence Responses were compared May 2019 \| Volume 6 \| Article 96 8 Farage Sensitive Skin of the Genitalia Farage E 4 \| Perceptions of sensitive skin among women and men in different age groups. Responses in the Cincinnati, Ohio study were evaluated based on age A Mantel-Haenszel chi-square analysis was conducted to determine if increasing age was associated with an increase in the prevalence of sensitive skin (any (A) Sensitive genital skin, (B) sensitive skin in general. [Data adapted from Farage (16, 35)]. FIGURE 4 \| Perceptions of sensitive skin among women and men in different age groups. Responses in the Cincinnati, Ohio study were evaluated based on age group. A Mantel-Haenszel chi-square analysis was conducted to determine if increasing age was associated with an increase in the prevalence of sensitive skin (any degree). (A) Sensitive genital skin, (B) sensitive skin in general. [Data adapted from Farage (16, 35)]. not signiﬁcant (p = 0.08) (Figure 5A). A directionally lower percentage of incontinent subjects described their genital skin as “very” sensitive compared to the control subjects (6.9 and 12.2%, respectively, p = 0.08). It is possible that incontinent individuals may attribute adverse sensory eﬀects or irritation to their incontinent status, rather than to the notion that they have sensitive skin of the genital area (36). With regards to sensitive skin in general (Figure 5B), there was no diﬀerence between the test populations in the proportion claiming some degree of sensitivity (82.8% or incontinent subjects, and 76.2% of to age matched control subjects who do not have incontinence. Results are presented in Figure 5. We expected the incontinent subjects might have an increased perception of sensitive skin in the genital area since these individuals may experience periodic wetness and may wear pads or incontinence products to control wetness. However, such was not the case. A directionally higher proportion of incontinent women reported some degree of sensitive genital skin (very, moderately or slightly) compared to the controls (86.2 and 68.3%, respectively), but the diﬀerence between this group and the age-matched control group was May 2019 \| Volume 6 \| Article 96 Frontiers in Medicine \| www.frontiersin.org 9 Sensitive Skin of the Genitalia Farage TABLE 4 \| Perceptions about skin responses due to relevant external factors among women with incontinence. Effects of Incontinence Incontinent (n = 29) Control (n = 42) p-value Total responses % Sensitive to factor Total responses % Sensitive to factor (A) ENVIRONMENTAL AND PHYSIOLOGIC CONDITIONS Rough fabrics 27 84% 38 81% 0.45 Hot weather 24 87% 34 83% 0.25 Stress 26 58% 36 47% 0.57 Menstrual cycle 8 13% 18 22% 1 Humid weather 27 41% 36 61% 0.38 Dry weather 27 71% 37 72% 0.28 Cold weather 27 100% 34 83% 0.033 (B) PERSONAL CARE ITEMS Soaps (bar or liquid) 28 57% 38 66% 0.72 Undergarments/clothing 28 47% 34 56% 0.81 Perfumes/colognes 22 68% 25 60% 0.71 Deodorants/antiperspirants 21 53% 27 59% 1 Toilet paper 27 37% 33 36% 0.81 (C) FEMININE PRODUCTS Menstrual pads 18 61% 17 53% 0.60 Feminine wipes 23 39% 20 60% 0.19 Douching products 14 28% 17 47% 0.79 Panty liners 24 38% 22 37% 1 Tampons 8 26% 10 20% 0.72 Responders were given lists of environmental factors and products and asked to indicate if these items ever caused skin irritation. A Mantel-Haenszel chi-square test was used to compare responses of the group claiming some degree of genital sensitivity to those whose genital skin was not sensitive. The p-value indicating signiﬁcance is in bold type. [Data adapted from Farage (17)]. TABLE 4 \| Perceptions about skin responses due to relevant external factors among women with incontinence. A) ENVIRONMENTAL AND PHYSIOLOGIC CONDITIONS Responders were given lists of environmental factors and products and asked to indicate if these items ever caused skin irritation. A Mantel-Haenszel chi-square test was used to compare responses of the group claiming some degree of genital sensitivity to those whose genital skin was not sensitive. The p-value indicating signiﬁcance is in bold type. [Data adapted from Farage (17)]. ental factors and products and asked to indicate if these items ever caused skin irritation. A Mantel-Haenszel chi-square test was used to g some degree of genital sensitivity to those whose genital skin was not sensitive. The p-value indicating signiﬁcance is in bold type. [Data skin there were signiﬁcant diﬀerences (p = 0.02), with 62% of premenopausal women claiming sensitivity compared to 54% of perimenopausal and 43% of post-menopausal women. control subjects, p = 0.50). Interestingly, the incontinent women were directionally more inclined to describe their skin in general as “very” or “moderately” sensitive (p = 0.014). There are no apparent reasons for this tendency. Effects of Incontinence We conducted a study to evaluate potential diﬀerences in biomolecular and physical measures of the urogenital skin among women in diﬀerent stages of life (59, 60). Subjects were categorized into three groups of 15 subjects each: the Pre- M group consisted of pre-menopausal women (mean age ± SD = 33.0 ± 6.4); the Post-M Non-HRT group consisted of post-menopausal women who were not receiving any hormone replacement therapy and who exhibited a vaginal atrophy score ≥6 and vaginal pH ≥5 (60.7 ± 3.6); and the Post-M HRT group consisted of post-menopausal women who had been taking some form of hormone replacement therapy for a minimum of 1 year (60.5 ± 3.6). Evaluations were conducted on three genital sites; introitus, labia minora and labia majora. Physical measures at these sites included skin temperature and pH. In addition, sequential tape strips were used to collect material for the quantitative analysis of a variety of biomarkers and cytokines. We also collected information about perceived sensitive skin and urogenital symptoms. A large proportion of both study groups perceived each external factor as causing skin responses on some occasions (Table 4). With regards to environmental and physiologic factors, there were no signiﬁcant diﬀerences between groups except for “cold weather” (Table 4A). Note that “menstrual cycle” was not applicable to most of the individuals in this study since the study population was age 50 and above, therefore, there were few responses reported for this factor. For the personal care items and feminine products (Tables 4B,C, respectively), there were no diﬀerences between incontinent and control groups (17). EFFECTS OF MENOPAUSE: DIFFERENCES IN BIOMOLECULAR AND PHYSICAL MEASURES OF THE UROGENITAL SKIN Falcone et al. reported the results of a digital questionnaire distributed to a population of women aged 20–65 years old (58). Among the 278 responders, 121 were premenopausal, 55 were perimenopausal and 102 were post-menopausal. Sensitive skin of the face was reported by 54% of responders, while sensitive skin of the genital area was reported by only 10.8%. Interestingly, the responses for skin in the genital area did not diﬀer signiﬁcantly for life stage. However, for sensitive facial Group sizes in this study were relatively small, resulting in a low likelihood of signiﬁcant diﬀerences between groups or observations. However, some interesting trends emerged that provide directions for further investigation regarding an understanding of sensitive skin. Figure 6 shows results of the perception of sensitive genital skin, and the presence of subjective symptoms in the three May 2019 \| Volume 6 \| Article 96 Frontiers in Medicine \| www.frontiersin.org 10 Farage Sensitive Skin of the Genitalia Farage FIGURE 5 \| Perceptions of sensitive skin among women with incontinence. The sensitive skin questionnaire was administered to women who suffered from light urinary incontinence age 50 and above. Responses were compared to age matched control subjects who do not have incontinence. The percentage of participants claiming any degree of skin sensitivity, and the percentage giving each degree of response (i.e., very, moderately, or slightly) are plotted. Results were compared for any degree of sensitivity using a Chi-square analysis, and for all three degrees of sensitivity using a Mantel-Haenszel chi-square analysis. (A) Sensitive genital skin, (B) sensitive skin in general. [Data adapted from Farage (17)]. test groups. The proportion of this population who perceived they had sensitive skin of the genital area was smaller in this compared to 58% (shown in Table 1). There is no apparent reason for this diﬀerence; however, it may be related to the FIGURE 5 \| Perceptions of sensitive skin among women with incontinence. The sensitive skin questionnaire was administered to women who suffered from light urinary incontinence age 50 and above. Responses were compared to age matched control subjects who do not have incontinence. The percentage of participants claiming any degree of skin sensitivity, and the percentage giving each degree of response (i.e., very, moderately, or slightly) are plotted. Results were compared for any degree of sensitivity using a Chi-square analysis, and for all three degrees of sensitivity using a Mantel-Haenszel chi-square analysis. EFFECTS OF MENOPAUSE: DIFFERENCES IN BIOMOLECULAR AND PHYSICAL MEASURES OF THE UROGENITAL SKIN For each group, the mean pH for those who did not have sensitive genital skin was compared to those declaring sensitive skin. The number declaring sensitive genital within each group was: Pre-M, 5; Post-M Non-HRT, 4; Post-M HRT, 5. A mixed linear model was used to analyze the pH at different anatomic sites. None of the comparisons were signiﬁcantly different (i.e., p ≤0.05). FIGURE 6 \| Changing perceptions after menopause. A study was conducted to evaluate potential differences in biomolecular and physical measures of the urogenital skin among women in different stages of life. Participants were asked about perceptions of sensitive genital skin, and about speciﬁc subjective symptoms (i.e., external and vaginal dryness, external and vaginal itch, or difﬁculties with intercourse). The groups (15 each) consisted of women who were: pre-menopausal (Pre-M), l h l h (P M N HRT) d l i i h l h (P M HRT) Th FIGURE 6 \| Changing perceptions after menopause. A study was conducted to evaluate potential differences in biomolecular and physical measures of the urogenital skin among women in different stages of life. Participants were asked about perceptions of sensitive genital skin, and about speciﬁc subjective symptoms (i.e., external and vaginal dryness, external and vaginal itch, or difﬁculties with intercourse). The groups (15 each) consisted of women who were: pre-menopausal (Pre-M), post-menopausal on no hormone replacement therapy (Post-M Non-HRT), and post-menopausal receiving hormone replacement therapy (Post-M HRT). The proportion of individuals in each test group claiming any degree of sensitive genital skin or any of the subjective symptoms is plotted. Pairwise comparisons were conducted using Fisher’s exact test. FIGURE 7 \| Comparison of skin pH at genital sites among women with sensitive genital skin. In the study to evaluate biomolecular and physical measures of the urogenital skin, evaluations of skin pH were conducted at the following anatomic sites; vaginally, at the introitus, the labia minora, and the labia majora. For each group, the mean pH for those who did not have sensitive genital skin was compared to those declaring sensitive skin. The number declaring sensitive genital within each group was: Pre-M, 5; Post-M Non-HRT, 4; Post-M HRT, 5. A mixed linear model was used to analyze the pH at different anatomic sites. None of the comparisons were signiﬁcantly different (i.e., p ≤0.05). FIGURE 7 \| Comparison of skin pH at genital sites among women with sensitive genital skin. EFFECTS OF MENOPAUSE: DIFFERENCES IN BIOMOLECULAR AND PHYSICAL MEASURES OF THE UROGENITAL SKIN (A) Sensitive genital skin, (B) sensitive skin in general. [Data adapted from Farage (17)]. FIGURE 5 \| Perceptions of sensitive skin among women with incontinence. The sensitive skin questionnaire was administered to women who suffered from light urinary incontinence age 50 and above. Responses were compared to age matched control subjects who do not have incontinence. The percentage of participants claiming any degree of skin sensitivity, and the percentage giving each degree of response (i.e., very, moderately, or slightly) are plotted. Results were compared for any degree of sensitivity using a Chi-square analysis, and for all three degrees of sensitivity using a Mantel-Haenszel chi-square analysis. (A) Sensitive genital skin, (B) sensitive skin in general. [Data adapted from Farage (17)]. test groups. The proportion of this population who perceived they had sensitive skin of the genital area was smaller in this study compared to previous studies in the same geographic area (Cincinnati, Ohio), i.e., about one third in this study compared to 58% (shown in Table 1). There is no apparent reason for this diﬀerence; however, it may be related to the small sample size (15 individuals per group). As expected, symptoms associated with vulvovaginal atrophy (i.e., dryness, May 2019 \| Volume 6 \| Article 96 Frontiers in Medicine \| www.frontiersin.org 11 Farage Farage Sensitive Skin of the Genitalia FIGURE 6 \| Changing perceptions after menopause. A study was conducted to evaluate potential differences in biomolecular and physical measures of the urogenital skin among women in different stages of life. Participants were asked about perceptions of sensitive genital skin, and about speciﬁc subjective symptoms (i.e., external and vaginal dryness, external and vaginal itch, or difﬁculties with intercourse). The groups (15 each) consisted of women who were: pre-menopausal (Pre-M), post-menopausal on no hormone replacement therapy (Post-M Non-HRT), and post-menopausal receiving hormone replacement therapy (Post-M HRT). The proportion of individuals in each test group claiming any degree of sensitive genital skin or any of the subjective symptoms is plotted. Pairwise comparisons were conducted using Fisher’s exact test. FIGURE 7 \| Comparison of skin pH at genital sites among women with sensitive genital skin. In the study to evaluate biomolecular and physical measures of the urogenital skin, evaluations of skin pH were conducted at the following anatomic sites; vaginally, at the introitus, the labia minora, and the labia majora. EFFECTS OF MENOPAUSE: DIFFERENCES IN BIOMOLECULAR AND PHYSICAL MEASURES OF THE UROGENITAL SKIN In the study to evaluate biomolecular and physical measures of the urogenital skin, evaluations of skin pH were conducted at the following anatomic sites; vaginally, at the introitus, the labia minora, and the labia majora. For each group, the mean pH for those who did not have sensitive genital skin was compared to those declaring sensitive skin. The number declaring sensitive genital within each group was: Pre-M, 5; Post-M Non-HRT, 4; Post-M HRT, 5. A mixed linear model was used to analyze the pH at different anatomic sites. None of the comparisons were signiﬁcantly different (i.e., p ≤0.05). May 2019 \| Volume 6 \| Article 96 Frontiers in Medicine \| www.frontiersin.org 12 Farage Sensitive Skin of the Genitalia Farage FIGURE 8 \| Content of IL-1α, IL-1ra, and the ratio of IL-1ra/IL-1α at genital sites among women with sensitive genital skin. In the study to evaluate biomolecular and physical measures of the urogenital skin, sequential tape strips were used to collect material for the quantitative analysis of a variety of biomarkers and cytokines. Results of analyses of: (A) IL-1α, (B) IL-1ra, and (C) the ratio of IL-1ra/IL-1α are presented for each group. For each group, the mean value for those who did not have sensitive genital skin was compared to those declaring sensitive skin using a mixed linear model. (A) Mean IL-1α, (B) mean IL-1ra, (C) mean of the ratios of IL-1ra/IL-1α. FIGURE 8 \| Content of IL-1α, IL-1ra, and the ratio of IL-1ra/IL-1α at genital sites among women with sensitive genital skin. In the study to evaluate biomolecular and physical measures of the urogenital skin, sequential tape strips were used to collect material for the quantitative analysis of a variety of biomarkers and cytokines. Results of analyses of: (A) IL-1α, (B) IL-1ra, and (C) the ratio of IL-1ra/IL-1α are presented for each group. For each group, the mean value for those who did not have sensitive genital skin was compared to those declaring sensitive skin using a mixed linear model. (A) Mean IL-1α, (B) mean IL-1ra, (C) mean of the ratios of IL-1ra/IL-1α. close to that of the Pre-M group. Frontiers in Medicine \| www.frontiersin.org EFFECTS OF MENOPAUSE: DIFFERENCES IN BIOMOLECULAR AND PHYSICAL MEASURES OF THE UROGENITAL SKIN Results of analyses of the Pre-M and Post-M Non-HRT groups are presented for those who did not have sensitive genital skin compared to those declaring sensitive skin using a mixed linear model. (A) Mean histamine, (B) mean histidine, (C) mean of the ratios of histamine/histidine. FIGURE 9 \| Content of histamine, histidine, and the ratio of histidine/histamine at genital sites among women with sensitive genital skin. In the study to evaluate biomolecular and physical measures of the urogenital skin, histamine, histidine, and the ratio of histidine/histamine were evaluated. Results of analyses of the Pre-M and Post-M Non-HRT groups are presented for those who did not have sensitive genital skin compared to those declaring sensitive skin using a mixed linear model. (A) Mean histamine, (B) mean histidine, (C) mean of the ratios of histamine/histidine. The cytokine IL-1α is produced by epithelial cells, and the normal human epidermis acts as a major reservoir of this material. Regulated cytokine expression is essential to the quality and function of the epidermal barrier, and deregulation of this complex signaling mechanism can result in multiple consequences in skin barrier function (61). The cytokine IL- 1ra functions as a competitive inhibitor to block the response to IL-1α (62). There is evidence that levels of IL-1α and IL-1ra measured in the stratum corneum may be related to inﬂammation. Hirao et al. reported that the stratum corneum of an area of skin unexposed to sunlight, i.e., the inner side of the upper arm, contained more IL-1α than a sun exposed area, i.e., the face. In contrast, the IL-1ra content was reversed, with the sun-exposed area containing higher amounts than the unexposed area (63). The ratio of IL-1ra to IL-1α was over 100 in the sun-exposed area, and only 8 in the unexposed area, leading to the conclusion that IL- 1ra activity was predominant in sun exposed areas, and IL- 1α was predominant in unexposed areas. These same authors reported that the IL-1α content in the unexposed site increased with age, while the content of IL-1ra decreased, resulting in an age-dependent decrease in the IL-1ra/IL-1α ratio (63). In infants suﬀering from diaper rash, Perkins et al. reported a positive correlation between IL-1ra levels recovered from the buttocks and diaper rash severity (64). EFFECTS OF MENOPAUSE: DIFFERENCES IN BIOMOLECULAR AND PHYSICAL MEASURES OF THE UROGENITAL SKIN As mentioned, diﬀerences among this small test population did not reach signiﬁcance, but an interesting area for future research with a larger number of subjects would be an investigation into whether pH is related to sensitive genital skin, and the impact of HRT on the pH of genital tissues other than the vagina. itch and diﬃculties with intercourse) were reported by very few of the Pre-M subjects. Compared to the Pre-M group, a directionally higher proportion of both Post-M groups (HRT and Non-HRT) reported external and vaginal dryness, and diﬃculties with intercourse. The diﬀerences between the Pre-M group and the Post-M HRT group were signiﬁcant for vaginal dryness (p = 0.035) and intercourse diﬃculties (p = 0.006). External itch was reported by a signiﬁcantly higher proportion of the Post-M Non-HRT group compared to both the Pre-M and Post-M HRT groups (p = 0.035 for each comparison). The content of IL-1α, IL-1ra, and the ratio of IL-1ra/IL-1α were evaluated in the study groups (Figure 8) (60). In both post- menopausal study groups, the IL-1α content recovered from tape stripping at the introitus was signiﬁcantly higher in women claiming sensitive genital skin (Post-M Non-HRT, p = 0.002; Post-M HRT, p = 0.004) (Figure 8A). For those claiming sensitive skin in the Post-M Non-HRT group, the IL-1α content was also signiﬁcantly higher at the labia minora (p = 0.01). There were no signiﬁcant diﬀerences in the IL-1ra content or the ratio of IL-1ra/IL-1α when individuals with sensitive genital skin were compared to those without (Figures 8B,C, respectively). The pH was evaluated at 4 anatomic sites; vaginally and at the introitus, the labia minora, and the labia majora (60). Diﬀerences in pH were small (not reaching statistical signiﬁcance), however within each test group, women who claimed sensitive genital skin tended to demonstrate a higher pH vaginally and at the introitus compared to those who did not claim sensitive skin (Figure 7). As expected, HRT appeared to result in a vaginal pH that was May 2019 \| Volume 6 \| Article 96 13 Farage Sensitive Skin of the Genitalia Farage FIGURE 9 \| Content of histamine, histidine, and the ratio of histidine/histamine at genital sites among women with sensitive genital skin. In the study to evaluate biomolecular and physical measures of the urogenital skin, histamine, histidine, and the ratio of histidine/histamine were evaluated. Frontiers in Medicine \| www.frontiersin.org UNIQUE HABITS AND PRACTICES EFFECTING GENITAL SKIN There are several aesthetic practices that can impact the skin of the genital area. Pubic hair grooming, mostly consisting of hair removal, is the most common practice. Recent investigations indicate that well-over half the population in the United States engage in the practice. Table 5 summarizes three studies conducted on grooming practices in the United States. Two were conducted using female panelists (70, 71), and one was conducted among women and men (72). All three studies determined that the most popular methods of pubic hair removal among women is a non-electric razor (∼60–90% of groomers). In the study that included men as subjects it was determined that both non-electric and electric razors were commonly used for pubic hair removal (30.1 and 35.9%, respectively) (72). Histamine is commonly associated with itch in a dose- dependent manner. Generally, we did not see an increase in histamine levels among the Post-M Non-HRT group of women who had a higher number of complaints of itching. This provides an indirect indication that itching experienced by these women may be caused by other biochemical mediators associated with itch, and/or other stimuli, such as dryness (66). Several roles have been identiﬁed for histamine that are related to sexual function (65). Histamine receptors are important in brain areas involved in sexual arousal (67). As a neurotransmitter, histamine levels are related to sexual desire; a decrease in histamine causes a decrease in sexual desire, and an increase causes the reverse (68). Histamine has local eﬀects on smooth muscle and blood vessels critical to physiological sexual arousal (69). In women, this involves an increase in clitoral cavernosal artery inﬂow and an increase in clitoral intracavernous pressure that leads to tumescence and extrusion of the clitoris (68). Engorgement of the genital vascular network increases pressure inside the vaginal capillaries and results in lubrication of the epithelial surface of the vaginal wall (68). Histamine also causes the sexual ﬂush that occurs during arousal. Orgasm is triggered when histamine is released from the mast cells in the genitals. For some women who fail to achieve sexual pleasure and orgasm the problem may be a result of a biochemical imbalance related to histamine and histidine. Further, histamine may be an important All studies reported that the occurrence and frequency of pubic hair grooming is higher among younger individuals (70, 71, 73). In a study of university students conducted in 2015 by Butler et al. EFFECTS OF MENOPAUSE: DIFFERENCES IN BIOMOLECULAR AND PHYSICAL MEASURES OF THE UROGENITAL SKIN The ratio of IL-1ra/IL- 1α for sun-exposed skin (i.e., skin on the face and lower leg) was signiﬁcantly higher (3–6 times, respectively) than skin that was minimally sun-exposed (upper back, underarm, upper leg) (64). Histamine is derived from the decarboxylation of the amino acid histidine (65). An altered ratio of histamine to histidine may indicate a change in the induction of histidine Frontiers in Medicine \| www.frontiersin.org May 2019 \| Volume 6 \| Article 96 14 Sensitive Skin of the Genitalia Farage Farage TABLE 5 \| Pubic hair grooming habits. DeMaria and Berenson (70) Rowen et al. (71) Truesdale et al. (72) Adults 16–40 Adults 18–65 Adults 18–65 Year(s) study conducted 2010–2011 2012–2013 2014 Number of subjects 1,677 women 3,316 women 3,372 women 4,198 men Number of subjects 1,677 3,316 3,372 4,198 Prevalence of grooming 71.7% 83.8% 83.5% 66.5% Method Non-electric razor 92.2% 61.0% 70.5% 30.1% Electric razor Not given 12.0% 14.3% 35.9% Scissors 27.6% 17.5% 17.4% 15.7% Waxing 19.5% 4.6% 6.6% 0.4% Depilatory creams 22.3% 1.2% Not given Not given Laser or electrolysis 13.9% 0.8% 1.1% 0.3% Other Not given 1.7% 3.1% 2.0% Pubic hair dye 2.8% Not given Not given Not given biomarker for genital tissue health regarding blood perfusion and sexual function. biomarker for genital tissue health regarding blood perfusion and sexual function. decarboxylase or a shift in the equilibrium between these two materials. Overall histamine and histidine levels did diﬀer signiﬁcantly when individuals with perceived sensitive genital skin were compared to non-sensitive individuals (Figures 9A,B). Previously we reported that the ratio of histidine to histamine was signiﬁcantly higher at the introitus and labia majora in individuals with perceived sensitive genital skin compared to individuals who were non-sensitive (60). However, after this publication appeared, corrected statistical analyses indicated that, although individuals with sensitive skin tended to have higher ratios of histidine to histamine, the diﬀerences were not signiﬁcant (Figure 9C). Frontiers in Medicine \| www.frontiersin.org CONCLUSION Complications from removing pubic hair are generally minor and include abrasions, itching, cuts, and rash (74, 75). DaMaria et al. reported that a majority (over 90%) of complications related to pubic hair removal occurred among women who had shaved with a razor (75). Sensitive skin is a real phenomenon aﬀecting a large proportion of the population, and it is becoming increasingly clear that individuals can have diﬀerent perceptions about the sensitivity of their skin based on anatomic site. For women, sensitive skin of the genital area can have an adverse impact on daily life and activities. Aging can contribute to the prevalence and symptoms of sensitive skin due to the normal changes that occur to epidermal structure and function. Added to that, aging results in an increased likelihood of incontinence among women, and the inevitable onset of menopause with several changes. Other types of hair removal methods include waxing, depilatory creams, laser hair removal and electrolysis. Several depilatories and home waxing products are formulated speciﬁcally for use on the “bikini line” and not on other areas of the genitalia. Use on areas other than the bikini line, such as the vulva, can lead to irritation. Electrolysis and laser hair removal are newer approaches to permanent hair removal and should only be done by professionals. As the population ages, it will become increasingly important to understand the phenomenon of sensitive skin to develop eﬀective therapies for those who suﬀer from it. We are only beginning to understand the physiological basis for this condition. Painstaking evaluation of the physical and biochemical properties of sensitive skin is a next step in illuminating the mechanisms and causes of this condition. Some individuals resort to dyeing since pubic hair tends to be darker than hair color and grays with age. Home hair coloring products are not formulated for use on the vulva and would likely cause irritation if used for that purpose. A professional colorist and a dye formulated for facial hair are the best choices. Other aesthetic modiﬁcations include piercings and tattoos. Little is published on complications resulting from these practices. Genital piercings can provide an environment for local microﬂora to multiply, potentially increasing sexually transmitted infections (76). Male genital piercings have been associated with complications such as urethral rupture or obstruction, paraphimosis, and scar formation. REFERENCES 10. Misery L, Myon E, Martin N, Consoli S, Nocera T, Taieb C. Sensitive skin: epidemiological approach and impact on quality of life in France. In: Berardesca E, Fluhr JW, Maibach HI, editors. Sensitive Skin Syndrome. New York, NY: Taylor and Francis (2006). p. 181–91. 1. Saint-Martory C, Roguedas-Contios AM, Sibaud V, Degouy A, Schmitt AM, Misery L. Sensitive skin is not limited to the face. Br J Dermatol. (2008) 158:130–3. doi: 10.1111/j.1365-2133.2007.08280.x 1. Saint-Martory C, Roguedas-Contios AM, Sibaud V, Degouy A, Schmitt AM, Misery L. Sensitive skin is not limited to the face. Br J Dermatol. (2008) 158:130–3. doi: 10.1111/j.1365-2133.2007.08280.x 11. Löﬄer H, Dickel H, Kuss O, Diepgen TL, Eﬀendy I. Characteristics of self- estimated enhanced skin susceptibility. Acta Derm Venereol. (2001) 81:343–6. doi: 10.1080/000155501317140052 2. Berardesca E, Fluhr JW, Maibach HI. What is sensitive skin? In: Berardesca E, Fluhr JW, Maibach HI, editors. Sensitive Skin Syndrome. New York, NY: Taylor and Francis (2006). p. 1–6. doi: 10.3109/9781420004601 12. Sparavigna A, Di Pietro A, Setaro M. ‘Healthy skin’: signiﬁcance and results of an Italian study on healthy population with particular regard to ‘sensitive’ skin. Int J Cosmet Sci. (2005) 27:327–31. doi: 10.1111/j.1467-2494.2005. 00287.x 3. Marriott M, Holmes J, Peters L, Cooper K, Rowson M, Basketter DA. The complex problem of sensitive skin. Contact Dermatitis. (2005) 53:93–9. doi: 10.1111/j.0105-1873.2005.00653.x 4. Misery L, Ständer S, Szepietowski JC, Reich A, Wallengren J, Evers AW, et al. Deﬁnition of sensitive skin: an expert position paper from the special interest group on sensitive skin of the international forum for the study of itch. Acta Derm Venereol. (2017) 97:4–6. doi: 10.2340/00015555-2397 13. Farage MA, Bowtell P, Katsarou A. Self-diagnosed sensitive skin in women with clinically diagnosed atopic dermatitis. Clin Med Dermatol. (2008) 2:21–8. 14. Willis CM, Shaw S, De Lacharriere O, Baverel M, Reiche L, Jourdain R, et al. Sensitive skin: an epidemiological study. Br J Dermatol. (2001) 145:258–63. doi: 10.1046/j.1365-2133.2001.04343.x 5. Berardesca E, Farage M, Maibach H. Sensitive skin: an overview. Int J Cosmet Sci. (2013) 35:2–8. doi: 10.1111/j.1468-2494.2012.00754.x 15. Jourdain R, de Lacharriere O, Bastien P, Maibach HI. Ethnic variations in self- perceived sensitive skin: epidemiological survey. Contact Dermatitis. (2002) 46:162–9. doi: 10.1034/j.1600-0536.2002.460307.x 6. Farage MA, Berardesca E, Maibach HI. Sensitive skin: a valid syndrome of multiple origins. In: Wilhelm K-P, Zhai H, Maibach H, editors. I.Marzulli and Maibach’s Dermatotoxicology, Eighth Edition. London, Informa Healthcare, Inc (2012). p. 238–47. doi: 10.3109/9781841848570.031 16. UNIQUE HABITS AND PRACTICES EFFECTING GENITAL SKIN the prevalence was about 95% of women and 85% of men in the younger group (74). Rowen et al. reported that the most common sites of hair removal for women were around the vagina (75.1%), above the vagina (73.9%), and inner thighs (54.2%) (71). For women the most common reason given for hair removal was hygienic purposes (over 50%), followed by attractiveness (31%), partner preference (21%), and ease of oral sex (20%) were also given (71). The majority of male groomers remove the hair above the penis (82–91%, depending on age group), followed by the scrotum (57–71%), penile shaft (51–61%), and inner thigh (25–45%) (73). The majority of men report grooming in preparation for sexual activity (50–72%, depending on age group), followed by hygiene (61%), and routine care (44%) (73). May 2019 \| Volume 6 \| Article 96 15 Farage Sensitive Skin of the Genitalia FUNDING The Procter and Gamble Co. CONCLUSION Female genital piercings may damage condoms or dislodge diaphragms during sexual intercourse, thereby leading to a higher risk of pregnancy. AUTHOR CONTRIBUTIONS The author conﬁrms being the sole contributor of this work and has approved it for publication. REFERENCES (2008) 300:1311–6. doi: 10.1001/jama.300.11.1311 30. Farage MA, Maibach HI. Tissue structure and phyisology of the vulva. In: Farage MA, Maibach HI, editors. The Vulva: Anatomy, Physiology and Pathology, 1st Edition. New York, NY: Informa Healthcare (2006). p. 9–26. doi: 10.1201/9781420005318 53. Botlero R, Urquhart DM, Davis SR, Bell RJ. Prevalence and incidence of urinary incontinence in women: review of the literature and investigation of methodological issues. Int J Urol. (2008) 15:230–4. doi: 10.1111/j.1442-2042.2007.01976.x 31. Farage MA. Vulvar susceptibility to contact irritants and allergens: a review. Arch Gynecol Obstet. (2005) 272:167–72. doi: 10.1007/s00404-005-0732-4 54. Brown JS, Grady D, Ouslander JG, Herzog AR, Varner RE, Posner SF. Prevalence of urinary incontinence and associated risk factors in postmenopausal women. Heart and Estrogen/Progestin Replacement Study (HERS) research group. Obstet Gynecol. (1999) 94:66–70. doi: 10.1097/00006250-199907000-00013 32. Schober J, Cooney T, PfaﬀD, Mayoglou L, Martin-Alguacil N. Innervation of the labia minora of prepubertal girls. J Pediatr Adolesc Gynecol. (2010) 23:352–7. doi: 10.1016/j.jpag.2010.03.009 33. Misery L, Sibaud V, Ambronati M, Macy G, Boussetta S, Taieb C. Sensitive scalp: does this condition exist? An epidemiological study. Contact Dermatitis. (2008) 58:234–8. doi: 10.1111/j.1600-0536.2007.01288.x 55. Roberts RO, Jacobsen SJ, Reilly WT, Pemberton JH, Lieber MM, Talley NJ. Prevalence of combined fecal and urinary incontinence: a community-based study. J Am Geriatr Soc. (1999) 47:837–41. doi: 10.1111/j.1532-5415.1999.tb03841.x 34. Misery L, Rahhali N, Ambonati M, Black D, Saint-Martory C, Schmitt AM, et al. Evaluation of sensitive scalp severity and symptomatology by using a new score. J Eur Acad Dermatol Venereol. (2011) 25:1295–8. doi: 10.1111/j.1468-3083.2010.03968.x 56. Harrison GL, Memel DS. Urinary incontinence in women: its prevalence and its management in a health promotion clinic. Br J Gen Pract. (1994) 44:149–52 35. Farage MA. Does sensitive skin diﬀer between men and women? Cutan Ocul Toxicol. (2010) 29:153–63. doi: 10.3109/15569521003774990 57. Diokno AC, Estanol MV, Ibrahim IA, Balasubramaniam M. Prevalence of urinary incontinence in community dwelling men: a cross sectional nationwide epidemiological survey. Int Urol Nephrol. (2007) 39:129–36. doi: 10.1007/s11255-006-9127-0 36. Farage MA. Perceptions of sensitive skin of the genital area. In: Surber C, Elsner P, Farage MA, editors. Topical Applications and the Mucosa. Basel: Karger (2011). p. 142–54. doi: 10.1159/000321066 58. Falcone D, Richters RJ, Uzunbajakava NE, Van Erp PE, Van De Kerkhof PC. Sensitive skin and the inﬂuence of female hormone ﬂuctuations: results from a cross-sectional digital survey in the Dutch population. Eur J Dermatol. (2017) 27:42–8. doi: 10.1111/ced.13133 37. Farage MA. REFERENCES Farage MA. How do perceptions of sensitive skin diﬀer at diﬀerent anatomical sites? An epidemiological study. Clin Exp Dermatol. (2009) 34:e521–30. doi: 10.1111/j.1365-2230.2009.03487.x 7. Farage MA. Perceived sensitive skin at diﬀerent anatomic sites. In: Honari G, Anderson R, Maibach H, editors. Sensitive Skin Syndrome, 2nd Edition. Boca Raton, FL: CRC Press, Taylor & Francis Group (2017). p. 117–35. doi: 10.1201/9781315121048-22 17. Farage MA. Perceptions of sensitive skin: women with urinary incontinence. Arch Gynecol Obstet. (2009) 280:49–57. doi: 10.1007/s00404-008- 0870-6 18. Misery L, Sibaud V, Merial-Kieny C, Taieb C. Sensitive skin in the American population: prevalence, clinical data, and role of the dermatologist. Int J Dermatol. (2011) 50:961–7. doi: 10.1111/j.1365-4632.2011.04884.x 8. Morizot F, Guinot C, Lopez S, Le Fur I, Tschachler E, Wood C. Sensitive skin: analysis of symptoms, perceived causes and possible mechanisms. Cosmet Toiletries. (2000) 115:83–9. 19. Farage MA, Miller KW, Wippel AM, Berardesca E, Misery L. Sensitive skin in the United states: survey of regional diﬀerences. Family Med Medical Sci Res. (2013) 2:1–8. doi: 10.4172/2327-4972.1000112 9. Misery L, Myon E, Martin N, Verriere F, Nocera T, Taieb C. [Sensitive skin in France: an epidemiological approach]. Ann Dermatol Venereol. (2005) 132:425–9. doi: 10.1016/S0151-9638(05)79303-0 May 2019 \| Volume 6 \| Article 96 Frontiers in Medicine \| www.frontiersin.org 16 Farage Farage Sensitive Skin of the Genitalia 20. Kamide R, Misery L, Perez-Cullell N, Sibaud V, Taieb C. Sensitive skin evaluation in the Japanese population. J Dermatol. (2013) 40:177–81. doi: 10.1111/1346-8138.12027 42. Robinson MK. Age and gender as inﬂuencing factors in skin sensitivity. In: Berardesca E, Fluhr JW, Maibach HI, editors. Sensitive Skin Syndrome. New York, NY: Taylor & Francis (2006). p. 169–80. 21. Kim YR, Cheon HI, Misery L, Taieb C, Lee YW. Sensitive skin in Korean population: an epidemiological approach. Skin Res Technol. (2018) 24:229–34. doi: 10.1111/srt.12418 43. Robinson MK. Population diﬀerences in acute skin irritation responses. Race, sex, age, sensitive skin and repeat subject comparisons. Contact Dermatitis. (2002) 46:86–93. doi: 10.1034/j.1600-0536.2002.460205.x 44. Cua AB, Wilhelm KP, Maibach HI. Cutaneous sodium lauryl sulphate irritation potential: age and regional variability. Br J Dermatol. (1990) 123:607–13. doi: 10.1111/j.1365-2133.1990.tb01477.x 22. Hernández-Blanco D, Castanedo-Cázares JP, Ehnis-Pérez A, Jasso-Ávila I, Conde-Salazar L, Torres-Álvarez B. Prevalence of sensitive skin and its biophysical response in a Mexican population. World J Dermatol. (2013) 2:1–7. doi: 10.5314/wjd.v2.i1.1 45. Grove GL, Duncan S, Kligman AM. Eﬀect of ageing on the blistering of human skin with ammonium hydroxide. Br J Dermatol. REFERENCES (1982) 107:393–400. doi: 10.1111/j.1365-2133.1982.tb00382.x 23. Xu F, Yan S, Wu M, Li F, Sun Q, Lai W, et al. Self-declared sensitive skin in China: a community-based study in three top metropolises. J Eur Acad Dermatol Venereol. (2013) 27:370–5. doi: 10.1111/j.1468-3083.2012.04648.x 46. Lejman E, Stoudemayer T, Grove G, Kligman AM. Age diﬀerences in poison ivy dermatitis. Contact Dermatitis. (1984) 11:163–7. doi: 10.1111/j.1600-0536.1984.tb00964.x 24. Farage MA, Mandl CP, Berardesca E, Maibach HI. Sensitive skin in China. J Cosmet Dermatol Sci Appl. (2012) 2:184–95. doi: 10.4236/jcdsa.2012.23035 j 47. Guinot C, Malvy D, Mauger E, Ezzedine K, Latreille J, Ambroisine L, et al. Self-reported skin sensitivity in a general adult population in France: data of the SU.VI.MAX cohort. J Eur Acad Dermatol Venereol. (2006) 20:380–390. doi: 10.1111/j.1468-3083.2006.01455.x 25. Taieb C, Auges M, Georgescu V, Perez Cullell N, Misery L. Sensitive skin in Brazil and Russia: an epidemiological and comparative approach. Eur J Dermatol. (2014) 24:372–6. doi: 10.1684/ejd.2014.2367 j 26. Misery L, Boussetta S, Nocera T, Perez-Cullell N, Taieb C. Sensitive skin in Europe. J Eur Acad Dermatol Venereol. (2009) 23:376–81. doi: 10.1111/j.1468-3083.2008.03037.x 48. Hagglund D, Olsson H, Leppert J. Urinary incontinence: an unexpected large problem among young females. Results from a population-based study. Fam Pract. (1999) 16:506–9. doi: 10.1093/fampra/16.5.506 27. Tagami H. Location-related diﬀerences in structure and function of the stratum corneum with special emphasis on those of the facial skin. Int J Cosmet Sci. (2008) 30:413–34. doi: 10.1111/j.1468-2494.2008.00459.x 49. Jolleys JV. Reported prevalence of urinary incontinence in women in a general practice. Br Med J. (1988) 296:1300–2. doi: 10.1136/bmj.296.6632.1300 50. Thomas TM, Plymat KR, Blannin J, Meade TW. Prevalence of urinary incontinence. Br Med J. (1980) 281:1243–5. doi: 10.1136/bmj.281.6250.1243 28. Ya-Xian Z, Suetake T, Tagami H. Number of cell layers of the stratum corneum in normal skin - relationship to the anatomical location on the body, age, sex and physical parameters. Arch Dermatol Res. (1999) 291:555–9. doi: 10.1007/s004030050453 51. Farage MA, Aronstein WS, Miller KW, Karram M, Katz M, Hertzman B. A disposable intravaginal device for the management of stress urinary incontinence. Open Women’s Health J. (2011) 5:16–21. doi: 10.2174/1874291201105010016 29. Farage MA, Maibach HI. The vulvar epithelium diﬀers from the skin: implications for cutaneous testing to address topical vulvar exposures. Contact Dermatitis. (2004) 51:201–9. doi: 10.1111/j.0105-1873.2004. 00444.x 52. Nygaard I, Barber MD, Burgio KL, Kenton K, Meikle S, Schaﬀer J, et al. Prevalence of symptomatic pelvic ﬂoor disorders in US women. JAMA. REFERENCES Perceptions of sensitive skin: changes in perceived severity and associations with environmental causes. Contact Dermatitis. (2008) 59:226– 32. doi: 10.1111/j.1600-0536.2008.01398.x 38. Farage MA. Perceptions of sensitive skin with age. In: Farage MA, Miller KW, Maibach HI, editors. Textbook of Aging Skin. Berlin; Heidelberg: Springer- Verlag (2010). p. 1027–46. doi: 10.1007/978-3-540-89656-2_95 59. Farage MA, Wehmeyer K, Fadayel G, Carpenter S, Cheng R, Wang B, et al. Urogenital biomolecular and physical measures in pre- and post-menopausal women. J Clin Gynecol Obstet. (2015) 4:237–250. doi: 10.14740/jcgo347e 39. Farage MA, Miller KW, Elsner P, Maibach HI. Characteristics of the aging skin. Adv Wound Care. (2013) 2:5–10. doi: 10.1089/wound.2011. 0356 60. Farage MA, Cheng R, Maibach MI. Possible correlation between self-reported sensitive skin and physical and chemical biomarkers. J J Expt Derm Res. (2015) 1:10–4. doi: 10.1007/978-3-642-27814-3_64-2 61. Hanel KH, Cornelissen C, Luscher B, Baron JM. Cytokines and the skin barrier. Int J Mol Sci. (2013) 14:6720–45. doi: 10.3390/ijms14046720 40. Jaﬀerany M, Huynh TV, Silverman MA, Zaidi Z. Geriatric dermatoses: a clinical review of skin diseases in an aging population. Int J Dermatol. (2012) 51:509–22. doi: 10.1111/j.1365-4632.2011.05311.x 62. Borg M, Calleja-Agius J. The eﬀect of cytokines on skin during menopause. In: Farage MA, Miller KW, Fugate-Woods, Maibach MI, editors. Skin, Mucosa and Menopause; Management of Clinical Issues. Berlin, Springer-Verlag. (2015). p. 53–70. doi: 10.1007/978-3-662-44080-3_5 41. Farage MA, Miller KW, Elsner P, Maibach HI. Intrinsic and extrinsic factors in skin ageing: a review. Int J Cosmet Sci. (2008) 30:87–95. doi: 10.1111/j.1468-2494.2007.00415.x May 2019 \| Volume 6 \| Article 96 Frontiers in Medicine \| www.frontiersin.org 17 Farage Farage Sensitive Skin of the Genitalia 72. Truesdale MD, Osterberg EC, Gaither TW, Awad MA, Elmer-DeWitt MA, Sutcliﬀe S, et al. Prevalence of pubic hair grooming-related injuries and identiﬁcation of high-risk individuals in the United States. JAMA Dermatol. (2017) 153:1114–21. doi: 10.1001/jamadermatol. 2017.2815 63. Hirao T, Aoki H, Yoshida T, Sato Y, Kamoda H. Elevation of interleukin 1 receptor antagonist in the stratum corneum of sun-exposed and ultraviolet B-irradiated human skin. J Invest Dermatol. (1996) 106:1102–7. doi: 10.1111/1523-1747.ep12340143 64. Perkins MA, Osterhues MA, Farage MA, Robinson MK. A noninvasive method to assess skin irritation and compromised skin conditions using simple tape adsorption of molecular markers of inﬂammation. Skin Res Technol. (2001) 7:227–37. doi: 10.1034/j.1600-0846.2001.70405.x 73. Gaither TW, Awad MA, Osterberg EC, Rowen TS, Shindel AW, Breyer BN. Prevalence and motivation: pubic hair grooming among men in the United States. REFERENCES Am J Mens Health. (2017) 11:620–40. doi: 10.1177/1557988316661315 65. Lieberman P. The basics of histamine biology. Ann Allergy Asthma Immunol. (2011) 106:S2–5. doi: 10.1016/j.anai.2010.08.005 74. Butler SM, Smith NK, Collazo E, Caltabiano L, Herbenick D. Pubic hair preferences, reasons for removal, and associated genital symptoms: comparisons between men and women. J Sex Med. (2015) 12:48–58. doi: 10.1111/jsm.12763 66. Stander S, SteinhoﬀM, Schmelz M, Weisshaar E, Metze D, Luger T. Neurophysiology of pruritus: cutaneous elicitation of itch. Arch Dermatol. (2003) 139:1463–70. doi: 10.1001/archderm.139.11.1463 75. DeMaria AL, Flores M, Hirth JM, Berenson AB. Complications related to pubic hair removal. Am J Obstet Gynecol. (2014) 210:528.e1–5. doi: 10.1016/j.ajog.2014.01.036 67. Devidze N, Lee AW, Zhou J, PfaﬀDW. CNS arousal mechanisms bearing on sex and other biologically regulated behaviors. Physiol Behav. (2006) 88:283–93. doi: 10.1016/j.physbeh.2006.05.030 76. Lee B, Vangipuram R, Petersen E, Tyring SK. Complications associated with intimate body piercings. Dermatol Online J. (2018) 24. 68. Meston CM, Frohlich PF. The neurobiology of sexual function. Arch Gen Psychiatry. (2000) 57:1012–30. doi: 10.1001/archpsyc.57.11.1012 Frontiers in Medicine \| www.frontiersin.org Conﬂict of Interest Statement: MF is an employee of The Procter and Gamble Company. 69. Adaikan PG, Karim SM. Male sexual dysfunction during treatment with cimetidine. Br Med J. (1979) 1:1282–3. doi: 10.1136/bmj.1.6173.1282-d Conﬂict of Interest Statement: MF is an employee of The Procter and Gamble Company. 70. DeMaria AL, Berenson AB. Prevalence and correlates of pubic hair grooming among low-income Hispanic, Black, and White women. Body Image. (2013) 10:226–31. doi: 10.1016/j.bodyim.2013.01.002 Copyright © 2019 Farage. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. 71. Rowen TS, Gaither TW, Awad MA, Osterberg EC, Shindel AW, Breyer BN. Pubic hair grooming prevalence and motivation among women in the United States. JAMA Dermatol. (2016) 152:1106–13. doi: 10.1001/jamadermatol.2016.2154 May 2019 \| Volume 6 \| Article 96 Frontiers in Medicine \| www.frontiersin.org 18
https://openalex.org/W2104787654	https://jfootankleres.biomedcentral.com/counter/pdf/10.1186/s13047-014-0050-9	English	null	Clinical outcomes following ankle fracture: a cross-sectional observational study	Journal of foot and ankle research	2,014	cc-by	6,241	RESEARCH Open Access JOURNAL OF FOOT AND ANKLE RESEARCH JOURNAL OF FOOT AND ANKLE RESEARCH Segal et al. Journal of Foot and Ankle Research 2014, 7:50 http://www.jfootankleres.com/content/7/1/50 JOURNAL OF FOOT AND ANKLE RESEARCH JOURNAL OF FOOT AND ANKLE RESEARCH © 2014 Segal et al.; licensee BioMed Central. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. * Correspondence: ganitm@apostherapy.com 1AposTherapy Research Group, 1 Abba Even Blvd, Herzliya 46733, Israel Full list of author information is available at the end of the article Clinical outcomes following ankle fracture: a cross-sectional observational study Ganit Segal1*, Avi Elbaz1, Alon Parsi1, Ziv Heller2, Ezequiel Palmanovich2, Meir Nyska2, Zeev Feldbrin3 and Benjamin Kish2 Abstract Background: The purpose of the current study was to examine objective and subjective differences between three severity groups of ankle fractures patients compared to healthy controls. Methods: This was a case-controlled study. 92 patients with an ankle fracture injury of which 41 patients were eligible to participate in the study. 72 healthy people served as controls. All patients underwent a computerized gait test, completed self-assessment questionnaires (The Foot and Ankle Outcome Score (FAOS) and the SF-36), evaluated with the American Foot and Ankle Score (AOFAS) form and completed the 6-min walk test. The control group performed a computerized gait test and completed the SF-36 health survey. Results: All ankle fracture patients presented compromised gait patterns and limb symmetry compared to controls (p < 0.05). Ankle fracture patients also had lower SF-36 scores compared to controls (p < 0.05). Significant differences were found between the unimalleolar group compared to the bimalleolar and trimalleolar groups in most parameters, except for the FAOS scores. There were no significant differences between the bimalleolar fracture group and the trimalleolar fracture groups. Conclusions: Although all fracture severity classification groups presented a compromised gait pattern and worse clinical symptoms compared to controls, it seems that patients with a unimalleolar fracture is a different group compared to bimalleolar and trimalleolar fracture. Furthermore, it seems that bimalleolar fracture and trimalleolar fracture affect the gait pattern and clinical symptoms to an equal extent, at least in the short-term. Trial registration: NCT01127776. Trial registration: NCT01127776. Keywords: Ankle fracture severity, Gait, Clinical outcomes Keywords: Ankle fracture severity, Gait, Clinical outcomes Some concluded that a fracture severity classification is a consistent predictor of functional outcome following surgery [8-10]. However, recent work by Egol et al. con- cluded that the type of fracture had no influence on func- tional recovery [11]. Most studies used self-assessment questionnaires and functional scores to evaluate the func- tional status of the patient post an ankle fracture surgery. Although questionnaires are considered a valid method of assessment, they are subjective, and objective methods of evaluation are warranted. Segal et al. Journal of Foot and Ankle Research 2014, 7:50 http://www.jfootankleres.com/content/7/1/50 Page 2 of 7 All patients were informed of the study procedure in- cluding its purpose, protocol and any known risks and were asked to sign a consent form that was approved by the ethics committee. [20]. To the best of our knowledge, there is limited infor- mation regarding the changes in gait patterns following ankle fracture. We found one study by Becker et al. who evaluated gait symmetry post ankle fracture surgery. They concluded that after 18 months of surgery, gait symmetry in plantar pressure distribution was achieved. They did not find a difference between fracture severity groups [16]. Their study was done on a young population and months post-surgery. The purpose of the current study was to examine objective (gait analysis) and subjective (pain, function and quality of life) differences between three se- verity groups of ankle fractures patients immediately with weight-bearing allowance compared to healthy controls. Gait analysis A computerized mat was used to measure spatiotempo- ral gait parameters (GAITRite® system, CIR Systems Inc. Peekskill, NY, USA) [22,23]. During the gait test, all pa- tients walked barefoot at a self-selected speed. Patients walked 3 meters before and after the walkway mat to allow sufficient acceleration and deceleration time out- side the measurement area. Each gait test included 6 walks and the mean value of the 6 walks was calculated for each of the following parameters: velocity (m/s); step length (cm); single limb support (SLS) phase (% Gait cycle). involved ‐ uninvolved involved þ uninvolved ð Þ=2 100 involved ‐ uninvolved involved þ uninvolved ð Þ=2 100 A symmetry index value of zero represents perfect symmetry and up to 5% difference between limbs was considered normal [24]. Patients Ninety-two patients with acute ankle fracture were re- ferred to the study between December 2010 and August 2013. All patients were treated operatively with open re- duction and internal fixation according to AO/ASIF methods [21], and were instructed to avoid weight- bearing for 6 weeks. Patients who had a syndesmosis in- jury were treated with a 3.5 mm 4 cortex screw, which was removed 3 months post-surgery. Patients were re- cruited to the study during their follow-up examination with the orthopedic surgeon, who offered them to join the study. Patients were contacted by the research team, and once weight-bearing was allowed, patients came to a therapy center for a first assessment. Exclusion criteria were: additional injury apart from the ankle fracture, other musculoskeletal disorder, neurological problems, any condition that prevents the patients from performing a gait analysis test or complete self-assessment question- naires. Out of the ninety-two patients, 41 participated. Main reason for not entering the study was that patients did not want to commit to a long-term follow-up study. Other reasons were pregnancy, vision problem, recruiting military service, and seeking other medical intervention. Background Ankle fractures are one of the most common injuries of the lower limb [1]. There has been a constant increase in ankle fracture rates amongst young, active patients as well as in the elderly population over the last several de- cades [2,3]. Operative treatment of ankle fracture includes open reduction and internal fixation [4], followed by immobilization and rehabilitation [5-7]. Ankle fractures severity can be defined and classified to three sub-groups including unimalleolar, bimalleolar and trimalleolar fractures. Several studies have examined the differences between severity groups in regard to functional outcomes and showed conflicting results. Gait analysis is widely used to characterize functional performance of different populations [12-15]. It is also used as an outcome measure for decision making and for evaluating different treatments [16-19]. Recently, func- tional severity classification for patients with knee osteo- arthritis, which is based on gait analysis, was presented Segal et al. Journal of Foot and Ankle Research 2014, 7:50 http://www.jfootankleres.com/content/7/1/50 Segal et al. Journal of Foot and Ankle Research 2014, 7:50 http://www.jfootankleres.com/content/7/1/50 Methods This study is part of a wider, double blind, randomized controlled trial examining the effect of a new rehabilitation intervention for patients following ankle fracture. Ethic approval was obtained from by the Helsinki Committee of the participating medical center. The trial registra- tion number is NCT01127776. Temporal distance (T-D) symmetry was calculated for SLS and step length using the formula: involved ‐ uninvolved involved þ uninvolved ð Þ=2 100 Questionnaires P i Patients were asked to complete two self-assessment questionnaires. The Foot and Ankle Outcome Score (FAOS) was used to evaluate patients symptoms [25]. This questionnaire comprises 42 Likert scale questions. Five sub-categories are calculated including symptoms (7 questions), pain (9 questions), function (17 questions), sport performance (5 questions) and quality of life (4 ques- tions). Results range from 0 to 100. A score of 0 indicates a poor score and a score of 100 indicates a best score. The Short Form (SF)-36 Health Survey (SF-36) was used to evaluate quality of life [26]. Posner et al. have concluded that the SF-36 is a useful tool to assess out- comes post ankle fracture [27]. The SF-36 is scored be- tween 0 and 100, with 0 indicating the worst quality of life and 100 indicating the best quality of life. The question- naire contains 36 questions of which 8 sub-categories are calculated: physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, mental health. Furthermore, 2 summarizing scores, physical health score (PHC) and a mental health score (MHS) are also calculated. Seventy-two healthy volunteers served as controls. This group was part of a larger database of healthy indi- viduals that was collected by the current study re- searchers, at the study’s treatment site which is a private therapy center. Patients were healthy employees, care- givers and family members. This group was matched for age, gender and body mass index (BMI) and included healthy people without a history of any musculoskeletal problems and neurological problems. The American Orthopaedic Foot and Ankle Score (AOFAS) was used to evaluate the clinical condition of the patients [28]. This questionnaire was completed by trained physiotherapist. The AOFAS is scored between 0 and 100, with 0 indicating worst clinical condition and 100 indicating best clinical condition. Page 3 of 7 Segal et al. Journal of Foot and Ankle Research 2014, 7:50 http://www.jfootankleres.com/content/7/1/50 of the patients had both a deltoid ligament injury and a syndesmosis injury, none had a deltoid ligament injury, and 4 patients (33%) had syndesmosis damage. In the bimalleolar fracture group one patient (6%) had both a deltoid ligament and syndesmosis damage, 4 patients (26%) had syndesmosis damage and none of the patients had solely a deltoid ligament injury. Results Forty-one patients with an ankle fracture met inclusion criteria. There mean age (sd) was 47.3 (14.6) years. 12 patients (29%) had unimalleolar fracture, 15 patients (37%) had bimalleolar fracture and 14 patients (34%) had trimalleolar fracture. Furthermore, patients were also classified with deltoid ligament injury, syndesmosis injury or both. In the unimalleolar fracture group none Questionnaires P i In the trimalleolar fracture group two patients (14%) had both a deltoid ligament and syndesmosis damage, 6 patients (42%) had syndesmosis damage and none of the patients had solely a deltoid ligament injury. There were no significant dif- ferences between groups in regard to prevalence of soft tissue injury (p = 0.328, p = 0.640 and p = 0.654 for del- toid ligament injury, syndesmosis injury or a combin- ation of both, respectively). In an additional analysis we examined the differences in all measured variables be- tween patients with syndesmosis injury and patients without syndesmosis injury and another comparison be- tween patients with deltoid ligament injury and patients without and did not find significant differences between groups. Data collection ll kl f All Ankle fracture patients came to a private therapy center and underwent the following assessments: the medical records of the patients were scanned and saved in the patient’s file; anthropometric measures of height, weight and age were captured. Each patient was asked to perform a gait analysis test and to complete the two self-assessment questionnaires. A trained physiotherap- ist completed the AOFAS and measured the ankle joint range of motion (ROM) in the sagittal (plantar flexion/ dorsiflexion) and coronal plane (inversion/eversion). In addition, patients were asked to complete the 6-min walk test. All patients were instructed to refrain from taking pain medications, including paracetamol and non-steroidal anti-inflammatory drugs, for a period of 3 days prior to the clinical and gait evaluation. There were no significant differences between ankle fracture severity groups in regard to age, gender distri- bution and BMI. Furthermore, there were no significant differences between ankle severity groups in the time frame between injury and first assessment. Mean time (sd) of assessment for the unimalleolar, bimalleolar and trimalleolar groups were 66.1 (28.2), 68.1 (15.6) and 64.5 (17.4) days, respectively. Seventy-two healthy people matched for age, BMI and gender distribution served as controls. There mean (sd) age was 47.1 (17.7) years. Pa- tients’ characteristics are summarized in Table 1. The healthy population had only spatiotemporal gait analysis and SF-36 self-assessment questionnaire. Statistical analysis Data were analyzed with IBM SPSS software version 21.0 and were presented as frequencies and percentages for categorical variables and as mean and standard devi- ation for all gait spatiotemporal parameters and self- evaluation questionnaires, followed by 95% confidence interval for the mean. Non-parametric one-sample Kolmogorov-Smirnov tests were calculated to compare the observed cumulative distribution function for the continuous variables with the Normal theoretical distri- bution. To demonstrate the differences in spatiotempo- ral gait parameters and self-evaluation questionnaires within the study groups, one-way ANOVA tests were performed. Once significant differences between groups’ means were determined, post hoc tests and pairwise multiple comparisons (Bonferroni test) were performed to determine which means differ. There were no significant differences between ankle fracture groups in ankle ROM in the sagittal plane and in the frontal plane, except for plantarflexion angle of the op- erated leg. Mean (sd) dorsiflexion angle of the operated leg was 0.8 (6.7) degrees, −3.3 (6.2) degrees and −0.8 (7.6) degrees for the unimalleolar, bimalleolar and trimalleolar groups respectively (p = 0.365). Mean (sd) plantarflexion angle of the operated leg was 46.7 (5.8) degrees, 44.6 (8.4) degrees and 40.6 (7.5) degrees for the unimalleolar, bimalleolar and trimalleolar groups respectively (p = 0.014). Mean (sd) inversion angle of the operated leg was 8.8 (3.8) degrees, 6.7 (3.7) degrees and 5.6 (3.6) degrees for the unimalleolar, bimalleolar and trimalleolar groups respect- ively (p = 0.058). Mean (sd) eversion angle of the operated leg was −0.4 (3.3) degrees, 0.0 (6.0) degrees and 2.5 (4.0) degrees for the unimalleolar, bimalleolar and trimalleolar groups respectively (0.893). Functional test All patients performed the 6-min walk test which is a valid test that evaluates the functional status of the pa- tient [29]. Patient is asked to walk the longest possible distance during 6 minutes and the total distance is mea- sured. A low score indicate a short walking distance (i.e. worse functional status), whereas a high score indicate longer walking distance (i.e. better functional status). Gait analysis f d Significant differences were found between groups in all gait parameters including gait velocity, involved and un- involved step length, involved and uninvolved SLS. All Segal et al. Journal of Foot and Ankle Research 2014, 7:50 http://www.jfootankleres.com/content/7/1/50 Segal et al. Journal of Foot and Ankle Research 2014, 7:50 http://www.jfootankleres.com/content/7/1/50 Page 4 of 7 Table 1 Patients characteristics Unimalleolar Bimalleolar Trimalleolar Healthy controls p N (F/M) 12 (5/7) 15 (10/5) 14 (9/5) 72 (36/36) 0.437 Age (years) 50.3 (13.2) 47.9 (16.3) 44.1 (14.2) 47.1 (17.7) 0.815 Height 171.5 (7.2) 165.0 (6.2) 169.8 (9.5) 168.6 (9.4) 0.335 Weight 79.8 (16.8) 71.6 (13.2) 80.3 (15.7) 73.1 (14.1) 0.199 BMI (kg/m2) 27.0 (4.5) 26.7 (5.0) 28.1 (6.3) 25.5 (3.4) 0.185 Time from injury (days) 70.5 (26.0) 68.1 (15.6) 64.5 (17.4) - 0.812 Table 1 Patients characteristics three fracture severity groups were significantly below the normal range. Patients with unimalleolar ankle frac- ture walked significantly faster compared to bimalleolar fracture (p = .016), but not compared to trimalleolar fracture (p = .239). They also had a significant longer step length in the uninvolved leg compared to the bimal- leolar and trimalleolar groups (p = .002 and p = .041, re- spectively). Finally, the unimalleolar ankle fracture group had significant higher SLS values in the involved leg compared to both the bimalleolar group and trimalleolar group (p = .002 and p = .009, respectively). There were no significant differences in all gait parameters between the bimalleolar ankle fracture group and the trimalleolar ankle fracture group. Table 2 summarizes the differences in gait patterns between groups. three fracture severity groups were significantly below the normal range. Patients with unimalleolar ankle frac- ture walked significantly faster compared to bimalleolar fracture (p = .016), but not compared to trimalleolar fracture (p = .239). They also had a significant longer step length in the uninvolved leg compared to the bimal- leolar and trimalleolar groups (p = .002 and p = .041, re- spectively). Finally, the unimalleolar ankle fracture group had significant higher SLS values in the involved leg compared to both the bimalleolar group and trimalleolar group (p = .002 and p = .009, respectively). There were no significant differences in all gait parameters between the bimalleolar ankle fracture group and the trimalleolar ankle fracture group. Table 2 summarizes the differences in gait patterns between groups. all 8 sub-scale categories and in the summerising scores (PHS and MHS). Functional test Significant differences were found between ankle frac- ture groups in the 6-min walk test (p = .024). As fracture severity increased the walking distance of the patient de- creased. Patients with unimalleolar fracture walked a mean ± sd distance of 376.3 ± 136.0 m, patients with bimalleolar fracture walked a mean ± sd distance of 250.3 ± 145.4 m and patients with trimalleolar fracture walked a mean ± sd distance of 201.2 ± 179.1. Significant Gait analysis f d There were no significant differences between ankle fracture groups in all SF-36 sub-scale categories. Figure 1 illustrates the differences between ankle fracture groups and healthy controls. The AOFAS clinical assessment form was statistically different between groups (p = .017). Patients with unim- alleolar fracture had a mean ± sd score of 59.0 ± 18.0, pa- tients with bimalleolar fracture had a mean ± sd score of 37.7 ± 15.4 and patients with trimalleolar fracture had a mean ± sd score of 40.4 ± 20.0. Significant differences were found between the unimalleolar fracture group and the bimalleolar and trimalleolar fracture groups (p = .011 and p = .034, respectively). There were no significant dif- ferences between the bimalleolar fracture group and the trimalleolar fracture group (p =1.000). An additional analysis of the gait patterns examined limb symmetry. Significant asymmetry was found in step length and SLS in all fracture groups but not in healthy controls. Step length asymmetry was 162% (p = .002), 119% (p = .003), and 131% (p = .001) for unimalleolar, bimalleolar and trimalleolar respectively. SLS asymmetry was 63% (p = .004), 256% (p = .001) and 189% (p = .001) for unimalleolar, bimalleolar and trimalleolar, respectively. There were no significant differences between fracture groups in FAOS total score and its 5 sub-scale categories. Abbreviations: SL step length, SLS single limb support, GC gait cycle. Significance was set to p < 0.05. SLS single limb support, GC gait cycle. Significance was set to p < 0.05. Questionnaires SF-36 health survey scores were significantly different between groups. Significant differences were found be- tween all 3 ankle fracture groups and healthy controls in Table 2 Differences in gait patterns between ankle fracture groups and controls Unimalleolar Bimalleolar Trimalleolar Healthy controls p Velocity (cm/s) 74.8 (29.0) 48.2 (21.6) 56.7 (23.8) 118.0 (20.2) p < 0.001 [56.4-93.2] [36.2-60.1] [42.9-70.4] [113.3-122.8] Involved SL (cm) 61.6 (11.5) 41.6 (12.9) 48.1 (13.5) 62.9 (8.0) p < 0.001 [44.3-59.0] [34.5-48.8] [40.3-55.9] [61.1-64.8] Uninvolved SL (cm) 43.0 (14.7) 26.2 (12.4) 30.4 (13.9) 62.8 (8.3) p < 0.001 [33.6-52.3] [19.4-33.1] [22.4-38.5] [60.9-64.8] Involved SLS (% GC) 29.6 (6.8) 21.6 (6.2) 22.5 (7.6) 39.9 (1.5) p < 0.001 [25.3-33.9] [18.2-25.0] [18.1-26.9] [39.5-40.2] Uninvolved SLS (% GC) 33.9 (5.7) 37.5 (5.5) 36.9 (7.2) 39.9 (1.5) p < 0.001 [30.3-37.5] [34.5-40.5] [32.8-41.0] [39.6-40.3] Abbreviations: SL step length, SLS single limb support, GC gait cycle. Significance was set to p < 0.05. Table 2 Differences in gait patterns between ankle fracture groups and controls Segal et al. Journal of Foot and Ankle Research 2014, 7:50 http://www.jfootankleres.com/content/7/1/50 Page 5 of 7 Figure 1 SF-36 scores in the 3 ankle fracture groups and controls. Abbreviation: PHS-Physical Health Score; MHS-Mental Health Score. Significant values were found between health controls and all 3 ankle fracture groups. There were no significant differences between the ankle fracture groups. p-value was set to p < 0.05. We found that all ankle fracture groups had a compro- mised gait pattern and poorer quality of life compared to healthy controls. Significant differences were also found within the ankle fracture groups in most parameters, espe- cially between patients with unimalleolar fracture com- pared to bimalleolar and trimalleolar fracture. Patients with an ankle fracture injury, which were treated surgically with immobilization period, presented compromised gait pattern and clinical symptoms. This is not surprising as this group of participants was exam- ined immediately once weight-bearing activities were ap- proved (2–3 months post ankle injury). We aimed to characterize the differences between ankle fracture se- verity groups early in the rehabilitation process and hy- pothesized that there will be a linear correlation between fracture severity group and functional disability (i.e. pa- tients with unimalleolar will present better results com- pared to bimalleolar fracture and trimalleolar fracture and that patients with bimalleolar fracture will present better results compared to trimalleolar fracture). Our hypothesis was only partially supported with the study results. Questionnaires Regarding their gait pattern, patients with unim- alleolar fracture presented better gait results compared to patients with bimalleolar and trimalleolar fracture in most gait parameters. But, patients with bimalleolar frac- ture were not statistically different compared to trimal- leolar fracture patients. Interestingly, patients with bimalleolar fracture presented a slightly worse gait pat- tern compared to trimalleolar fracture patients. Further- more, patients with unimalleolar fracture had better clinical score and functional score compared to bimal- leolar fracture and trimalleolar fracture (AFAS, 6-min walk test). Similar to the gait results, there were no sig- nificant differences between the bimalleolar fracture group and trimalleolar fracture. Overall it seems that pa- tients with unimalleolar fracture are at a better functional condition immediately with weight-bearing permission compared to bimalleolar and trimalleolar fracture. This was determined with both objective and subjective mea- sures in order to make this characterization sound and valid. Moreover, it seems that there are no significant dif- ferences between patients with bimalleolar fracture and patients with trimalleolar fracture. Based on the current study results it may be postulated that unimalleolar frac- ture is a different injury than bimalleolar and trimalleolar fracture and can be considered as a mild injury. Further- more, it may be postulated that both bimalleolar fracture and trimalleolar fracture affect the functional condition of the patient to the same extend and should be considered as equal. It is important to note that there were no signifi- cant differences between ankle fracture groups in the prevalence of soft tissue injuries. Furthermore, there were no significant differences between patients with a soft tis- sue injury and patients without a soft tissue injury in all Figure 1 SF-36 scores in the 3 ankle fracture groups and Figure 1 SF 36 scores in the 3 ankle fracture groups and controls. Abbreviation: PHS-Physical Health Score; MHS-Mental Health Score. Significant values were found between health controls and all 3 ankle fracture groups. There were no significant differences between the ankle fracture groups. p-value was set to p < 0.05. differences in walking distance were found between pa- tients with unimalleolar fracture and patients with tri- malleolar fracture (p = .020). There were no significant differences between patients with unimalleolar fracture and patients with bimalleolar fracture (p = .128), and be- tween patients with bimalleolar fracture and patients with trimalleolar fracture (p =1.000). Discussion The functional condition of patients following ankle fracture has been well examined, but results remain un- clear as researchers reported contradicting findings [8-11]. A long-term follow-up study of patients following unimalleolar and bimalleolar ankle fracture found that more than half of the patients still report pain, stiffness and swelling, and almost half of them had functional dis- abilities [30]. Although age was a dominant discrimin- ator, fracture severity type should also be considered. A previous study has evaluated the gait patterns and sym- metry in patients following ankle fracture, however no significant differences were found between fracture se- verity groups [16]. They evaluated only young popula- tion and months post-surgery. A two-year follow-up study on the functional outcomes and quality of life of patients with type B ankle fracture showed that nearly 60% of the patients had good clinical outcomes, however patients’ sensation of full recovery was reported by only 37%, 40% had work-related problems and 60% com- plained of ankle related problems with leisure or sports activities. Furthermore, they found that quality of life was negatively affected in patients with ankle fractures two-year after the injury [27]. The current study charac- terized the gait patterns and clinical symptoms of pa- tients following ankle fracture compared to controls and evaluated the differences between ankle fracture severity groups (unimalleolar, bimalleolar and trimalleolar fracture). Segal et al. Journal of Foot and Ankle Research 2014, 7:50 http://www.jfootankleres.com/content/7/1/50 Page 6 of 7 Segal et al. Journal of Foot and Ankle Research 2014, 7:50 http://www.jfootankleres.com/content/7/1/50 measured variables. Based on these findings it may be as- sumed that the presence of a soft tissue injury is not a co- variate to the results presented above. Our results partially support previous results that have reported no significant differences in functional outcomes between fracture sever- ity classifications [8-10]. In contrast, our results also sup- port the results of Tejwani et al. who also concluded that the functional outcome for patients with a bimalleolar fracture is worse than that for patients with a unimalleolar fracture [31]. They did not, however, include a group of patients with trimalleolar fracture. Future research should examine whether this trend changes in the long-term. It could be that fracture severity will have a long-term effect and that patients with trimalleolar fracture will have a slower rehabilitation period compared to patients with bimalleolar fracture. References 1. Lash N, Horne G, Fielden J, Devane P: Ankle fractures: functional and lifestyle outcomes at 2 years. ANZ J Surg 2002, 72(10):724–730. 1. Lash N, Horne G, Fielden J, Devane P: Ankle fractures: functional and lifestyle outcomes at 2 years. ANZ J Surg 2002, 72(10):724–730. 2. Bauer M, Bengnér U, Johnell O, Redlund-Johnell I: Supination-eversion fractures of the ankle joint: changes in incidence over 30 years. Foot Ankle 1987, 8(1):26–28. 3. Court-Brown CM, McBirnie J, Wilson G: Adult ankle fractures–an increasing problem? Acta Orthop Scand 1998, 69(1):43–47. 4. Tile M: Fractures of the Ankle. In The Rationale of Operative Fracture Care. 2nd edition. Edited by Schatzker J, Tile M. Berlin: Springer; 1996:523–561. 5. Donatto KC: Ankle fractures and syndesmosis injuries. Orthop Clin North Am 2001, 32(1):79–90. 6. Shaffer MA, Okereke E, Esterhai JL, Elliott MA, Walker GA, Yim SH, Vandenborne K: Effects of immobilization on plantar-flexion torque, fatigue resistance, and functional ability following an ankle fracture. Phys Ther 2000, 80(8):769–780. 7. Vandenborne K, Elliott MA, Walter GA, Abdus S, Okereke E, Shaffer M, Tahernia D, Esterhai JL: Longitudinal study of skeletal muscle adaptations during immobilization and rehabilitation. Muscle Nerve 1998, 21(8):1006–1012. Discussion Furthermore, future research may also use these measures to evaluate intervention outcomes. as tools to evaluate treatment outcomes. Furthermore, based on the results of the present study it may be sug- gested that although rehabilitation following ankle fracture is essential for all types of fracture severity, they should be personally fitted. Bimalleolar and trimalleolar fracture groups may need a more extensive rehabilitation program as they present lower scores compared to unimalleolar patients. Authors’ contributions GS - Conception and design, analysis of data, drafting the article, final approval. AE - Conception and design, revising the article, final approval. AP - Conception and design, analysis of data, revising the article, final approval. ZH - Analysis of data, revising the article, final approval. EP - Analysis of data, revising the article, final approval. MN - Conception and design, revising the article, final approval. ZF - Conception and design, revising the article, final approval. BK - Conception and design, revising the article, final approval. All authors read and approved the final manuscript. This study had some limitations. First, this study was applied to a relatively small sample groups. Although the study had strong power to detect differences between the ankle fracture groups and the control group, it had lower statistical power for comparing the ankle fracture groups within themselves. Nevertheless, it seems that even with a sample size of 60 patients in each group (bimalleolar and trimalleolar), there is not enough power to detect statistical significant differences (the power is below 80%). This implies that the fact that there were no significant differences between bimalleolar fracture and trimalleolar fracture groups is not due to small sample size, but rather a true reflection of these groups. Future research should examine the changes in gait patterns and clinical symptoms in larger cohort groups. Secondly, only spatiotemporal gait analysis was applied. Evaluating gait differences in a comprehensive 3 dimensional gait analysis could have added additional and wider informa- tion regarding the changes in gait patterns between the ankle fracture groups. Nevertheless, we sought of using a simple, objective measuring tool that can be imple- mented in any clinic with immediate results. Thirdly, this was a one-session evaluation of patients immediately following ankle fracture. Future research should examine long-term differences in gait patterns and clinical out- comes between severity groups. Author details 1 1AposTherapy Research Group, 1 Abba Even Blvd, Herzliya 46733, Israel. 2Department of Orthopedic Surgery, Meir Medical Center, Kfar-Saba, Israel. 3Department of Orthopaedic Surgery, Wolfson Medical center, Holon, Israel, affiliated to the Sacker Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel. 1AposTherapy Research Group, 1 Abba Even Blvd, Herzliya 46733, Israel. 2Department of Orthopedic Surgery, Meir Medical Center, Kfar-Saba, Israel. 3Department of Orthopaedic Surgery, Wolfson Medical center, Holon, Israel, affiliated to the Sacker Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel. Received: 6 July 2014 Accepted: 7 November 2014 Received: 6 July 2014 Accepted: 7 November 2014 Received: 6 July 2014 Accepted: 7 November 2014 Abbreviations BMI: Body mass index; SLS: Single limb support; FAOS: The Foot and Ankle Outcome Score; PHC: Physical health score; MHS: Mental health score; AOFAS: The American Orthopaedic Foot and Ankle Score; ROM: Range of motion. Competing interests The authors declare that they have no competing interests. 11. Egol KA, Tejwani NC, Walsh MG, Capla EL, Koval KJ: Predictors of short- term functional outcome following ankle fracture surgery. J Bone Joint Surg Am 2006, 88(5):974–979. 10. Lindsjö U: Operative treatment of ankle fracture-dislocations. A follow-up study of 306/321 consecutive cases. Clin Orthop Relat Res 1985, 199:28–38. 9. Hancock MJ, Herbert RD, Stewart M: Prediction of outcome after ankle fracture. J Orthop Sports Phys Ther 2005, 35(12):786–792. 12. Allet L, Armand S, Golay A, Monnin D, de Bie RA, de Bruin ED: Gait characteristics of diabetic patients: a systematic review. Diabetes Metab Res Rev 2008, 24(3):173–191. Conclusions 8. Broos PL, Bisschop AP: Operative treatment of ankle fractures in adults: correlation between types of fracture and final results. Injury 1991, 22(5):403–406. 8. Broos PL, Bisschop AP: Operative treatment of ankle fractures in adults: correlation between types of fracture and final results. Injury 1991, 22(5):403–406. Patients with ankle fracture present altered gait patterns and clinical symptoms compared to healthy controls. Pa- tients with unimalleolar fracture present significant bet- ter results compared to bimalleolar and trimalleolar ankle fracture. There were no significant differences be- tween bimalleolar fracture and trimalleolar fracture patients immediately post injury with weight-bearing approval. Such characterization was done while using both objective and subjective measures. Using these parameters may also serve 9. Hancock MJ, Herbert RD, Stewart M: Prediction of outcome after ankle fracture. J Orthop Sports Phys Ther 2005, 35(12):786–792. 10. Lindsjö U: Operative treatment of ankle fracture-dislocations. A follow-up study of 306/321 consecutive cases. Clin Orthop Relat Res 1985, 199:28–38. 11. Egol KA, Tejwani NC, Walsh MG, Capla EL, Koval KJ: Predictors of short- term functional outcome following ankle fracture surgery. J Bone Joint Surg Am 2006, 88(5):974–979. term functional outcome following ankle fracture surgery. J Bone Joint Surg Am 2006, 88(5):974–979. 12. Allet L, Armand S, Golay A, Monnin D, de Bie RA, de Bruin ED: Gait characteristics of diabetic patients: a systematic review. Diabetes Metab Res Rev 2008, 24(3):173–191. 12. Allet L, Armand S, Golay A, Monnin D, de Bie RA, de Bruin ED: Gait characteristics of diabetic patients: a systematic review. Diabetes Metab Res Rev 2008, 24(3):173–191. Page 7 of 7 Page 7 of 7 Segal et al. Journal of Foot and Ankle Research 2014, 7:50 http://www.jfootankleres.com/content/7/1/50 13. Barton CJ, Levinger P, Menz HB, Webster KE: Kinematic gait characteristics associated with patellofemoral pain syndrome: a systematic review. Gait Posture 2009, 30(4):405–416. 14. Zuka-Nowak K, Ogrodzka K, Chwała W, Niedźwiedzki L, Niedźwiedzki T: Kinetic and kinematic characteristics of natural velocity gait in anterior cruciate ligament-deficient patients. Int J Rehabil Res 2013, 36(2):152–161. 15. Kaufman KR, Hughes C, Morrey BF, Morrey M, An KN: Gait characteristics of patients with knee osteoarthritis. J Biomech 2001, 34(7):907–915. 16. Becker HP, Rosenbaum D, Kriese T, Gerngross H, Claes L: Gait asymmetry following successful surgical treatment of ankle fractures in young adults. Clin Orthop Relat Res 1995, 311:262–269. 17. Fuller DA, Keenan MA, Esquenazi A, Whyte J, Mayer NH, Fidler-Sheppard R: The impact of instrumented gait analysis on surgical planning: treatment of spastic equinovarus deformity of the foot and ankle. Foot Ankle Int 2002, 23(8):738–743. 18. Foley N, Murie-Fernandez M, Speechley M, Salter K, Sequeira K, Teasell R: Does the treatment of spastic equinovarus deformity following stroke with botulinum toxin increase gait velocity? A systematic review and meta-analysis. Eur J Neurol 2010, 17(12):1419–1427. 19. Theologis T, Stebbins J: The use of gait analysis in the treatment of pediatric foot and ankle disorders. Foot Ankle Clin 2010, 15(2):365–382. 20. Elbaz A, Mor A, Segal G, Debi R, Shazar N, Herman A: Novel classification of knee osteoarthritis severity based on spatiotemporal gait analysis. Osteoarthritis Cartilage 2014, 22(3):457–463. 21. Michelson J, Solocoff D, Waldman B, Kendell K, Ahn U: Ankle fractures. The Lauge-Hansen classification revisited. Clin Orthop Relat Res 1997, 345:198–205. 22. Bilney B, Morris M, Webster K: Concurrent related validity of the GAITRite walkway system for quantification of the spatial and temporal parameters of gait. Gait Posture 2003, 17(1):68–74. 23. Menz HB, Latt MD, Tiedemann A, Mun San Kwan M, Lord SR: Reliability of the GAITRite walkway system for the quantification of temporo-spatial parameters of gait in young and older people. Gait Posture 2004, 20(1):20–25. 24. Kim CM, Eng JJ: Symmetry in vertical ground reaction force is accompanied by symmetry in temporal but not distance variables of gait in persons with stroke. Gait Posture 2003, 18(1):23–28. 25. Roos EM, Brandsson S, Karlsson J: Validation of the foot and ankle outcome score for ankle ligament reconstruction. Foot Ankle Int 2001, 22(10):788–794. 26. Ware JE, Sherbourne CD: The MOS 36-item short-form health survey (SF-36). I Conceptual framework and item selection. Med Care 1992, 30(6):473–483. 27. Segal et al. Journal of Foot and Ankle Research 2014, 7:50 http://www.jfootankleres.com/content/7/1/50 Posner S, Nasell H, Bergman B, Tornkvist H: Functional outcome and quality of life in patients with type B ankle fractures: a two-year follow-up study. J Orthop Rauma 1999, 13(5):363–368. 28. Kitaoka HB, Alexander IJ, Adelaar RS, Nunley JA, Myerson MS, Sanders M: Clinical rating systems for the ankle-hindfoot, midfoot, hallux, and lesser toes. Foot Ankle Int 1994, 15(7):349–353. 28. Kitaoka HB, Alexander IJ, Adelaar RS, Nunley JA, Myerson MS, Sanders M: Clinical rating systems for the ankle-hindfoot, midfoot, hallux, and lesser toes. Foot Ankle Int 1994, 15(7):349–353. 29. Enright PL, McBurnie MA, Bittner V, Tracy RP, McNamara R, Arnold A, Newman AB, Cardiovascular Health Study: The 6-min walk test: a quick measure of functional status in elderly adults. Chest 2003, 123(2):387–398. 29. Enright PL, McBurnie MA, Bittner V, Tracy RP, McNamara R, Arnold A, Newman AB, Cardiovascular Health Study: The 6-min walk test: a quick measure of functional status in elderly adults. Chest 2003, 123(2):387–398. 30. Nilsson MG, Jonsson K, Ekdahl SC, Eneroth M: Unsatisfactory outcome following surgical intervention of ankle fractures. Foot Ankle Surg 2005, 11:11–16. 30. Nilsson MG, Jonsson K, Ekdahl SC, Eneroth M: Unsatisfactory outcome following surgical intervention of ankle fractures. Foot Ankle Surg 2005, 11:11–16. 31. Tejwani NC, McLaurin TM, Walsh M, Bhadsavle S, Koval KJ, Egol KA: Are outcomes of bimalleolar fractures poorer than those of lateral malleolar fractures with medial ligamentous injury? J Bone Joint Surg Am 2007, 89(7):1438–1441. 31. Tejwani NC, McLaurin TM, Walsh M, Bhadsavle S, Koval KJ, Egol KA: Are outcomes of bimalleolar fractures poorer than those of lateral malleolar fractures with medial ligamentous injury? J Bone Joint Surg Am 2007, 89(7):1438–1441. Segal et al. Journal of Foot and Ankle Research 2014, 7:50 http://www.jfootankleres.com/content/7/1/50 Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: Submit your next manuscript to BioMed Central and take full advantage of: doi:10.1186/s13047-014-0050-9 Cite this article as: Segal et al.: Clinical outcomes following ankle fracture: a cross-sectional observational study. Journal of Foot and Ankle Research 2014 7:50. • Convenient online submission
https://openalex.org/W1977921870	https://europepmc.org/articles/pmc4174675?pdf=render	English	null	Predicting disease associations via biological network analysis	BMC bioinformatics	2,014	cc-by	11,918	RESEARCH ARTICLE Open Access © 2014 Sun et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Correspondence: natasha@imperial.ac.uk 1Department of Computing, Imperial College London, London, SW7 2AZ, UK Full list of author information is available at the end of the article Abstract Background: Understanding the relationship between diseases based on the underlying biological mechanisms is one of the greatest challenges in modern biology and medicine. Exploring disease-disease associations by using system-level biological data is expected to improve our current knowledge of disease relationships, which may lead to further improvements in disease diagnosis, prognosis and treatment. Results: We took advantage of diverse biological data including disease-gene associations and a large-scale molecular network to gain novel insights into disease relationships. We analysed and compared four publicly available disease-gene association datasets, then applied three disease similarity measures, namely annotation-based measure, function-based measure and topology-based measure, to estimate the similarity scores between diseases. We systematically evaluated disease associations obtained by these measures against a statistical measure of comorbidity which was derived from a large number of medical patient records. Our results show that the correlation between our similarity measures and comorbidity scores is substantially higher than expected at random, confirming that our similarity measures are able to recover comorbidity associations. We also demonstrated that our predicted disease associations correlated with disease associations generated from genome-wide association studies significantly higher than expected at random. Furthermore, we evaluated our predicted disease associations via mining the literature on PubMed, and presented case studies to demonstrate how these novel disease associations can be used to enhance our current knowledge of disease relationships. Conclusions: We present three similarity measures for predicting disease associations. The strong correlation between our predictions and known disease associations demonstrates the ability of our measures to provide novel insights into disease relationships. Keywords: Disease classification, Network analysis, Graph theory, Topology, Protein-protein interact that are characterized by distinct patterns of pathway acti- vation [1]. However, a common mechanism may lead to different phenotypes. For example, a mutation at the β- globin locus may lead to sickle-cell anemia with different phenotypes such as bony infarcts, acute chest syndrome and stroke [2]. Predicting disease associations via biological network analysis Kai Sun1, Joana P Gonçalves1, Chris Larminie2 and Nataša Pržulj1 Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 Open Access Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 function-based measure and topology-based measure, to estimate similarity scores between diseases. The disease associations obtained by the three measures were sys- tematically evaluated against the standard disease clas- sification system, namely the International Classification of Diseases, 9th revision (ICD-9) [23], and a statistical measure of comorbidity derived from a large number of medical patient records. In addition, we evaluated our predicted disease associations by using disease associ- ations generated from GWAS studies, which represent one of the most robust routes for identifying causal rela- tionships between genes and diseases. To our knowledge, this is the first time comorbidity and GWAS data have been used to evaluate computationally predicted disease associations. can be a genetic association (i.e., mutations in that gene may lead to that disease), or a connection inferred from other aspects. Disease-gene associations can be obtained from large-scale knowledge-bases such as the Online Mendelian Inheritance in Man (OMIM) [3]. Early studies used text mining to infer similarities between phenotypes contained in OMIM, and found those similarities were positively correlated with a number of measures of gene functions [4] and could be used to predict disease-causing genes [5]. Also by using OMIM, Goh et al. [6] constructed the human diseasome by connecting diseases that share a disease-causing gene. Other types of biological data such as biological pathways [7], gene expression data [8,9], biomedical ontologies [10,11], and genome-wide associ- ation study (GWAS) data [12-14], have also been used to improve the current understanding of disease relation- ships from different aspects. Recently, networks have been used to model large-scale biological data, and network topology is beginning to provide insights into diseases and their associations [6,15-17]. By considering the inter- connectivity of biomolecules in the cell, the topology of biological networks is expected to have various biological and clinical applications [18,19]. In the rest of this paper, we will start with a descrip- tion of the biological data we analysed, followed by details of our methodology of measuring disease associations. Then we will show and discuss the evaluation of dis- ease associations predicted by our similarity measures against known disease associations derived from ICD-9, comorbidity data and GWAS data. Finally, we will present case studies to demonstrate the ability of our similarity measures to predict novel disease associations. Biological data Three types of biological data were used in this study: protein-protein interactions (PPIs), Gene Ontology (GO) annotations and disease-gene associations. Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 pp Despite these advances, early studies have several lim- itations when inferring disease associations from bio- logical data. First, some studies only considered several specific diseases, rather than giving a global comparison among all diseases (e.g., [9,12-14]). This is the case for GWAS-based studies, since a small number of GWAS studies have been completed to date in a relatively small proportion of the total disease population. Furthermore, most studies solely used OMIM as the source of disease- gene association data. OMIM is a catalogue of mendelian disorders and as a result, most diseases are annotated with few genes in OMIM [20]. Limitations of using OMIM have also been discussed previously [21,22]. Finally, most computationally predicted disease associations were not systematically evaluated due to the difficulty in identify- ing a suitable benchmark of known disease associations. In particular, most studies were only able to validate part of their results by comparing them with phenotypic sim- ilarities (e.g., [12]) or mining the literature manually (e.g., [13]). A comparison of previous studies can be found in Table 1. Background d Correct diagnosis is critical for effective treatment and prevention of disease. As a result, disease classification has become a key cornerstone of modern medicine. Dis- ease may be classified by any one of a number of criteria: topographic, anatomic, pathological, physiological, etio- logical, juristic, epidemiological or statistical approaches. However, without considering the molecular mechanisms driving diseases, such knowledge is limited and can even be misleading. For example, a common phenotype can be caused by different underlying mechanisms, such as breast cancer, which can be divided into several subgroups During the past decade, a wealth of biological data has been generated from various large-scale genomic studies, prompting the scientific community to gain deeper insight into disease relationships based on their underlying bio- logical mechanisms. Various types of biological data have been used to infer associations between diseases. One of the most commonly used biological data is disease-gene association. In a broad definition, a disease-gene asso- ciation is a connection reported in the literature, which Page 2 of 13 Page 2 of 13 PPI network We modelled PPI data as a network. A network or graph G(V, E) consists of two types of elements, a set V of nodes and a set E ⊆V × V of edges connecting them. A PPI network models the physical interaction among proteins in the cell, in which a node represents a pro- tein, and an undirected edge exists between a pair of nodes if their corresponding proteins can physically bind to each other. Currently available PPIs are mostly yielded from various high throughput proteomics experiments, such as yeast two-hybrid screening (e.g., [24]) and affin- ity capture mass spectrometry (e.g., [25]). We constructed a human PPI network using data obtained from BioGRID [26] version 3.1.93 (released in October 2012). All self- loops, duplicate interactions were removed since we con- sidered only simple, undirected graphs. We also removed the cross-species interactions (i.e., interactions between human proteins and proteins of other species) because we focused on the physical interactions between human proteins in our study. The PPI network we constructed contained 11,375 nodes and 66,317 edges, while its largest connected component contained 11,261 nodes and 66,253 edges. Note that the second largest connected component In our study, we used diverse biological data from a number of repositories to gain novel insights into the relationship of over 500 known human diseases by con- sidering their underlying biological mechanisms. We used disease-gene associations obtained from four different sources to avoid the bias introduced by a single dataset. Moreover, we took advantage of the topology of a large- scale molecular network to examine its use for inferring disease associations. We applied three different disease similarity measures, namely annotation-based measure, Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 Page 3 of 13 Table 1 Comparison of studies on inferring disease-disease associations Data Size Evaluation van Driel et al. (2006) [4] OMIM 5132 phenotypes in OMIM Comparing results with genotypic similarities Lage et al. (2007) [5] OMIM 7000 OMIM record pairs Evaluating results against the overlap of the OMIM record pairs Goh et al. (2007) [6] OMIM 1284 OMIM diseases Analysing network topological properties Huang et al. (2009) [12] GWAS 7 diseases Comparing results with phenotypic similarities Li and Agarwal (2009) [7] Pubmed abstracts, biological pathways 1028 diseases in MeSH Comparing results with MeSH classification Kim et al. PPI network (2009) [13] GWAS 53 clinical traits related to severe asthma Mining the literature manually Hu and Agarwal (2009) [8] Expression data 645 diseases in MeSH Comparing results with MeSH classification Suthram et al. (2010) [9] Expression data, PPI 54 diseases Evaluating results against genetic similarities Lewis et al. (2011) [14] GWAS 61 diseases Comparing results with Huang et al. (2009) results Mathur and Dinakarpandian et al. (2007) [10] DO annotation, GO annotation 36 diseases (for evaluation) Evaluating results using 68 curated disease associations Our study Disease-gene associations, GO annotation, PPI 543 ICD-9 diseases Evaluating results against ICD-9 classification, comorbidity, and genetic similarities derived from GWAS data The comparison is based on the data used to derive associations (denoted by ‘Data’), number of diseases evaluated (denoted by ‘Size’) and benchmarks used for evaluation (denoted by ‘Evaluation’). The number of diseases evaluated in our study is computed as the union of diseases annotated in the four disease-gene association datasets we analysed, given in Figure 1. Table 1 Comparison of studies on inferring disease-disease associations The comparison is based on the data used to derive associations (denoted by ‘Data’), number of diseases evaluated (denoted by ‘Size’) and benchmarks used for evaluation (denoted by ‘Evaluation’). The number of diseases evaluated in our study is computed as the union of diseases annotated in the four disease-gene association datasets we analysed given in Figure 1 The comparison is based on the data used to derive associations (denoted by ‘Data’), number of diseases evaluated (denoted by ‘Size’) and benchmarks used for evaluation (denoted by ‘Evaluation’). The number of diseases evaluated in our study is computed as the union of diseases annotated in the four disease-gene association datasets we analysed, given in Figure 1. only contained 5 nodes and 5 edges. There were also 7 iso- lated triangles and 43 isolated edges in the PPI network. The presentence of these small components may be due to the incompleteness of the PPI data. In addition, the topol- ogy of these small components is not as informative as that of the largest connected component. For these reasons, we only used the largest connected component of the PPI network in our analysis. genes in the human genome. The degree of a disease is the number of genes associated with that disease, while the degree of a gene is the number of diseases annotated with that gene. PPI network We used four disease-gene association datasets obtained from different sources: OMIM, Com- parative Toxicogenomics Database (CTD) [28], Functional Disease Ontology annotations (FunDO) [29] and Human Genome Epidemiology Network (HuGENet) [30]. Among these datasets, OMIM, CTD, and FunDO contain curated associations, while HuGENet contains computationally inferred associations. Details of these disease-gene associ- ation datasets are described below. GO annotations Genes are annotated with GO terms to represent their biological properties [27]. All GO terms are organised in three domains: cellular component, molecular function and biological process. We downloaded the ontology file and annotations of Homo sapiens from the Gene Ontol- ogy database (http://www.geneontology.org) in November 2012. We removed annotations with evidence code ‘Inferred from Electronic Annotation’ (IEAs), since IEAs are computationally inferred annotations which have not been reviewed by curators. In total, we collected 171,888 annotations between 13,166 genes and 10,787 GO terms. • OMIM is considered to be the best-curated resource of known phenotype-genotype relationships, and it has been used in various disease-related studies (discussed in the Background section). We downloaded the OMIM database in November 2012. In total, it contains 3,537 diseases (annotated by OMIM IDs), 2,862 genes and 4,337 disease-gene associations. • CTD provides scientific data describing relationships between chemicals, genes, and human diseases, with the goal of improving the understanding of environmental chemicals’ effects on human health. It Function-based measure The function-based similarity measure used both GO term annotations and disease-gene associations to esti- mate the similarity score between a pair of diseases. We first propagated the GO annotations upwards through the GO hierarchy, i.e., when a gene was annotated with a GO term, we assumed associations between the gene and the term’s parents. For each disease Di annotated in a spe- cific disease-gene association dataset, we then identified the set of GO terms that were overrepresented within GDi, denoted by GODi. The statistical significance (p-value) of the enrichment of a GO term was computed according to the hypergeometric distribution for sampling without replacement, and was corrected for multiple testing using the Benjamini-Hochberg test. Only overrepresented GO terms from the ‘biological process’ domain of GO and hav- ing a p-value less than 0.05 were considered to be in GODi. For a pair of diseases Di and Dj, we computed the Jaccard index of GODi and GODj as their function-based similarity score, defined as: • HuGENet is known as an integrated knowledge-base on human genome epidemiology. The Phenopedia collection [30] of HuGENet contains disease-gene associations obtained by text-mining of abstracts on PubMed using machine learning techniques. Disease- gene association data were downloaded via HuGE Navigator in September 2012. We obtained 353,883 associations between 2,387 diseases (annotated by Unified Medical Language System (UMLS) [32] Concept Unique Identifiers (CUIs)) and 11,915 genes. • HuGENet is known as an integrated knowledge-base on human genome epidemiology. The Phenopedia collection [30] of HuGENet contains disease-gene associations obtained by text-mining of abstracts on PubMed using machine learning techniques. Disease- gene association data were downloaded via HuGE Navigator in September 2012. We obtained 353,883 associations between 2,387 diseases (annotated by Unified Medical Language System (UMLS) [32] Concept Unique Identifiers (CUIs)) and 11,915 genes. Since disease names or IDs used in these datasets are based on different labelling schemes, we mapped all dis- ease names or IDs to ICD-9 codes, for the purpose of comparing these datasets and further evaluation (also see the Results and discussion section for details). We used the mapping manually constructed by [6] and [33] to convert OMIM IDs to ICD-9 codes, and used the corre- sponding mapping provided in Disease Ontology version 3 (the latest stable version of DO, released in May 2007) to map DO IDs, MeSH terms and UMLS CUIs to ICD- 9 codes. Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 Page 4 of 13 Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 Function-based measure In total, 1,467 OMIM IDs in OMIM, 423 MeSH terms in CTD, 806 DO IDs in FunDO and 693 UMLS CUIs in HuGENet were mapped to ICD-9 codes. Simfunction(Di, Dj) = \|GODi ∩GODj\| \|GODi ∪GODj\|. (2) (2) (2) Topology-based measure Many studies have shown the relationship between topo- logical properties of proteins in the PPI network and the involvement of proteins in diseases [6,34,35]. Topological similarities of proteins in a PPI network are considered as a complementary information to sequence similarities [36]. Thus in this study, we took advantage of the topol- ogy of the human PPI network along with disease-gene association data to examine the use of network topology for uncovering novel disease associations. In particular, we proposed a measure to estimate the similarity score between a pair of diseases based on the topological simi- larity of their annotated genes. Annotation-based measure contains both curated and inferred disease-gene associations, but we only used curated associations as they have higher confidence than inferred associations. Disease-gene associations directly derived from OMIM were excluded to reduce the dependency between datasets. We downloaded the data from CTD in November 2012 and obtained 17,754 associations between 2,761 diseases (annotated by Medical Subject Heading (MeSH) terms [31]) and 5,828 genes. The annotation-based measure solely used the informa- tion obtained from disease-gene association data. We applied the Jaccard index, which is known as a standard method for comparing the similarity between two sets, to estimate the similarity score between diseases as fol- lows. Let GDi be the set of genes associated with a disease Di. We computed the annotation-based similarity score of two diseases Di and Dj as the Jaccard index (or Jaccard similarity coefficient) of GDi and GDj: • FunDO contains disease-gene associations extracted from the NCBI Gene Reference Into Function (GeneRIF) database. A GeneRIF is a brief statement about the function of a gene, along with information of its association with diseases. We downloaded the latest stable version of FunDO (released in October 2008) and obtained 1,854 diseases (annotated by Disease Ontology (DO) terms), 4,781 genes and 28,442 disease-gene associations. Simannotation(Di, Dj) = \|GDi ∩GDj\| \|GDi ∪GDj\|. (1) (1) Disease-gene associations Disease-gene associations can be modelled as a graph con- taining both known human diseases and disease-related Comorbidity associations of diseases Comorbidity associations of diseases y The availability of electronic patient records facilitates studies into disease comorbidity, which indicates the potential for co-occurrence of two given diseases in the same individual. Comorbidity can be considered as a type of disease association derived from electronic medical record, but the underlying driver for comorbidity may be very different from one another. Comorbidity and its correlation with other types of disease associations such as genetic associations [45] and evolutionary associations [46] have previously been studied. Unlike these studies, we used comorbidity data to evaluate disease associa- tions predicted by our similarity measures. Comorbidity associations were downloaded from the Human Disease Network (HuDiNe, [47]), which were obtained from the disease history of 32 million American patients. Diseases were annotated using ICD-9 codes in HuDiNe, and as many diseases in patient records were not specific enough to map to 4-digit or 5-digit codes, we used the comorbid- ity data annotated using 3-digit level ICD-9 codes for our analysis. The strength of comorbidity association between a pair of diseases can be measured by the Relative Risk and φ-correlation [47]. Because comorbidity associations quantified by φ-correlation were reported to contain more connections across different ICD-9 categories [47], we chose φ-correlation as the measure of comoridity. The φ- correlation score between Di and Dj was defined as the Pearson’s correlation for binary variables, given by: We calculated the signature similarity of each pair of genes in the human PPI network. Note that the network has an edge density (the proportion of the number of edges to the maximum possible number of edges) of 0.001, which for its size (11,261 nodes and 66,253 edges) is dense enough to avoid low edge density regions in which the topology of networks is unstable (see [44] for details). Here we extended the use of graphlet-based method to measure disease similarities. We introduced two terms to quantify the topology-based similarity score between dis- eases Di and Dj. The first term, denoted by AllSig, is the maximum of the signature similarity between a gene in GDi and a gene in GDj: φ(Di, Dj) = CijN −PiPj PiPj(N −Pi)(N −Pj) (7) (7) where Cij is the number of individuals affected by both Di and Dj, N is the total number of individuals in the population, Pi and Pj are the prevalences of Di and Dj respectively. Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 Page 5 of 13 Page 5 of 13 The second term, denoted by ShareSig, focuses on the topological similarity between genes shared with both diseases: nodes are topologically identical to each other, and such identical nodes are said to belong to the same automor- phism orbit [38]. The graphlet signature of a node u is a 73-dimensional vector, whose ith element ui counts the number of times the node u is touched by the particular automorphism orbit i [39]. According to [39], the signa- ture similarity of a pair of nodes u and v is defined as: ShareSig(Di, Dj) = max gm̸=gn gm∈GDi∩GDj gn∈GDi∩GDj SigSim(gm, gn). (5) (5) Finally we defined the topology-based similarity score between Di and Dj as the average of these two terms: SigSim(u, v) = 1 − 1 72 i=0 wi × 72 i=0 wi× \| log(ui + 1) −log(vi + 1)\| log(max{ui, vi} + 2) (3) SigSim(u, v) = 1 − 1 72 i=0 wi × 72 i=0 wi× \| log(ui + 1) −log(vi + 1)\| log(max{ui, vi} + 2) (3) Simtopology(Di, Dj) = 1 2 × (ShareSig(Di, Dj) + AllSig(Di, Dj)). (6) (6) (3) AllSig(Di, Dj) = max gm∈GDi gn∈GDj SigSim(gm, gn). (4) Evaluation where wi is a weight assigned to orbit i defined as 1 − log(oi)/log(73) (oi is the dependency count of orbit i, see [39] for details). SigSim(u, v) ranges between 0 and 1, where the value of 1 means that the two nodes, u and v, are considered to be topologically identical. This measure is a highly constraining measure of local topological simi- larity between two nodes in a network as it compares the nodes based on local structures of their neighbourhoods, which describe their interconnectivities out to a distance of four [39]. Signature similarities have been applied to measure the topological similarities between proteins in a PPI network [34,36,39-43]. It has been shown that topo- logically similar proteins are likely to belong to the same protein complexes, perform the same biological functions, be localised in the same subcellular compartments and have the same tissue expressions [39]. Signature similar- ities have also been used to relate the network structure around a protein in a PPI network to homology [36] and its involvement in diseases [34]. For these reasons, we hypothesize that the topology around disease genes in the PPI network can reflect the underlying biological mechanisms of diseases. Disease similarity measures We applied three similarity measures to estimate simi- larity scores between diseases. These measures include standard methods (i.e., Jaccard index) and novel mea- sures proposed in this study (i.e., graphlet-based mea- sure). Considering the information used in calculation, the similarity score of a pair of diseases was measured in three different ways: annotation-based, function-based and topology-based. We applied a graphlet-based method to assess the topo- logical similarity of genes in the human PPI network. A graphlet is defined as a small, connected and induced sub- graph of a larger network [37]. Within each graphlet, some GWAS data GWAS is a powerful method to identify genetic variations associated with diseases and is one of the most robust routes for identifying causal relationships between genes and diseases [48,49]. GWAS studies examine the genome for single-nucleotide polymorphisms (SNPs) that occur more frequently in people with a particular disease than in people without it. GWAS studies have enabled exploration of gene association in complex diseases in a systematic way on a genome scale. Whilst individual studies are extremely powerful, only a small number of diseases have been studied thus far using GWAS. Hence the GWAS database as a whole is only able to contribute a relatively small component to the overall knowledge base of gen- eral disease-gene associations. For this reason, we did not use GWAS data as a source of disease-gene association to measure disease similarity scores, but used them to eval- uate our predicted disease associations. We downloaded GWAS data from the National Human Genome Research Institute (NHGRI) GWAS catalog [50] in May 2013. This resource collects significant associations between traits (or diseases) and SNPs from the literature. Similar to [51], we only considered highly confident associations with p-value lower than 10−7. We also eliminated not replicated associations to minimise false-positives. For all disease-SNP associations in our analysis, we used the corresponding disease-gene associations reported by the authors in the original publications as recorded in the GWAS Catalog. After mapping diseases to ICD-9 codes, we obtained 1,756 genetic associations (from 478 publica- tions) between 126 diseases and 1,298 genes. Interestingly, the overlap among the four disease-gene association datasets is unexpectedly small, as shown in Figure 1. While a considerable number of diseases (120 diseases in total, that is, 50.21%, 47.43%, 26.20% and 33.33% of diseases annotated in OMIM, CTD, FunDO and HuGENet, respectively) have gene annotations in all four datasets, few disease-gene associations (159 associ- ations in total, that is, 7.05%, 1.99%, 0.92% and 0.11% of associations in OMIM, CTD, FunDO and HuGENet, respectively) can be found in all datasets. Additional file 1: Figure S1 further demonstrates the difference between these datasets according to the degree distribution of dis- eases. In general, these distributions follow power law distributions, indicating that most human diseases are associated with only a few disease genes, while a small number of diseases relate to many genes. Comorbidity associations of diseases A φ-correlation higher than 0 indicates the co-occurrence of Di and Dj is more frequently than AllSig(Di, Dj) = max gm∈GDi gn∈GDj SigSim(gm, gn). (4) (4) Page 6 of 13 Page 6 of 13 Sun et al. BMC Bioinformatics 2014, 15:304 Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 http://www.biomedcentral.com/1471-2105/15/304 expected by random. The statistical significance of φ- correlation was determined by using a t-test, expected by random. The statistical significance of φ- correlation was determined by using a t-test, in common. We mapped all disease names or IDs anno- tated in these datasets to ICD-9 codes for a correct com- parison (see the Methods section for more details). If several diseases were mapped to a common ICD-9 code, we assigned the union of genes associated with those diseases to that ICD-9 code. In order to evaluate our measures using comorbidity data, we further limited the ICD-9 codes to 3-digit level. We are aware that noise may be introduced when merging diseases into 3-digit level. Generally speaking, a 3-digit level ICD-9 code is always associated with more than one disease, thus the average degree of diseases increased after mapping. Note that it is possible that two diseases may share clinical traits but have different 3-digit level ICD-9 codes, e.g., acute bron- chitis (ICD-9: 466) and chronic bronchitis (ICD-9: 491). However, in most cases if two diseases have different ICD- 9 codes at 3-digit level, they always have different clinical phenotypes and they are unlikely to share similarity traits. t = φ√n −2 1 −φ2 (8) t = φ√n −2 1 −φ2 (8) where n = max(Pi, Pj) is the number of observations used to calculate φ. We used significant associations at 5% level (t ≥1.96) for our analyses. where n = max(Pi, Pj) is the number of observations used to calculate φ. We used significant associations at 5% level (t ≥1.96) for our analyses. GWAS data However, this scale-free topology may also be an artifact of sampling: several diseases are better studied than others [52]. We notice that in OMIM, most diseases are associated with fewer genes compared with other datasets. The average number of genes associated with a disease in OMIM is 9.43, while in the two other curated datasets CTD and FunDO, these numbers are 31.59 and 37.80. On the other hand, on average a disease in HuGENet is annotated with more than 300 genes: HuGENet has a higher false posi- tive rate compared to other datasets, since its associations were derived from computational predictions rather than manual curations. Correlation with comorbidity is no single standard and systematic method to assess the quality of these data. Therefore, to gain a more comprehensive view of human diseases and to test the robustness of our methods, we used all four disease-gene association datasets along with the intersection/union of the three curated datasets in further computation and evaluation. As the goal of our study is to uncover novel disease associ- ations that may reflect common underlying mechanisms, we are more interested in the associations between dis- eases that belong to different ICD-9 categories. For this reason, we systematically evaluated our similarity mea- sures against a statistical measure of comorbidity. We say two diseases are associated according to comorbidity if they are reported to have a significant co-occurrence in the same individual. In particular, their φ-correlation score should be higher than a chosen threshold and statis- tically significant at 5% level. Additional file 1: Figure S2 shows the distribution of φ-correlation scores for all pairs of diseases we analysed. Note that even though the comor- bidity associations we used for evaluation contained dis- ease associations across different ICD-9 categories, there was overlap between associations derived from ICD-9 and comorbidity associations. For example, the associa- tion between diabetes mellitus and obesity was supported by both ICD-9 classification and comorbidity data. Since ICD-9 and comorbidity describe the relationship between diseases from different aspects, we believe the evalua- tions against ICD-9 classification and comorbidity do not contradict each other, but are complementary to each other. Results and discussion Comparison of disease gene association datasets We analysed four different disease-gene association datasets: three curated datasets, namely OMIM, CTD and FunDO, and one computationally predicted dataset, HuGENet (details of these datasets can be found in the Methods section). Although these datasets focus on dif- ferent aspects of the connections between diseases and genes, they are not fully independent since information contained in these datasets is extracted from the litera- ture. For example, disease-gene associations contained in CTD and FunDO were extracted from 9,269 and 48,436 publications respectively, and they have 799 publications The difference and inconsistency discussed above indi- cate that currently available disease-gene association datasets are still noisy and incomplete. The incomplete- ness may be due to the focus of the datasets and the nature of the curation process. For example, OMIM mainly focuses on mendelian diseases and traits. Mean- while, many false positives may be introduced by text- mining the literature (e.g., HuGENet). However, there Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 Sun et al. BMC Bioinformatics 2014, 15:304 Page 7 of 13 Figure 1 The overlap of datasets. The overlap of diseases (denoted by ‘D’), genes (denoted by ‘G’) and their associations (denoted by ‘A’) between the four disease-gene association datasets we analysed. Boxes on the left list the sizes of the datasets. The size of the intersection of the datasets is marked in bold. Figure 1 The overlap of datasets. The overlap of diseases (denoted by ‘D’), genes (denoted by ‘G’) and their associations (denoted by ‘A’) between the four disease-gene association datasets we analysed. Boxes on the left list the sizes of the datasets. The size of the intersection of the datasets is marked in bold. Evaluation of similarity measures Correlation with ICD-9 The results obtained by these measures were first eval- uated against the standard disease classification system ICD-9. We say that two diseases are associated according to ICD-9, if they are classified under the same ICD-9 cat- egory. For example, diabetes mellitus (ICD-9 code: 250) and thyroiditis (ICD-9 code: 245) are classified under the same category ‘endocrine, nutritional and metabolic diseases, and immunity disorders’. To investigate the cor- relation between our similarity measures and the ICD-9 classification, we tested whether a pair of diseases from the same ICD-9 category tends to have a higher similar- ity score than diseases from different ICD-9 categories (Table 2). Since similarity scores obtained by our measures are not normally distributed, we used a non-parametric test, namely the Mann-Whitney U test, to assess the sta- tistical significance (p-value). Our results show that for all three similarity measures and all four disease-gene associ- ation datasets, similarity scores of diseases from the same ICD-9 category are significantly higher than those from different ICD-9 categories. To assess the ability of our measures to uncover highly confident comorbidity associations, we used Receiver Operating Characteristic (ROC) curves, in which we plot- ted the True Positive Rate (TPR, also known as sensitiv- ity) versus the False Positive Rate (FPR, also known as Sun et al. Evaluation of similarity measures Correlation with ICD-9 BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 Page 8 of 13 Page 8 of 13 Table 2 Evaluation of our measures against ICD-9 classification Data Group Annotation-based Function-based Topology-based OMIM Same 0.0114 ± 0.0665 0.0355 ± 0.0892 0.4349 ± 0.1101 Different 0.0010 ± 0.0139 0.0118 ± 0.0314 0.3996 ± 0.0760 P-value 1.2785 ×10−13 1.0423 ×10−52 2.1257 ×10−54 CTD Same 0.0361 ± 0.1590 0.0728 ± 0.1754 0.4863 ± 0.1770 Different 0.0050 ± 0.0274 0.0333 ± 0.0662 0.4408 ± 0.1368 P-value 1.4887 ×10−23 1.4040 ×10−9 2.0240 ×10−25 FunDO Same 0.0418 ± 0.1344 0.0991 ± 0.1611 0.5560 ± 0.2214 Different 0.0100 ± 0.0262 0.0549 ± 0.0830 0.4952 ± 0.1636 P-value 1.7609 ×10−144 9.6708 ×10−100 2.7037 ×10−90 HuGENet Same 0.0931 ± 0.1798 0.2470 ± 0.2123 0.8031 ± 0.2248 Different 0.0438 ± 0.0566 0.1881 ± 0.1522 0.7837 ± 0.2292 P-value 1.4585 ×10−74 9.9053 ×10−72 4.5910 ×10−14 Intersection Same 0.0338 ± 0.1511 0.0593 ± 0.1907 0.3826 ± 0.1131 Different 0.0024 ± 0.0329 0.0089 ± 0.0428 0.3496 ± 0.1020 P-value 2.2667 ×10−2 2.7448 ×10−4 5.4716 ×10−4 Union Same 0.0350 ± 0.1179 0.0963 ± 0.1463 0.5680 ± 0.2226 Different 0.0085 ± 0.0219 0.0583 ± 0.0818 0.5042 ± 0.1716 P-value 1.3493 ×10−211 7.1478 ×10−113 4.1709 ×10−141 Numbers in the table are similarity scores between diseases from the same ICD-9 categories, compared with those from different ICD-9 categories. P-values are calculated by using the Mann−Whitney U test. Table 2 Evaluation of our measures against ICD-9 classification Data Group Annotation-based Numbers in the table are similarity scores between diseases from the same ICD-9 categories, compared with those from different ICD-9 categories. P-values are calculated by using the Mann−Whitney U test. Additional file 1: Figure S5 and Table 3 show the ROC curves and Area Under Curve (AUC) values obtained by the three disease similarity measures. To illustrate that our results cannot be obtained by chance, we assigned a randomised score which was drawn from the same distri- bution of the similarity scores to each pair of diseases, and evaluated associations derived from these randomised scores against comorbidity. We show that the correla- tion between our similarity measures and comorbidity scores is substantially higher than expected at random for 1−specificity) for different thresholds of similarity score. Evaluation of similarity measures Correlation with ICD-9 The φ-correlation threshold was set to 0.06 (the same threshold was used in [47]), and all diseases annotated with at least 3 genes were evaluated. Average AUC values obtained by using randomised scores are shown by numbers in brackets (standard deviations are not shown in the table due to space limitation). Each evaluation test was run 30 times to compute the statistics reported in the table. Table 3 Evaluation of our measures against comorbidity Numbers in the table are AUC values obtained by evaluating the three disease similarity measures against comorbidity associations. The φ-correlation threshold was set to 0.06 (the same threshold was used in [47]), and all diseases annotated with at least 3 genes were evaluated. Average AUC values obtained by using randomised scores are shown by numbers in brackets (standard deviations are not shown in the table due to space limitation). Each evaluation test was run 30 times to compute the statistics reported in the table. Numbers in the table are AUC values obtained by evaluating the three disease similarity measures against comorbidity associations. The φ-correlation threshold was set to 0.06 (the same threshold was used in [47]), and all diseases annotated with at least 3 genes were evaluated. Average AUC values obtained by using randomised scores are shown by numbers in brackets (standard deviations are not shown in the table due to space limitation). Each evaluation test was run 30 times to compute the statistics reported in the table. 3.2.112) in June 2014 to re-examine whether we obtained the same results when we used the latest biological data. In total, the OMIM data contained 4,002 diseases (anno- tated by OMIM IDs), 3,218 genes and 4,816 disease-gene associations. The PPI network we constructed contained 14,089 nodes and 126,891 edges. By re-computing the similarity scores and evaluating the results against comor- bidity on these latest biological data, we showed that we were able to obtain results (shown in Additional file 1: Figure S4) that agree with the ones reported in Table 3 and Additional file 1: Figure S5. These results further validated the robustness of our approach. all disease-gene association datasets we analysed. In par- ticular, diseases yielding a high similarity score are very likely to have comorbidity associations, thus confirming that our measures are able to uncover known comorbidity relationships. Evaluation of similarity measures Correlation with ICD-9 TPR is defined as the fraction of true positives (that is, all pairs of diseases having a similarity score higher than a chosen threshold and having comorbidity association) out of the positives (all pairs of diseases having comorbidity association), while FPR is defined as the fraction of false positives (all pairs of diseases having a similarity score higher than a chosen threshold but having no comorbid- ity association) out of the negatives (all pairs of diseases excluding those having comorbidity association). Figure 2, Figure 2 Evaluation against comorbidity. ROC curves obtained by evaluating the three disease similarity measures against comorbidity. Due to space limitations, only ROC curves of FunDO are shown here (see Additional file 1: Figure S5 for ROC curves of other datasets). The φ-correlation threshold was set to 0.06 (the same threshold was used in [47]). We evaluated diseases annotated with at least 1, 3, 5, 7, 10, 15 genes, shown by curves with different colours in each plot. Figure 2 Evaluation against comorbidity. ROC curves obtained by evaluating the three disease similarity measures against comorbidity. Due to space limitations, only ROC curves of FunDO are shown here (see Additional file 1: Figure S5 for ROC curves of other datasets). The φ-correlation threshold was set to 0.06 (the same threshold was used in [47]). We evaluated diseases annotated with at least 1, 3, 5, 7, 10, 15 genes, shown by curves with different colours in each plot. Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 Page 9 of 13 Page 9 of 13 Table 3 Evaluation of our measures against comorbidity Data Annotation-based Function-based Topology-based OMIM 0.8009 ± 0.0277 (0.5740) 0.8694 ± 0.0073 (0.5120) 0.8495 ± 0.0011 (0.5044) CTD 0.7849 ± 0.0164 (0.5404) 0.7316 ± 0.0046 (0.5047) 0.7949 ± 0.0042 (0.5203) FunDO 0.7426 ± 0.0088 (0.4672) 0.7142 ± 0.0017 (0.4940) 0.7497 ± 0.0016 (0.5031) HuGENet 0.7563 ± 0.0001 (0.5084) 0.8185 ± 0.0001 (0.4987) 0.7153 ± 0.0015 (0.4922) Intersection 0.9925 ± 0.0001 (0.6013) 0.9802 ± 0.0001 (0.5081) 0.9958 ± 0.0041 (0.4664) Union 0.8225 ± 0.0045 (0.4704) 0.7491 ± 0.0001 (0.4999) 0.7939 ± 0.0022 (0.5008) Average 0.8194 ± 0.0837 (0.5270) 0.8106 ± 0.0930 (0.5029) 0.8163 ± 0.0907 (0.4979) Numbers in the table are AUC values obtained by evaluating the three disease similarity measures against comorbidity associations. Evaluation of similarity measures Correlation with ICD-9 p While varying the φ-correlation threshold, we obtained higher AUC values for higher thresholds (the ROC curves are not shown in the paper due to space limitations). For example, when the φ-correlation threshold was set to 0.06 (49 comorbidity pairs), the AUC value was 0.7580 ± 0.0024 (using the topology-based measure and FunDO as the source of disease-gene associations). When the φ-correlation threshold was set to 0.08 (33 comorbid- ity pairs) and 0.10 (25 comorbidity pairs), the AUC value increased to 0.7669 ± 0.0027 and 0.7996 ± 0.0060, respectively. This indicates our similarity measures tend to detect strong comorbidity associations with high φ- correlation. Meanwhile, when we decreased the number of false negatives in the comorbidity data by lowering the φ-correlation threshold from 0.06 to 0.02, the AUC values we obtained were still higher than expected at random. For example, when the φ-correlation threshold was set to 0.04 (93 comorbidity pairs) and 0.02 (300 comorbid- ity pairs), the AUC values we obtained were 0.7064 ± 0.0019 and 0.6017 ± 0.0015, respectively. These results suggest our similarity measures are robust to high false negatives in the comorbidity data. Better ROC curves can also be obtained by evaluating diseases annotated with higher numbers of genes (Additional file 1: Figure S5). From Table 3, we observed that best performances of our similarity measures are achieved by using highly confident curated disease-gene associations (i.e. the intersection set of OMIM, CTD and FunDO), with AUC values higher than 0.98. Correlation with GWAS data f h d h We further examined the correlation between our pre- dicted disease associations and currently available highly confident GWAS data (see the Methods section for details) to see whether our findings are supported by GWAS studies. A gene is said to be associated with a disease according to GWAS, if the occurrence of genetic variants (SNPs) within that gene is significantly higher in people with that disease than in people without it. We say that two diseases are associated according to GWAS if they share at least one gene in GWAS data. Since disease-gene associations collected in the four datasets we analysed were extracted from the literature, genetic associations reported in GWAS studies may also be col- lected in these datasets. To avoid bias in evaluation, we chose FunDO as the source of disease-genes associations, as it has few overlap with GWAS data. In particular, since most GWAS data were published after FunDO’s last stable release (October 2008), only 42 out of 48,436 publications in FunDO were also found in GWAS data. We removed disease-gene associations collected from the common 42 publications before computing similarity scores between diseases using FunDO. Similar to our eval- uation against comorbidity, we used ROC curve analysis to assess the ability of our similarity measures to recover Note that our approach is robust to the incompleteness presented in disease-gene association datasets and PPI networks. We downloaded the disease-gene association data from OMIM and the PPI data from BioGRID (version Page 10 of 13 Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 disease associations derived from GWAS (Table 4). For each of the three measures, we found that the correla- tion between our similarity measures and GWAS data is substantially higher than expected at random. This result further confirms the validity of our methods. ease pairs as our predicted disease associations, 14% ∼ 38% of the predictions are supported by all three similarity measures. In the topology-based measure, we used two terms, namely AllSig(Di, Dj) and ShareSig(Di, Dj), to measure the topological simialrity of disease genes. Since the term AllSig(Di, Dj) is defined as the maximum of the signature similarity between a gene associated with disease Di and a gene associated with disease Dj, we have AllSig(Di, Dj) = 1 if the two diseases Di and Dj have at least one com- mon genes. Correlation with GWAS data f h d h The term ShareSig(Di, Dj) is defined as the maximum of the signature similarity between genes that are shared between diseases Di and Dj, thus we have ShareSig(Di, Dj) = 0 if the two diseases share no genes. Therefore, the topology-based similarity score for a pair of diseases that share genes is always higher than a pair of diseases that do not share genes. To assess the contribu- tion of the two terms, AllSig and ShareSig, in predicting disease associations, we evaluated the performance of the topology-based similarity measure for predicting comor- bidity associations by solely using AllSig and ShareSig as the disease similarity score. The good performance of the topology-based similarity measure is mainly attributed to the term AllSig when using OMIM or CTD as the disease-gene association dataset (Additional file 1: Table S3). Since in these two datasets, only 2.69% (OMIM) and 16.62% (CTD) disease pairs have common genes, we have Simtopology = AllSig for most disease pairs. On the other hand, the good performance of the topology-based mea- sure is mainly caused by ShareSig when using FunDO or HuGENet, as in these two datasets 31.41% (FunDO) and 80.57% (HuGENet) of disease pairs have common genes. Comparison of similarity measures The three similarity measures, namely annotation-based measure, function-based measure, and topology-based measure, use different biological information to predict disease associations. For a pair of diseases, the annotation- based measure estimates their similarity score based on the overlap of their annotated genes, while the function- based measure estimates their similarity score based on the overlap of their associated biological functions derived from GO annotations. The topology-based mea- sure makes use of the topology information derived from the underlying PPI network, and estimates disease simi- larity scores based on the topological similarity of their annotated genes. Based on our evaluation, the three simi- larity measures perform well in recovering known disease associations. Note that since all three measures compare diseases based on information derived from their associ- ated genes, the three measures are not independent from each other. Diseases that have many shared genes are likely to have common biological processes and have high topo- logical similarities. In addition, a part of the GO annota- tions is inferred from PPIs (i.e., annotations with evidence code ‘inferred from physical interactions’). However, even though dependency between the three measures exists, the three measures uncover different aspects of disease- disease associations. In fact, the predictions derived from them can differ from each other, demonstrating that the three measures give different insights despite being dependent. Additional file 1: Figure S3 shows the overlap of disease associations predicted by the three measures. When considering the top 5% of the most associated dis- Our similarity measures are sensitive to the noise in disease-gene association data. We notice that pre- diction performances of our similarity measures gen- erally decrease with the increase of noise level, thus using the intersection of curated disease-gene association datasets results in the best performance when predicting comorbidity associations (Table 3 and Additional file 1: Figure S5). Both the annotation-based measure and the topology-based measure have better performances by using curated disease-gene associations (i.e., OMIM, CTD and FunDO) than computationally predicted associations (i.e., HuGENet). However, the function-based measure obtains lower AUC values for curated datasets CTD and FunDO than the two other similarity measures, but higher AUC values for HuGENet. In this regard, the function- based measure may be more appropriate for analysing predicted datasets, while the annotation-based measure and topology-based measure may be more appropriate for analysing curated datasets. Comparison of similarity measures Table 4 Evaluation of our measures against GWAS Data Annotation- Function- Topology- based based based F/G 0.7224 ± 0.0010 0.6781 ± 0.0001 0.6863 ± 0.0009 (0.4945) (0.4968) (0.5005) Common 0.7527 ± 0.0010 0.7147 ± 0.0001 0.7555 ± 0.0020 (0.4926) (0.5005) (0.4951) Numbers in the table are AUC values obtained by evaluating the three disease similarity measures against disease associations derived from highly confident GWAS data. Only diseases annotated with at least 3 genes were evaluated. ‘F/G’ are diseases having associated genes in both FunDO and GWAS data (99 diseases in total). ‘Common’ are diseases having associated genes in all four disease-gene association datasets (given in Figure 1) and GWAS data (50 diseases in total). Average AUC values obtained by using randomised scores are shown by numbers in brackets (standard deviations are not shown in the table due to space limitation). Each evaluation test was run 30 times to compute the statistics reported in the table. Table 4 Evaluation of our measures against GWAS Data Annotation- Function- Topology- based based based F/G 0.7224 ± 0.0010 0.6781 ± 0.0001 0.6863 ± 0.0009 (0.4945) (0.4968) (0.5005) Common 0.7527 ± 0.0010 0.7147 ± 0.0001 0.7555 ± 0.0020 (0.4926) (0.5005) (0.4951) Numbers in the table are AUC values obtained by evaluating the three disease similarity measures against disease associations derived from highly confident GWAS data. Only diseases annotated with at least 3 genes were evaluated. ‘F/G’ are diseases having associated genes in both FunDO and GWAS data (99 diseases in total). ‘Common’ are diseases having associated genes in all four disease-gene association datasets (given in Figure 1) and GWAS data (50 diseases in total). Average AUC values obtained by using randomised scores are shown by numbers in brackets (standard deviations are not shown in the table due to space limitation). Each evaluation test was run 30 times to compute the statistics reported in the table. Table 4 Evaluation of our measures against GWAS Numbers in the table are AUC values obtained by evaluating the three disease similarity measures against disease associations derived from highly confident GWAS data. Only diseases annotated with at least 3 genes were evaluated. ‘F/G’ are diseases having associated genes in both FunDO and GWAS data (99 diseases in total). ‘Common’ are diseases having associated genes in all four disease-gene association datasets (given in Figure 1) and GWAS data (50 diseases in total). Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 to estimate similarity scores, for a pair of diseases shar- ing few genes, their annotation-based similarity score may be low, even if their annotated genes are closely related. In particular, the annotation-based measure is highly affected by the occurrence of pleiotropic genes (genes that cause multiple phenotypes) in the dataset. We obtained the list of 802 pleiotropic genes from the OMIM Morbidmap by identifying genes that associated with more than one disease (similar approach was used in [53]). To examine the influence of pleiotropic genes on our measures, we excluded these genes from OMIM and evaluated the performances of our similarity measures against comorbidity. Note that when pleiotropic genes were excluded from OMIM, there were no disease pairs that had any common genes. Therefore, the annotation- based similarity score for a pair of diseases became 0 in this case and no predictions could be derived from the annotation-based measure. On the other hand, since both the function-based measure and the topology-based mea- sure use additional data sources (GO annotations or net- work topology) to estimate similarity scores, they are less affected by pleiotropic genes. AUC values obtained by the function-based measure and the topology-based measure dropped to 0.7816 and 0.7199 respectively, after remov- ing pleitropic genes from OMIM. These results show the contribution of similarities between specific genes (genes associated with only one disease) to the prediction perfor- mances of our similarity measures. heart disease [54]. Table 5 lists the top 10 diseases asso- ciated with DM using the topology-based measure and FunDO as the source of disease-gene associations. Results obtained by other measures and data are not shown here due to space limitations. p Among these 10 diseases, both ovarian dysfunction (ICD-9 code: 256) and obesity (ICD-9 code: 278) are clas- sified under the same ICD-9 catalogue ‘Endocrine, nutri- tional and metabolic diseases, and immunity disorders’ with DM. In addition, both obesity and essential hyperten- sion (ICD-9 code: 401) have highly confident comorbidity associations with DM. Note that among all disease pairs that we analysed, only 0.74% of them have a φ-correlation score higher than 0.06. Therefore, the φ-correlation scores reported in the case study (see Additional file 1: Table S4 and Additional file 1: Table S5 for details) are relatively high compared with the φ-correlation scores of all disease pairs. Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 Moreover, 6 out of 10 associations are supported by the GWAS data, e.g., rheumatoid arthritis shares 8 genes with DM according to GWAS data. Apart from the above, associations between DM and the remaining diseases listed in the table are considered as novel associations predicted by the topology-based measure. We evaluated the top 14 novel associations via mining the literature on PubMed (see Additional file 1: Table S4 for details). We are able to confirm all of these associations, includ- ing surprising associations such as DM and ‘other cerebral Since disease-gene association datasets were obtained by different research groups and approaches, good per- formances for all datasets confirm the robustness of our similarity measures in predicting disease associations. In addition, the topology-based measure is also robust to the noise and incompleteness presented in PPI networks. We evaluated this by using PPI data obtained from dif- ferent releases of BioGRID database (see Additional file 1: Table S1 for details). Generally speaking, the performance of the topology-based measure slightly decreases when using early PPI networks (Additional file 1: Table S2). However, AUC values obtained by using these early PPI networks are still substantially higher than expected at random. These results suggest that the ability of the topology-based measure to predict disease-disease associ- ations may increase with more accurate and complete PPI data. Table 5 List of the top 10 diseases associated with DM Rank Code Disease name Reference 1 239 Neoplasms of unspecified PMID: 23639840 nature 2 155 Malignant neoplasm of liver GWAS and intrahepatic bile ducts 3 710 Diffuse diseases of connective GWAS tissue 4 714 Rheumatoid arthritis and other GWAS inflammatory polyarthropathies 5 256 Ovarian dysfunction ICD-9, GWAS 5 278 Overweight, obesity and ICD-9, comorbidity, other hyperalimentation GWAS 7 401 Essential hypertension Comorbidity 8 295 Schizophrenic disorders PMID: 17474808 9 282 Hereditary hemolytic anemias GWAS 10 289 Other diseases of blood and PMID: 11727971 blood-forming organs Table 5 List of the top 10 diseases associated with DM Comparison of similarity measures Average AUC values obtained by using randomised scores are shown by numbers in brackets (standard deviations are not shown in the table due to space limitation). Each evaluation test was run 30 times to compute the statistics reported in the table. Numbers in the table are AUC values obtained by evaluating the three disease similarity measures against disease associations derived from highly confident GWAS data. Only diseases annotated with at least 3 genes were evaluated. ‘F/G’ are diseases having associated genes in both FunDO and GWAS data (99 diseases in total). ‘Common’ are diseases having associated genes in all four disease-gene association datasets (given in Figure 1) and GWAS data (50 diseases in total). Average AUC values obtained by using randomised scores are shown by numbers in brackets (standard deviations are not shown in the table due to space limitation). Each evaluation test was run 30 times to compute the statistics reported in the table. The annotation-based measure is straightforward, but has relatively good performance according to our evalu- ation. However, as it only uses disease-gene associations Page 11 of 13 Page 11 of 13 4. van Driel MA, Bruggeman J, Vriend G, Brunner HG, Leunissen JA: A text-mining analysis of the human phenome. Eur J Human Genet 2006, 14(5):535–542. 5. Lage K, Karlberg EO, Størling ZM, Ólason PI, Pedersen AG, Rigina O, Hinsby AM, Tümer Z, Pociot F, Tommerup N, Moreau Y, Brunak S: A human phenome-interactome network of protein complexes implicated in genetic disorders. Nat Biotechnol 2007, 25(3):309–316. 6. Goh KI, Cusick ME, Valle D, Childs B, Vidal M, Barabasi AL: The human disease network. Proc Nat Acad Sci 2007, 104(21):8685–8690. 7. Li Y, Agarwal P: A pathway-based view of human diseases and disease relationships. PloS one 2009, 4(2):e4346. 8. Hu G, Agarwal P: Human disease-drug network based on genomic expression profiles. PLoS One 2009, 4(8):e6536. 9. Suthram S, Dudley JT, Chiang AP, Chen R, Hastie TJ, Butte AJ: Network-based elucidation of human disease similarities reveals common functional modules enriched for pluripotent drug targets. PLoS Comput Biol 2010, 6(2):e1000662. 10. Mathur S, Dinakarpandian D: Finding disease similarity based on implicit semantic similarity. J Biomed Inform 2012, 45(2):363–371. 11. Žitnik M, Janji´c V, Larminie C, Zupan B, Pržulj N: Discovering disease-disease associations by fusing systems-level molecular data. Sci Rep 2013, 3(3202). doi:10.1038/srep03202. 12. Huang W, Wang P, Liu Z, Zhang L: Identifying disease associations via genome-wide association studies. BMC Bioinformatics 2009, 10(Suppl 1):S68. 13. Kim S, Sohn KA, Xing EP: A multivariate regression approach to association analysis of a quantitative trait network. Bioinformatics 2009, 25(12):i204–i212. 14. Lewis SN, Nsoesie E, Weeks C, Qiao D, Zhang L: Prediction of disease and phenotype associations from genome-wide association studies. PloS One 2011, 6(11):e27175. 15. Lee DS, Park J, Kay K, Christakis N, Oltvai Z, Barabási AL: The implications of human metabolic network topology for disease comorbidity. Proc Nat Acad Sci 2008, 105(29):9880–9885. 16. Milenkovi´c T, Memiševi´c V, Bonato A, Pržulj N: Dominating biological networks. PloS one 2011, 6(8):e23016. Acknowledgements We thank members of GlaxoSmithKline (GSK) Computational Biology group, specifically Dr. Hannah Tipney and Dr. Peter Woollard for their helpful comments. This work was supported by the European Research Council (ERC) Starting Independent Researcher Grant 278212, the National Science Foundation (NSF) Cyber-Enabled Discovery and Innovation (CDI) OIA-1028394, ARRS project J1-5454, GSK Research & Development Ltd., and the Serbian Ministry of Education and Science Project III44006. Another case study (Parkinson’s disease, ICD-9 code: 332) can be found in the Additional file 1. Abbreviations p p 12. Huang W, Wang P, Liu Z, Zhang L: Identifying disease associations via genome-wide association studies. BMC Bioinformatics 2009, 10(Suppl 1):S68. GWAS: Genome-Wide Association Studies; OMIM: Online Mendelian Inheritance in Man; ICD: International Classification of Diseases; PPI: Protein-Protein Interaction; GO: Gene Ontology; IEA: Inferred from Electronic Annotation; CTD: Comparative Toxicogenomics Database; FunDO: Functional Disease Ontology; HuGENet: Human Genome Epidemiology Network; MeSH: Medical Subject Heading; GeneRIF: Gene Reference Into Function; DO: Disease Ontology; SNP: Single-Nucleotide Polymorphism; ROC: Receiver Operating Characteristic; TPR: True Positive Rate; FPR: False Positive Rate; AUC: Area Under Curve; DM: Diabetes Mellitus. 15. Lee DS, Park J, Kay K, Christakis N, Oltvai Z, Barabási AL: The implications of human metabolic network topology for disease comorbidity. Proc Nat Acad Sci 2008, 105(29):9880–9885. 15. Lee DS, Park J, Kay K, Christakis N, Oltvai Z, Barabási AL: The implications of human metabolic network topology for disease comorbidity. Proc Nat Acad Sci 2008, 105(29):9880–9885. Authors’ contributions Designed the experiments: KS, JG, CL, NP. Collected the data: KS, JG. Performed the experiments: KS. Interpreted the results: KS, JG, CL, NP. Wrote the manuscript: KS, CL, NP. Conceived and directed the research: NP. All authors read and approved the final manuscript. Additional file Additional file 1: Supplementary information. The Supplementary information file contains all additional figures and tables mentioned in the manuscript. 11. Žitnik M, Janji´c V, Larminie C, Zupan B, Pržulj N: Discovering disease-disease associations by fusing systems-level molecular data. Sci Rep 2013, 3(3202). doi:10.1038/srep03202. 11. Žitnik M, Janji´c V, Larminie C, Zupan B, Pržulj N: Discovering disease-disease associations by fusing systems-level molecular data. Sci Rep 2013, 3(3202). doi:10.1038/srep03202. p p 12. Huang W, Wang P, Liu Z, Zhang L: Identifying disease associations via genome-wide association studies. BMC Bioinformatics 2009, 10(Suppl 1):S68. Case studies To demonstrate how our similarity measures can be used for uncovering novel disease associations, we present a case study for diabetes mellitus (DM, ICD-9 code: 250). DM is a metabolic disease that affects the body’s ability to produce or use insulin, a hormone for regulating carbo- hydrates. It causes hyperglycemia and may lead to severe consequences such as brain damage, amputations and The top 10 diseases associated with DM were inferred using the topology-based similarity measure and FunDO as the source of disease-gene associations. Only diseases annotated in all four disease-gene association datasets are listed in the table. For a disease associated with DM according to ICD-9, comorbidity or GWAS, we added the supported evidence to the reference (the last column). The remaining disease associations were validated via mining the literature on PubMed (http://www.ncbi.nlm.nih.gov/pubmed), and for each disease only one reference (shown by PubMed ID) was listed in the table due to space limitation. Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 Page 12 of 13 Page 12 of 13 Page 12 of 13 Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 degenerations’ (ICD-9 code: 331). This result highlights the power of our approaches to identify novel associa- tions between diseases. Further exploration of potential underlying mechanisms shared by these diseases may lead to improvement in disease diagnosis, prognosis and treatment. Conclusions In this study, we gained novel insights into the relationship between human diseases by considering their molecu- lar causes and underlying physical interactions. We used information derived from latest biological data, including disease-gene associations, gene functions and the topol- ogy of the human PPI network in our analysis. We applied three different measures to estimate the similarity score of diseases, and these measures were systematically eval- uated against ICD-9 classification system, a statistical measure of comorbidity and GWAS data. Our results showed the correlation between associations predicted by our measures and known disease associations, and we also demonstrated the use of our measures in discovering novel disease associations and validated it via literature curation. Author details 1 1Department of Computing, Imperial College London, London, SW7 2AZ, UK. 2Computational Biology, GlaxoSmithKline, Stevenage, Hertfordshire, SG1 2NY, UK. Received: 9 December 2013 Accepted: 19 August 2014 Published: 17 September 2014 References 1. Gatza ML, Lucas JE, Barry WT, Kim JW, Wang Q, Crawford MD, Datto MB, Kelley M, Mathey-Prevot B, Potti A, Nevins JR: A pathway-based classification of human breast cancer. Proc Nat Acad Sci 2010, 107(15):6994–6999. 1. Gatza ML, Lucas JE, Barry WT, Kim JW, Wang Q, Crawford MD, Datto MB, Kelley M, Mathey-Prevot B, Potti A, Nevins JR: A pathway-based classification of human breast cancer. Proc Nat Acad Sci 2010, 107(15):6994–6999. 107(15):6994 6999. 2. Loscalzo J, Kohane I, Barabasi AL: Human disease classification in the postgenomic era: a complex systems approach to human pathobiology. Mol Syst Biol 2007, 3. doi:10.1038/msb4100163. 3. Hamosh A, Scott AF, Amberger JS, Bocchini CA, McKusick VA: Online Mendelian Inheritance in Man (OMIM), a knowledgebase of human genes and genetic disorders. Nucleic Acids Res 2005, 33(suppl 1):D514–D517. 2. Loscalzo J, Kohane I, Barabasi AL: Human disease classification in the postgenomic era: a complex systems approach to human pathobiology. Mol Syst Biol 2007, 3. doi:10.1038/msb4100163. 2. Loscalzo J, Kohane I, Barabasi AL: Human disease classification in the postgenomic era: a complex systems approach to human pathobiology. Mol Syst Biol 2007, 3. doi:10.1038/msb4100163. y y 3. Hamosh A, Scott AF, Amberger JS, Bocchini CA, McKusick VA: Online Mendelian Inheritance in Man (OMIM), a knowledgebase of human genes and genetic disorders. Nucleic Acids Res 2005, 33(suppl 1):D514–D517. y y 3. Hamosh A, Scott AF, Amberger JS, Bocchini CA, McKusick VA: Online Mendelian Inheritance in Man (OMIM), a knowledgebase of human genes and genetic disorders. Nucleic Acids Res 2005, 33(suppl 1):D514–D517. Novel disease associations uncovered in this study can be further used to improve our understanding of disease classification. For example, a human disease network that models the relationship of diseases can be constructed based on these similarity measures, and computational approaches, such as clustering, can be applied to detect communities in the disease network. This may provide the opportunity to redefine the current disease classification and further lead to improvements in disease diagnosis, prognosis and treatment. 4. van Driel MA, Bruggeman J, Vriend G, Brunner HG, Leunissen JA: A text-mining analysis of the human phenome. Eur J Human Genet 2006, 14(5):535–542. Competing interests Th h d l h 16. Milenkovi´c T, Memiševi´c V, Bonato A, Pržulj N: Dominating biological networks. PloS one 2011, 6(8):e23016. The authors declare that they have no competing interests. Page 13 of 13 Sun et al. BMC Bioinformatics 2014, 15:304 http://www.biomedcentral.com/1471-2105/15/304 17. Janji´c V, Pržulj N: The core diseasome. Mol BioSyst 2012, 8(10):2614–2625. 39. Milenkoviæ T, Pržulj N: Uncovering biological network function via graphlet degree signatures. Cancer Inform 2008, 6:257. 18. Ideker T, Sharan R: Protein networks in disease. Genome Res 2008, 18(4):644–652. 40. Milenkovi´c T, Ng WL, Hayes W, Pržulj N: Optimal network alignment with graphlet degree vectors. Cancer Informat 2010, 9:121. 19. Barabási AL, Gulbahce N, Loscalzo J: Network medicine: a network-based approach to human disease. Nat Rev Genet 2011, 12:56–68. 41. Ho H, Milenkovi´c T, Memiševi´c V, Aruri J, Pržulj N, Ganesan AK: Protein interaction network topology uncovers melanogenesis regulatory network components within functional genomics datasets. BMC Syst Biol 2010, 4:84. 20. Bauer-Mehren A, Bundschus M, Rautschka M, Mayer MA, Sanz F, Furlong LI: Gene-disease network analysis reveals functional modules in mendelian, complex and environmental diseases. PloS one 2011, 6(6):e20284. 42. Kuchaiev O, Milenkovi´c T, Memiševi´c V, Hayes W, Pržulj N: Topological network alignment uncovers biological function and phylogeny. J R Soc Interface 2010, 7(50):1341–1354. 21. Becker KG, Barnes KC, Bright TJ, Wang SA: The genetic association database. Nat Genet 2004, 36(5):431–432. 43. Kuchaiev O, Pržulj N: Integrative network alignment reveals large regions of global network similarity in yeast and human. Bioinformatics 2011, 27(10):1390–1396. database. Nat Genet 2004, 36(5):431–432. 22. Oti M, Huynen MA, Brunner HG: Phenome connections. Trends Genet 2008, 24(3):103–106. 44. Hayes W, Sun K, Pržulj N: Graphlet-based measures are suitable for biological network comparison. Bioinformatics 2013, 29(4):483–491. 23. ICD 9 CM: The International Classification of Diseases. 9. Rev: Clinical Modification.; Vol. 1: Diseases: Tabular List.; Vol. 2: Diseases: Alphabetic Index.; Vol 3: Procedures: Tabular List and Alphabetic Index. US Government Printing Office, 1980. http://www.who.int/classifications/ icd/en/. 45. Davis DA, Chawla NV: Exploring and exploiting disease interactions from multi-relational gene and phenotype networks. PloS one 2011, 6(7):e22670. 24. Stelzl U, Worm U, Lalowski M, Haenig C, Brembeck FH, Goehler H, Stroedicke M, Zenkner M, Schoenherr A, Koeppen S, Timm J, Mintzlaff S, Abraham C, Bock N, Kietzmann S, Goedde A, Toksöz E, Droege A, Krobitsch S, Korn B, Birchmeier W, Lehrach H, Wanker EE: A human protein-protein interaction network: a resource for annotating the proteome. Cell 2005, 122(6):957–968. 46. Competing interests Th h d l h Sci Res Essay 2009, 4(5):367–373. 29. Osborne J, Flatow J, Holko M, Lin S, Kibbe W, Zhu L, Danila M, Feng G, Chisholm R: Annotating the human genome with disease ontology. BMC Genomics 2009, 10(Suppl 1):S6. doi:10.1186/1471-2105-15-304 Cite this article as: Sun et al.: Predicting disease associations via biological network analysis. BMC Bioinformatics 2014 15:304. 30. Yu W, Clyne M, Khoury MJ, Gwinn M: Phenopedia and Genopedia: disease-centered and gene-centered views of the evolving knowledge of human genetic associations. Bioinformatics 2010, 26:145–146. 31. Lipscomb CE: Medical subject headings (MeSH). Bull Med Library Assoc 2000, 88.3:265. http://www.nlm.nih.gov/mesh/. 32. Bodenreider O: The unified medical language system (UMLS): integrating biomedical terminology. Nucleic Acids Res 2004, 32(1):D267–D270. http://www.nlm.nih.gov/research/umls/. Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit 33. Park J, Lee DS, Christakis NA, Barabási AL: The impact of cellular networks on disease comorbidity. Mol Syst Biol 2009, 5:1–7. 34. Milenkovi´c T, Memiševi´c V, Ganesan AK, Pržulj N: Systems-level cancer gene identification from protein interaction network topology applied to melanogenesis-related functional genomics data. J R Soc Interface 2010, 7(44):423–437. 35. Competing interests Th h d l h Park S, Yang JS, Kim J, Shin YE, Hwang J, Park J, Jang SK, Kim S: Evolutionary history of human disease genes reveals phenotypic connections and comorbidity among genetic diseases. Sci Rep 2012, 2:757. 47. Hidalgo CA, Blumm N, Barabási AL, Christakis NA: A dynamic network approach for the study of human phenotypes. PLoS Comput Biol 2009, 5(4):e1000353. p 25. Krogan NJ, Cagney G, Yu H, Zhong G, Guo X, Ignatchenko A, Li J, Pu S, Datta N, Tikuisis AP, Punna T, Peregrín-Alvarez JM, Shales M, Zhang X, Davey M, Robinson MD, Paccanaro A, Bray JE, Sheung A, Beattie B, Richards DP, Canadien V, Lalev A, Mena F, Wong P, Starostine A, Canete MM, Vlasblom J, Wu S, Orsi C, et al.: Global landscape of protein complexes in the yeast Saccharomyces cerevisiae. Nature 2006, 440(7084):637–643. 48. Hirschhorn JN, Daly MJ: Genome-wide association studies for common diseases and complex traits. Nat Rev Genet 2005, 6(2):95–108. 49. McCarthy MI, Abecasis GR, Cardon LR, Goldstein DB, Little J, Ioannidis JP, Hirschhorn JN: Genome-wide association studies for complex traits: consensus, uncertainty and challenges. Nat Rev Genet 2008, 9(5):356–369. 26. Stark C, Breitkreutz BJ, Reguly T, Boucher L, Breitkreutz A, Tyers M: BioGRID: a general repository for interaction datasets. Nucleic Acids Res 2006, 34(suppl 1):D535–D539. 50. Hindorff LA, MacArthur J, Morales J, Junkins HA, Hall PN, Klemm AK, Manolio TA: A catalog of published genome-wide association studies. 2011. http://www.genome.gov/gwastudies/. 27. Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, Davis AP, Dolinski K, Dwight SS, Eppig JT, Harris MA, Hill DP, Issel-Tarver L, Kasarskis A, Lewis S, Matese JC, Richardson JE, Ringwald M, Rubin GM, Sherlock G: Gene Ontology: tool for the unification of biology. Nat Genet 2000, 25:25–29. 51. Sanseau P, Agarwal P, Barnes MR, Pastinen T, Richards JB, Cardon LR, Mooser V: Use of genome-wide association studies for drug repositioning. Nat Biotechnol 2012, 30(4):317–320. 52. Han JDJ, Dupuy D, Bertin N, Cusick ME, Vidal M: Effect of sampling on topology predictions of protein-protein interaction networks. Nat Biotechnol 2005, 23(7):839–844. 28. Davis AP, King BL, Mockus S, Murphy CG, Saraceni-Richards C, Rosenstein M, Wiegers T, Mattingly CJ: The comparative toxicogenomics database: update 2011. Nucleic Acids Res 2011, 39(suppl 1):D1067–D1072. 53. Chavali S, Barrenas F, Kanduri K, Benson M: Network properties of human disease genes with pleiotropic effects. BMC Syst Biol 2010, 4:78. 54. Samreen R: Diabetes mellitus: a review. Competing interests Th h d l h Sarajli´c A, Janji´c V, Stojkovi´c N, Radak D, Pr˙zulj N: Network topology reveals key cardiovascular disease genes. PloS One 2013, 8(8):e71537. 36. Memiševi´c V, Milenkovi´c T, Pržulj N: Complementarity of network and sequence structure in homologous proteins. J Integrative Bioinform 2010, 7(3):135. 37. Pržulj N, Corneil DG, Jurisica I: Modeling interactome: scale-free or geometric? Bioinformatics 2004, 20(18):3508–3515. 38. Pržulj N: Biological network comparison using graphlet degree distribution. Bioinformatics 2007, 23(2):e177–e183.
https://openalex.org/W4288086092	https://cadmus.eui.eu/bitstream/1814/66408/2/stanoeva_chapter%20Margins%20for%20Maneuver.pdf	English	null	Margins for Manoeuvre in Cold War Europe	Routledge eBooks	2,019	public-domain	10,448	The Imperative of Opening to the West and the Impact of the 1968 Crisis: Bulgaria’s Cooperation with Denmark and West Germany in the 1960s Elitza Stanoeva Introduction Exemplifying small-state diplomacy in the straitjacket of Soviet geopolitics, the foreign policy of socialist Bulgaria was a matter of precarious ‘manoeuvring’ – a word that Todor Zhivkov, head of state and party until 1989, often used to self-congratulate his political shrewdness.1 Towards the West, his was a strategy of manoeuvring national interests in uncharted waters around the ‘icebergs’ of superpower geopolitics. While Cold War historiography usually emphasises Bulgaria’s unfaltering allegiance to the Soviet Union, thereby dismissing its foreign policy as a reiteration of Soviet positions, the argument in this chapter is that Bulgaria’s alignment in the Soviet sphere of influence acted as both a propeller and a brake for its ambitions towards the West. Fear of remaining more insulated than the other people’s democracies in Eastern Europe propelled its engagement with Western countries and gave it a competitive urge vis-à-vis its allies. Being a smaller power – both in a geopolitical and an economic sense – shaped Bulgaria’s foreign policy in terms of goals, opportunities and constraints, but also pressures to adapt, at times sacrificing ideological orthodoxy for economic benefits or vice versa, within the changing parameters of the Cold War. While the party leadership manoeuvred along the fine line of national interests and Soviet allegiance, the operatives in the Bulgarian foreign services were trying to maintain their own room for manoeuvre against the interference of political dogmatism. Economic cooperation with the West, the primary engine of the Bulgarian diplomatic pursuits outside the Eastern bloc, allowed the foreign services to take the lead in the 1960s, a period when East–West relations made a shift from ‘hostile confrontation to antagonistic cooperation.’2 In this decade, 1 when the party leadership was denied direct access to Western governments, the state administration was responsible for the initial diplomatic advances. Implementing Bulgaria’s foreign policy towards the West, ministers and technocrats enjoyed enough room for manoeuvre to infuse this policy with their pragmatic priorities and to steer the nascent cooperation in directions that would still guide intergovernmental relations once the party took a more active role in the 1970s. This chapter examines Bulgaria’s parallel political and economic relations with two countries across the ‘Iron Curtain’, the Federal Republic of Germany (FRG) and Denmark: their (re)activation in the early 1960s, peak around the mid-1960s, deterioration in the aftermath of the crackdown on the Prague Spring in 1968, and the subsequent efforts at recovery. Introduction This comparison has several important premises. Firstly, while the FRG was by far the most important economic partner of Bulgaria in the West, Denmark in contrast was one of the smallest among the European nations. Comparing Bulgaria’s foreign policy towards the two countries demonstrates how strongly the imperative of opening to the West shaped its outreach and how much the concerns of lagging behind the rest of the Soviet bloc steered its course. Secondly, while Denmark, an ardent supporter of East–West rapprochement, was at the sideline of the Cold War confrontation, the FRG was at the very centre of it. Therefore, Bulgaria’s engagement with the two countries evolved within profoundly different constellations of opportunities and limitations for small-state diplomacy. Shaped by divergent considerations of costs and benefits, Bulgaria’s overtures towards Denmark and the FRG differed both on the level of policy-making (in the degree of party oversight) and on the level of policy-implementation (in the scope of autonomous initiative held by state operatives). The comparison therefore exposes not only the range of Bulgaria’s strategies for manoeuvre in its diplomacy vis-à-vis Western countries but also the range of interpretations of foreign policy objectives within its party and state apparatuses. Furthermore, it challenges 2 the common assumption that the Bulgarian Communist Party, Balgarska Komunisticheska Partiya (BKP) and its general secretary held all decision-making powers in foreign as well as economic policy whereas the state institutions were demoted to a passive role.3 This chapter’s contribution to the existing scholarship, therefore, is to delineate the scope of Bulgaria’s small-state diplomacy outside of Soviet tutelage and then to demonstrate that by infusing it with the pragmatic objectives of economic cooperation, state institutions managed to diminish its ideological dogmatism. Bulgaria’s cooperation with the West was prompted by the post-Stalinist economic reforms which aimed to modernise the country’s industrial base. While Bulgarian research focuses on the domestic impacts of these reforms, little attention is given to their links with Bulgaria’s foreign trade and the entailed pursuits of economic partnerships abroad.4 In fact, their success was tied, from the onset, to an increase of both costly imports of production equipment, industrial inputs and knowhow from technologically advanced Western economies and of high-value exports to capitalist markets that could replenish the hard-currency reserves.5 This called for the expansion of trade volume and diversification of traditional economic partnerships. Introduction As much as it kept in line with a general reformist thrust across the bloc, this strategy also stirred competition with fraternal regimes over lucrative transfers of Western technology and access to the more demanding Western markets. In their competitive undertakings, the Bulgarian foreign trade service worked hard to restore ties with old partners like (West) Germany, and to reach out into new markets like Denmark. Bulgaria’s cooperation with the West was prompted by the post-Stalinist economic reforms which aimed to modernise the country’s industrial base. While Bulgarian research focuses on the domestic impacts of these reforms, little attention is given to their links with Bulgaria’s foreign trade and the entailed pursuits of economic partnerships abroad.4 In fact, their success was tied, from the onset, to an increase of both costly imports of production equipment, Through the prism of small-state diplomacy in the Cold War context, the comparison below aims to gauge the interests that motivated Bulgaria’s opening to the West as well as the conjunctural opportunities and structural limitations it faced in carving out a foreign policy course against the backdrop of growing pan-European cooperation.6 This study is based on declassified documents from the Central State Archive of Bulgaria: Politburo resolutions and 3 documentation compiled by the party’s Department ‘Foreign Policy and International Relations’ – contained in the BKP repositories (fund 1B); bilateral agreements ratified by the Council of Ministers (CM, f.136); internal correspondence and memos of meetings with foreign officials as well as annual reports and other information from the Ministry of Foreign Affairs (MFA, f.1477), the Ministry of Foreign Trade (MFT, f.259) and the Commission for Economic and Scientific-Technical Cooperation (f.1244). Reports of the Radio Free Europe research unit (Vera and Donald Blinken Open Society Archive) are also consulted to assess the repercussions of Bulgarian domestic and international activities in the West. Mapping synchronous negotiations and comparable milestones in Bulgaria’s bilateral relations with Denmark and West Germany, and contextualising them vis-à-vis external geopolitical processes and internal agenda-setting, facilitates more general insights about Bulgaria’s diplomacy and its room for manoeuvre in a particularly dynamic period of the Cold War. Introduction Finally, investigating the parallel document trails of the party and the state apparatus (and multiple sub-levels) reveals less straightforward links between (party) decision-making and (state) policy-implementation than the ‘party–state’ paradigm suggests, and brings to the fore a more ambiguous match of ideology and pragmatism than the ‘Soviet satellite’ perspective allows for. decision-making and (state) policy-implementation than the ‘party–state’ paradigm suggests, and brings to the fore a more ambiguous match of ideology and pragmatism than the ‘Soviet satellite’ perspective allows for. Bulgaria’s pursuit of bilateral contacts with Denmark in the 1960s Under the imperative of opening to the West, Denmark – a country of minimal prior contact – entered the scope of Bulgaria’s foreign-policy pursuits in the late 1950s. Distant not only politically but also geographically, Denmark would subsequently become a partner that would allow Bulgaria some room for independent foreign-policy initiatives, as bilateral contacts were safely detached from any sensitive geopolitical issue, and hence developed under the Soviet radar. Yet, the opportunities it presented for Bulgaria’s diplomacy were to a large extent secured because of their intrinsic limitations. In their entry into Denmark, the Bulgarian 4 foreign services followed in the steps of East European countries with stable historical contacts there, particularly the Soviet Union, Poland and East Germany. This simultaneously constrained the breadth of Bulgarian deals with Denmark as secondary in importance to other socialist regimes, and infused the underlying cooperation strategy with competitiveness vis-à- vis the rest of the bloc. In this strategy, foreign trade was both a means and an end. On the one hand, trade was anticipated to break the ice in bilateral communication. On the other hand, its growth alone was an incentive in light of Bulgaria’s weak position in international trade. Denmark, for its part, was responsive to the Bulgarian advances inasmuch as they chimed with its foreign policy of rapprochement with the East.7 While keeping the door open due to its staunch support for détente, the Danish government did not have much vested interest in the potential practical gains of economic cooperation with Bulgaria. In Bulgaria’s opening to the West, Denmark did not hold a top priority, but it was nevertheless an opportunity not to miss. The initial contacts were thus delegated to lower-level institutional entities. Although several tracks of fostering partnerships in Denmark were activated, each was developed by a limited number of individuals representing their respective state institutions. The small teams they formed could operate side by side in Denmark without disrupting each other’s agendas. First to break the ice in Bulgarian diplomacy towards Denmark were the foreign trade services. Bulgaria’s pursuit of bilateral contacts with Denmark in the 1960s In 1960, the MFT opened a trade representation in Copenhagen (Bultarg).8 At the time, Bulgaria still did not have a full-fledged diplomatic representation – as of 1957 the ambassador in Sweden was also accredited to Denmark and Norway.9 Enjoying a degree of autonomy under its loose ties to the Stockholm mission, the Bulgarian trade representatives in Copenhagen were tasked not only with building networks of economic cooperation from scratch but also with laying the grounds for intergovernmental contacts. Yet, for the envoys, expansion of economic relations would become their primary objective. 5 What characterised the overall diplomatic strategy of Bulgaria towards Denmark – opportunities to operate free of ideological pressures and economic competitiveness vis-à-vis allies – would shape also the operation of the Copenhagen trade office on the micro-level of the Bulgarian state apparatus. Its lower rank within the foreign trade sector allowed its employees greater autonomy from centralised oversight. But their office also had less political pull within the command chain of foreign trade. In the inevitable competition over high- quality exports resulting from Bulgaria’s limited production capabilities, Bultarg Copenhagen was at a disadvantage compared to larger and politically prioritised missions as the one in West Germany. Accordingly, when the Copenhagen mission strove to convince the administration at home of the untapped potential of economic cooperation with Denmark, it partly acted upon its own micro-institutional stakes, trying to expand its own room for manoeuvre within the foreign-trade sector. As Denmark did not open a reciprocal trade service and handled diplomatic relations with Bulgaria through its embassy in Warsaw, the Bulgarian trade envoys brokered and maintained contacts with private firms and state institutions in Denmark. The incipient partnerships were highly personalised and shaped by a shared business ethos, which however, would often be challenged by the failure of Bulgarian enterprises to deliver upon agreed terms. In ensuing disputes, Bultarg Copenhagen tended to take the side of the Danish party: even in controversial cases it often blamed organisations at home when reporting to the MFT. While the commercial exchange with Denmark might have been negligible (below 0.1 percent of Bulgaria’s total trade volume), its growth was nevertheless an impressive accomplishment for the small foreign-trade service dedicated to the Danish market. Bulgaria’s pursuit of bilateral contacts with Denmark in the 1960s Starting from zero, the commodity flow between the two countries accounted for 0.6 million USD in 1960, then doubled its value by 1962.10 That year, a long-term trade agreement was negotiated with a validity of three years, subsequently renewed with an extended span of five years 6 (1966-1970).11 Under its arrangements, Bulgarian exports obtained a higher level of liberalisation and trade grew swiftly reaching approximately 5 million USD by 1968 but also generating a stable trade deficit for Bulgaria.12 Another track of economic cooperation with Denmark was tourism. Soon after the MFT opened a trade mission in Copenhagen, the Bulgarian tourist administration created its own affiliate there.13 Although Denmark was not the main provider of foreign tourists for the Bulgarian seaside resorts, its market was perceived as particularly promising due to Scandinavia’s high standards of living which rendered the expenditure of Scandinavian vacationers relatively higher than that of other Westerners.14 Just like initial economic partnerships were delegated to lower-ranking operatives with business expertise, the first political contacts between Bulgaria and Denmark were forged by peripheral actors with more discreet political profiles. In this area, the BKP lent the pioneering role to its subservient coalition partner, the Bulgarian Agrarian People’s Union, Balgarski Zemedelski Naroden Sayuz (BZNS). Void of any independent involvement in domestic policy-making, BZNS’ international activities were encouraged under a tightly supervised mandate and targeted specifically Western political parties and organisations that the communists could not approach.15 One of the testing grounds for this strategy was indeed the Scandinavian region and the first official visits in Denmark were all done by BZNS delegations. In 1964, a leading figure at the BZNS, Lalyu Ganchev, was dispatched to Stockholm to serve as ambassador to all four Scandinavian countries – a post of unprecedented importance for the union.16 As this proved to be a viable diplomatic channel, in 1965 the Politburo ratified BZNS’ international outreach as an integral part of Bulgaria’s foreign policy.17 Same year, the first Danish governmental visit in Sofia was undertaken by the Minister of Agriculture. 7 7 Active under a limited diplomatic mandate, the BZNS nevertheless promoted its agenda in the early contacts with Denmark. Bulgaria’s pursuit of bilateral contacts with Denmark in the 1960s The practical goals on his agenda for the trip contained several Bulgarian requests rejected by Krag (like the most- favoured nation clause).27 Reports from the BZNS on their parallel line of communication also signalled that the moment was ripe for negotiating better terms of commerce. During yet another visit in Copenhagen in April 1968, their delegation was received at the Danish Parliament where the new Prime Minister Hilmar Baunsgaard emphasised the trade links between the two countries as a propeller of rapprochement.28 Finally, the talks between Bashev and Krag included arrangements for a visit of Zhivkov to Copenhagen – which would have been the first visit ever of a Bulgarian prime minister to a Scandinavian country.26 His trip, moreover, was to include a second stop in Oslo. For the BKP, such a diplomatic breakthrough in the West was a PR stunt to bolster the international prestige of the country and its leader. From the viewpoint of the Bulgarian economic services, however, Zhivkov’s forthcoming visit was perceived as much as a diplomatic accomplishment as an opportunity for further economic gains. The practical goals on his agenda for the trip contained several Bulgarian requests rejected by Krag (like the most- favoured nation clause).27 Reports from the BZNS on their parallel line of communication also signalled that the moment was ripe for negotiating better terms of commerce. During yet another visit in Copenhagen in April 1968, their delegation was received at the Danish Parliament where the new Prime Minister Hilmar Baunsgaard emphasised the trade links between the two countries as a propeller of rapprochement.28 Bulgaria’s pursuit of bilateral contacts with Denmark in the 1960s The embassy in Stockholm saw Denmark as an agricultural country that in the post-war years had succeeded to mechanise its agriculture and transform its economy in a way that Bulgaria had long aspired and consistently failed to do.18 As agriculture still held the major share of the exports of socialist Bulgaria, its economic cooperation with Denmark focused on a transfer of knowhow and technology for optimising Bulgarian farming and stockbreeding. Steering economic cooperation towards agricultural modernisation, the BZNS was in fact building upon the legacy of its activities in Scandinavia before the war.19 Bilateral contacts ascended further up the ladder in 1967 with an exchange of visits by the Bulgarian Minister of Foreign Affairs, Ivan Bashev, and his Danish counterpart Jens Otto Krag, who also held the post of Prime Minister.20 The diplomatic accomplishments comprised an Agreement on Economic, Industrial and Technological Cooperation and an agreement for visa-free travel between the two countries.21 The latter was anticipated to bring more Danish vacationers at the moment when international tourism became recognised as a key source of hard-currency income for the Bulgarian economy.22 Further agreements for road transportation, consular services and cultural cooperation were put in preparation.23 In his meeting with Todor Zhivkov, Krag emphasised the need for de-escalation of tensions between East and West stating, reportedly, that the ‘old thesis of peaceful coexistence does not satisfy the requirements of today, active mutual cooperation is needed’.24 Despite the warm tone of the conversation with the party leader, the MFA placed at the top of its report to the Politburo, the prospects for further economic cooperation and the pressing issues of the Bulgarian trade deficit. Although Krag acknowledged Bulgaria’s troubles, he declined Zhivkov’s request for bank credits arguing that Denmark was itself an importer of capital.25 8 8 Finally, the talks between Bashev and Krag included arrangements for a visit of Zhivkov to Copenhagen – which would have been the first visit ever of a Bulgarian prime minister to a Scandinavian country.26 His trip, moreover, was to include a second stop in Oslo. For the BKP, such a diplomatic breakthrough in the West was a PR stunt to bolster the international prestige of the country and its leader. From the viewpoint of the Bulgarian economic services, however, Zhivkov’s forthcoming visit was perceived as much as a diplomatic accomplishment as an opportunity for further economic gains. Bulgaria’s responsiveness to the FRG’s diplomatic overtures in the 1960s Unlike its lack of prior contacts with Denmark, Bulgaria had a long history of binding its foreign policy to the German state. Before and during World War II, the Bulgarian economy was pegged to the German Reich. On the eve of the war, the bilateral commerce accounted for almost 70 percent of Bulgaria’s exports and imports – a much greater economic dependency compared to the German neighbours in the East (Hungary – below 50 percent, Poland and Czechoslovakia – below 25 percent).29 This legacy, though politically sensitive, made it only natural for Bulgaria to look towards post-war West Germany for a boost to its economy. Economic ties between the two countries carried on after the war but their pronounced expansion occurred in the 1960s.30 In contrast to Bulgarian–Danish cooperation where both the initiative and the institutional commitment resided mainly on the Bulgarian side, the FRG took the lead in Bulgarian–West German relations. And while the Bulgarian diplomatic 9 strategy towards Denmark was limited in both the breadth of its goals and the resources it could mobilise to reach them, in its overtures towards Bulgaria, West Germany applied a cohesive diplomatic agenda with long-term objectives and sufficient incentives to pursue them. Yet, while Bulgarian–Danish bilateralism could develop outside of Soviet interference, West Germany was at the centre of superpower politics and its relations with the socialist states were under close Soviet supervision. Although all socialist countries intensified their economic transactions with the FRG in the 1960s, diplomatic relations were out of the question. The erection of the Berlin Wall in 1961 prompted the FRG to seek some form of rapprochement with the Soviet bloc instead of containment, yet the post-war status quo remained a thorny issue of no compromise.31 A main vehicle of the ensuing probes was ‘diplomacy through trade’, a continuation of a long existing German tradition of Osthandel.32 Intertwining diplomacy and trade however could not circumvent the unresolved questions concerning post-war borders and territorial integrity, war reparations and German minority rights. Unlike the GDR, Poland and Czechoslovakia, Although all socialist countries intensified their economic transactions with the FRG in the 1960s, diplomatic relations were out of the question. Bulgaria’s responsiveness to the FRG’s diplomatic overtures in the 1960s The erection of the Berlin Wall in 1961 prompted the FRG to seek some form of rapprochement with the Soviet bloc instead of containment, yet the post-war status quo remained a thorny issue of no compromise.31 A main vehicle of the ensuing probes was ‘diplomacy through trade’, a continuation of a long existing German tradition of Osthandel.32 Intertwining diplomacy and trade however could not circumvent the unresolved questions concerning post-war borders and territorial integrity, war reparations and German minority rights. Unlike the GDR, Poland and Czechoslovakia, Bulgaria did not have outstanding issues with West Germany, which opened an easier path for the latter’s ‘diplomacy through trade’.33 In 1960, when the GDR urged the fraternal governments to release a joint declaration condemning West German military revanchism, the BKP’s Politburo added a special note in its internal resolution in support of the proposal that Bulgaria (along with Romania) was ‘not an immediate target of West Germany’s territorial revanchist aspirations’.34 Bulgaria did not have outstanding issues with West Germany, which opened an easier path for the latter’s ‘diplomacy through trade’.33 In 1960, when the GDR urged the fraternal governments to release a joint declaration condemning West German military revanchism, the BKP’s Politburo added a special note in its internal resolution in support of the proposal that Bulgaria (along with Romania) was ‘not an immediate target of West Germany’s territorial revanchist aspirations’.34 Bulgaria did not have outstanding issues with West Germany, which opened an easier path for the latter’s ‘diplomacy through trade’.33 In 1960, when the GDR urged the fraternal governments to release a joint declaration condemning West German military revanchism, the BKP’s Politburo added a special note in its internal resolution in support of the proposal that Bilateral commercial relations began to expand already in 1956 when the two countries signed a trade protocol. By 1960, the trade volume increased tenfold compared to 1950 when it stood at 4.7 million USD. Yet, at this peak, Bulgarian imports from West Germany twice surpassed the value of its exports, thereby causing a deficit of 18.7 million USD.35 While the Bulgarian regime did not have much room for manoeuvre in its policy towards the FRG, it 10 10 clearly recognised the economic benefits of possible relations. Bulgaria’s responsiveness to the FRG’s diplomatic overtures in the 1960s Over the next years, Bulgaria reduced its purchases from the FRG in a painstaking effort for a more balanced exchange, while keeping up its exports thanks to the goodwill of the West German authorities. As a result, for three consecutive years, Bulgaria ran a trade surplus, though its total did not come anywhere near that of West Germany in 1960 alone.36 In the same period, however, Bulgaria started generating huge invisible earnings on the German market through its program of international tourism.37 In March 1964, Bulgaria and the FRG ratified their first post-war trade agreement, which provided the legal framework for the reciprocal opening of official trade missions with consular functions in Sofia and Frankfurt, an expansion in size and status of already operational trade bureaus.38 This move forward in bilateral relations was still within the bounds of acceptable contacts: in fact, Bulgaria was the fourth bloc member to exchange trade missions with West Germany.39 In terms of personnel as well as weight within the MFT, the Frankfurt office by far surpassed the one in Copenhagen. Yet, it had neither the broad prerogatives of the trade team in Denmark in nurturing partnerships, nor the opportunities to operate independently of political interference. Its work was conditioned mainly by political assignments and it was political expediency that determined the nominations of employees, not economical expertise as in Copenhagen. Moreover, the executive oversight was shared by the MFT and MFA, which subordinated the activities of the mission to clashing institutional interests.40 In economic as well as diplomatic matters, the initiative remained on the German side, both in Frankfurt and in Sofia. While Denmark never opened a reciprocal service in Bulgaria, the FRG’s foreign trade representatives stationed in Sofia would actively seek to expand their access to state institutions beyond the MFT and promote further cooperation in non-economic fields. Sensitive to the dangers of soft diplomacy, Bulgarian bureaucrats tried hard to prevent 11 exposure without severing the communication. Following instructions from the party top, they systematically declined to stage West German cultural events in Bulgaria and were reluctant to facilitate scientific exchange under the scholarships offered by German academic foundations.41 By 1965, bilateral trade exceeded 100 million USD and Bulgaria spiralled into a trade deficit that spiked to 91 million USD in 1966.42 While this caused concerns within the foreign-trade services, profits from German tourists went through a comparable trend in reverse. Bulgaria’s responsiveness to the FRG’s diplomatic overtures in the 1960s Soon after the opening of trade missions, the tourist administration gave the green light to charter flights from the FRG and both the number of tourists and the revenues quadrupled from 1963 to 1964.43 For Germans separated by the Wall, Bulgarian resorts were a middle ground to meet in the holiday season.44 In 1965 when the influx of foreign guests in Bulgaria reached 1 million, the two countries accounting for the highest numbers of visitors were the GDR and the FRG.45 That year, the FRG also became Bulgaria’s fourth biggest trade partner (after the Soviet Union, the GDR and Czechoslovakia) in addition to its top position among the capitalist countries.46 Meanwhile, various forms of economic cooperation developed bypassing the trade mission in Frankfurt: especially, the joint ventures with Bulgarian majority rights.47 Modelled on capitalist enterprises and operating outside of strict governmental control, these so-called ‘trans-border firms’ authorised at the party apex often went rogue and competed with the official foreign-trade organisations for export contingents. Between 1965 and 1968, ten Bulgarian firms were created in West Germany, an exceptionally high concentration. Not only did they keep the trade mission in the dark about their deals, thus violating their legal obligations, as the trade attaché would complain, but they also blemished the reputation of the Bulgarian commerce with shady business practices.48 At the same time, important business negotiations with West German companies surpassed the level of the trade mission. Instead, 12 12 they were brokered by high-ranking state officials and their conclusion often involved top political participation. In line with the West German strategy of Osthandel, such negotiations supplied corporate entities with opportunities to forge political contacts. For example, the concern ‘Krupp’ initiated its negotiations for the delivery of an entire factory in personal communication with Todor Zhivkov and upon their successful conclusion in 1965 Zhivkov honoured the general manager with a protocol reception.49 While towing the Soviet line on diplomatic non-recognition, Bulgaria tried to preserve the bon ton at any cost, lured by the benefits of growing economic cooperation with the FRG. Bulgaria’s responsiveness to the FRG’s diplomatic overtures in the 1960s The Bulgarian border patrols showed no mercy to intercepted fugitives creating a chilling death toll of ‘tourist accidents’ as such failed escapes were officially reported.56 However, the Bulgarian authorities were quite lenient with West German passport holders, many recent refugees from the GDR themselves, who were quickly released. As a rule, Bulgaria declined the GDR’s requests for extradition of its former citizens and sent detainees back to the FRG.57 for diplomatic normalisation.54 The new government also stepped up the efforts to expand its economic footprint in Bulgaria. By 1967, the FRG was second only to the Soviet Union in Bulgaria’s imports, responsible for 10 percent of the total, and the single largest tourist provider, accounting for approximately 100,000 visitors per year.55 In this area, the Bulgarian regime was also a bit too accommodating to the annoyance particularly of the GDR. Vacationing in Bulgaria was one way for East Germans to flee into neighbouring NATO members, Greece and Turkey, and their fugitive attempts were often aided by West German citizens. The Bulgarian border patrols showed no mercy to intercepted fugitives creating a chilling death toll of ‘tourist accidents’ as such failed escapes were officially reported.56 However, the Bulgarian authorities were quite lenient with West German passport holders, many recent refugees from the GDR themselves, who were quickly released. As a rule, Bulgaria declined the GDR’s requests for extradition of its former citizens and sent detainees back to the FRG.57 Vacationing in Bulgaria was one way for East Germans to flee into neighbouring NATO members, Greece and Turkey, and their fugitive attempts were often aided by West German citizens. The Bulgarian border patrols showed no mercy to intercepted fugitives creating a chilling death toll of ‘tourist accidents’ as such failed escapes were officially reported.56 Bulgaria declined the GDR’s requests for extradition of its former citizens and sent detainees back to the FRG.57 In January 1967, Romania established diplomatic relations with the FRG in breach of the unified bloc stance, causing much havoc in the Warsaw Pact. Bulgaria’s responsiveness to the FRG’s diplomatic overtures in the 1960s Chancellor Erhard’s Peace Note of March 1966 stirred negative responses from other Warsaw Pact members, but Bulgaria and Romania kept silent.50 At the 1966 International Fair in Plovdiv, Bulgaria’s major forum for trade negotiations, the FRG had the largest exhibition surpassing even the USSR.51 Rolf Lahr, undersecretary for economic affairs at the Foreign Office, visited Bulgaria during the fair and held talks with three members of government – Ivan Bashev and his colleagues Ivan Budinov, Minister of Foreign Trade, and Lachezar Avramov, Minister without Portfolio.52 Lahr pushed for concessions in the field of cultural exchanges on the back of expanding economic relations. He also handed Budinov an official invitation to Bonn which would have been the first Bulgarian governmental visit there. In press comments following his trip, Lahr underlined that he had not discussed diplomatic normalisation, but the MFA’s internal reports construed the talks as a ‘cautious probing for establishing diplomatic relations.’53 While towing the Soviet line on diplomatic non-recognition, Bulgaria tried to preserve the bon ton at any cost, lured by the benefits of growing economic cooperation with the FRG. If in the early 1960s Bulgaria could extract high economic advantages from the FRG at small political costs, after 1966 when the Grand Coalition government formulated its Neue Ostpolitik, Bulgaria found itself in a more difficult position to uphold its economic interests while not yielding politically. In the eyes of the new West German leadership, Bulgaria (along with Romania) came up as a potential weak-link in the bloc’s line of resistance to its overtures 13 13 for diplomatic normalisation.54 The new government also stepped up the efforts to expand its economic footprint in Bulgaria. By 1967, the FRG was second only to the Soviet Union in Bulgaria’s imports, responsible for 10 percent of the total, and the single largest tourist provider, accounting for approximately 100,000 visitors per year.55 In this area, the Bulgarian regime was also a bit too accommodating to the annoyance particularly of the GDR. Vacationing in Bulgaria was one way for East Germans to flee into neighbouring NATO members, Greece and Turkey, and their fugitive attempts were often aided by West German citizens. Bulgaria’s responsiveness to the FRG’s diplomatic overtures in the 1960s A month later, the FRG proposed an identical treaty to the Bulgarian government via the trade mission in Sofia.58 The risk of Bulgaria following suit raised concerns in Poland and the GDR, and was countered by direct Soviet pressure on the Bulgarian establishment during the Warsaw Pact foreign ministers’ meeting in February 1967.59 Two months later, the Politburo adopted a resolution formulating the acceptable scope of bilateral relations: delaying negotiations on diplomatic normalisation, yet ‘searching for new opportunities and paths […] for a favourable development of the economic relations between the two countries’.60 Additionally, the Politburo approved a number of concessions proposed by the MFA: a green light for Budinov’s visit in Bonn; permission for selected West German cultural initiatives in Bulgaria; better protocol treatment of the trade envoys; instructions to mass media to restrain from 14 14 direct attacks on West German statesmen. Finally, the party leadership backed issuing official assurances that under no circumstances would Bulgaria deport West German citizens to third countries.61 Accordingly, the Deputy Chairman of the Committee on Tourism gave an interview to the West German media only to stress Bulgaria’s refusal of extraditions to the GDR – in contrast to other socialist countries.62 After Yugoslavia also resumed diplomatic relations with the FRG in early 1968, Soviet pressure on Bulgaria intensified.63 Bulgaria got back in line curtailing not just its diplomatic receptiveness, but to an extent also its economic openness to West Germany. The damage of 1968 on Bulgaria’s bilateral relations As part of this recovery, in May 1969 the Politburo ratified 15 a blueprint for Bulgaria’s foreign policy that highlighted relations with small and neutral capitalist states in Europe, including the Scandinavian countries.67 During the suspension of political contacts with Denmark, the Bulgarian regime fell back on tested approaches in building trust, like BZNS’ participation in agrarian forums and political conferences, and initiated negotiations in less politicised domains.68 In May 1969, the two governments signed a cultural agreement fulfilling one of the commitments made during Krag’s visit in 1967 as well as a protocol for cooperation between Danish and Bulgarian broadcasters.69 The embassy in Copenhagen dedicated much efforts to reschedule Zhivkov's visit, but the Danish government kept deflecting such approaches.70 This historic event would ultimately occur two years after the initial date, in September 1970, only after Denmark resumed relations with the Soviet Union.71 a blueprint for Bulgaria’s foreign policy that highlighted relations with small and neutral capitalist states in Europe, including the Scandinavian countries.67 During the suspension of political contacts with Denmark, the Bulgarian regime fell back on tested approaches in building trust, like BZNS’ participation in agrarian forums and political conferences, and initiated negotiations in less politicised domains.68 In May 1969, the two governments signed a cultural agreement fulfilling one of the commitments made during Krag’s visit in 1967 as well as a protocol for cooperation between Danish and Bulgarian broadcasters.69 The embassy in Copenhagen dedicated much efforts to reschedule Zhivkov's visit, but the Danish government kept deflecting such approaches.70 This historic event would ultimately occur two Subsequently, the MFA declared Zhivkov’s trip – with visits in Copenhagen and Oslo as originally planned and also Reykjavik – a glaring success. The ministry’s summary report to the Politburo highlighted the proximity in positions, at times complete concord, between the Bulgarian leader and his hosts. Furthermore, it ascertained favourable conditions for enhancing bilateral relations in terms of small-state diplomacy: ‘In these countries, apparently, desire for greater […] autonomy from the big Western countries is growing, fears of “the Eastern threat” are diminishing, and certain tendencies for greater activity in the relations with socialist countries are breaking through.’72 This optimistic assessment was quite far from the concerns of the envoys in Copenhagen who were preoccupied with the threat of raising trade barriers as a consequence of Denmark’s prospective membership in a customs union. The damage of 1968 on Bulgaria’s bilateral relations Whereas 1966–1967 marked a peak in Bulgaria’s opening to the West, relations would soon deteriorate. The Czechoslovak events in the spring of 1968 terminated by the Soviet-led military intervention in August, which was vehemently supported by the Bulgarian regime, caused turmoil in Bulgaria’s domestic and foreign policy. Bulgarian–Danish relations suffered as a consequence of Bulgaria’s participation in the invasion. Zhivkov’s long-planned visit to Norway and Denmark scheduled to take place a month later was cancelled at short notice by the two Scandinavian governments in coordination.64 The decisive governmental stance led to negative coverage of Bulgaria in the Danish press, which hurt ongoing economic negotiations with public visibility. This particularly affected tourism as Bulgaria’s largest Danish partner annulled its entire program in the Black Sea resorts indefinitely.65 As the tourist business was vulnerable to public opinion, the company announced its cancelation as a political decision. The boycott of Bulgarian organisations spread down the chain leading to further suspensions of economic deals.66 Facing the repercussions for its hard-line, the BKP mobilised the diplomatic corps to mend the country's international image. The damage of 1968 on Bulgaria’s bilateral relations New obstacles for Bulgarian exports were first flagged with regard to a Danish proposal for a Nordic customs union, ‘Nordek’, which was eventually aborted, and then by Denmark’s accession to the European Communities in January 1973. Subsequently, the MFA declared Zhivkov’s trip – with visits in Copenhagen and Oslo as originally planned and also Reykjavik – a glaring success. The ministry’s summary report to the Politburo highlighted the proximity in positions, at times complete concord, between the Bulgarian leader and his hosts. Furthermore, it ascertained favourable conditions for enhancing bilateral relations in terms of small-state diplomacy: ‘In these countries, 16 Bulgaria’s relations with the FRG also went through a downfall and a slow recovery. Yet, their deterioration was inflicted by the Bulgarian government under the growing pressure by its allies. In its change of tone, the party leadership took a cue from East Germany, which propped up its allegations of the FRG’s revanchist ambitions after the Prague Spring.73 The Bulgarian establishment, weary of its own exposure to West Germany, took to heart this antagonist rhetoric and Todor Zhivkov became an energetic mouthpiece against the FRG in contrast to his earlier restraint. Unlike Denmark, West Germany was not bent on closing the door to Bulgaria and even the press coverage in the aftermath of the invasion was not uniformly negative. Although bashing publications constituted the bulk of ‘foreign-media propaganda on Bulgaria’ that the trade mission in Frankfurt forwarded home, the envoys also flagged a number of articles speculating about subtle signs of Bulgaria’s divergence from the bloc along the path of Romania and Yugoslavia. This combined with the restrained positions of West German media close to the government and the business circles led them to conclude that ‘the federal government after all retains some room for tactical manoeuvres with regard to our country’.74 Bulgaria’s relations with the FRG also went through a downfall and a slow recovery. Yet, their deterioration was inflicted by the Bulgarian government under the growing pressure by its allies. The damage of 1968 on Bulgaria’s bilateral relations In its change of tone, the party leadership took a cue from East Germany, which propped up its allegations of the FRG’s revanchist ambitions after the Prague Spring.73 The Bulgarian establishment, weary of its own exposure to West Germany, took to heart this propaganda on Bulgaria’ that the trade mission in Frankfurt forwarded home, the envoys also flagged a number of articles speculating about subtle signs of Bulgaria’s divergence from the bloc along the path of Romania and Yugoslavia. This combined with the restrained positions of West German media close to the government and the business circles led them to conclude that ‘the federal government after all retains some room for tactical manoeuvres with regard to our country’.74 Despite Zhivkov’s show of loyalty to Moscow, Bulgaria’s official course towards West Germany did not undergo substantive modifications. Development of beneficial economic ties remained its primary rationale, while restraint in cultural contacts was reiterated.75 What changed was tightening central control over all forms of cooperation, yet oversight was not carried directly by the party but was delegated to the MFA.76 Bilateral trade was reduced but this was motivated not so much by political expedience as by concerns over the growing trade deficit. Although the FRG lost its leading position among Bulgaria’s capitalist trade partners to Italy, it remained the premier importer of Bulgarian production in the West.77 Moreover, business with West Germany did not seem to be hurt by retaliatory actions and in fact 1968 saw an increase in tourism to Bulgaria: new contracts with large travel agencies were signed 17 17 and a massive advertisement campaign for Bulgarian resorts was carried across the FRG. As a result, now the FRG alone accounted for half of Bulgaria’s hard-currency revenues from international tourism.78 While not interfering directly in economic cooperation with West Germany, the BKP was nevertheless determined to sever any other contacts that could imply diplomatic normalisation. In addition to declining West German cultural activities in Bulgaria, the Politburo disseminated confidential instructions for excluding the trade attaché from official events in Sofia. Yet, as the previous decade of economic advances had proven, diplomacy and trade could not evolve separately. The damage of 1968 on Bulgaria’s bilateral relations Heinz Herrmann, the head of the West German trade mission, reported this breach of protocol and his complaint entered the agenda of the bilateral session for ratification of the annual trade protocol in 1969.79 A month later he was recalled and the FRG delayed dispatching a successor for almost a year leaving the mission under ad interim direction of Herrmann’s deputy. The signal did not escape the attention of the Bulgarian foreign service and prompted them to push for recovery of the strained relations. In this period, a recurrent theme in reports and proposals from the MFA and the MFT would be their concerns of lagging behind the rest of the bloc in the cooperation with West Germany. The party, seemingly tenacious in its hard-line, was not as invested in defending the bloc’s stance vis-à-vis West Germany as Zhivkov’s rhetoric might have indicated. Although the BKP urgently instructed all state institutions to be on alert for any mention of West Berlin in their communication with the FRG, it also remained surprisingly oblivious to the significance of the heated issues related to the post-war status quo.80 For example, in 1969 when Bulgaria was still under Soviet pressure to cool off political openness, a deputy prime minister sent a telegram addressed ‘West Berlin, FRG’. The federal postal service redirected it to the trade mission in Frankfurt with a stamp ‘wrong address’, causing great embarrassment in the foreign service.81 18 18 After Willy Brandt became Chancellor in October 1969, West Germany stepped up its diplomatic pursuits in the Soviet bloc. Brandt’s willingness for concessions on the unresolved questions with bloc members led to the successful conclusion of treaties with the Soviet Union and Poland in 1970, the GDR in 1972 and Czechoslovakia in 1973. During these lengthy and difficult negotiations, however, bilateral talks with Hungary and Bulgaria were put on hold. As neither country had outstanding issues with the FRG, their treaties with West Germany were rendered conditional on the successful conclusion of the more sensitive talks to give additional leverage to their allies on the negotiating table.82 Thus, while hitherto the lack of outstanding issues placed Bulgaria in the fore of the West German diplomatic strategy towards the East, now it pushed her at the back of the line together with Hungary.83 The detrimental effects of this change were immediately noted by the Bulgarian foreign services. The damage of 1968 on Bulgaria’s bilateral relations In December 1969, the West German trade attaché a.i. was received in the MFA upon his request to discuss a concert in Sofia of a West German symphony orchestra. During the conversation, he veered into a detailed expose of the forthcoming negotiations between his government and the other socialist regimes. The meeting’s memo was subsequently appended with an unusually lengthy note by a high-ranking official at the MFA. Ordering the memo’s broader circulation among economic ministries and party sections, he made a strong recommendation for concessions on the cultural ban. Arguing that relations between the Soviet Union and the FRG were warming up, his note concluded: ‘We should avoid remaining the only passive country that shows no change regarding the FRG’.84 Under ministerial pressure, in early 1971 the Politburo moderated the guidelines for Bulgaria’s foreign policy towards the FRG also allowing cultural initiatives as long as they did not have ‘mass and propaganda character’.85 While the foreign services were trying to ease the hard-line to preserve the advantages of opening to West Germany, Todor Zhivkov kept his zeal in bashing the West German 19 19 government. In international declarations, he often overstepped the line set by Moscow leading Radio Free Europe to surmise: ‘there have been occasions when Bulgaria’s rigidity vis-à-vis the Federal Republic would appear to have been slightly embarrassing to her allies’ rather than ‘an enunciation of a common Communist stance’.86 Such erratic improvisations of his ran contrary to the endorsed governmental policy towards the FRG and caused much disarray among his own foreign services.87 Whether he embraced the role of Moscow’s ‘attack dog’ or simply lost track with the trend across the bloc, Zhivkov’s antagonistic performance undermined not only the efforts of the state administration but also the decisions of his Politburo. government. The damage of 1968 on Bulgaria’s bilateral relations In international declarations, he often overstepped the line set by Moscow leading Radio Free Europe to surmise: ‘there have been occasions when Bulgaria’s rigidity vis-à-vis the Federal Republic would appear to have been slightly embarrassing to her allies’ rather than ‘an enunciation of a common Communist stance’.86 Such erratic improvisations of his ran contrary to the endorsed governmental policy towards the FRG and caused much disarray among his own foreign services.87 Whether he embraced the role of Moscow’s ‘attack dog’ or simply lost track with the trend across the bloc, Zhivkov’s antagonistic performance undermined not only the efforts of the state administration but also the decisions of his Politburo. Backed by both party and state, at this point, the Bulgarian foreign services were working hard on a diplomatic breakthrough with Bonn. The impending signing of a new long-term trade agreement was planned as an occasion for the first governmental visit in Bonn by Minister of Foreign Trade Lachezar Avramov. During this visit in February 1971, he held a number of top-level meetings, including with Chancellor Brandt and Minister of Foreign Affairs Walter Scheel.88 Part of Avramov’s agenda was preparing the ground for future diplomatic relations.89 During the preliminary negotiations, two issues caused disagreements: the insistence of Bonn to include West Berlin under the provisions, which Bulgaria did not feel at liberty to accept, and the insistence of Sofia on the most-favoured nation clause. The compromise on the first issues was to define the territorial scope of the agreement monetarily, by reference to the Deutschmark – thus covering West Berlin de jure without mentioning it explicitly. Such a solution was indeed to the satisfaction of the Bulgarian side because West Berlin also fell within its economic interests. Once this formulation was accepted, Bonn conceded the Bulgarian request for most-favoured nation.90 And two years later, in December 1973, Bulgaria and the FRG finally established diplomatic relations.91 Conclusion 20 20 Throughout the 1960s, Bulgaria aspired to take its place in the process of pan-European cooperation as bridge building between East and West opened the path to détente. While not short of ambition, its room for manoeuvre in diplomatic pursuits towards the West was constrained by Moscow’s superpower politics. Yet, the limitations for Bulgaria’s small-state diplomacy also presented an opportunity: the opportunity for state officials with a technocratic rather than an ideological outlook to take the lead and subordinate the strategy of this opening to the pragmatic interests of economic cooperation. The interplay between economic exchange and interstate diplomacy took different shapes depending on how particular bilateral relations with Western countries fit in the geopolitical constellation of the Cold War at a given moment, thus also determining the margins for manoeuvre of Bulgaria’s economic interests vis-à-vis the bloc’s political pressures. Throughout the 1960s, Bulgaria aspired to take its place in the process of pan-European cooperation as bridge building between East and West opened the path to détente. While not short of ambition, its room for manoeuvre in diplomatic pursuits towards the West was constrained by Moscow’s superpower politics. Yet, the limitations for Bulgaria’s small-state diplomacy also presented an opportunity: the opportunity for state officials with a technocratic rather than an ideological outlook to take the lead and subordinate the strategy of this opening to the pragmatic interests of economic cooperation. The interplay between economic exchange and interstate diplomacy took different shapes depending on how particular bilateral relations with Western countries fit in the geopolitical constellation of the Cold War at a given moment, thus also determining the margins for manoeuvre of Bulgaria’s economic interests vis-à-vis the bloc’s political pressures. The comparative study of Bulgaria’s bilateral relations with Denmark and the FRG reveals these margins of manoeuvre in their full range: from an advance unburdened by Soviet scrutiny and party interference where lower-ranking actors of a mind to do business enjoyed initiative, to a precarious venturing into the minefield of the superpower confrontation where economic incentives would time and again clash with ideological allegiance. Being a small state itself, Denmark could offer little benefits to the struggling Bulgarian economy in contrast to the West German economic powerhouse. Conclusion Bilateral relations with the FRG were not only far more asymmetrical but they were actively shaped by the latter’s ‘policy of using trade as the narrow end of a wedge for opening Eastern Europe to West German influence’.92 Consequentially, Bulgaria’s interaction with the FRG carried much higher political risks than its bridge-building towards Denmark, but its economic incentives were also significantly larger. In the nascent Bulgarian–Danish relations of the early 1960s, economic cooperation and foreign trade in particular emerged as a safe zone to launch bilateral contacts, contacts driven 21 by Bulgarian economic interests and enabled by Denmark’s support for détente. Although bilateral trade would remain negligible, its function surpassed economic pragmatism. Trade was instrumental in opening a channel for intergovernmental contacts and keeping it open when conventional diplomacy was hampered as would be the case after the crackdown of Prague Spring in 1968. Conversely, in Bulgaria’s initial opening to the FRG, economic cooperation was inevitably leading the Bulgarian regime to the hotspot of diplomatic normalisation where it would face the opposition of its allies. In its grasping for economic advantages, Bulgaria heavily relied on the West German diplomatic investment in the East- bound trade and its own foreign-trade services were denied much room for action. The events in 1968 complicated the terrain of East–West cooperation, especially for a smaller socialist state like Bulgaria. After the invasion of Czechoslovakia, the approaches of its foreign services were stalled by both political retaliation from Western countries and ideological caution within the Bulgarian party apparatus. Although the pace of cooperation slowed, the internal bargaining between the party and the state apparatus – accordingly, political dogmatism and economic pragmatism – intensified. Ultimately, economic interests took the upper hand. Albeit with more precarious manoeuvring and less leverage, the Bulgarian foreign services not only regained access to both the Danish and West German governments in the early 1970s, but signed ground-breaking treaties like diplomatic normalisation with the FRG. While the negotiations required much more persistence and adaptations on their side than earlier, the experience of bridge-building and the evidence of the economic benefits of bilateral contacts gained in the past decade certainly increased their readiness for concessions and determination to stretch their room for manoeuvre. 1 Tom Junes und Elitza Stanoeva, ‘Todor C. Schiwkow (1911–1998): Bulgarischer Kommunismus zwischen Acknowledgement 22 The research leading to this article is part of the project PanEur1970s, which has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (Grant Agreement No. 669194). I am thankful to the editors as well as my colleagues Sune Bechmann Pedersen and Maximilian Graf for their critical input on this chapter. 23 24 24 nationaler Unabhängigkeit und Loyalität gegenüber der Sowjetunion’, in: Martin Sabrow und Susanne Schattenberg (Hgs), Die letzten Generalsekretäre: Kommunistische Herrschaft im Spätsozialismus (Berlin 2018) 37–63, 55 2 Gottfried Niedhart, ‘East-West Conflict: Short Cold War and Long Détente An Essay on Terminology and Periodization’, in: Oliver Bange and Poul Villaume (eds), The Long Détente: Changing Concepts of Security and Cooperation in Europe, 1950s-1980s (New York 2017) 19–30, 22. 2 Gottfried Niedhart, ‘East-West Conflict: Short Cold War and Long Détente An Essay on Terminology and Periodization’, in: Oliver Bange and Poul Villaume (eds), The Long Détente: Changing Concepts of Security and Cooperation in Europe, 1950s-1980s (New York 2017) 19–30, 22. 3 Alexander Vezenkov, Vlastovite strukturi na Balgarskata Komunisticheska Partiya 1944-1989 (Sofia 2008); Hristo Hristov, Todor Zhivkov: biografiya (Sofia, 2009), ch. 11; Stephanos Katsikas, ‘A Comparative Overview on the Foreign Policy-making of Albania, Bulgaria and Yugoslavia during the 1980’s’, Bulgarian Historical Review 3:4 (2006) 106–144, 109. 3 Alexander Vezenkov, Vlastovite strukturi na Balgarskata Komunisticheska Partiya 1944-1989 (Sofia 2008); Hristo Hristov, Todor Zhivkov: biografiya (Sofia, 2009), ch. 11; Stephanos Katsikas, ‘A Comparative Overview on the Foreign Policy-making of Albania, Bulgaria and Yugoslavia during the 1980’s’, Bulgarian Historical Review 3:4 (2006) 106–144, 109. 4 Martin Ivanov, Reformatorstvo bez reformi: Politicheska ikonomiya na balgarskiya komunizam 1963–1989 g. (Sofia 2008); Iliyana Marcheva, Politikata za stopanska modernizatsiya v Balgariya po vreme na Studenata voyna (Sofia, 2016). In contrast, John Lampe emphasises the ‘large relative size of Bulgaria's foreign trade’ compared to the other East European countries. John Lampe, The Bulgarian Economy in the Twentieth Century (London 1986) 177–178. 4 Martin Ivanov, Reformatorstvo bez reformi: Politicheska ikonomiya na balgarskiya komunizam 1963–1989 g. (Sofia 2008); Iliyana Marcheva, Politikata za stopanska modernizatsiya v Balgariya po vreme na Studenata voyna (Sofia, 2016). In contrast, John Lampe emphasises the ‘large relative size of Bulgaria's foreign trade’ compared to the other East European countries. John Lampe, The Bulgarian Economy in the Twentieth Century (London 1986) 177–178. 5 Ibid, 179–181. nationaler Unabhängigkeit und Loyalität gegenüber der Sowjetunion’, in: Martin Sabrow und Susanne Schattenberg (Hgs), Die letzten Generalsekretäre: Kommunistische Herrschaft im Spätsozialismus (Berlin 2018) 37–63, 55. itburo Protocol 319A, 2/11/1965, TsDA/f.1B/op.6/a.e.6037, 1–3. Acknowledgement See also Suvi Kansikas, Socialist Countries Face the European Community: Soviet-Bloc Controversies over East–West Trade (Frankfurt a.M. 2014). 6 Angela Romano and Federico Romero, ‘European Socialist Regimes Facing Globalisation and European Co- operation: Dilemmas and Responses – Introduction’, European Review of History: Revue europeenne d’histoire 21:2 (2014) 157–164. 7 Hans Branner, ‘Options and Goals in Danish European Policy Since 1945: Explaining Small State Behavior and Foreign Policy Change’, in: Hans Branner and Morten Kelpstrup (eds), Denmark’s Policy towards Europe after 1945: History, Theory and Options (Odense 2000) 333–380, 350. 8 Information: development of economic relations between Bulgaria and Denmark 27/06/1968, Tsentralen Darzhaven Arhiv (TsDA), f.1477/op.25/a.e.867, 17. 9 Materials on the Danish-Bulgarian diplomatic relations were provided by the Danish embassy in Sofia for which I extend my gratitude. 10 Report by P.Valchev, head of Bultarg Copenhagen, 26/01/1976, TsDA/f.1244/op.1/a.e.8142, 69; Information: development of economic relations between Bulgaria and Denmark 27/06/1968, TsDA/f.1477/op.25/a.e.867, 18. 11 Information: contractual-legal relations between Bulgaria and Denmark 20/10/1976, TsDA/f.259/op.39/a.e.231, 150. 12 Information: implementation of Politburo’s Resolution 168 (18/04/1967) and Resolution 430 (31/10/1967) on development of Bulgarian-Danish relations, TsDA/f.1477/op.26/a.e.1015, 4; Information: political, economic, industrial and scientific-technical relations between Bulgaria and Denmark, Feb.1973, TsDA/f.1244/op.1/a.e.6781, 5. 13 Elitza Stanoeva, ‘Exporting Holidays: Bulgarian International Tourism on the Scandinavian Market in the 1960s and 1970s’, in: Sune Bechmann Pedersen and Christian Noack (eds), Tourism and Travel during the Cold War: Negotiating Tourist Experiences across the Iron Curtain (London forthcoming). 14 Emil Georgiev, ‘Sastoyanie i perspektivi na zapadnoevropeyskiya turisticheski pazar’, Vanshna targoviya 3 (1969), 10–13, 12. 15 Kostadin Grozev and Jordan Baev, ‘Bulgaria, Balkan Diplomacy and the Road to Helsinki’, in: Oliver Bange and Gottfried Niedhart (eds), Helsinki 1975 and the Transformation of Europe (New York 2008) 160–173, 165. 16 Jordan Baev, ‘Two Poles of the Cold War Confrontation in Europe: Bulgaria and the Nordic Countries’, online publication by the Cold War History Research Center (Budapest), available at: http://www.coldwar.hu/html/en/publications/two_poles.html. 25 18 Information: cooperation between Bulgaria and Denmark in agriculture [1970], TsDA/f.1477/op.27/a.e.1116, 47–52. 18 Information: cooperation between Bulgaria and Denmark in agriculture [1970], TsDA/f.1477/op.27/a.e.1116, 47–52. 19 I am thankful to Stefan Detchev for bringing this historical link to my attention. 19 I am thankful to Stefan Detchev for bringing this historical link to my attention. 19 I am thankful to Stefan Detchev for bringing this historical link to my attention. 20 Information: political relations between Bulgaria and Denmark [1968], TsDA/f.1477/op.25/a.e.867, 113–114. Acknowledgement 20 Information: political relations between Bulgaria and Denmark [1968], TsDA/f.1477/op.25/a.e.867, 113–114. 20 Information: political relations between Bulgaria and Denmark [1968], TsDA/f.1477/op.25/a.e.867, 113–114. 21 CM Ordinance 277, 22/09/1967, TsDA/f.136/op.44/a.e.326; CM Resolution 379, 30/08/1967, TsDA/f.136/op.45/a.e.360. 21 CM Ordinance 277, 22/09/1967, TsDA/f.136/op.44/a.e.326; CM Resolution 379, 30/08/1967, TsDA/f.136/op.45/a.e.360. 22 Stanoeva, ‘Exporting Holidays’. 22 Stanoeva, ‘Exporting Holidays’. 23 Politburo Protocol 430A, 31/10/1967, TsDA/f.1B/op.6/a.e.6931, 2–3. 24 Letter from MFA to Politburo 7/09/1967, TsDA/f.1B/op.6/a.e.6931, 53. 26 Memorandum by L.Ganchev, ambassador in Stockholm, 31/05/1968, TsDA/f.1477/op.25/a.e.867, 2 27 Program for expanding commercial, industrial and scientific–technical cooperation between Bulgaria and Denmark, TsDA/f.1477/op.26/a.e.1015, 16–17; Memo: questions MFT considers expedient for discussion during T.Zhivkov’s visit in Denmark 29/07/1968, TsDA/f.1477/op.25/a.e.867, 21. 28 Information: BZNS delegation’s visit at small landowners’ congress in Denmark 10/04/1968, TsDA/f.1477/op.25/a.e.870, 6. 29 Michael Kreile, Osthandel und Ostpolitik: Zum Verhältnis von Außenpolitik und Außenwortschaft in den Beziehungen der Bundesrepublik Deutschland zu den RGW-Ländern (Tübingen 1977) 27. 30 Jordan Baev, ‘The Establishment of Bulgarian–West German Diplomatic Relations within the Coordinating Framework of the Warsaw Pact’, Journal of Cold War Studies 18:3 (2016) 158–180, 159–161. 31 Hope Harrison, ‘The Berlin Wall, Ostpolitik and Détente’, Bulletin of the German Historical Institute supplement 1 (2004) 5–18, 8. 32 Robert Mark Spaulding, Osthandel and Ostpolitik: German Trade Policies in Eastern Europe from Bismarck to Adenauer (Provedance 1997). 33 Csaba Békés, ‘The Warsaw Pact, the German Question and the Birth of the CSCE Process, 1961–1970’, in: Bange and Niedhart, Helsinki 1975, 113–128, 113–114. 34 Letter from K.Lukanov to Politburo 24/08/1960, TsDA/f.1B/op.6/a.e.4248, 50. 35 Information: commodity turnover and economic ties between Bulgaria and FRG 3/12/1970, TsDA/f.1477/op.27/a.e.941, 28. 36 Information: trade relations between Bulgaria and FRG and trade mission’s operation, May 1969, TsDA/f.1477/op.25/a.e.761, 15. 37 Information: development of tourist links between Bulgaria and FRG 26/12/1969, TsDA/f.1477/op.25/a.e.779, 17. 38 Information: trade relations between Bulgaria and FRG and trade mission’s operation, May 1969, TsDA/f.1477/op.25/a.e.761, 15. 26 39 Bulgaria: Situation report 23/01/1964, Vera and Donald Blinken Open Society Archive (HU OSA), 300-8-47- 17-6, 3. 40 Annual report of Frankfurt trade mission for 1968, May 1969, TsDA/f.1477/op.25/a.e.761, 40–42; Information: Ministry’s activities on Bulgaria’s political, economic and cultural relations with FRG [1969], TsDA/f.1477/op.26/a.e.886, 13–15. 41 Information on FRG, Aug.1970, TsDA/f.1477/op.26/a.e.886, 71. 41 Information on FRG, Aug.1970, TsDA/f.1477/op.26/a.e.886, 71. 42 Information: commodity turnover and economic ties between Bulgaria and FRG 3/12/1970, TsDA/f.1477/op.27/a.e.941, 28. 43 Information: development of tourist links between Bulgaria and FRG 26/12/1969, TsDA/f.1477/op.25/a.e.779, 17. Acknowledgement 44 Vacationland Bulgaria: The tourist boom 24/02/1966, HU OSA/300-8-3-15923, 3. 45 Bulgaria: Situation report 28/12/1965, HU OSA/300-8-47-20-41, 2. 46 Bulgaria: Situation report 27/10/1965, HU OSA/300-8-47-20-28, 3. 47 Hristo Hristov, Imperiyata na zadgranichnite firmi: sazdavane, deynost i iztochvane na druzhestvata s balgarsko uchastie zad granitsa 1961–2007 (Sofia 2009). 48 Annual report of Frankfurt trade mission for 1968, May 1969, TsDA/f.1477/op.25/a.e.761, 32–34. 49 Letter from B.Beitz, General Manager of Krupp, to T.Zhivkov 20/09/1963, TsDA/f.1B/op.6/a.e.5250, 8–9; Bulgaria: Situation report 1/04/1965, HU OSA/300-8-47-19-21, 3. 50 Zhivkov’s Party Congress report 21/11/1966, HU OSA/300-8-3-698, 5. 51 Ibid 6 50 Zhivkov’s Party Congress report 21/11/1966, HU OSA/300-8-3-698, 5. Bulgaria: Situation Report 27/09/1966, HU OSA/300-8-47-21-58, 4. 52 Bulgaria: Situation Report 27/09/1966, HU OSA/300-8-47-21-58, 4. 53 Strictly confidential information to Politburo 28/02/1967, TsDA/f.1B/op.6/a.e.6655, 30; Memo: conversation with FRG’s trade representative Herrmann 7/04/1967, TsDA/f.1B/op.81A/a.e.912, 2. 54 Self-interest versus orthodoxy: A study of Bulgaria’s relationship with the Federal Republic of Germany 4/05/1970, HU OSA/300-8-3-13561, 2; Information on FRG, Aug.1970, TsDA/f.1477/op.26/a.e.886, 64. 55 Bulgaria: Situation report 12/10/1967, HU OSA/300-8-47-22-62, 1; Bulgaria prepares for the tourist season 11/03/1968, HU OSA/300-8-3-741, 3. 56 For a documentalist account, see Kapka Kassabova, Border: A Journey to the Edge of Europe (London 2017) 87–106. 57 Bulgaria: Situation report 24/08/1967, HU OSA/300-8-47-22-51, 2. 57 Bulgaria: Situation report 24/08/1967, HU OSA/300-8-47-22-51, 2. 58 Strictly confidential information to Politburo 28/02/1967, TsDA/f.1B/op.6/a.e.6655, 32. 59 Baev, ‘The Establishment of Bulgarian–West German Diplomatic Relations’, 164. 60 Politburo Protocol 156A, 13/04/1967, TsDA/f.1B/op.6/a.e.6655, 22. Strictly confidential information to Politburo 28/02/1967, TsDA/f.1B/op.6/a.e.6655, 34–38. 62 Bulgaria: Situation report 19/09/1967, HU OSA/300-8-47-22-55, 3; Bulgaria prepares for the tourist season 11/03/1968, HU OSA/300-8-3-741, 3. 27 64 Letter from Bulgarian embassy in Oslo to MFA 23/08/1968, TsDA/f.1477/op.25/a.e.867, 221–222. 65 Information: development of tourist relations between Bulgaria and Denmark 26/12/1969, TsDA/f.1477/op.26/a.e.1015, 10. 66 Information: economic, industrial and technical cooperation between Bulgaria and Denmark 17/12/1969, TsDA/f.1477/op.26/a.e.1015, 7. 67 Chief guidelines and main tasks of Bulgaria’s foreign policy, 1969, TsDA/f.1B/op.35/a.e.704, 65. 68 Letter from I.Vrazhilov, chargé d’affaires a.i. in Copenhagen, to MFA 2/07/1971, TsDA/f.1477/op.27/a.e.1083, 7–8. 69 Cable from Bulgarian embassy in Copenhagen 6/05/1969, TsDA/f.1477/op.25/a.e.889, 1; Information: contractual-legal relations between Bulgaria and Denmark 20/10/1976, TsDA/f.259/op.39/a.e.231, 150. 70 Memorandum by V.Savov, chargé d’affaires in Copenhagen, to I.Bashev 24/11/1969, TsDA/f.1477/op.25/a.e.875, 102–103. 71 Letter to BKP Central Committee 11/12/1969, TsDA/f.1477/op.26/a.e.1001, 1–2. 72 Report by H.Traykov, Deputy Minister of Foreign Affairs, to Politburo 2/10/1970, TsDA/f.1B/op.35/a.e.1727, 14. posal by MFA to Politburo 27/10/1970, TsDA/f.1B/op.35/a.e.1927, 243. Acknowledgement 73 Self-interest versus orthodoxy: A study of Bulgaria’s relationship with the Federal Republic of Germany 4/05/1970, HU OSA/300-8-3-13561, 9–10; 74 Information by Petar Mezhdurechki, Bulgarian trade mission in FRG [Apr.1969], TsDA/f.1477/op.25/a.e.762, 27. 75 Action measures on bilateral relations with FRG, Oct.1968, TsDA/f.1477/op.25/a.e.779, 6. 76 Circular letter by MFA, Jun.1969, TsDA/f.1477/op.25/a.e.781, 7. 77 Information: commodity turnover and economic ties between Bulgaria and FRG 3/12/1970, TsDA/f.1477/op.27/a.e.941, 19. 78 Information: development of tourist links between Bulgaria and FRG 26/12/1969, TsDA/f.1477/op.25/a.e.779, 17. 79 Letter by H.Herrmann to P.Todorov, Chairman of Committee on Tourism, 30/06/1969, TsDA/f.1477/op.25/a.e.780, 34. Circular letter of MFA, Jun.1969, TsDA/f.1477/op.25/a.e.781, 7. 81 Letter from S.Zhelenski, deputy trade representative in Frankfurt, to MFA 13/09/1969, TsDA/f.1477/op.25/a.e.789, 7. 82 Oldřich Tůma, ‘The Difficult Path to the Establishment of Diplomatic Relations between Czechoslovakia and the Federal Republic of Germany’, in: Carole Fink and Bernd Schaefer (eds), Ostpolitik, 1969-1974: European and Global Responses (Cambridge 2009) 58–79, 58–59. 83 Békés, ‘The Warsaw Pact’, 118. 84 Memo by Z.Zelenogradski, deputy head of Section IV at MFA, 12/12/1969, TsDA/f.1477/op.25/a.e.764, 56. 85 Proposal by MFA to Politburo 27/10/1970, TsDA/f.1B/op.35/a.e.1927, 243. 85 Proposal by MFA to Politburo 27/10/1970, TsDA/f.1B/op.35/a.e.1927, 243. 28 86 Self-interest versus orthodoxy: A study of Bulgaria’s relationship with the Federal Republic of Germany 4/05/1970, HU OSA/300-8-3-13561, 13–14. 87 Letter from S.Zhelenski, deputy trade representative in Frankfurt, to MFA 2/10/1970, TsDA/f.1477/op.26/a.e.884, 50–51. 88 Memorandum: L.Avramov’s meetings with W.Brandt and W.Scheel, 15/02/1971, TsDA/f.1477/op.27/a.e.941, 76–83. 89 Information on FRG, Aug.1970), TsDA/f.1477/op.26/a.e.886, 77–80. 90 Memo: preliminary talks between governmental delegations of Bulgaria and FRG on Long-term Trade Agreement, TsDA/f.1477/op.26/a.e.870, 39; Report by L.Avramov regarding ratification of Long-term Trade Agreement, TsDA/f.1477/op.27/a.e.977, 36–39. 91 Report by P.Mladenov, Minister of Foreign Affairs, 18/12/1973, TsDA/f.1477/op.29/a.e.853, 43. 92 Spaulding, Osthandel and Ostpolitik 371. 92 Spaulding, Osthandel and Ostpolitik 371. 29
https://openalex.org/W4390521815	https://anthor.org/index.php/anthor/article/download/250/169	English	null	Comparison of Educational Theories: Perspectives of Carol Dweck and Howard Gardner in Developing Individual Potential	Anthor	2,023	cc-by-sa	6,014	Comparison Comparison First R First Received: 8 Juli 2023 Abstrak Kata Kunci: Teori Pemikiran Bertumbuh, Teori Kecerdasan Majemuk, Pengembangan potensi individu, Inklusivitas pendidikan, Perbandingan teori pendidikan. Penelitian ini bertujuan untuk membandingkan Teori Pemikiran Bertumbuh (Growth Mindset) oleh Carol Dweck dan Teori Kecerdasan Majemuk (Multiple Intelligences) oleh Howard Gardner dalam mengembangkan potensi individu. Penelitian ini menggunakan metode studi literatur dan analisis perbandingan untuk membandingkan Teori Pemikiran Bertumbuh (Growth Mindset) oleh Carol Dweck dan Teori Kecerdasan Majemuk (Multiple Intelligences) oleh Howard Gardner. Dalam upaya mengoptimalkan potensi individu, pemikiran bertumbuh menekankan pentingnya sikap mental yang berkembang, memandang kegagalan sebagai kesempatan belajar, dan menciptakan lingkungan belajar yang mendukung. Sementara itu, kecerdasan majemuk mengakui keberagaman kecerdasan dan mengadvokasi pengembangan semua jenis kecerdasan dalam pendidikan. Dalam pendidikan modern yang inklusif, penerapan kedua teori ini dapat memotivasi siswa untuk tumbuh dan belajar, mengembangkan kepercayaan diri, dan mengoptimalkan potensi mereka di berbagai bidang kecerdasan. Melalui pendekatan ini, pendidikan modern dapat menciptakan lingkungan yang merangsang, memfasilitasi perkembangan holistik, dan menghasilkan individu yang siap menghadapi tantangan masa depan. Dengan memahami perbedaan dan persamaan antara kedua teori ini, pendidikan dapat mengadopsi pendekatan yang inklusif dan berfokus pada pengembangan potensi penuh setiap individu, memastikan bahwa siswa diberdayakan untuk mencapai keberhasilan akademik dan perkembangan pribadi yang optimal. Keywords: Growth Mindset Theory, Multiple Intelligences Theory, Individual Potential Development, Inclusive Education, Educational Theory Comparison. COMPARISON OF EDUCATIONAL THEORIES: PERSPECTIVES OF CAROL DWECK AND HOWARD GARDNER IN DEVELOPING INDIVIDUAL POTENTIAL Arshy Prodyanatasari1, Mustofa Aji Prayitno2, Ferdinand Salomo Leuwol3, Siti Aminah4, Maskur5 Institut Ilmu Kesehatan Bhakti Wiyata Kediri1, Institut Agama Islam Negeri Ponorogo2, Universitas Pattimura3, Universitas Islam Kadiri Kediri4, Sekolah Tinggi Agama Islam Walisembilan Semarang5 arshy.prodyanatasari@iik.ac.id1, mustofaajiprayitno@gmail.com2*, eddieleuwol0@gmail.com3, sitiaminah@uniska-kediri.ac.id4, maskur2106128401@gmail.com5 INTRODUCTION Education is a crucial aspect of individual development (H. Gardner, 2006). In the pursuit of reaching their maximum potential, education should be based on in-depth theories that can effectively aid in developing individual capabilities (Lolang et al., 2023). Moreover, by integrating evidence-based educational theories, educators can create a supportive and stimulating learning environment that fosters growth, resilience, and a lifelong passion for acquiring knowledge (Darmawati.R et al., 2023). Two prominent theories in this field are Carol Dweck's theory of mindset and Howard Gardner's theory of multiple intelligences. Carol Dweck's mindset theory primarily focuses on individuals' beliefs about the nature of intelligence and abilities. According to Dweck, there are two types of mindsets that influence how individuals perceive themselves and approach challenges. The first is a fixed mindset, in which individuals tend to view intelligence and abilities as fixed and unchangeable. The second is a growth mindset, in which individuals believe that intelligence and abilities can be developed through effort, practice, and perseverance. On the other hand, Howard Gardner's theory of multiple intelligences emphasizes that every individual possesses diverse types of intelligence that extend beyond verbal or logical-mathematical intelligence. Gardner identified seven types of intelligences: verbal-linguistic, logical-mathematical, visual- spatial, bodily-kinesthetic, musical, interpersonal, and intrapersonal. According to Gardner, effective education should acknowledge the diversity of these intelligences and provide opportunities for individuals to develop the intelligences that reside within them. In the context of modern education, understanding both theories has significant implications. Implementing the concept of Growth Mindset can motivate students to persevere, overcome challenges, and develop holistically. On the other hand, understanding Multiple Intelligences enables educators to design learning experiences that recognize and leverage the diversity of students' intelligences. However, despite their relevance in education, the differences and similarities between the Growth Mindset Theory and Multiple Intelligences Theory have not been fully understood. Therefore, this research aims to analyze the differences and similarities between these theories in individual potential development, as well as their implications and relevance in modern education. By deepening our understanding of these theories, we can identify the best approaches to create an inclusive learning environment that motivates and supports the full development of each student's potential. This article will explore the comparison between Carol Dweck's and Howard Gardner's educational theories in the context of developing individual potential. INTRODUCTION The comparison will involve central concepts from both theories, such as mindset in facing challenges and the diversity of intelligences in the learning context. The article will analyze the similarities and differences between Dweck's and Gardner's perspectives, as well as their relevance in developing individual potential within the educational context. By understanding these different approaches, it is hoped that readers can gain deeper insights into how to perceive individual potential within the realm of education. With this understanding, educators and mentors can develop more effective approaches to assist each individual in reaching their full potential, overcoming potential obstacles, and cultivating strong self-confidence in the process of learning and growth. By applying appropriate approaches, educators and mentors can create an inclusive and motivating learning environment, empowering students to achieve academic success and develop their personal abilities with a strong sense of confidence. Abstract This research aims to compare Carol Dweck's Growth Mindset Theory and Howard Gardner's Multiple Intelligences Theory in the context of developing individual potential. The study utilizes literature review and comparative analysis as the research method to compare the two theories. In the pursuit of optimizing individual potential, the growth mindset emphasizes the importance of a developing mindset, viewing failures as opportunities for learning, and creating a supportive learning environment. On the other hand, multiple intelligences theory acknowledges the diversity of intelligences and advocates for the development of all types of intelligence in education. In inclusive modern education, the application of both theories can motivate students to grow and learn, develop self-confidence, and optimize their potential in various areas of intelligence. Through this approach, modern education can create a stimulating environment, facilitate holistic development, and produce individuals who are prepared to face future challenges. By understanding the differences and similarities between these two theories, education can adopt an inclusive approach that focuses on the full development of each individual's potential, ensuring that students are empowered to achieve academic success and optimal personal growth. 725 \| ANTHOR: Education and Learning Journal Volume 2 Nomor 6, 2023 e-ISSN: 2963 – 198X p-ISSN: 2963 – 2498 Corresponding Author: Mustofa Aji Prayitno PAI, Institut Agama Islam Negeri Ponorogo, Ponorogo, Jawa Timur, Indonesia: mustofaajiprayitno@gmail.com Copyright © 2023 Arshy Prodyanatasari, Mustofa Aji Prayitno, Ferdinand Salomo Leuwol, Siti Aminah, Maskur This work is licensed under Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) Corresponding Author: Mustofa Aji Prayitno PAI, Institut Agama Islam Negeri Ponorogo, Ponorogo, Jawa Timur, Indonesia: mustofaajiprayitno@gmail com e-ISSN: 2963 – 198X p-ISSN: 2963 – 2498 725 \| ANTHOR: Education and Learning Journal Volume 2 Nomor 6, 2023 Comparison of Educational Theories: Perspectives of Carol Dweck and Howard Gardner in Developing Individual Potential 726 \| ANTHOR: Education and Learning Journal Volume 2 Nomor 6, 2023 RESULTS AND DISCUSSIONS Carol Dweck and Growth Mindset Theory Carol Dweck and Growth Mindset Theory Carol Dweck is a renowned educational psychologist who has made significant contributions in the fields of developmental psychology and education (Prinz et al., 2021). One of her major contributions is the development of the concept of "Growth Mindset" and “Fixed Mindset.” The Mindset theory posits that individuals have two basic patterns of thinking that influence how they approach challenges, failures, and successes (Dweck & Yeager, 2019). Essentially, there are two mindsets that a person can possess (Haimovitz & Dweck, 2017). Firstly, the growth mindset is characterized by individuals who believe that abilities and intelligence can be developed through effort, practice, and experience. They see failures as opportunities to learn and grow, and they believe that success can be achieved through perseverance, effort, and the right strategies. People with a growth mindset tend to embrace challenges enthusiastically and are willing to try new things. Secondly, the fixed mindset where the individuals tend to believe that one's abilities and intelligence are fixed and cannot be changed. They view failures as signs of inadequacy and avoid challenges that might reveal their limitations. People with a fixed mindset often feel threatened by the success of others and tend to avoid the effort required for growth. According to Dweck, it is crucial for educators to understand the differences between these mindsets and play a role in helping students develop a Growth Mindset (Yeager & Dweck, 2020). By nurturing this mindset, educators can create a supportive, motivating, and empowering learning environment. Dweck believes that by cultivating this mindset in students, educators can create a positive and motivating learning environment that encourages students to take risks, embrace challenges, and strive beyond their perceived limitations (Dweck, 2014). Through effort and perseverance, students can develop their abilities to the fullest. In education, the application of Growth Mindset involves shifting the focus from mere achievement to the process of individual learning and growth (Subotnik et al., 2011). This approach encourages students to view failures as a normal part of learning and as opportunities for self-improvement. Growth Mindset provides a foundation for students to approach challenges with perseverance, set realistic goals, and employ effective strategies to achieve success. RESEARCH METHODS In this study, the methods employed are literature review and comparative analysis. The first step involves identifying relevant sources on Carol Dweck's Growth Mindset Theory and Howard Gardner's Multiple Intelligences Theory. These sources include books, scientific articles, educational journals, and other reputable resources discussing both theories. Subsequently, the researcher selects the most relevant and up-to-date sources for analysis, based on their reliability, accuracy, and currency of information presented. Relevant sources are obtained by conducting searches using relevant keywords and utilizing special features provided by various academic databases such as Google Scholar, Scopus, ERIC, JSTOR, and other academic sources. The search for theory sources is conducted using relevant keywords such as "Growth Mindset Theory," "Multiple Intelligences Theory," "Carol Dweck," "Howard Gardner," "education," "learning," and other keywords that are most relevant to the research topic. The researcher then carefully reads and studies the content of each source, noting key points related to the comparison between the two theories. In the comparative analysis, the researcher compares and contrasts the concepts, implications, relevance, and applications of both theories in modern education. By considering the findings from the analyzed sources, the researcher formulates conclusions regarding the comparison between the e-ISSN: 2963 – 198X p-ISSN: 2963 – 2498 Comparison of Educational Theories: Perspectives of Carol Dweck and Howard Gardner in Developing Individual Potential Comparison of Educational Theories: Perspectives of Carol Dweck and Howard Gardner in Developing Individual Potential Growth Mindset Theory and Multiple Intelligences Theory, as well as their implications and relevance in the context of modern education. Growth Mindset Theory and Multiple Intelligences Theory, as well as their implications and relevance in the context of modern education. 727 \| ANTHOR: Education and Learning Journal Volume 2 Nomor 6, 2023 Howard Gardner and Multiple Intelligences Theory Howard Gardner, a psychologist and professor at the Harvard Graduate School of Education, is known for his theory of Multiple Intelligences (Cavas & Cavas, 2020). Gardner argues that intelligence is not a single entity that can be measured by one IQ test, but rather a combination of various distinct intelligences (H. E. Gardner, 2000). According to the Theory of Multiple Intelligences, each individual possesses multiple and diverse intelligences that can be developed (H. Gardner & Hatch, 1989). Gardner identified eight different types of intelligences, these multiple intelligences, including linguistic, logical-mathematical, spatial, musical, bodily-kinesthetic, interpersonal, intrapersonal, and naturalistic intelligences, allow individuals to excel in different areas and have diverse strengths and ways of learning (Davis et al., 2011). Here is the explanation of each intelligence according to Gardner's Theory of Multiple Intelligences: p g 1. Linguistic Intelligence: This refers to the ability to use language effectively, both spoken and written. Individuals with linguistic intelligence excel in areas such as reading writing and public speaking p g 1. Linguistic Intelligence: This refers to the ability to use language effectively, both spoken and written. p g 1. Linguistic Intelligence: This refers to the ability to use language effectively, both spoken and written. Individuals with linguistic intelligence excel in areas such as reading, writing, and public speaking. 2. Logical-Mathematical Intelligence: This intelligence involves logical thinking, reasoning, and problem- solving skills. People with this intelligence are adept at analyzing patterns, making connections, and solving complex mathematical problems. 3. Visual-Spatial Intelligence: This intelligence involves the ability to perceive and manipulate visual information accurately. Individuals with strong visual-spatial intelligence excel in areas such as visualizing objects, maps, and designs. 4. Musical Intelligence: This intelligence relates to musical abilities, including recognizing musical patterns, pitch, rhythm, and composition. People with musical intelligence have a strong appreciation for music and often exhibit exceptional musical talents. p 5. Interpersonal Intelligence: This intelligence involves understanding and interacting effectively with others. Individuals with strong interpersonal intelligence excel in social situations, are empathetic, and have good communication skills. 6. Intrapersonal Intelligence: This intelligence refers to self-awareness and understanding one's own emotions, strengths, and weaknesses. Individuals with intrapersonal intelligence possess a deep understanding of themselves and have strong introspective capabilities. 7. Bodily-Kinesthetic Intelligence: This intelligence involves physical coordination, dexterity, and the ability to use one's body effectively. People with strong bodily-kinesthetic intelligence excel in activities such as sports, dance, and other physical pursuits. 7. RESULTS AND DISCUSSIONS e-ISSN: 2963 – 198X p-ISSN: 2963 – 2498 727 \| ANTHOR: Education and Learning Journal Volume 2 Nomor 6, 2023 Comparison of Educational Theories: Perspectives of Carol Dweck and Howard Gardner in Developing Individual Potential Comparison of Educational Theories: Perspectives of Carol Dweck and Howard Gardner in Developing Individual Potential 728 \| ANTHOR: Education and Learning Journal Volume 2 Nomor 6, 2023 RESULTS AND DISCUSSIONS In the classroom context, educators can implement Growth Mindset by providing constructive feedback to students, giving them opportunities for reflection and identifying areas of development, and offering appropriate support and guidance to help them reach their goals. Additionally, it is important for educators to create a safe and inclusive classroom environment where students feel comfortable taking risks and sharing their thoughts. The application of Growth Mindset also influences the way educators assess and praise students (Dweck, 2008). Praise should focus on effort, perseverance, and progress rather than solely on the end result. By providing appropriate praise, educators help students understand that success is not solely a result of innate intelligence but is a product of the effort and perseverance they put in. In order to implement Growth Mindset in education, it is important for educators to provide praise that focuses on effort, struggle, and student progress, rather than just the final outcome. This way, students are trained to see failure as a stepping stone towards growth and learn from their experiences. The implications of Growth Mindset in modern education are significant. By applying this approach, educators can help students develop a positive attitude towards learning, enhance intrinsic motivation, and assist them in reaching their full potential. Furthermore, understanding Growth Mindset can improve the classroom climate, reduce the stigma of failure, and foster healthy self-confidence in students. To create positive change in education, it is crucial for educators, parents, and other stakeholders to understand the concept of Growth Mindset and consistently apply it. By empowering students to develop this mindset, they will have a powerful tool to overcome obstacles, achieve success, and grow into individuals who reach their full potential. p Growth Mindset has gained widespread recognition in the field of education and is utilized by many educators to enhance student motivation, achievement, and success. This concept has important implications in shaping students' mindsets, helping them develop self-confidence, resilience, and a lifelong commitment to growth and learning. By fostering a growth mindset in students, educators can create a positive and supportive learning environment that encourages students to embrace challenges, learn from failures, and ultimately reach their full potential. Howard Gardner and Multiple Intelligences Theory Bodily-Kinesthetic Intelligence: This intelligence involves physical coordination, dexterity, and the ability to use one's body effectively. People with strong bodily-kinesthetic intelligence excel in activities such as sports, dance, and other physical pursuits. 8. Naturalistic Intelligence: This intelligence relates to the ability to observe and understand the natural world. Individuals with strong naturalistic intelligence have a keen awareness and appreciation of nature, as well as the ability to recognize patterns and classify living things. 8. Naturalistic Intelligence: This intelligence relates to the ability to observe and understand the natural world. Individuals with strong naturalistic intelligence have a keen awareness and appreciation of nature, as well as the ability to recognize patterns and classify living things. Gardner argues that everyone possesses a unique combination of these intelligences, and it is crucial for education to recognize and develop these intelligences holistically (Booth & O’Brien, 2008). He emphasizes that education should focus on the development of all types of intelligences, rather than prioritizing linguistic and logical-mathematical intelligences alone. By acknowledging and nurturing the diverse range of intelligences in individuals, education can better cater to the varied learning styles and strengths of students, leading to more effective and inclusive learning experiences. The Theory of Multiple Intelligences has significant implications in education (Kezar, 2001). Gardner suggests that effective education should incorporate various learning approaches that reflect the diversity of students' intelligences and learning styles. In practice, this means creating a learning environment that recognizes the diversity of students' intelligences and facilitates the development of these intelligences through the use of diverse teaching methods and assessments. By understanding the concept of Multiple Intelligences, educators can design inclusive and engaging learning experiences for students. This approach allows students to discover their own strengths and develop their potential in various areas of intelligence. By valuing the diversity of intelligences, educators can motivate students, build self-confidence, and enhance students' academic and personal achievements. Howard Gardner and the Theory of Multiple Intelligences have had a significant impact in the field of education, transforming the way we understand intelligence and engage students in the learning process. The theory emphasizes the importance of viewing each student as a unique individual with the potential to develop in various areas of intelligence. This perspective encourages educators to adopt a more personalized and differentiated approach to teaching, tailoring instructional methods and activities to suit the diverse intelligences present in their classrooms. Howard Gardner and Multiple Intelligences Theory g p imilarities between Carol Dweck's and Howard Gardner's theories imilarities between Carol Dweck's and Howard Gardner's theories In comparing Carol Dweck's Growth Mindset Theory and Howard Gardner's Multiple Intelligences Theory, there are significant similarities that highlight the importance of developing individual potential e-ISSN: 2963 – 198X p-ISSN: 2963 – 2498 728 \| ANTHOR: Education and Learning Journal Volume 2 Nomor 6, 2023 Comparison of Educational Theories: Perspectives of Carol Dweck and Howard Gardner in Developing Individual Potential through effort, recognition of diversity in intelligence, and emphasis on a supportive educational environment. Both theories underscore the belief that intelligence and abilities can be nurtured through dedication and practice, while also acknowledging that individuals possess unique combinations of intelligences that should be cultivated holistically. Here are the prominent points of similarity between Carol Dweck's Growth Mindset Theory and Howard Gardner's Multiple Intelligences Theory: through effort, recognition of diversity in intelligence, and emphasis on a supportive educational environment. Both theories underscore the belief that intelligence and abilities can be nurtured through dedication and practice, while also acknowledging that individuals possess unique combinations of intelligences that should be cultivated holistically. Here are the prominent points of similarity between Carol Dweck's Growth Mindset Theory and Howard Gardner's Multiple Intelligences Theory: g , g y g , p pp environment. Both theories underscore the belief that intelligence and abilities can be nurtured through dedication and practice, while also acknowledging that individuals possess unique combinations of intelligences that should be cultivated holistically. Here are the prominent points of similarity between Carol Dweck's Growth Mindset Theory and Howard Gardner's Multiple Intelligences Theory: p g g p q intelligences that should be cultivated holistically. Here are the prominent points of similarity between Carol Dweck's Growth Mindset Theory and Howard Gardner's Multiple Intelligences Theory: 1. Understanding that potential can be developed: The main similarity between both theories is the belief that individual potential can be developed through effort, practice, and experience. Dweck focuses on the development of one's mindset and the belief that one's abilities can grow and develop over time. Gardner, on the other hand, emphasizes the recognition that every individual possesses various intelligences that can be developed and expanded. 2. Avoiding a static understanding of abilities: Both Dweck and Gardner reject the notion that a person's abilities are predetermined or static. Howard Gardner and Multiple Intelligences Theory Dweck emphasizes the importance of seeing intelligence as something that can grow and be enhanced through effort and perseverance. Gardner proposes the view that intelligence is not a fixed, singular entity but consists of various intelligences that can be developed through experience and learning. 3. Emphasizing the importance of facing challenges and failures: Both Dweck and Gardner highlight the importance of viewing challenges and failures as opportunities for learning and growth. In the Growth Mindset, Dweck emphasizes the need to embrace failure as a normal part of the learning process and sees it as a step towards growth. Gardner, in the Multiple Intelligences Theory, encourages the recognition that challenges help develop different intelligences, and that failure is an integral part of the learning process. 4. Focus on holistic individual development: Both theories emphasize the importance of a holistic approach in developing individual potential. Dweck and Gardner both reject approaches that overly focus on one aspect of intelligence or another. Dweck promotes an approach that engages the entire mindset and beliefs of the individual, while Gardner advocates for the appreciation and development of all types of intelligence within an individual. 5. Emphasizing the importance of a supportive environment: Both Dweck and Gardner acknowledge that the environment plays a crucial role in individual development. Dweck highlights the importance of creating an environment that provides constructive feedback, empowers, and motivates students to strive for better. Gardner emphasizes the need for education to create an inclusive learning environment that recognizes various types of intelligence, allowing all students to develop their potential. 5. Emphasizing the importance of a supportive environment: Both Dweck and Gardner acknowledge that the environment plays a crucial role in individual development. Dweck highlights the importance of creating an environment that provides constructive feedback, empowers, and motivates students to strive for better. Gardner emphasizes the need for education to create an inclusive learning environment that recognizes various types of intelligence, allowing all students to develop their potential. While there are similarities between these two theories, it is important to note that they also have differences in their focus and context. Dweck is more focused on the development of individual mindset and beliefs, while Gardner focuses on the recognition and development of various types of intelligence. 729 \| ANTHOR: Education and Learning Journal Volume 2 Nomor 6, 2023 Howard Gardner and Multiple Intelligences Theory However, both Growth Mindset and Multiple Intelligences Theory have significant implications in education, and these two theories can complement each other in creating a supportive learning environment that motivates students to develop their potential optimally. Implications and Relevance in Modern Education The implications and relevance of Carol Dweck's Theory of Growth Mindset and Howard Gardner's Theory of Multiple Intelligences in modern education are highly significant. The Theory of Growth Mindset offers significant implications in shaping students' attitudes and motivation towards learning. By teaching students that abilities and intelligence can be developed through effort and perseverance, modern education can effectively motivate students to strive for their full potential. Through Growth Mindset, students are taught to see failure as an opportunity to learn and grow, rather than a permanent indicator of failure. This helps students develop self-confidence, resilience, and enthusiasm in facing educational challenges. Additionally, Growth Mindset creates a positive and inclusive learning environment. By providing constructive feedback, educators can build students' self-belief and enthusiasm for continuous development. A supportive and motivating learning environment encourages students to take risks, face challenges, and actively participate in the learning process. In modern education, the Growth Mindset approach becomes relevant in shaping an inclusive school culture, where every student feels valued, encouraged, and supported in developing their potential. On the other hand, the Theory of Multiple Intelligences offers different implications in terms of diverse curriculum design and varied teaching approaches. By recognizing the diversity of individual intelligences, educators can design learning experiences that reflect various types of intelligence. This enables students to develop their full potential in different areas of intelligence, including linguistic, logical-mathematical, visual-spatial, musical, interpersonal, intrapersonal, kinesthetic, and naturalistic. In modern education, the Multiple Intelligences approach provides relevance in creating an inclusive curriculum that offers diverse choices and teaching approaches, allowing each student to learn effectively according to their intelligence. 1. Growth Mindset by Carol Dweck: 1. Growth Mindset by Carol Dweck: a. Enhancing learning motivation: The Growth Mindset approach helps increase students' learning motivation by teaching them that abilities and intelligence can develop through effort and perseverance. By motivating students to see failure as an opportunity to learn and grow, modern education can encourage students to continuously strive for their full potential. a. Enhancing learning motivation: The Growth Mindset approach helps increase students' learning motivation by teaching them that abilities and intelligence can develop through effort and perseverance. By motivating students to see failure as an opportunity to learn and grow, modern education can encourage students to continuously strive for their full potential. b. Differences between Carol Dweck's and Howard Gardner's theories When comparing Carol Dweck's theory of Growth Mindset and Howard Gardner's theory of Multiple Intelligences, there are significant differences that encompass the conceptual focus, emphasized aspects, application areas, measurement focus, and application in the context of education. Each theory offers distinct perspectives on how individuals perceive intelligence and potential for growth, leading to diverse approaches in educational practices and strategies. Here are the prominent points of differences between Carol Dweck's Growth Mindset Theory and Howard Gardner's Multiple Intelligences Theory: y p g y 1. Conceptual Focus: Dweck's theory focuses on individuals' thought patterns and beliefs regarding their abilities and intelligence. Growth Mindset emphasizes that abilities can be developed through effort, practice, and experience, viewing failure as an opportunity for learning and growth. On the other hand, Gardner's theory focuses on the diversity of individual intelligences. Gardner identifies eight different types of intelligence and emphasizes that each individual possesses a unique combination of these intelligences. This theory recognizes that intelligence is not limited to verbal-linguistic and logical- mathematical intelligence. 1. Conceptual Focus: Dweck's theory focuses on individuals' thought patterns and beliefs regarding their abilities and intelligence. Growth Mindset emphasizes that abilities can be developed through effort, practice, and experience, viewing failure as an opportunity for learning and growth. On the other hand, Gardner's theory focuses on the diversity of individual intelligences. Gardner identifies eight different types of intelligence and emphasizes that each individual possesses a unique combination of these intelligences. This theory recognizes that intelligence is not limited to verbal-linguistic and logical- mathematical intelligence. g 2. Emphasized Aspects: Dweck highlights the importance of developing a positive attitude towards learning and challenges. The main focus is on changing individuals' thought patterns and beliefs related to their abilities and potential. Meanwhile, Gardner emphasizes the importance of recognizing and developing various types of intelligence in education. This theory highlights the diversity of individual intelligences and emphasizes the need for education to encompass and appreciate all types of intelligence. e-ISSN: 2963 – 198X p-ISSN: 2963 – 2498 729 \| ANTHOR: Education and Learning Journal Volume 2 Nomor 6, 2023 Comparison of Educational Theories: Perspectives of Carol Dweck and Howard Gardner in Developing Individual Potential 3. Application Areas: Dweck's theory is primarily applied in the context of education and personal development. Differences between Carol Dweck's and Howard Gardner's theories Growth Mindset is used to create a supportive and motivating learning environment, encouraging students to develop the belief that their abilities can grow through effort and perseverance. On the other hand, Gardner's theory has broader implications and is applied in various fields, including education, assessment, and curriculum design. This approach highlights the importance of recognizing and developing the diversity of individual intelligences within the learning experience. 4. Measurement Focus: Growth Mindset primarily focuses on changing individuals' thought patterns and beliefs, which are difficult to measure directly. Measurement often involves observing students' behaviors and thought patterns in facing challenges and failure. On the other hand, Multiple Intelligences theory focuses on recognizing and developing various types of intelligence. Intelligence measurement may involve tests, observations, and performance assessments designed to identify individuals' strengths in specific intelligence areas. g p g 5. Application in the Context of Education: Growth Mindset emphasizes the importance of providing constructive feedback, creating a supportive environment, and encouraging students to develop the belief that they can learn and grow through effort and perseverance. On the other hand, Multiple Intelligences theory influences curriculum design centered around the recognition and development of various types of intelligence. This approach encourages the use of diverse teaching strategies to facilitate the development of all types of individual intelligence. 5. Application in the Context of Education: Growth Mindset emphasizes the importance of providing constructive feedback, creating a supportive environment, and encouraging students to develop the belief that they can learn and grow through effort and perseverance. On the other hand, Multiple Intelligences theory influences curriculum design centered around the recognition and development of various types of intelligence. This approach encourages the use of diverse teaching strategies to facilitate the development of all types of individual intelligence. While there are differences between Carol Dweck's and Howard Gardner's theories, both have important implications in education. Growth Mindset emphasizes the importance of changing individuals' thought patterns and beliefs regarding their abilities, while Multiple Intelligences theory recognizes and develops the diversity of individual intelligences. These two theories can complement each other and help create an inclusive, motivating, and supportive learning environment that fosters the holistic development of students' potential. 730 \| ANTHOR: Education and Learning Journal Volume 2 Nomor 6, 2023 Implications and Relevance in Modern Education Creating a positive learning environment: By introducing the concept of Growth Mindset, educators can create a positive and inclusive learning environment. Praise that focuses on effort and perseverance can build students' self-confidence and enthusiasm in facing challenges, while constructive feedback helps them view mistakes as a step towards growth. e-ISSN: 2963 – 198X p-ISSN: 2963 – 2498 730 \| ANTHOR: Education and Learning Journal Volume 2 Nomor 6, 2023 Comparison of Educational Theories: Perspectives of Carol Dweck and Howard Gardner in Developing Individual Potential c. Overcoming the stigma of failure: By introducing Growth Mindset, modern education can change the perception of failure. Students will be more open to taking risks, facing challenges, and learning from their mistakes because they understand that failure is a natural part of the learning process and can bring them closer to success. c. Overcoming the stigma of failure: By introducing Growth Mindset, modern education can change the perception of failure. Students will be more open to taking risks, facing challenges, and learning from their mistakes because they understand that failure is a natural part of the learning process and can bring them closer to success. g 2. Multiple Intelligences by Howard Gardner: 2. Multiple Intelligences by Howard Gardner: a. Recognition of intelligence diversity: The theory of Multiple Intelligences acknowledges that every individual possesses different types of intelligence. In modern education, this has significant implications in creating an inclusive curriculum and offering a variety of learning experiences that value students' diverse intelligences. This allows students to develop their potential in various areas of intelligence. b. Individual-based instruction: By understanding the theory of Multiple Intelligences, educators can design diverse and tailored learning experiences based on students' individual intelligences. By leveraging different intelligences, educators can create a challenging environment that effectively meets each student's learning needs. c. Full potential development: The theory of Multiple Intelligences emphasizes the importance of developing all types of intelligence in students. It provides opportunities for students to discover their strengths and talents, enabling them to realize their full potential in various areas of intelligence. Thus, modern education can help students feel valued and succeed in an environment that recognizes and utilizes their intelligences comprehensively. By implementing the theories of Multiple Intelligences and Growth Mindset, modern education can provide more diverse, inclusive, and engaging learning experiences for students. CONCLUSION In comparing the theories of Growth Mindset by Carol Dweck and Multiple Intelligences by Howard Gardner in the development of individual potential, both have significant implications in modern education. The Growth Mindset theory emphasizes the importance of developing a positive attitude towards learning, overcoming the stigma of failure, and creating a supportive learning environment that motivates students to develop their potential. On the other hand, the Multiple Intelligences theory recognizes the diversity of individual intelligences and emphasizes the importance of developing all types of intelligences in education. Through diverse and inclusive teaching approaches, every student has the opportunity to develop their full potential in various areas of intelligence. To create an inclusive, holistic, and relevant modern education, the approaches of Growth Mindset and Multiple Intelligences provide crucial foundations. By applying both theories, modern education can create a positive learning environment, motivate students to continually grow and learn, and provide opportunities for individuals to achieve their full potential in various aspects of intelligence. Implications and Relevance in Modern Education The implications of both theories lead to the development of well-rounded students, where each individual has the opportunity to develop their full potential in various areas of intelligence and gain strong self-confidence. Furthermore, these approaches also help overcome the stigma of failure, enhance students' intrinsic motivation, and build a positive and supportive learning environment. In the ever-evolving modern education landscape, the application of both theories provides a crucial foundation in shaping an inclusive, holistic, and relevant educational approach for every student. 731 \| ANTHOR: Education and Learning Journal Volume 2 Nomor 6, 2023 732 \| ANTHOR: Education and Learning Journal Volume 2 Nomor 6, 2023 e-ISSN: 2963 – 198X p-ISSN: 2963 – 2498 REFERENCES Booth, R., & O’Brien, P. J. (2008). An Holistic Approach for Counsellors: Embracing Multiple Intelligences. International Journal for the Advancement of Counselling, 30(2), 79–92. https://doi.org/10.1007/s10447-008-9046-0 p g Cavas, B., & Cavas, P. (2020). Multiple Intelligences Theory—Howard Gardner (pp. 405–418). https://doi.org/10.1007/978-3-030-43620-9_27 p g Davis, K., Christodoulou, J., Seider, S., & Gardner, H. E. (2011). The Theory of Multiple Intelligences. Cambridge University Press. g y Dweck, C. S. (2008). Mindsets and Math/Science Achievement. Carnegie Corporation of New York g p Dweck, C. S. (2014). Teachers’ Mindsets: “Every Student has Something to Teach Me.” Edu Horizons, 93(2), 10–15. https://doi.org/10.1177/0013175X14561420 Dweck, C. S., & Yeager, D. S. (2019). Mindsets: A View from Two Eras. Perspectives on Psychological Science, 14(3), 481–496. https://doi.org/10.1177/1745691618804166 p g Gardner, H. (2006). The Development and Education of the Mind. Routledge. https://doi.org/10.4324/9780203019689 Gardner H E (2000) Intelligence Reframed: Multiple Intelligences for the 21st Century Basic Book 731 \| ANTHOR: Education and Learning Journal Volume 2 Nomor 6, 2023 e-ISSN: 2963 – 198X p-ISSN: 2963 – 2498 Comparison of Educational Theories: Perspectives of Carol Dweck and Howard Gardner in Developing Individual Potential Gardner, H., & Hatch, T. (1989). Educational Implications of the Theory of Multiple Intelligences. Educational Researcher, 18(8), 4–10. https://doi.org/10.3102/0013189X018008004 Haimovitz, K., & Dweck, C. S. (2017). The Origins of Children’s Growth and Fixed Mindsets: New Research and a New Proposal. Child Development, 88(6), 1849–1859. https://doi.org/10.1111/cdev.12955 Kezar, A. (2001). Theory of Multiple Intelligences: Implications for Higher Education. Innovative Higher Education, 26(2), 141–154. https://doi.org/10.1023/A:1012292522528 Lolang, E., Rais, R., Oualeng, A., & Prayitno, M. A. (2023). Analysis of Educational Messages in The Lion King Movie: Perspectives on Character Education and Environmental Conservation. Competitive: Journal of Education, 2(2), 122–130. Prinz, A., Zeeb, H., Flanigan, A. E., Renkl, A., & Kiewra, K. A. (2021). Conversations with Five Highly Successful Female Educational Psychologists: Patricia Alexander, Carol Dweck, Jacquelynne Eccles, Mareike Kunter, and Tamara van Gog. Educational Psychology Review, 33(2), 763–795. https://doi.org/10.1007/s10648-020-09552-y Subotnik, R. F., Olszewski-Kubilius, P., & Worrell, F. C. (2011). Rethinking Giftedness and Gifted Education. Psychological Science in the Public Interest, 12(1), 3–54. https://doi.org/10.1177/1529100611418056 Yeager, D. S., & Dweck, C. S. (2020). What Can be Learned from Growth Mindset Controversies? American Psychologist, 75(9), 1269–1284. https://doi.org/10.1037/amp0000794. e-ISSN: 2963 – 198X p-ISSN: 2963 – 2498 e-ISSN: 2963 – 198X p-ISSN: 2963 – 2498
https://openalex.org/W2909300313	https://chemrxiv.org/engage/api-gateway/chemrxiv/assets/orp/resource/item/60c73f1f337d6c6cabe264de/original/enhancing-electrocatalytic-co2-reduction-using-a-system-integrated-approach-to-catalyst-discovery.pdf	English	null	CO<sub>2</sub> reduction on gas-diffusion electrodes and why catalytic performance must be assessed at commercially-relevant conditions	Energy & environmental science	2,019	cc-by	13,982	Enhancing electrocatalytic CO2 reduction using a system-integrated 1 approach to catalyst discovery 2 3 Thomas Burdyny,1 and Wilson A. Smith,1 4 1Materials for Energy Conversion and Storage, Department of Chemical Engineering, Delft 5 University of Technology, 2629 HZ Delft, The Netherlands 6 7 () Correspondence and requests for materials can be addressed to Thomas Burdyny 8 (t.e.burdyny@tudelft.nl) and Wilson A. Smith (W.smith@tudelft.nl) 9 10 11 Electrocatalytic CO2 reduction has the dual-promise of neutralizing carbon emissions in 12 the near future, while providing a long-term pathway to create energy-dense chemicals 13 and fuels from atmospheric CO2. The field has advanced immensely in recent years, 14 taking significant strides towards commercial realization. Catalyst innovations have 15 played a pivotal role in these advances, with a steady stream of new catalysts providing 16 gains in CO2 conversion efficiencies and selectivities of both C1 and C2 products. 17 Comparatively few of these catalysts have been tested at commercially-relevant current 18 densities (~200 mA/cm2) due to transport limitations in traditional testing configurations 19 and a research focus on fundamental catalyst kinetics, which are measured at 20 substantially lower current densities. A catalyst’s selectivity and activity, however, have 21 been shown to be highly sensitive to the local reaction environment, which changes 22 drastically as a function of reaction rate. As a consequence of this, the surface properties 23 of many CO2 reduction catalysts risk being optimized for the wrong operating 24 conditions. The goal of this article is to communicate the substantial impact of reaction 25 rate on catalytic behaviour and the CO2 reduction reaction. In brief, this work motivates 26 high current density catalyst testing as a necessary step to properly evaluate materials 27 for electrochemical CO2 reduction, and to accelerate the technology toward its 28 envisioned application of neutralizing CO2 emissions on a global scale. 29 Enhancing electrocatalytic CO2 reduction using a system-integrated 1 approach to catalyst discovery 2 3 Thomas Burdyny,1 and Wilson A. Smith,1 4 1Materials for Energy Conversion and Storage, Department of Chemical Engineering, Delft 5 University of Technology, 2629 HZ Delft, The Netherlands 6 7 () Correspondence and requests for materials can be addressed to Thomas Burdyny 8 (t.e.burdyny@tudelft.nl) and Wilson A. Smith (W.smith@tudelft.nl) 9 10 11 Electrocatalytic CO2 reduction has the dual-promise of neutralizing carbon emissions in 12 the near future, while providing a long-term pathway to create energy-dense chemicals 13 and fuels from atmospheric CO2. The field has advanced immensely in recent years, 14 taking significant strides towards commercial realization. Catalyst innovations have 15 played a pivotal role in these advances, with a steady stream of new catalysts providing 16 gains in CO2 conversion efficiencies and selectivities of both C1 and C2 products. 17 Comparatively few of these catalysts have been tested at commercially-relevant current 18 densities (~200 mA/cm2) due to transport limitations in traditional testing configurations 19 and a research focus on fundamental catalyst kinetics, which are measured at 20 substantially lower current densities. A catalyst’s selectivity and activity, however, have 21 been shown to be highly sensitive to the local reaction environment, which changes 22 drastically as a function of reaction rate. As a consequence of this, the surface properties 23 of many CO2 reduction catalysts risk being optimized for the wrong operating 24 conditions. The goal of this article is to communicate the substantial impact of reaction 25 rate on catalytic behaviour and the CO2 reduction reaction. In brief, this work motivates 26 high current density catalyst testing as a necessary step to properly evaluate materials 27 for electrochemical CO2 reduction, and to accelerate the technology toward its 28 envisioned application of neutralizing CO2 emissions on a global scale. 29 Enhancing electrocatalytic CO2 reduction using a system-integrated 1 approach to catalyst discovery 2 Under an applied potential and in the presence of an appropriate catalyst, carbon-dioxide 30 (CO2) and water can be electrocatalytically converted into syngas (CO + H2), ethylene (C2H4), 31 methane (CH4), ethanol (C2H5OH) and formate (HCOOH) among other products. The 32 collective market size of these reduction products is >500 Mton/year, indicating the potential 33 scale of a commercially competitive CO2 electrolyzer;1 in the process, and of utmost 34 importance and urgency, this process may aid in reducing fossil fuels by supplanting current 35 production routes. Excitingly, CO2 electroreduction catalysts have shown enough promise that 36 we are beginning to see the first steps towards commercial application of the technology, 37 including more and more start-ups (see OPUS12, CERT, Dioxide Materials) and established 38 companies (Siemens) focusing on system design and engineering. As a result, researchers are 39 now targeting lower overall cell potentials by improving other parts of the conversion unit,2–5 40 while looking to more efficiently integrate electrochemical CO2 conversion units with 41 upstream and downstream processes.6 Future devices will also need to demonstrate stable 42 long-term operation (>20,000 hours) at substantial current densities (>200 mA/cm2) to 43 minimize the capital-expenditure of a conversion unit to economically-compelling levels.1,7,8 44 Due to the low solubility of CO2 in aqueous-fed systems that limits CO2 conversion to current 45 densities of ~35 mA/cm2,9–11 researchers have turned to pressurized electrolytes and 46 gas-diffusion layer-based systems to supply enough CO2 to the catalyst layer to sustain higher 47 current densities. Despite these capabilities, an overwhelming percentage (>95%)12,13 of 48 fundamental studies and catalytic materials are still developed, tested and characterized in 49 classical H-cell configurations, where current densities are limited. The local catalytic 50 environment, and subsequently the energetics of the reactions occurring on a catalyst’s 51 surface, are known however to be highly sensitive to changes in reaction rate. Therefore a 52 fundamental question remains: how representative are the conclusions from aqueous-fed 53 systems that are constrained to ~35 mA/cm2 when the goal is to achieve >200 mA/cm2? 54 Under an applied potential and in the presence of an appropriate catalyst, carbon-dioxide 30 (CO2) and water can be electrocatalytically converted into syngas (CO + H2), ethylene (C2H4), 31 methane (CH4), ethanol (C2H5OH) and formate (HCOOH) among other products. Enhancing electrocatalytic CO2 reduction using a system-integrated 1 approach to catalyst discovery 2 To aid in the analysis we draw upon recent literature findings from electrochemical 58 experiments, transport phenomena and Density-Functional Theory (DFT) modelling. 59 Due to the promise of electrochemical CO2 reduction technology, and a lack of selective and 60 efficient cathode materials, a large fraction of the field has undergone a global, 61 multidisciplinary effort over the last decade to find new and better catalysts. The search is 62 complicated by the large number of surface factors impacting activity including intermediate 63 binding energy,14,15 (via coordination16,17 and site availability18–20), packing,21,22 kinetic supply 64 of reactants,23 desorption of products,24 adsorbate-adsorbate interactions,25 etc. The urgency 65 of these efforts is illustrated by the large number of material-centric review papers published 66 in the last year alone on catalyst development/optimization.10,26–30 Here, we define the catalyst 67 Due to the promise of electrochemical CO2 reduction technology, and a lack of selective and 60 efficient cathode materials, a large fraction of the field has undergone a global, 61 multidisciplinary effort over the last decade to find new and better catalysts. The search is 62 complicated by the large number of surface factors impacting activity including intermediate 63 binding energy,14,15 (via coordination16,17 and site availability18–20), packing,21,22 kinetic supply 64 of reactants,23 desorption of products,24 adsorbate-adsorbate interactions,25 etc. The urgency 65 of these efforts is illustrated by the large number of material-centric review papers published 66 in the last year alone on catalyst development/optimization.10,26–30 Here, we define the catalyst 67 as the surface on which CO2 is reduced. As with any catalytic process, however, the local and 68 system reaction environments play equally important roles in efficiently driving specific 69 reactions, while suppressing unwanted competing ones. Many researchers have reported the 70 extreme sensitivity of the reaction to changes in local pH,31–35 electrolytes36,37 and cations38–41 71 (easily illustrated by replacing K+ with Na+). The ‘catalyst’ that notably reduces the energy 72 barrier for CO2 reduction processes is then very much a combined material and environmental 73 effect, rather than that of the catalyst’s surface (composition, coordination, 74 nano/mesostructure) alone. 75 as the surface on which CO2 is reduced. As with any catalytic process, however, the local and 68 system reaction environments play equally important roles in efficiently driving specific 69 reactions, while suppressing unwanted competing ones. Enhancing electrocatalytic CO2 reduction using a system-integrated 1 approach to catalyst discovery 2 The 32 collective market size of these reduction products is >500 Mton/year, indicating the potential 33 scale of a commercially competitive CO2 electrolyzer;1 in the process, and of utmost 34 importance and urgency, this process may aid in reducing fossil fuels by supplanting current 35 production routes. Excitingly, CO2 electroreduction catalysts have shown enough promise that 36 we are beginning to see the first steps towards commercial application of the technology, 37 including more and more start-ups (see OPUS12, CERT, Dioxide Materials) and established 38 companies (Siemens) focusing on system design and engineering. As a result, researchers are 39 now targeting lower overall cell potentials by improving other parts of the conversion unit,2–5 40 while looking to more efficiently integrate electrochemical CO2 conversion units with 41 upstream and downstream processes.6 Future devices will also need to demonstrate stable 42 long-term operation (>20,000 hours) at substantial current densities (>200 mA/cm2) to 43 minimize the capital-expenditure of a conversion unit to economically-compelling levels.1,7,8 44 Due to the low solubility of CO2 in aqueous-fed systems that limits CO2 conversion to current 45 densities of ~35 mA/cm2,9–11 researchers have turned to pressurized electrolytes and 46 gas-diffusion layer-based systems to supply enough CO2 to the catalyst layer to sustain higher 47 current densities. Despite these capabilities, an overwhelming percentage (>95%)12,13 of 48 fundamental studies and catalytic materials are still developed, tested and characterized in 49 classical H-cell configurations, where current densities are limited. The local catalytic 50 environment, and subsequently the energetics of the reactions occurring on a catalyst’s 51 surface, are known however to be highly sensitive to changes in reaction rate. Therefore a 52 fundamental question remains: how representative are the conclusions from aqueous-fed 53 This article seeks to shed light on this question by summarizing how the local reaction 55 environment is known to vary as a function of current density, and how these changes may 56 impact reactions occurring on a catalyst’s surface when pushed to commercial current 57 densities. To aid in the analysis we draw upon recent literature findings from electrochemical 58 experiments, transport phenomena and Density-Functional Theory (DFT) modelling. 59 This article seeks to shed light on this question by summarizing how the local reaction 55 environment is known to vary as a function of current density, and how these changes may 56 impact reactions occurring on a catalyst’s surface when pushed to commercial current 57 densities. Enhancing electrocatalytic CO2 reduction using a system-integrated 1 approach to catalyst discovery 2 Many researchers have reported the 70 extreme sensitivity of the reaction to changes in local pH,31–35 electrolytes36,37 and cations38–41 71 (easily illustrated by replacing K+ with Na+). The ‘catalyst’ that notably reduces the energy 72 barrier for CO2 reduction processes is then very much a combined material and environmental 73 effect, rather than that of the catalyst’s surface (composition, coordination, 74 nano/mesostructure) alone. 75 The above distinction, while central to any catalytic process, warrants particular attention here 76 due to the unique peculiarities of electrocatalytic CO2 reduction in aqueous solvents. 77 Specifically, while the local environment directly influences reaction pathways and kinetics, 78 the reduction reaction itself greatly disturbs the local environment. At the root of this 79 reaction-driven sensitivity is the requirement for both CO2 and protons (H+) to participate in 80 the CO2 reduction process. The ever-present, and in many cases more thermodynamically 81 favourable, hydrogen evolution reaction (HER) then simultaneously competes with CO2 82 reduction for protons and electrons. At extremely low current densities (<1 mA/cm2), these 83 protons can be supplied to either reaction directly by hydronium contained within the local 84 electrolyte (Fig. 1a).42 As hydronium is depleted, water reduction fills the role as a hydrogen 85 source while the unused hydroxide molecule generated as a by-product, rapidly increases the 86 local electrolyte pH (Fig. 1c).43,44 At slightly more moderate CO2 reduction current densities 87 in aqueous-fed systems (~35 mA/cm2 for C1 products, ~100 mA/cm2 for C2 products32,36), 88 CO2 ultimately becomes depleted by a combination of diffusion limitations from the bulk 89 electrolyte, and the now unfavourable bicarbonate-equilibrium conditions as a result of the 90 Fig. 1: The relative effect of current density on the reaction (a) proton source, (b) concentration of CO2 and (c) pH at the surface of a CO2 reduction catalyst. Fig. 1: The relative effect of current density on the reaction (a) proton source, (b) concentration of CO2 and (c) pH at the surface of a CO2 reduction catalyst. increased local pH.45,46 Thus, as the reaction proceeds from 0 mA/cm2 to CO2-depletion (Fig. 91 1b), the surface coverage and binding energies of key intermediates on a catalyst’s surface are 92 ultimately impacted through these changes in the local environment;44,47,48 not dissimilar from 93 the surface effects incurred by varying material composition, structure or morphology. Enhancing electrocatalytic CO2 reduction using a system-integrated 1 approach to catalyst discovery 2 The 94 activity of a catalyst is then identified by the environment around it, as much as its physical 95 make-up. 96 increased local pH.45,46 Thus, as the reaction proceeds from 0 mA/cm2 to CO2-depletion (Fig. 91 1b), the surface coverage and binding energies of key intermediates on a catalyst’s surface are 92 ultimately impacted through these changes in the local environment;44,47,48 not dissimilar from 93 the surface effects incurred by varying material composition, structure or morphology. The 94 activity of a catalyst is then identified by the environment around it, as much as its physical 95 make-up. 96 The inherent sensitivity of the reaction to changing local conditions is seen in literature to 97 directly and indirectly drive many of the experimental choices in electrolyte type and 98 concentration. Unsurprisingly, a catalyst can perform exceptionally well or poorly depending 99 on the medium in which it is tested. The importance of the electrolyte medium on catalytic 100 behaviour is most clearly displayed through the near ubiquitous use of low KHCO3 101 concentrations in reports of high selectivity C2 production on nanostructured49–51 and oxide- 102 derived Cu52–55 in H-cells. Here the poor buffering capacity of the electrolyte causes the pH 103 close to the electrode to quickly increase at low current densities (<20 mA/cm2), helping to 104 promote C2 products and suppress the competing CH4 and H2 reactions.31,56 The morphology 105 of the catalyst can be used to provide similar cumulative effects,52 which ultimately makes 106 properly separating the direct contributions of intrinsic catalytic activity and the local 107 environment even more difficult. To this point, the field has collectively learned to manipulate 108 both catalysts and local electrochemical conditions as needed to optimize CO2 reduction 109 performance metrics at primarily indiscriminate current densities. Given this widely- 110 implemented knowledge that the environment is critical to catalytic behaviour, and that these 111 conditions change as a function of current density, it is our perspective that the environment at 112 commercially-relevant current densities (>200 mA/cm2) should be used as the criterion for 113 assessing catalytic performance and suitability. 114 115 Effect of cell configuration and reaction rate on CO2 reduction environments 116 Effect of cell configuration and reaction rate on CO2 reduction environments 116 The number of CO2 electroreduction experiments reported at elevated current densities is 117 rapidly increasing, owed in part to the maturity of the field as well as the observed 118 performance benefits. Recent work has reviewed various electrochemical architectures 119 capable of delivering enough CO2 to the catalyst to reach current densities >200 mA/cm2,12,13 120 summarizing in detail the impacts of different components and configurations on system 121 performance. While we refer the reader to these publications for specific technical advances 122 in the field, we will briefly summarize several commonly-used architectures (Fig. 2) and 123 aspects of their operation to illustrate how the reaction environment around the catalyst layer 124 is influenced by elevated current densities. 125 Fig. 2: Comparative mass transport phenomena in commonly-used electrochemical CO2 reduction configurations. (a) Cell view for an H-cell configuration with a catalyst deposited on a solid substrate, (b) Cell view for a catalyst deposited on a gas-diffusion layer with a flowing catholyte channel, (c) Cell view for a catalyst deposited on a gas-diffusion layer with a non-flowing catholyte, (d) species transport to and from a catalyst layer in which CO2 is supplied via diffusion from the bulk electrolyte on the microscale (~50 µm). (e) a CO2 reduction catalyst layer deposited onto a hydrophobic substrate with CO2 diffusion from a nearby gas-liquid interface (~50 nm). Liquid species diffuse to the ion- exchange membrane through either a bulk flowing electrolyte or a solid-supported electrolyte layer. Fig. 2: Comparative mass transport phenomena in commonly-used electrochemical CO2 reduction configurations. (a) Cell view for an H-cell configuration with a catalyst deposited on a solid substrate, (b) Cell view for a catalyst deposited on a gas-diffusion layer with a flowing catholyte channel, (c) Cell view for a catalyst deposited on a gas-diffusion layer with a non-flowing catholyte, (d) species transport to and from a catalyst layer in which CO2 is supplied via diffusion from the bulk electrolyte on the microscale (~50 µm). (e) a CO2 reduction catalyst layer deposited onto a hydrophobic substrate with CO2 diffusion from a nearby gas-liquid interface (~50 nm). Liquid species diffuse to the ion- exchange membrane through either a bulk flowing electrolyte or a solid-supported electrolyte layer. In each of the described configurations CO2 is supplied to a catalyst layer that is fully or 126 partially immersed in a conductive electrolyte. Effect of cell configuration and reaction rate on CO2 reduction environments 116 This CO2 can diffuse through the 127 hydrodynamic boundary layer of a saturated bulk electrolyte as in a standard H-cell 128 configuration (Fig. 2a and d),45,46,57 or from a nearby gas-phase with a much shorter diffusion 129 pathway (Fig. 2b, c and e).58–61 Configurations where CO2 is provided from the gas-phase use 130 a gas-diffusion layer to form a gas-liquid interface adjacent to the catalyst layer. 131 Electrochemical CO2 reactors using gas-diffusion layers have historically been used in diverse 132 configurations,13,62–64 but range primarily from a Kenis-type reactor with a flowing 133 catholyte58,65–73 (Fig. 2b), to membrane electrode assemblies2,5,12,74,75 which directly combine 134 a gas-diffusion layer, catalyst and ion exchange membrane into one unit (Fig. 2c). These latter 135 configurations are occasionally denoted as ‘catholyte-free’ or gas-phase electrolysis due to the 136 lack of a flowing electrolyte between the catalyst layer and ion-exchange membrane. Liquid 137 water, however, is reported to be present in the porous catalyst layer in the majority of cases. 138 Further, research has shown that without the direct presence of a solid-supported electrolyte, 139 CO2 reduction selectivity can be heavily penalized.2,5,64 The two cathode variations shown in 140 Fig. 2b and c then similarly involve the diffusion of CO2 across a gas-liquid interface and 141 through a thin electrolyte to a porous catalyst layer (Fig. 2e), where evidence suggests that the 142 reaction occurs primarily in the aqueous phase rather than at a three-phase solid-liquid-gas 143 interface. Water management is essential to maintain a stable gas-liquid equilibrium as both 144 flooding of the gas-diffusion layer and evaporation of the catholyte will impact CO2 transport 145 to the catalyst layer.5,70,76 A catalyst can be applied to the gas-diffusion layer via drop-casting, 146 airbrushing, electrodeposition, compression, or incorporation into the porous layer itself. 147 The primary difference between these architectures is a roughly 3-order of magnitude 148 reduction in the CO2 diffusion pathway to the surface of the catalyst: from ~50 μm in an H- 149 cell to ~50 nm using a gas-diffusion layer (Fig. 2b,c), which allows for the increased 150 Further, research has shown that without the direct presence of a solid-supported electrolyte, 139 CO2 reduction selectivity can be heavily penalized.2,5,64 The two cathode variations shown in 140 Fig. 2b and c then similarly involve the diffusion of CO2 across a gas-liquid interface and 141 through a thin electrolyte to a porous catalyst layer (Fig. Effect of cell configuration and reaction rate on CO2 reduction environments 116 Furthermore, for slightly acidic CO2-saturated 165 electrolytes, the solvent itself can also act as the sole proton source (via H3O+) at current 166 densities <1 mA/cm2 until becoming depleted and replaced by water reduction (Fig. 1a). It is 167 then entirely possible for a CO2 reduction catalyst to be effective within the Tafel region but 168 exhibit slower kinetics at higher potentials/current densities if the water reduction step to 169 liberate protons becomes rate-determining. Within this low current density region, where the 170 most important electrochemical characterizations of a material take place, the reaction 171 environment is then extremely sensitive, complicating analysis of intrinsic catalyst behaviour 172 and the reaction mechanisms at play. 173 The first observation from these curves is that the CO2 concentration and pH in the electrolyte 154 at low current densities is relatively similar for both configurations (Fig. 3). This is to be 155 expected as the decreased CO2 diffusion pathway does not affect the maximum availability of 156 CO2 in an electrolyte, and OH- diffusion is moderated by the catholyte. In each case a sudden 157 increase in local pH is observed for the weakly-buffered electrolytes (1 M KCl, 158 0.1 M KHCO3) as the hydroxide generated as a by-product of water-splitting cannot diffuse 159 away fast enough or be immediately buffered by the solution (Fig. 3b and d). Shown here, and 160 in recent experimental work by the Koper group,42 the electrode pH can in fact shift by as 161 much as 6 units within the first 1 - 2 mA/cm2. Not only could the reaction itself be altered by 162 this shift, but large pH differences between the reference and working electrodes in this 163 current density range complicate determination of the equilibrium potential, and subsequently, 164 product onset potentials and Tafel slopes. Furthermore, for slightly acidic CO2-saturated 165 electrolytes, the solvent itself can also act as the sole proton source (via H3O+) at current 166 densities <1 mA/cm2 until becoming depleted and replaced by water reduction (Fig. 1a). It is 167 then entirely possible for a CO2 reduction catalyst to be effective within the Tafel region but 168 exhibit slower kinetics at higher potentials/current densities if the water reduction step to 169 liberate protons becomes rate-determining. Effect of cell configuration and reaction rate on CO2 reduction environments 116 2e), where evidence suggests that the 142 reaction occurs primarily in the aqueous phase rather than at a three-phase solid-liquid-gas 143 interface. Water management is essential to maintain a stable gas-liquid equilibrium as both 144 flooding of the gas-diffusion layer and evaporation of the catholyte will impact CO2 transport 145 to the catalyst layer.5,70,76 A catalyst can be applied to the gas-diffusion layer via drop-casting, 146 airbrushing, electrodeposition, compression, or incorporation into the porous layer itself. 147 The primary difference between these architectures is a roughly 3-order of magnitude 148 reduction in the CO2 diffusion pathway to the surface of the catalyst: from ~50 μm in an H- 149 cell to ~50 nm using a gas-diffusion layer (Fig. 2b,c), which allows for the increased 150 maximum current densities reported in literature. Using a simple 1D reaction-diffusion model 151 developed previously,45,46,57–59 we can also approximate the similarities and differences in the 152 local catalytic environments as a function of reaction rate for both cases. 153 maximum current densities reported in literature. Using a simple 1D reaction-diffusion model 151 developed previously,45,46,57–59 we can also approximate the similarities and differences in the 152 local catalytic environments as a function of reaction rate for both cases. 153 y The first observation from these curves is that the CO2 concentration and pH in the electrolyte 154 at low current densities is relatively similar for both configurations (Fig. 3). This is to be 155 expected as the decreased CO2 diffusion pathway does not affect the maximum availability of 156 CO2 in an electrolyte, and OH- diffusion is moderated by the catholyte. In each case a sudden 157 increase in local pH is observed for the weakly-buffered electrolytes (1 M KCl, 158 0.1 M KHCO3) as the hydroxide generated as a by-product of water-splitting cannot diffuse 159 away fast enough or be immediately buffered by the solution (Fig. 3b and d). Shown here, and 160 in recent experimental work by the Koper group,42 the electrode pH can in fact shift by as 161 much as 6 units within the first 1 - 2 mA/cm2. Not only could the reaction itself be altered by 162 this shift, but large pH differences between the reference and working electrodes in this 163 current density range complicate determination of the equilibrium potential, and subsequently, 164 product onset potentials and Tafel slopes. Effect of cell configuration and reaction rate on CO2 reduction environments 116 Within this low current density region, where the 170 most important electrochemical characterizations of a material take place, the reaction 171 environment is then extremely sensitive, complicating analysis of intrinsic catalyst behaviour 172 and the reaction mechanisms at play. 173 In our reaction-diffusion model, the maximum predicted CO2 reduction current densities in 174 In our reaction-diffusion model, the maximum predicted CO2 reduction current densities in 174 the aqueous-fed system are again <35 mA/cm2 for two-electron reduction processes (Fig. 3b). 175 In our reaction-diffusion model, the maximum predicted CO2 reduction current densities in 174 the aqueous-fed system are again <35 mA/cm2 for two-electron reduction processes (Fig. 3b). 175 As shown here in Fig. 3d, and in previous gas-diffusion layer modelling work, we can see 176 however that the maximum current densities are much higher in the case of a gas-diffusion 177 layer as a result of the reduced CO2 diffusion distance; here all electrolytes are capable of 178 sustaining current densities over the proposed 200 mA/cm2 minimum. This agrees well with 179 experimental literature where current densities substantially higher than 200 mA/cm2 have 180 been reported, including the first report using gas-diffusion layers for CO2 reduction from 181 Cook et al. in 1990.62 In recent work by Dinh et al., for example, current densities of 1 A/cm2 182 were reached in 1 M KOH at 1 atm with an overall CO2 reduction selectivity over 90%.70 183 While these current densities might not be desired economically due to the corresponding 184 increase in cell potential required,1,7 it indicates that substantial CO2 reduction reaction rates 185 are possible even if the CO2 partial pressure is reduced. This is important for future large- 186 scale devices (>100 cm2 catalyst areas) where CO2 partial pressure will decrease as it is 187 converted in the device. It is also worth noting that the proposed 200 mA/cm2 minimum 188 current density is defined primarily for two-electron reduction products. For multi-carbon 189 products requiring more electrons per CO2 molecule converted, this current density limit must 190 be higher to have the same molecular yield per unit area. 191 A final extremely important takeaway from Fig. 3 is that the pH near the electrode will be 192 inevitably high at commercially-viable current densities (>200 mA/cm2) in all of the most 193 commonly-used electrolytes, regardless of the choice of catalyst or electrolyte (Fig. 3d). Effect of cell configuration and reaction rate on CO2 reduction environments 116 2D/3D effects, morphology, partial 212 pressure of CO2 in the gas-diffusion layer, etc.). A more detailed representation of CO2 213 diffusion through a porous electrode structure, for instance, may provide further insight but 214 requires consideration of pore sizes, structures and additives,77 which are catalyst-specific and 215 beyond the scope of this article. These studies are particularly needed for zero-gap membrane 216 electrode assemblies where the extremely-low catholyte volume is expected to make the local 217 reaction conditions more sensitive to the specific device configuration and the properties of 218 the ion-exchange membrane. 219 how various factors may influence the reaction (e.g. 2D/3D effects, morphology, partial 212 pressure of CO2 in the gas-diffusion layer, etc.). A more detailed representation of CO2 213 diffusion through a porous electrode structure, for instance, may provide further insight but 214 requires consideration of pore sizes, structures and additives,77 which are catalyst-specific and 215 beyond the scope of this article. These studies are particularly needed for zero-gap membrane 216 electrode assemblies where the extremely-low catholyte volume is expected to make the local 217 reaction conditions more sensitive to the specific device configuration and the properties of 218 the ion-exchange membrane. 219 Impact of high current densities on CO2 reduction catalyst testing 220 In the previous section we discussed the impact that both hydrogen evolution and CO2 221 reduction have on the local environment up to commercially-relevant current densities. This 222 section discusses the important opposite side of the equation, how does access to surplus CO2 223 and the predicted local environment at higher current densities impact the reactions occurring 224 on the surface of the catalyst. More importantly, can we use this understanding to design even 225 better catalysts or conditions that may advance CO2 electroreduction performance further. 226 In the previous section we discussed the impact that both hydrogen evolution and CO2 221 reduction have on the local environment up to commercially-relevant current densities. This 222 section discusses the important opposite side of the equation, how does access to surplus CO2 223 and the predicted local environment at higher current densities impact the reactions occurring 224 on the surface of the catalyst. More importantly, can we use this understanding to design even 225 better catalysts or conditions that may advance CO2 electroreduction performance further. 226 In catalyst design the surface of a material is altered as a means of modifying catalytic 227 behaviour. Effect of cell configuration and reaction rate on CO2 reduction environments 116 Due 194 to the extremely large quantity of hydroxide generated as a by-product of water reduction and 195 limitations in transporting hydroxide away from the electrode, we predict the pH within the 196 catalyst layer will greater than 12, even for a 1 M KHCO3 buffered electrolyte. While the use 197 of a 0.1 M KHCO3 electrolyte can, and has, been used in an H-cell to approach the local pH 198 values reached at high current densities (Fig. 2b), these conditions are reached only when CO2 199 is largely depleted and over a small current density/potential range. It is worth noting that the 200 Fig. 3: Simplified predictions of the electrode concentration of CO2 and pH for commonly-used electrolytes as a function of current density in an (a,b) H-cell with a 50 µm CO2 diffusion thickness and, (c,d) a gas-diffusion layer with a 50 nm CO2 diffusion thickness and liquid diffusion layer of 200 µm. A Faradaic efficiency of CO2-to-CO of 90% is assumed. Fig. 3: Simplified predictions of the electrode concentration of CO2 and pH for commonly-used electrolytes as a function of current density in an (a,b) H-cell with a 50 µm CO2 diffusion thickness and, (c,d) a gas-diffusion layer with a 50 nm CO2 diffusion thickness and liquid diffusion layer of 200 µm. A Faradaic efficiency of CO2-to-CO of 90% is assumed. locally alkaline conditions could potentially be avoided by using an acidic electrolyte, but the 201 reaction kinetics for CO2 would have to outweigh the heavily-favoured hydrogen evolution 202 rate in acidic media. In any of these cases all changes in local pH will also ultimately have to 203 be paid for in the overall cell potential, regardless of the locally corrected cathode potential. In 204 brief, the results presented in Fig. 3d show a convergence towards local conditions that have 205 far-reaching implications for not only catalytic activity, but the stability and maximum 206 efficiency of an entire CO2 reduction system, as discussed in later sections. 207 From existing knowledge about CO2 reduction, we can infer that the thermodynamics and 208 kinetics of reactions on a catalytic surface will be impacted by testing under these elevated 209 reaction conditions. Further modelling and direct-measurement experimental studies of the 210 local electrolyte environment are of course warranted, however, to even better understand 211 how various factors may influence the reaction (e.g. Effect of cell configuration and reaction rate on CO2 reduction environments 116 By changing a surface’s composition, nanostructure or even substrate, the binding 228 energies of molecules to the surface and the reaction rate of different pathways can be tuned 229 to promote certain reactions, and hopefully, suppress unwanted ones. In electrochemical CO2 230 reduction, the local environment can provide a similar function. Numerous experimental 231 studies have particularly noted the tendency for higher local pH conditions to favour CO and 232 multi-carbon products while suppressing H2 and CH4 on metals such as Ag and Cu.32,36 As 233 shown in Fig. 3d, when operating at higher current densities this effect is pushed to the 234 extreme, even in highly buffered solutions. An important place to start is then to discuss the 235 effect of pH on the selectivity of both CO2 reduction products and hydrogen evolution. 236 Due to its role as the primary competing reaction to CO2 reduction, H2 evolution is one of the 237 most important products to consider. Under basic conditions the H2 evolution reaction 238 proceeds through direct water reduction and the Volmer-Tafel or Volmer-Heyrovsky steps 239 (see EIS Fig. S2). For commonly-used metals such as Cu, Au and Ag, the Volmer step is 240 particularly sluggish in basic conditions due to both poor water dissociation properties and 241 weak H binding energies, which is pushed even further to the right and away from the peak 242 of the classical volcano plot under higher pH conditions.78,79 Simultaneous CO2 reduction 243 occurring on a catalyst’s surface also tends to further suppress H2 activity by weakening 244 hydrogen binding, occupying surface sites and consuming protons.25,80 For catalysts on the 245 left side of the hydrogen binding energy volcano curve, however, hydrogen evolution under 246 locally basic conditions may increase for the same fundamental reasons. 247 Due to its role as the primary competing reaction to CO2 reduction, H2 evolution is one of the 237 most important products to consider. Under basic conditions the H2 evolution reaction 238 proceeds through direct water reduction and the Volmer-Tafel or Volmer-Heyrovsky steps 239 (see EIS Fig. S2). Effect of cell configuration and reaction rate on CO2 reduction environments 116 For commonly-used metals such as Cu, Au and Ag, the Volmer step is 240 particularly sluggish in basic conditions due to both poor water dissociation properties and 241 weak H binding energies, which is pushed even further to the right and away from the peak 242 of the classical volcano plot under higher pH conditions.78,79 Simultaneous CO2 reduction 243 occurring on a catalyst’s surface also tends to further suppress H2 activity by weakening 244 hydrogen binding, occupying surface sites and consuming protons.25,80 For catalysts on the 245 left side of the hydrogen binding energy volcano curve, however, hydrogen evolution under 246 locally basic conditions may increase for the same fundamental reasons. 247 For the most commonly-used set of materials, access to >200 mA/cm2 current densities can 248 then provide a secondary means of suppressing hydrogen evolution by weakening hydrogen 249 binding energies under higher local pH conditions. A secondary, non-material approach for 250 suppressing hydrogen is particularly important for multi-carbon product formation where, 251 despite achieving 100-fold C2:C1 ratios, many of the best catalysts when operated in an H- 252 cell are constantly dogged by a persistent 20-30% H2 selectivity.62,63,68,69 When these same 253 catalyst are operated under elevated current densities in the configurations and electrolytes 254 described here (Fig. 2 and 3), the selectivity towards target products could be increased purely 255 by penalizing hydrogen evolution, rather than necessarily promoting CO2 reduction activity. 256 From Fig. 3 one can expect that any pH-dependent suppression of H2 would be a function of 257 current density and buffering strength, with the simultaneous requirement that the 258 overpotentials needed for CO2 reduction are also competitive with H2 evolution. 259 For the most commonly-used set of materials, access to >200 mA/cm2 current densities can 248 then provide a secondary means of suppressing hydrogen evolution by weakening hydrogen 249 binding energies under higher local pH conditions. A secondary, non-material approach for 250 suppressing hydrogen is particularly important for multi-carbon product formation where, 251 despite achieving 100-fold C2:C1 ratios, many of the best catalysts when operated in an H- 252 cell are constantly dogged by a persistent 20-30% H2 selectivity.62,63,68,69 When these same 253 catalyst are operated under elevated current densities in the configurations and electrolytes 254 described here (Fig. 2 and 3), the selectivity towards target products could be increased purely 255 by penalizing hydrogen evolution, rather than necessarily promoting CO2 reduction activity. Effect of cell configuration and reaction rate on CO2 reduction environments 116 256 From Fig. 3 one can expect that any pH-dependent suppression of H2 would be a function of 257 current density and buffering strength, with the simultaneous requirement that the 258 overpotentials needed for CO2 reduction are also competitive with H2 evolution. 259 The local pH environment can also separately influence the energetics of different CO2 260 reduction products. For catalysts capable of producing only CO and H2, the suppression of H2 261 The local pH environment can also separately influence the energetics of different CO2 260 reduction products. For catalysts capable of producing only CO and H2, the suppression of H2 261 can lead to high CO selectivities. On a Ag catalyst CO formation under alkaline conditions 262 has been also been observed to be produced almost immediately following the equilibrium 263 potential of -0.11 V vs RHE58,67, indicating that the reaction itself is improved. The behaviour 264 of CO formation on Cu also differs from that of a more neutral H-cell environment. In H-cell 265 tests CO selectivities of >20% are observed on Cu and Cu-derivatives only at very low 266 current densities (<5 mA/cm2) before being supplanted by CO2 reduction to methane, 267 ethylene and ethanol. Under alkaline conditions performed in a gas-diffusion layer, high CO 268 selectivities appear over a much broader range (0-200 mA/cm2)70,73. This suggests that the 269 binding energy of CO on Cu may be weakened under alkaline conditions, promoting faster 270 desorption of the formed CO intermediate more than under neutral conditions.70 The 271 selectivity towards CO does eventually decrease in favour of higher order products similar to 272 what is observed in an H-cell, but at much higher overall current densities. 273 can lead to high CO selectivities. On a Ag catalyst CO formation under alkaline conditions 262 has been also been observed to be produced almost immediately following the equilibrium 263 potential of -0.11 V vs RHE58,67, indicating that the reaction itself is improved. The behaviour 264 of CO formation on Cu also differs from that of a more neutral H-cell environment. In H-cell 265 tests CO selectivities of >20% are observed on Cu and Cu-derivatives only at very low 266 current densities (<5 mA/cm2) before being supplanted by CO2 reduction to methane, 267 ethylene and ethanol. Under alkaline conditions performed in a gas-diffusion layer, high CO 268 selectivities appear over a much broader range (0-200 mA/cm2)70,73. Effect of cell configuration and reaction rate on CO2 reduction environments 116 This suggests that the 269 binding energy of CO on Cu may be weakened under alkaline conditions, promoting faster 270 desorption of the formed CO intermediate more than under neutral conditions.70 The 271 selectivity towards CO does eventually decrease in favour of higher order products similar to 272 what is observed in an H-cell, but at much higher overall current densities. 273 Similar to H2 evolution, methane formation on Cu has been shown to be suppressed by 274 increased local pH on oxide-derived samples. If locally alkaline (pH>12) conditions are 275 indeed unavoidable with the currently-used electrolytes, then these results suggest that an 276 alternate reaction mechanism or catalyst may be needed to realize selective methane 277 formation at elevated current densities. An interesting fundamental result would then be the 278 demonstration of a catalyst capable of selective methane formation under alkaline or locally 279 alkaline conditions. Inversely, for multi-carbon product formation on Cu, higher local pH 280 conditions have been experimentally demonstrated31,73,83 to be an important factor for 281 promoting carbon-carbon coupling. The higher observed activity toward multi-carbon species 282 at lower potentials could be due to the imporved CO onset potentials, changes to the binding 283 energy of CO, a direct effect of the local conditions on the energetics of the coupling step 284 itself, or a cumulative effect of multiple factors. There does not appear to be any strong 285 Similar to H2 evolution, methane formation on Cu has been shown to be suppressed by 274 increased local pH on oxide-derived samples. If locally alkaline (pH>12) conditions are 275 indeed unavoidable with the currently-used electrolytes, then these results suggest that an 276 alternate reaction mechanism or catalyst may be needed to realize selective methane 277 formation at elevated current densities. An interesting fundamental result would then be the 278 demonstration of a catalyst capable of selective methane formation under alkaline or locally 279 alkaline conditions. Inversely, for multi-carbon product formation on Cu, higher local pH 280 conditions have been experimentally demonstrated31,73,83 to be an important factor for 281 promoting carbon-carbon coupling. The higher observed activity toward multi-carbon species 282 at lower potentials could be due to the imporved CO onset potentials, changes to the binding 283 energy of CO, a direct effect of the local conditions on the energetics of the coupling step 284 itself, or a cumulative effect of multiple factors. Effect of cell configuration and reaction rate on CO2 reduction environments 116 There does not appear to be any strong 285 correlations between pH and product distribution after C-C coupling, however, as most 286 studies report similar alkane to alkene ratios as in lower current density H-cell experiments. 287 correlations between pH and product distribution after C-C coupling, however, as most 286 studies report similar alkane to alkene ratios as in lower current density H-cell experiments. 287 While several studies have operated at elevated current densities using membrane electrode 288 assemblies or neutral-pH catholytes such as KCl and bicarbonate-based salts,2,5,12,13,73,84 a 289 larger number of gas-diffusion layer experiments have used KOH directly as a bulk 290 catholyte,58,65–68,70,71,85 with much of the original CO2 reduction research pioneered by the 291 Kenis Group. By using an alkaline catholyte directly, the impact of a higher pH environment 292 on catalyst performance can be probed across both low and high current densities.65–68,85 In a 293 1 M KOH environment, CO2 reduction products have been observed on Cu, Au and Ag 294 catalysts at earlier overall onset potentials than in neutral conditions, with current densities of 295 >100 mA/cm2 having been reached for CO, C2H4 and ethanol at more anodic potentials 296 than -0.6 V vs. RHE and with <10% H2 selectivities.61,65,67,70,86 These experiments, however, 297 While several studies have operated at elevated current densities using membrane electrode 288 assemblies or neutral-pH catholytes such as KCl and bicarbonate-based salts,2,5,12,13,73,84 a 289 larger number of gas-diffusion layer experiments have used KOH directly as a bulk 290 catholyte,58,65–68,70,71,85 with much of the original CO2 reduction research pioneered by the 291 Kenis Group. By using an alkaline catholyte directly, the impact of a higher pH environment 292 on catalyst performance can be probed across both low and high current densities.65–68,85 In a 293 1 M KOH environment, CO2 reduction products have been observed on Cu, Au and Ag 294 catalysts at earlier overall onset potentials than in neutral conditions, with current densities of 295 >100 mA/cm2 having been reached for CO, C2H4 and ethanol at more anodic potentials 296 than -0.6 V vs. RHE and with <10% H2 selectivities.61,65,67,70,86 These experiments, however, 297 do not pay the same overpotential price associated with the large local pH swing from neutral 298 to alkaline conditions, which are not taken into account when cathode potential are reported 299 versus a reversible hydrogen electrode (RHE). Effect of cell configuration and reaction rate on CO2 reduction environments 116 This is in part due to the necessity for multi-carbon products to follow 321 from CO (Fig. S2) and C-C coupling which requires both sufficient potentials and current 322 densities.48,86,91 Testing novel catalysts at elevated current densities with less CO2 limitations 323 would allow the surface coverage of the CO reaction intermediate to be maximized over a 324 wider current and potential range without being kinetically-limited by CO2 availability. This 325 is in contrast to current H-cell environments where peak C2 selectivities are often observed 326 only at singular operating conditions (i.e. at a specific potential, current density and pH). 327 Access to a larger operating window then allows more attention to be placed on modifying 328 catalysts to specifically alter CO2 reduction product selectivity between higher-order products 329 such as alkenes vs. alcohols and C2 vs C3 products. Finally, operating under an excess CO2 330 environment reduces the overpotential losses associated with transport limitations (sometimes 331 called concentration polarizations). As a result, plots of voltage versus log(j) have been 332 observed to remain linear even up to 300 mA/cm2, helping to remove one of the barriers to 333 gauging intrinsic catalyst activity.17 334 A second important property of the local reaction environment at commercial-relevant current 311 densities is access to excess CO2. While we have imposed a current density of 200 mA/cm2 as 312 a threshold to reach, from Fig. 3c we can see that additional unreacted CO2 surrounding the 313 catalyst provides the capacity for even higher current densities. An increased CO2 314 concentration, even at elevated current densities, helps to kinetically ensure that CO2 rather 315 than protons are able to populate the reaction surface. More importantly, the reaction will not 316 be hindered by a deficit of CO2, even at higher local pH conditions. While single-carbon CO2 317 reduction products such as CO,73 HCOOH88,89 and CH490 can reach relatively high 318 selectivities even at lower current densities in an H-cell configuration, the highest Faradaic 319 efficiencies reported for multi-carbon products typically appear at current densities where 320 CO2 is almost depleted. This is in part due to the necessity for multi-carbon products to follow 321 from CO (Fig. Effect of cell configuration and reaction rate on CO2 reduction environments 116 Additionally, the interaction between 300 unreacted CO2 and hydroxide is problematic for overall stability, as described in the following 301 section. Using an alkaline electrolyte for testing and characterizing catalyst performance is 302 however a promising means of mimicking the local environment of high current densities 303 while being able to measure catalytic activity even at lower current densities. 304 A final consideration for our prediction that many electrolytes will be forced towards high 305 local pH conditions pertains to catalyst stability. Depending on the properties and composition 306 of a given catalyst, such conditions may result in the dissolution of metals into solution or the 307 removal of some species from the surface. While this has not been observed for many of the 308 Cu and Ag catalysts tested in gas-diffusion layer configurations to date, it should be a 309 consideration in the design of new catalysts. 310 A final consideration for our prediction that many electrolytes will be forced towards high 305 local pH conditions pertains to catalyst stability. Depending on the properties and composition 306 of a given catalyst, such conditions may result in the dissolution of metals into solution or the 307 removal of some species from the surface. While this has not been observed for many of the 308 Cu and Ag catalysts tested in gas-diffusion layer configurations to date, it should be a 309 consideration in the design of new catalysts. 310 A second important property of the local reaction environment at commercial-relevant current 311 densities is access to excess CO2. While we have imposed a current density of 200 mA/cm2 as 312 a threshold to reach, from Fig. 3c we can see that additional unreacted CO2 surrounding the 313 catalyst provides the capacity for even higher current densities. An increased CO2 314 concentration, even at elevated current densities, helps to kinetically ensure that CO2 rather 315 than protons are able to populate the reaction surface. More importantly, the reaction will not 316 be hindered by a deficit of CO2, even at higher local pH conditions. While single-carbon CO2 317 reduction products such as CO,73 HCOOH88,89 and CH490 can reach relatively high 318 selectivities even at lower current densities in an H-cell configuration, the highest Faradaic 319 efficiencies reported for multi-carbon products typically appear at current densities where 320 CO2 is almost depleted. Effect of cell configuration and reaction rate on CO2 reduction environments 116 S2) and C-C coupling which requires both sufficient potentials and current 322 densities.48,86,91 Testing novel catalysts at elevated current densities with less CO2 limitations 323 would allow the surface coverage of the CO reaction intermediate to be maximized over a 324 wider current and potential range without being kinetically-limited by CO2 availability. This 325 is in contrast to current H-cell environments where peak C2 selectivities are often observed 326 only at singular operating conditions (i.e. at a specific potential, current density and pH). 327 Access to a larger operating window then allows more attention to be placed on modifying 328 catalysts to specifically alter CO2 reduction product selectivity between higher-order products 329 such as alkenes vs. alcohols and C2 vs C3 products. Finally, operating under an excess CO2 330 environment reduces the overpotential losses associated with transport limitations (sometimes 331 called concentration polarizations). As a result, plots of voltage versus log(j) have been 332 observed to remain linear even up to 300 mA/cm2, helping to remove one of the barriers to 333 i i t i i t l t ti it 17 334 While the exact implications of high current density catalyst testing will vary slightly with 335 materials, we can assert that the local environment will differ greatly from the bulk electrolyte 336 with a substantial chance of changing important surface kinetics and the observed catalytic 337 activity. Tuning catalysts to optimize morphology or surface binding energies for low current 338 density characterization may then risk optimizing the catalyst for the wrong environment 339 unless higher current conditions can be appropriately mimicked. Performing such 340 experiments adds an additional degree of complexity due to the need for researchers to adopt 341 either a pressurized system or a gas-diffusion layer to supply additional CO2 (Fig. 2), in 342 addition to developing new catalysts. We believe, however, that this to be an essential step to 343 make the best (highest activity, selectivity and stability) CO2 reduction catalyst possible. 344 Fortunately, the range of experimental reports already performed at higher current densities 345 indicate that CO2 reduction is typically improved versus operating in the same electrolyte at 346 lower current densities, at least in part due to increased CO2 availability and suppression of 347 pH-independent products. Effect of cell configuration and reaction rate on CO2 reduction environments 116 We may then be resigned to the use of neutral-pH catholytes which to date 370 would represent an increase in expected cathode overpotentials. Further, overall cell potentials 371 will be higher due to the need for the oxygen evolution reaction to occur in a similar pH 372 electrolyte, or by using a bipolar membrane to maintain an alkaline anolyte. While CO2- 373 hydroxide interactions are typically only considered as a critical issue for alkaline catholytes 374 such as KOH, systems using neutral electrolytes should also aim to balance the generated 375 hydroxide ions with protons generated by the anode reaction.92 Even in a neutral-pH medium 376 current densities less than the CO2 limiting case where higher concentrations of CO2 exist in 360 solution. Our reaction-diffusion model shows that this interaction can decrease the pH within 361 the porous catalyst layer by 1-2 units at lower current densities depending on the 362 replenishment rate from a bulk liquid phase (Fig. 3d).17 While this pH change can be 363 accounted for when trying to ascertain the intrinsic activity of the catalyst, a portion of the 364 CO2 crossing the gas-liquid interface will be converted to bicarbonate upon interacting with 365 hydroxide and then carbonate.61 Not only does this decrease CO2 utilization, but over a long 366 enough operating time it will destroy the expensive KOH catholyte, itself energy-intensively 367 produced through electrochemical reactions. At the moment there is no engineering solution 368 to completely mitigate this effect even at small scales, let alone a more complex >100 cm2 369 reactor design. We may then be resigned to the use of neutral-pH catholytes which to date 370 would represent an increase in expected cathode overpotentials. Further, overall cell potentials 371 will be higher due to the need for the oxygen evolution reaction to occur in a similar pH 372 electrolyte, or by using a bipolar membrane to maintain an alkaline anolyte. While CO2- 373 hydroxide interactions are typically only considered as a critical issue for alkaline catholytes 374 such as KOH, systems using neutral electrolytes should also aim to balance the generated 375 hydroxide ions with protons generated by the anode reaction.92 Even in a neutral-pH medium 376 current densities less than the CO2 limiting case where higher concentrations of CO2 exist in 360 solution. Effect of cell configuration and reaction rate on CO2 reduction environments 116 348 Impact of high current densities on system design 349 Up to this point we have discussed the impacts of configuration and reaction conditions on the 350 CO2 reduction reaction and vice versa. While understanding the CO2 reduction reaction and 351 catalyst over a range of conditions is important, our preferred choice of catalyst and catholyte 352 will have implications and constraints for the system as a whole. Further catalyst testing in a 353 more commercial-type reactor will help to bring operational issues to the forefront of the field 354 and may result in new ingenious scientific and engineering solutions to these issues. 355 One such issue pertains to the twist of fate that the most efficient CO2 reduction conditions 356 appear to occur when a highly-alkaline electrolyte is used as a catholyte. While an alkaline 357 catholyte may provide optimal cathode performance, it comes at the cost of system stability 358 due to the interaction between unreacted CO2 and hydroxide in the electrolyte, particularly at 359 One such issue pertains to the twist of fate that the most efficient CO2 reduction conditions 356 appear to occur when a highly-alkaline electrolyte is used as a catholyte. While an alkaline 357 catholyte may provide optimal cathode performance, it comes at the cost of system stability 358 due to the interaction between unreacted CO2 and hydroxide in the electrolyte, particularly at 359 current densities less than the CO2 limiting case where higher concentrations of CO2 exist in 360 solution. Our reaction-diffusion model shows that this interaction can decrease the pH within 361 the porous catalyst layer by 1-2 units at lower current densities depending on the 362 replenishment rate from a bulk liquid phase (Fig. 3d).17 While this pH change can be 363 accounted for when trying to ascertain the intrinsic activity of the catalyst, a portion of the 364 CO2 crossing the gas-liquid interface will be converted to bicarbonate upon interacting with 365 hydroxide and then carbonate.61 Not only does this decrease CO2 utilization, but over a long 366 enough operating time it will destroy the expensive KOH catholyte, itself energy-intensively 367 produced through electrochemical reactions. At the moment there is no engineering solution 368 to completely mitigate this effect even at small scales, let alone a more complex >100 cm2 369 reactor design. Effect of cell configuration and reaction rate on CO2 reduction environments 116 379 the system’s electrolyte balance would similarly be steadily shifted away from the initial 377 condition due our CO2 reagent that can influence pH; in this case, however, the electrolyte 378 could be externally regenerated without additional energy input. 379 Another cell design issue with using alkaline electrolytes is the need for anion exchange 380 membranes, which have comparatively slower ion transport than proton exchange membranes 381 and overall limited mobility for both bicarbonate and carbonate anions.11,81 For these reasons 382 a large amount of research is being undertaken to improve OH-, HCO3- and CO32- transport 383 through anion exchange membranes, with a fair amount of work done by Dioxide Materials’ 384 and their Sustainion© membrane which has shown 1000’s of hours of stability, albeit under 385 specific operating conditions.25,82 Without sufficient bicarbonate/carbonate transport through 386 the membrane, the concentration of buffering ions will increase over time, forcing co-ion 387 transport through the membrane and resulting in electrolyte precipitation, destroying the 388 membrane and/or the gas-diffusion layer. Without solutions to these issues it will be difficult 389 for either membrane electrode assemblies or alkaline catholytes to be paired with anion 390 exchange membranes in a practical CO2 reduction device. 391 An additional reaction constraint brought on by the need for >200 mA/cm2 operation pertains 392 to the practical choice of electrolytes, independent of their impact on catalytic activity. Using 393 this minimum current density as a target threshold we can approximate the expected ohmic 394 losses of commonly-used electrolytes at 25°C, regardless of the catalyst or substrate used. 395 Assuming a combined catholyte and anolyte thickness of 3 mm, for example, it is clear that 396 certain electrolytes will cause prohibitive ohmic losses (Fig. 4). The 0.1 M KHCO3 electrolyte 397 used in the majority of the highest selectivity C2H4 studies, for instance, results in heating 398 losses of 6 V at 200 mA/cm2, five-fold larger than the thermodynamic cell potential when 399 using an oxygen-evolving anode (Fig. S3). Such a low conductivity electrolyte can then never 400 be used in a functioning system unless the electrolyte pathway between the anode and cathode 401 An additional reaction constraint brought on by the need for >200 mA/cm2 operation pertains 392 to the practical choice of electrolytes, independent of their impact on catalytic activity. Effect of cell configuration and reaction rate on CO2 reduction environments 116 Our reaction-diffusion model shows that this interaction can decrease the pH within 361 the porous catalyst layer by 1-2 units at lower current densities depending on the 362 replenishment rate from a bulk liquid phase (Fig. 3d).17 While this pH change can be 363 accounted for when trying to ascertain the intrinsic activity of the catalyst, a portion of the 364 CO2 crossing the gas-liquid interface will be converted to bicarbonate upon interacting with 365 hydroxide and then carbonate.61 Not only does this decrease CO2 utilization, but over a long 366 enough operating time it will destroy the expensive KOH catholyte, itself energy-intensively 367 produced through electrochemical reactions. At the moment there is no engineering solution 368 to completely mitigate this effect even at small scales, let alone a more complex >100 cm2 369 reactor design. We may then be resigned to the use of neutral-pH catholytes which to date 370 would represent an increase in expected cathode overpotentials. Further, overall cell potentials 371 will be higher due to the need for the oxygen evolution reaction to occur in a similar pH 372 electrolyte, or by using a bipolar membrane to maintain an alkaline anolyte. While CO2- 373 hydroxide interactions are typically only considered as a critical issue for alkaline catholytes 374 such as KOH, systems using neutral electrolytes should also aim to balance the generated 375 hydroxide ions with protons generated by the anode reaction.92 Even in a neutral-pH medium 376 Fig. 4: Expected ohmic losses as a function of current density for commonly-used electrolytes in an electrochemical cell with a combined 3 mm catholyte and anolyte thickness at 25 °C. Fig. 4: Expected ohmic losses as a function of current density for commonly-used electrolytes in an electrochemical cell with a combined 3 mm catholyte and anolyte thickness at 25 °C. the system’s electrolyte balance would similarly be steadily shifted away from the initial 377 condition due our CO2 reagent that can influence pH; in this case, however, the electrolyte 378 could be externally regenerated without additional energy input. 379 the system’s electrolyte balance would similarly be steadily shifted away from the initial 377 condition due our CO2 reagent that can influence pH; in this case, however, the electrolyte 378 could be externally regenerated without additional energy input. Effect of cell configuration and reaction rate on CO2 reduction environments 116 Using 393 this minimum current density as a target threshold we can approximate the expected ohmic 394 losses of commonly-used electrolytes at 25°C, regardless of the catalyst or substrate used. 395 Assuming a combined catholyte and anolyte thickness of 3 mm, for example, it is clear that 396 certain electrolytes will cause prohibitive ohmic losses (Fig. 4). The 0.1 M KHCO3 electrolyte 397 used in the majority of the highest selectivity C2H4 studies, for instance, results in heating 398 losses of 6 V at 200 mA/cm2, five-fold larger than the thermodynamic cell potential when 399 using an oxygen-evolving anode (Fig. S3). Such a low conductivity electrolyte can then never 400 be used in a functioning system unless the electrolyte pathway between the anode and cathode 401 is eliminated or greatly reduced, as in the case of membrane electrode assemblies (Fig. 2c).2,5 402 These ohmic drops also put into perspective how other cell losses may be more influential 403 than further decreases to cathode overpotentials. It is worth noting that the result in Fig. 4 404 becomes even more pronounced when the presumed even higher current densities needed for 405 multi-carbon products such as C2H4 are considered. Finally, separate from our discussion 406 motivating the testing of catalysts at elevated current densities, this result highlights the need 407 to test catalysts in higher conductivity electrolytes to see the effect that a higher concentration 408 of supporting ions may or may not impact a catalyst’s activity. 409 is eliminated or greatly reduced, as in the case of membrane electrode assemblies (Fig. 2c).2,5 402 These ohmic drops also put into perspective how other cell losses may be more influential 403 than further decreases to cathode overpotentials. It is worth noting that the result in Fig. 4 404 becomes even more pronounced when the presumed even higher current densities needed for 405 multi-carbon products such as C2H4 are considered. Finally, separate from our discussion 406 motivating the testing of catalysts at elevated current densities, this result highlights the need 407 to test catalysts in higher conductivity electrolytes to see the effect that a higher concentration 408 of supporting ions may or may not impact a catalyst’s activity. 409 A final notable difference to performing CO2 electroreduction in an H-cell versus a gas- 410 diffusion layer configuration is subtler. Effect of cell configuration and reaction rate on CO2 reduction environments 116 As has been recently demonstrated, under high current 411 density operation the CO2 reduction reaction is confined to a relatively small portion of a 412 porous catalyst layer and takes place on the backside of the material (Fig. 5a).70 This is in 413 contrast to an H-cell where CO2 diffuses to the catalyst layer from the front-side bulk 414 electrolyte (Fig. 5a). Much like the effects of varying morphology and porosity in H-cell 415 catalyst studies allowed for performance to be improved and varied, the reversed flow 416 direction of CO2 transport allows for unique gas-diffusion and catalyst layer engineering to 417 take place. One recent example is placing an inactive material on top of the catalyst layer 418 (Fig. 5a) to provide an additionally conductive layer. While this layer is used as a current 419 collector, it may also help to prevent both catalyst restructuring that can change catalytic 420 behaviour as well as the deposition of contaminants on the active catalyst surface. 421 Contaminants are a problem largely unique to CO2 reduction due to the tendency for most 422 electrolytes to contain minute concentrations of Ni, Fe or Co, metals capable of preferentially 423 promoting hydrogen evolution even at low potentials and are a primary source of catalyst 424 deactivation.2,95,96 One could further imagine layering catalyst structures on gas-diffusion 425 layers for either unique catalyst designs or supportive layers that provide new functionalities. 426 Fig. 5: (a) Schematic of the directional change in CO2 transport for fully-aqueous and gas-diffusion layer CO2 reduction catalysts. (b) Potential advantages of a change in the CO2 transport and reaction direction inside a catalyst layer. A buried catalyst layer may be inherently more stable and protected from both contaminants and restructuring while still having access to CO2. Fig. 5: (a) Schematic of the directional change in CO2 transport for fully-aqueous and gas-diffusion layer CO2 reduction catalysts. (b) Potential advantages of a change in the CO2 transport and reaction direction inside a catalyst layer. A buried catalyst layer may be inherently more stable and protected from both contaminants and restructuring while still having access to CO2. While many possibilities have yet to be explored for CO2 reduction, a gas-diffusion layer 427 provides a porous electrode structure which fundamentally changes both reagent and product 428 transport pathways compared to the impermeable supports traditionally used in fully-aqueous 429 reactors. Conflicts of Interest 447 There are no conflicts of interest to declare. 448 There are no conflicts of interest to declare. 448 Effect of cell configuration and reaction rate on CO2 reduction environments 116 As adoption increases substantial opportunity then exists for further innovative 430 catalyst-support integrations to improve either stability or performance. The subtle operational 431 difference also means the traditionally described benefits of catalyst nanostructuring (for 432 increased surface area, mass transport, facet exposure, etc.) have to be somewhat reassessed, 433 which may be difficult as the primary active surface is no longer easily accessible to surface 434 characterization techniques. 435 While many possibilities have yet to be explored for CO2 reduction, a gas-diffusion layer 427 provides a porous electrode structure which fundamentally changes both reagent and product 428 transport pathways compared to the impermeable supports traditionally used in fully-aqueous 429 reactors. As adoption increases substantial opportunity then exists for further innovative 430 catalyst-support integrations to improve either stability or performance. The subtle operational 431 difference also means the traditionally described benefits of catalyst nanostructuring (for 432 increased surface area, mass transport, facet exposure, etc.) have to be somewhat reassessed, 433 which may be difficult as the primary active surface is no longer easily accessible to surface 434 characterization techniques. 435 Conclusion 436 In summary, in this article we have shown that the minimum current density requirements for 437 future commercial systems will ultimately drive CO2 reduction away from the operating 438 conditions where much of the cutting-edge catalyst research has been performed. 439 Conclusion 436 In summary, in this article we have shown that the minimum current density requirements for 437 future commercial systems will ultimately drive CO2 reduction away from the operating 438 conditions where much of the cutting-edge catalyst research has been performed. 439 In summary, in this article we have shown that the minimum current density requirements for 437 future commercial systems will ultimately drive CO2 reduction away from the operating 438 conditions where much of the cutting-edge catalyst research has been performed. 439 Acknowledging these realities and testing state-of-the-art catalysts under these more realistic 440 operating conditions will be important not only for further improving performance metrics 441 such as selectivity, activity and stability, but to accelerate research towards commercial 442 applications that are exceedingly needed sooner rather than later. It is encouraging that there 443 has been a recent noticeable shift in literature towards more system-integrated testing 444 platforms for electrochemical CO2 reduction, and we hope that this new perspective further 445 motivates adoption and helps incite new discoveries. 446 Acknowledgements 449 This work was supported by contributions from the European Research Council in the form of 450 a ERC Starting Grant (WUTANG) provided to WAS. The authors greatly acknowledge the 451 stimulating research conversations with Dr. Cao-Thang Dinh, Dr. Ali Seifitokaldani, Dr. Md 452 Kibria, Dr. Recep Kas, Dr. Christine Gabardo, Dr. Nathan Nesbitt, Jonathan Edwards and 453 Divya Bohra, as well as Dr. David Sinton and Dr. Edward Sargent, that over time have 454 ll d hi i l f i i 4 This work was supported by contributions from the European Research Council in the form of 450 a ERC Starting Grant (WUTANG) provided to WAS. The authors greatly acknowledge the 451 stimulating research conversations with Dr. Cao-Thang Dinh, Dr. Ali Seifitokaldani, Dr. Md 452 Kibria, Dr. Recep Kas, Dr. Christine Gabardo, Dr. Nathan Nesbitt, Jonathan Edwards and 453 Divya Bohra, as well as Dr. David Sinton and Dr. Edward Sargent, that over time have 454 allowed this article to come to fruition. 455 allowed this article to come to fruition. 455 References 456 457 References 456 457 457 1 M. Jouny, W. Luc and F. Jiao, Ind. Eng. Chem. Res., 2018, 57, 2165–2177. 458 2 Y. C. Li, D. Zhou, Z. Yan, R. H. Gonçalves, D. A. Salvatore, C. P. Berlinguette and T. E. Mallouk, 459 ACS Energy Lett., 2016, 1, 1149–1153. 460 3 H.-R. “Molly” Jhong, F. R. Brushett and P. J. A. Kenis, Adv. Energy Mater., 2013, 3, 589–599. 461 4 T. Li, Y. Cao, J. He and C. P. Berlinguette, ACS Cent. Sci., 2017, 3, 778–783. 462 5 D. A. Salvatore, D. M. Weekes, J. He, K. E. Dettelbach, Y. C. Li, T. E. Mallouk and C. P. 463 Berlinguette, ACS Energy Lett., 2018, 3, 149–154. 464 6 T. Haas, R. Krause, R. Weber, M. Demler and G. Schmid, Nat. Catal., 2018, 1, 32–39. 465 7 S. Verma, B. Kim, H.-R. “Molly” Jhong, S. Ma and P. J. A. Kenis, ChemSusChem, 2016, 9, 1972– 466 1979. 467 8 J. M. Spurgeon and B. Kumar, Energy Environ. Sci., 2018, 11, 1536–1551. 468 9 K. Jiang, S. Siahrostami, T. Zheng, Y. Hu, S. Hwang, E. Stavitski, Y. Peng, J. Dynes, M. 469 Gangisetty, D. Su, K. Attenkofer and H. Wang, Energy Environ. Sci., , DOI:10.1039/C7EE03245E. 470 10 W. Zhang, Y. Hu, L. Ma, G. Zhu, Y. Wang, X. Xue, R. Chen, S. Yang and Z. Jin, Adv. Sci., 2018, 471 5, n/a-n/a. 472 11 F. Li, L. Chen, G. P. Knowles, D. R. MacFarlane and J. Zhang, Angew. Chem. Int. Ed., 56, 505– 473 509. 474 12 D. M. Weekes, D. A. Salvatore, A. Reyes, A. Huang and C. P. Berlinguette, Acc. Chem. Res., 2018, 475 51, 910–918. 476 13 B. Endrődi, G. Bencsik, F. Darvas, R. Jones, K. Rajeshwar and C. Janáky, Prog. En 477 Sci., 2017, 62, 133–154. 478 14 D. Kim, J. Resasco, Y. Yu, A. M. Asiri and P. Yang, Nat. Commun., 2014, 5, 4948. 479 15 X. Liu, J. Xiao, H. Peng, X. Hong, K. Chan and J. K. Nørskov, Nat. Commun., 2017, 8, 15438. 480 16 H. Mistry, R. Reske, Z. Zeng, Z.-J. Zhao, J. Greeley, P. Strasser and B. R. Cuenya, J. Am. Chem. 481 Soc., 2014, 136, 16473–16476. 482 17 R. Reske, H. Mistry, F. Behafarid, B. Roldan Cuenya and P. Strasser, J. Am. Chem. Soc., 2014, 483 136, 6978–6986. 484 18 W. Zhu, Y.-J. Zhang, H. Zhang, H. Lv, Q. Li, R. Michalsky, A. References 456 457 A. Peterson and S. Sun, J. Am. 485 Chem. Soc., 2014, 136, 16132–16135. 486 19 W. Zhu, R. Michalsky, Ö. Metin, H. Lv, S. Guo, C. J. Wright, X. Sun, A. A. Peterson and S. Sun, J. 487 Am. Chem. Soc., 2013, 135, 16833–16836. 488 20 S. Liu, X.-Z. Wang, H. Tao, T. Li, Q. Liu, Z. Xu, X.-Z. Fu and J.-L. Luo, Nano Energy, 2018, 45, 489 456–462. 490 21 S. Sen, D. Liu and G. T. R. Palmore, ACS Catal., 2014, 4, 3091–3095. 491 22 Loiudice Anna, Lobaccaro Peter, Kamali Esmail A., Thao Timothy, Huang Brandon H., Ager Joel 492 W. and Buonsanti Raffaella, Angew. Chem. Int. Ed., 2016, 55, 5789–5792. 493 23 A. S. Hall, Y. Yoon, A. Wuttig and Y. Surendranath, J. Am. Chem. Soc., 2015, 137, 14834–14837. 494 24 X. Min and M. W. Kanan, J. Am. Chem. Soc., 2015, 137, 4701–4708. 495 23 A. S. Hall, Y. Yoon, A. Wuttig and Y. Surendranath, J. Am. Chem. Soc., 2015, 137, 14834–14837. 494 49 24 X. Min and M. W. Kanan, J. Am. Chem. Soc., 2015, 137, 4701–4708. 25 E. R. Cave, C. Shi, K. P. Kuhl, T. Hatsukade, D. N. Abram, C. Hahn, K. Chan and T. F. Jaramillo, 496 ACS Catal., 2018, 3035–3040. 497 26 B. Khezri, A. C. Fisher and M. Pumera, J. Mater. Chem. A, 2017, 5, 8230–8246. 498 27 X. Duan, J. Xu, Z. Wei, J. Ma, S. Guo, S. Wang, H. Liu and S. Dou, Adv. Mater., 2017, 29, n/a-n/a. 499 28 J. Wu, Y. Huang, W. Ye and Y. Li, Adv. Sci., 2017, 4, n/a-n/a. 500 27 X. Duan, J. Xu, Z. Wei, J. Ma, S. Guo, S. Wang, H. Liu and S. Dou, Adv. Mater., 2017, 29, 28 J. Wu, Y. Huang, W. Ye and Y. Li, Adv. Sci., 2017, 4, n/a-n/a. 28 J. Wu, Y. Huang, W. Ye and Y. Li, Adv. Sci., 2017, 4, n/a-n/a. 500 29 He Jingfu, Johnson Noah J. J., Huang Aoxue and Berlinguette Curtis P., ChemSusChem, 2017, 11, 501 48–57. 502 30 F. Li, D. R. MacFarlane and J. Zhang, Nanoscale, , DOI:10.1039/C7NR09620H. 503 31 K. J. P. Schouten, E. Pérez Gallent and M. T. M. Koper, J. Electroanal. Chem., 20 504 32 Y. Pang, T. Burdyny, C.-T. Dinh, M. G. Kibria, J. Z. Fan, M. Liu, E. H. Sargent and D. Sinton, 505 Green Chem., 2017, 19, 4023–4030. References 456 457 506 33 Pander James E., Ren Dan, Huang Yun, Loo Nicholas Wei Xian, Hong Samantha Hui Lee and Yeo 507 Boon Siang, ChemElectroChem, 2017, 5, 219–237. 508 34 Y. Zhang, X. Zhang, A. M. Bond and J. Zhang, Phys. Chem. Chem. Phys., 2018, 20, 5936–5941. 509 35 C. F. C. Lim, D. A. Harrington and A. T. Marshall, Electrochimica Acta, 2017, 238, 56–63. 36 R. Kas, R. Kortlever, H. Yılmaz, M. T. M. Koper and G. Mul, ChemElectroChem, 2015, 2, 354– 511 358. 512 37 A. S. Varela, M. Kroschel, T. Reier and P. Strasser, Catal. Today, 2016, 260, 8–13. 38 J. Resasco, L. D. Chen, E. Clark, C. Tsai, C. Hahn, T. F. Jaramillo, K. Chan and A. T. Bell, J. Am. 514 Chem. Soc., 2017, 139, 11277–11287. 515 39 O. Ayemoba and A. Cuesta, ACS Appl. Mater. Interfaces, 2017, 9, 27377–27382. 516 40 E. Pérez-Gallent, G. Marcandalli, M. C. Figueiredo, F. Calle-Vallejo and M. T. M. Koper, J. Am. 517 Chem. Soc., 2017, 139, 16412–16419. 518 41 M. R. Singh, Y. Kwon, Y. Lum, J. W. Ager and A. T. Bell, J. Am. Chem. Soc., 2016, 138, 13006– 519 13012. 520 42 H. Ooka, M. C. Figueiredo and M. T. M. Koper, Langmuir, 2017, 33, 9307–9313. 521 43 J. D. Goodpaster, A. T. Bell and M. Head-Gordon, J. Phys. Chem. Lett., 2016, 7, 1471–1477. 522 44 M. R. Singh, J. D. Goodpaster, A. Z. Weber, M. Head-Gordon and A. T. Bell, Proc. Natl. Acad. 523 Sci., 2017, 114, E8812–E8821. 524 45 N. Gupta, M. Gattrell and B. MacDougall, J. Appl. Electrochem., 2006, 36, 161–172. 46 M. R. Singh, E. L. Clark and A. T. Bell, Phys. Chem. Chem. Phys., 2015, 17, 18924–18936 47 T. Cheng, H. Xiao and W. A. Goddard, Proc. Natl. Acad. Sci., 2017, 114, 1795–1800. 48 J. H. Montoya, C. Shi, K. Chan and J. K. Nørskov, J. Phys. Chem. Lett., 2015, 6, 2032–2037 528 49 Ma Ming, Trześniewski Bartek J., Xie Jie and Smith Wilson A., Angew. Chem. Int. Ed., 2016, 55, 529 9748–9752. 530 50 D. Kim, C. S. Kley, Y. Li and P. Yang, Proc. Natl. Acad. Sci., 2017, 114, 10560–10565. 51 P. D. Luna, R. Quintero-Bermudez, C.-T. Dinh, M. B. Ross, O. S. Bushuyev, P. Todorović, T. 532 Regier, S. O. Kelley, P. Yang and E. H. Sargent, Nat. Catal., 2018, 1, 103–110. 533 52 H. Mistry, A. S. Varela, C. 70 C.-T. Dinh, T. Burdyny, M. G. Kibria, A. Seifitokaldani, C. M. Gabardo, F. P. G. de Arquer, A. 565 Kiani, J. P. Edwards, P. D. Luna, O. S. Bushuyev, C. Zou, R. Quintero-Bermudez, Y. Pang, D. 566 Sinton and E. H. Sargent, Science, 2018, 360, 783–787. 567 References 456 457 S. Bonifacio, I. Zegkinoglou, I. Sinev, Y.-W. Choi, K. Kisslinger, E. A. 534 Stach, J. C. Yang, P. Strasser and B. R. Cuenya, Nat. Commun., 2016, 7, 12123. 535 53 A. D. Handoko, C. W. Ong, Y. Huang, Z. G. Lee, L. Lin, G. B. Panetti and B. S. Yeo, J. Phys. 536 Chem. C, 2016, 120, 20058–20067. 537 54 D. Ren, Y. Deng, A. D. Handoko, C. S. Chen, S. Malkhandi and B. S. Yeo, ACS Catal., 2015, 5, 538 2814–2821. 539 55 Y. Lum, B. Yue, P. Lobaccaro, A. T. Bell and J. W. Ager, J. Phys. Chem. C, 2017, 121, 14191– 540 14203. 541 56 R. Kortlever, J. Shen, K. J. P. Schouten, F. Calle-Vallejo and M. T. M. Koper, J. Phys. Chem. Lett., 542 2015, 6, 4073–4082. 543 57 T. Burdyny, P. J. Graham, Y. Pang, C.-T. Dinh, M. Liu, E. H. Sargent and D. Sinton, ACS Sustain. 544 Chem. Eng., 2017, 5, 4031–4040. 545 58 C. M. Gabardo, A. Seifitokaldani, J. P. Edwards, C.-T. Dinh, T. Burdyny, M. G. Kibria, C. P. 546 O’Brien, E. H. Sargent and D. Sinton, Energy Environ. Sci., , DOI:10.1039/C8EE01684D. 547 59 A. Seifitokaldani, C. M. Gabardo, T. Burdyny, C.-T. Dinh, J. P. Edwards, M. G. 548 Bushuyev, S. O. Kelley, D. Sinton and E. H. Sargent, J. Am. Chem. Soc., 2018, 140, 3 549 60 L.-C. Weng, A. T. Bell and A. Z. Weber, Phys. Chem. Chem. Phys., 2018, 20, 16973–16984. 550 61 C.-T. Dinh, F. P. Garcia de Arquer, D. Sinton and E. H. Sargent, ACS Energy Lett., , 551 DOI:10.1021/acsenergylett.8b01734. 552 62 R. L. Cook, R. C. MacDuff and A. F. Sammells, J. Electrochem. Soc., 1990, 137, 607–608. 553 63 H. Yano, F. Shirai, M. Nakayama and K. Ogura, J. Electroanal. Chem., 2002, 519, 93–100. 64 C. Delacourt, P. L. Ridgway, J. B. Kerr and J. Newman, J. Electrochem. Soc., 2008, 155, B42–B49. 555 65 S. Ma, R. Luo, J. I. Gold, A. Z. Yu, B. Kim and P. J. A. Kenis, J. Mater. Chem. A, 2016, 4, 8573– 556 8578. 557 66 B. Kim, F. Hillman, M. Ariyoshi, S. Fujikawa and P. J. A. Kenis, J. Power Sources, 2016, 312, 558 192–198. 559 67 S. Verma, X. Lu, S. Ma, R. I. Masel and P. J. A. Kenis, Phys. Chem. Chem. Phys., 2016, 18, 7075– 560 7084. 561 68 E. J. Dufek, T. E. y Q Sinton and E. H. Sargent, Science, 2018, 360, 783–787. References 456 457 Lister and M. E. McIlwain, Electrochem. Solid-State Lett., 2012, 15, B48– 69 T. T. H. Hoang, S. Ma, J. I. Gold, P. J. A. Kenis and A. A. Gewirth, ACS Catal., 2017, 7, 3313– 563 3321. 564 70 C.-T. Dinh, T. Burdyny, M. G. Kibria, A. Seifitokaldani, C. M. Gabardo, F. P. G. de Arquer, A. 565 Kiani, J. P. Edwards, P. D. Luna, O. S. Bushuyev, C. Zou, R. Quintero-Bermudez, Y. Pang, D. 566 Sinton and E. H. Sargent, Science, 2018, 360, 783–787. 567 71 T.-T. Zhuang, Z.-Q. Liang, A. Seifitokaldani, Y. Li, P. D. Luna, T. Burdyny, F. Che, F. Meng, Y. 568 Min, R. Quintero-Bermudez, C. T. Dinh, Y. Pang, M. Zhong, B. Zhang, J. Li, P.-N. Chen, X.-L. 569 Zheng, H. Liang, W.-N. Ge, B.-J. Ye, D. Sinton, S.-H. Yu and E. H. Sargent, Nat. Catal., 2018, 1, 570 421–428. 571 72 E. Irtem, T. Andreu, A. Parra, M. D. Hernández-Alonso, S. García-Rodríguez, J. M. Riesco-García, 572 G. Penelas-Pérez and J. R. Morante, J. Mater. Chem. A, 2016, 4, 13582–13588. 573 73 J.-J. Lv, M. Jouny, W. Luc, W. Zhu, J.-J. Zhu and F. Jiao, Adv. Mater., 0, 1803111. 574 74 Z. Liu, R. I. Masel, Q. Chen, R. Kutz, H. Yang, K. Lewinski, M. Kaplun, S. Luopa and D. R. Lutz, 575 J. CO2 Util., 2016, 15, 50–56. 576 75 W. Lee, Y. E. Kim, M. H. Youn, S. K. Jeong and K. T. Park, Angew. Chem. Int. Ed., 57, 6883– 577 6887. 578 76 P. Jeanty, C. Scherer, E. Magori, K. Wiesner-Fleischer, O. Hinrichsen and M. Fleischer, J. CO2 579 Util., 2018, 24, 454–462. 580 77 Analysis of Oxygen-Transport Diffusion Resistance in Proton-Exchange-Membrane Fuel Cells, 581 http://jes.ecsdl.org/content/158/4/B416.abstract, (accessed September 17, 2018). 582 78 W. Sheng, M. Myint, J. G. Chen and Y. Yan, Energy Environ. Sci., 2013, 6, 1509–1512. 79 J. K. Nørskov, T. Bligaard, A. Logadottir, J. R. Kitchin, J. G. Chen, S. Pandelov and U. Stimming, 584 J. Electrochem. Soc., 2005, 152, J23–J26. 585 80 Y.-J. Zhang, V. Sethuraman, R. Michalsky and A. A. Peterson, ACS Catal., 2014, 4, 3742–3 81 K. Jiang, R. B. Sandberg, A. J. Akey, X. Liu, D. C. Bell, J. K. Nørskov, K. Chan and H. Wang, Nat. 587 Catal., 2018, 1, 111–119. 588 82 D. Ren, J. Fong and B. S. Yeo, Nat. Commun., 2018, 9, 925. 589 82 D. Ren, J. Fong and B. S. 96 A. Wuttig and Y. Surendranath, ACS Catal., 2015, 5, 4479–4484. References 456 457 Yeo, Nat. Commun., 2018, 9, 925. 589 83 M. G. Kibria, C.-T. Dinh, A. Seifitokaldani, P. D. Luna, T. Burdyny, R. Quintero‐Bermudez, M. B. 590 Ross, O. S. Bushuyev, F. P. G. de Arquer, P. Yang, D. Sinton and E. H. Sargent, Adv. Mater., 0, 591 1804867. 592 84 A. D. Castillo, M. Alvarez‐Guerra and A. Irabien, AIChE J., 60, 3557–3564. 593 85 S. Ma, R. Luo, S. Moniri, Y. Lan and P. J. A. Kenis, J. Electrochem. Soc., 2014, 161, F1124– 594 F1131. 595 86 S. Ma, M. Sadakiyo, R. Luo, M. Heima, M. Yamauchi and P. J. A. Kenis, J. Power Sources, 2016, 596 301, 219–228. 597 87 Q. Li, J. Fu, W. Zhu, Z. Chen, B. Shen, L. Wu, Z. Xi, T. Wang, G. Lu, J. Zhu and S. Sun, J. Am. 598 Chem. Soc., 2017, 139, 4290–4293. 599 88 X. Zheng, P. De Luna, F. P. García de Arquer, B. Zhang, N. Becknell, M. B. Ross, Y. Li, M. N. 600 Banis, Y. Li, M. Liu, O. Voznyy, C. T. Dinh, T. Zhuang, P. Stadler, Y. Cui, X. Du, P. Yang and E. 601 H. Sargent, Joule, 2017, 1, 794–805. 602 89 A. Klinkova, P. De Luna, C.-T. Dinh, O. Voznyy, E. M. Larin, E. Kumacheva and E. H. Sargent, 603 ACS Catal., 2016, 6, 8115–8120. 604 90 H. Hashiba, H. K. Sato, S. Yotsuhashi, K. Fujii, M. Sugiyama and Y. Nakano, Sustain. Energy 605 Fuels, 2017, 1, 1734–1739. 606 91 Y. Huang, A. D. Handoko, P. Hirunsit and B. S. Yeo, ACS Catal., 2017, 7, 1749–1756. 607 92 S. Hernández, M. A. Farkhondehfal, F. Sastre, M. Makkee, G. Saracco and N. Russo, Green Chem., 608 2017, 19, 2326–2346. 609 93 J. R. Varcoe, P. Atanassov, D. R. Dekel, A. M. Herring, M. A. Hickner, P. A. Kohl, A. R. 610 Kucernak, W. E. Mustain, K. Nijmeijer, K. Scott, T. Xu and L. Zhuang, Energy Environ. Sci., 611 2014, 7, 3135–3191. 612 94 J. J. Kaczur, H. Yang, Z. Liu, S. D. Sajjad and R. I. Masel, Front. Chem., , 613 DOI:10.3389/fchem.2018.00263. 614 95 Y. Hori, H. Konishi, T. Futamura, A. Murata, O. Koga, H. Sakurai and K. Oguma, Electrochimica 615 Acta, 2005, 50, 5354–5369. 616 96 A. Wuttig and Y. Surendranath, ACS Catal., 2015, 5, 4479–4484. 617 618
https://openalex.org/W2970816008	https://europepmc.org/articles/pmc6961776?pdf=render	English	null	The Paradoxical Effect of Deep Brain Stimulation on Memory	Aging and disease	2,020	cc-by	9,041	The Paradoxical Effect of Deep Brain Stimulation on Memory ABSTRACT: Deep brain stimulation (DBS) is a promising treatment for many memory-related disorders including dementia, anxiety, and addiction. However, the use of DBS can be a paradoxical conundrum— dementia treatments aim to improve memory, whereas anxiety or addiction treatments aim to suppress maladaptive memory. In this review, the key hypotheses on how DBS affects memory are highlighted. We consolidate the findings and conclusions from the current research on the effects of DBS on memory in attempt to make sense of the bidirectional nature of DBS in disrupting and enhancing memory. Based on the current literature, we hypothesize that the timing of DBS plays a key role in its contradictory effects, and therefore, we propose a consolidated model of how DBS can both disrupt and enhance memory. Key words: memory, neuromodulation, deep brain stimulation, dementia, anxiety, addiction of the brain. This provides the benefit of high spatial and temporal specificity as compared to most other neuromodulation techniques, however the major inherent downside to this is its invasiveness. Although the nature of how DBS works is rather complex, the overarching principle is to modulate the firing of neurons in highly specific brain regions with high temporal resolution through electrical stimulation, which makes it a prime technique for altering memory systems. Much research has focussed on applying DBS to alter memory as possible treatments for learning and memory-related disorders such as Alzheimer’s disease and other dementias [4–7], anxiety-related disorders [8–13], and addiction disorders [14–16]. However, the application of DBS has created an interesting paradox in which the mode of treatments for dementias and treatments for anxiety or addiction appear to oppose each other. Treatments for dementias aim to enhance memory, whereas treatments for anxiety or addiction aim to dampen or obliterate maladaptive Memories define who we are, they affect the way we think, the way we interact with each other, and the way we interact with the world. Memory systems can be affected by diseases manifesting as memory-related disorders. For example, dementia leads to a loss of memory that eventually affects daily functioning, and diseases such as anxiety and addiction have roots in maladaptive learning and memories [1, 2]. Research related to treatments for memory-related disorders has mainly focused on modulating memories and neurological systems [3]. Deep brain stimulation (DBS) is a minimally invasive neuromodulation surgical technique. Copyright: © 2019 Tan SZK et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Correspondence should be addressed to: Dr. Lee Wei Lim, Neuromodulation Laboratory, School of Biomedical Sciences, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong SAR. China. Email: drlimleewei@gmail.com. Volume 11, Number 1; 179-190, February 2020 http://dx.doi.org/10.14336/AD.2019.0511 Review The Paradoxical Effect of Deep Brain Stimulation on Memory This method typically involves a stereotaxic surgery in which burr holes are drilled into the skull of patients/animals, and an electrode, typically made of platinum-iridium or stainless steel, is implanted in the desired region of the brain. Electrodes can then be connected to stimulators that provide electrical currents directly to the targeted location Correspondence should be addressed to: Dr. Lee Wei Lim, Neuromodulation Laboratory, School of Biomedical Sciences, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong SAR. China. Email: drlimleewei@gmail.com. Copyright: © 2019 Tan SZK et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. 179 ISSN: 2152-5250 The paradoxical effects of DBS Tan SZK., et al memories. While treatment of anxiety or addiction can be based on improving extinction memory, the somewhat contradictory notion of DBS application, with some studies showing improvement of memories [17–20], whereas others showed disruption of memories [21–26], strongly suggest that DBS can function through both means. Even though these studies might have targeted different brain regions or used different stimulation parameters, the key question remains as to how one technique can produce opposing outcomes. In this review, we examined studies on the mechanisms of DBS to come up with a hypothesis on how DBS is able to produce such contradictory effects on memory. hippocampal neurogenesis and restore corticosterone- induced suppressed neurogenesis [38–40]. Stone et al., [41] showed that entorhinal cortex DBS increased neurogenesis in the dentate gyrus (DG) of the hippocampus, which was suggested to be a mechanism of its pro-cognitive effects as observed in the long-term spatial memory as shown by the Morris water maze test. Similarly, our lab showed that medial prefrontal cortex (mPFC) DBS in middle-aged rats upregulated neurogenesis-associated genes and enhanced hippocampal cell proliferation, which was strongly correlated with enhanced memory performance in both the long- and short-term memory as shown by the novel- object recognition test and Morris water maze [7]. Forniceal DBS has also been shown to induce changes in the expression of genes related to neurogenesis [42]. Overall, there appears to be strong evidence suggesting that DBS is able to improve memory through increased neurogenesis in the hippocampus. However, other studies have suggested the effects of DBS might be independent of neurogenesis. The Paradoxical Effect of Deep Brain Stimulation on Memory For example, forniceal DBS has been shown to improve spatial memory in the Morris water maze without evidence of stimulation-induced neuro- genesis [43]. In addition, various other mechanisms of memory enhancement have been proposed (discussed later). Similar to the somewhat contradictory nature of DBS, hippocampal neurogenesis has also been linked to forgetting [44, 45]. On the surface, this might explain the dual mechanism of DBS in both disrupting and enhancing memory, but the disruption of memory by DBS appears to be almost immediate, whereas neurogenesis-related forgetting appears to occur over time indicating a long- term mechanism. Lastly, there have been some debates about the role of neurogenesis in memory, and more recently, some have argued about the presence of adult neurogenesis in humans (whereas all the above studies have been conducted in rodents) [46, 47]. Although DBS appears to be able to increase neurogenesis that might explain some of these effects, this did not satisfactorily explain all of the effects, and thereby suggesting other mechanisms are in play. Mechanisms The application of electrical stimulation in the brain, which is the precursor to DBS, has long been used and studied, with evidence of its use appearing as early as the 1900s [27]. Despite this long history, the mechanism of how DBS affects memories is still unclear [28]. Early hypotheses of the mechanisms of DBS suggested it worked by generating a temporary neural activity lesion [29]. This was partly due to similarities in its effects to ablative surgery in the treatment of Parkinson’s disease, and electrophysiological studies further showed there was a reduction in neuronal spike activity in the area of stimulation [30, 31]. However, more recent studies looking beyond the area of stimulation have suggested that the lesion hypothesis is overly simplistic, rather the mechanism of how DBS exerts its effect is more likely to be based on changes in a wider network of downstream targets [29, 32]. We highlight key prevailing theories on the mechanisms in which DBS might be able to affect memory. Neurogenesis The hippocampus plays a crucial role in the formation and retention of many types of memories, which is thought to happen through synaptic plasticity [33, 34]. Since Joseph Altman’s discovery of adult neurogenesis in the hippocampus [35], researchers have debated about its involvement in learning and memory, with one side arguing that it plays a major role and the other side arguing that it is a developmental by-product [36, 37]. Both sides of the argument have been previously written about [36] and will not be covered in this review, instead, we will focus on the possibility that DBS exerts its effects through neurogenic mechanisms. Aging and Disease • Volume 11, Number 1, February 2020 Neurotransmitters In another study, NAc DBS was shown to release dopamine in the prefrontal cortex [52]. However, these findings are controversial as it has also been reported that NAc DBS had effects that were local and not in the PFC [60]. Our lab showed a similar direct involvement of dopaminergic transmission in the ability of mPFC DBS to disrupt consolidation of memories, with ventral hippocampal (vHPC) dopamine 2 receptors playing a crucial role [57]. The study of dopamine in memory can, however, be challenging as changes in dopaminergic transmission appear to occur during memory tasks [58, 61], and paradoxical results can easily be seen due to the existence of optimal dopamine/dopamine receptor levels for certain cognitive functions [61, 62]. This paradox might be related to the contradictory nature of DBS in enhancing and disrupting memory. Interestingly, NAc DBS was shown to increase the release of the three major monoamines (dopamine, serotonin, and norepinephrine) in the PFC [52]. Similarly, we showed mPFC DBS caused changes in the transmission and metabolism of serotonin and dopamine in the vHPC [57]. The complexity of the release of monoamines may in part underlie the contradictory nature of DBS, suggesting the observed effects of DBS on memory are likely mediated through multiple neurotransmitters rather than a single type. Regardless, we have previously argued that DBS of the mPFC would be the optimal target for enhancing memory Acetylcholine, a neurotransmitter best known for its role in neuromuscular junctions, also plays a key role in learning and memory in the brain [66, 67]. More recently, acetylcholine has been implicated in the mechanism of DBS. Hescham et al., showed that forniceal stimulation was able to rescue spatial and discrimination memory in a rodent model in which impairment was induced by muscarinic acetylcholine receptor antagonist scopolamine and increase acetylcholine levels in the hippocampus, suggesting the involvement of neurogenesis in memory enhancement [5, 43, 68]. Liu et al., [69] further showed that intermittent stimulation of the nucleus basalis of Meynert in a non-human primate improved working memory, and this could be rendered ineffective by either nicotinic or muscarinic receptor antagonists, suggesting a crucial role of acetylcholine in the effects of DBS on memory enhancement. The role of acetylcholine in the hippocampus is complex. Both modelling and experimental studies have shown that acetylcholine inhibits consolidation of memory in the CA3 region [70]. Neurotransmitters There is mounting evidence that DBS exerts its effects by evoking changes in distal neural activity through axonal activation [48], and researchers have begun to study changes in neurotransmitters in various efferent targets during DBS as potential mechanisms for these effects. In this section, we will highlight three groups of neurotransmitters, namely the monoamines, acetylcholine, and the glutamate, which are suggested to be involved in the effects of DBS on memory. High frequency stimulation (HFS) of various targets related to the hippocampus have been shown to increase neurogenesis. For example, in rodent studies, DBS of the anterior thalamus has been shown to increase Aging and Disease • Volume 11, Number 1, February 2020 180 The paradoxical effects of DBS Tan SZK., et al Tan SZK., et al [7, 63], and its bidirectional connections between the hippocampus and the amygdala [64, 65] make it a prime target for disrupting memory. Further studies on the changes in monoamine transmission during mPFC DBS could prove crucial in understanding how DBS can both enhance and disrupt memories. Monoamines (e.g., dopamine, serotonin, and norepinephrine) are a group of neurotransmitters containing an amino group connected to an aromatic ring by a carbon-carbon chain. Monoamines have been strongly linked to learning and memory, as well as mood and anxiety-related disorders [49-51], making them a prime target for treatments. It has been shown that DBS has the ability to modulate the transmission of monoamines in rodent models [52]. Hamani et al., [53] showed that mPFC DBS increased serotonin levels in the hippocampus. The same group also showed that mPFC DBS, while effective in a depression model, depended on an intact serotonergic system, regardless of BDNF levels in the hippocampus [54]. Although these studies used models of depression rather than memory, the serotonergic system in the hippocampus has been shown to play a huge role in memory [55, 56], suggesting possible effects of DBS on memory through serotonergic transmission. Our group recently showed that mPFC DBS disrupted consolidation of fear memories and induced changes in serotonergic transmission in the hippocampus [57]. Similarly, changes in dopaminergic transmission have been implicated in the effects of DBS. Falowski et al., [58] showed that DBS of the nucleus accumbens (NAc) caused a decrease in tyrosine hydroxylase and dopamine in the prefrontal cortex (PFC), a structure known to be crucial for long-term memory [59]. Aging and Disease • Volume 11, Number 1, February 2020 Neurotransmitters However, the role of acetylcholine in CA1 is controversial with research showing both inhibition and activation of the Schaffer collateral pathway, and its effects were suggested to be time-dependent with cholinergic input causing either long-term potentiation (LTP) or short-term depression depending on the time of activation [71]. Similarly, the role of acetylcholine in the DG is complicated, as it has been shown to either impair or enhance LTP that depending on the individual subtypes of acetylcholine receptor [72-75]. To further complicate the matter, the effects of DBS on acetylcholine is complex; in rodents, the pool of acetylcholine in the hippocampus is limited and peaks after 20 minutes of continuous fornix stimulation before declining [68]. Overall, the complexity of acetylcholine and its various receptors could also potentially explain the contradictory nature of the effects of DBS on memory. Regardless, further studies on how DBS modulates acetylcholine and how this, in turn, affects memory will be crucial to fully understand this mechanism. Glutamate is the main excitatory neurotransmitter in the brain and plays major roles in learning and memory through both fast (ionotropic receptors) and slow modulatory mechanisms (metabotropic receptors) [76– 78]. As expected, both ionotropic and metabotropic receptors, specifically in the hippocampus, have been shown to play important roles in memory processes [79– 81]. However, the role of glutamate on the effects of DBS Aging and Disease • Volume 11, Number 1, February 2020 181 The paradoxical effects of DBS Tan SZK., et al is controversial. Multiple animal studies have shown that DBS causes local increases in glutamate [82–84], but forniceal stimulation has been shown to have beneficial effects on spatial memory without increasing glutamate levels in the hippocampus [68]. We showed that mPFC DBS disrupted consolidation of fear memory in rats and decreased glutamate levels and various glutamate receptor expressions in the vHPC [57]. Regardless, the literature on how DBS affects glutamate transmission, particularly in memory tasks, are surprisingly scarce. Overall, the complex role of glutamate in memory and the lack of understanding of how DBS affects glutamatergic systems make it difficult to elucidate the role of glutamate on the effects of DBS on memory. However, given the ability of glutamate to both enhance and disrupt memory [85, 86], its involvement in DBS would be worth exploring. memory. Electrical potentials Given the immense complexity of neurotransmitter involvement in DBS, an arguably more consolidated method of study would be to measure the overall changes in electrical potential caused by DBS, which could be seen as a summation of changes in neurotransmitters. In this regard, researchers have studied neural oscillations in relation to learning and memory. Increased gamma oscillations in the hippocampus, for example, have been shown to predict successful encoding of new verbal memories and retrieval of memories, which were distinguishable from incorrect responses [87]. Furthermore, phase synchronisation in gamma band activity has been shown to be important in encoding memory [88, 89]. Similarly, theta oscillations have been shown to be involved in memory, and synchronisation between the hippocampus and other related parts of the brain were found to be important [90], although literature on this is scarce. The question remains as to how DBS affects these waves and hence memory. In a review by Lee et al., [90] they suggested that DBS might enhance memory through mimicking the oscillatory patterns of memory. They further suggested that this same mechanism could account for impairment of memory by DBS applied at a frequency or high amplitude that would interfere with memory encoding. Suthana et al., [20] showed that entorhinal cortex DBS in human patients could reset theta oscillations and increase phase stability in the hippocampus together with enhancement of Neurotransmitters Entorhinal cortex DBS in humans has also been shown to increase theta-gamma coupling, hinting at a potential mechanism where gamma frequency DBS might function to modulate theta frequency oscillations in order to modulate memory [90]. Recently, Kim et al., [91] showed that in humans theta burst stimulation between nodes, as identified through intracranial electro- encephalography during a memory task, was able to impair retrieval of memory. They suggested that this could happen through perturbation of endogenous theta rhythm to disrupt the ongoing memory processes. Another way of studying the electrical impact of DBS on memory is to examine its effects on the engram, a hypothetical representation of the physical/biochemical storage of memories encoded in neurons in structures like the hippocampus [92, 93]. As DBS exerts effects on distal neural activity through axonal activation, it is highly likely that electrical signals from DBS directly affect neurons in the engram. We have previously suggested that DBS is able to add more information into the engram that, in turn, might be able to partially disrupt, and/or suppress, or abolish the engram/engram nodes, which could potentially be a mechanism of how DBS disrupts memories [63]. If we consider that DBS is able to increase LTP in the hippocampus [94], then this could also function to increase the synaptic weight of inputs that strengthens the engram. However, overloading LTP has also been shown to impair memory [95], which could be a mechanism for “knocking out” nodes in the engram that disrupts memory during consolidation. The outcome of the engram then depends on how and when DBS is conducted, which will be discussed in the next section. There is a wide array of neurotransmitters implicated in memory, and the effects of DBS on memory are likely to involve a complex combination of neurotransmitters. This might explain the paradoxical nature of DBS on memory and also explain the difficulty in fully understanding the mechanisms of how DBS exerts its effects. Aging and Disease • Volume 11, Number 1, February 2020 Timing We have yet to directly address how DBS is able to both enhance and disrupt memory. In this section, we will discuss the timing of DBS as a possible explanation for this paradoxical effect. Given the complexity of human memories, different genetic and environmental backgrounds, and ethical considerations in performing implantation surgery on humans, animal models could represent an alternative way to study the effects of DBS on memory. Unfortunately, animal studies on the effects of DBS on memory are limited. Among the relevant animal studies that we identified, most found DBS had beneficial effects on memory (Table 1) with the exception of two studies, one by our group that found mPFC DBS disrupted consolidation of fear memory [57] and another that found anterior thalamic DBS impaired contextual fear memory [96]. Interestingly, our group also found that mPFC DBS rescued memory impairments and Hamani et al., found that anterior thalamic DBS had beneficial effects [40, 57]. 182 The paradoxical effects of DBS Tan SZK., et al Tan SZK., et al Looking at the protocols in these studies might give us the first clue on the nature of DBS. Impairments occurred when DBS was administered during/directly after the behaviour paradigm, whereas beneficial effects occurred when DBS was administered days before the behavioural tests were conducted (though it should be noted that acute DBS before behavioural tests seem to also have a beneficial effects, albeit not as long lasting as chronic procedures [7]). Protocols from other studies also showed an emerging trend that improvements in memory occurred when the stimulation was performed before the behavioural experiments rather than during or after, with the exception of one study by Hescham et al., which found forniceal stimulation improved memory during a behavioural task [5]. However, it should be noted that different stimulation parameters were used in six consecutive sessions, and the study might not have considered cumulative effects [5]. We, therefore, hypothesize that DBS during or after a memory task disrupts either the acquisition or consolidation of memory, whereas stimulation before the memory task is beneficial due to synaptic plasticity that enhances memory. Notably, anterior thalamic DBS during but not after a behavioural task impaired memory [96], suggesting an impact on acquisition; whereas mPFC DBS after but not during a behavioural task impaired memory [57], suggesting an impact on consolidation. Timing Overall, these findings highlight the complexities of DBS on memory in which both timing and the target of stimulation play major roles in the outcome. Table 1. Non-exhaustive list of rodent studies looking at the effects of Deep Brain Stimulation on memory. Target Study Stimulation Parameters Paradigm Results Ventromedial prefrontal cortex Liu et al., 2015 [7] Single 1-h stimulation 30 mins prior to behaviour testing Morris Water Maze, Novel Object Recognition Only short-term memory improvement Daily 1-h stimulation for 4 weeks, 30 mins prior to behaviour testing Morris Water Maze, Novel Object Recognition Long-lasting benefits to memory Tan et al., 2019 [57] Single 15-min stimulation during consolidation Fear Conditioning Disruption of memory Forniceal area Sweet et al., 2010 [124] Traumatic Brain Injury (TBI) model (also non-TBI), stimulation 15 min before and during testing Delayed non-match- to-sample swim T- maze No significant difference in non-TBI animals Hescham et al., 2013 [5] 6 consecutive sessions with different parameters, 2 mins before and during behaviour testing Object Location Task Specific memory benefits in certain parameters (did not consider cumulative effects) Hao et al., 2015 [94] Rett syndrome mice, daily 1-h stimulation for 2 weeks, not stimulated during behaviour days Morris Water Maze, Contextual Fear Rescue of impaired memory Hescham et al., 2016 [43] Single 6-h stimulation, behaviour testing 30 days after stimulation Morris Water Maze Improvement in memory Entorhinal cortex Stone et al., 2011 [41] Single 30 to 120-min stimulation, behaviour testing 10 weeks after Morris Water Maze Improvement in memory Xia et al., 2017 [4] Alzheimer's mice model, single 1-h stimulation, behaviour testing 1,3,6 weeks post-stimulation Morris Water Maze, Contextual Fear Improvement later at 3 & 6 weeks but not at 1 week Anterior thalamus Hamani et al., 2010 [96] Stimulation during behaviour testing Contextual Fear Impaired memory Stimulation immediately after behaviour testing (unknown time) Contextual Fear No significant difference Hamani et al., 2011 [40] Cortisone-treated rats, single 1-h stimulation, behaviour testing 4/28 days after stimulation Non-Matching-to- Sample Rescue of impaired memory xhaustive list of rodent studies looking at the effects of Deep Brain Stimulation on memory. disruption, high-frequency optogenetic stimulation of amygdala projections to the PFC has been shown to disrupt consolidation but not acquisition of memory [97], which might hint at how DBS can disrupt memory. Some There are a few pieces of evidence that back up our hypothesis. Aging and Disease • Volume 11, Number 1, February 2020 Conclusions Studies looking at the effects of DBS on memory have been relatively scarce and have confounding and contradictory results. Although there have been some animal and human studies attempting to understand the effects of DBS on memory, little has been done to consolidate the findings. In this review, we studied the current literature on the effects of DBS on memory and proposed a consolidated model on the possible mechanisms of how DBS can both disrupt and enhance memory, which suggests time is a major factor in the bidirectional memory effects. The current treatments for anxiety/addiction disorders and dementia all have problems in their efficacy. For example, the most common treatment for both anxiety and addiction disorders is cognitive-behavioural therapy (CBT) in which maladaptive memories are targeted, but CBT has unfortunately been shown to be ineffective in the long-term with many patients relapsing [102–106]. This has led to a situation where many patients undergoing CBT are unable to maintain the benefits of the therapy Timing Compared to the study by Jacobs et al., [21] that had a larger number of patients, more independent observations, more appropriate tasks, and perhaps most importantly, used only 5-s stimulations, the study by Suthana et al., [20] used stimulation at 50 Hz (gamma wave) with variable duration depending on the task time of each patient, which suggests longer stimulation and multiple stimulations could increase LTP with related improvements. Interestingly, Kim et al., [91] showed that patients with memory impairment through theta-burst stimulation reduced theta phase coupling, which suggests the stimulation perturbed the endogenous theta phase leading to impairment of memory processes. Regardless, there appears to be strong evidence from human studies that gamma wave stimulation (high frequency) during memory tasks impairs memory processes, which supports our hypothesis stated above. In terms of improved memory, a beneficial effect of DBS elicited through changes in plasticity or neurogenesis (as discussed above) might explain why benefits are seen when stimulation is applied before a behavioural task. Given the chronic nature of mental illness and the long-term protocols of the DBS stimulations, it is likely that therapeutic benefits result from longer-term changes in plasticity [48, 98, 99]. Some human studies on DBS appear to contradict this hypothesis, as DBS during the memory task itself was observed to improve memory [18-20]. However, these improvements might appear to be related to theta waves mechanisms, as mentioned in the electrical section above. Animal studies suggest the mechanisms involve long- term memory improvements that can be achieved typically with stimulation prior to the memory task. It is, therefore possible that chronic DBS, through the chronic release of neurotransmitters, could increase neurogenesis or increase LTP benefiting memory. Furthermore, electrical mimicking of oscillatory patterns of the memory could serve to reinforce memories. More studies are needed to substantiate these claims. Aging and Disease • Volume 11, Number 1, February 2020 Timing Although optogenetic stimulation is mechanistically different from DBS, it shares the same concept of axonal activation. In terms of memory Aging and Disease • Volume 11, Number 1, February 2020 183 The paradoxical effects of DBS Tan SZK., et al various mechanisms is shown in Figure 1. To simplify the model, we represent the effects of DBS on memory through the stimulation of the mPFC, which we previously argued is an optimal target for both enhancing [7, 100] and disrupting memory [63] due to its bidirectional connections with the hippocampus [65]. and amygdala [64], both of which has implications on memories. In particular, the prelimbic and/or infralimbic regions of mPFC would be ideal targets because of the ‘limbic’ inputs [101]. In this model, short-term release of neurotransmitters together with electrical axonal activation would lead to a situation where memory processes occurring in the hippocampus might be disrupted at the point of stimulation (possibly at either acquisition or consolidation, or both) hence impairing the memory, the theoretical underpinnings of this topic can be found in our other review paper [63]. If, however, DBS is applied beforehand, ideally chronically [7], then processes involving neurotransmitter release and electrical mimicking of oscillatory patterns would increase LTP/plasticity in the hippocampus, leading to a situation where memory is enhanced in the long-term. As discussed above, neurogenesis might play a role in memory enhancement, though it is unlikely that it plays any role in memory disruption. Based on this model, the application of DBS to disrupt memory would optimally involve acute stimulation at a precise timing/stage of memory processing, whereas the application of DBS to enhance memory would optimally involve more chronic/longer applications of DBS to enhance memory systems. human studies have shown similar results in which DBS of the entorhinal region or hippocampus during encoding of memory caused a decrease in memory performance [21, 23, 25]. Conversely, some studies showed that stimulation during encoding improved memory [18–20]. Interestingly, the above studies that showed improvement in memory used theta wave stimulation [18, 19] or attributed it to theta phase resetting [20]. Key methodological differences could explain some of these findings. Consolidated model We propose a consolidated model to incorporate all the findings presented in this review. The circuitry of how DBS can both disrupt and enhance memory through its Aging and Disease • Volume 11, Number 1, February 2020 184 The paradoxical effects of DBS Tan SZK., et al [107]. Although both pharmacological treatments and optimisation of CBT protocols have been explored to address these pitfalls [108–113], issues of efficacy and safety, and problems of exacerbation persist [109, 114– 118]. Similarly, current treatments for dementia such as Alzheimer’s Disease focus on pharmacological interventions to treat symptoms rather than directly slowing down or stopping the neuronal damage [119]. However, the treatment effects tend to diminish over time [120] and there are issues with drug toxicity, and lack of significant therapeutic effects have resulted in high failure rates [121, 122]. DBS offers a potentially more effective treatment, but the unknowns and contradictory nature of some of the findings in the current studies (as discussed in this review) have made translation to the clinic incredibly difficult. [107]. Although both pharmacological treatments and optimisation of CBT protocols have been explored to address these pitfalls [108–113], issues of efficacy and safety, and problems of exacerbation persist [109, 114– 118]. Similarly, current treatments for dementia such as Alzheimer’s Disease focus on pharmacological interventions to treat symptoms rather than directly slowing down or stopping the neuronal damage [119]. However, the treatment effects tend to diminish over time Figure 1. Consolidated model on how DBS can disrupt and enhance memory. In this model, DBS is applied to the mPFC, a target previously shown to be ideal for both disruption and enhancement of memory. This results in downstream effects in the hippocampus, including effects on brainwaves, neurotransmitters, and possibly neurogenesis, leading to either disruption or enhancement of memory depending on how and when DBS is applied. Figure 1. Consolidated model on how DBS can disrupt and enhance memory. In this model, DBS is applied to the mPFC, a target previously shown to be ideal for both disruption and enhancement of memory. This results in downstream effects in the hippocampus, including effects on brainwaves, neurotransmitters, and possibly neurogenesis, leading to either disruption or enhancement of memory depending on how and when DBS is applied. Given the invasive nature of DBS, studies on humans have proven incredibly difficult. Consolidated model Animal studies allow us to study the effects of neuromodulation on memory in a more controlled environment, and furthermore, it also allows us to study the molecular mechanisms through terminal experiments. Given the relatively unknown nature of DBS on memory, elucidating how 1) the time of application, 2) length of stimulations (chronic or acute), and 3) optimal stimulation targets can affect neuromodulation in animal models would be crucial before testing in humans. Although much of the human experience cannot be modelled by animals, ethical concerns on the changes that DBS may have on agency and personality [123], which would be affected by memories, this means that we need to conduct more rigorous animal studies before its transitioning to human studies. Overall, although DBS holds immense potential in treating dementia and memory-related disorders, the contradictory nature of DBS on memory requires further study. Nevertheless, with major advancements in neuroscientific methodologies, we are optimistic in understanding the underlying mechanisms of DBS, with the ultimate goal of translation to the clinic. Aging and Disease • Volume 11, Number 1, February 2020 References [14] Luigjes J, Van Den Brink W, Feenstra M, Van Den Munckhof P, Schuurman PR, Schippers R, et al. (2012). Deep brain stimulation in addiction: A review of potential brain targets. Mol Psychiatry, 17(6):572–583. [1] Breuer J, Freud S (1893). On The Psychical Mechanism of Hysterical Phenomena : Preliminary Communication from Studies on Hysteria. Stand Ed Complet Psychol Work Sigmund Freud, Vol II Stud Hysteria:1–17. [15] [15] Müller UJ, Voges J, Steiner J, Galazky I, Heinze HJ, Möller M, et al. (2013). Deep brain stimulation of the nucleus accumbens for the treatment of addiction. Ann N Y Acad Sci, 1282(1):119–128. [2] Robbins TW, Ersche KD, Everitt BJ (2008). Drug addiction and the memory systems of the brain. Ann N Y Acad Sci, 1141:1–21. [16] [16] Pierce RC, Vassoler FM (2013). Deep brain stimulation for the treatment of addiction: Basic and clinical studies and potential mechanisms of action. Psychopharmacology (Berl), 229(3):487–491. [3] Temel Y, Hescham SA, Jahanshahi A, Janssen MLF, Tan SKH, van Overbeeke JJ, et al. (2012). Neuromodulation in Psychiatric Disorders. Int Rev Neurobiol, 107:283-314. [17] [17] Fell J, Staresina BP, Do Lam ATA, Widman G, Helmstaedter C, Elger CE, et al. (2013). Memory modulation by weak synchronous deep brain stimulation: A pilot study. Brain Stimul, 6(3):270–273. [4] [4] Xia F, Yiu A, Stone SSD, Oh S, Lozano AM, Josselyn SA, et al. (2017). Entorhinal cortical deep brain stimulation rescues memory deficits in both young and old mice genetically engineered to model Alzheimer’s disease. Neuropsychopharmacology, 42(13):2493– 2503. [18] [18] Titiz AS, Hill MRH, Mankin EA, Aghajan ZM, Eliashiv D, Tchemodanov N, et al. (2017). Theta-burst microstimulation in the human entorhinal area improves memory specificity. Elife, 6:1–18. [5] [5] Hescham S, Lim LW, Jahanshahi A, Steinbusch HWM, Prickaerts J, Blokland A, et al. (2013). Deep brain stimulation of the forniceal area enhances memory functions in experimental dementia: The role of stimulation parameters. Brain Stimul, 6(1):72–77. [19] [19] Miller JP, Sweet JA, Bailey CM, Munyon CN, Luders HO, Fastenau PS (2015). Visual-spatial memory may be enhanced with theta burst deep brain stimulation of the fornix: A preliminary investigation with four cases. Brain, 138(7):1833–1842. [6] [6] Hescham S, Lim LW, Jahanshahi A, Blokland A, Temel Y (2013). Deep brain stimulation in dementia- related disorders. Neurosci Biobehav Rev, 37(10):2666–2675. [20] [20] Suthana N, Haneef Z, Stern J, Mukamel R, Behnke E, Knowlton B, et al. (2012). Acknowledgements The scientific work was funded by the Hong Kong Research Grant Council (RGC-ECS 27104616), and The Aging and Disease • Volume 11, Number 1, February 2020 185 Tan SZK., et al The paradoxical effects of DBS University of Hong Kong URC Supplementary Funding (102009728). M, Javidnazar D, Fakhraei N, et al. (2017). Effect of deep brain stimulation of the amygdala on post- traumatic stress disorder syndrome-induced by contextual fear conditioning in rat: Role of c-Fos protein and corticosterone hormone. Brain Stimul, 10(2):348. Aging and Disease • Volume 11, Number 1, February 2020 Statement of Disclosure and Conflict of Interest The authors declare no conflict of interest or financial interest that may bias the data interpretation and presentation of this work. [13] [13] Langevin J-P, De Salles AAF, Kosoyan HP, Krahl SE (2010). Deep brain stimulation of the amygdala alleviates post-traumatic stress disorder symptoms in a rat model. J Psychiatr Res, 44(16):1241–5. [14] References Memory Enhancement and Deep-Brain Stimulation of the Entorhinal Area. N Engl J Med, 366(6):502–10. [7] [7] Liu A, Jain N, Vyas A, Lim LW (2015). Ventromedial prefrontal cortex stimulation enhances memory and hippocampal neurogenesis in the middle-aged rats. Elife, 2015(4):1–21. [21] [21] Jacobs J, Miller J, Lee SA, Coffey T, Watrous AJ, Sperling MR, et al. (2016). Direct Electrical Stimulation of the Human Entorhinal Region and Hippocampus Impairs Memory. Neuron, 92(5):983– 990. [8] [8] Reznikov R, Binko M, Nobrega JN, Hamani C (2016). Deep Brain Stimulation in Animal Models of Fear, Anxiety and Post-Traumatic Stress Disorder. Neuropsychopharmacology, 41(12):2810-2817. [22] [22] Merkow MB, Burke JF, Ramayya AG, Sharan AD, Sperling MR, Kahana MJ (2017). Stimulation of the human medial temporal lobe between learning and recall selectively enhances forgetting. Brain Stimul, 10(3):645–650. [9] [9] Rodriguez-Romaguera J, Monte FHM Do, Quirk GJ (2012). Deep brain stimulation of the ventral striatum enhances extinction of conditioned fear. Proc Natl Acad Sci, 109:8764–8769. [23] ( ) [23] Lacruz ME, Valentín A, Seoane JJG, Morris RG, Selway RP, Alarcón G (2010). Single pulse electrical stimulation of the hippocampus is sufficient to impair human episodic memory. Neuroscience, 170(2):623– 632. [10] [10] Milad MRR, Quirk GJJ (2002). Neurons in medial prefrontal cortex signal memory for fear extinction. Nature, 420(6911):70–74. [11] [11] Sui L, Huang S, Peng B, Ren J, Tian F, Wang Y (2014). Deep brain stimulation of the amygdala alleviates fear conditioning-induced alterations in synaptic plasticity in the cortical-amygdala pathway and fear memory. J Neural Transm, 121(7):773–782. [24] [24] Halgren E, Wilson CL (1985). Recall deficits produced by afterdischarges in the human hippocampal formation and amygdala. Electroencephalogr Clin Neurophysiol, 61(5):375–380. [25] [12] Coleshill SG (2004). Material-Specific Recognition Memory Deficits Elicited by Unilateral Hippocampal Coleshill SG (2004). Material-Specific Recognition Memory Deficits Elicited by Unilateral Hippocampal , ( ) [12] Mokhtari Hashtjin M, Pirzad Jahromi G, Sadr SS, Tat [25] , ( ) 12] Mokhtari Hashtjin M, Pirzad Jahromi G, Sadr SS, Tat [25] Coleshill SG (2004). Mat M D fi it Eli it d b Aging and Disease • Volume 11, Number 1, February 2020 186 Tan SZK., et al The paradoxical effects of DBS Electrical Stimulation. J Neurosci, 24(7):1612–1616. Electrical Stimulation. J Neurosci, 24(7):1612–1616. Gundry M, Hao S, et al. (2018). Forniceal deep brain stimulation induces gene expression and splicing changes that promote neurogenesis and plasticity. Elife, 7:e34031. [26] [26] Halgren E, Wilson CL, Stapleton JM (1985). References Autoradiographic and histologicalevidence of postnatal hippocampal i i J C N l 124(3) 319 335 [49] Gottfries CG (1990). Neurochemical aspects on aging and diseases with cognitive impairment. J Neurosci Res, 27(4):541–547. and memory. Nat Rev Neurosci, 5(1):45–54. [35] Altman J, Das GD (1965). Autoradiographic and histologicalevidence of postnatal hippocampal neurogenesis in rats. J Comp Neurol, 124(3):319–335. [35] Altman J, Das GD (1965). Autoradiographic and histologicalevidence of postnatal hippocampal neurogenesis in rats. J Comp Neurol, 124(3):319–335. [50] Peters R (2006). Ageing and the brain. Postgr Med J, 82:84–88. g p [36] Leuner B, Gould E, Shors TJ (2002). Is There A Link Between Adult Neurogenesis and Learning? Hippocampus, 12(5):578–584. [51] Johansen JP, Cain CK, Ostroff LE, Ledoux JE (2011). Molecular mechanisms of fear learning and memory. Cell, 147(3):509–524. [52] [37] Epp JR, Chow C, Galea LAM (2013). Hippocampus- dependent learning influences hippocampal neurogenesis. Front Neurosci, 7:57 [52] Van Dijk A, Klompmakers AA, Feenstra MGP, Denys D (2012). Deep brain stimulation of the accumbens increases dopamine, serotonin, and noradrenaline in the prefrontal cortex. J Neurochem, 123(6):897–903. [38] Brummelte S, Galea LAM (2010). Chronic high corticosterone reduces neurogenesis in the dentate gyrus of adult male and female rats. Neuroscience, 168(3):680–690. [53] Hamani C, Diwan M, Macedo CE, Brandão ML, Shumake J, Gonzalez-Lima F, et al. (2010). Antidepressant-Like Effects of Medial Prefrontal Cortex Deep Brain Stimulation in Rats. Biol Psychiatry, 67(2):117–124. [39] Toda H, Hamani C, Fawcett AP, Hutchison WD, Lozano AM (2008). The regulation of adult rodent hippocampal neurogenesis by deep brain stimulation. J Neurosurg, 108(1):132–138. [54] [54] Hamani C, MacHado DC, Hipólide DC, Dubiela FP, Suchecki D, MacEdo CE, et al. (2012). Deep brain stimulation reverses anhedonic-like behavior in a chronic model of depression: Role of serotonin and brain derived neurotrophic factor. Biol Psychiatry, 71(1):30–35. [40] g [40] Hamani C, Stone SS, Garten A, Lozano AM, Winocur G (2011). Memory rescue and enhanced neurogenesis following electrical stimulation of the anterior thalamus in rats treated with corticosterone. Exp Neurol, 232(1):100–104. [55] Meneses A (2015). Serotonin, neural markers and memory. Front Pharmacol, 6:143. [41] [41] Stone SSD, Teixeira CM, DeVito LM, Zaslavsky K, Josselyn S a., Lozano a. M, et al. (2011). Stimulation of Entorhinal Cortex Promotes Adult Neurogenesis and Facilitates Spatial Memory. J Neurosci, 31(38):13469– 13484. m [56] C re 19 [57] T Cavallaro S (2008). Genomic analysis of serotonin receptors in learning and memory. Behav Brain Res, 195(1):2–6. Aging and Disease • Volume 11, Number 1, February 2020 References Human medial temporal-lobe stimulation disrupts both formation and retrieval of recent memories. Brain Cogn, 4(3):287–295. [43] Hescham S, Temel Y, Schipper S, Lagiere M, Schönfeld L-M, Blokland A, et al. (2016). Fornix deep brain stimulation induced long-term spatial memory independent of hippocampal neurogenesis. Brain Struct Funct, 222(2):1069-1075. [27] Gildenberg PL (2005). Evolution of neuromodulation. Stereotact Funct Neurosurg, 83(2–3):71–79. [28] Kringelbach ML, Jenkinson N, Owen SLF, Aziz TZ (2007). Translational principles of deep brain stimulation. Nat Rev Neurosci, 8(8):623–635. [44] ( ) [44] Akers KG, Martinez-canabal A, Restivo L, Yiu AP, Cristofaro A De, Hsiang HL, et al. (2014). Hippocampal Neurogenesis Regulates Forgetting During Adulthood and Infancy. Science, 344(6184):598–602. [29] [29] McIntyre CC, Anderson RW (2016). Deep brain stimulation mechanisms: the control of network activity via neurochemistry modulation. J Neurochem, 139:338–345. [30] [45] [30] Benazzouz A, Gao DM, Ni ZG, Piallat B, Bouali- Benazzouz R, Benabid AL (2000). Effect of high- frequency stimulation of the subthalamic nucleus on the neuronal activities of the substantia nigra pars reticulata and ventrolateral nucleus of the thalamus in the rat. Neuroscience, 99(2):289–295. [45] Frankland PW, Kohler S, Josselyn S (2013). Hippocampal neurogenesis and forgetting. Trends Neurosci, 36(9):497–503. [46] Sorrells SF, Paredes MF, Cebrian-Silla A, Sandoval K, Qi D, Kelley KW, et al. (2018). Human hippocampal neurogenesis drops sharply in children to undetectable levels in adults. Nature, 555(7696):377–381. [31] [31] Dostrovsky JO, Levy R, Wu JP, Hutchison WD, Tasker RR, Lozano AM (2000). Microstimulation-Induced Inhibition of Neuronal Firing in Human Globus Pallidus. J Neurophysiol, 84(1):570–574. [47] [47] Boldrini M, Fulmore CA, Tartt AN, Simeon LR, Pavlova I, Poposka V, et al. (2018). Human Hippocampal Neurogenesis Persists throughout Aging. Cell StemCell, 22(5):589–599. [32] Ashkan K, Rogers P, Bergman H, Ughratdar I (2017). Insights into the mechanisms of deep brain stimulation. Nat Rev Neurol, 13(9):548–554. [48] [48] Abulseoud OA, Knight EJ, Lee KH (2012). Chapter 18: Neurotransmitter Release During Deep Brain Stimulation. Deep Brain Stimulation: A New Frontier in Psychiatry, eds Denys D, Feenstra MGP, Schuurman R (Springer), pp 193–204. [33] Martin SJ, Grimwood PD, Morris RGM (2000). Synaptic plasticity and memory: An Evaluation of the Hypothesis. Annu Rev Neurosci, 23:649–711. [49] [34] Lamprecht R, LeDoux J (2004). Structural plasticity and memory. Nat Rev Neurosci, 5(1):45–54. [34] Lamprecht R, LeDoux J (2004). Structural plasticity and memory. Nat Rev Neurosci, 5(1):45–54. [35] Altman J, Das GD (1965). References [57] Tan SZK, Poon CH, Chan Y-S, Lim LW (2019). Deep Brain Stimulation of the Ventromedial Prefrontal [42] [42] Pohodich AE, Yalamanchili H, Raman AT, Wan Y-W, Aging and Disease • Volume 11, Number 1, February 2020 187 Tan SZK., et al The paradoxical effects of DBS nicotinic acetylcholine receptor agonist TC-2559 impairs long-term potentiation in the dentate gyrus in vivo. Neurosci Lett, 406(3):183–188. Cortex Disrupts Consolidation of Fear Memories. bioRxiv, 537514. [58] [58] Falowski SM, Sharan A, Reyes BAS, Sikkema C, Szot P, Van Bockstaele EJ (2011). An evaluation of neuroplasticity and behavior after deep brain stimulation of the nucleus accumbens in an animal model of depression. Neurosurgery, 69(6):1281–90. [74] Sawada S, Ohno-shosaku T, Yamamoto C (1994). Augmenting action of nicotine on population spikes in the dentate gyrus of the guinea pig. Neurosci Res, 20:317–322. [59] [75] p g y [59] Kitamura T, Ogawa SK, Roy DS, Okuyama T, Morrissey MD, Smith LM, et al. (2017). Engrams and circuits crucial for systems consolidation of a memory. Science, 356(6333):73–78. [75] Matsuyama S, Matsumoto a, Enomoto T, Nishizaki T (2000). Activation of nicotinic acetylcholine receptors induces long-term potentiation in vivo in the intact mouse dentate gyrus. Eur J Neurosci, 12(10):3741– 3747. [60] [60] Sesia T, Bulthuis V, Tan S, Lim LW, Vlamings R, Blokland A, et al. (2010). Deep brain stimulation of the nucleus accumbens shell increases impulsive behavior and tissue levels of dopamine and serotonin. Exp Neurol, 225(2):302–309. [76] [76] Riedel G, Platt B, Micheau J (2003). Glutamate receptor function in learning and memory. Behav Brain Res, 140(1–2):1–47. [77] [77] Izquierdo I (1991). Role of NMDA receptors in memory. Trends PharmacolSci, 12(1987):128–129. [61] [61] Zbukvic IC, Park CHJ, Ganella DE, Lawrence AJ, Kim JH (2017). Prefrontal Dopaminergic Mechanisms of Extinction in Adolescence Compared to Adulthood in Rats. Front Behav Neurosci, 11:32. [78] [78] Duncan JR, Lawrence AJ (2012). The role of metabotropic glutamate receptors in addiction: Evidence from preclinical models. Pharmacol Biochem Behav, 100(4):811–824. [62] [62] Cools R, D’Esposito M (2011). Inverted-U shaped dopamine actions on human working memory and cognitive control. Biol Psychiatry, 69(12):e113–e125. [79] [79] Panaceau M, Gustafsson B (1997). NMDA receptor dependence of the input specific NMDA receptor- independent LIT in the hippocampal CA 1 region. Brain Res, 752:255–260. [63] Tan SZK, Sheng V, Chan Y-S, Lim LW (2019). Eternal Sunshine of the Neuromodulated Mind: Altering Fear Memories Through Neuromodulation. Exp Neurol, 314:9–19. References [80] [80] Tan SZK, Ganella DE, Dick ALW, Duncan JR, Ong- Palsson E, Bathgate RAD, et al. (2015). Spatial Learning Requires mGlu5 Signalling in the Dorsal Hippocampus. Neurochem Res, 40(6):1303–1310. [64] Marek R, Strobel C, Bredy TW, Sah P (2013). The amygdala and medial prefrontal cortex: partners in the fear circuit. J Physiol, 591(10):2381–91. [81] [81] Riedel G, Reymann KG (1996). Metabotropic glutamate receptors in hippocampal long-term potentiation and learning and memory. Acta Physiol Scaninavia, 157(1911):1–19. [65] Jin J, Maren S (2015). Prefrontal-Hippocampal Interactions in Memory and Emotion. Front Syst Neurosci, 9:170. [82] [66] Blokland A (1995). Acetylcholine: a neurotransmitter for learning and memory? Brain Res Rev, 21(3):285– 300. [82] Jiménez-Sánchez L, Castañé A, Pérez-Caballero L, Grifoll-Escoda M, López-Gil X, Campa L, et al. (2016). Activation of AMPA Receptors Mediates the Antidepressant Action of Deep Brain Stimulation of the Infralimbic Prefrontal Cortex. Cereb Cortex, 26(6):2778–2789. [67] Hasselmo ME (2009). The Role of Acetylcholine in Learning and Memory Michael. Curr Opin Neurobiol, 16(6):710–715. [83] [83] Agnesi F, Blaha CD, Lin J, Lee KH (2010). Local glutamate release in the rat ventral lateral thalamus evoked by high-frequency stimulation. J Neural Eng, 7(2):26009. [68] Hescham S, Jahanshahi A, Schweimer J V., Mitchell SN, Carter G, Blokland A, et al. (2015). Fornix deep brain stimulation enhances acetylcholine levels in the hippocampus. Brain Struct Funct, 221(8):1–6. [83] [84] [69] [69] Liu R, Crawford J, Callahan PM, Terry A V., Constantinidis C, Blake DT (2017). Intermittent Stimulation of the Nucleus Basalis of Meynert Improves Working Memory in Adult Monkeys. Curr Biol, 27(17):2640-2646.e4. [84] [84] Tawfik VL, Chang SY, Hitti FL, Roberts DW, Leiter JC, Jovanovic S, et al. (2010). Deep brain stimulation results in local glutamate and adenosine release: Investigation into the role of astrocytes. Neurosurgery, 67(2):367–375. [85] [70] [70] Hasselmo ME (1999). Neuromodulation: Acetylcholine and memory consolidation. Trends Cogn Sci, 3(9):351–359. [85] Staubli U, Rogers G, Lynch G (1994). Facilitation of glutamate receptors enhances memory. Proc Natl Acad Sci U S A, 91(2):777–81. [71] Haam J, Yakel JL (2017). Cholinergic modulation of the hippocampal region and memory function. J Neurochem, 142:111–121. [86] Dennis TS, Perrotti LI (2015). Erasing Drug Memories Through the Disruption of Memory Reconsolidation: A Review of Glutamatergic Mechanisms. J Appl Biobehav Res, 20(3):101–129. [72] Luo L, Chen W-H, Wang M, Zhu D-M, She J-Q, Ruan D-Y (2008). Aging and Disease • Volume 11, Number 1, February 2020 [77] [78] References Modulation of long-term potentiation by individual subtypes of muscarinic acetylcholine receptor in the rat dentate gyrus. Hippocampus, 18(10):989–995. [87] Sederberg PB, Schulze-Bonhage A, Madsen JR, Bromfield EB, McCarthy DC, Brandt A, et al. (2007). Hippocampal and neocortical gamma oscillations predict memory formation in humans. Cereb Cortex, 17(5):1190–1196. [73] ( ) [73] Wang Y, Sherwood JL, Lodge D (2006). The α4β2 [73] Wang Y, Sherwood JL, Lodge D (2006). The α4β2 Aging and Disease • Volume 11, Number 1, February 2020 188 Tan SZK., et al The paradoxical effects of DBS [88] inhibition? Why both, of course. Learn Mem, 13(2):108–109. [88] Fell J, Ludowig E, Rosburg T, Axmacher N, Elger CE (2008). Phase-locking within human mediotemporal lobe predicts memory formation. Neuroimage, 43(2):410–419. [105] [105] Bouton ME, Bolles RC (1979). Contextual control of the extinction of conditioned fear. Learn Motiv, 10(4):445–466. [89] [89] Fell J, Klaver P, Lehnertz K, Grunwald T, Schaller C, Elger CE, et al. (2001). Human memory formation is accompanied by rhinal-hippocampal coupling and decoupling. Nat Neurosci, 4(12):1259–1264. [106] Baum M (1988). Spontaneous recovery from the effects of flooding (exposure) in animals. Behav Res Ther, 26(2):185–186. [90] [107] McNally RJ (2007). Mechanisms of exposure therapy: How neuroscience can improve psychological treatments for anxiety disorders. Clin Psychol Rev, 27(6):750–759. p g [90] Lee H, Fell J, Axmacher N (2013). Electrical engram: How deep brain stimulation affects memory. Trends Cogn Sci, 17(11):574–584. [91] [91] Kim K, Schedlbauer A, Rollo M, Karunakaran S, Ekstrom AD, Tandon N (2018). Network-based brain stimulation selectively impairs spatial retrieval. Brain Stimul, 11(1):213–221. [108] de Kleine R a, Rothbaum BO, van Minnen A (2013). Pharmacological enhancement of exposure-based treatment in PTSD: A qualitative review. Eur J Psychotraumatol, 4:eCollection. [92] Ramirez S, Liu X, Lin P-A, Suh J, Pignatelli M, Redondo RL, et al. (2013). Creating a False Memory in the Hippocampus. Science, 341(6144):387–391. [109] Farach FJ, Pruitt LD, Jun JJ, Jerud AB, Zoellner LA, Roy-Byrne PP (2012). Pharmacological treatment of anxiety disorders: Current treatments and future directions. J Anxiety Disord, 26(8):833–843. [93] Ramirez S, Liu X, MacDonald CJ, Moffa A, Zhou J, Redondo RL, et al. (2015). Activating positive memory engrams suppresses depression-like behaviour. Nature, 522(7556):335–339. [110] [110] Kindt M, Soeter M, Vervliet B (2009). Beyond extinction: erasing human fear responses and preventing the return of fear. Nat Neurosci, 12(3):256– 8. [94] Hao S, Tang B, Wu Z, Ure K, Sun Y, Tao H, et al. (2015). References Forniceal deep brain stimulation rescues hippocampal memory in Rett syndrome mice. Nature, 526(7573):430–4. [111] [111] Brunet A, Orr SP, Tremblay J, Robertson K, Nader K, Pitman RK (2008). Effect of post-retrieval propranolol on psychophysiologic responding during subsequent script-driven traumatic imagery in post-traumatic stress disorder. J Psychiatr Res, 42(6):503–506. [95] Moser EI, Krober KA, Moser M-B, Morris RGM (1998). Impaired Spatial Learning after Saturation of Long-Term Potentiation. Science, 281:2038–2042. [112] Monfils M-H, Cowansage KK, Klann E, LeDoux JE (2009). Extinction-reconsolidation boundaries: key to persistent attenuation of fear memories. Science, 324(5929):951–955. [96] Hamani C, Dubiela FP, Soares JCK, Shin D, Bittencourt S, Covolan L, et al. (2010). Anterior thalamus deep brain stimulation at high current impairs memory in rats. Exp Neurol, 225(1):154–162. [113] Schiller D, Monfils M-H, Raio CM, Johnson DC, Ledoux JE, Phelps E (2010). Preventing the return of fear in humans using reconsolidation update mechanisms. Nature, 463(7277):49–53. [97] Klavir O, Prigge M, Sarel A, Paz R, Yizhar O (2017). Manipulating fear associations via optogenetic modulation of amygdala inputs to prefrontal cortex. Nat Neurosci, 20(6):836-844. [114] Torregrossa MM, Taylor JR (2013). Learning to forget: Manipulating extinction and reconsolidation processes to treat addiction. Psychopharmacology (Berl), 226(4):659–672. [98] Lujan JL, Chaturvedi A, McIntyre CC (2008). Tracking the mechanisms of deep brain stimulation for neuropsychiatric disorders. Front Biosci, 13:5892–904. [99] [99] Greenberg BD, Malone DA, Friehs GM, Rezai AR, Kubu CS, Malloy PF, et al. (2006). Three-year outcomes in deep brain stimulation for highly resistant obsessive-compulsive disorder. Neuropsychopharmacology, 31(11):2384–2393. [115] Klucken T, Kruse O, Schweckendiek J, Kuepper Y, Mueller EM, Hennig J, et al. (2016). No evidence for blocking the return of fear by disrupting reconsolidation prior to extinction learning. Cortex, 79:112–122. [100] Lim LW, Prickaerts J, Huguet G, Kadar E, Hartung H, Sharp T, et al. (2015). Electrical stimulation alleviates depressive-like behaviors of rats: investigation of brain targets and potential mechanisms. Transl Psychiatry, 5(3):e535. [116] Merlo E, Milton AL, Goozee ZY, Theobald DE, Everitt BJ (2014). Reconsolidation and extinction are dissociable and mutually exclusive processes: behavioral and molecular evidence. J Neurosci, 34(7):2422–2431. [101] Hoover WB, Vertes RP (2007). Anatomical analysis of afferent projections to the medial prefrontal cortex in the rat. Brain Struct Funct, 212(2):149–179. [117] Pedreira ME, Maldonado H (2003). Protein synthesis subserves reconsolidation or extinction depending on reminder duration. Neuron, 38(6):863–869. [102] Rescorla RA, Heth CD (1975). Reinstatement of Fear to an Extinguished Conditioned Stimulus. J Exp Psychol Anim Behav Process,104(1):88–96. Aging and Disease • Volume 11, Number 1, February 2020 [121] Cummings J (2010). What can be inferred from the interruption of the semagacestat trial for treatment of Alzheimer’s disease? Biol Psychiatry, 68(10):876–878. [124] Sweet JA, Eakin KC, Munyon CN, Miller JP (2014). Improved learning and memory with theta-burst stimulation of the fornix in rat model of traumatic brain injury. Hippocampus, 24(12):1592–1600. y y ( ) [122] Cummings JL, Morstorf T, Zhong K (2014). Alzheimer’s disease drug-development pipeline: few candidates, frequent failures. Alzheimers Res Ther, Jr R (2008). Current Pharmacologic Treatment of Dementia : A Clinical Practice Guideline from the American College of Physicians and the American. Ann Intern Med, (148):370–378. Jr R (2008). Current Pharmacologic Treatment of Dementia : A Clinical Practice Guideline from the American College of Physicians and the American. Ann Intern Med, (148):370–378. Aging and Disease • Volume 11, Number 1, February 2020 References [118] Eisenberg M, Kobilo T, Berman DE, Dudai Y (2003). Stability of Retrieved Memory: Inverse Correlation with Trace Dominance. Science, 301:1102–1104. [103] Bouton ME (2002). Context, ambiguity, and unlearning: Sources of relapse after behavioral extinction. Biol Psychiatry, 52(10):976–986. [119] Alzheimer’s Association (2015). Alzheimer’s disease facts and figures. Alzheimer’s Dement, 12(4):88. [120] Qaseem A, Snow V, Cross Jr JT, Forciea MA, Hopkins g: Sources of relapse after behavioral Biol Psychiatry, 52(10):976–986. (2006). Is extinction of fear erasure or [119] Alzheimer’s Association (2015). Alzheimer’s disease facts and figures. Alzheimer’s Dement, 12(4):88. [120] Qaseem A, Snow V, Cross Jr JT, Forciea MA, Hopkins [119] Alzheimer’s Association (2015). Alzheimer’s disease facts and figures. Alzheimer’s Dement, 12(4):88. y y ( ) [104] Barad M (2006). Is extinction of fear erasure or g [120] Qaseem A, Snow V, Cro y y [104] Barad M (2006). Is extinction of fear erasure or Aging and Disease • Volume 11, Number 1, February 2020 189 The paradoxical effects of DBS Tan SZK., et al Jr R (2008). Current Pharmacologic Treatment of Dementia : A Clinical Practice Guideline from the American College of Physicians and the American. Ann Intern Med, (148):370–378. [123] Goering S, Klein E, Dougherty DD, Widge AS (2017). Staying in the Loop: Relational Agency and Identity in Next-Generation DBS for Psychiatry. AJOB Neurosci, 8(2):59–70. [121] Cummings J (2010). What can be inferred from the interruption of the semagacestat trial for treatment of Alzheimer’s disease? Biol Psychiatry, 68(10):876–878. [124] Sweet JA, Eakin KC, Munyon CN, Miller JP (2014). Improved learning and memory with theta-burst stimulation of the fornix in rat model of traumatic brain injury. Hippocampus, 24(12):1592–1600. y y [122] Cummings JL, Morstorf T, Zhong K (2014). Alzheimer’s disease drug-development pipeline: few candidates, frequent failures. Alzheimers Res Ther, Aging and Disease • Volume 11, Number 1, February 2020 190
https://openalex.org/W3007254872	https://abpnrevista.org.br/index.php/site/article/download/835/755	Portuguese	null	Filosofia africana desde saberes ancestrais femininos: bordando perspectivas de descolonização do SER-TÃO que há em nós	Revista da ABPN	2,020	cc-by	8,732	2 Doutora em Educação (UFC); Mestra em Educação (UFBA); Bacharela e Licenciada em Filosofia (UECE). Faz parte do NACE (Núcleo das Africanidades Cearenses / UFC), AAFROCEL (Academia Afrocearense de Letras), GRIÔ: cultura popular, ancestralidade africana e educação (UFBA), ABPN (Associação Brasileira de Pesquisadores Negr@s) e Rede Brasileira de Mulheres Filósofas. E-mail: adilmachado@yahoo.com.br DOI: 10.31418/2177-2770.2020.v12.n.31.p27-47 \| ISSN 2177-2770 Licenciado sob uma Licença Creative Commons DOI: 10.31418/2177-2770.2020.v12.n.31.p27-47 \| ISSN 2177-2770 Licenciado sob uma Licença Creative Commons 1 Conhecimento, corpos, afetos, sentidos. FILOSOFIA AFRICANA DESDE SABERES ANCESTRAIS FEMININOS: BORDANDO PERSPECTIVAS DE DESCOLONIZAÇÃO DO SER-TÃO1 QUE HÁ EM NÓS Adilbênia Freire Machado2 Adilbênia Freire Machado2 Resumo: Esse artigo tem o intuito de dialogar com os saberes ancestrais femininos desde a filosofia africana com intuito de refletir acerca da descolonização do ser-tão que há em nós, tendo as filosofias da ancestralidade e do encantamento como fios que tecem esses diálogos formativos. Dialogamos desde saberes tecidos por mulheres negras que bordam experiências coletivas, irmanadas, ancestrais e encantadas desde com-partilhas de seus dons, suas vivências, experiências e saberes. Compreendendo nossos corpos como templos ancestrais, nos fortalecendo coletivamente, nos encantando e lutando diariamente pela descolonização do ser-tão desde uma relação comunitária e ancestral, delineada pela natureza e o tempo, sabendo que é fundante inter-PRETAR o cotidiano, os acontecimentos, insurgir e transformar... transfor-A-MAR! O encantamento tecido pela ancestralidade é um ato contínuo de transfor-AMAR e a filosofia africana é oriunda desse encantamento. Palavras-Chave: Filosofia Africana; Saberes Ancestrais Femininos; Transfor-AMAR; Descolonização do ser-tão. Natureza e Tempo. knowing that it is fundamental to inter-PRETAR the daily, the events, to rise and transform ... transform-A-MAR! The enchantment woven by ancestry is a continuous act of transfor-AMAR and African philosophy comes from this enchantment. knowing that it is fundamental to inter-PRETAR the daily, the events, to rise and transform ... transform-A-MAR! The enchantment woven by ancestry is a continuous act of transfor-AMAR and African philosophy comes from this enchantment. Key-words: African philosophy; Female Ancestral Knowledge; Transfor-AMAR; D l i ti f th tã N t d Ti g y transform ... transform-A-MAR! The enchantment woven by ancestry is a continuous act of transfor-AMAR and African philosophy comes from this enchantment. Key-words: African philosophy; Female Ancestral Knowledge; Transfor-AMAR; Decolonization of the ser-tão. Nature and Time. Key-words: African philosophy; Female Ancestral Knowledge; Transfor-AMAR Decolonization of the ser-tão. Nature and Time. FILOSOFÍA AFRICANA DEL CONOCIMIENTO ANCESTRAL FEMENINO: INCRUSTANDO PERSPECTIVAS DE DESCOLONIZACIÓN DEL SER-TÃO QUE HAY EN NOSOTROS Resumen: Este artículo tiene como objetivo dialogar con el conocimiento ancestral femenino de la filosofía africana para reflexionar sobre la descolonización del ser-tão, por lo que existe en nosotros, teniendo las filosofías de ascendencia y encanto como los hilos que tejen estos diálogos formativos. Dialogamos a partir del conocimiento tejido por mujeres negras que bordan experiencias colectivas, fraternales, ancestrales y encantadas, al compartir sus dones, sus experiencias, experiencias y conocimientos. Entender nuestros cuerpos como templos ancestrales, fortalecernos colectivamente, encantarnos y luchar diariamente por la descolonización del ser-tão, desde una comunidad y una relación ancestral, esbozada por la naturaleza y el tiempo, sabiendo que es fundamental inter-PRETAR lo cotidiano, los eventos, para elevarse. y transformar... transform-A-MAR! El encanto tejido por la ascendencia es un acto continuo de transformación (transfor-AMAR) e la filosofía africana proviene de este encanto. Palabras-clave: filosofía africana; Conocimiento ancestral femenino; Transfor-AMAR; Descolonización del ser-tão; Naturaleza y tiempo. AFRICAN PHILOSOPHY FROM FEMALE ANCESTRAL KNOWLEDGE: EMBEDDING PERSPECTIVES OF DECOLONIZATION OF THE SER-TÃO THAT IS IN US Abstract: This article aims to dialogue with female ancestral knowledge from African philosophy in order to reflect on the decolonization of the ser-tão that exists in us, having the philosophies of ancestry and enchantment as the threads that weave these formative dialogues. We dialogue from knowledge woven by black women who embroider collective, brotherly, ancestral and enchanted experiences from sharing their gifts, their experiences, experiences and knowledge. Understanding our bodies as ancestral temples, strengthening us collectively, enchanting us and fighting daily for the decolonization of the ser-tão from a community and ancestral relationship, delineated by nature and time, 27 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 Mots-clés: philosophie africaine; Connaissances ancestrales féminines; Transfor- AMAR; Décolonisation de Ser-tão. Nature et temps. 3 Esse artigo é alinhavado por poemas do livro “Poemas da recordação e outros movimentos” de Conceição Evaristo. Abrem e fecham, fecham ou abrem, não importa, eles dizem com poesia o que intento dizer no texto, assim, não explicarei o sentido deles em cada lugar onde se encontram, afinal, cada uma de nós é um poema na imensidão e cada uma de nós sentimos de um jeito. Sinta e leia do seu jeito. Vozes Mulheres, EVARISTO, 2017, pp. 24-25. PHILOSOPHIE AFRICAINE ISSUE DU SAVOIR ANCESTRAL FÉMININ: INTÉGRER DES PERSPECTIVES DE DÉCOLONISATION DU SER-TÃO QUI EST EN NOUS Résumé: Cet article vise à dialoguer avec les connaissances ancestrales féminines de la philosophie africaine afin de réfléchir sur la décolonisation de l'être-donc (ser-tão) qui existe en nous, en ayant les philosophies de l'ascendance et de l'enchantement comme fil conducteur qui tissent ces dialogues formatifs . Nous dialoguons à partir de connaissances tissées par des femmes noires qui brodent des expériences collectives, fraternelles, ancestrales et enchantées en partageant leurs dons, leurs expériences, leurs expériences et leurs connaissances. Comprendre nos corps comme des temples ancestraux, nous fortifier collectivement, nous enchanter et nous battre quotidiennement pour la décolonisation de l'être-donc (ser-tão) à partir d'une relation communautaire et ancestrale, définie par la nature et le temps, sachant qu'il est fondamental inter-PRETAR le quotidien, les événements, de s'élever et transformer ... transfor-A-MAR! L'enchantement tissé par l'ascendance est un acte continu de transfor-AMAR et la philosophie africaine vient de cet enchantement. Mots-clés: philosophie africaine; Connaissances ancestrales féminines; Transfor- AMAR; Décolonisation de Ser-tão. Nature et temps. 28 28 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 V O Z E S - M U L H E R E S 3 Conceição Evaristo A voz de minha bisavó ecoou criança nos porões do navio. Ecoou lamentos de uma infância perdida. A voz de minha avó ecoou obediência aos brancos-donos de tudo. A voz de minha mãe ecoou baixinho revolta no fundo das cozinhas alheias debaixo das trouxas roupagens sujas dos brancos pelo caminho empoeirado rumo à favela. A minha voz ainda ecoa versos perplexos com rimas de sangue e fome. A voz de minha filha recolhe todas as nossas vozes recolhe em si as vozes mudas caladas engasgadas nas gargantas. A voz de minha filha recolhe em si a fala e o ato. O ontem – o hoje – o agora. Na voz de minha filha se fará ouvir a ressonância O eco da vida-liberdade. A voz de minha filha recolhe em si a fala e o ato. O ontem – o hoje – o agora. Na voz de minha filha se fará ouvir a ressonância O eco da vida-liberdade. Pensar desde as filosofias africanas em terras brasileiras é dialogar / aprender / ensinar / construir / sentir / ser desde valores culturais africanos que nos constituíram, numa perspectiva interdisciplinar e transversal, ou seja, diálogo com a própria filosofia, a literatura, a sociologia, a política, a história, a geografia, a religião, a espiritualidade, as artes, os estudos sobre as mulheres, sobre raça e classe, etc., no intuito de descolonização dos saberes, desde um corpo inteiro e de maneira plural. É uma construção plural, diversa, entretanto, percebo que há uma invisibilidade das mulheres negras na “divulgação” da filosofia africana. Poucos textos chegam a nós, raramente em português, ou seja, conhecemos poucas filósofas africanas. Assim, proponho pensar a filosofia africana, em terras brasileiras, implicando- se com a descolonização do conhecimento e com uma perspectiva antissexista, desde os saberes ancestrais femininos que encantam o pensamento africano e de sua diáspora. Saberes tecidos, fundamentalmente, por mulheres negras que bordam suas experiências coletivas, irmanadas, ancestrais e encantadas, desde com-partilhas de seus dons, suas vivências, experiências e saberes. A filosofia de um modo geral é patriarcal, sexista, machista, a filosofia africana não foge a isso, portanto, é necessário forçarmos a passagem, pois, como diz Conceição Evaristo (2018, on-line, grifos meus): as mulheres negras estão mobilizadas desde sempre, mas esse protagonismo não era apontado, não era reconhecido. Se eu conto a história, por exemplo, de minha mãe, de minha tia e de outras mulheres negras que me antecederam, você vai ver que, a partir dos seus espaços de vida, essas mulheres se posicionam e sempre se posicionaram, de uma forma ou de outra. Se você for pensar na memória ancestral brasileira, por exemplo no candomblé, as grandes guardiãs foram as mulheres, as grandes mães-de-santo, as grandes cuidadoras de orixás são mulheres. Então, me parece que essa movimentação, essa atuação, essa procura de formas defensivas, de formas de resistência e também de formas de ataque, as mulheres negras construíram isso ao longo dos séculos. E hoje esse protagonismo é reconhecido através da nossa própria imposição. De um modo geral, o que nós conquistamos não foi porque a sociedade resolveu nos abrir a porta. Foi porque realmente forçamos a passagem. SOPROS INICIAIS... 3 Esse artigo é alinhavado por poemas do livro “Poemas da recordação e outros movimentos” de Conceição Evaristo. Abrem e fecham, fecham ou abrem, não importa, eles dizem com poesia o que intento dizer no texto, assim, não explicarei o sentido deles em cada lugar onde se encontram, afinal, cada uma de nós é um poema na imensidão e cada uma de nós sentimos de um jeito. Sinta e leia do seu jeito. Vozes Mulheres, EVARISTO, 2017, pp. 24-25. 29 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 As mulheres negras carregam em si o encantamento das sabedorias ancestrais, compreendem a necessidade da reinvenção da vida e de encontrar novos / outros caminhos, como nos ensina Duzu-Querença quando, pela escrevivência de Conceição Evaristo, nos diz que “era preciso reinventar a vida, encontrar novos caminhos” (EVARISTO, 2016, pp. 36-37). São mulheres guerreiras que tiveram, e continuam tendo, a “coragem de esquecer os lamentos” (LIMA, 2017, p. 23) e construir possibilidades de mundos melhores, aprender a “perder a memória da dor” (Ibid, p. 88) por meio de 30 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 linguagens diferentes, linguagens femininas, irmanadas, fundamentais para nosso fortalecimento pessoal e coletivo, comunitário, pois que somos fruto de nossa ancestralidade, de quem veio antes e da nossa comunidade, nossa subjetividade é tecida pela comunidade que se tece desde nossa subjetividade. Tessituras ancestrais de nossas mais velhas que desejam nos ajudar a “caminhar sem medo, encontrar dignidade, pisar confiante, assim como fiz com noss[a]s antepassad[a]s arrancad[a]s a força da África. [...] desacelere o passo, vai encontrar muita boniteza nessa vida, a Terra é como uma mãe, ela nos acolhe sempre” (Ibid, p. 91). li Linguagens criativas e livres, própria das pessoas que têm a terra, o chão como fonte de ser e viver, re-existir. O ser-tão que nos tece, ou seja, o conhecimento, o corpo, os afetos e os sentidos que nos tecem. Poesia da oralidade tecendo uma filosofia ancestral, cosmoencantamento. Assim, transformam cotidianamente os espaços onde pisam, de onde são, onde se fortalecem e que fortalecem. Mulheres movimento e em movimento, pois o movimento é a própria vida, é a força vital, é o axé! Movimento é o feminino tecendo a vida. Mulheres tecidas pela terra, pelos terreiros, pelo chão, pois se ligar ao chão é também sinal de humildade, de reconhecimento do respeito que devemos aos nossos antepassados e à origem das coisas. [...] nessas culturas, tradicionalmente, eram e ainda são comuns atos como os de sentar no chão, comer no chão, deitar rente ao chão (em esteira), e tirar os sapatos para pisar o chão de espaços sagrados e/ou dignos de particular respeito. (PETIT, 2015, p. 91). se ligar ao chão é também sinal de humildade, de reconhecimento do respeito que devemos aos nossos antepassados e à origem das coisas. [...] nessas culturas, tradicionalmente, eram e ainda são comuns atos como os de sentar no chão, comer no chão, deitar rente ao chão (em esteira), e tirar os sapatos para pisar o chão de espaços sagrados e/ou dignos de particular respeito. (PETIT, 2015, p. 91). Portanto, pensar a filosofia africana desde os saberes ancestrais femininos é voltar-se ao nosso chão, é aprender desde a esteira, como nos ensina Paulina Chiziane (2014, on-line) ao nos dizer que: O defeito de muitos intelectuais africanos, hoje, é depender de matérias preparadas por outras pessoas. Durante mais de 500 anos os africanos foram ensinados a renegar-se. Quando se fala de ser negro, africano, conhecer a sua própria cultura, a pessoa salta em respeito dos dogmas que vêm com igrejas, religiões e formas de conhecimento estrangeiras, que ensinam que os africanos estão nas trevas, que o curandeirismo é coisa do diabo e que um bom cristão não deve tocar nisso. Objectivo: embranquecer as vossas mentes, obrigar-vos a curvarem-se diante deles, esvaziar os vossos bolsos, já pobres de natureza. Mas uma coisa é certa: nunca haverá uma filosofia africana, apenas a partir de um livro da Europa. A intelectualidade africana começa na esteira. A esteira é o nosso chão de pertencimento, são as nossas experiências, nossos saberes, nossos valores, nossos sentidos. Histórias tecidas por bocas e ouvidos dóceis, ou seja, ouvidos que desejam ouvir e aprender com a ancestralidade, desde a ancestralidade, 31 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 e assim ouvir não apenas com os ouvidos, mas com os sentidos, de corpo inteiro. Memórias ancestrais tecidas pelo nosso viver. Memórias em movimento para descolonizar, transfor-Amar. i O pensamento africano se caracteriza pela ideia do corpo comprometido com os fenômenos da natureza. Nessa perspectiva, nos colocamos na relação com as energias da natureza do cosmo de modo a vivenciá-las também no próprio corpo. Exu Obará, Exu, rei do corpo, é o que anima, embeleza e revitaliza. Para cada conjunto de célula que morre por dia, Obará faz nascer outras tantas que nos mantêm a vida. É ele que mantém vivo nas pessoas o impulso para troca de afetos e o desejo de gozos para que jamais se acabe a vida na terra. E quando o ser de cada um exulta o prazer e a vida, Exu se move infinitamente sem a contagem inflexível do tempo que limitaria os movimentos do corpo. Exu é o que faz o jogo do universo e nele estão contidas as infinitas possibilidades como a aleatoriedade do movimento, a vagueza e a desorganização. Nele está contida também a turbulência que [o ser humano] vive como um refazer contínuo da ação e do pensamento. Encanto! Movimento! Vida... refazendo continuamente a ação e o pensamento para não nos deixar cair na tentação da morte de quem somos, de nossa memória ancestral. Sabemos que o racismo, as imagens negativizadas impostas a nós, frequentemente nos deixam “devastad[a]s pela raiva reprimida, nos sentimos exaust[a]s, desesperançad[a]s e, às vezes, simplesmente de coração partido. Essas lacunas na nossa psique são os espaços nos quais penetram a cumplicidade irrefletida, a raiva autodestrutiva, o ódio e o desespero paralisante (hooks, 2019, p. 36). Compreender nossos corpos como templos ancestrais é nos fortalecer contra o racismo e seu poder destrutivo, é compreender que o “corpo negro é criativo em sua existência espiritual” (CARNEIRO, 2006, p. 24). É encantar-se e, assim, implicar-se com o estar no mundo. Portanto, resistir é nos permitir re-existir, resistindo de formas criativas, desde uma relação comunitária e ancestral, pois, “o corpo marca e recria gestos e culturas que vêm de longe... e tem o sentido de afirmação da vida espiritual em sua espessura histórica” (Idem). ANCESTRALIDADE E ENCANTAMENTO: TESSITURAS DO TRANSFOR- AMAR Em nossos corpos encontramos memórias de nossas ancestrais, nossos corpos são memórias, a pele da cor da noite do nosso país é sua / nossa grande força. Nosso corpo é tecido pela ancestralidade que nos permite ser / estar no mundo. Esse corpo é cultural, carregado de histórias, de filosofias e sentidos. O corpo negro é tecido pela pedagogia da ancestralidade, pois “estabelece uma ruptura provocada pela decolonialidade” (KIUSAM, 2019, on-line). Isso significa que não se trata mais de falar pelo corpo, mas proporcionar situações para que o próprio corpo fale por si, alimentado pela cultura vivida na e pela carne. Esse corpo-templo que se (re) significa na e para a resistência, com o propósito de se tornar um corpo-templo-resistência – porque resistir às atrocidades também é sagrado –, acaba por estar conectado com a realidade vivida na coletividade, em seu entorno e, dessa forma, é um corpo capaz de sobreviver às intempéries sociais (Idem). não se trata mais de falar pelo corpo, mas proporcionar situações para que o próprio corpo fale por si, alimentado pela cultura vivida na e pela carne. Esse corpo-templo que se (re) significa na e para a resistência, com o propósito de se tornar um corpo-templo-resistência – porque resistir às atrocidades também é sagrado –, acaba por estar conectado com a realidade vivida na coletividade, em seu entorno e, dessa forma, é um corpo capaz de sobreviver às intempéries sociais (Idem). Somos plurais, diversas, porém, um único povo. E a relação coletiva, o comunitarismo tecido pela ancestralidade é que permite o fortalecimento da nossa existência, que permite a resistência, a re-existência, pois que “resistir às atrocidades também é sagrado”, posto sermos tecidas por um “corpo-templo-resistência” (Idem). Nosso corpo é vivo, “... é aberto para o mundo e, por isso, vulnerável a ele. O sagrado não é algo exterior ao corpo imprimindo-lhe uma negatividade, não se reduz a objetos e não é alcançado pela renúncia ao corpo e às coisas do mundo. O corpo transa e entra em transe. Relaciona-se e luta”. (CARNEIRO, 2006, pp. 28/29). Existimos de corpo inteiro desde o pensamento africano, somos parte da natureza, só existimos em relação com tudo o que existe no mundo, pois, como nos ensina Vanda Machado (2010, pp. 14/15, grifos meus): 32 32 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 A descolonização dos sentidos implica na descolonização dos afetos, dos corpos e essa descolonização é perpassada pela descolonização do pensamento, descolonização epistemológica, pois o conhecimento fora usado, e continua, para nos inferiorizar, para nos desumanizar, para nos negar enquanto seres humanos e seres humanos que produzem conhecimento, cultura, seres que amam e que merecem ser amados. A desumanização, concordando com Fanon (apud WALSH, 2009, p. 34), é “componente central da colonização” que não resiste sem o opressor e o oprimido, tirando, assim, do oprimido a capacidade de acreditar que podemos ser amadas e podemos amar. É necessário, portanto, transfor-AMAR essa perspectiva. O encantamento é a possibilidade dessa transformação, do transfor-AMAR, ele nos faz acreditar que “a vontade de amar tem representado um ato de resistência [...]. Mas ao fazer essa escolha, muit[a]s de nós descobrimos nossa incapacidade de dar e receber amor” (hooks, 2006, p. 189). Assim, diante do desencantamento imposto é fundante confiarmos em nossas ancestrais, pedir sabedoria e cultivar o amor, encantá-lo e assim responder as “dores do mundo” de forma amorosa e isso não quer dizer que estamos dando a outra face ao tapa, mas que acreditamos na potência da transformação oriunda da ancestralidade e do encantamento em nosso ser-viver. TransformAÇÃO coletiva, compreendendo que “a dor e a mágoa não expressada doem em nossas almas, e está diretamente ligada ao nosso sentido geral de seca espiritual e tumulto emocional, para não mencionar as muitas doenças que experimentamos em nossas vidas” (SOMÉ apud ROCHA, 2014, on-line). Não podemos esquecer que essa dor não é individual, “é de todo o grupo. Experimentamos uma sensação coletiva, para que uma pessoa não tenha que suportar sozinha todo o peso do sofrimento” (Idem). Assim, acredito que o futuro do nosso mundo depende muito da maneira como administramos nossa dor e tristeza. As expressões positivas da nossa dor são terapêuticas. No entanto, a falta de expressão da nossa dor ou sua incorreta gestão está na raiz da infelicidade geral e da depressão, algo que também provoca guerras e crimes. Há coisas que podemos fazer na sociedade para ajudar a curar. Podemos começar a aceitar nossa própria tristeza e sofrimento do outro. (Idem). Desse modo, ao termos nossa espiritualidade entranhada em nosso viver / ser, é importante pedirmos sabedoria às nossas ancestrais para transformarmos as agressões oriundas do racismo (assim como do epistemicídio, do sexismo, do genocídio, etc.) em potência para desconstrução do desencantamento do mundo que nos adoece, assim como de suas armadilhas, com intuito de compreendermos que: numa sociedade onde prevalece a supremacia d[a]s [pessoas] branc[a]s, a vida d[a]s [pessoas] negr[a]s é permeada por questões políticas que explicam a interiorização do racismo e de um sentimento de inferioridade. Esses sistemas de dominação são mais eficazes quando alteram nossa habilidade de querer e amar. Nós negr[a]s temos sido profundamente ferid[a]s, como a gente diz, “ferid[a]s até o coração”, e essa ferida emocional que carregamos afeta nossa capacidade de sentir e consequentemente, de amar. Somos um povo ferido. Feridos naquele lugar que poderia conhecer o amor, que estaria amando (HOOKS, 2006, pp. 188/189). numa sociedade onde prevalece a supremacia d[a]s [pessoas] branc[a]s, a vida d[a]s [pessoas] negr[a]s é permeada por questões políticas que explicam a interiorização do racismo e de um sentimento de inferioridade. Esses sistemas de dominação são mais eficazes quando alteram nossa habilidade de querer e amar. Nós negr[a]s temos sido profundamente ferid[a]s, como a gente diz, “ferid[a]s até o coração”, e essa ferida emocional que carregamos afeta nossa capacidade de sentir e consequentemente, de amar. Somos um povo ferido. Feridos naquele lugar que poderia conhecer o amor, que estaria amando (HOOKS, 2006, pp. 188/189). 33 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 compreender a necessidade da escuta sensível, da cura oriunda do compartilhar dores, anseios, medos, traumas... Acredito que lutar contra o racismo, o epistemicídio, o patriarcado, o sexismo, a xenofobia, é uma tarefa difícil, porém, sem amor, sem sensibilidade, sem empatia, sem escuta sensível (MACHADO, 2019), sem encantamento não é possível, assim como não será possível a destruição dessas “dores do mundo”. E tudo isso não implica ausência de conflitos, mas aprender com eles, usá-los para potencializar as existências, para desconstruir e transformar e não para destruir. Transfor- AMAR. Não venceremos o racismo, o patriarcado, etc., usando as armas coloniais, as armas patriarcais, venceremos utilizando nossas armas e capacidades de reinvenções, ressignificações, reconstruções, re-existências pautadas pela ancestralidade, pelo encantamento tecido pela força vital que nos tece, pelo templo que somos nós. A cura é comunitária! Nossa ancestral encantada Makota Valdina Pinto (2018, fonte oral) nos ensina que “uma pessoa, não é só uma pessoa, aquela pessoa tem história, de ancestralidade, de tudo que vem até chegar naquela pessoa”. Somos uma comunidade inteira. Somos em comunidade. Uma comunidade ancestral: Não sou eu, não é você, é algo que transcende a mim, a você. Eles estão aqui presente, essa minha ancestralidade que eu tenho preservada está aqui comigo. Mas, está aqui (no espaço), você não vê, eu não vejo, não toca, mas eles estão presente hoje, da mesma maneira que há séculos atrás. Eu acredito! (Idem). Um dos maiores propósitos da Lei 10.639 / 2003, assim como das Relações Étnico-Raciais, da Filosofia Africana, Afrodiaspórica, Afrorreferenciada, é fortalecer nosso pertencimento, mostrar nossa história e nossa origem e nos dar alegria (lida como potência para a vida) de dizer quem somos e de onde viemos, pois “aqui, aprendemos a não saber o que somos e, sobretudo, o que devemos querer ser. Temos sido ensinados a usar a miscigenação ou a mestiçagem como carta de alforria do estigma da negritude” (CARNEIRO, 2011, p. 64). Sabemos que “uma das características do racismo é a maneira pela qual ele aprisiona o outro em imagens fixas e estereotipadas” (Ibid, p. 70), assim, nos é imposto “dúvidas metafísicas que assombram a racialidade no Brasil” (Ibid, p. 71). Reconhecer nosso pertencimento negro é, também, um ato político! Encantamento Ancestral! Não é uma atitude impensada, pois o encantamento não nos permite, é uma atitude para a vida: implicação ética, estética, política, social, cultural, amorosa. Consciência racial! Estética do Bem-viver! Engajamento! Aceitar a dor, compreendê-la nos leva à empatia, respeito por nós e por todas as pessoas, reconhecimento de nossas dores, das dores das outras pessoas, das dores coletivas, inclusive das dores que perpassam o tempo, as dores de nossas ancestrais arrancadas de suas terras, suas famílias, seus países, seus pertencimentos. Aceitar a dor é 34 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 nossas origens e do encantamento por ela. O encantamento por nossa ancestralidade africana nos leva a seguirmos numa luta engajada por nosso direito à vida, à existência em sua totalidade. Desse modo, é necessário não perdermos de vista a construção da consciência política, social, cultural, amorosa, estética, filosófica, de pertencimento, também para a construção de uma sociedade democrática, onde o bem-viver marque nossas existências, nos permitindo viver plenamente nossa cidadania e assim o respeito às diferenças aparece como primordial. Portanto, compreendemos que a descolonização perpassa a escuta sensível, a percepção do todo, o cuidado a cada segundo com o que o colonizador (e o patriarcado construído por eles) e o racismo entranham em nosso modo de ser e estar no mundo. Assim, afirma-se que reconhecer nosso pertencimento negro africano fortalece, potencializa e multiplica os sentidos de nossa existência em toda sua potência, atingindo outras pessoas, sabendo que a “luta contra o racismo e as desigualdades raciais, assim como a afirmação da identidade negra são processos complexos, desafiadores e que precisam ser desenvolvidos de forma enfática, persistente e contundente” (GOMES, 2018, p. 113). Por isso, é fundante existirmos em nossas pesquisas, falarmos desde nós mesmas, desde nossas experiências, fragilidades, dúvidas, curiosidades, lutas, conquistas, pois o racismo, a colonialidade afeta profundamente nossas subjetividades, desde o ser- tão que nos tece. A afirmação de si é afirmação de todas as nossas ancestrais... de quem já veio, de quem aqui está e de quem virá. Afinal, nossos passos vêm de longe. Desse modo, temos responsabilidade em contribuir para os processos de descolonização, assim, para encarar essas feridas, para curá-las, as pessoas negras progressistas e nossos aliados nessa luta devem estar comprometidas em realizar os esforços de intervir criticamente no mundo das imagens e transformá-lo, conferindo uma posição de destaque em nossos movimentos políticos de libertação e autodefinição. [...] a questão da raça e da representação não se restringe apenas a criticar o status quo. É também uma questão de transformar as imagens, criar alternativas, questionar quais tipos de imagens subverter, apresentar alternativas críticas e transformar, nossas visões de mundo e nos afastar de pensamentos dualistas acerca do bom e do mau. A Lei 10.639 / 2003 e o ensino para as relações étnico-raciais, tecidas pelas filosofias africanas, estão implicadas em provocar mudanças desde o re-conhecimento de 35 35 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 35 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 4 EVARISTO, 2017, pp. 119-120. Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 É fundamental unirmos nossas forças para nos curarmos, potencializarmos cada vez mais nossos modos de ser e estar no mundo. Nos encantarmos! Mesmo diante de tantas dores, negações, agressões, nossa ancestralidade não deixa o encantamento acabar, mas é necessário ouvirmos essa ancestralidade que há nós... E em mim, minhas ancestrais têm transformado o desencantamento do mundo em potência para viver, ser, criar, encantar, encantar-se... Sem encantamento eu não sou, portanto, insisto, resisto, busco potencializar o feminino insurgente que há em mim! O feminino que há em nós é o que permite o transfor-A-MAR ser resistência tecida por nossa espiritualidade, por nossa ancestralidade. A p e s a r d a s a c o n t e c ê n c i a s d o b a n z o Conceição Evaristo4 [...] a descolonização como um processo político é sempre uma luta para nos definir internamente, e que vai além do ato de resistência à dominação, estamos sempre no processo de recordar o passado, mesmo enquanto criamos novas formas de imaginar e construir o futuro. (hooks, 2019, pp. 36-37). Portanto, é fundante trabalharmos metodologias, currículos, referenciais afro para que possamos despertar ancestralidades adormecidas, encantamentos silenciados, calados... Desse modo, faz-se necessário que nossas escritas, nossas pesquisas, nossas 36 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 leituras estejam impregnados de vida, de axé, de luta, nossas teorias gritam, choram, dançam, flertam, seduzem, cantam e encantam... somos de corpo inteiro, nosso corpo fala, nossos textos são escritas de nossos corpos, nossas pesquisas são tecidas por nossas experiências e saberes... Escrevivências. l i A p e s a r d a s a c o n t e c ê n c i a s d o b a n z o Conceição Evaristo4 Apesar das acontecências do banzo há de nos restar a crença na precisão de viver e a sapiente leitura das entre-falhas da linha-vida Apesar das acontecências do banzo há de nos restar a crença na precisão de viver e a sapiente leitura das entre-falhas da linha-vida Apesar de... uma fé há de nos afiançar de que, mesmo estando nós entre rochas, não haverá pedra a nos entupir o caminho. Das acontecências do banzo A pesar sobre nós, Há de nos aprumar a coragem. Murros em ponta de faca (valem) afiam os nossos desejos neutralizando o corte da lâmina. Das acontecências do banzo brotará em nós o abraço à vida e seguiremos nossas rotas de sal e mel por entre Salmos, Axés e Aleluias 37 37 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 TESSITURAS ANCESTRAIS FEMININAS: NATUREZA E TEMPO Desde o pensamento africano, afrorreferenciado, existimos de corpo inteiro, somos parte da natureza, só existimos em relação com o que existe no mundo, assim, só é possível pensar de corpo inteiro, o cognitivo e sensorial, razão e emoção. Pluralidades de sentidos das nossas vivências, das nossas escrevivências, das nossas ações pedagógicas, das pretagogias, dos referenciais teóricos e metodológicos afrorreferenciados, das filosofias africanas, das filosofias da ancestralidade e do encantamento... Pois, como nos diz Vanda Machado (2013, p. 52): o pensamento africano não separa, não hierarquiza. Corpo, membro, memória, tradição, sentidos, imaginário, símbolos, signos, espiritualidade e as vivências cotidianas, tudo faz parte de uma tradição na sua multidimensionalidade que não se presta a explicação reduzida, a categoria que fragmenta sentido. As filosofias africanas, da ancestralidade e do encantamento são tecidas, bordadas por seres vivos em todas as suas dimensões, seres integrais, engajadas com uma liberdade concreta, com justiça. Corpo, mente e espírito! Somos seres de sentidos, sensações, percepções, afetos! Vanda Machado mostra que o fazer / saber é compreendido na experiência, no afetar-se, no desenho tecido desde o conhecimento, a vida e a poiésis de encantamento... somos conhecimento oriundo da arte de viver, do viver a arte, do celebrar a vida. Pois, A herança cultural, o conjunto de saberes, o mito, o canto, a dança, os provérbios, as diversas narrativas vivenciadas ampliam a percepção que ajuda a compreender a vida em sua interdependência como um enredo que permite dar significados a todos os acontecimentos do mundo em todos os tempos. Este é o sentido que traspassa da história para a solidariedade (MACHADO, 2013, p. 43). Por isso é fundante escrever, contar, filosofar desde nossas histórias. Escrevivências, pois cada palavra é cheia de sentidos, “é um jeito de encantar, a palavra não é só um jeito de se comunicar, a palavra é encanto, tem vida, tem uma liga”, como Vanda Machado me ensinou em uma com-versa na sua cozinha, em sua casa localizada em Salvador, em 27 de Setembro de 2017. É com a palavra que se educa, que se aprende, que se expressa, que se diz, que se faz. A ancestralidade nos ensina que “tudo está no presente. Todo ensinamento pela história está no presente para ser entregue em forma de vivências” (MACHADO, 2013, p. 66), ou seja, todo ensinamento é processo formativo. Isso implica que a ancestralidade se atualiza continuamente na própria vivência, na 38 Revista da ABPN • v. TESSITURAS ANCESTRAIS FEMININAS: NATUREZA E TEMPO 12, n. 31 • jan – fev 2020, p. 27-47 experiência, na formação pessoal que se faz no coletivo. É o saber, viver, experienciar, fazer. Sandra Petit (2015, p. 123) nos diz que: iê na tradição oral africana existe uma relação íntima entra palavra e o fazer, como explica Vansina (1982, p. 157): “quase em toda parte, a palavra tem um poder misterioso, pois palavras criam coisas. Isso, pelo menos, é o que prevalece na maioria das civilizações africanas. Os Dogon sem dúvida expressaram esse nominalismo da forma mais evidente; nos rituais constatamos em toda parte que o nome é a coisa, e que ‘dizer’ é ‘fazer’”. Nessa perspectiva do dizer é fazer, do viver como ação contínua do pensamento, percebemos a relação outra de nossa existência, de nosso ser desde um tempo outro e uma percepção outra da natureza. Sermos de corpo inteiro é ser numa relação intrínseca com a natureza e o tempo tecido pela ancestralidade. A natureza é o que nos borda, o que nos permite ser, viver. Sem ela não somos, não criamos. Ela é a própria vida, movimento da existência. A natureza é a ancestralidade que habita em nós e que nos permite ser, como Makota Valdina Pinto ensina: Ancestralidade pra mim é tudo o que veio antes de mim. Então, a natureza é a minha ancestralidade. [...] E a natureza não foi o [ser humano] que fez, [ser humano] veio depois que toda a natureza foi criada pra dar boa intenção de vida pro [ser humano]. Então, a minha ancestralidade é toda a natureza que foi criada pela primeira semente viva que iniciou esse mundo. [...] Minha ancestralidade é a natureza. (PINTO, 2017, fonte oral). A natureza é fonte para nosso caminhar, nosso viver. Para nós que estamos encruzilhadas com o pensamento africano, afrorreferenciado, a terra é o nosso alimento, é fonte. Na relação íntima com a natureza podemos “aprender/partilhar/ensinar através de experiências condizentes com as referências de matriz africana e as expressões da cultura na diáspora”, nos diz Kellynia Alves (2015, p. 38). Desse modo, possibilita-se “uma proposta educativa que seja transmitida pelo corpo, pela ancestralidade, pela cultura, pela estética africana e pela tradição oral” (Idem). Dona Toinha, mulher negra, quilombola, liderança da comunidade Água Preta (Tururu – CE), diz que a natureza: É tudo. Primeiro Deus, segundo a natureza. A natureza é a ancestralidade em nós, o feminino criando, possibilitando a vida. A natureza representa os saberes ancestrais femininos! Nossos corpos expressam o tempo da natureza, pois o tempo ancestral é o tempo da natureza. Makota Valdina Pinto (2010, fonte oral), nos ensina que: O tempo para a cultura banto, a cultura do Congo, é muito profundo, porque a formação de tudo, eles imaginam que muito tempo passou. O tempo teve um tempo que não foi o nosso tempo, o tempo do humano, que é a ancestralidade. Quando a gente fala, usando a linguagem mais conhecida, que orixá é ancestral, o Inquice é ancestral, Vodun é ancestral, é porque eles vieram antes. [...] Porque a natureza veio antes do ser humano, quando surgiu o protótipo do ser humano. O tempo é o espaço onde somos, de onde viemos, tempo é chão. O tempo é a natureza que nos tece, nos borda. O tempo é o antes, o agora, o depois: A gente é resultado de toda essa natureza criada antes e que a gente encontrou. E esse tempo é um tempo que a gente não dá conta, a gente especula, a gente cria, as várias culturas criam histórias, se cria lendas para falar desse tempo, mas ninguém tava lá para saber, então, tudo é história, criada. Mas, é um tempo que é hoje também. Ai é que tá, que é o x da questão. Esse tempo ancestral é o tempo de hoje também, se a gente vive porque a gente precisa até do vento, que é o tempo, que é tempo, não o tempo, mas que é tempo... Tempo é vento, vento é tempo. Quem é que não precisa do vento? Quem é que vive sem respirar? Ninguém... (Idem). Não existimos fora da natureza, não somos sem o tempo. Somos espaços e potência de vida. O tempo conduz nossas andanças, encruzilhando encontros / encantos. O tempo é como uma mágica, é a relação com o mistério como disse, em uma com-versa na cozinha, Dona Toinha (17 de Agosto de 2019): “Tem coisas que só o tempo para explicar. Sem palavras. Mas a gente acaba entendendo”. TESSITURAS ANCESTRAIS FEMININAS: NATUREZA E TEMPO [...] Natureza é vida mesmo, [...] é tudo o que a gente precisa para viver, sem a natureza a gente não consegue... tem o ar que respiramos, tem a água que vem da terra, as plantas, as plantas medicinais que a gente usa para fazer remédio. [...] tudo meu era remédio caseiro [...] e eu aconselho as pessoas também a usarem remédios caseiros. (2010, fonte oral) 39 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 o temor e a coragem, se expunha ao incerto vaivém da vida. o temor e a coragem, se expunha ao incerto vaivém da vida. A mulher mirou-se no espelho de suas águas: A mulher mirou-se no espelho de suas águas: p - dos pingos lágrimas à plenitude da vazante. E no fluxo e refluxo de seu eu viu o tempo se render. Viu os dias gastos em momentos renovados d’esperança nascitura. Viu seu ventre eterno grávido, Salpicado de mil estrias, (só as contáveis estrelas) em revitalizado brilho. p - dos pingos lágrimas à plenitude da vazante. E no fluxo e refluxo de seu eu viu o tempo se render. Viu os dias gastos em momentos renovados d’esperança nascitura. Viu seu ventre eterno grávido, Salpicado de mil estrias, (só as contáveis estrelas) em revitalizado brilho. - dos pingos lágrimas E viu nos infindos filetes de sua pele desenhos-louvores nasciam do tempo de todas as eras em que a voz-mulher na rouquidão de seu silêncio de tanto gritar acordou o tempo no tempo. E viu nos infindos filetes de sua pele desenhos-louvores nasciam do tempo de todas as eras em que a voz-mulher na rouquidão de seu silêncio de tanto gritar acordou o tempo no tempo. E só, só ela, a mulher, alisou as rugas dos dias e sapiente adivinhou: não, o tempo não lhe fugiu entre os dedos, ele se guardou de uma mulher a outra... E só, não mais só, recolheu o só da outra, da outra, da outra... fazendo solidificar uma rede de infinitas jovens linhas cosidas por mãos ancestrais e rejubilou-se com o tempo guardado no templo de seu eternizado corpo. N a m u l h e r , o t e m p o . . . Conceição Evaristo5 A mulher mirou-se no espelho do tempo, mil rugas (só as visíveis) sorriram, perpendiculares às linhas das dores. Amadurecidos sulcos atravessavam o opaco e o fulgor de seus olhos em que a íris, entre A mulher mirou-se no espelho do tempo, mil rugas (só as visíveis) sorriram, perpendiculares às linhas das dores. Amadurecidos sulcos atravessavam o opaco e o fulgor de seus olhos em que a íris, entre 5 EVARISTO, 2017, pp. 38-40. 40 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 DESCOLONIZAÇÃO DO SER-TÃO QUE HÁ EM NÓS: IN-CONCLUSÕES A reinvenção da vida tecida pela ancestralidade e potencializada pelo encantamento é delineada pela ética do cuidado, pois é necessário cuidar de si mesma e 41 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 de quem está a nossa volta, é o cuidar comunitário, alimentar o ntu, ou seja, nossa força vital que só é no coletivo. Esse cuidado é perpassado pela natureza e pelo tempo ancestral. d A vida desde a cosmopercepção africana é delineada pelo coletivo, onde “o ato de educar passa pela experiência de preparar a construção de outra geração e a construção de cada um[a] em particular” (MACHADO, 2013, p. 22). Um viver coletivo onde a comunidade na qual estamos inseridas é o lugar que alimenta nossa existência, é “o espírito, a luz-guia [...], é onde as pessoas se reúnem para realizar um objetivo específico, para ajudar os outros a realizarem seu propósito e para cuidar uma das outras” (SOMÉ, 2003, p. 35), é onde “as pessoas vão compartilhar seus dons e recebem as dádivas dos outros” (Idem). Fonte de força, alimento do espírito, das nossas energias, pois “...enquanto nos preparamos para a nossa reexistência negra, somos confortados pelo pulsar do coração da mãe comunidade” (MACHADO, 2013, p.115). Desse modo, descolonizar o ser-tão que há em nós é falar desde os sentidos, as energias que tecem nosso existir, nossas escrevivências, compreendendo o conhecimento como gestado e parido por nossas experiências, trajetórias, histórias, vivências, assim, é necessário mergulhar em nossos corações, nossas emoções e ouvir seu ritmo (SOMÉ, 2003) para nos fortalecer, para descolonizar o ser-tão. É sabido que a força das mulheres africanas, das mulheres negras, perpassa o tempo e o espaço, uma força presente no cotidiano, no nosso falar, dançar, cantar, ouvir, fazer, em nosso paladar, nas religiões de matriz africana, no nosso modo de acolher, de ser. Para a moçambicana Paulina Chiziane (2016, p. 08), a mulher é comparada à terra “porque lá é o centro da vida. Da mulher emana a força mágica da criação. Ela é abrigo no período da gestação. É alimento no princípio de todas as vidas. Ela é prazer, calor, conforto de todos os seres humanos na superfície da terra”. A mulher negra enraizou o “ensinamento da ética do cuidado” (CARNEIRO, 2006, p. 35) em nosso solo. É assim “a luta da mulher negra para participar, ter voz e vez, enfim, exercer em toda a plenitude o seu direito à cidadania, sem deixar de se dizer e se ver conforme suas tradições” (THEODORO, 1996, p. 132), seu jeito de ser / estar / sentir o mundo, compartilhando sua potência de ser-tão. Potência feminina! Sueli Carneiro (2018, p. 112) diz: “compartilhar é um verbo que as mulheres conjugam em maior escala do que os homens... Tempo ancestral feminino, de potência criativa: “Aprendemos a administrar a escassez e como Cristo temos multiplicado o pão em nossas mesas. [...] Com isso aprendemos sobre solidariedade e fraternidade. Contribuições que temos a dar a um Tempo Feminino”. (Idem). Nos reeducamos, reeducamos pessoas negras e brancas, assim como de outros pertencimentos étnico-raciais (GOMES, 2018). Somos resultado de vários elementos, de uma interação constante com a natureza, com os seres viventes e não viventes, com o tempo ancestral, portanto, “compreender o pensamento africano passa pela necessidade de apreensão de outras realidades. O ser humano não foi construído de um único elemento da natureza. A construção foi de um ser síntese do mundo, síntese de elementos cósmicos”. (MACHADO, 2013, p. 101). Somos parte do todo e o todo de uma parte, estamos interligadas, somos parte de uma grande teia! O encontro com a filosofia africana, e assim com as filosofias da ancestralidade e do encantamento, fortalece diariamente a compreensão de que estamos em constante aprendizado e somos eternas aprendizes, pois o conhecimento é fruto dos acontecimentos e esses dos movimentos e não há vida sem movimento. É importante inter-PRETAR o cotidiano, o que está à nossa volta, os discursos, os contextos, os acontecimentos e ver as possibilidades de insurgir e transformar... transfor-A-MAR! Transfor-A-MAR é um ato de resistência e espiritualidade! Em tempos tão desumanos, amar é um ato de resistência, de re-existência. Amar como potência para a vida, empatia, amor ancestral, encantado, implicado com o bem-viver, com as pessoas, com a humanidade, com a natureza! Somos a continuidade de uma “força majestosa”, como afrodescendentes recebemos “uma herança de feminilidade que, desde o início, nunca foi definida pelos atributos subjetivos de pureza, fragilidade e vulnerabilidade (ou seja, dependência), mas sempre pela autodeterminação orientada para a ação” (BANKOLE, 2009, p. 274). Somos frutos de um legado de força e “essa história persistente preenche volumes e está contida 43 Revista da ABPN • v. 12, n. DESCOLONIZAÇÃO DO SER-TÃO QUE HÁ EM NÓS: IN-CONCLUSÕES Entretanto, é fundante demarcar que o cuidado aqui não é negação de si, abnegação, renúncia, mas um cuidado demarcado por luta, por força, pela maternidade entranhada mesmo em quem nunca pariu, pelo deixar-se chorar para que a dor traga força, ensine, fortaleça... Sueli Carneiro (2018, p. 112) nos diz: Fomos educadas para cuidar dos outros, de nossos companheiros, de nossos filhos, de nossos pais. Durante séculos a obrigatoriedade desses cuidados foram fatores de opressão. Mas de dentro dessa opressão desenvolvemos um forte sentimento de compaixão que nos permite hoje cuidar do mundo, reeducá-lo sem dor e sem opressão. 42 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 31 • jan – fev 2020, p. 27-47 na sabedoria atemporal dos provérbios, adágios, assim como na filosofia de nossas avós” (Idem). b d O feminino está em tudo, em todos os lugares, ele é a possibilidade de criar, de nascer, é escuta, sensibilidade, motor da existência, inclusive, do próprio mundo, somos resistência, re-existência! O feminino é a energia do encantamento, pois é quem permite o existir com ética, amorosidade, cuidado, numa relação contínua com o Outro, com a natureza, é o que dá vida, permite a vida, dá sentido, assim, é a ancestralidade perpassando e criando sentidos, encantando. Implicação, resistência, cuidado, encanto... Assim, vamos bordar propostas de descolonização potencializando o feminino que há em nós. A vida é um ato de encantamento, de reflexão e luta por nossa própria humanidade, portanto, voltar-se para nossas histórias, contadas desde nós mesmas, desde nossas ancestralidades e vivências de nossos saberes é ressignificar e potencializar nossa existência negada, valorizando e potencializando a vida desde nossos corpos. Insurgências próprias de corpos que carregam toda uma linhagem, corpos-templos ancestrais, encantados. Portanto, ao nos encantarmos, não abrimos mão de compreender, aprender e apreender desde as diversidades, desde as encruzilhadas, desde nossos corpos, nossos saberes diversos, múltiplos, plurais. Pois nossas identidades são históricas, políticas, culturais, comunitárias... experienciadas. São os saberes ancestrais femininos que tecem o encanto pela vida em comunidade, o feminino é o útero do mundo, a potência da vida comunitária. O encantamento é coletivo, ancestral. E é esse encantamento tecido pela ancestralidade que potencializa a filosofia africana na ação contínua da descolonização do ser-tão que tece nossa existência. E u - M u l h e r Conceição Evaristo6 E u - M u l h e r Conceição Evaristo6 Uma gota de leite me escorre entre os seios. Uma mancha de sangue me enfeita entre as pernas. Meia palavra mordida me foge da boca. Vagos desejos insinuam esperanças. Eu-mulher em rios vermelhos inauguro a vida. Vagos desejos insinuam esperanças. Eu-mulher em rios vermelhos inauguro a vida. 6 EVARISTO, 2017, p. 23. 44 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 HOOKS, bell. Olhares negros: raça e representação. Tradução de Stephanie Borges. São Paulo: Elefante, 2019. HOOKS, bell. Vivendo de amor. In: WERNECK, Jurema; MENDONÇA, Maisa; C.WHITE Evelyn (Orgs). Traduzido por Maisa Mendonça, Marilena Agostini e Maria Cecília MacDowell dos Santos. O livro da saúde das mulheres negras: nossos passos vêm de longe. 2. ed. – Rio de Janeiro: Pallas / Criola, 2006. LIMA, Aidil Araújo. Mulheres Sagradas. Cachoeira – Ba, Portuário Atelier Editorial, 2017. MACHADO, Adilbênia Freire. Filosofia africana: ética de cuidado e de pertencimento ou uma poética de encantamento. Número Especial “Filosofia Africana: pertencimento, resistência e educação”: Problemata: R. Intern. Fil. V. 10. n. 2 (2019), p. 56-75. MACHADO, Vanda. Pele da Cor da Noite. Salvador: EDUFBA, 2013. MACHADO, Vanda. Exu: o senhor dos caminhos e das alegrias. VI Enecult, encontro de estudos multidisciplinares em culturas. Facom-UFBA – Salvador / Bahia / Brasil. 25 a 27 de Maio de 2010. OLIVEIRA, Kiusam Regina de. Pedagogia da Ancestralidade. In: Revista eonline, postado em 18 de Julho de 2019. Visitado em 08 de Agosto de 2019. Vide: https://www.sescsp.org.br/online/artigo/13431_PEDAGOGIA+DA+ANCESTRALIDADE PETIT, Sandra Haydée. Pretagogia: Pertencimento, Corpo-Dança Afroancestral e Tradição Oral – Contribuições do Legado Africano para a Implementação da Lei nº 10.639/2003. Fortaleza: EdUECE, 2015. SOMÉ, Sobonfu. O Espírito da Intimidade: ensinamentos ancestrais africanos sobre relacionamentos.SP: Odysseus Editora, 2003. SOMÉ, Sobonfu. Aceitar a dor: Quando banhar-se em lágrimas cura as feridas mais profundas. Texto foi publicado em Africaneando. Revista de actualidade y experiencias, nº. 09, 2012. www.oozebap.org/africaneando. Tradução de Aline Matos da Rocha, março de 2014. Acesso em Outubro de 2014. THEODORO, Helena. Mito e Espiritualidade: mulheres negras. Rio de Janeiro: Pallas ed., 1996. WALSH, Catherine. Interculturalidade Crítica e Pedagogia Decolonial: in-surgir, re-existir e re- viver. In: CANDAU, Vera Maria. Educação Intercultural na América Latina: entre concepções, tensões e propostas. 2009. REFERÊNCIAS BIBLIOGRÁFICAS ALVES, Maria Kellynia Farias. Resistência Negra no Círculo de Cultura Sociopoético: pretagogia e produção didática para a implementação da lei 10.639/03 no projovem urbano. Dissertação (mestrado) em Educação. Universidade Federal do Ceará. Programa de Pós- Graduação em Educação. Fortaleza, 2015. 159f. BANKOE, Katherine. Mulheres africanas nos Estados Unidos. In: NASCIMENTO, Elisa Larkin (Org.). Afrocentricidade: uma abordagem epistemológica inovadora. São Paulo: Selo Negro, 2009. (Sankofa: matrizes africanas da cultura brasileira; 4). CARNEIRO, Sueli. Escritos de uma vida. Belo Horizonte (MG): Letramento, 2018. CARNEIRO, Sueli. Racismo, sexismo e desigualdade no Brasil. São Paulo: Selo Negro, 2011. CARNEIRO, Fernanda. Nossos Passos Vêm de Longe. In: WERNECK, Jurema; MENDONÇA, Maisa; WHITE, Evelyn C (Orgs). O livro da saúde das mulheres negras: nossos passos vêm de longe. 2. Ed. – Rio de Janeiro: Pallas / Criola, 2006. CHIZIANE, Paulina. Eu, mulher... por uma nova visão do mundo. 2.ed. Belo Horizonte: Nandyala, 2016. CHIZIANE, Paulina. Não haverá filosofia africana a partir de um livro da Europa, 2014. Disponível em: https://www.geledes.org.br/nao-havera-filosofia-africana-partir-de-um-livro-da- europa/. EVARISTO, Conceição. Conceição Evaristo: “O que nós conquistamos não foi porque a sociedade abriu a porta, mas porque forçamos a passagem”. Entrevista Concedida a Kamille Viola. Revista Marie Claire, Maio de 2018. https://revistamarieclaire.globo.com/Mulheres-do- Mundo/noticia/2018/05/conceicao-evaristo-o-que-nos-conquistamos-nao-foi-porque-sociedade- abriu-porta-mas-porque-forcamos-passagem.html . Acesso em 26 de maio de 2018. EVARISTO, Conceição. Poemas da recordação e outros movimentos. Rio de Janeiro: Malê, 2017. EVARISTO, Conceição. Poemas da recordação e outros movimentos. Rio de Janeiro: Malê, 2017. EVARISTO, Conceição. Olhos d’água. 1. Ed. – Rio de Janeiro: Pallas: Fundação Biblioteca Nacional, 2016. EVARISTO, Conceição. Olhos d’água. 1. Ed. – Rio de Janeiro: Pallas: Fundação Biblioteca Nacional, 2016. 45 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 GOMES, Nilma Lino. Por uma indignação antirracista e diaspórica: negritude e afrobrasilidade em tempos de incertezas. Revista da Associação Brasileira de Pesquisadores/as Negros/as (ABPN), [S.l.], v. 10, n. 26, p. 111-124, out. 2018. ISSN 2177-2770. Disponível em: <http://abpnrevista.org.br/revista/index.php/revistaabpn1/article/view/642>. Acesso em: 20 jan. 2020. HOOKS, bell. Olhares negros: raça e representação. Tradução de Stephanie Borges. São Paulo: Elefante, 2019. HOOKS, bell. Olhares negros: raça e representação. Tradução de Stephanie Borges. São Paul Elefante, 2019. FONTES ORAIS DONA TOINHA. COM-VERSAS. Entrevista realizada em Água Preta, 17 e 18 de Agosto de 2010. VANDA MACHADO. COM-VERSAS NA COZINHA. Entrevista (com-versa) realizada em Salvador, 27 de Setembro de 2017. 46 VANDA MACHADO. COM-VERSAS NA COZINHA. Entrevista (com-versa) realizada em Salvador, 27 de Setembro de 2017. 46 46 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47 PINTO, Makota Valdina. Seja mestre de si. 2018.Vide: https://www.youtube.com/results?search_query=TPSM_Conexão%7C+Seja+mestre+de+si+- +makota+valdina 47 Revista da ABPN • v. 12, n. 31 • jan – fev 2020, p. 27-47
https://openalex.org/W4313245337	https://ojs.uniska-bjm.ac.id/index.php/AIJP/article/download/8562/4842	Indonesian	null	PENERAPAN APLIKASI “SMART CLASS” SEBAGAI MEDIA PEMBELAJARAN INOVATIF BAGI GURU SEKOLAH DASAR	Jurnal Pengabdian Al-Ikhlas Universitas Islam Kalimantan Muhammad Arsyad Al Banjary	2,022	cc-by	1,850	ABSTRAK Perkembangan teknologi informasi di pasca pandemi saat ini, telah membawa perubahan yang sangat besar bagi kemajuan dunia pendidikan. Seiring dengan perkembangan tersebut, media pembelajaran juga banyak mengalami perkembangan. Akan tetapi, fakta yang terjadi di lapangan, banyak para guru di SDN 3 Banjaranyar Kecamatan Tanjunganom Kabupaten Nganjuk dalam proses mengajarnya belum sepenuhnya kreatif dan inovatif, sehingga minat belajar siswa-siswi di SDN 3 Banjaranyar Tanjunganom tersebut kurang termotivasi dalam menuntut ilmu. Berdasarkan permasalahan tersebut, maka dapat dilaksanakan kegiatan pengabdian kepada masyarakat berupa pemanfaatan media pembelajaran berupa aplikasi “smart class” berbasis macromedia flash pada proses belajar mengajar. Metode yang digunakan pada kegiatan pengabdian kepada masyarakat ini menggunakan model pengembangan instruksional, yang meliputi tahap perencanaan, tahap sosialisasi dan tahap pelatihan dengan tujuan guru SDN 3 Banjaranyar Tanjunganom Nganjuk dapat menguasai macromedia flash untuk digunakan sebagai media pembelajaran. Adapun instrumen yang digunakan pada saat kegiatan pengabdian kepada masyarakat, diantaranya observasi lapangan, diskusi, dokumentasi dan pendampingan yang disertai forum diskusi kelompok. Hasil yang telah dicapai dari kegiatan pengabdian kepada masyarakat ini adalah memberikan tambahan pengetahuan dan wawasan tentang pemanfaatan media pembelajaran berupa aplikasi “smart class” berbasis macromedia flash pada proses belajar mengajar kepada siswa-siswi yang ada di SDN 3 Banjaranyar Tanjunganom Nganjuk. j g g j ata Kunci : media pembelajaran, aplikasi smart class, macromedia flash. Kata Kunci : media pembelajaran, aplikasi smart class, macromedia flash metode pembelajaran secara daring ataupun metode pembelajaran berupa media pembelajaran yang inovatif. Salah satu komponen yang memiliki peran penting dalam kegiatan pembelajaran adalah media pembelajaran (Rahman et al., 2020). Kasus seperti ini yang dialami oleh SD Negeri 3 Banjaranyar yang berlokasi di PENERAPAN APLIKASI “SMART CLASS” SEBAGAI MEDIA PEMBELAJARAN INOVATIF BAGI GURU SEKOLAH DASAR Dian Efytra Yuliana1, Yudo Bismo Utomo2, Danang Erwanto1, Mohammad Rangga Nur Faizin1 1Fakultas Teknik, Teknik Elektro, Universitas Islam Kadiri Kediri 2Fakultas Teknik, Teknik Komputer, Universitas Islam Kadiri Kediri E-mail : dianefytra@uniska-kediri.ac.id Perkembangan Perkembangan teknologi informasi di pasca pandemi saat ini, telah membawa perubahan yang sangat besar bagi kemajuan dunia pendidikan. Seiring dengan perkembangan tersebut metode pembelajaran juga banyak mengalami perkembangan (Napitupulu, 2020; Suyadi, 2022). Baik berupa ISSN : 2461-0992 ISSN : 2461-0992 Jurnal Pengabdian Al-Ikhlas Volume 8 Nomor 2, Desember 2022 j g g j Kata Kunci : media pembelajaran, aplikasi smart class, macromedia flash. teknologi 153 ISSN : 2461-0992 ISSN : 2461-0992 Jurnal Pengabdian Al-Ikhlas Volume 8 Nomor 2, Desember 2022 untuk diaplikasikan pada kegiatan belajar mengajar di kelas. Karena Guru merupakan kunci utama terwujudnya pencapaian pencapaian tujuan pembelajaran (Rahmawati, Krisanjaya and Azmin, 2022). Akan tetapi, sebagian guru di SD Negeri 3 Banjaranyar masih menggunakan cara yang lama dan terlalu monoton, yaitu metode ceramah di depan kelas, yang membuat siswa dan siswi di sekolah tersebut menjadi bosan dalam menerima materi yang disampaikan oleh guru (Dewi, 2018; Satriani, 2018; Wulandari, Handayani and Prasetyo, 2019). Permasalahan tersebut yang sedang dialami oleh para guru yang ada di SD Negeri 3 Banjaranyar saat ini. Jalan Majapahit, Desa Banjar Anyar, Kecamatan Tanjunganom, Kabupaten Nganjuk, Jawa Timur. Fakta yang terjadi di lapangan, banyak para guru di SD Negeri 3 Banjaranyar dalam proses belajar mengajar belum sepenuhnya inovatif dan kurang menarik, sehingga minat belajar siswa maupun siswi di sekolah tersebut kurang termotivasi dalam belajarnya. Selain itu, di sekolahan tersebut sudah terpasang proyektor di tiap-tiap kelas, akan tetapi belum sepenuhnya dimanfaatkan oleh para guru di SD Negeri 3 Banjaranyar Kecamatan Tanjunganom Kabupaten Nganjuk tersebut. Hal ini dikarenakan terbatasnya kemampuan guru dalam mengopersikan perangkat tersebut. Untuk proses belajar mengajar yang ada di SD Negeri 3 Banjaranyar dapat dilihat pada gambar berikut ini. Sebagai wujud ikut serta dalam mengembangkan IPTEK di Perguruan Tinggi, maka kami berinisiatif pada kegiatan pengabdian kepada masyarakat kali ini mengadakan kegiatan dengan tema pemanfaatan aplikasi Smart Class sebagai media pembelajaran yang berbasis Macromedia Flash, sehingga dengan adanya media pembelajaran tersebut proses belajar mengajar yang disampaikan oleh guru ke siswa dan siswi anak didiknya lebih kreatif dan inovatif. Gambar 1. Proses Belajar Mengajar di SD 3 Banjaranyar Gambar 1. Proses Belajar Mengajar di SD 3 Banjaranyar Seorang Guru dituntut untuk mengikuti perkembangan teknologi 154 154 Jurnal Pengabdian Al-Ikhlas Volume 8 Nomor 2, Desember 2022 Jurnal Pengabdian Al-Ikhlas Volume 8 Nomor 2, Desember 2022 ISSN : 2461-0992 ISSN : 2461-0992 Di tahap ini, kami melakukan identifikasi masalah terlebih dahulu dengan instrument yang digunakan adalah observasi lapangan, dengan cara mendatangi tempat lokasi yang akan dilaksanakan kegiatan pengabdian. Setelah itu, kami menggunakan instrument berikutnya berupa diskusi dengan mitra pengabdian, yaitu para guru yang ada di SD Negeri 3 Banjaranyar, dengan fokus permasalahan yang dihadapi oleh mitra pengabdian. Kemudian kita melakukan pengkajian terhadap permasalahan yang sedang dihadapi oleh mitra tersebut. teknologi Macromedia flash itu sendiri merupakan sebuah program yang didesain khusus oleh macromedia, dengan kemampuan untuk membuat animasi 2 dimensi yang handal dan ringan, sehingga macromedia flash banyak digunakan untuk menghasilkan efek animasi pada website, media pembelajaran yang interaktif dan yang lainnya (Meilinda, Nuraisyah and Senjayawati, 2019; Qusyairi, 2019; Alim, 2021; Nurhasanah, 2021). Manfaat lain, dengan adanya aplikasi smart class sebagai media pembelajaran yang berbasis Macromedia Flash, supaya minat belajar siswa dan siswi di SD Negeri 3 Banjaranyar menjadi menarik dan termotivasi dalam menuntut ilmu. Gambar 2. Diskusi Dengan Mitra Pengabdian KEGIATAN PENGABDIAN Metode yang digunakan pada kegiatan pengabdian kepada masyarakat ini menggunakan model pengembangan instruksional, yang meliputi tahap perencanaan kegiatan, tahap sosialisasi dan tahap pelatihan. Uraian dari ketiga tahap tersebut adalah sebagai berikut: Gambar 2. Diskusi Dengan Gambar 2. Diskusi Dengan Mitra Pengabdian Mitra Pengabdian 2. Tahap kedua dilakukan tahap sosialisasi. Di tahap ini, kami melakukan sosialisasi tentang pentingnya pemanfaatan media pembelajaran aplikasi “smart class” berbasis macromedia flash bagi guru yang ada di SD Negeri 3 1. Pada tahap pertama dilakukan tahap perencanaan kegiatan pengabdian. 155 Jurnal Pengabdian Al-Ikhlas Volume 8 Nomor 2, Desember 2022 ISSN : 2461-0992 Gambar 4. Tahap Pelatihan Hari Pertama Gambar 5. Tahap Pelatihan Hari Kedua Banjaranyar. Hasil dari sosialisasi ini adalah semua peserta antusias dalam mengikuti sosialisasi ini. Hal ini dikarenakan mitra pengabdian mendapatkan pengetahuan terhadap pemanfaatan aplikasi “smart class” berbasis macromedia flash pada proses belajar mengajar di kelas. Gambar 4. Tahap Pelatihan Hari Pertama Gambar 4. Tahap Pelatihan Hari Pertama Pertama Gambar 3. Tahap Sosialisasi Gambar 5. Tahap Pelatihan Hari Kedua Gambar 5. Tahap Pelatihan Hari Kedua Gambar 3. Tahap Sosialisasi 3. Setelah dilakukan sosialisasi, tahap ketiga atau tahap terakhir dari kegiatan pengabdian ini adalah tahap pelatihan. Di tahap pelatihan ini, kami memberikan pelatihan selama 2 (dua) hari kepada mitra pengabdian tentang bagaimana cara membuat media pembelajaran berupa aplikasi “smart class” berbasis macromedia flash yang menarik supaya minat belajar siswa dan siswi yang ada di SD Negeri 3 Banjaranyar menjadi termotivasi dalam menuntut ilmu. Gambar 5. Tahap Pelatihan Hari Kedua 3. Setelah dilakukan sosialisasi, tahap ketiga atau tahap terakhir dari kegiatan pengabdian ini adalah tahap pelatihan. Di tahap pelatihan ini, kami memberikan pelatihan selama 2 (dua) hari kepada mitra pengabdian tentang bagaimana cara membuat media pembelajaran berupa aplikasi “smart class” berbasis macromedia flash yang menarik supaya minat belajar siswa dan siswi yang ada di SD Negeri 3 Banjaranyar menjadi termotivasi dalam menuntut ilmu. KEGIATAN Hasil yang telah dicapai dari kegiatan pengabdian kepada masyarakat ini adalah sebagai berikut: 1. Mitra pengabdian telah memanfaatkan media pembelajaran yang inovatif berupa aplikasi “smart class” yang berbasis macromedia flash, karena dengan menggunakan aplikasi “smart class” ini proses belajar mengajar di kelas menjadi lebih menarik. 156 156 ISSN : 2461-0992 Jurnal Pengabdian Al-Ikhlas Volume 8 Nomor 2, Desember 2022 Gambar 6. Halaman Awal Aplikasi “Smart Class” Gambar 7. Halaman Utama Aplikasi “Smart Class” Gambar 8. Halaman Keluar Aplikasi “Smart Class” Gambar 9. Ketrampilan Mitra Mengoperasikan Aplikasi Gambar 6. Halaman Awal Aplikasi “Smart Class” Gambar 9. Ketrampilan Mitra Mengoperasikan Aplikasi Gambar 6. Halaman Awal Aplikasi “Smart Class” Gambar 9. Ketrampilan Mitra Mengoperasikan Aplikasi Gambar 9. Ketrampilan Mitra Mengoperasikan Aplikasi “Smart Class” Mengoperasikan Aplikasi Gambar 7. Halaman Utama Aplikasi “Smart Class” Gambar 7. Halaman Utama Aplikasi “Smart Class” 3. Mitra pengabdian telah memperoleh tambahan pengetahuan dalam menerapkan media pembelajaran berupa aplikasi “smart class” berbasis macromedia flash dalam proses belajar mengajar di kelas, sehingga minat belajar siswa dan siswi di sekolahan tersebut menjadi menarik dan termotivasi dalam menuntut ilmu. Gambar 7. Halaman Utama Aplikasi “Smart Class” Gambar 7. Halaman Utama Aplikasi “Smart Class” Gambar 8. Halaman Keluar Aplikasi “Smart Class” KESIMPULAN Kesimpulan yang dapat diambil dari kegiatan pengabdian masyarakat yang telah dilaksanakan adalah dengan menerapkan media pembelajaran berupa aplikasi “smart class” berbasis macromedia flash, telah membantu mitra pengabdian, yaitu para guru yang ada di SD Negeri 3 Banjaranyar Kecamatan Tanjunganom Kabupaten Nganjuk dalam proses belajar mengajar di kelas, sehingga minat belajar siswa dan siswi di sekolahan tersebut menjadi Gambar 8. Halaman Keluar Aplikasi “Smart Class” 2. Mitra pengabdian sudah terampil dalam menggunakan dan mengoperasikan sendiri media pembelajaran berupa aplikasi “smart class” berbasis macromedia flash. 157 ISSN : 2461-0992 ISSN : 2461-0992 Jurnal Pengabdian Al-Ikhlas Volume 8 Nomor 2, Desember 2022 menarik dan termotivasi dalam menuntut ilmu. PKM Bidang Ilmu Pendidikan), 2(3). Available at: https://doi.org/10.54371/ainj.v2i3 .69. DAFTAR PUSTAKA Qusyairi, L.A.H. (2019) ‘Pembuatan Media Pembelajaran Fisika Berbasis Macromedia Flash’, ISLAMIKA, 1(1). Available at: https://doi.org/10.36088/islamika .v1i1.184. Alim, S.N. (2021) ‘Pengembangan Media Pembelajaran Berbasis Multimedia Menggunakan Macromedia Flash dalam Fiqih Islam Materi Pokok Sholat dan Zakat’, Munaddhomah: Jurnal Manajemen Pendidikan Islam, 2(1). Available at: https://doi.org/10.31538/munadd homah.v2i1.66. Rahman, F. et al. (2020) ‘Pelatihan Pembuatan Presentasi Interaktif Menggunakan Prezi Pada Guru Di SD IT Anak Sholeh Mandiri Banjarmasin’, Jurnal Pengabdian Al-Ikhlas, 6(2). Available at: https://doi.org/10.31602/jpaiunis ka.v6i2.3894. Dewi, E.R. (2018) ‘Metode Pembelajaran Modern Dan Konvensional Pada Sekolah Menengah Atas’, PEMBELAJAR: Jurnal Ilmu Pendidikan, Keguruan, dan Pembelajaran, 2(1). Available at: https://doi.org/10.26858/pembela jar.v2i1.5442. Rahmawati, A., Krisanjaya, K. and Azmin, G.G. (2022) ‘Pengembangan Pembelajaran Menggunakan QR Code Untuk Meningkatkan Literasi Siswa Di SIT Al Haraki Depok’, Jurnal Pengabdian Al-Ikhlas, 8(1). Available at: https://doi.org/10.31602/jpaiunis ka.v8i1.4471. Meilinda, N.V., Nuraisyah, L.F. and Senjayawati, E. (2019) ‘Implementasi Media Pembelajaran Menggunakan Aplikasi Macromedia Flash 8 Pada Materi Bangun Ruang Sisi Datar’, Journal on Education, 1(3). Satriani, S. (2018) ‘Inovasi Pendidikan: Metode Pembelajaran Monoton ke Pembelajaran Variatif (Metode Ceramah Plus)’, Jurnal Ilmiah Iqra’, 10(1). Available at: https://doi.org/10.30984/jii.v10i1. 590. Napitupulu, R.M. (2020) ‘Dampak pandemi Covid-19 terhadap kepuasan pembelajaran jarak jauh’, Jurnal Inovasi Teknologi Pendidikan, 7(1). Available at: https://doi.org/10.21831/jitp.v7i1. 32771. Suyadi, S. (2022) ‘Dampak Pembelajaran Daring pada Perkembangan Psikososial Anak Saat Pandemi COVID-19’, Jurnal Obsesi : Jurnal Pendidikan Anak Usia Dini, 6(4). Available at: https://doi.org/10.31004/obsesi.v 6i4.2349. Nurhasanah, E. (2021) ‘Pengembangan Multimedia Pembelajaran Sejarah Perkembangan Islam Berbasis Macromedia Flash untuk Meningkatkan Hasil Belajar Mahasiswa’, Ainara Journal (Jurnal Penelitian dan Wulandari, A., Handayani, P. and 158 Jurnal Pengabdian Al-Ikhlas ISSN : 2461-0992 Volume 8 Nomor 2, Desember 2022 Prasetyo, D.R. (2019) ‘Pembelajaran Ilmu Pengetahuan Alam Berbasis EMC (Education Mini Club) sebagai Solusi Menghadapi Tantangan Pendidikan dI Era Revolusi Industri 4.0’, Thabiea : Journal Of Natural Science Teaching, 2(1). Available at: https://doi.org/10.21043/thabiea. v2i1.5498. ISSN : 2461-0992 ISSN : 2461-0992 Jurnal Pengabdian Al-Ikhlas Volume 8 Nomor 2, Desember 2022 Prasetyo, D.R. (2019) ‘Pembelajaran Ilmu Pengetahuan Alam Berbasis EMC (Education Mini Club) sebagai Solusi Menghadapi Tantangan Pendidikan dI Era Revolusi Prasetyo, D.R. (2019) ‘Pembelajaran Ilmu Pengetahuan Alam Berbasis EMC (Education Mini Club) sebagai Solusi Menghadapi Tantangan Pendidikan dI Era Revolusi Industri 4.0’, Thabiea : Journal Of Natural Science Teaching, 2(1). Available at: https://doi.org/10.21043/thabiea. v2i1.5498. Industri 4.0’, Thabiea : Journal Of Natural Science Teaching, 2(1). Available at: https://doi.org/10.21043/thabiea. v2i1.5498. 159 159
https://openalex.org/W2938403062	https://www.nature.com/articles/s41598-020-57500-1.pdf	English	null	Multiplexed and Inducible Gene Modulation in Human Pluripotent Stem Cells by CRISPR Interference and Activation	bioRxiv (Cold Spring Harbor Laboratory)	2,019	cc-by	11,287	A multiplexed gRNA piggyBac transposon system facilitates efficient induction of CRISPRi and CRISPRa in human pluripotent stem cells OPEN Dane Z. Hazelbaker1, Amanda Beccard1, Gabriella Angelini1, Patrizia Mazzucato1, Angelica Messana1, Daisy Lam1, Kevin Eggan1,2 & Lindy E. Barrett1,2* CRISPR-Cas9-mediated gene interference (CRISPRi) and activation (CRISPRa) approaches hold promise for functional gene studies and genome-wide screens in human pluripotent stem cells (hPSCs). However, in contrast to CRISPR-Cas9 nuclease approaches, the efficiency of CRISPRi/a depends on continued expression of the dead Cas9 (dCas9) effector and guide RNA (gRNA), which can vary substantially depending on transgene design and delivery. Here, we design and generate new fluorescently labeled piggyBac (PB) vectors to deliver uniform and sustained expression of multiplexed gRNAs. In addition, we generate hPSC lines harboring AAVS1-integrated, inducible and fluorescent dCas9-KRAB and dCas9-VPR transgenes to allow for accurate quantification and tracking of cells that express both the dCas9 effectors and gRNAs. We then employ these systems to target the TCF4 gene in hPSCs and assess expression levels of the dCas9 effectors, individual gRNAs and targeted gene. We also assess the performance of our PB system for single gRNA delivery, confirming its utility for library format applications. Collectively, our results provide proof-of-principle application of a stable, multiplexed PB gRNA delivery system that can be widely exploited to further enable genome engineering studies in hPSCs. Paired with diverse CRISPR tools including our dual fluorescence CRISPRi/a cell lines, this system can facilitate functional dissection of individual genes and pathways as well as larger-scale screens for studies of development and disease. CRISPR-Cas9 systems have revolutionized genome editing in myriad cell types and organisms and ushered the development of variant technologies that utilize dCas9 fused to epigenetic modifiers which can be localized to a gene of interest upon expression of a gRNA1–3. Two such approaches are CRISPRi, which fuses dCas9 to tran- scriptional repressors, such as the KRAB domain3, and CRISPRa which fuses dCas9 to transcriptional activators, such as the chimeric VPR domain2. These tools can be deployed for both single and multiplexed gene manip- ulation and allow modulation of gene expression in the absence of cellular toxicity caused by Cas9-mediated DNA double-strand breaks4. CRISPRi/a set-ups have been used successfully in studies of cellular programming5, cellular reprogramming6,7, in vivo gene manipulation8, enhancer screens9, chemical screens10, and whole-genome genetic interaction mapping studies11. When targeting populations of cells, gene repression through CRISPRi is reported to be more homogeneous and efficient compared to Cas9 nuclease12. www.nature.com/scientificreports www.nature.com/scientificreports www.nature.com/scientificreports 1Stanley Center for Psychiatric Research, Broad Institute of MIT and Harvard, Cambridge, MA, 02142, USA. 2Department of Stem Cell and Regenerative Biology, Harvard University, Cambridge, MA, 02138, USA. email: lbarrett@broadinstitute.org Scientific Reports \| (2020) 10:635 \| https://doi.org/10.1038/s41598-020-57500-1 Results Generation of AAVS1 integrated and doxycycline-inducible dCas9-KRAB and dCas9-VPR hPSC lines. To create stable CRISPRi and CRISPRa hPSC lines, we cloned and introduced all-in-one cassettes containing S. pyogenes dCas9 fused to the KRAB repressor domain21 or VPR activation domain2 into the AAVS1 safe-harbor locus of the XY embryonic stem cell line H122 via a TALEN-mediated gene-trap approach that confers neomycin (G418) resistance to cells upon on-target integration12,16 (Fig. 1a). In both constructs, dCas9-KRAB and dCas9-VPR expression is driven by the TRE3G doxycycline inducible promoter (Takara Bio) and fused to Enhanced Green Fluorescent Protein (EGFP) transcriptional reporters by an IRES sequence (dCas9-KRAB) or a T2A self-cleaving peptide sequence (dCas9-VPR). Following selection with G418, dCas9-KRAB and dCas9-VPR clones were assessed for EGFP expression and genotyped by junction PCR (Supplementary Fig. S1). From these data, dCas9-KRAB and dCas9-VPR clones were expanded and con- firmed to have normal karyotypes (data not shown).ii i y yp To validate inducible and reversible dCas9 expression in our CRISPRi/a hPSC lines, we first quantified EGFP fluorescence by flow cytometry following 48 hours of doxycycline treatment. Doxycycline led to strong induc- tion of EGFP fluorescence, reaching 99% in both dCas9-VPR and dCas9-KRAB hPSC lines (+48 h; Figs. 1b,c, Supplementary Fig. S1). 168 hours after washout of doxycycline, EFGP fluorescence levels dropped to back- ground levels in both the dCas9-KRAB and dCas9-VPR lines (−168 h; Fig. 1b,c, Supplementary Fig. S1). We also observed strong induction of dCas9-KRAB and dCas9-VPR protein expression after doxycycline induc- tion (Fig. 1d,e, Supplementary Fig. S1) and loss of detectable dCas9 expression by 120 hours post-washout in dCas9-KRAB cells and by 72 hours in dCas9-VPR cells (Fig. 1d,e, Supplementary Fig. S1). The increased stability of dCas9-KRAB and EGFP protein in dCas9-KRAB cells in comparison to dCas9-VPR cells may be due to the presence of the WPRE (Woodchuck Hepatitis Virus Post-transcriptional Response Element) in the 3′UTR of the dCas9-KRAB construct (Fig. 1a), which has been reported to increase transcript stability23. Similar to previous reports, dCas9 protein was not detected in the absence of doxycycline12 (Fig. 1d,e, Supplementary Fig. S1). These data confirm that our AAVS1-integrated dCas9-KRAB and dCas9-VPR constructs exhibit robust induction and reversibility of dCas9 expression in hPSCs. Identification of the relevant transcriptional start site of TCF4 in hPSCs. To assess the potency of our dCas9-KRAB and dCas9-VPR systems for gene repression and activation, we chose to target the TCF4 gene in hPSCs as an example. www.nature.com/scientificreports/ persistent and uniform expression of dCas9 effectors and gRNA across cell populations, an important considera- tion both in single gene studies and whole-genome screens.hf persistent and uniform expression of dCas9 effectors and gRNA across cell populations, an important considera- tion both in single gene studies and whole-genome screens.hf g g g There is limited data on the stability of dCas9 effectors12 and studies report variability in the induction and expression of different promoters in different loci due to de novo DNA methylation13. Further, gRNA delivery and expression require optimization in order to fully capitalize on the multiplexing potential of CRISPRi/a. With regard to gRNA delivery, previous studies have utilized transfection and selection of plasmid DNA12,14,15 transient transfection of in vitro transcribed gRNA16,17, lentiviral integration17 or piggyBac transposon-based integration18. In particular, piggyBac (PB) delivery methods have the advantages of being easy to clone and deliver into hPSCs and carry substantially larger payload compared to lentiviral vectors19,20. As a result, PB vectors are particularly applicable for studies of parallel pathways or polygenic disease, enabling the perturbation of many genes with a single delivery vehicle at minimal cost. g y Here, we developed a new piggyBac vector system to enable rapid cloning and stable delivery of multiple gRNAs for CRISPRi/a applications. We coupled this system with a set of hPSC lines harboring genomically inte- grated and inducible dCas9-KRAB and dCas9-VPR, including a dual-fluorescent readout to readily quantify cells that express both gRNAs and dCas9 variants in a population. We then quantified expression levels of the effector components as well as a targeted gene, TCF4, at both the transcript and protein levels. We also assessed the per- formance of our PB system for single gRNA delivery targeting TCF4 and POGZ. Our results confirm the utility of the dual-fluorescent readout and multiplexed or single PB gRNA delivery system for CRISPRi/a that can be broadly employed in hPSCs for gene perturbation studies. A multiplexed gRNA piggyBac transposon system facilitates efficient induction of CRISPRi and CRISPRa in human pluripotent stem cells OPEN Indeed, while Cas9-nuclease strategies have been employed in genome-wide screens, they are limited by heterogeneity in the targeted cell populations, which may include a significant number of wild-type cells alongside cells with mixtures of indels that produce partial loss or gain of function phenotypes, or truncated gene products which can complicate interpreta- tions12. Furthermore, CRISPRi/a offers the potential for conditional gene perturbation, allowing for the functional study of essential genes3 and reversibility of phenotypes. However, unlike genetic knockout by CRISPR-Cas9 that requires a single indel formation event to permanently disrupt gene function, successful CRISPRi/a requires 1Stanley Center for Psychiatric Research, Broad Institute of MIT and Harvard, Cambridge, MA, 02142, USA. 2Department of Stem Cell and Regenerative Biology, Harvard University, Cambridge, MA, 02138, USA. email: lbarrett@broadinstitute.org Scientific Reports \| (2020) 10:635 \| https://doi.org/10.1038/s41598-020-57500-1 www.nature.com/scientificreports/ Results Thus, taking into account our protein and transcript Scientific Reports \| (2020) 10:635 \| https://doi.org/10.1038/s41598-020-57500-1 www.nature.com/scientificreports/ a AAVS1 TALENS 5' Arm NEOR TetON CAG EGFP IRES KRAB-dCAS9 TRE3G SA AAVS1 AAVS1 WPRE HA NLS 5' Arm NEOR TetON CAG EGFP IRES KRAB-dCAS9 TRE3G SA AAVS1 AAVS1 WPRE HA 5' Arm NEOR TetON CAG EGFP IRES KRAB-dCAS9 TRE3G SA AAVS1 AAVS1 WPRE HA NLS 5' Arm NEOR TetON CAG EGFP IRES KRAB-dCAS9 TRE3G SA AAVS1 AAVS1 WPRE HA G418 selection dCas9-KRAB integration a a b d 230 180 116 kDa 40 AAVS1 TALENS 5' Arm NEOR TetON CAG EGFP IRES KRAB-dCAS9 TRE3G SA AAVS1 AAVS1 WPRE HA NLS 5' Arm NEOR TetON CAG EGFP IRES KRAB-dCAS9 TRE3G SA AAVS1 AAVS1 WPRE HA 5' Arm NEOR TetON CAG EGFP IRES KRAB-dCAS9 TRE3G SA AAVS1 AAVS1 WPRE HA NLS 5' Arm NEOR TetON CAG EGFP IRES KRAB-dCAS9 TRE3G SA AAVS1 AAVS1 WPRE HA 5' Arm NEOR TetON CAG EGFP TRE3G SA AAVS1 AAVS1 5' Arm NEOR TetON CAG EGFP TRE3G SA AAVS1 AAVS1 NLS FLAG NLS T2A AAVS1 TALENS KRAB-dCAS9 2x NLS 5' Arm NEOR TetON CAG EGFP TRE3G SA AAVS1 AAVS1 5' Arm NEOR TetON CAG EGFP TRE3G SA AAVS1 AAVS1 NLS FLAG NLS T2A KRAB-dCAS9 dCAS9 2x NLS VPR dCas9-VPR integration dCAS9 VPR −Dox +24h +48h −24h −72h −120h −168h α-Cas9 α-GAPDH dCas9-KRAB: EGFP+ cells G418 selection G418 selection dCas9-KRAB integration dCas9-KRAB protein expression 0 20 40 60 80 100 Control +48h −120h −168h % EGFP+ Cells a b d 230 180 116 kDa 40 230 180 116 kDa 40 AAVS1 TALENS 5' Arm NEOR TetON CAG EGFP IRES KRAB-dCAS9 TRE3G SA AAVS1 AAVS1 WPRE HA NLS 5' Arm NEOR TetON CAG EGFP IRES KRAB-dCAS9 TRE3G SA AAVS1 AAVS1 WPRE HA 5' Arm NEOR TetON CAG EGFP IRES KRAB-dCAS9 TRE3G SA AAVS1 AAVS1 WPRE HA NLS 5' Arm NEOR TetON CAG EGFP IRES KRAB-dCAS9 TRE3G SA AAVS1 AAVS1 WPRE HA 5' Arm NEOR TetON CAG EGFP TRE3G SA AAVS1 AAVS1 5' Arm NEOR TetON CAG EGFP TRE3G SA AAVS1 AAVS1 NLS FLAG NLS T2A AAVS1 TALENS KRAB-dCAS9 2x NLS 5' Arm NEOR TetON CAG EGFP TRE3G SA AAVS1 AAVS1 5' Arm NEOR TetON CAG EGFP TRE3G SA AAVS1 AAVS1 NLS FLAG NLS T2A KRAB-dCAS9 dCAS9 2x NLS VPR dCas9-VPR integration dCAS9 VPR −Dox +24h +48h −24h −72h −120h −168h −Dox +24h +48h −24h −72h −120h−168h c e α-Cas9 α-Cas9 α-GAPDH α-GAPDH dCas9-KRAB: EGFP+ cells G418 selection G418 selection dCas9-KRAB integration dCas9-VPR: EGFP+ cells dCas9-VPR protein expression dCas9-KRAB protein expression 0 20 40 60 80 100 Control +48h −120h −168h 0 20 40 60 80 100 % EGFP+ Cells % EGFP+ Cells Control +48h −120h −168h Figure 1. Results (e) dCas9-VPR protein expression before doxycycline treatment (-Dox), after 24 and 48 hours doxycycline treatment (+24 h, +48 h) and after washout of doxycycline for 24, 72, 120, and 168 hours (−24 h, −72 h, −120 h, −168 h). Images in d,e derived from Wes analysis (ProteinSimple). analysis above, we opted to target exon 3b and designed three TCF4 CRISPRi gRNAs (i1, i2, i3) and three TCF4 CRISPRa gRNAs (a1, a2, and a3) that fall within the optimal window of 300 bp from the TSS3,29 using the CRISPR-ERA guide selection tool30 (Supplementary Fig. S2). Design and delivery of multi-gRNA piggyBac vectors in hPSCs. A noted strength of CRISPRi and CRISPRa is the ability to deliver multiple gRNAs for enhanced targeting of one or several genes in the absence of DNA damage31. To facilitate stable delivery of multiple gRNAs in hPSCs, we designed a new vector that incorpo- rates the efficiency and ease of the piggyBac (PB) transposase system32 with a multiplex gRNA cloning system33. To do so, we cloned either the three CRISPRi gRNAs (i1, i2, i3) or CRISPRa gRNAs (a1, a2, a3) targeting TCF4 into individual vectors and sequentially assembled the final PB vector including mRFP and blasticidin resistance via Golden Gate and Gateway cloning (Fig. 2c). Of note, while we opted to introduce three gRNAs per PB vector, the parental vectors allow for the cloning of up seven gRNAs in tandem array33 that can readily be introduced into our PB vectors. Additionally, FRT sites flank the mRFP and blasticidin cassettes (Fig. 2c) to facilitate removal of these selectable features with FLP recombinase and allow for future PB re-targeting events. Using our optimized workflow, multiple gRNAs can be cloned into PB vectors and confirmed via BamHI restriction digest within one week (Supplementary Fig. S2 and Methods). We next co-transfected the multi-gRNA PB vectors along with a plasmid encoding the piggyBac transposase into dCas9-KRAB and dCas9-VPR hPSC lines. Following selection with blasticidin, individual dCas9-KRAB and dCas9-VPR clones were isolated and screened for high levels of uniform mRFP fluorescence (Fig. 2d). We then selected and expanded two independent dCas9-KRAB clones (dCas9-KRAB-PB clones K1 and K2) and two independent dCas9-VPR clones (dCas9-VPR-PB clones V1 and V2) and assessed integrated PB copy number via droplet digital PCR (ddPCR) of the mRFP cassette in both the CRISPRi and CRISPRa PB vectors. Results Generation and validation of AAVS1-integrated inducible dCas9-KRAB and dCas9-VPR hPSC Figure 1. Generation and validation of AAVS1-integrated inducible dCas9-KRAB and dCas9-VPR hPSC lines. (a) Schematic overview of AAVS1 targeting strategy in H1 hPSCs with TRE3G-driven dCas9-KRAB (left) or dCas9-VPR (right) cassettes and TALENs that target AAVS1 and confer G418 resistance upon on-target integration. (b) Quantification of flow cytometry analysis of EGFP fluorescence in dCas9-KRAB cells after 48 hours of doxycycline treatment (+48 h) followed by removal of doxycycline for 120 (−120 h) and 168 hours (−168 h) in comparison to no GFP control H1 cells (control). (c) Quantification of flow cytometry analysis of EGFP fluorescence in dCas9-VPR cells after 48 hours of doxycycline treatment (+48 h) followed by removal of doxycycline for 120 (−120 h) and 168 hours (−168 h) in comparison to no GFP control H1 cells (control). (d) dCas9-KRAB protein expression in absence of doxycycline (−Dox), after 24 and 48 hours doxycycline treatment (+24 h, +48 h), and after washout of doxycycline for 24, 72, 120, and 168 hours (−24 h, −72 h, −120 h, −168 h). (e) dCas9-VPR protein expression before doxycycline treatment (-Dox), after 24 and 48 hours doxycycline treatment (+24 h, +48 h) and after washout of doxycycline for 24, 72, 120, and 168 hours (−24 h, −72 h, −120 h, −168 h). Images in d,e derived from Wes analysis (ProteinSimple). Figure 1. Generation and validation of AAVS1-integrated inducible dCas9-KRAB and dCas9-VPR hPSC lines. (a) Schematic overview of AAVS1 targeting strategy in H1 hPSCs with TRE3G-driven dCas9-KRAB (left) or dCas9-VPR (right) cassettes and TALENs that target AAVS1 and confer G418 resistance upon on-target integration. (b) Quantification of flow cytometry analysis of EGFP fluorescence in dCas9-KRAB cells after 48 hours of doxycycline treatment (+48 h) followed by removal of doxycycline for 120 (−120 h) and 168 hours (−168 h) in comparison to no GFP control H1 cells (control). (c) Quantification of flow cytometry analysis of EGFP fluorescence in dCas9-VPR cells after 48 hours of doxycycline treatment (+48 h) followed by removal of doxycycline for 120 (−120 h) and 168 hours (−168 h) in comparison to no GFP control H1 cells (control). (d) dCas9-KRAB protein expression in absence of doxycycline (−Dox), after 24 and 48 hours doxycycline treatment (+24 h, +48 h), and after washout of doxycycline for 24, 72, 120, and 168 hours (−24 h, −72 h, −120 h, −168 h). Results Generation and validation of AAVS1-integrated inducible dCas9-KRAB and dCas9-VPR hPSC lines. (a) Schematic overview of AAVS1 targeting strategy in H1 hPSCs with TRE3G-driven dCas9-KRAB (left) or dCas9-VPR (right) cassettes and TALENs that target AAVS1 and confer G418 resistance upon on-target integration. (b) Quantification of flow cytometry analysis of EGFP fluorescence in dCas9-KRAB cells after 48 hours of doxycycline treatment (+48 h) followed by removal of doxycycline for 120 (−120 h) and 168 hours (−168 h) in comparison to no GFP control H1 cells (control). (c) Quantification of flow cytometry analysis of EGFP fluorescence in dCas9-VPR cells after 48 hours of doxycycline treatment (+48 h) followed by removal of doxycycline for 120 (−120 h) and 168 hours (−168 h) in comparison to no GFP control H1 cells (control). (d) dCas9-KRAB protein expression in absence of doxycycline (−Dox), after 24 and 48 hours doxycycline treatment (+24 h, +48 h), and after washout of doxycycline for 24, 72, 120, and 168 hours (−24 h, −72 h, −120 h, −168 h). (e) dCas9-VPR protein expression before doxycycline treatment (-Dox), after 24 and 48 hours doxycycline treatment (+24 h, +48 h) and after washout of doxycycline for 24, 72, 120, and 168 hours (−24 h, −72 h, −120 h, −168 h). Images in d,e derived from Wes analysis (ProteinSimple). 5' Arm NEOR TetON CAG EGFP TRE3G SA AAVS1 AAVS1 5' Arm NEOR TetON CAG EGFP TRE3G SA AAVS1 AAVS1 NLS FLAG NLS T2A AAVS1 TALENS KRAB-dCAS9 2x NLS 5' Arm NEOR TetON CAG EGFP TRE3G SA AAVS1 AAVS1 5' Arm NEOR TetON CAG EGFP TRE3G SA AAVS1 AAVS1 NLS FLAG NLS T2A KRAB-dCAS9 dCAS9 2x NLS VPR dCas9-VPR integration dCAS9 VPR G418 selection dCas9-VPR integration dCas9-VPR integration TALENS G418 selection b d 230 180 116 kDa 40 −Dox +24h +48h −24h −72h −120h −168h α-Cas9 α-GAPDH dCas9-KRAB: EGFP+ cells dCas9-KRAB protein expression 0 20 40 60 80 100 Control +48h −120h −168h % EGFP+ Cells b dCas9-KRAB: EGFP+ cells 0 20 40 60 80 100 Control +48h −120h −168h % EGFP+ Cells b d dCas9-KRAB protein expression 230 180 116 kDa 40 −Dox +24h +48h −24h −72h −120h −168h α-Cas9 α-GAPDH c dCas9-VPR: EGFP+ cells 0 20 40 60 80 100 % EGFP+ Cells Control +48h −120h −168h 230 180 116 kDa 40 −Dox +24h +48h −24h −72h −120h−168h e α-Cas9 α-GAPDH dCas9-VPR protein expression c e dCas9-VPR protein expression Figure 1. Results TCF4 plays important roles in development and TCF4 gene dysfunction has been implicated in multiple neurodevelopmental diseases including Pitt-Hopkins syndrome and schizophre- nia by GWAS24–26. Importantly, TCF4 has multiple alternatively-spliced transcripts27 making it critical to identify the most relevant TCF4 isoform and its corresponding transcriptional start site (TSS) to target with CRISPRi/a. Generally speaking, functional gRNA design for CRISPRi/a applications has the added challenge that TSSs may not be well annotated for a given cell type. We therefore first carried out western blot analysis with a recombinant antibody (ab217668, Abcam) that targets a fragment present in all known TCF4 isoforms (amino acids 350–500) and observed two protein bands at approximately 62 kDa (band a) and 72 kDa (band b) (Fig. 2a, Supplementary Fig. S2). To determine the TSSs of potential transcripts that upon translation could generate peptides within the size range of the observed bands in Fig. 2a, we utilized exon-specific RT-qPCR in H1 hPSCs with primers targeting candidate TSS-harboring exons. (Fig. 2b). As shown in Fig. 2b, RT-qPCR analysis revealed exon 3b to be the most dominantly expressed exon, which is the leading exon of the TCF4 3b transcript. Translation of the TCF4 3b transcript is reported to generate the canonical TCF4-B isoform27 of approximately 72 kDa corresponding to band b in Fig. 2a. Importantly, the FANTOM5 promoterome (fantom. gsc.riken.jp/zenbu), which is commonly used to identify the TSS for gRNA design28,29, annotated exon 3d as harboring the dominant TSS based on pooled CTSS tracks (Supplementary Fig. S2). However, we did not observe detectable expression of exon 3d via RT-qPCR in H1 hPSCs using two different primer pairs (data not shown, primers listed in Supplementary Table S1) thus underscoring the importance of validating the relevant TSS for a given cell type in single gene studies. Results Our ddPCR analysis revealed approximately 14 and 31 PB copies in dCas9-KRAB-PB clones K1 and K2, respectively, and approximately 13 and 34 PB copies dCas9-VPR-PB clones V1 and V2, respectively (Fig. 2e). We also confirmed Scientific Reports \| (2020) 10:635 \| https://doi.org/10.1038/s41598-020-57500-1 www.nature.com/scientificreports/ dCas9-KRAB-PB a b e TCF4 Exon Expression (relative to GAPDH) 3b 1 4 3 5b 5 6 7a 7b 7 3c 4a 21 2 3 exon 5b d PB Genomic Copy Number mRFP copies/cell CAG CAG FRT mRFP T2A BlastR U6 gRNA 1 U6 gRNA 2 U6 gRNA 3 ITR attB2 piggyBac transposase FRT ITR genomic DNA CAG CAG ITR Blasticidin Selection U6 gRNA 1 U6 gRNA 2 U6 gRNA 3 attB1 FRT mRFP T2A BlastR U6 gRNA 1 U6 gRNA 2 U6 gRNA 3 attB2 ITR attB1 Golden Gate assembly U6 gRNA 1 U6 gRNA 2 U6 gRNA 3 attL2 attL1 Gateway cloning + c DAPI mRFP FRT T2A FRT 100 μm 100 μm 100 μm 100 μm 100 μm 100 μm 100 μm 100 μm 100 μm 100 μm 100 μm 100 μm attL1 attL2 3d 0 0.001 0.002 0.003 0.004 0.005 exon 3b * exon 3c exon 4a 4 5 6 exon 7a exon 7b −RT 0 10 20 30 40 K1 K2 dCas9-KRAB-PB V1 V2 dCas9-VPR-PB CRISPRi vector 14.2 31.1 13.1 33.9 CRISPRa vector α-TCF4 α-GAPDH merge DAPI mRFP merge DAPI mRFP merge DAPI mRFP merge clone K1 clone K2 clone V1 clone V2 dCas9-VPR-PB Potential TSS locations of TCF4 protein isoforms 75 kDa a b 72 kDa 72 kDa 69 kDa 67 kDa 64 kDa 50 kDa 8a 54 kDa 2 83 kDa Figure 2. Design and delivery of multi-gRNA PB vectors for CRISPRi and CRISPRa targeting of the TCF4 gene. (a) Representative western blot of TCF4 protein expression in hPSCs. Letters a and b denote the two observed bands. (b) Top, Schematic of the TCF4 gene exons containing potential TSS locations for the putative protein isoforms shown in (a). Numbers below exons correspond to the primer pairs used for RT-qPCR. Exon locations are modified from Sepp et al.27, with approximate predicted molecular weights of resulting protein isoform(s) listed above each starting exon in italics, with tested exons highlighted in red. The asterisks under exon 3d indicates there was no detectable PCR signal for this exon with two separate primer pairs (data not shown). Results Bottom, Exon-specific expression of TCF4 transcripts from RT-qPCR of two independent H1 hPSC samples and normalized to GAPDH. -RT lane denotes the no reverse transcriptase control cDNA reaction. Data is shown as mean +/− s.e.m. of two independent biological samples. (c) Overview of multi-gRNA PB vector cloning, delivery, and selection. (d) Representative images of mRFP fluorescence in dCas9-KRAB-PB clones K1 and K2 (top panels) and dCas9-VPR-PB clones V1 and V2 (bottom panels). Cells are counterstained with DAPI (blue). Scale bar = 100 μm. (e) PB vector copy number in dCas9-KRAB and dCas9-VPR clones as determined by ddPCR quantification of mRFP gene. Data is shown as the mean of three experiments with error bars as +/− s.e.m. b TCF4 Exon Expression (relative to GAPDH) 3b 1 4 3 5b 5 6 7a 7b 7 3c 4a 21 2 3 exon 5b d 3d 0 0.001 0.002 0.003 0.004 0.005 exon 3b * exon 3c exon 4a 4 5 6 exon 7a exon 7b −RT Potential TSS locations of TCF4 protein isoforms 72 kDa 72 kDa 69 kDa 67 kDa 64 kDa 8a 54 kDa 2 83 kDa a α-TCF4 α-GAPDH 75 kDa a b 50 kDa b a e PB Genomic Copy Number mRFP copies/cell CAG CAG FRT mRFP T2A BlastR U6 gRNA 1 U6 gRNA 2 U6 gRNA 3 ITR attB2 piggyBac transposase FRT ITR genomic DNA CAG CAG ITR Blasticidin Selection U6 gRNA 1 U6 gRNA 2 U6 gRNA 3 attB1 FRT mRFP T2A BlastR U6 gRNA 1 U6 gRNA 2 U6 gRNA 3 attB2 ITR attB1 Golden Gate assembly U6 gRNA 1 U6 gRNA 2 U6 gRNA 3 attL2 attL1 Gateway cloning + c FRT T2A FRT attL1 attL2 0 10 20 30 40 K1 K2 dCas9-KRAB-PB V1 V2 dCas9-VPR-PB CRISPRi vector 14.2 31.1 13.1 33.9 CRISPRa vector c dCas9-KRAB-PB d DAPI mRFP 100 μm 100 μm 100 μm 100 μm 100 μm 100 μm merge DAPI mRFP merge clone K1 clone K2 d e 100 μm 100 μm 100 μm 100 μm 100 μm 100 μm DAPI mRFP merge DAPI mRFP merge clone V1 clone V2 dCas9-VPR-PB Figure 2. Design and delivery of multi-gRNA PB vectors for CRISPRi and CRISPRa targeting of the TCF4 gene. (a) Representative western blot of TCF4 protein expression in hPSCs. Letters a and b denote the two observed bands. Results (b) Top, Schematic of the TCF4 gene exons containing potential TSS locations for the putative protein isoforms shown in (a). Numbers below exons correspond to the primer pairs used for RT-qPCR. Exon locations are modified from Sepp et al.27, with approximate predicted molecular weights of resulting protein isoform(s) listed above each starting exon in italics, with tested exons highlighted in red. The asterisks under exon 3d indicates there was no detectable PCR signal for this exon with two separate primer pairs (data not shown). Bottom, Exon-specific expression of TCF4 transcripts from RT-qPCR of two independent H1 hPSC samples and normalized to GAPDH. -RT lane denotes the no reverse transcriptase control cDNA reaction. Data is shown as mean +/− s.e.m. of two independent biological samples. (c) Overview of multi-gRNA PB vector cloning, delivery, and selection. (d) Representative images of mRFP fluorescence in dCas9-KRAB-PB clones K1 and K2 (top panels) and dCas9-VPR-PB clones V1 and V2 (bottom panels). Cells are counterstained with DAPI (blue). Scale bar = 100 μm. (e) PB vector copy number in dCas9-KRAB and dCas9-VPR clones as determined by ddPCR quantification of mRFP gene. Data is shown as the mean of three experiments with error bars as +/− s.e.m. that both dCas9-KRAB-PB and dCas9-VPR-PB cells harboring moderate levels of integrated PBs (i.e., 13–34 copies) maintained pluripotency and tri-lineage potential (Supplementary Fig. S2). Thus, our new PB vectors facilitate both rapid cloning and efficient delivery of multiple gRNAs into hPSCs. Quantification of CRISPRi/a component expression in hPSCs. To quantify expression levels of both the dCas9 effector and individual gRNAs, we treated dCas9-KRAB-PB and dCas9-VPR-PB clones with doxycycline for 0, 24, and 48 hours and collected replicate and matched samples for side-by-side clonal anal- ysis via flow cytometry, western blot, and RT-qPCR (Fig. 3a). For CRISPRi, flow cytometric quantification of EGFP fluorescence of the two independent dCas9-KRAB-PB clones showed high levels of EGFP fluorescence (99.9% for both K1 and K2 clones) and mRFP expression fluorescence (100% for both K1 and K2 clones) after 48 hours of doxycycline induction (Fig. 3b, Supplementary Fig. S3) indicating robust and uniform expression of dCas9-KRAB and gRNA. Results a a c gRNA Transcripts (log10 gRNA/GAPDH) CRISPRi gRNAs dCas9-KRAB protein α-Cas9 α-GAPDH −Dox 24h 48h 24h 48h −Dox mRFP+ K1 K2 b dCas9-KRAB-PB clones K1 + K2 dCas9-VPR-PB clones V1 + V2 −Dox +Dox 24h +Dox 48h Flow Cytometry Western Blot RT-qPCR d 10-2 10-1 10-3 10-4 10-5 10-6 −RT −Dox 24h 48h i1 i2 i3 −Dox 24h 48h i1 i2 i3 i1 i2 i3 i1 i2 i3 i1 i2 i3 i1 i2 i3 K1 K2 1 0 20 40 60 80 100 −Dox 24h 48h 24h 48h −Dox %Fluorescent cells EGFP+ Clone K1 mRFP+ EGFP+ Clone K2 0 20 40 60 80 100 %Fluorescent cells −Dox 24h 48h 24h 48h −Dox a dCas9-KRAB-PB clones K1 + K2 dCas9-VPR-PB clones V1 + V2 −Dox +Dox 24h +Dox 48h Flow Cytometry Western Blot RT-qPCR dCas9-KRAB protein α-Cas9 α-GAPDH −Dox 24h 48h 24h 48h −Dox K1 K2 c mRFP+ b d 0 20 40 60 80 100 −Dox 24h 48h 24h 48h −Dox %Fluorescent cells EGFP+ Clone K1 b c d d mRFP+ EGFP+ Clone K2 0 20 40 60 80 100 %Fluorescent cells −Dox 24h 48h 24h 48h −Dox f −Dox 24h 48h 24h 48h −Dox V1 V2 α-Cas9 α-GAPDH dCas9-VPR protein e f 0 20 40 60 80 100 %Fluorescent cells −Dox 24h 48h 24h 48h −Dox mRFP+ EGFP+ Clone V1 e g −RT −Dox 24h 48h a1 a2 a3 −Dox 24h 48h a1 a2 a3 a1 a2 a3 a1 a2 a3 a1 a2 a3 a1 a2 a3 1 10 10-1 10-2 10-3 10-4 10-5 V1 V2 CRISPRa gRNAs gRNA Transcripts (log10 gRNA/GAPDH) 0 20 40 60 80 100 %Fluorescent cells −Dox 24h 48h 24h 48h −Dox mRFP+ EGFP+ Clone V2 g Figure 3. Assessment of CRISPRi and CRISPRa component expression. (a) Experimental overview for Figure 3. Assessment of CRISPRi and CRISPRa component expression. (a) Experimental overview for activation and repression of TCF4 in dCas9-KRAB-PB and dCas9-VPR-PB clones (b) Quantification of flow cytometry analysis of EGFP and mRFP fluorescence in dCas9-KRAB-PB clones K1 and K2 in absence of doxycycline (−Dox) and in presence of doxycycline for 24 and 48 hours. (c) Western blot analysis of dCas9- KRAB protein in dCas9-KRAB-PB clones K1 and K2 at indicated time-points. (d) RT-qPCR analysis of CRISPRi gRNAs i1, i2, and i3 in dCas9-KRAB-PB clones K1 and K2. Results In direct congruence with the EGFP and mRFP fluorescence data, we observe strong Scientific Reports \| (2020) 10:635 \| https://doi.org/10.1038/s41598-020-57500-1 Scientific Reports \| (2020) 10:635 \| https://doi.org/10.1038/s41598-020-57500-1 www.nature.com/scientificreports/ a c e gRNA Transcripts (log10 gRNA/GAPDH) CRISPRi gRNAs f g dCas9-KRAB protein α-Cas9 α-GAPDH −Dox 24h 48h 24h 48h −Dox −Dox 24h 48h 24h 48h −Dox mRFP+ K1 K2 V1 V2 b dCas9-KRAB-PB clones K1 + K2 dCas9-VPR-PB clones V1 + V2 −Dox +Dox 24h +Dox 48h Flow Cytometry Western Blot RT-qPCR d α-Cas9 α-GAPDH 10-2 10-1 10-3 10-4 10-5 10-6 −RT −Dox 24h 48h i1 i2 i3 −Dox 24h 48h i1 i2 i3 i1 i2 i3 i1 i2 i3 i1 i2 i3 i1 i2 i3 K1 K2 1 −RT −Dox 24h 48h a1 a2 a3 −Dox 24h 48h a1 a2 a3 a1 a2 a3 a1 a2 a3 a1 a2 a3 a1 a2 a3 1 10 10-1 10-2 10-3 10-4 10-5 V1 V2 CRISPRa gRNAs gRNA Transcripts (log10 gRNA/GAPDH) 0 20 40 60 80 100 −Dox 24h 48h 24h 48h −Dox dCas9-VPR protein %Fluorescent cells EGFP+ Clone K1 mRFP+ EGFP+ Clone K2 0 20 40 60 80 100 %Fluorescent cells −Dox 24h 48h 24h 48h −Dox 0 20 40 60 80 100 %Fluorescent cells 0 20 40 60 80 100 %Fluorescent cells −Dox 24h 48h 24h 48h −Dox −Dox 24h 48h 24h 48h −Dox mRFP+ EGFP+ Clone V1 mRFP+ EGFP+ Clone V2 Figure 3. Assessment of CRISPRi and CRISPRa component expression. (a) Experimental overview for activation and repression of TCF4 in dCas9-KRAB-PB and dCas9-VPR-PB clones (b) Quantification of flow cytometry analysis of EGFP and mRFP fluorescence in dCas9-KRAB-PB clones K1 and K2 in absence of doxycycline (−Dox) and in presence of doxycycline for 24 and 48 hours. (c) Western blot analysis of dCas9- KRAB protein in dCas9-KRAB-PB clones K1 and K2 at indicated time-points. (d) RT-qPCR analysis of CRISPRi gRNAs i1, i2, and i3 in dCas9-KRAB-PB clones K1 and K2. (e) Quantification of flow cytometry analysis of EGFP and mRFP fluorescence in dCas9-VPR-PB clones V1 and V2 at indicated time-points. (f) Western blot analysis of dCas9-VPR protein level in dCas9-VPR-PB clones at indicated time-points. (g) RT- qPCR analysis of CRISPRa gRNAs a1, a2, and a3 expression in dCas9-VPR-PB clones V1 and V2. Results (e) Quantification of flow cytometry analysis of EGFP and mRFP fluorescence in dCas9-VPR-PB clones V1 and V2 at indicated time-points. (f) Western blot analysis of dCas9-VPR protein level in dCas9-VPR-PB clones at indicated time-points. (g) RT- qPCR analysis of CRISPRa gRNAs a1, a2, and a3 expression in dCas9-VPR-PB clones V1 and V2. Figure 3. Assessment of CRISPRi and CRISPRa component expression. (a) Experimental overview for activation and repression of TCF4 in dCas9-KRAB-PB and dCas9-VPR-PB clones (b) Quantification of flow cytometry analysis of EGFP and mRFP fluorescence in dCas9-KRAB-PB clones K1 and K2 in absence of doxycycline (−Dox) and in presence of doxycycline for 24 and 48 hours. (c) Western blot analysis of dCas9- KRAB protein in dCas9-KRAB-PB clones K1 and K2 at indicated time-points. (d) RT-qPCR analysis of CRISPRi gRNAs i1, i2, and i3 in dCas9-KRAB-PB clones K1 and K2. (e) Quantification of flow cytometry analysis of EGFP and mRFP fluorescence in dCas9-VPR-PB clones V1 and V2 at indicated time-points. (f) Western blot analysis of dCas9-VPR protein level in dCas9-VPR-PB clones at indicated time-points. (g) RT- qPCR analysis of CRISPRa gRNAs a1, a2, and a3 expression in dCas9-VPR-PB clones V1 and V2. induction of dCas9-KRAB protein (Fig. 3c, Supplementary Fig. S3) and high levels of all three CRISPRi gRNAs by gRNA-specific RT-qPCR (Fig. 3d) in both K1 and K2 clones. Indeed, we observed gRNA expression levels between 1% and nearly 100% of the levels of GAPDH transcripts. These results indicate that PB vectors provide a consistent and reproducible means to express multiple gRNAs across cells in a population using a single delivery vehicle. In the case of CRISPRa, flow cytometric quantification of EGFP fluorescence of the two independent dCas9-VPR-PB clones revealed intermediate levels of EGFP fluorescence (82% for clone V1 and 53% for clones V2; Fig. 3e, Supplementary Fig. S3). The decreased levels of EGFP expression in dCas9-VPR-PB clones contrasts Scientific Reports \| (2020) 10:635 \| https://doi.org/10.1038/s41598-020-57500-1 www.nature.com/scientificreports/ with the high levels of EGFP expression in the parental dCas9-VPR clone (Fig. 1c), perhaps indicating a CRISPRa gRNA- or PB-specific effect as prolonged doxycycline treatment in dCas9-VPR cells without integrated PB vectors did not result in reduced EGFP levels (data not shown). However, mRFP expression remained high in dCas9-VPR-PB cells, as quantified by flow cytometry (100% for both clone V1 and V2). Direct assessment of dCas9-VPR protein levels in clones V1 and V2 revealed strong induction upon doxycycline treatment (Fig. Results 3f, Supplementary Fig. S3), albeit to a lesser extent compared with dCas9-KRAB-PB clones. Again, RT-qPCR anal- ysis showed robust and uniform expression of all three CRISPRa gRNAs (Fig. 3g) further confirming the mul- tiplexing capacity of the PB system. Interestingly, the expression levels of CRISPRi and CRISPRa gRNAs i2 and a2 increased 4–6 fold upon expression of dCas9 by doxycycline treatment (Fig. 3d,g). It is possible that the pres- ence of dCas9 selectively increases gRNA stability by binding particular gRNAs with high affinity and protecting them from degradation, perhaps by masking the 5′ end of the gRNA, as suggested by previous studies34. These results demonstrate that both dCas9-effectors and multiplex gRNAs are efficiently expressed in our CRISPRi and CRISPRa hPSC lines. Quantification of TCF4 repression and activation at the transcript and protein level. Having established robust expression of our effector components, we next sought to quantify levels of repression and activation of a target gene in hPSCs. To quantify the efficiency of TCF4 repression by CRISPRi, we first analyzed TCF4 3b transcript levels by RT-qPCR in dCas9-KRAB-PB clonal pairs. As shown in Fig. 4a, doxycycline induc- tion resulted in rapid and significant repression of TCF4 transcripts in both clones K1 and K2 (average decrease of 18-fold at 24 hours and 200-fold at 48 hours). Comparison of TCF4 transcript levels in clone K1 and K2 revealed that clone K2 displayed more rapid repression, suggesting that CRISPRi potency may titrate with PB copy num- ber (Fig. 2e). As expected, TCF4 protein was significantly reduced in both dCas9-KRAB K1 and K2 clones by 48 hours (Fig. 4b, Supplementary Fig. S4). By comparison, targeting of TCF4 for activation in dCas9-VPR-PB clones (V1 and V2) resulted in a 1.8- and 1.6-fold average increase in transcript levels after 24 hours and 48 hours doxycycline treatment, respectively (Fig. 4c). In dCas9-VPR-PB cells TCF4 protein levels were also significantly increased after 48 hours of doxycycline induction (Fig. 4d, Supplementary Fig. S4). Assessing compatibility for single gRNA delivery. To assess the performance of our PB system for delivering single gRNAs, such as those that might be used in library format, we cloned individual PB vectors con- taining one CRISPRi gRNA targeting TCF4, one CRISPRa gRNA targeting TCF4, one CRISPRi gRNA targeting POGZ, or one CRISPRa gRNA targeting POGZ (Fig. 4e). Results We then co-transfected the single-gRNA PB vectors along with a plasmid encoding the piggyBac transposase into the relevant dCas9-KRAB and dCas9-VPR hPSC lines. Following selection with blasticidin and expansion, cells were treated with doxycycline for 0 or 48 hours for western blot analysis (Fig. 4f,g, Supplementary Fig. S4). TCF4 and POGZ protein levels were each significantly reduced using a single gRNA in dCas9-KRAB cells (Fig. 4f,g, Supplementary Fig. S4). In the case of dCas9-VPR cells, TCF4 and POGZ protein levels both trended toward an increase, although the results did not reach statisti- cal significance. The variability in gene modulation within each system is likely due to the relative efficiencies of individual gRNAs. As the single TCF4 gRNAs (Fig. 4f) were also included in the multiplexed system (Fig. 4b,d) these data also indicate that the inclusion of additional gRNAs can lead to additional gene modulation. Discussion CRISPRi/a systems hold great potential for exploring gene function and dissecting human disease mechanisms in hPSCs and hPSC-derived cell types, such as cardiomyocytes12 and neurons17. Benefits of CRISPRi/a over knock- out strategies utilizing Cas9 nuclease include the ability to conditionally perturb essential and multiple genes in the absence of DNA damage and genetic instability35,36. However, in contrast to gene perturbation by gene knockout with CRISPR-Cas9, gene modulation by CRISPRi and CRISPRa approaches are dependent on sustained expression of the dCas9 effector and gRNA. pf g Here, we developed a set of tools to facilitate multiplexed CRISPR-mediated gene modulation in hPSCs. We find that our integrated dCas9-KRAB and dCas9-VPR constructs allow for reproducible and reversible induction of dCas9 alongside EGFP in the vast majority of cells, consistent with previous reports12. To facilitate stable and multiplex gRNA expression, we designed and validated a drug-selectable piggyBac vector with constitutive mRFP fluorescence to visualize and track gRNA-expressing cells. Thus, dual monitoring of both EGFP and mRFP fluo- rescence allows for quantification of the percentage of CRISPRi/a competent cells in a population.i qi p g p p p With regard to gRNA expression, we find that CAG-promoter driven PB vectors support sustained gRNA and reporter expression in hPSCs. Importantly, we confirmed high expression levels of six independently cloned gRNAs by direct assessment via RT-qPCR in the multiplexed system, demonstrating that PB vectors provide a dependable, rapid and inexpensive delivery vehicle for transgene expression. Specifically, we anticipate these vec- tors will be useful for rapid and multiplexed expression of gRNAs in hPSCs for perturbation analysis at the single gene and whole genome levels, both in CRISPRi/a contexts, as presented here, and in CRISPR knockout schemes with Cas9 nuclease or Cas9-fused base editors37. Indeed, our dCas9-VPR construct and PB vectors have been successfully employed in another hPSC line, 18a, to target the XIST gene (Fukuda et al., unpublished results), supporting the reproducibility of our system with additional cell lines and gene targets. pp g p y y g g Collectively, our newly designed gRNA PB vectors deliver robust, sustained gRNA expression in hPSCs. Further, the coupling of these tools with our dual-fluorescence dCas9-KRAB and dCas9-VPR systems facilitates accurate quantification and tracking of CRISPRi/a components across cells in a population. We anticipate these tools will facilitate both single and multiplexed gene perturbation studies and screens in hPSCs and other cell types for functional interrogation of development and disease. Discussion Scientific Reports \| (2020) 10:635 \| https://doi.org/10.1038/s41598-020-57500-1 www.nature.com/scientificreports/ dCas9-KRAB-PB: TCF4 3b transcript c a b d dCas9-KRAB-PB: TCF4 protein −Dox 24h 48h −Dox 24h 48h −RT TCF4 Transcript (relative to GAPDH) 0 0.001 0.002 0.003 0.004 0.005 * ** −Dox 24h 48h K1 K2 Average K1/K2 dCas9-VPR-PB: TCF4 3b transcript Clone K2 dCas9-VPR-PB: TCF4 protein Clone V1 Clone K1 −Dox α-TCF4 α-GAPDH replicate 1 replicate 2 replicate 3 24h 48h −Dox 24h 48h −Dox 24h 48h α-TCF4 replicate 1 replicate 2 replicate 3 α-GAPDH −Dox 24h 48h −Dox 24h 48h −Dox 24h 48h Clone V2 α-TCF4 replicate 1 replicate 2 replicate 3 α-GAPDH −Dox 24h 48h −Dox 24h 48h −Dox 24h 48h TCF4 Transcripts (norm. Discussion to GAPDH) 0 0.001 0.002 0.003 0.004 0.005 0.006 * ** −Dox 24h 48h −Dox 24h 48h −RT −Dox 24h 48h V1 V2 Average V1/V2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0.0 0.2 0.4 0.6 0.8 1.0 1.2 −Dox 24h 48h TCF4 Protein Levels (relative to GAPDH) Clone K1 −Dox 24h 48h TCF4 Protein Levels (relative to GAPDH) 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 −Dox 24h 48h TCF4 Protein Levels (relative to GAPDH) Clone V1 Clone K2 −Dox 24h 48h TCF4 Protein Levels (relative to GAPDH) 0.0 2.0 4.0 6.0 8.0 10.0 12.0 Clone V2 e f * ** * * ** * p = 0.0003 p = 0.0013 p = 0.0082 p = 0.0290 p = 0.0037 p = 0.0681 replicate 1 replicate 2 replicate 3 −Dox 24h 48h −Dox 24h 48h −Dox 24h 48h α-TCF4 α-GAPDH single gRNA: POGZ protein −Dox 48h replicate 1 2 3 −Dox 48h −Dox 48h α-GAPDH α-POGZ POGZ - KRAB single gRNA replicate 1 2 3 −Dox 48h −Dox 48h −Dox 48h α-GAPDH α-POGZ POGZ - VPR single gRNA POGZ - VPR single gRNA CAG CAG FRT mRFP T2A BlastR U6 gRNA ITR attB2 FRT ITR attB1 single gRNA: TCF4 protein replicate 1 2 3 −Dox 48h −Dox 48h −Dox 48h α-GAPDH α-TCF4 piggyBac transposase genomic DNA Blasticidin Selection + CAG CAG ITR FRT mRFP T2A BlastR U6 gRNA attB2 ITR attB1 FRT T2A FRT g TCF4 - KRAB single gRNA TCF4 - VPR single gRNA 0.0 0.2 0.4 0.6 0.8 1.0 1.2 −Dox 48h −Dox 48h 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.6 1.4 POGZ - KRAB single gRNA 0.0 0.2 0.4 0.6 0.8 1.0 1.2 −Dox 48h TCF4 - KRAB single gRNA p = 0.0265 * p = 0.0097 p = 0.3069 p = 0.0074 p = 0.0017 p = 0.0210 p = 0.0086 * p = 0.0903 p = 0.0324 p = 0.0754 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.6 1.4 1.8 2.0 TCF4 - VPR single gRNA −Dox 48h 1 2 3 −Dox 48h −Dox 48h −Dox 48h α-GAPDH α-TCF4 replicate POGZ Protein Levels (relative to GAPDH) TCF4 Protein Levels (relative to GAPDH) POGZ Protein Levels (relative to GAPDH) TCF4 Protein Levels (relative to GAPDH) ** * * Figure 4. Gene repression and activation at the transcript and protein levels. Discussion to GAPDH) 0 0.001 0.002 0.003 0.004 0.005 0.006 * −Dox 24h 48h −Dox 24h 48h −RT −Dox 24h 48h V1 V2 Average V1/V2 p = 0.0210 p = 0.0086 d dCas9-VPR-PB: TCF4 protein Clone V1 α-TCF4 replicate 1 replicate 2 replicate 3 α-GAPDH −Dox 24h 48h −Dox 24h 48h −Dox 24h 48h Clone V2 α-TCF4 replicate 1 replicate 2 replicate 3 α-GAPDH −Dox 24h 48h −Dox 24h 48h −Dox 24h 48h 24h 48h age V1/V2 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 −Dox 24h 48h TCF4 Protein Levels (relative to GAPDH) Clone V1 −Dox 24h 48h TCF4 Protein Levels (relative to GAPDH) 0.0 2.0 4.0 6.0 8.0 10.0 12.0 Clone V2 * * ** p = 0.0082 p = 0.0290 p = 0.0037 p = 0.0681 p = 0.0210 p = 0.0086 * * d dCas9-VPR-PB: TCF4 protein Clone V1 α-TCF4 replicate 1 replicate 2 replicate 3 α-GAPDH −Dox 24h 48h −Dox 24h 48h −Dox 24h 48h Clone V2 α-TCF4 replicate 1 replicate 2 replicate 3 α-GAPDH −Dox 24h 48h −Dox 24h 48h −Dox 24h 48h h V2 d c 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 −Dox 24h 48h TCF4 Protein Levels (relative to GAPDH) Clone V1 −Dox 24h 48h TCF4 Protein Levels (relative to GAPDH) 0.0 2.0 4.0 6.0 8.0 10.0 12.0 Clone V2 * * ** p = 0.0082 p = 0.0290 p = 0.0037 p = 0.0681 * * f single gRNA: TCF4 protein replicate 1 2 3 −Dox 48h −Dox 48h −Dox 48h α-GAPDH α-TCF4 TCF4 - KRAB single gRNA TCF4 - VPR single gRNA 1 2 3 −Dox 48h −Dox 48h −Dox 48h α-GAPDH α-TCF4 replicate f e CAG CAG FRT mRFP T2A BlastR U6 gRNA ITR attB2 FRT ITR attB1 piggyBac transposase genomic DNA Blasticidin Selection + CAG CAG ITR FRT mRFP T2A BlastR U6 gRNA attB2 ITR attB1 FRT T2A FRT e 0.0 0.2 0.4 0.6 0.8 1.0 1.2 −Dox 48h TCF4 - KRAB single gRNA p = 0.0265 * p = 0.0754 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.6 1.4 1.8 2.0 TCF4 - VPR single gRNA −Dox 48h TCF4 Protein Levels (relative to GAPDH) TCF4 Protein Levels (relative to GAPDH) single gRNA: POGZ protein −Dox 48h replicate 1 2 3 −Dox 48h −Dox 48h α-GAPDH α-POGZ POGZ - KRAB single gRNA replicate 1 2 3 −Dox 48h −Dox 48h −Dox 48h α-GAPDH α-POGZ POGZ - VPR single gRNA g g POGZ - VPR single gRNA 0.0 0.2 0.4 0.6 0.8 1.0 1.2 −Dox 48h −Dox 48h 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.6 1.4 POGZ - KRAB single gRNA p = 0.0097 p = 0.3069 * POGZ Protein Levels (relative to GAPDH) POGZ Protein Levels (relative to GAPDH) Figure 4. Discussion dCas9-KRAB-PB: TCF4 3b transcript a b dCas9-KRAB-PB: TCF4 protein −Dox 24h 48h −Dox 24h 48h −RT TCF4 Transcript (relative to GAPDH) 0 0.001 0.002 0.003 0.004 0.005 * ** −Dox 24h 48h K1 K2 Average K1/K2 Clone K2 Clone K1 −Dox α-TCF4 α-GAPDH replicate 1 replicate 2 replicate 3 24h 48h −Dox 24h 48h −Dox 24h 48h 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0.0 0.2 0.4 0.6 0.8 1.0 1.2 −Dox 24h 48h TCF4 Protein Levels (relative to GAPDH) Clone K1 −Dox 24h 48h TCF4 Protein Levels (relative to GAPDH) Clone K2 * ** * p = 0.0003 p = 0.0013 replicate 1 replicate 2 replicate 3 −Dox 24h 48h −Dox 24h 48h −Dox 24h 48h α-TCF4 α-GAPDH p = 0.0074 p = 0.0017 p = 0.0903 p = 0.0324 ** dCas9-KRAB-PB: TCF4 3b transcript a −Dox 24h 48h −Dox 24h 48h −RT TCF4 Transcript (relative to GAPDH) 0 0.001 0.002 0.003 0.004 0.005 * ** −Dox 24h 48h K1 K2 Average K1/K2 p = 0.0074 p = 0.0017 b dCas9-KRAB-PB: TCF4 protein 8h 2 Clone K2 Clone K1 −Dox α-TCF4 α-GAPDH replicate 1 replicate 2 replicate 3 24h 48h −Dox 24h 48h −Dox 24h 48h 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0.0 0.2 0.4 0.6 0.8 1.0 1.2 −Dox 24h 48h TCF4 Protein Levels (relative to GAPDH) Clone K1 −Dox 24h 48h TCF4 Protein Levels (relative to GAPDH) Clone K2 * ** * p = 0.0003 p = 0.0013 replicate 1 replicate 2 replicate 3 −Dox 24h 48h −Dox 24h 48h −Dox 24h 48h α-TCF4 α-GAPDH 7 p = 0.0903 p = 0.0324 ** b a c d dCas9-VPR-PB: TCF4 3b transcript TCF4 Transcripts (norm. Discussion Gene repression and activation at the transcript and protein levels. (a) RT-qPCR analysis of TCF4 3b transcript levels at indicated time points in dCas9-KRAB-PB clones K1 and K2 separately and averaged. Data is shown as mean +/− s.e.m. p = 0.0074, p = 0.0017. (b) Western blot analysis of TCF4 protein in dCas9-KRAB-PB clones K1 and K2 in biological triplicate at indicated time-points with quantification shown on the right. p = 0.0003, p = 0.0013 and p = 0.0324. (c) RT-qPCR analysis of TCF4 3b transcript levels at indicated time points in dCas9-VPR-PB clones V1 and V2 separately and averaged. Data is shown as mean +/− s.e.m. p = 0.0210, *p = 0.0086. (d) Western blot analysis of TCF4 protein in dCas9-VPR-PB clones V1 and V2 in biological triplicate at indicated time-points with quantification shown on the right. p = 0.0037, p = 0.02901, p = 0.0082. (e) Overview of single-gRNA PB vector cloning, delivery, and selection. (f) Western blot analysis of TCF4 protein with single gRNAs in dCas9-KRAB and dCas9-VPR cells in biological triplicate with quantification shown on the right. p = 0.0265. (g) Western blot analysis of POGZ protein with single gRNAs in dCas9-KRAB and dCas9-VPR cells in biological triplicate with quantification shown on the right. p = 0.0097. In all Western blot panels, expression levels are normalized to GAPDH. Figure 4. Gene repression and activation at the transcript and protein levels. (a) RT-qPCR Discussion (a) RT-qPCR analysis of TCF4 3b transcript levels at indicated time points in dCas9-KRAB-PB clones K1 and K2 separately and averaged. Data is shown as mean +/− s.e.m. p = 0.0074, *p = 0.0017. (b) Western blot analysis of TCF4 protein in dCas9-KRAB-PB clones K1 and K2 in biological triplicate at indicated time-points with quantification shown on the right. p = 0.0003, p = 0.0013 and p = 0.0324. (c) RT-qPCR analysis of TCF4 3b transcript levels at indicated time points in dCas9-VPR-PB clones V1 and V2 separately and averaged. Data is shown as mean +/− s.e.m. p = 0.0210, *p = 0.0086. (d) Western blot analysis of TCF4 protein in dCas9-VPR-PB clones V1 and V2 in biological triplicate at indicated time-points with quantification shown on the right. p = 0.0037, p = 0.02901, p = 0.0082. (e) Overview of single-gRNA PB vector cloning, delivery, and selection. (f) Western blot analysis of TCF4 protein with single gRNAs in dCas9-KRAB and dCas9-VPR cells in biological triplicate with quantification shown on the right. p = 0.0265. (g) Western blot analysis of POGZ protein with single gRNAs in dCas9-KRAB and dCas9-VPR cells in biological triplicate with quantification shown on the right. *p = 0.0097. In all Western blot panels, expression levels are normalized to GAPDH. Methods Plasmid construction. To generate the dCas9-KRAB-IRES-EGFP AAVS1 targeting plasmid pT077, paren- tal plasmid pHR-TRE3G-KRAB-dCas9-IRES-GFP (a gift from Jesse Engreitz, Broad Institute) was cloned by Gibson assembly into the backbone fragment of plasmid pGEP116 that contains AAVS1 homology arms and the doxycycline-responsive activator rTTA driven by a constitutive CAG promoter38. To generate the dCas9-VPR- T2A-EGFP AAVS1 targeting plasmid (pT076), the dCas9-VPR cassette from plasmid SP-dCas9-VPR2 (Addgene Scientific Reports \| (2020) 10:635 \| https://doi.org/10.1038/s41598-020-57500-1 www.nature.com/scientificreports/ 63798) was fused by Gibson assembly with a PCR fragment containing a T2A-EGFP-NLS cassette (from plasmid PT059) and cloned into plasmid pGEP116. Oligonucleotides (IDT) corresponding to gRNA target sequences (Supplemental Table S1) were cloned via BpiI into pX330S-2 and pX330S-3 and a third vector pGEP179_pX330K (this study) according to kit instructions33 (Addgene Kit#1000000055). The pGEP179_pX330K plasmid is a mod- ified entry vector generated by cloning the BsaI-pU6-sgRNA-BsaI fragment from pX330A-1 × 333 into a slightly modified MCS of the attL-containing entry vector, pENTR1A (Invitrogen). These gRNA-containing pX330S and pGEP179_pX330K plasmids were then assembly by Golden Gate cloning to form a single entry vector. This entry vector was then cloned by Gateway cloning into the piggyBac donor destination plasmid pGEP163 that contains piggyBac ITRs for transposase-mediated insertion, a CAG promoter driving an mRFP-T2A-BLASTR cassette, and attR sites Gateway cloning to create CRISPRi multi-gRNA plasmid pPN441 and CRISPRa multi-gRNA plasmid pPN440. Donor plasmid pGEP163 was constructed by fusing a fragment of plasmid PB-CA39 (Addgene 20960) containing the piggyBac ITRs and a CAG promoter with a synthetic gene block containing FRT-mRFP-T2A- BLASTR-SV40 pA-FRT (IDT). The versions of pPN441 and pPN440 plasmids used in this study were initially created by an earlier cloning strategy that was replaced by the strategy described above to generate the same pPN441 and pPN440 plasmids more rapidly. BamHI digests of pPN441 and pPN440 were performed according to manufacturer’s instructions (New England Biolabs). Single gRNAs were cloned into pGEP179_pX330K and this entry vector was then cloned by Gateway cloning into the piggyBac donor destination plasmid pGEP163, as described above, for generation of the final pPN458 (TCF4 VPR), pPN459 (TCF4 KRAB), pPN460 (POGZ VPR) and pPN461 (POGZ KRAB) constructs. Cell culture and gene targeting. The human embryonic stem cell line H140 (WA01) was obtained com- mercially from WiCell Research Institute (Madison, WI). The use of H1 was approved by Harvard University’s ESCRO committee and methods were carried out in accordance with approved guidelines. Methods Stem cells were grown in either mTeSR1 medium (Stem Cell Technologies 05850) or StemFlex medium (ThermoFisher A3349401) on Geltrex (Life Technologies A1413301) coated plates under conditions previously described41. Throughout cultur- ing, cells were tested to confirm the absence of mycoplasma contamination (Lonza MycoAlert LT07-418). To inte- grate the dCas9-KRAB and dCas9-VPR constructs into the AAVS1 locus, 2.5 × 106 cells were co-transfected with 10ug of pT077 (KRAB) or pT076 (VPR), 1.5 µg AAVS1 TALEN L (Addgene 59025) and 1.5 µg AAVS1 TALEN R (Addgene 59026) via the Neon Electroporation System (ThermoFisher) at 1050 mV, 30 ms, 2 pulses. For the first round of clonal selection, the transfected cells were plated at low-density (8,000 cells in a 10 cm dish) under G418 selection (50 ug/ml, Gibco 10131035) to allow for single-cell colony formation (~10 days). Importantly, cells with the dCas9-KRAB and dCas9-VPR cassettes are kept under selection with G418 for the duration of culture and experiments to protect against shutdown of the AAVS1 integrated transgenes. In this strategy, colonies are picked and deposited into a 96-well plate and when sufficiently dense, the 96-well plate is triplicated to create 3 plates of identical clones. Plate 1 is frozen for storage, plate 2 is treated with doxycycline (Sigma, D9891-25g) at a final concentration of 2 µg/ml 24 hours after duplication for visualization of EGFP + colonies (with high levels of EGFP expression serving as a proxy for high dCas9 expression), and plate 3 is maintained for expansion and banking of EGFP + colonies (n = 6) while the analysis of plate 2 is performed. For integration of the multiplex PB vectors, 2.5 × 106 dCas9-KRAB and dCas9-VPR cells were transfected with 5 µg of pPN441 (CRISPRi multi-gRNA plas- mid) and 5 µg of pPN440 (CRISPRa multi-gRNA plasmid), respectively, with 1 µg of transposase plasmid (System Biosciences #PB210PA-1) under conditions described above. 24 hours after transfection, cells are treated with blasticidin at a final concentration of 2 µg/ml for 12–15 days to select for positive piggyBac integrants and allow clearing of free plasmid. Genomic DNA for PCR-based genotyping and piggyBac copy number analysis by ddPCR was isolated via the DNeasy Blood and Tissue Kit (Qiagen 69504). For doxycycline induction of dCas9-KRAB and dCas9-VPR, cells are treated with 2 µg/ml doxycycline and pelleted at indicated time points. Western blot analysis. Methods Droplets were generated using a QX100 Droplet Generator and PCR was per- formed on a C1000 Touch thermal-cycler (Bio-Rad) followed by sample streaming onto a QX100 Droplet Reader (Bio-Rad). Quantification was performed with QuantaSoft software. Primer sequences are listed in Table S1. RT-qPCR. Total RNA from hPSCs was extracted using an RNeasy Mini Kit (Qiagen). Reverse Transcription cDNA synthesis reactions were performed on 0.2–2 μg total RNA with iScript cDNA synthesis kit (BioRad) according to manufacturer’s instructions. Quantitative PCR reactions were performed the iTaq Universal SYBR Green Supermix (BioRad) and quantified by the ΔΔcT method on a CFX384 Real-Time System (Bio-Rad). All primers that yielded detectable PCR product had amplification efficiencies between 95–110%. Primer sequences are listed in Table S1. Embryoid body differentiation and immunostaining. Embryoid bodies (EBs) were generated as pre- viously described41. For immunostaining, hPSC colonies and EBs were fixed with 4% paraformaldehyde in PBS for 15 mins at room temperature (RT), blocked and permeabilized with 0.1% TritonX-100 and 4% serum in PBS for 1 hr at RT and incubated with the appropriate primary antibody at RT. Following primary antibody incuba- tion, cells were washed with PBS and incubated with the appropriate secondary antibody (Alexa Fluor 488 or 594, 1:500, Invitrogen) for 1 hr. Cells were then washed with PBS and incubated with DAPI before imaging at 20X magnification. The following primary antibodies were used: OCT4 (R&D Systems AF1759; 1:250), SSEA-4 (SCBT SC21704; 1:250), TRA-1-60 (SCBT SC21705; 1:200), AFP (Sigma A8452; 1:250), SMA (Sigma A2547; 1:2000), β-III Tubulin (R&D Systems MAB1195; 1:3000). Methods To isolate protein for western blot analysis, hPSCs were lysed using Pierce IP lysis buffer (Life Technologies 87787) with protease inhibitors (Sigma Aldrich 11836153001). 20 μg total pro- tein, as determined by Pierce BCA Protein Assay kit (Thermo Scientific 23227), was loaded onto Bolt 4–12% NuPAGE Bis-Tris Plus gels (Invitrogen). Gels were transferred overnight at 4 °C to nitrocellulose membranes in 1X NuPAGE transfer buffer (Invitrogen) with 10% methanol. The following antibodies were used for western blot analysis: Cas9 (Diagenode C15310258, 1:1000) TCF4 (Abcam ab217668, 1:500), POGZ (Abcam ab167408, 1:1000), GAPDH (EMD MAB374; 1:2000), α-rabbit HRP-linked F(ab’)2 (GE Life Sciences NA9340; 1:5000) and α-mouse HRP-linked F(ab’)2 (GE Life Sciences NA9310; 1:5000). Blots were visualized by chemiluminescence with the SuperSignal West Femto kit (Pierce) and imaged and quantified with a ChemiDoc MP Imaging System (BioRad). For quantification of Cas9 protein in dCas9-KRAB and dCas9-VPR parental clones, 1.2 μg total protein was analysed with Cas9 (Diagenode C15310258, 1:400) and GAPDH (EMD MAB374; 1:50) antibodies using the Wes capillary immunoassay system (ProteinSimple). Flow cytometry. Flow cytometry was performed at the Broad Institute Flow Facility on a CytoFLEX flow cytometer (Beckman Coulter). Cells were treated with 10 mM ROCK inhibitor (Y-27632) for 4 to 6 hours prior to analysis. For each experiment, 100,000 events were recorded and analyzed with FCS Express 6 software (De Novo Software). Genomic DNA isolation and genotyping PCR and ddPCR. Genomic DNA (gDNA) was extracted from hPSCs with the DNeasy Blood and Tissue kit according to manufacturer’s instructions (Qiagen). For gen- otyping of WT AAVS1 in dCas9-KRAB and dCas9-VPR clones, PCR of gDNA was performed with primer pair GE222 and GE668. For genotyping of gene targeted AAVS1 in dCas9-KRAB cells, PCR was performed with primer pair GE222 and GE586 for 5′ junctions and primer pair GE819 and GE668 for 3′ junctions. For genotyping Scientific Reports \| (2020) 10:635 \| https://doi.org/10.1038/s41598-020-57500-1 www.nature.com/scientificreports/ gene targeted AAVS1 in dCas9-VPR cells, PCR was performed with primer pair GE222 and GE332 for 5′ junc- tions and primer pair GE233 and GE668 for 3′ junctions. For ddPCR of gDNA to quantify piggyBac copy number in dCas9-KRAB-PB and dCas9-VPR-PB clones, 20 μl reactions were prepared with ddPCR Supermix for Probes (no dUTP) (Bio-Rad, #1863024) with probes specific to mRFP and control gene RPP30 according to manufac- turer’s instructions (Bio-Rad). Data availability y All plasmids generated in this study including all-in-one dCas9-KRAB and dCas9-VPR targeting plasmids and multiplexed PB gRNA delivery systems will be deposited in Addgene.org. Cell lines will be made available upon request with appropriate institution approvals and following WiCell requirements for cell line distribution. Received: 22 July 2019; Accepted: 30 December 2019; Published: xx xx xxxx Published: xx xx xxxx Scientific Reports \| (2020) 10:635 \| https://doi.org/10.1038/s41598-020-57500-1 Referencesh 1. Adli, M. The CRISPR tool kit for genome editing and beyond. Nature Communications 1–13, https://doi.org/10.1038/s41467-018- 04252-2 (2018).fi 1. Adli, M. The CRISPR tool kit for genome editing and beyond. Nature Communications 1–13, https://doi.org/10.1038/s41467 04252-2 (2018).fi 2. Chavez, A. et al. Highly efficient Cas9-mediated transcriptional programming. Nat Meth 12, 326–328 (2015). fi 3. Gilbert, L. A. et al. Genome-Scale CRISPR-Mediated Control of Gene Repression and Activation. Cell 159, 647–661 (2014). l b d d ll 4. Aguirre, A. J. et al. Genomic Copy Number Dictates a Gene-Independent Cell Response to CRISPR/Cas9 Targeting. Cance Discovery 6, 914–929 (2016). 5. Kearns, N. A. et al. Cas9 effector-mediated regulation of transcription and differentiation in human pluripotent stem cells Development 141, 219–223 (2013). p 6. Weltner, J. et al. Human pluripotent reprogramming with CRISPR activators. Nature Communications, 1–12, https://doi org/10.1038/s41467-018-05067-x (2018). g 7. Liu, Y. et al. CRISPR Activation Screens Systematically Identify Factors that Drive Neuronal Fate and Reprogramming. Stem Cell 1–23, https://doi.org/10.1016/j.stem.2018.09.003 (2018). p g j 8. Zhou, H. et al. In vivo simultaneous transcriptional activation of multiple genes in the brain using CRISPR–dCas9- activator transgenic mice. Nature Neuroscience 1–13, https://doi.org/10.1038/s41593-017-0060-6 (2018). 9. Fulco, C. P. et al. Systematic mapping of functional enhancer–promoter connections with CRISPR interference. Science 354 769–773 (2016). ( ) 10. Jost, M. et al. Combined CRISPRi/a-Based Chemical Genetic Screens Reveal that Rigosertib Is a Microtubule- Destabilizing Agent. Molecular Cell 68, 210–223.e6 (2017). 0. Jost, M. et al. Combined CRISPRi/a-Based Chemical Genetic Screens Reveal that Rigosertib Is a Microtubule- Destabilizing Agent Molecular Cell 68, 210–223.e6 (2017). 1 Horlbeck M A et al Mapping the Genetic Landscape of Human Cells Cell 1 38 https://doi org/10 1016/j cell 2018 06 010 (2018) 12. Mandegar, M. A. et al. CRISPR Interference Efficiently Induces Specific and Reversible Gene Silencing in Human iPSCs. Stem Cell 1–42, https://doi.org/10.1016/j.stem.2016.01.022 (2016). fi 1–42, https://doi.org/10.1016/j.stem.2016.01.022 (2016). p g j 3. Bertero, A. et al. Optimized inducible shRNA and CRISPR/Cas9 platforms for in vitrostudies of human development using hPSCs Development 143, 4405–4418 (2016). p ( ) 14. Heman-Ackah, S. M., Bassett, A. R. & Wood, M. J. A. Precision Modulation of Neurodegenerative Disease- Related Gene Expression in Human iPSC-Derived Neurons. Sci Rep 1–12, https://doi.org/10.1038/srep28420 (2016). ed Neurons. Sci Rep 1–12, https://doi.org/10.1038/srep28420 (2016 p p g p ( ) 15. Balboa, D. et al. Conditionally Stabilized dCas9 Activator for Controlling Gene Expression in Human Cell Reprogramming and Differentiation. Stem Cell Reports 5, 448–459 (2015). y Differentiation. Stem Cell Reports 5, 448–459 (2015). f p 6. González, F. et al. An iCRISPR Platform for Rapid, Multiplexable, and Inducible Genome Editing in Human Pluripotent Stem Cells Cell Stem Cell 15, 215–226 (2014). 17. Ho, S.-M. et al. Evaluating Synthetic Activation and Repression of Neuropsychiatric-Related Genes in hiPSC-Derived NPCs, Neurons, and Astrocytes. Stem Cell Reports 9, 615–628 (2017). y p 18. Li, S., Zhang, A., Xue, H., Li, D. & Liu, Y. One-Step piggyBac Transposon-Based CRISPR/Cas9 Activation of Multiple Genes. Molecular Therapy: Nucleic Acid 8, 64–76 (2017).fi h py 9. Wang, G. et al. Efficient, footprint-free human iPSC genome editing by consolidation of Cas9/CRISPR and piggyBac technologies Nat Protoc 12, 88–103 (2016). 20. Schertzer, M. D. et al. A piggyBac-based toolkit for inducible genome editing in mammalian cells. bioRxiv 1–29, https://doi. org/10.1101/448803 (2018). g ( ) 1. Gilbert, L. A. et al. CRISPR-Mediated Modular RNA-Guided Regulation of Transcription in Eukaryotes. Cell 154, 442–451 (2013) 2 Thomson J A et al Embryonic stem cell lines derived from human blastocysts Science 282 1145 1147 (1998) g ( ) 21. Gilbert, L. A. et al. CRISPR-Mediated Modular RNA-Guided Regulation of Transcription in Eukaryotes. Cell 154, 442–45 21. Gilbert, L. A. et al. CRISPR-Mediated Modular RNA-Guided Regulation of Transcription in Eukaryotes. Cell 154, 442–451 (2013). 22. Thomson, J. A. et al. Embryonic stem cell lines derived from human blastocysts. Science 282, 1145–1147 (1998).f g p y ( ) 22. Thomson, J. A. et al. Embryonic stem cell lines derived from human blastocysts. Science 282, 1145–1147 (1998).f Thomson, J. A. et al. Embryonic stem cell lines derived from human h , J y y , ( ) 3. Zufferey, R., Donello, J. E., Trono, D. & Hope, T. J. Woodchuck Hepatitis Virus Posttranscriptional Regulatory Element Enhance Expression of Transgenes Delivered by Retroviral Vectors. J. Virol. fi 1–42, https://doi.org/10.1016/j.stem.2016.01.022 (2016). Sci Rep 4, 2–6 (2014). y p 34. Jiang, F. & Doudna, J. A. CRISPR–Cas9 Structures and Mechanisms. Annu. Rev. Biophys. 46, 505–529 (2017). 35. Ihry, R. J. et al. p53 inhibits CRISPR–Cas9 engineering in human pluripotent stem cells. Nature Medicine 1–16, https:/ org/10.1038/s41591-018-0050-6 (2018). 36. Kosicki, M., Tomberg, K. & Bradley, A. Repair of double-strand breaks induced by CRISPR–Cas9 leads to large deletions and complex rearrangements. Nature Publishing Group 1–10, https://doi.org/10.1038/nbt.4192 (2018).fi 7. Billon, P. et al. CRISPR-Mediated Base Editing Enables Efficient Disruption of Eukaryotic Genes through Induction of STOP Codons. Molecular Cell 67, 1068–1079.e4 (2017). 38. Sellgren, C. M. et al. Increased synapse elimination by microglia in schizophrenia patient-derived models of synaptic pruning. Nature Neuroscience 1–19, https://doi.org/10.1038/s41593-018-0334-7 (2019).i p g ( ) 39. Woltjen, K. et al. piggyBac transposition reprograms fibroblasts to induced pluripotent stem cells. Nature 458, 766 EP – j p ggy p p gi p p 40. Thomson, J. A. Embryonic Stem Cell Lines Derived from Human Blastocysts. Science 282, 1145–1147 (1998). h 1. Hazelbaker, D. Z. et al. A Scaled Framework for CRISPR Editing of Human Pluripotent Stem Cells to Study Psychiatric Disease. Stem Cell Reports 9, 1315–1327 (2017). Competing interestsh p g The authors declare no competing interests. fi 1–42, https://doi.org/10.1016/j.stem.2016.01.022 (2016). 73, 2886 (1999). 9 www.nature.com/scientificreports/ 24. Jung, M. et al. Analysis of the expression pattern of the schizophrenia-risk and intellectual disability gene TCF4 in the developing and adult brain suggests a role in development and plasticity of cortical and hippocampal neurons. 1–15, https://doi.org/10.1186/ s13229-018-0200-1 (2018). ( ) 25. Quednow, B. B., Brzózka, M. M. & Rossner, M. J. Transcription factor 4 (TCF4) and schizophrenia: integrating the animal an human perspective. Cellular and Molecular Life Sciences 71, 2815–2835 (2014). p p f , ( ) 26. Ripke, S. et al. Biological insights from 108 schizophrenia-associated genetic loci. Nature 511, 421–427 (2014). p p f ( ) 26. Ripke, S. et al. Biological insights from 108 schizophrenia-associated genetic loci. Nature 511, 421–427 (2014). 27. Sepp, M., Kannike, K., Eesmaa, A., Urb, M. & Timmusk, T. Fun 27. Sepp, M., Kannike, K., Eesmaa, A., Urb, M. & Timmusk, T. Functional Diversity of Human Basic Helix-Loo Factor TCF4 Isoforms Generated by Alternative 5′ Exon Usage and Splicing. PLoS ONE 6, e22138–14 (2011) pp y p Factor TCF4 Isoforms Generated by Alternative 5′ Exon Usage and Splicing. PLoS ONE 6, e22138–14 (2011).fi y g p g , ( ) 28. Radzisheuskaya, A., Shlyueva, D., Müller, I. & Helin, K. Optimizing sgRNA position markedly improves the efficiency of CRISPR/ dCas9 mediated transcriptional repression Nucleic Acids Res 44 e141 e141 (2016) y g p g 28. Radzisheuskaya, A., Shlyueva, D., Müller, I. & Helin, K. Optimizing sgRNA position markedly improves the efficiency of CRISPR/ 28. Radzisheuskaya, A., Shlyueva, D., Müller, I. & Helin, K. Optimizing sgRNA position markedly improves the efficien dCas9-mediated transcriptional repression Nucleic Acids Res 44 e141–e141 (2016) p p 29. Doench, J. G. Am I ready for CRISPR? A user’s guide to genetic screens. Nature Publishing Group 19, 67–80 (2017). l h d l f d d d 30. Liu, H. et al. CRISPR-ERA: a comprehensive design tool for CRISPR-mediated gene editing, repression and activation. Bioinformatics 31, 3676–3678 (2015). f 31. Jusiak, B., Cleto, S., Perez-Piñera, P. & Lu, T. K. Engineering Synthetic Gene Circuits in Living Cells with CRISPR Technology. Trends in Biotechnology 34, 535–547 (2016). gy 2. Chen, Y.-T. et al. PiggyBacTransposon-Mediated, Reversible Gene Transfer in Human Embryonic Stem Cells. Stem Cells and Development 19, 763–771 (2010). p 33. Sakuma, T., Nishikawa, A., Kume, S., Chayama, K. & Yamamoto, T. Multiplex genome engineering in human cells using all-in-one CRISPR/Cas9 vector system. Author contributions D.Z.H., K.E. and L.E.B. conceived and designed the study. A.B., G.A., P.M., A.M., D.L. and D.Z.H. performed the xperiments and data analysis. D.Z.H. and L.E.B. wrote the manuscript with input from all coauthors. Acknowledgementsh g This project was funded by the Stanley Center for Psychiatric Research at the Broad Institute. We thank Robert Ihry, Katie Worringer, Ajamete Kaykas (Novartis), Jesse Engreitz (Broad Institute), and Alejandro Chavez (Columbia University) for sharing of plasmids and cloning suggestions. We thank Tõnis Timmusk (Tallinn University of Technology) for advice on TCF4 isoform identification. We thank members of the Barrett laboratory for advice and suggestions. We thank the Broad Institute Flow Facility for experimental support. Author contributions © The Author(s) 2020 Additional information Supplementary information is available for this paper at https://doi.org/10.1038/s41598-020-57500-1. Supplementary information is available for this paper at https://doi.org/10.1038/s41598-020-57500-1. Supplementary information is available for this paper at https://doi.org/10.1038/s41598-020-57500-1. Correspondence and requests for materials should be addressed to L.E.B. Reprints and permissions information is available at www.nature.com/reprints. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2020 Scientific Reports \| (2020) 10:635 \| https://doi.org/10.1038/s41598-020-57500-1 10
https://openalex.org/W4388731394	https://www.e3s-conferences.org/articles/e3sconf/pdf/2023/86/e3sconf_pdsed2023_01004.pdf	English	null	Entropy approach for managing the socio-economic development of the region	E3S web of conferences	2,023	cc-by	5,782	1 Introduction In the age of the economic globalization, the management of the socio-economic development of a region or territory becomes even more pressing because, in addition to meeting social objectives, they are aimed to preserve the integrity and identity of the country, region and territory. Anymore, there are many theories and approaches to regional development [1] and their contradictions and divergence are obvious. Therefore, there is a basic urge to develop an integrated theory of regional development. Attempts to develop an integrated theory of regional development are already underway, as an example, the article by E. Medeiros [2], wherein he tried to develop a “theory of everything” for regional development, which would summarize all actual approaches for managing the socio- economic development of the region. Entropy approach for managing the socio- economic development of the region V.F. Islamutdinov1,2,* 1Urgench State University, Urgench, Uzbekistan 2Kazan Innovative University named after V.G. Timiryasov, Kazan, Russia Abstract. The article is focused on actual theories and approaches for managing the socio-economic development of the region. And also strengths and weaknesses of the actual theories and approaches of regional development are shown. The article studies the specifics of regional development in Europe, the USA, the Russian Federation and the countries of Central Asia. Entropy approach to study the economic phenomena and processes is offered as an integrated one. It is shown that all actual approaches to regional development can be explained under the entropy approach as special cases. Opportunities for practical application of entropy approach for managing the socio-economic development of the region are shown. As well as the strengths and weaknesses of the entropy approach and the opportunities for its further development are pointed out * Corresponding author: isvad74@gmail.com E3S Web of Conferences 449, 01004 (2023) PDSED 2023 E3S Web of Conferences 449, 01004 (2023) PDSED 2023 https://doi.org/10.1051/e3sconf/202344901004 © The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0 (https://creativecommons.org/licenses/by/4.0/). Actual management theories of socio-economic development of the region All theories of regional development can be formed into two large groups: economic-centric and semi-economic. Economic-centric theories of socio-economic development of the region give priority to endogenous economic factors of development of the region (Table 1). * Corresponding author: isvad74@gmail.com © The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0 (https://creativecommons.org/licenses/by/4.0/). https://doi.org/10.1051/e3sconf/202344901004 E3S Web of Conferences 449, 01004 (2023) PDSED 2023 Table 1. Economic-centric theories of socio-economic development of the region Theoretical paradigm Research objectives (problems) Reasons/Development factors Theory of firm/ production location Optimal location alternatives for production facilities Regional development depends significantly on firms / transport costs in the region Traditional neoclassical theory Finding out endogenous and exogenous factors of economic development of the region Regional development is mainly influenced by factors related to productive capacity, including free trade between regions, perfect competition and information, exogenously determined technological progress and balanced growth path leading to approach of growth rates between regions / security and productivity (“Borts and Stein model” and “theory of factor price alignment”) Keynesian theory Finding out the exogenous factors of economic development of the region Regional development is largely driven by demand / production of goods and services (“export base theory” and “input-output theory”) Theory of functional development Finding out the regional development management tools Regional development can be achieved by using selected regional resources to create generative growth. Actual management theories of socio-economic development of the region Thus, it is possible to relocate the region to higher steps of development around a core function closely linked to the investment / funds from leading regions, reduced imports of goods and services and reinvestment at the local level Stadial theory Finding out the steps of economic development and its impact on progress rates of this development Regional development gets through five steps: (1) traditional, (2) opportunity for rise, (3) rise, (4) maturity and (5) mass consumption / technology, investment flows, physical infrastructure, and skilled labor Non-equilibrium theory Finding out the dynamic development factors of the region Regional development is created by unbalanced forces / production structure, technology, infrastructure, market dynamics (“dynamic Schumpeter disbalance”, “regional life cycle theory” and “product life cycle theory”) Cumulative growth theory Explanation for unbalanced regional growth The main driver of economic growth is accumulated growth experience and base effect Regional growth theory Determination of the economic quantity and economy structure of the region Progress of economic growth depends on balanced and optimal structure of the region’s economy ED 2023 2 E3S Web of Conferences 449, 01004 (2023) PDSED 2023 https://doi.org/10.1051/e3sconf/202344901004 Neoclassical endogenous growth theory Finding out mainly endogenous factors of economic development of the region Regional development depends not only on long-term regional growth through capital and labor accumulation, but also on endogenous economic factors / human capital, Research and Development (effort), innovation, extension and technology Source: made up on basis of [1, P.125; 2, P.5; 3] Regional development depends not only on long-term regional growth through capital and labor accumulation, but also on endogenous economic factors / human capital, Research and Development (effort), innovation, extension and technology Regional development depends not only on long-term regional growth through capital and labor accumulation, but also on endogenous economic factors / human capital, Research and Development (effort), innovation, extension and technology Source: made up on basis of [1, P.125; 2, P.5; 3] Economic-centric growth theory attempts to explain the growth and development of the region’s economy just by economic conditions and factors: location of economic activity, trade balance, resource availability, capital, structure of economy, etc. Advantages of economic-centric theory includes: y - simplicity and comprehensibility for most economists; - simplicity and comprehensibility for most economists; - econometric data and calculations; - open and easily explained causal relationships. Actual management theories of socio-economic development of the region Its weaknesses: - limitations of the conditions and factors of regional development considered; - failure to explain several regional development phenomena and imbalances. limitations of the conditions and factors of regional development considered; - failure to explain several regional development phenomena and imbalances. Semi-economic theory of socio-economic development of the region attempts to link economic development with various non-economic factors (Table 2). Table 2. Semi-economic theories of socio-economic development of the region Table 2. Semi-economic theories of socio-economic development of the region Theoretical paradigm Research objectives (problems) Reasons/Development factors Innovation diffusion theory Addressing the emergence and diffusion of innovation Development and introduction of innovation, process of its diffusion in the region are discussed as the main factor of development of the region Core-periphery theory Finding out the impact of territorial heterogeneity on economic development Regional development is inherently uneven, as it implies under the advanced (leading) regions and less-developed (lagging) regions / resource endowments (“cumulative causation theory”, “pole/growth center theory” and “central location theory”). Development axis theory Determination of transport routes connecting the most important industrial centers Transport connectivity of the region and transport costs are the main drivers of development Infrastructure theory Determining the required level of infrastructure development The development of the region is defined by sophistication of all types of infrastructure: market, logistics, energy, transport, etc. New Economic Geography Theory / Agglomeration Theory Finding out the impact of migration processes on economic development Regional development depends on a balance between centripetal (agglomerate) and centrifugal (diffusive) forces determining the degree and form of regional concentration of economic activity 3 E3S Web of Conferences 449, 01004 (2023) PDSED 2023 https://doi.org/10.1051/e3sconf/202344901004 Complex theory Connecting the unified theory and approach of location and regional problems of economy The rate and quality of economic growth are determined by a combination of factors, both economic and non-economic: institutional, behavioural, etc. Source: made up on basis of [1, P. 125; 2, P. 5; 3] Advantages of semi-economic theories of regional development: - finding out the full range of factors affecting the growth and development of the region’s economy; - attempts to explain phenomena and inconsistencies of economic-centric theory. Its weaknesses: - arbitrariness in non-economic factors option affecting the economic development of the region; - imbalance and contradiction of these theories. - imbalance and contradiction of these theories. 2.2 Main approaches to the management of socio-economic development of the region Theory generalization of economic development of the region allowed to find out the main approaches to the management of socio-economic development of the region (Table 3). Table 3. Synthesis of actual approaches and models of socio-economic development of the reg Approach Approach content Development models in the approach Infrastructural development The first thing we have to do for the economic development of the region is to create the infrastructure base Infrastructure Telecommunications Socio-cultural services Productive capital Accessibility features Balanced development The main condition for the growth of the region’s economy is a balanced growth and development of industries and areas of the economy Urban polycentrism Territorial cohesion Policy integration Spatial planning Spatial connectivity Cluster development Inclusive development / Community development To develop the region under globalization, the high priority is not to lose local advantages, but to try to develop them Smart Specialization / Using Comparative Advantages Balanced bottom-up / top- down approach Collaborative approaches Community led enterprise Endogenous assets Knowledge-based development Under the formation of an innovative economy, the most important resource for the development of the region is knowledge- based assets Human capital assets Innovation Entrepreneurship Business environment Intellectual capital 4 E3S Web of Conferences 449, 01004 (2023) PDSED 2023 https://doi.org/10.1051/e3sconf/202344901004 Institutional development First of all, for the economic development of the region, it is necessary to create institutional assumptions, i.e. “rules of the game” Administrative capacity Institutional capacity / density and change Social capital Territorial cooperation Leadership Sustainable development The main condition for the development of the region’s economy is balanced development of economic, social and environmental components Closed-loop economy Social environmental consciousness Environmental protection Global governance sustainability Sustainability of global spatial planning Source: made up on basis of [2, P. 14] Source: made up on basis of [2, P. 14] Indeed, the diversity of approaches to managing the socio-economic development of the region cannot be fully reduced to the above approaches. But anymore these approaches and models are most often used in regional development practices. The strengths and weaknesses of actual approaches to managing the socio-economic development of the region are shown in Table 4. Table 4. 2.3 Regional specificity While there are generally accepted approaches to the management of socio-economic development of the regions’ economies, it should be noted as part of management impacts for development of socio-economic system of regions must be considered the specifics of regional development in the various macroregions of the planet Earth. Generalization of the planetary experience of regional development was made in a monograph [4]. The features of regional policy in different countries by type and level of development, the experience of which for various reasons is interesting for Russia are focused on this summary activity. Special priority should be given to such issues as the motives of regional policy, its legislative and organizational support, tools. Interaction of regional policy and other components of state regulation of economy, including inter-budgetary relations, regional (territorial) planning, various directions of sectoral policy was studied. Today, the specifics of regional development of European countries are described in more details since actual regional economy as a science originated in Europe. For example, the paper [5] shows a method for disaggregating national trade data in Europe at the regional level. This makes it possible to develop trade indicators at the regional level, including the identified export strengths. Identification of industries in which the region performs a powerful line of business does double service in the development of industrial and regional policies. First, the identification of successful structures at the industry and regional levels helps to better get micro- and meso-competitiveness, as well as the scope and cases of political interference. Secondly, knowledge of the spatial distribution of competitive industries and the required location factors is necessary for a differentiated view of future economic development and choice of policy instruments. The article [6] studies the influence to form the Single European Market on local development. The main research issue was whether some parts of Europe have improved their position in the internal trade of the EU through better use of their comparative advantages (productivity, factor endowment...) and scale economy. g (p y ) y The paper [7] reviews and provides examples of how local development in areas or regions of advanced industrial countries and countries in transition is addressing the negative impacts of globalization. 2.2 Main approaches to the management of socio-economic development of the region The strengths and weaknesses of actual approaches to managing the socio-economic development of the region Approach Strengths Weaknesses Infrastructural development It is logical to start with the material and technical base of economic development Good infrastructure does not guarantee successful development Balanced development It avoids distortions in the development of the region It is inapplicable under the catching-up development because distortions are unavoidable Inclusive development / Community development Caring for local resources and communities If there are no comparative strengths in the region, it is not applicable Knowledge-based development It allows the development of the region even at a high base It is inapplicable under the catching-up development, because there is no required scientific and technical base Institutional development Starting with the basic economic conditions of development Not applicable at a very low initial income level of the population, because the development of the institutional environment takes place only in a co-evolution with the development of the economy Sustainable development Environmental accounting and social management When catching-up development is weak, because insufficient economic resources to solve all problems simultaneously, you need to choose priorities Source: by author based on research Table 4. The strengths and weaknesses of actual approaches to managing the socio-economic development of the region As shown in Table 4, each of the approaches to the socio-economic development of the region has its own strengths and weaknesses, depending on the type of region to which these approaches are applied. Some of the current approaches are weak in catching-up development regions. 5 https://doi.org/10.1051/e3sconf/202344901004 E3S Web of Conferences 449, 01004 (2023) PDSED 2023 E3S Web of Conferences 449, 01004 (2023) PDSED 2023 E3S Web of Conferences 449, 01004 (2023) PDSED 2023 https://doi.org/10.1051/e3sconf/202344901004 The articles [11, 12, 13] found out and generalized the factors of regional development of Russia in 1998-2014 based on an analysis of successful regional examples and econometric methods. In general, development (economic productivity) is determined by natural conditions: advantageous geographical location, raw materials and agro-climatic resources but realizing the potential also depended on the effectiveness of government policies to reduce investment risks and use human capital. p p The specificity of the Central Asian regional development is most underexplored, but progress is evident. For example, this paper [14] touches on the administrative and territorial division of the Republic of Kazakhstan, provides a comparison by size of the regions, and presents features of their development. The authors focus on the integrated solution of regional development problems. As a research problem, the authors attempted to assess and systematize the features of regional development in Kazakhstan. Characteristics of socio- economic development of regions are studied and given. Taxonomic levels of research and scientists view, which made a great contribution to the development of theoretical concepts of regional science, were analyzed. The article proves the priority and effectiveness of the application model “smart specialization” to the development process of Kazakhstan regions. pp p p p g This report [15] based on the study of world experience summarizes the approaches and tools to stimulate the economic development of the country regions, suggests the author’s vision of the approach to their selection under increased external impacts on regions and uncertainty. The practice of state regional economic policy in Kyrgyzstan has been analyzed, the approach and some tools to increase its effectiveness have been offered. Article [16] analyses comparative indicators of regions of Uzbekistan. The study concluded that the regions with the least developed industrial infrastructure were the poorest. These include the Namangan region (the share of industry in the GRP of this region is only 14.7%), Surkhandarya region (8.2%), Andijan (20.8%) and Fergana (24.8%) regions. Hence, to increase the economic potential of these regions and reduce their dependence on higher budgets, it is necessary to develop industrial production in these regions. Publication [17] describes the change in Uzbekistan’s regional policy, including foreign economic relations. E3S Web of Conferences 449, 01004 (2023) PDSED 2023 It is shown that progression of foreign economic interests of the country has become the number one challenge for the development of regions in modern conditions: increase of exports, attraction of foreign investments and advanced technologies, improvement of transport and transit situation in the country and development of tourism. 2.3 Regional specificity Employment is the central determinant of socio-economic well- being, so regions seek for effective local development strategies aimed at promoting economic growth and employment growth by process of combining a number of elements from many different local individuals and based on local resources, need and characteristics. y The specifics of regional development in the United States of America are also detailed. For example, article [8] provides an empirical study related and absolute specializated in US regions. Analyzing local agglomerations in time, the authors of the article found that the growth of absolute specialization is positively related to salary, while the change in relative concentration has no significant relation to the dynamics of salary. The authors of this article [9] made an empirical study of how the concentration of employment in industries and professions in US regions has changed over time and how regional specialization has developed. Results show that indices of industrial concentration and specialization have decreased, while indices of concentration of occupation and specialization have increased. Specifics of regional development in the Russian Federation described in the paper [10]. The study shows the function of federal policy of regional development as a key tool for implementation aimed at improving the spatial structure of the Russian economy. Assessments of the economic differentiation of the Russian regions and justifications for maintaining the significance of «balancing policy» as one of the priorities of spatial structure regulation of the Russian economy are given; the system of institutions and tools of modern policy of regional development, as well as its legal bases is studied. 6 6 6 3.1. The reality of the entropy approach to the management of socio-economic development of the region Some opportunities for the development of the integrated theory of socio-economic development of the region based on the entropy approach are given in the articles [22, 23]: - the properties of socio-economic systems derived from the properties of living systems, Some opportunities for the development of the integrated theory of socio-economic development of the region based on the entropy approach are given in the articles [22, 23]: p g py pp g - the properties of socio-economic systems derived from the properties of living systems, which in turn are properties of all open dissipative systems: complexity, hierarchy, emergence, role of information; - the properties of socio-economic systems derived from the properties of living systems, which in turn are properties of all open dissipative systems: complexity, hierarchy, emergence, role of information; - the features of entropy dynamics in socio-economic systems, which are that these systems are able to regulate the entropy level within themselves by using substance, energy and information from the environment, and to accumulate negentropy as stock of goods, raw materials, and products; - multiple entropy dichotomy in the economy, showing that individuals/firms have to monitor and regulate the level of both current entropy at once, or death/decay, and entropy (uncertainty) of the future. Besides, entropy can be private (entropy of an individual) and public (entropy of a complex socio-economic system) between which it is also necessary to switch locus of control. Well, there is another dichotomy in objective entropy, which reaches itself out in real energy collapse and/or increasing uncertainty of the future, and subjective entropy felt by human senses. When managing the socio-economic development of the region, the main type of entropy that will serve as a reference when making management decisions is public entropy because the region is a set of individuals, firms, institutions, industries, and territories. The main principle of the entropy approach to the management of socio-economic development of the region is reduction of the current entropy of the region expressed in social, economic and environmental problems, as well as in disproportionality, randomness of development. If there is no managerial influence, the entropy of the socio-economic system of the region is regularly increasing according to the second law of thermodynamics. If there is no managerial influence, it may result in the critical level of current entropy, i.e. the degradation of the region or even the disintegration of the whole socio-economic system. 3.1. The reality of the entropy approach to the management of socio-economic development of the region Entropy approach to study the economic phenomena began to develop lately. Its fact consists to use terminology and mathematical apparatus of synergetics and econophysics upon studying and explanation of socio-economic phenomena and processes. N. Georgescu- Roegen was the first economist to show the entropic nature of economic processes [18]. He formulated several highly important abstracts: formulated several highly important abstracts: - economic process is an entropy process; - economic process is an entropy process; - a man struggles for low entropy, and the economic deficit is a reflection of the entropy law as the most economical one of all natural laws; - economic activity, in fact, serves as an extension and complement to the biological evolution of man. The words V.I. Maevsky “The evolution of living systems is related to a decrease of entropy, and to an increase of energy potential of these systems are better to get the entropic The words V.I. Maevsky “The evolution of living systems is related to a decrease of entropy, and to an increase of energy potential of these systems are better to get the entropic 7 7 E3S Web of Conferences 449, 01004 (2023) PDSED 2023 https://doi.org/10.1051/e3sconf/202344901004 reality of socio-economic systems. It reaches out in improving living systems, the complexity of their functions and structure... there are two types of evolution, one of which determines the movement of living matter, the other having the opposite arrow of time – the movement of non-living matter” [19, P.6]. g [ ] J. Chen offered an entropy theory of value based on the scarcity/rarity of resources required for people to reduce entropy in economic systems [20, 21]. 3.1. The reality of the entropy approach to the management of socio-economic development of the region Each management impact (activity/tool) reduces current entropy and moves away from the critical level. If the offered regional policy activity or tool reduces the current entropy of the region more than it increases, then it is effective (Figure 1). Fig. 1. Entropy dynamics of the current status of the socio-economic system of the region. t H Hкри Degradation, decay Fig. 1. Entropy dynamics of the current status of the socio-economic system of the region. Moreover, to manage the socio-economic development of the region effectively, it must be considered the entropy dynamics of the future. However, it should be noted that if current entropy should be reduced, future entropy management requires keeping it in some optimal range. In other words, it is impossible to allow both this entropy to decrease to minimal 8 8 8 E3S Web of Conferences 449, 01004 (2023) PDSED 2023 https://doi.org/10.1051/e3sconf/202344901004 values, because it goes to show a lack of possible options for management decisions, and to allow its increase above a certain limit, otherwise the number of variants of possible scenarios becomes so great that creates entropy of choice, i.e. it is impossible to choose the best option actually due to the high level of uncertainty (Figure 2). values, because it goes to show a lack of possible options for management decisions, and to allow its increase above a certain limit, otherwise the number of variants of possible scenarios becomes so great that creates entropy of choice, i.e. it is impossible to choose the best option actually due to the high level of uncertainty (Figure 2). t H Hmax Impossibility of choice Hmin Lack of choice Fig. 2. Entropy dynamics of the future status of the socio-economic system of the region. Fig. 2. Entropy dynamics of the future status of the socio-economic system of the region. Synthetic capabilities of entropy approach to the management of socio-economic development of the region. One of the criteria thereof that a new theory or hypothesis to be viable, is how this new theory is in keeping with the past theory and approaches. The entropy approach to the socio- economic development of the region can be harmonized with ready-made approaches by allowing all actual approaches to be seen as ad hoc entropy cases (Table 5). Table 5. 3.1. The reality of the entropy approach to the management of socio-economic development of the region Actual approaches to managing the socio-economic development of the region in terms of the entropy approach Approach Entropy based explanation Infrastructural development Infrastructural development lowers the current entropy of the socio-economic system of the region based on improving the material and technical conditions for individuals and firms, providing them with the substance, energy and information required to lower the current entropy Balanced development Improving the balance of development lowers the randomness of this development, increases degree of order in the system, which is equivalent to lowering the current entropy Inclusive development / Community development The process of globalization increases the current entropy of the region, draining resources and human capital, and globalization leads to an excessive lowering of the entropy of the future, leaving the region with no options for development other than integration into the global agenda. Local resource and community development lowers the entropy effects of globalization. Knowledge-based development Knowledge provides a basis for the emergence of new options for the development of the socio-economic system of the region, allowing to move the entropy of the future from the lower border, in which there is a management deadlock Institutional development An advanced institutional environment controls a mix of options for individuals, thereby lowering the uncertainty of the future, and thus lowering an excessive entropy of the future Sustainable development Integrating the environmental and social dimensions of development lowers current entropy by improving the environment and addressing social challenges. It also allows you Table 5. Actual approaches to managing the socio-economic development of the region in terms of the entropy approach able 5. Actual approaches to managing the socio-economic development of the region in terms of the entropy approach 9 E3S Web of Conferences 449, 01004 (2023) PDSED 2023 https://doi.org/10.1051/e3sconf/202344901004 to prevent the growth of entropy of the future by cutting off the most dangerous development options. to prevent the growth of entropy of the future by cutting off the most dangerous development options. Source: by author based on research As you can see in the table, all current approaches to the management of the socio- economic development of the region can somehow be explained from the point of view of the entropy approach. Hence, the hypothesis that actual approaches can be taken as ad hoc cases of an entropy approach may be seen as proved. 3.2. Useful results of the entropy approach So, useful result is another criterion for the validity of a theory or hypothesis. Therefore, you need to show how the entropy approach allows you to choose the best management solutions. Setting Priorities of economic development is one of the directions of using approach results to managing socio-economic development in practice. The entropy approach offers a universal criterion for Setting Priorities: each of the priorities should provide either a lowering in the current entropy of the region, or regulation of the entropy of the future within an allowed range. Another area of useful results is the choice of economic development activities and tools. Here, by analogy with the Setting Priorities development: if the offered activity or tool lowers the current entropy or somehow controls the entropy of the future, then such an activity should be seen effective. However, in contrast to development priorities, the entropy effect of the tools and activities should also be weighed against the entropy cost of their implementation. If the activity increases current entropy to a greater degree than it reduces, then its use is ineffective. Another useful result is the consideration of regional development. Each region has its own level and dynamics of entropy, so when Setting Priorities it is necessary to choose those that provide the maximum lowering of the current entropy of the region or do not lead to the release of entropy of the future beyond the allowed limits. The same thing is when choosing activities and tools: the entropy effect of their using should be calculated based on the entropy level of the region at this step of development. 5 Conclusion The number of theories describing the socio-economic development of the region is quite high. All theories can be roughly divided into economic-centric and semi-economic. Attempts are being made to form an integrated theory of regional development. Based on the theory of regional development, six most common approaches to the management of the socio-economic development of the regions have emerged: infrastructural development, balanced development, Inclusive development / Community development, knowledge-based development, institutional development, and sustainable development. There are strengths and weaknesses to each of the approaches, but they are poorly coordinated. The entropy approach to the study of economic processes and phenomena appeared lately, but it can be offered as an integrated approach that will support integrate all actual approaches to regional development and address them as a special case. The entropy approach has not just theoretical but also practical results, as it allows a more conscious approach to choose the priorities of regional development, tools and activities of regional strategies and programmes, and also including the specific development of particular regions. g Entropy approach to managing the socio-economic development of the region has its strengths and weaknesses, but has very great prospects to use and to develop. 4 Discussion Indeed, the entropy approach to managing the socio-economic development of the region is unused for most economists, both theorists and practitioners. Although for practitioners, in our opinion, it is intuitive, because practitioners can not explain over and over again why they prefer a particular tool or activity. It is believed that practitioners just intuitively determine the entropic effectiveness of these activities. The advantages of the entropy approach can be attributed: The advantages of the entropy approach can be attributed: - intuitive comprehension for regional management practitioners; - synthetic capabilities to form an integrated approach to the management of the socio- economic development of the region; p g - possibility to develop new tools and activities based on the entropy approach. Disadvantages of the entropy approach: - aversion to economic theorists, especially economic mainstream; - force to master synergy and environmental terminology and methodology. Opportunities for further development of the entropy approach consist in the formation of a mathematical apparatus for level calculating and entropy dynamics of the socio- 10 E3S Web of Conferences 449, 01004 (2023) PDSED 2023 https://doi.org/10.1051/e3sconf/202344901004 economic system of the region and its integrated parts, harmonization of actual regional development terminology and methodology with the entropy approach. economic system of the region and its integrated parts, harmonization of actual regional development terminology and methodology with the entropy approach. References 1. A.G. Guseynov, Fundamental Research, 8, 124-132 (2014) [In Russian] 2. E. Medeiros, European Journal of Spatial Development, 19(5), 1-24 (2022) 3. Andrea Ascani, Riccardo Crescenzi, Simona Iammarino, Regional Economic Development: A Review, SEARCH WP01/03 (2012) http://www.ub.edu/searchproject/wp-content/uploads/2012/02/WP-1.3.pdf 4. A.V. Kuznetsova, O.V. Kuznetsova, Regional policy: foreign experience and Russian realities (M.: IMEMO RAN, 2015) [In Russian] 5. Alexander Cordes, Birgit Gehrke, Christian Rammer, Roman Römisch, Paula Schliessler and Pia Wassmann, Identifying Revealed Comparative Advantages in an EU Regional Context, Research Report 412 (The Vienna Institute for International Economic Studies, 2016) 6. I. Frías, A. Iglesias, I. Neira, Regional specialization and trade patterns in Europe, Working Paper Series Economic Development, 46 (University of Santiago de Compostela, 2000) 7. Maarten Keune, Regions, Regional Institutions and Regional Development, Seed working paper, 16 (Series on Globalization, Area-based Enterprise Development and Employment, International Labour Office Geneva, 2001) 8. Thomas Kemeny, Michael Storper, Specialization and Regional Economic Development, SERC DISCUSSION PAPER 121, 1-38 (2012) 9. Antoine Gervais, James R. Markusen, Anthony J. Venables, Regional Specialization: From the Geography of Industries to the Geography of Jobs (2022) https://ssrn.com/abstract=4092368 or http://dx.doi.org/10.2139/ssrn.4092368 11 https://doi.org/10.1051/e3sconf/202344901004 E3S Web of Conferences 449, 01004 (2023) PDSED 2023 10. E. M. Buchwald, The policy of regional development in modern Russia: priorities, institutions and tools: Scientific work (M.: IE RAN, 2018) [In Russian] 11. S.P. Zemtsov, Yu.A. Smelov, Journal of the New Economic Association, 4 (40), 84– 108 (2019) [In Russian] 12. S. V. Khusainova, S. Y. Bakhvalov, European Research Studies Journal, 21(1), 588- 601 (2018) 13. R. B. Gabdulkhakov, A. L. Poltarykhin, O. M. Tsukanova, & Y. M. Avdeev, Siberian Journal of Life Sciences and Agriculture, 13(6), 339-361(2021) doi:10.12731/2658- 6649-2021-13-6-339-361 14. N.A. Kurmanov, A.T. Uskelenova, N.V. Nikiforova, A.Zh. Satbaeva, Bulletin of the University Turan, (4), 19-27 (2021) https://doi.org/10.46914/1562-2959-2021-1-4-19- 27 [In Russian] 15. Iya Degtyareva, Approaches and tools to stimulate the economic development of the regions of the state in the face of increased external influences and uncertainty, (University of Central Asia, Institute of Public Administration and Politics, Notes IGUP, 3, 2015) [In Russian] 16. L.N. Khazratkulova, Modern development of the economy of Uzbekistan in the context of regions, In the collection: Actual problems and prospects for the development of state statistics in modern conditions, Collection of materials of the V International scientific and practical conference, 2, 170-173 (2019) [In Russian] 17. References Doniyor Kurbanov, Uzbekistan’s new regional policy: Implications for international cooperation in wider Eurasia, Uzbekistan in the context of Regional Security and Global Change, (GCRF COMPASS University of World Economy and Diplomacy. Conference Proceedings, 2019) 26-28. 18. N. Georgescu-Roegen, The entropy law and the economic process (Cambridge, MA: Harvard University Press, 1971) 19. V.I. Maevsky, Introduction to evolutionary macroeconomics (M., Publishing house "Japan today", 1997) [In Russian] 20. J. Chen, An Entropy Theory of Value, SSRN Electronic Journal, (2002) DOI: 10.2139/ssrn.307442. 21. J. Chen, J.K. Galbraith, An Entropy Theory of Value: With Reflections on the Arrow- Debreu Model, SSRN Electronic Journal (2022) Available at SSRN: https://ssrn.com/abstract=4137452 or http://dx.doi.org/10.2139/ssrn.4137452 22. V.F. Islamutdinov, L.N. Salimov, V.M. Kurikov, N.A. Gareeva, Academy of Strategic Management Journal, 20 (4), 1-7 (2021) 23. V.F. Islamutdinov, V.M. Kurikov, Sibirskaya finansovaya shkola, 2, 168-178 (2022) [in Russian] 12
https://openalex.org/W4226518251	https://link.springer.com/content/pdf/10.1007/s12686-022-01263-6.pdf	English	null	Successful storage of Trichomonas gallinae on Whatman FTA cards following culture	Conservation genetics resources	2,022	cc-by	4,461	Abstract Logistical constraints concerning parasite sample storage can hinder progress with the discovery of genetic variation on a global scale. New storage methods are being developed to address this, but require testing in order to understand how widely applicable these methods are. Whatman FTA cards have been tested previously under laboratory conditions for storing low- concentration Trichomonas gallinae isolates with the conclusion that they are not suitable, but have not been tested under field conditions. Here, we conducted a field-test, comparing FTA cards with storage in ethanol for T. gallinae samples col- lected and cultured from wild Columbiformes in Africa using standard field methods, before transportation to the UK. After 6 months storage, both methods resulted in an overall prevalence of 100% following PCR amplification (n = 59), suggesting that FTA cards are suitable for estimation of T. gallinae prevalence. However, samples stored in ethanol produced more, and longer, sequences than those stored on FTA cards. These data suggest storage in ethanol is preferable for the acquisition of high quality genetic strain data, but that FTA cards can be used successfully to ascertain infection prevalence and identify parasite strains under field conditions. Keywords Trichomonas gallinae · Whatman FTA cards · Ethanol · Storage · Genetic variation Successful storage of Trichomonas gallinae on Whatman FTA cards following culture Rebecca C. Thomas1 · Jenny C. Dunn2,3 · Chris J. Orsman2 · Antony J. Morris2 · Helen Hippe Keith C. Hamer1 · Simon J. Goodman1 Received: 29 November 2021 / Accepted: 18 March 2022 / Published online: 2 April 2022 © The Author(s) 2022 * Rebecca C. Thomas rebeccaclairethomas@gmail.com Conservation Genetics Resources (2022) 14:225–229 https://doi.org/10.1007/s12686-022-01263-6 Conservation Genetics Resources (2022) 14:225–229 https://doi.org/10.1007/s12686-022-01263-6 METHODS AND RESOURCES ARTICLE Introduction 2016). Continuing these efforts on a global scale could pave the way for T. gallinae to be developed as a model system for understanding host-parasite relationships. To facilitate this goal, refinements to sample storage immediately after col- lection are required to render the process easier in areas that lack the facilities to carry out the currently adopted protocol International research efforts screening wild birds for Tricho- monas gallinae and identifying genetic strain composition is contributing greatly to the characterization of intra-specific variation and can be used to understand transmission dynam- ics during epidemics (Chi et al. 2013; Ermgassen et al. Keith C. Hamer k.c.hamer@leeds.ac.uk 1 School of Biology, Irene Manton Building, University of Leeds, Leeds LS2 9JT, UK 2 RSPB Centre for Conservation Science, Royal Society for the Protection of Birds, The Lodge, Potton Road, Sandy, Bedfordshire SG19 2DL, UK 3 School of Life Sciences, University of Lincoln, Joseph Banks Laboratories, Lincoln LN6 7TS, UK 4 NERC Biomolecular Analysis Facility, Department of Animal and Plant Sciences, University of Sheffield, Western Bank, Sheffield S10 2TN, UK 5 Natural England, Suite D, Unex House, Bourges Boulevard, Peterborough PE1 1NG, UK * Rebecca C. Thomas rebeccaclairethomas@gmail.com * Jenny C. Dunn jdunn@lincoln.ac.uk * Simon J. Goodman s.j.goodman@leeds.ac.uk Chris J. Orsman Chris.orsman@rspb.org.uk Antony J. Morris tony.morris@rspb.org.uk Helen Hipperson H.Hipperson@sheffield.ac.uk Philip V. Grice phil.grice@naturalengland.org.uk * Rebecca C. Thomas rebeccaclairethomas@gmail.com * Jenny C. Dunn jdunn@lincoln.ac.uk * Simon J. Goodman s.j.goodman@leeds.ac.uk Chris J. Orsman Chris.orsman@rspb.org.uk Antony J. Morris tony.morris@rspb.org.uk Helen Hipperson H.Hipperson@sheffield.ac.uk Philip V. Grice phil.grice@naturalengland.org.uk Keith C. Hamer k.c.hamer@leeds.ac.uk 1 School of Biology, Irene Manton Building, University of Leeds, Leeds LS2 9JT, UK 2 RSPB Centre for Conservation Science, Royal Society for the Protection of Birds, The Lodge, Potton Road, Sandy, Bedfordshire SG19 2DL, UK 3 School of Life Sciences, University of Lincoln, Joseph Banks Laboratories, Lincoln LN6 7TS, UK 4 NERC Biomolecular Analysis Facility, Department of Animal and Plant Sciences, University of Sheffield, Western Bank, Sheffield S10 2TN, UK 5 Natural England, Suite D, Unex House, Bourges Boulevard, Peterborough PE1 1NG, UK (0121 3456789) 3 226 Conservation Genetics Resources (2022) 14:225–229 for T. gallinae storage. This problem is currently limiting sample collection on a global scale. extractions performed 48 h, two and three weeks later (Holt et al. 2015). Sample collection Five dove species were caught in an enclosed area of aca- cia scrubland at a site near Sandiara, west Senegal (14° 25′ 53.9″ N, 16° 47′ 35.0″ W) during January and February 2015. This site is an enclosed area that has been protected from grazing, allowing semi-mature woodland to form, con- sisting mostly of acacia bushes used by roosting doves at high densities. European Turtle doves Streptopelia turtur, Laughing doves Spilopelia senegalensis, Namaqua doves Oena capensis, Vinaceous doves Streptopelia vinacea and Black-billed Wood doves Turtur abyssinicus were caught using whoosh nets and mist nets (Redfern and Clark 2001) and placed into cloth bags prior to processing. All birds were ringed using individually numbered colour rings, and aged and sexed where possible (Redfern and Clark 2001). The current storage protocol for T. gallinae requires the parasite to be isolated from culture media using centrifu- gation and subsequently re-suspended in phosphate buff- ered saline (PBS) where it is frozen at -20 °C until DNA extraction (Riley et al. 1992; Lennon et al. 2013), which requires any field site to have both a centrifuge and freezer. Whatman FTA cards (GE Healthcare Life Sciences, UK; hereafter referred to as FTA cards) are available in a range of technology formats (FTA classic card, mini card, micro card and Elute card). All formats capture nucleic acids and preserve them at room temperature making this system ideal for collection of samples in the field and preservation during storage and transportation. Storing T. gallinae isolates on FTA Elute cards has previously been tested in laboratory, but not field, conditions. Two different concentrations (10 and 100 trichomonads/40 µl) of two T. gallinae isolates were inoculated onto FTA Elute cards, in triplicate, with DNA p All birds were swabbed for T. gallinae using a moistened sterile viscose swab that was inserted into the oral cavity and used to swab the mouth cavity, oesophagus and crop. The swab was inoculated into an individual InPouch TF culture kit (Biomed Diagnostics, Oregon), sealed and incubated at room temperature (ranging from 25 to 30 °C) for 7–11 days in order to culture any parasites present. After the incuba- tion period, 0.5 ml of the culture media was transferred to a FTA Classic card using a separate disposable pipette for each sample, and allowed to air dry before being placed into an envelope with a desiccant sachet, ready for transport and subsequent storage. Introduction Only three samples (one at the weaker concen- tration and two from the stronger concentration) from one isolate extracted after 48 h were detected as positive from PCR (Holt et al. 2015). This suggested that FTA Elute cards are not a suitable storage solution for T. gallinae, at least not in the long-term or for samples of low concentration (Holt et al. 2015). However, FTA cards have not been tested under field conditions, where the requirement for culture of parasites prior to isolation (Dunn et al. 2016) may lead to higher concentrations of parasites present in media, and thus an increase in subsequent detection rates.f p g Trichomonas gallinae is a microaerophilic protozoan par- asite that resides in the upper digestive tract of birds (Stabler 1954). The clinical presentation of infection, trichomonosis, has been responsible for epidemics in wild finch popula- tions across Europe (Neimanis et al. 2010; Peters and Lud- wichowski 2010; Robinson et al. 2010; Lawson et al. 2011b; Ganas et al. 2014; Zadravec et al. 2017). Mortality from trichomonosis has led to a change in conservation status to endangered for the UK population of greenfinch Cardualis chloris, (Stanbury et al. 2017). A clonal strain of T. gallinae (A1) has been identified by morphological and subsequent molecular analysis as a causative agent of trichomonosis in European birds (Robinson et al. 2010; Lawson et al. 2011a). Globally, however, it is not the sole etiological agent of trich- omonosis outbreaks, as a different variant (A2) was isolated from Band-tailed Pigeons Patagioenas fasciata during epi- demics of the disease in California (Girard et al. 2014b). A new species of Trichomonas (T. stableri n.sp.) was also detected during these epidemics confirming a non-clonal eti- ology of avian trichomonosis in Band-tailed Pigeons (Girard et al. 2014a). Multiple additional—apparently non-patho- genic—strains of T. gallinae also circulate in wild bird pop- ulations (Thomas 2017). None of these discoveries would have been possible without identifying the parasite strains using molecular techniques. It is becoming standard practice to include this type of analysis for the accurate and reliable identification of etiological agents of infection, but expan- sion of geographical and temporal monitoring is imperative for progress within this field. In this context, T. gallinae has features that lend it to being a model system, such as being widespread, abundant and easily assayed. Introduction However, further refinement of the assay protocol is required to remove bar- riers currently experienced by research groups attempting to sample the parasite without access to a laboratory. Here, we trialled two different methods of storing T. gal- linae samples collected from African Columbidae and cul- tured before transport from Senegal to the UK for genetic analysis. After incubation in culture media, each sample was split between FTA Classic cards and ethanol. After 6 months of storage, we extracted DNA and compared PCR amplifica- tion and DNA sequencing success between the two methods. Results Fifty-nine samples collected from European Turtle doves (n = 44), Laughing doves (n = 4), Namaqua doves (n = 9), Vinaceous doves (n = 1) and Black-billed Wood doves (n = 1) were stored on both Whatman FTA card and in ethanol. The efficacy of storage methods for subsequent genetic analy- sis was compared based on PCR and Sanger sequencing results (Table 1). T. gallinae prevalence was 100% for both methods. Sample collection The remaining 2 ml of culture media was transferred to three 1.5 ml eppendorf tubes, leaving enough room for an equal volume of absolute ethanol to be 1 3 3 227 Conservation Genetics Resources (2022) 14:225–229 Samples were purified and sequenced at the NERC Biomo- lecular Analysis Facility (NBAF Sheffield) on the ABI3730 DNA Analyser (Applied Biosystems, CA, USA). subsequently added to each tube. FTA cards and eppendorf tubes were stored at room temperature for 6–27 days before transport to the UK. On arrival in the UK, FTA cards were stored at room temperature, and eppendorfs were stored at − 20 °C until further processing. DNA extraction Sequences returned from Sanger sequencing were manually assessed for sequencing errors, trimmed, and the forward and reverse sequences aligned in BioEdit (Hall 2005). The NCBI-BLAST algorithm (Altschul et al. 1997) was used to determine the closest sequence match for each sequence. If the sequence was of poor quality (less than 70% query cover) it was removed from the analysis. A Mann–Whitney U test was used to test for differences in sequence length between the two storage methods. Samples in ethanol were centrifuged at 13,000 rpm for 10 min and the supernatant removed to leave a parasite pellet. One eighth of the circle (approx. 2 mm × 2 mm) that held the parasite sample on the FTA card was cut out using a sterile pair of scissors and placed into an eppendorf tube. The scissors were disinfected between each sample with 10% bleach and thoroughly dried. DNA was extracted from all samples using a modified ammonium acetate method (Nicholls et al. 2000). Briefly, the parasite pellet was digested overnight in digestion buffer (20 mM EDTA, 50 mM Tris, 120 mM NaCl, 1% SDS, pH 8.0) with 50 µg of Proteinase K (Sigma-Aldrich, UK). Ammonium acetate (4 M) was then used to precipitate out the proteins and etha- nol precipitated out the DNA. The resulting DNA pellet was dissolved in 20–50 µl low TE buffer, depending on the size of the pellet, in a water bath at 65 °C. The extracted DNA was stored at − 20 °C. PCR amplification of the ITS1/ 5.8S/ ITS2 ribosomal region and sequencing Comparing results from the same sample, the ethanol storage method obtained sequences from 15 samples that the FTA card storage method did not. The FTA card method obtained one sequence from a sample for which the ethanol storage method failed. Where sequence reads were obtained for samples from both storage techniques (n = 30), there were no conflicts on strain identification, with six strains of Trichomonas sp. identified; consistent with additional stud- ies from the same site (Thomas 2017; Young et al. unpubl. data) and reported in more detail elsewhere (Thomas et al. 2022). Overall, the samples stored in ethanol gave longer sequence reads than the samples stored on FTA cards (Mann Whitney-U, W = 111, p < 0.001; Fig. 1). The primers TFR1 [f] (5′-TGCTTCAGTTCAGCGGGT CTTCC-3′) and TFR2 [r] (5′-CGGTAGGTGAACCTGCCG TTGG-3′) (Gaspar da Silva et al. 2007) were used to target a 400 bp length of the ITS1/5.8S/ITS2 ribosomal region of the genome (hereafter referred to as the ITS region). The PCR recipe consisted of 0.8 × Qiagen Multiplex PCR Master Mix (Qiagen, Hilden, Germany), 0.5 µM each of forward and reverse primer (Sigma-Aldrich, UK), and 1 µl of template DNA in a total reaction volume of 10 µl. A positive and neg- ative control were included in each PCR run to confirm suc- cessful amplification and lack of contamination respectively. A touchdown PCR thermal cycling program was performed as follows: 15 min at 95 °C followed by 11 cycles of 1 min at 94 °C, 30 s at 66 °C (decreasing by 1 °C every cycle until 56 °C), 1 min at 72 °C; then 24 cycles of 1 min at 94 °C, 30 s at 55 °C and 1 min at 72 °C with a final elongation step of 10 min at 72 °C. PCR reactions were run on either a Gene Amp 9700 PCR system (Applied Biosystems, Foster City, CA) or a DNA Engine Tetrad 2 (Bio-Rad Laboratories Inc, CA, USA). PCR products were electrophoresed through a 1% agarose gel stained with ethidium bromide and visualised by UV light. The presence of an amplicon at the expected product size indicated the presence of T. gallinae infection. If the result was negative, the PCR was repeated to confirm. Discussion The samples stored in ethanol performed better than the samples stored on FTA cards in terms of being able to obtain a DNA sequence and the quality of DNA sequence data. In contrast to the findings of Holt et al. (2015), we were able to amplify T. gallinae DNA from FTA cards up to six months after the samples had been collected, although we cannot compare the concentrations of trichomonads that were ini- tially inoculated onto the FTA cards in each case as this was beyond the scope of our study, and more representative of 1 1 3 Conservation Genetics Resources (2022) 14:225–229 228 A ‘Good’ quality result from PCR is a strong band visualised on gel, ‘Poor’ quality is either a faint band or non-distinct band; overall prevalence was 100%. A ‘Good’ quality result from sequencing is both directions of sequencing run being assembled and DNA matched to Trichomonas sp., ‘Poor’ quality is sequence fail- ure or poor quality sequence (small/non-existent peaks or double peaks on chromatogram) Stage Ethanol stored samples FTA card stored samples Good quality (%) Poor quality (%) Total N Good quality (%) Poor quality (%) Total N PCR #1 88 12 59 76 24 59 Sequence #1 73 27 53 47 53 45 PCR #2 100 0 27 97 3 39 Sequence #2 52 48 27 23 77 39 A ‘Good’ quality result from PCR is a strong band visualised on gel, ‘Poor’ quality is either a faint band or non-distinct band; overall prevalence was 100%. A ‘Good’ quality result from sequencing is both directions of sequencing run being assembled and DNA matched to Trichomonas sp., ‘Poor’ quality is sequence fail- ure or poor quality sequence (small/non-existent peaks or double peaks on chromatogram) Fig. 1 Mean ± SE length (bp) of Trichomonas sp. sequences produced following analysis of samples stored in ethanol or on FTA cards recovering the DNA from the matrix than simple DNA elu- tion. Despite the improved performance of FTA cards in pre- serving T. gallinae DNA under field conditions, we recom- mend that cultured T. gallinae media should be mixed with an equal amount of ethanol for storage and transportation from a field site to a laboratory to maximise the acquisition of genetic strain information. Discussion However, if the acquisition and/or transport of ethanol is logistically difficult, storage on FTA cards following culture is a viable alternative for strain identification, and for estimation of infection preva- lence, although we recommend a repeat PCR on any nega- tive samples. The storage of T. gallinae in this form has not been trialled beyond 6 months so it is advisable to perform DNA extraction within this time frame.i Fig. 1 Mean ± SE length (bp) of Trichomonas sp. sequences produced following analysis of samples stored in ethanol or on FTA cards This solution has simplified the process for T. gallinae storage after growth in culture media. Collection of T. gallinae samples from wild bird populations is no longer restricted to sites with access to laboratory facilities or field workers with the necessary laboratory skills. Research into the prevalence of T. gallinae infection in wild bird popu- lations and the strains circulating in these populations can hence be pursued on a global scale. Furthermore, this pro- tocol greatly eases T. gallinae sample storage post-collec- tion which, in addition to the use of the InPouch culture kits, results in a very simple sample collection and storage system—an important feature for any model host-parasite system. field conditions. It is possible that much higher concentra- tions than 10 or 100 trichomonads per 40 µl were inoculated onto our FTA cards, and elsewhere it has been demonstrated that parasite culture is an essential step to identify sub-clin- ical T. gallinae infection in wild birds (Dunn et al. 2016). Holt et al. (2015) suggested the use of FTA cards would not be sensitive enough to detect sub-clinical infections but none of the African species of Columbidae in our study had clinical signs of trichomonosis, showing that it is possible to diagnose sub-clinical infection using FTA cards. Holt et al. (2015) used a different protocol to initially process the FTA cards storing T. gallinae. They used FTA Elute cards, which allow the direct elution of DNA through a series of pulse centrifugation steps and washes whereas the FTA Classic cards require a DNA extraction protocol such as was used here. However, a previous study comparing efficiency of DNA extractions from multiple types of preservation paper found no significant differences between storage on FTA Classic or FTA Elute cards, when the same DNA extraction method was used (Love Stowell et al. Discussion (Patagioenas fasciata monilis). Int J Parasitol Parasites Wildl 3:32–40. https://doi.org/10.1016/j.ijppaw.2013.12.002 (Patagioenas fasciata monilis). Int J Parasitol Parasites Wildl 3:32–40. https://doi.org/10.1016/j.ijppaw.2013.12.002 Girard Y, Rogers K, Woods L et al (2014b) Dual-pathogen etiology of avian trichomonosis in a declining band-tailed pigeon population. Infect Genet Evol 24:146–156. https://doi.org/10.1016/j.meegid. 2014.03.002 Funding RCT was funded by the Royal Society for the Protection of Birds and Natural England through the Action for Birds in England (AfBiE) partnership; fieldwork and sample collection was funded by the RSPB. CJO, AJM and PVG are employed by the funders and con- tributed to data collection, reviewing and editing of the manuscript. Hall T (2005) BioEdit: biological sequence alignment editor for Win95/98/NT/2K/XP Holt J, Purple K, Gerhold R (2015) Use of FTA technology for detec- tion of Trichomonas gallinae. Vet Parasitol 212:396–399. https:// doi.org/10.1016/j.vetpar.2015.07.017 Data availability Data will be deposited in an Open Access Repository upon manuscript acceptance. Lawson B, Cunningham A, Chantrey J et al (2011a) A clonal strain of Trichomonas gallinae is the aetiologic agent of an emerging avian epidemic disease. Infect Genet Evol 11:1638–1645. https://doi. org/10.1016/j.meegid.2011.06.007 Declarations Lawson B, Robinson R, Neimanis A et al (2011b) Evidence of spread of the emerging infectious disease, finch trichomonosis, by migrating birds. EcoHealth 8:143–153. https://doi.org/10.1007/ s10393-011-0696-8 Conflict of interest The authors have no relevant financial or non-fi- nancial interests to disclose. Ethical approval Ethical approval for this work was granted by the RSPB Ethics committee. All birds were released back to the wild fol- lowing sample collection. Lennon R, Dunn J, Stockdale J et al (2013) Trichomonad parasite infection in four species of Columbidae in the UK. Parasitology 140:1368–1376 Love Stowell S, Bentley E, Gagne R et al (2018) Optimal DNA extractions from blood on preservation paper limits conservation genomic but not conservation genetic applications. J Nat Conserv 46:89–96. https://doi.org/10.1016/j.jnc.2018.09.004 Open Access This article is licensed under a Creative Commons Attri- bution 4.0 International License, which permits use, sharing, adapta- tion, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Neimanis A, Handeland F, Isomursu B et al (2010) First report of epizootic trichomoniasis in wild finches (Family Fringillidae) in Southern Fennoscandia. Avian Dis 54:136–141. https://doi.org/ 10.1080/03079457.2014.896120 Nicholls J, Double M, Rowell D, Magrath R (2000) The evolution of cooperative and pair breeding in thornbills. J Avian Biol 2:165–176i Peters M, Ludwichowski I (2010) Trichomoniasis in Greenfinches Car- duelis chloris and other passerine birds (Passeriformes) in German in 2009—an attempt to strike a balance. Vogelwelt 131:207–212 df l k i l i i h f i h l in 2009—an attempt to strike a balance. Vogelwelt 131:207–212 Redfern C, Clark J (2001) Ringers’ Manual. British Trust for Ornithol- in 2009—an attempt to strike a balance. Vogelwelt 131:207–212 Redfern C, Clark J (2001) Ringers’ Manual. British Trust for Ornithol- ogy, Thetford Discussion 2018) The FTA card technology is designed to lyse cells, denature proteins and protect DNA from degradation. Our method of DNA extrac- tion from the FTA card may have been more efficient at field conditions. It is possible that much higher concentra- tions than 10 or 100 trichomonads per 40 µl were inoculated onto our FTA cards, and elsewhere it has been demonstrated that parasite culture is an essential step to identify sub-clin- ical T. gallinae infection in wild birds (Dunn et al. 2016). Holt et al. (2015) suggested the use of FTA cards would not be sensitive enough to detect sub-clinical infections but none of the African species of Columbidae in our study had clinical signs of trichomonosis, showing that it is possible to diagnose sub-clinical infection using FTA cards. Holt et al. (2015) used a different protocol to initially process the FTA cards storing T. gallinae. They used FTA Elute cards, which allow the direct elution of DNA through a series of pulse centrifugation steps and washes whereas the FTA Classic cards require a DNA extraction protocol such as was used here. However, a previous study comparing efficiency of DNA extractions from multiple types of preservation paper found no significant differences between storage on FTA Classic or FTA Elute cards, when the same DNA extraction method was used (Love Stowell et al. 2018) The FTA card technology is designed to lyse cells, denature proteins and protect DNA from degradation. Our method of DNA extrac- tion from the FTA card may have been more efficient at Acknowledgements Ringing and sampling procedures were carried out under a permit granted by the Direction des Eaux, Forêts, Chasses et de la Conservation des Sols. Samples were imported to the UK under import licence PATH/201/2012/2 from DEFRA. A positive control of T. gallinae was provided by Alrefaei et al. (University of East Anglia, finch epidemic strain, Type A). Thanks to an anonymous reviewer and to the Editor in Chief for comments that improved an earlier draft. Author contributions RCT: Conceptualization, Methodology, Inves- tigation, Formal analysis, Writing; JCD: Conceptualization, Formal analysis, Supervision, Writing; COO: Investigation, Writing—Review & Editing; AJM: Funding acquisition, HH: Methodology, Writ- ing – Review & Editing; Writing—Review & Editing; PVG: Fund- ing acquisition, Writing—Review & Editing; KCH: Supervision, 1 3 3 Conservation Genetics Resources (2022) 14:225–229 229 Writing—Review & Editing; SJG: Conceptualization, Supervision, Writing. Writing—Review & Editing; SJG: Conceptualization, Supervision, Writing. References Riley D, Roberts M, Takayama T, Krieger J (1992) Development of a polymerase chain reaction-based diagnosis of Trichomonas vagi- nalis. J Clin Microbiol 30:465–472 Altschul S, Madden T, Schäffer A et al (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search pro- grams. Nucleic Acids Res 25:3389–3402i Robinson R, Lawson B, Toms M et al (2010) Emerging infectious dis- ease leads to rapid population declines of common British birds. PLoS ONE 5:e12215. https://doi.org/10.1371/journal.pone.00122 15 Chi J, Lawson B, Durrant C et al (2013) The finch epidemic strain of Trichomonas gallinae is predominant in British non-passerines. Parasitology 140:1234–1245. https://doi.org/10.1017/S003118201 3000930 Stabler R (1954) Trichomonas gallinae: a review. Exp Parasitol 3:368–402 Dunn J, Stockdale J, McCubbin A et al (2016) Non-cultured faecal and gastrointestinal seed samples fail to detect Trichomonad infection in clinically and sub-clinically infected columbid birds. Conserv Genet Resour 8:97–99 Stanbury A, Brown A, Eaton M et al (2017) The risk of extinction for birds in Great Britain. Br Birds 110:502–517 Thomas R, Dunn J, Dawson D, et al (2022) Assessing rates of parasite coinfection and spatiotemporal strain variation via metabarcoding: insights for the conservation of European Turtle Doves Streptope- lia turtur. Mol Ecol. https://doi.org/10.1111/mec.16421 Ermgassen E, Durrant C, John S et al (2016) Detection of the Euro- pean epidemic strain of Trichomonas gallinae in finches, but not other non-columbiformes, in the absence of macroscopic disease. Parasitology 143:1294–1300 Thomas R (2017) Molecular epidemiology of Trichomonas gallinae in Turtle doves (Streptopelia turtur). PhD thesis, University of Leeds Ganas P, Jaskulska B, Lawson B et al (2014) Multi-locus sequence typing confirms the clonality of Trichomonas gallinae isolates cir- culating in European finches. Parasitology 141:652–661. https:// doi.org/10.1017/S0031182013002023 Zadravec M, Slavec B, Krapež U et al (2017) Trichomonosis outbreak in a flock of canaries (Serinus canaria f. domestica) caused by a finch epidemic strain of Trichomonas gallinae. Vet Parasitol 239:90–93. https://doi.org/10.1016/j.vetpar.2017.04.001 Gaspar da Silva D, Barton E, Bunbury N et al (2007) Molecular iden- tity and heterogeneity of trichomonad parasites in a closed avian population. Infect Genet Evol 7:433–440. https://doi.org/10. 1016/j.meegid.2007.01.002 Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. j g Girard Y, Rogers K, Gerhold R et al (2014a) Trichomonas stableri n. sp., an agent of trichomonosis in Pacific Coast band-tailed pigeons 1 3
W4298021842.txt	https://wiki.lyrasis.org/spaces/flyingpdf/pdfpageexport.action?pageId=230825353	en		Feasibility Study for a Copyright Education Center	null	2,017	cc-by	78	Feasibility Study for a Copyright Education Center Grant DOI: https://doi.org/10.54514/i9f5s Grantee: Columbia University Lead Investigator: Kristin Kelly Co-Lead Investigator: Rina Elster Pantalony Keywords: University library, Copyright, Fair use (Copyright), Electronic information resources--Fair use (Copyright) Year Awarded: 2017 Amount: $30,000 Related Items: Final Report: https://doi.org/10.48609/s6tg-vb13 Presentation Slides: https://doi.org/10.48609/jx0y-2c28 Description: One of the first Catalyst Fund grants given, this project conducted a feasibility study to determine the viability and sustainability of a copyright education center for libraries, archives and museums.
https://openalex.org/W2750920106	https://www.biorxiv.org/content/biorxiv/early/2017/08/31/183210.full.pdf	English	null	Detection and accurate False Discovery Rate control of differentially methylated regions from Whole Genome Bisulfite Sequencing	bioRxiv (Cold Spring Harbor Laboratory)	2,017	cc-by	19,892	. CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint ; https://doi.org/10.1101/183210 doi: bioRxiv preprint Detection and accurate False Discovery Rate control of differentially methylated regions 1 from Whole Genome Bisulfite Sequencing 2 3 Keegan D. Korthauer1,2, Sutirtha Chakraborty3, Yuval Benjamini4, and Rafael A. Irizarry1,2 4 1Department of Biostatistics & Computational Biology, Dana-Farber Cancer Institute 5 2Department of Biostatistics, Harvard T.H. Chan School of Public Health 6 3Novartis 7 4Department of Statistics, Hebrew University 8 9 Summary 10 With recent advances in sequencing technology, it is now feasible to measure DNA methylation 11 at tens of millions of sites across the entire genome. In most applications, biologists are 12 interested in detecting differentially methylated regions, composed of multiple sites with 13 differing methylation levels among populations. However, current computational approaches for 14 detecting such regions do not provide accurate statistical inference. A major challenge in 15 reporting uncertainty is that a genome-wide scan is involved in detecting these regions, which 16 needs to be accounted for. A further challenge is that sample sizes are limited due to the costs 17 associated with the technology. We have developed a new approach that overcomes these 18 challenges and assesses uncertainty for differentially methylated regions in a rigorous manner. 19 Region-level statistics are obtained by fitting a generalized least squares (GLS) regression model 20 with a nested autoregressive correlated error structure for the effect of interest on transformed 21 methylation proportions. We develop an inferential approach, based on a pooled null 22 distribution, that can be implemented even when as few as two samples per population are 23 available. Here we demonstrate the advantages of our method using both experimental data and 24 Monte Carlo simulation. We find that the new method improves the specificity and sensitivity of 25 list of regions and accurately controls the False Discovery Rate (FDR). 26 Detection and accurate False Discovery Rate control of differentially methylated regions 1 from Whole Genome Bisulfite Sequencing 2 3 Detection and accurate False Discovery Rate control of differentially methylated regions 1 from Whole Genome Bisulfite Sequencing 2 3 Even though the most common application of the technology is to detect differentially 10 methylated regions (DMRs) between populations, most methods for analysis of WGBS 11 experiments focus on statistical differences for CpG loci one at a time (Akalin et al., 2012, 12 Dolzhenko and Smith, 2014, Lee and Morris, 2016, Park et al., 2014, Park and Wu, 2016). While 13 useful, approaches for identification of differentially methylated loci (DML) have many practical 14 limitations in both implementation and interpretation. Here, we discuss these limitations as well 15 as outline the challenges of performing inference at the region level. Finally, we introduce a 16 rigorous statistical approach that overcomes these challenges to construct de novo DMRs with 17 accurate FDR control. 18 Methods to identify DMLs in WGBS experiments are greatly hindered by the high- 19 dimensionality and low sample size setting that is common in high-throughput genomics studies. 20 The number of tests performed is equal to the number of loci analyzed, which is very large in 21 typical WGBS studies. In the human genome, for example, there are close to thirty million CpG 22 loci (Smith and Meissner, 2013). Further, DML methods generally do not account for the well- 23 known fact that measurements are spatially correlated across the genome (Leek et al 2010) and 24 Detection and accurate False Discovery Rate control of differentially methylated regions 1 from Whole Genome Bisulfite Sequencing 2 3 27 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint Keywords: bisulfite sequencing, differential methylation, false discovery rate, generalized least 1 squares, inference 2 3 1. Introduction 4 DNA methylation is an important epigenetic modification that plays a role in a wide variety of 5 biological processes. Numerous studies have been carried out to locate CpG loci where DNA 6 methylation may be involved in gene regulation, differentiation, and cancer. With recent 7 advances in sequencing technology such as Whole Genome Bisulfite Sequencing (WGBS), it is 8 now possible to measure DNA methylation at single base resolution across all CpGs in the 9 genome. Even though the most common application of the technology is to detect differentially 10 methylated regions (DMRs) between populations, most methods for analysis of WGBS 11 experiments focus on statistical differences for CpG loci one at a time (Akalin et al., 2012, 12 Dolzhenko and Smith, 2014, Lee and Morris, 2016, Park et al., 2014, Park and Wu, 2016). While 13 useful, approaches for identification of differentially methylated loci (DML) have many practical 14 limitations in both implementation and interpretation. Here, we discuss these limitations as well 15 as outline the challenges of performing inference at the region level. Finally, we introduce a 16 rigorous statistical approach that overcomes these challenges to construct de novo DMRs with 17 accurate FDR control. 18 3 1. Introduction 4 DNA methylation is an important epigenetic modification that plays a role in a wide variety of 5 biological processes. Numerous studies have been carried out to locate CpG loci where DNA 6 methylation may be involved in gene regulation, differentiation, and cancer. With recent 7 advances in sequencing technology such as Whole Genome Bisulfite Sequencing (WGBS), it is 8 now possible to measure DNA methylation at single base resolution across all CpGs in the 9 genome. 1. Introduction 4 experiments focus on statistical differences for CpG loci one at a time (Akalin et al., 2012, 12 Dolzhenko and Smith, 2014, Lee and Morris, 2016, Park et al., 2014, Park and Wu, 2016). While 13 useful, approaches for identification of differentially methylated loci (DML) have many practical 14 limitations in both implementation and interpretation. Here, we discuss these limitations as well 15 as outline the challenges of performing inference at the region level. Finally, we introduce a 16 rigorous statistical approach that overcomes these challenges to construct de novo DMRs with 17 accurate FDR control. 18 Methods to identify DMLs in WGBS experiments are greatly hindered by the high- 19 dimensionality and low sample size setting that is common in high-throughput genomics studies. 20 The number of tests performed is equal to the number of loci analyzed, which is very large in 21 typical WGBS studies. In the human genome, for example, there are close to thirty million CpG 22 loci (Smith and Meissner, 2013). Further, DML methods generally do not account for the well- 23 known fact that measurements are spatially correlated across the genome (Leek et al., 2010) and 24 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint instead treat measurements from all loci as independent. Correcting for multiple comparisons 1 without taking into account these correlations can result in a loss of power. 2 instead treat measurements from all loci as independent. Correcting for multiple comparisons 1 without taking into account these correlations can result in a loss of power. 2 Additionally, methods for assessing the significance of DMLs typically require large 3 sample sizes due to reliance on large sample approximations (Dolzhenko and Smith, 2014, 4 Hansen, Langmead and Irizarry, 2012, Hebestreit, Dugas and Klein, 2013, Lee and Morris, 5 2016). Although WGBS is the current gold standard for estimating whole genome methylation 6 profiles (Marx, 2016), cost limitations are still a barrier to acquiring more than a few individuals 7 per biological condition in many studies (Ziller et al., 2015). This is reflected in the study design 8 of major consortiums that aim to characterize the epigenome. 1. Introduction 4 CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: ioRxiv preprint DML approaches have also modeled WGBS count data using Binomial models (Saito, Tsuji and 1 Mituyama, 2014). However, Binomial models on their own cannot account for biological 2 variability within sample groups. In order to account for biological variability in count data, 3 Beta-Binomial models (Park et al., 2014, Sun et al., 2014) are a natural extension. However they 4 come at the cost of increased computational burden when testing millions of loci. 5 Beyond implementation challenges, DML approaches also suffer from limited 6 interpretability. In general, identifying DMRs is more biologically relevant than reporting DMLs. 7 Apart from the so-called ‘CpG traffic lights’ (Khamis et al., 2017), most individual CpG loci 8 likely do not have a large impact on epigenetic function on their own, but rather through a 9 biochemical modification that involves several loci. Most notably, regional DNA methylation 10 levels are correlated with the expression levels of nearby genes. Specifically, methylation gain is 11 associated with stable transcriptional silencing of nearby genes (Bird, 2002). In the context of 12 differential methylation analysis, Aryee et al. (2014) found that differentially expressed genes 13 were consistently more likely to be located near DMRs than DMLs. 14 While DML approaches may construct DMRs by chaining together neighboring 15 i ifi l i hi f h ill i ld f h i i l DML approaches have also modeled WGBS count data using Binomial models (Saito, Tsuji and 1 Mituyama, 2014). However, Binomial models on their own cannot account for biological 2 variability within sample groups. In order to account for biological variability in count data, 3 Beta-Binomial models (Park et al., 2014, Sun et al., 2014) are a natural extension. However they 4 come at the cost of increased computational burden when testing millions of loci. 5 Beta-Binomial models (Park et al., 2014, Sun et al., 2014) are a natural extension. However they 4 come at the cost of increased computational burden when testing millions of loci. 5 Beyond implementation challenges, DML approaches also suffer from limited 6 interpretability. In general, identifying DMRs is more biologically relevant than reporting DMLs. 1. Introduction 4 For example, WGBS experiments 9 in murine embryos carried out as part of the ENCODE project are limited to two biological 10 replicates per tissue type and developmental time point combination (He et al., 2017). In 11 addition, the number of biological replicates measured with WGBS in the UCSD Human 12 Reference Epigenome Mapping Project (Schultz et al., 2015) is also limited to 2-3 per tissue 13 type. As such, we aim to maximize power while controlling the false discovery rate even with 14 sample sizes as small as two samples per condition. 15 Methods for identifying DMLs also need to properly model count data that does not 16 conform to standard Gaussian models. This is in contrast to methylation array analysis, where 17 Gaussian models performed well (Jaffe et al., 2012). One option is to assume that methylation 18 proportions, defined as the number of methylated reads divided by the number of total reads 19 covering a given CpG locus, follow a normal distribution (Hansen, Langmead and Irizarry, 20 2012). However this assumption clearly does not hold when the total reads covering the CpG, 21 referred to as the coverage, is small, a common occurrence in these datasets. The approach also 22 ignores that variance of this proportion depends on the coverage. To overcome these limitations, 23 Methods for identifying DMLs also need to properly model count data that does not 16 conform to standard Gaussian models. This is in contrast to methylation array analysis, where 17 Gaussian models performed well (Jaffe et al., 2012). One option is to assume that methylation 18 proportions, defined as the number of methylated reads divided by the number of total reads 19 covering a given CpG locus, follow a normal distribution (Hansen, Langmead and Irizarry, 20 2012). However this assumption clearly does not hold when the total reads covering the CpG, 21 referred to as the coverage, is small, a common occurrence in these datasets. The approach also 22 ignores that variance of this proportion depends on the coverage. To overcome these limitations 23 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . 1. Introduction 4 7 Apart from the so-called ‘CpG traffic lights’ (Khamis et al., 2017), most individual CpG loci 8 likely do not have a large impact on epigenetic function on their own, but rather through a 9 biochemical modification that involves several loci. Most notably, regional DNA methylation 10 levels are correlated with the expression levels of nearby genes. Specifically, methylation gain is 11 associated with stable transcriptional silencing of nearby genes (Bird, 2002). In the context of 12 differential methylation analysis, Aryee et al. (2014) found that differentially expressed genes 13 were consistently more likely to be located near DMRs than DMLs. 14 While DML approaches may construct DMRs by chaining together neighboring 15 significant loci, this type of approach will not yield a proper assessment of the statistical 16 significance of the constructed regions, nor will the False Discovery Rate (FDR) be properly 17 controlled (Robinson et al., 2014). This is because controlling the FDR at the level of individual 18 loci is not the same as controlling FDR of regions, as has been noted in the context of peak 19 calling in ChIP-seq experiments (Lun and Smyth, 2014, Siegmund, Zhang and Yakir, 2011). 20 FDR correction at the level of individual loci means that the proportion of expected false positive 21 loci is controlled, not the proportion of false positive regions. Statistically, this is a critical point 22 since FDR control of DMR detection is not guaranteed under the DML setting. In fact, many 23 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint discoveries at the loci level may constitute only a single discovery. This means that a large 1 number of correct rejections at the loci level can inflate the denominator in the FDR calculation, 2 which will artificially lower the false discovery rate of loci as compared to regions (Figure 1). 3 We were motivated to develop a procedure to control FDR at the region level and provide an 4 accurate measure of statistical significance for each region. 1. Introduction 4 It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: ioRxiv preprint addition, challenges are presented by the complex statistical dependencies observed in 1 measurements from nearby loci (Benjamini, Taylor and Irizarry, 2016), as well as different 2 within group variability across loci (Hansen, Langmead and Irizarry, 2012). Some methods 3 ignore correlation across loci (Wen et al., 2016) or biological variability from sample to sample 4 (Saito, Tsuji and Mituyama, 2014, Wu et al., 2015). Not properly accounting for both of these 5 sources of variability in DNA methylation data, however, results in misleading conclusions or 6 loss of power. For a full review of DML and DMR methods, see Shafi et al. (2017). 7 To overcome the limitations and challenges detailed above, we propose a two-stage 8 approach that first detects candidate regions and then explicitly evaluates statistical significance 9 at the region level while accounting for known sources of variability. Candidate DMRs are 10 defined by segmenting the genome into groups of CpGs that show consistent evidence of 11 differential methylation. Because the methylation levels of neighboring CpGs are highly 12 correlated, we first smooth the signal to combat loss of power due to low coverage as done by 13 Hansen, Langmead and Irizarry (2012). In the second stage, we compute a statistic for each 14 candidate DMR that takes into account variability between biological replicates and spatial 15 correlation among neighboring loci. Significance of each region is assessed via a permutation 16 procedure which uses a pooled null distribution that can be generated from as few as two 17 biological replicates, and false discovery rate is controlled using the procedure of Benjamini and 18 Hochberg (1995). Code to reproduce the analyses presented in this paper is provided in 19 Supplementary material and the open-source R package dmrseq that implements the approach is 20 available on GitHub. 21 In Section 2, we provide a detailed description of the datasets used. 1. Introduction 4 5 Many recent computational approaches have been developed with the goal of identifying 6 DMRs, but most do not provide formal inference for regions (Hansen, Langmead and Irizarry, 7 2012, Saito, Tsuji and Mituyama, 2014, Wu et al., 2015, Yu and Sun, 2016) and instead join 8 together significant DMLs. This type of procedure will suffer from the problems outlined above. 9 Other approaches can perform inference at the region level, but only for predefined regions of 10 interest or fixed sliding windows (Hebestreit, Dugas and Klein, 2013, Sun et al., 2014). Though 11 useful in targeted settings such as Reduced Representation Bisulfite Sequencing (RRBS), or 12 when we have prior knowledge of the DMR size, they are not applicable to identifying DMRs of 13 arbitrary size from WGBS. Those methods that scan the genome for DMRs and provide 14 inference at the region level do not properly control FDR (Juhling et al., 2016, Wen et al., 2016). 15 This is evidenced, for example, by the FDRs reported in the simulation studies of Wen et al. 16 (2016), which were as high as 0.85 and widely varied across scenarios. Juhling et al. (2016) also 17 do not achieve accurate FDR control in simulation studies (see Section 4.1). 18 The challenge of performing inference at the region level is complicated by several 19 factors in addition to the challenges already discussed in the context of DML analysis. The first 20 challenge is in defining the region boundaries themselves. Without prior knowledge or 21 predefined regions, we need to construct data-driven regions. Calculating a test statistic for these 22 data-driven regions of varying sizes with a known null distribution is not straightforward. In 23 The challenge of performing inference at the region level is complicated by several 19 factors in addition to the challenges already discussed in the context of DML analysis. The first 20 challenge is in defining the region boundaries themselves. Without prior knowledge or 21 predefined regions, we need to construct data-driven regions. Calculating a test statistic for these 22 data-driven regions of varying sizes with a known null distribution is not straightforward. In 23 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. 1. Introduction 4 We describe the 22 addition, challenges are presented by the complex statistical dependencies observed in 1 measurements from nearby loci (Benjamini, Taylor and Irizarry, 2016), as well as different 2 within group variability across loci (Hansen, Langmead and Irizarry, 2012). Some methods 3 ignore correlation across loci (Wen et al., 2016) or biological variability from sample to sample 4 (Saito, Tsuji and Mituyama, 2014, Wu et al., 2015). Not properly accounting for both of these 5 sources of variability in DNA methylation data, however, results in misleading conclusions or 6 loss of power. For a full review of DML and DMR methods, see Shafi et al. (2017). 7 To overcome the limitations and challenges detailed above, we propose a two-stage 8 approach that first detects candidate regions and then explicitly evaluates statistical significance 9 at the region level while accounting for known sources of variability. Candidate DMRs are 10 defined by segmenting the genome into groups of CpGs that show consistent evidence of 11 differential methylation. Because the methylation levels of neighboring CpGs are highly 12 correlated, we first smooth the signal to combat loss of power due to low coverage as done by 13 Hansen, Langmead and Irizarry (2012). In the second stage, we compute a statistic for each 14 candidate DMR that takes into account variability between biological replicates and spatial 15 correlation among neighboring loci. Significance of each region is assessed via a permutation 16 procedure which uses a pooled null distribution that can be generated from as few as two 17 biological replicates, and false discovery rate is controlled using the procedure of Benjamini and 18 Hochberg (1995). Code to reproduce the analyses presented in this paper is provided in 19 Supplementary material and the open-source R package dmrseq that implements the approach is 20 available on GitHub. 21 In Section 2, we provide a detailed description of the datasets used. We describe the 22 methodological details of the approach and detail the data processing and analysis procedure in 23 To overcome the limitations and challenges detailed above, we propose a two-stage 8 approach that first detects candidate regions and then explicitly evaluates statistical significance 9 at the region level while accounting for known sources of variability. Candidate DMRs are 10 defined by segmenting the genome into groups of CpGs that show consistent evidence of 11 differential methylation. 1. Introduction 4 Because the methylation levels of neighboring CpGs are highly 12 correlated, we first smooth the signal to combat loss of power due to low coverage as done by 13 Hansen, Langmead and Irizarry (2012). In the second stage, we compute a statistic for each 14 candidate DMR that takes into account variability between biological replicates and spatial 15 correlation among neighboring loci. Significance of each region is assessed via a permutation 16 procedure which uses a pooled null distribution that can be generated from as few as two 17 biological replicates, and false discovery rate is controlled using the procedure of Benjamini and 18 Hochberg (1995). Code to reproduce the analyses presented in this paper is provided in 19 Supplementary material and the open-source R package dmrseq that implements the approach is 20 available on GitHub. 21 In Section 2, we provide a detailed description of the datasets used. We describe the 22 methodological details of the approach and detail the data processing and analysis procedure in 23 In Section 2, we provide a detailed description of the datasets used. We describe the 22 methodological details of the approach and detail the data processing and analysis procedure in 23 In Section 2, we provide a detailed description of the datasets used. We describe the 22 methodological details of the approach and detail the data processing and analysis procedure in 23 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: ioRxiv preprint Section 3. In Section 4, we present our findings using both experimental data and simulations. 1 We demonstrate that the proposed approach assigns greater statistical significance to regions that 2 have greater biological significance in terms of potential functional roles in the regulation of 3 gene expression. 1. Introduction 4 We also evaluate sensitivity and specificity of the approach by analyzing null 4 comparisons of samples from the same biological condition, with and without adding simulated 5 DMRs. We demonstrate that dmrseq has higher sensitivity than existing approaches and 6 accurately assesses statistical significance of regions through False Discovery Rate estimation. A 7 discussion of the advantages and limitations of the method are given in Section 5. 8 Section 3. In Section 4, we present our findings using both experimental data and simulations. 1 We demonstrate that the proposed approach assigns greater statistical significance to regions that 2 have greater biological significance in terms of potential functional roles in the regulation of 3 gene expression. We also evaluate sensitivity and specificity of the approach by analyzing null 4 comparisons of samples from the same biological condition, with and without adding simulated 5 DMRs. We demonstrate that dmrseq has higher sensitivity than existing approaches and 6 accurately assesses statistical significance of regions through False Discovery Rate estimation. A 7 discussion of the advantages and limitations of the method are given in Section 5. 8 9 2. Data Description 10 It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint 2.1 Simulated data 1 2.1 Simulated data 1 Two sets of simulated data were constructed: one representing a null comparison (with no 2 DMRs) and another containing simulated DMRs. To ensure that the simulated datasets closely 3 match the characteristics of the observed experimental data, they were generated based on 4 WGBS data from a study of human dendritic cells (Pacis et al., 2015). This study estimated 5 methylation profiles of human dendritic cells from six donors before and after infection with a 6 pathogen. The null comparison was constructed by randomly partitioning the six control samples 7 (before infection) into two groups of three samples each, denoted Simulation N3. The same is 8 done for a subset of four of the samples to evaluate performance when there are only two 9 samples in each population, denoted Simulation N2. 10 2.1 Simulated data 1 Two sets of simulated data were constructed: one representing a null comparison (with no 2 DMRs) and another containing simulated DMRs. To ensure that the simulated datasets closely 3 match the characteristics of the observed experimental data, they were generated based on 4 WGBS data from a study of human dendritic cells (Pacis et al., 2015). This study estimated 5 methylation profiles of human dendritic cells from six donors before and after infection with a 6 pathogen. The null comparison was constructed by randomly partitioning the six control samples 7 (before infection) into two groups of three samples each, denoted Simulation N3. The same is 8 done for a subset of four of the samples to evaluate performance when there are only two 9 samples in each population, denoted Simulation N2. 10 Starting with the null comparisons, 3,000 simulated DMRs were added to each dataset in 11 order to evaluate specificity and sensitivity. These are denoted Simulations D2 and D3 for two 12 and three samples per population, respectively. Briefly, a DMR is constructed by sampling a 13 cluster of neighboring CpGs and simulating the number of methylated reads, conditional on 14 observed coverage, for the samples from one population from a binomial distribution. The 15 binomial probabilities are equal to the observed methylation proportions plus or minus a 16 randomly sampled difference, which varies smoothly over the region according to a function 17 similar to the tricube kernel (Cleveland, 1979) (see Section 2.4 of the Supplementary materials). 2. Data Description 10 dmrseq is generally applicable to WGBS data which contains the counts for both methylated and 11 unmethylated reads mapping to each CpG loci. This information can be obtained from raw 12 sequencing reads using the mapping software Bismark (Krueger and Andrews, 2011), as 13 described in the Supplementary materials. Specifically, CpG loci that are covered by at least one 14 read in every sample should be used in the analysis. Other methods for analysis of WGBS data 15 recommend removing CpG sites that have only a few reads in each sample, and while processed 16 data of this form may be analyzed by our approach, it is important to note that this may result in 17 a loss of power to detect regions in low-coverage areas of the genome. 18 In this study, we use our approach to identify DMRs using publically available WGBS 19 data from two different case studies, as described below. We also evaluate sensitivity and 20 specificity of DMR methods by applying them to simulated data. Summary of coverage and 21 methylation values for all datasets used can be found in Table 1 and Supplementary Figure S2. 22 For more details on data processing, see Section 1 of the Supplementary materials. 23 In this study, we use our approach to identify DMRs using publically available WGBS 19 data from two different case studies, as described below. We also evaluate sensitivity and 20 specificity of DMR methods by applying them to simulated data. Summary of coverage and 21 methylation values for all datasets used can be found in Table 1 and Supplementary Figure S2. 22 For more details on data processing, see Section 1 of the Supplementary materials. 23 In this study, we use our approach to identify DMRs using publically available WGBS 19 data from two different case studies, as described below. We also evaluate sensitivity and 20 specificity of DMR methods by applying them to simulated data. Summary of coverage and 21 methylation values for all datasets used can be found in Table 1 and Supplementary Figure S2. 22 For more details on data processing, see Section 1 of the Supplementary materials. 23 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. 2.1 Simulated data 1 18 19 2.2 UCSD Human Reference Epigenome Mapping Project 20 Data from several human tissue samples from the UCSD Human Reference Epigenome Mapping 21 Project (Schultz et al., 2015) was used to identify DMRs related to tissue type. Specifically, four 22 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint tissues were selected for performing pairwise comparisons: (1) Heart, left ventricle, (2) Heart, 1 right ventricle, (3) Sigmoid colon, and (4) Small intestine. 2 3 2.3 Murine models of leukemia 4 In this study, marrow or thymus cells from two biological replicates form each of three different 5 murine lines were extracted and genome-wide methylation levels measured with WGBS. One 6 condition consisted of a wild-type control mouse. The other two had alterations in one or both of 7 the DNMT3a or FLT3 loci, both of which have previously demonstrated implications in the 8 development of leukemia (Pacis et al., 2015). The mouse model with a wild-type DNMT3a locus 9 and a duplication of the FLT3 locus has been shown to induce ALL. The mouse model with the 10 same duplication of the FLT3 locus as well as a knock out of DNMT3a has been shown to 11 induce the more lethal and aggressive AML. The DNMT3a also plays a role in promoting DNA 12 methylation, so it is of interest to characterize the resulting differences in methylated regions 13 among the control and two different leukemia models. 14 15 tissues were selected for performing pairwise comparisons: (1) Heart, left ventricle, (2) Heart, 1 right ventricle, (3) Sigmoid colon, and (4) Small intestine. 2 3 2.3 Murine models of leukemia 4 In this study, marrow or thymus cells from two biological replicates form each of three different 5 murine lines were extracted and genome-wide methylation levels measured with WGBS. One 6 condition consisted of a wild-type control mouse. 2.1 Simulated data 1 It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: ioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint In step 1, we detect candidate regions that contain multiple loci showing evidence of a difference 1 in the smoothed pooled methylation proportion between biological conditions. For simplicity of 2 presentation, we assume there are two biological conditions 𝑠∈1,2 , with sample indices 3 𝑗∈𝐶!(see Supplementary materials Section 2.7 for the case of more than two conditions). Let 4 𝑀!" be the number of methylated reads and 𝑈!" the number of unmethylated reads for locus 𝑖 of 5 sample 𝑗 from condition 𝑠. The coverage is denoted 𝑁!", where 𝑁!" = 𝑀!" + 𝑈!". The estimate of 6 the mean methylation proportion 𝜋!" for loci 𝑖 in condition 𝑠 is taken to be the sum of methylated 7 reads from all samples in that condition divided by the sum of all reads (i.e. the coverage) from 8 all samples in condition 𝑠: 9 𝜋!" = 𝑀!" !∈!! 𝑁!" !∈!! 𝜋!" = 𝑀!" !∈!! 𝑁!" !∈!! This leads to the following estimate of methylation proportion difference 𝛽! between condition 𝑠 10 and 𝑠′ at loci 𝑖: 11 This leads to the following estimate of methylation proportion difference 𝛽! between condition 𝑠 10 and 𝑠′ at loci 𝑖: 11 This leads to the following estimate of methylation proportion difference 𝛽! between condition 𝑠 10 and 𝑠′ at loci 𝑖: 11 𝛽! = 𝜋!" −𝜋!!! 𝛽! = 𝜋!" −𝜋!!! In order to give more weight to measurements with higher coverage, this estimate pools together 12 samples within the same condition. To account for biological variability between samples and 13 further reduce influence of observations with low coverage, smoothed individual loci estimates 14 𝛽! !"##$! are obtained using a local-likelihood smoother (Loader, 1999) with smoothing weights 15 𝑤! equal to the median coverage at loci 𝑖 scaled by the average Median Absolute Deviation 16 (MAD) within the sample groups 𝛿!: 17 𝑤! = !"#$%&! !!" !! , where 𝛿! = ! ! 𝑀𝑒𝑑𝑖𝑎𝑛!∈!! !!" !!" −𝑀𝑒𝑑𝑖𝑎𝑛!∈!! !!" !!" ! 2.1 Simulated data 1 The other two had alterations in one or both of 7 the DNMT3a or FLT3 loci, both of which have previously demonstrated implications in the 8 development of leukemia (Pacis et al., 2015). The mouse model with a wild-type DNMT3a locus 9 and a duplication of the FLT3 locus has been shown to induce ALL. The mouse model with the 10 same duplication of the FLT3 locus as well as a knock out of DNMT3a has been shown to 11 induce the more lethal and aggressive AML. The DNMT3a also plays a role in promoting DNA 12 methylation, so it is of interest to characterize the resulting differences in methylated regions 13 among the control and two different leukemia models. 14 15 3. Analysis Framework 16 A two-step procedure is carried out to (1) construct de novo candidate regions, and (2) score 17 candidate regions to quantify the effect of the covariate of interest on methylation level, and 18 evaluate statistical significance by comparing them to null regions. Here we detail each stage of 19 the approach. 20 21 3.1 Construction of candidate regions 22 tissues were selected for performing pairwise comparisons: (1) Heart, left ventricle, (2) Heart, 1 right ventricle, (3) Sigmoid colon, and (4) Small intestine. 2 In this study, marrow or thymus cells from two biological replicates form each of three different 5 murine lines were extracted and genome-wide methylation levels measured with WGBS. One 6 condition consisted of a wild-type control mouse. The other two had alterations in one or both of 7 the DNMT3a or FLT3 loci, both of which have previously demonstrated implications in the 8 development of leukemia (Pacis et al., 2015). The mouse model with a wild-type DNMT3a locus 9 and a duplication of the FLT3 locus has been shown to induce ALL. The mouse model with the 10 same duplication of the FLT3 locus as well as a knock out of DNMT3a has been shown to 11 induce the more lethal and aggressive AML. The DNMT3a also plays a role in promoting DNA 12 methylation, so it is of interest to characterize the resulting differences in methylated regions 13 among the control and two different leukemia models. 14 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. 2.1 Simulated data 1 18 19 This places more emphasis on observations with high coverage and low variability within sample 20 This places more emphasis on observations with high coverage and low variability within sample 20 group (see Section 2.1 of the Supplementary material for more details). 21 This places more emphasis on observations with high coverage and low variability within sample 20 group (see Section 2 1 of the Supplementary material for more details) 21 This places more emphasis on observations with high coverage and low variability within sample 20 group (see Section 2.1 of the Supplementary material for more details). 21 group (see Section 2.1 of the Supplementary material for more details). 21 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint Candidate regions are defined by segmenting the genome into groups of loci with a 1 smoothed and scaled pooled proportion difference 𝛽! !"##$! 𝜎!(𝛽!) in the same direction that is 2 greater than some threshold in absolute value (refer to Supplementary materials Section 2.2 for 3 more details). Maximum spacing between loci within a candidate region is controlled by a 4 predetermined value, and loci at the start and end of the region with low difference values are 5 trimmed (refer to Supplementary materials Section 2.3 for more details). The threshold value 6 should be chosen liberally so that it will more or less capture all of the true differences without 7 regard to false positives, as significance of the candidate regions is assessed in the next step. 8 9 3.2 Assessing significance of regions 10 In the second step, we assess the significance of candidate regions. This task is complicated by 11 the fact that the null statistics are calculated on an enriched set of regions. In general, the null 12 distribution generated by the type of selection procedure described in the previous section is not 13 known. 2.1 Simulated data 1 A natural approach would be to carry out a permutation test to control FWER (family- 14 wise error rate), which is done by Jaffe et al. (2012) to infer DMRs from array data. However, 15 this is not feasible when we have only a few samples per population as is most often the case 16 with WGBS. Thus, we set out to construct a statistic that can be comparable across the genome 17 so that the signal can be compared among regions. Such an exchangeable statistic allows us to 18 generate an approximate null distribution by pooling genomewide candidate regions detected 19 from permutations. 20 To generate an approximately exchangeable region statistic that measures the strength of 21 Candidate regions are defined by segmenting the genome into groups of loci with a To generate an approximately exchangeable region statistic that measures the strength of 21 methylation difference, we need to account for sources of variation that are known to vary across 22 the genome, including biological variability from sample to sample (Hansen, Langmead and 23 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint Irizarry, 2012), as well as covariance of nearby loci (Benjamini, Taylor and Irizarry, 2016). 1 Failing to do so may result in large test statistics just by chance for regions with high variability, 2 leading to increased FDR or decreased power. For example, if we use an area-based statistic 3 (Hansen, Langmead and Irizarry, 2012) or a mean difference statistic averaged across loci, power 4 to detect DMRs is greatly reduced in simulation studies (Supplementary Figure S5 and 5 Supplementary materials Section 4.1). 6 Since we need to compute the statistic over potentially hundreds of thousands of 7 candidate regions, we also favor an approach that provides efficient and stable estimation 8 procedures. For these reasons, we make use of generalized least squares (GLS) regression model 9 with a nested autoregressive correlated error structure for the effect of interest on transformed 10 methylation proportions, the advantages of which are described in detail in the next subsections. 11 12 3.2.1 Estimation of region statistics with Generalized Least Squares models 13 To account for sampling variability, we assume that methylation counts for region 𝑟 are 14 Binomially-distributed with probability 𝑝!"#, where 15 𝑀!"# \| 𝑁!"#, 𝑝!"# ~ 𝐵𝑖𝑛(𝑁!"#, 𝑝!"#). 16 To model biological variability, we allow the binomial proportion for samples in condition 17 𝑠∈1,2 to vary according to a beta distribution with shape parameters 𝛼!"# and 𝛽!"#, where 18 To model biological variability, we allow the binomial proportion for samples in condition 17 To model biological variability, we allow the binomial proportion for samples in condition 17 𝑠∈1,2 to vary according to a beta distribution with shape parameters 𝛼!"# and 𝛽!"#, where 18 𝑝!"! ~ 𝐵𝑒𝑡𝑎(𝛼!"#, 𝛽!"#). 19 𝑠∈1,2 to vary according to a beta distribution with shape parameters 𝛼!"# and 𝛽!"#, where 18 𝑝!"! ~ 𝐵𝑒𝑡𝑎(𝛼!"#, 𝛽!"#). biological conditions. 2 Candidate regions are defined by segmenting the genome into groups of loci with a This helps us to form 18 a statistic involving the transformed proportions that is exchangeable across regions that have 19 different mean methylation values. 20 We assume a linear effect on the arcsine link-transformed methylation proportion 21 parameters: 22 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available u Our approach models transformed methylation proportions using GLS to obtain an 1 approximation of the effect of interest. While directly modeling counts with either a Beta- 2 Binomial Generalized Linear Model (GLM) or a Generalized Linear Mixed Model (GLMM) 3 would allow us to accommodate complex covariance structures across samples and loci, it also 4 results in complex likelihoods that require iterative maximization for each candidate region. 5 Our approach models transformed methylation proportions using GLS to obtain an 1 approximation of the effect of interest. While directly modeling counts with either a Beta- 2 Binomial Generalized Linear Model (GLM) or a Generalized Linear Mixed Model (GLMM) 3 would allow us to accommodate complex covariance structures across samples and loci, it also 4 results in complex likelihoods that require iterative maximization for each candidate region. 5 Further, these procedures are subject to instability of estimation for methylation levels near the 6 boundaries (zero and one) or non-identifiability in the case of separation as they occur in GLM 7 (Gelman et al., 2008) and GLMM (Abrahantes and Aerts, 2012) estimation. GLS models, in 8 contrast, are efficient and stable to estimate due to the availability of approximate closed-form 9 parameter estimates. Though GLS does not model counts directly, we incorporate information 10 lost after transformation of methylation proportions through specification of a variance estimate 11 that depends on coverage. 12 Further, these procedures are subject to instability of estimation for methylation levels near the 6 boundaries (zero and one) or non-identifiability in the case of separation as they occur in GLM 7 (Gelman et al., 2008) and GLMM (Abrahantes and Aerts, 2012) estimation. GLS models, in 8 contrast, are efficient and stable to estimate due to the availability of approximate closed-form 9 parameter estimates. Though GLS does not model counts directly, we incorporate information 10 lost after transformation of methylation proportions through specification of a variance estimate 11 that depends on coverage. Candidate regions are defined by segmenting the genome into groups of loci with a 19 Let 𝜋!"# = !!"# !!"#!!!"# denote the mean of this Beta distribution. We are interested in estimating and 20 assessing the significance of the difference in mean methylation levels acro 21 assessing the significance of the difference in mean methylation levels across a region 𝑟 for two 21 biological conditions 22 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: ioRxiv preprint Our approach models transformed methylation proportions using GLS to obtain an 1 approximation of the effect of interest. While directly modeling counts with either a Beta- 2 Binomial Generalized Linear Model (GLM) or a Generalized Linear Mixed Model (GLMM) 3 would allow us to accommodate complex covariance structures across samples and loci, it also 4 results in complex likelihoods that require iterative maximization for each candidate region. 5 Further, these procedures are subject to instability of estimation for methylation levels near the 6 boundaries (zero and one) or non-identifiability in the case of separation as they occur in GLM 7 (Gelman et al., 2008) and GLMM (Abrahantes and Aerts, 2012) estimation. GLS models, in 8 contrast, are efficient and stable to estimate due to the availability of approximate closed-form 9 parameter estimates. Though GLS does not model counts directly, we incorporate information 10 lost after transformation of methylation proportions through specification of a variance estimate 11 that depends on coverage. 12 We choose the arcsine link function 𝑍!"# = 𝑎𝑟𝑐𝑠𝑖𝑛 2𝑀!"# 𝑁!"# −1 to obtain 13 transformed methylation proportions, as proposed by (Park and Wu, 2016) for DML analysis, for 14 its desirable ability to stabilize the dependence of the variance on the mean methylation level. 15 While the variance of methylation proportions 𝑀!"# 𝑁!"# depends on the mean parameter 𝜋!"#, 16 the variance of 𝑍!"# only depends on coverage 𝑁!"# and the dispersion of the Beta-Binomial 17 distribution (refer to Supplementary materials Section 2.6 for more details). Candidate regions are defined by segmenting the genome into groups of loci with a 12 We choose the arcsine link function 𝑍!"# = 𝑎𝑟𝑐𝑠𝑖𝑛 2𝑀!"# 𝑁!"# −1 to obtain 13 transformed methylation proportions, as proposed by (Park and Wu, 2016) for DML analysis, for 14 its desirable ability to stabilize the dependence of the variance on the mean methylation level. 15 While the variance of methylation proportions 𝑀!"# 𝑁!"# depends on the mean parameter 𝜋!"#, 16 the variance of 𝑍!"# only depends on coverage 𝑁!"# and the dispersion of the Beta-Binomial 17 distribution (refer to Supplementary materials Section 2.6 for more details). This helps us to form 18 a statistic involving the transformed proportions that is exchangeable across regions that have 19 different mean methylation values. 20 We choose the arcsine link function 𝑍!"# = 𝑎𝑟𝑐𝑠𝑖𝑛 2𝑀!"# 𝑁!"# −1 to obtain 13 transformed methylation proportions, as proposed by (Park and Wu, 2016) for DML analysis, for 14 its desirable ability to stabilize the dependence of the variance on the mean methylation level. 15 We assume a linear effect on the arcsine link-transformed methylation proportion 21 parameters: 22 We assume a linear effect on the arcsine link-transformed methylation proportion 21 parameters: 22 We assume a linear effect on the arcsine link-transformed methylation proportion 21 parameters: 22 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint 𝑎𝑟𝑐𝑠𝑖𝑛2𝜋!"# −1 = 𝛽!!"1[!!!] !! !!! + 𝛽!!𝑋! = 𝑿𝜷! Here 𝛽!!" are loci-specific intercept terms that account for variation on overall methylation levels 1 across the region, where 𝑙= 1, … 𝐿! and 𝐿! denotes the number of loci in region 𝑟. The 2 coefficient for the effect of interest (e.g. biological group) is 𝛽!!. We denote the design matrix as 3 𝑿 and the (𝐿! + 1)-length vector of all coefficients 𝛽!"!, 𝛽!"!, … , 𝛽!!!!, 𝛽!! as 𝜷!. This leads 4 to the following model for the transformed response 𝒁𝒓= 𝑍!!!, … , 𝑍!! Candidate regions are defined by segmenting the genome into groups of loci with a !" in region 𝑟 5 Here 𝛽!!" are loci-specific intercept terms that account for variation on overall methylation levels 1 across the region, where 𝑙= 1, … 𝐿! and 𝐿! denotes the number of loci in region 𝑟. The 2 coefficient for the effect of interest (e.g. biological group) is 𝛽!!. We denote the design matrix as 3 𝑿 and the (𝐿! + 1)-length vector of all coefficients 𝛽!"!, 𝛽!"!, … , 𝛽!!!!, 𝛽!! as 𝜷!. This leads 4 to the following model for the transformed response 𝒁𝒓= 𝑍!!!, … , 𝑍!! !" in region 𝑟 5 𝒁𝒓= 𝑿𝜷𝒓+ 𝜺𝒓 𝒁𝒓= 𝑿𝜷𝒓+ 𝜺𝒓 where we assume that 𝐸𝜺𝒓= 0 and 𝑉𝑎𝑟𝜺𝒓= 𝑽𝒓,which can be fit by GLS given an estimate 6 of the covariance matrix 𝑽𝒓. Since GLS allows arbitrary covariance structures, we use an 7 autoregressive correlation structure to account for the correlation of methylation levels among 8 nearby loci. To account for the dependence of the variance on coverage as mentioned above, we 9 use variance weights. More details on the specific structure and estimation of 𝑽𝒓 are given in the 10 next section. 11 With the above model, we assess the strength of the effect of the covariate of interest on 12 methylation level within region 𝑟 using the t-statistic 𝑡! from the Wald test of the null hypothesis 13 that 𝛽!! = 0. Parameter estimates and their standard errors are obtained with the `gls` function in 14 the `nlme` package (Pinheiro et al., 2017). Significance is evaluated by permutation using a 15 pooled null distribution as described in detail in Section 3.2.3. 16 17 3.2.2 Covariance of methylation levels within regions 18 In the estimation of the covariance matrix 𝑽𝒓, we take into account biological variability through 19 variance weighting, and correlation of nearby loci through an autocorrelation structure. The 20 variance weighting is done to account for the dependence of the variance of transformed values 21 With the above model, we assess the strength of the effect of the covariate of interest on 12 methylation level within region 𝑟 using the t-statistic 𝑡! from the Wald test of the null hypothesis 13 that 𝛽!! = 0. Parameter estimates and their standard errors are obtained with the `gls` function in 14 the `nlme` package (Pinheiro et al., 2017). Significance is evaluated by permutation using a 15 pooled null distribution as described in detail in Section 3.2.3. Candidate regions are defined by segmenting the genome into groups of loci with a 16 With the above model, we assess the strength of the effect of the covariate of interest on 12 methylation level within region 𝑟 using the t-statistic 𝑡! from the Wald test of the null hypothesis 13 that 𝛽!! = 0. Parameter estimates and their standard errors are obtained with the `gls` function in 14 the `nlme` package (Pinheiro et al., 2017). Significance is evaluated by permutation using a 15 pooled null distribution as described in detail in Section 3.2.3. 16 3.2.2 Covariance of methylation levels within regions 18 In the estimation of the covariance matrix 𝑽𝒓, we take into account biological variability through 19 variance weighting, and correlation of nearby loci through an autocorrelation structure. The 20 variance weighting is done to account for the dependence of the variance of transformed values 21 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint 𝑍!"# on coverage. This variance depends non-linearly on 𝑁!"# (Supplement Section 2.6), but in 1 order to enable efficient closed-form estimation with GLS, we further approximate it by 2 𝑍!"# on coverage. This variance depends non-linearly on 𝑁!"# (Supplement Section 2.6), but in 1 order to enable efficient closed-form estimation with GLS, we further approximate it by 2 𝑉𝑎𝑟(𝑍!"#) ≈ 𝜎!! 𝑁!"# 𝑉𝑎𝑟(𝑍!"#) ≈ 𝜎!! 𝑁!"# In addition, in order to construct a valid permutation test where the variance conditional on the 3 effect of interest is invariant to permutation, we assume this variance identical for all samples at 4 a given loci by approximating 𝑁!"# by 𝑚𝑒𝑑𝑖𝑎𝑛! 𝑁!"# = 𝑁!.!. 5 To model correlation of nearby loci, we use the flexible continuous autoregressive 6 correlation structure of order 1, abbreviated CAR(1). Under CAR(1), the correlation parameter 7 depends on the length of the interval between the two observations considered in the following 8 manner 9 𝜌! 𝜏= 𝑒!!! ! 𝜌! Candidate regions are defined by segmenting the genome into groups of loci with a where 𝜏 is the length of the interval between two observations and 𝜙! is the positive continuous- 10 time autoregressive coefficient (following the notation of Jones and Boadi-Boateng (1991)) for 11 region 𝑟. Thus, for subject 𝑗, the predicted methylation value for loci 𝑖 at location 𝑡!"# in region 𝑟 12 given the methylation value at loci 𝑖−1 is 13 𝑍!"# = 𝑍!!!,!" 𝑒!!! !!"#!!!!!,!" 𝑍!"# = 𝑍!!!,!" 𝑒!!! !!"#!!!!!,!" 𝑍!"# = 𝑍!!!,!" 𝑒!!! !!"#!!!!!,!" If the error variance of the CAR1 process is 𝜎!" ! = !!! !!.!, and we let the correlation structure be 14 If the error variance of the CAR1 process is 𝜎!" ! = !!.!, and we let the correlation structure be 14 nested within subject (i.e. such that observations from two subjects are independent), it follows 15 that the covariance matrix for a given sample can be written 16 nested within subject (i.e. such that observations from two subjects are independent), it follows 15 that the covariance matrix for a given sample can be written 16 nested within subject (i.e. such that observations from two subjects are independent), it follows 15 that the covariance matrix for a given sample can be written 16 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint 𝑽!" = 𝜎! ! 1 𝑁!.! 𝑒!!! !!!"!!!!" 𝑁!.!𝑁!.! ⋯ 𝑒!!! !!!"!!!!!" 𝑁!.!𝑁!!.! 𝑒!!! !!!"!!!!" 𝑁!.!𝑁!.! 1 𝑁!.! ⋯ 𝑒!!! !!!"!!!!!" 𝑁!.!𝑁!!.! ⋮ ⋮ ⋱ ⋮ 𝑒!!! !!!!"!!!!" 𝑁!.!𝑁!!.! 𝑒!!! !!!!"!!!!" 𝑁!.!𝑁!!.! ⋯ 1 𝑁!!.! and for two subjects 𝑗 and 𝑗′, 𝐶𝑜𝑣𝑍!"#, 𝑍!"!! = 0. 1 and for two subjects 𝑗 and 𝑗′, 𝐶𝑜𝑣𝑍!"#, 𝑍!"!! = 0. 1 The estimation of 𝜙!is computationally efficient to carry out on small to moderately sized 2 regions. However, for larger regions with more than 40 loci we use the slightly simpler AR(1) 3 correlation structure since it is many times faster to compute. Candidate regions are defined by segmenting the genome into groups of loci with a It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint coefficient of interest under CAR(1) generally converge to the estimates under AR(1) as the 1 number of loci increases (Supplementary Figure S1 and Section 2.5). 2 3 3.2.3 Permutation to generate a null set of regions 4 The values of the covariate of interest (e.g. biological group) are permuted and the previous steps 5 repeated in order to generate a set of statistics under the null hypothesis. Since the statistics 6 account for known sources of variation that would otherwise prevent to comparison of regions 7 across the genome, we can pool them together to form an approximate null distribution with as 8 few as two samples per population. The empirical p-value is calculated by comparing the 9 observed test statistics to the entire null set of statistics from all permutations. Control of FDR is 10 carried out by adjusting the p-values using the procedure of Benjamini and Hochberg (1995). 11 12 4. Results 13 For each of the datasets described in Section 2, we applied dmrseq, as well as three widely used 14 methods for DMR detection: BSmooth (Hansen, Langmead and Irizarry, 2012), DSS (Park and 15 Wu, 2016), and metilene (Juhling et al., 2016). Each approach was evaluated based on the 16 criteria detailed in the next subsections. For specific details on software implementation, refer to 17 the Supplementary materials (Section 3). 18 19 4.1 Simulation using dendritic cell data 20 Specificity was evaluated by identifying DMRs in null comparisons of two (N2) and three (N3) 21 samples per group. Sensitivity was evaluated by identifying simulated DMRs in comparisons of 22 two (D2) and three (D3) samples per group. Performance of each method is assessed by its 23 coefficient of interest under CAR(1) generally converge to the estimates under AR(1) as the 1 number of loci increases (Supplementary Figure S1 and Section 2.5). 2 coefficient of interest under CAR(1) generally converge to the estimates under AR(1) as the 1 number of loci increases (Supplementary Figure S1 and Section 2.5). 2 Candidate regions are defined by segmenting the genome into groups of loci with a This discrete formulation assumes 4 that observations are equally spaced, and that observations that are separated by lag 1 are 5 correlated with region-specific correlation parameter 𝜌!. In addition, observations that are 6 separated by 𝑚 positions are correlated by 𝜌! !. This results in a covariance matrix for 𝒁𝒓 from 7 region 𝑟, subject 𝑗 of 8 𝑽!" = 𝜎! ! ! !!.! !! !!.!!!.! ⋯ !! !!!! !!.!!!!.! !! !!.!!!.! ! !!.! ⋯ !! !!!! !!.!!!!.! ⋮ ⋮ ⋱ ⋮ !! !!!! !!.!!!!.! !! !!!! !!.!!!!.! ⋯ ! !!!.! 9 d i th t f t bj t 𝑗 d 𝑗′ 𝐶 𝑍 𝑍 0 10 𝑽!" = 𝜎! ! ! !!.! !! !!.!!!.! ⋯ !! !!!! !!.!!!!.! !! !!.!!!.! ! !!.! ⋯ !! !!!! !!.!!!!.! ⋮ ⋮ ⋱ ⋮ !! !!!! !!.!!!!.! !! !!!! !!.!!!!.! ⋯ ! !!!.! 9 and again we assume that for two subjects 𝑗 and 𝑗′, 𝐶𝑜𝑣𝑍!"#, 𝑍!"!! = 0. 10 and again we assume that for two subjects 𝑗 and 𝑗′, 𝐶𝑜𝑣𝑍!"#, 𝑍!"!! = 0. 10 The CAR(1) structure simplifies to the AR(1) process under certain conditions when 11 observations are equally spaced (Jones and Boadi-Boateng, 1991). Thus the discrete AR(1) can 12 be viewed as an approximation of the CAR(1) when correlations are positive and the two provide 13 increasingly more similar estimates as observations approach constant spacing. Indeed, when 14 comparing model fits under both correlation structures in simulated data, the t-statistics for the 15 The CAR(1) structure simplifies to the AR(1) process under certain conditions when 11 observations are equally spaced (Jones and Boadi-Boateng, 1991). Thus the discrete AR(1) can 12 be viewed as an approximation of the CAR(1) when correlations are positive and the two provide 13 increasingly more similar estimates as observations approach constant spacing. Indeed, when 14 comparing model fits under both correlation structures in simulated data, the t-statistics for the 15 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. 4.1 Simulation using dendritic cell data 20 Specificity was evaluated by identifying DMRs in null comparisons of two (N2) and three (N3) 21 samples per group. Sensitivity was evaluated by identifying simulated DMRs in comparisons of 22 two (D2) and three (D3) samples per group. Performance of each method is assessed by its 23 Specificity was evaluated by identifying DMRs in null comparisons of two (N2) and three (N3) 21 samples per group. Sensitivity was evaluated by identifying simulated DMRs in comparisons of 22 two (D2) and three (D3) samples per group. Performance of each method is assessed by its 23 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: ioRxiv preprint ability to identify as many of the simulated DMRs as possible, while identifying as few DMRs as 1 possible in the null comparison. 2 dmrseq did not identify any DMRs at the 0.05 level for the null comparisons N2 or N3 3 (Table 2). This remains true even when increasing the FDR threshold to 0.5 in both settings. In 4 contrast, metilene identified a small number of DMRs, DSS identified many hundreds, and 5 BSmooth tens of thousands using default settings (specific parameter specifications provided in 6 Supplementary materials Section 2.6). When applied to the datasets with simulated DMRs (D2 7 and D3), dmrseq is able to accurately control the False Discovery Rate, whereas metilene cannot 8 (Figure 2, Supplementary Figure S3). Note that analogous results cannot be obtained from DSS 9 or BSmooth, as there is no way to specify FDR level. 10 dmrseq did not identify any DMRs at the 0.05 level for the null comparisons N2 or N3 3 (Table 2). This remains true even when increasing the FDR threshold to 0.5 in both settings. In 4 contrast, metilene identified a small number of DMRs, DSS identified many hundreds, and 5 BSmooth tens of thousands using default settings (specific parameter specifications provided in 6 Supplementary materials Section 2.6). When applied to the datasets with simulated DMRs (D2 7 and D3), dmrseq is able to accurately control the False Discovery Rate, whereas metilene cannot 8 (Figure 2, Supplementary Figure S3). Note that analogous results cannot be obtained from DSS 9 or BSmooth, as there is no way to specify FDR level. 10 BSmooth and DSS identify similar numbers of False Positive regions in D2 and D3 11 compared to the null setting of N2 and N3, and far more than dmrseq and metilene (Table 3). 12 Although both BSmooth and DSS have favorable numbers of TPs, it is clear that this comes at 13 the expense of lack of control of FDR (Figure 3). Similarly, metilene has favorable numbers of 14 FPs, but this comes at the expense of low power. Further, even at similar observed FDR levels, 15 dmrseq achieves higher power levels than the alternative methods. 16 BSmooth and DSS identify similar numbers of False Positive regions in D2 and D3 11 compared to the null setting of N2 and N3, and far more than dmrseq and metilene (Table 3). 12 Although both BSmooth and DSS have favorable numbers of TPs, it is clear that this comes at 13 the expense of lack of control of FDR (Figure 3). ability to identify as many of the simulated DMRs as possible, while identifying as few DMRs as 1 possible in the null comparison. 2 Similarly, metilene has favorable numbers of 14 FPs, but this comes at the expense of low power. Further, even at similar observed FDR levels, 15 dmrseq achieves higher power levels than the alternative methods. 16 Although FDR thresholds are not available for BSmooth or DSS, we also investigated the 17 sensitivity and specificity of other settings beyond defaults of the thresholds at the single-loci 18 level (the loci t-statistic cutoff for BSmooth, and the loci p-value for DSS). Making these 19 thresholds more conservative generally reduced the numbers of False Positives, but once again 20 dmrseq was consistently able to identify more True Positives at similar numbers of False 21 Positives (See Supplementary Results and Figure S4). 22 Although FDR thresholds are not available for BSmooth or DSS, we also investigated the 17 sensitivity and specificity of other settings beyond defaults of the thresholds at the single-loci 18 level (the loci t-statistic cutoff for BSmooth, and the loci p-value for DSS). Making these 19 thresholds more conservative generally reduced the numbers of False Positives, but once again 20 dmrseq was consistently able to identify more True Positives at similar numbers of False 21 Positives (See Supplementary Results and Figure S4). 22 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint We also stress that although lower False Positive rates could be achieved in this 1 simulation study for BSmooth and DSS, individual loci thresholds do not correspond directly to 2 specific FDRs at the region level. As a result, in practice, one must choose a threshold either by 3 default settings, or by trial and error. 4 5 4.2 Human tissue and murine leukemia experimental data 6 The human tissue and murine leukemia studies were evaluated empirically based on the observed 7 association of DMRs with differential expression by RNA-seq. 4.2 Human tissue and murine leukemia experimental data 6 The human tissue and murine leukemia studies were evaluated empirically based on the observed 7 association of DMRs with differential expression by RNA-seq. Differentially expressed (DE) 8 genes were identified using DESeq2 version 1.14.1 (Love, Huber and Anders, 2014). To assess 9 functional relevance of the results, detected DMRs that overlap promoter regions of DE genes 10 were assessed for signal in the expected direction. Specifically, a DMR - DE gene pair is 11 expected to have higher methylation values in the sample group with lower expression. The odds 12 that the DMR and DE statistics are in opposing directions are calculated at various FDR cutoffs 13 for dmrseq and metilene to assess whether top-ranked DMRs are more likely to be biologically 14 relevant. The same is done for various cutoffs for the numbers of top-ranking regions by effect 15 size. Additionally, for each cutoff we calculate the number of CpGs covered and the proportion 16 of detected DMRs that are within 2kb (from the center of the region) of a promoter region of a 17 DE gene. 18 The human tissue and murine leukemia studies were evaluated empirically based on the observed 7 association of DMRs with differential expression by RNA-seq. Differentially expressed (DE) 8 genes were identified using DESeq2 version 1.14.1 (Love, Huber and Anders, 2014). To assess 9 functional relevance of the results, detected DMRs that overlap promoter regions of DE genes 10 were assessed for signal in the expected direction. Specifically, a DMR - DE gene pair is 11 expected to have higher methylation values in the sample group with lower expression. The odds 12 that the DMR and DE statistics are in opposing directions are calculated at various FDR cutoffs 13 for dmrseq and metilene to assess whether top-ranked DMRs are more likely to be biologically 14 relevant. The same is done for various cutoffs for the numbers of top-ranking regions by effect 15 size. Additionally, for each cutoff we calculate the number of CpGs covered and the proportion 16 of detected DMRs that are within 2kb (from the center of the region) of a promoter region of a 17 DE gene. 18 To qualitatively assess the ability of the dmrseq region-level summary statistic to rank 19 DMRs as compared to other methods, we display example regions from the human tissue and 20 murine leukemia studies. ability to identify as many of the simulated DMRs as possible, while identifying as few DMRs as 1 possible in the null comparison. 2 Differentially expressed (DE) 8 genes were identified using DESeq2 version 1.14.1 (Love, Huber and Anders, 2014). To assess 9 functional relevance of the results, detected DMRs that overlap promoter regions of DE genes 10 were assessed for signal in the expected direction. Specifically, a DMR - DE gene pair is 11 expected to have higher methylation values in the sample group with lower expression. The odds 12 that the DMR and DE statistics are in opposing directions are calculated at various FDR cutoffs 13 for dmrseq and metilene to assess whether top-ranked DMRs are more likely to be biologically 14 relevant. The same is done for various cutoffs for the numbers of top-ranking regions by effect 15 size. Additionally, for each cutoff we calculate the number of CpGs covered and the proportion 16 of detected DMRs that are within 2kb (from the center of the region) of a promoter region of a 17 DE gene. 18 To qualitatively assess the ability of the dmrseq region-level summary statistic to rank 19 DMRs as compared to other methods, we display example regions from the human tissue and 20 We also stress that although lower False Positive rates could be achieved in this 1 simulation study for BSmooth and DSS, individual loci thresholds do not correspond directly to 2 specific FDRs at the region level. As a result, in practice, one must choose a threshold either by 3 default settings, or by trial and error. 4 4.2 Human tissue and murine leukemia experimental data 6 These examples illustrate the increased variability of regions that are 21 highly ranked by naïve statistics but not dmrseq (Figure 5). We include a DMR with concordant 22 rankings that exhibits clear differences between two human tissue types (Figure 5A). In contrast, 23 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint the regions with discordant rankings between dmrseq q-value and mean difference (Figure 5B) 1 and area statistics (Figure 5C) exhibit considerable variability between samples or loci (See 2 Supplementary materials Section 2.8 for more details). 3 the regions with discordant rankings between dmrseq q-value and mean difference (Figure 5B) 1 and area statistics (Figure 5C) exhibit considerable variability between samples or loci (See 2 Supplementary materials Section 2.8 for more details). 3 4.2.1 Tissue specificity in human samples 5 For DSS, metiline, and dmrseq, the number of DMRs found (Table 4) parallels the numbers of 6 DE genes found by DESeq2 (Supplementary Table S2), but DSS generally found far more 7 DMRs and metline far fewer. For BSmooth, however, the number of DMRs identified was 8 similar for all comparisons. This happens because the cutoff for the individual loci statistics is set 9 by default at a quantile of the observed statistics, resulting in a similar number of loci being 10 deemed significant. 11 4.2.1 Tissue specificity in human samples 5 For DSS, metiline, and dmrseq, the number of DMRs found (Table 4) parallels the numbers of 6 DE genes found by DESeq2 (Supplementary Table S2), but DSS generally found far more 7 DMRs and metline far fewer. For BSmooth, however, the number of DMRs identified was 8 similar for all comparisons. This happens because the cutoff for the individual loci statistics is set 9 by default at a quantile of the observed statistics, resulting in a similar number of loci being 10 deemed significant. 4.2.2 DNMT3a loss in murine leukemia models 1 4.2.2 DNMT3a loss in murine leukemia models 1 In the murine leukemia models, dmrseq finds the most DMRs in the comparison of AML and the 2 control (Table 5), which is also the comparison for which the most DE genes were identified (see 3 Supplementary Table S4). In contrast, DSS and metline both find the most DMRs in the 4 comparison with the fewest DE genes identified, and BSmooth identified similar numbers of 5 DMRs in each comparison, each with far more DMRs than the other methods. 6 DMRs in each comparison, each with far more DMRs than the other methods. 6 The murine leukemia DMRs found by dmrseq are enriched for inverse associations with 7 DE genes, and this enrichment is stronger for DMRs with lower FDRs (Figure 4). Additionally, 8 enrichment is generally stronger than that of BSmooth, DSS, and metline. While metiline also 9 provides an FDR estimate, there is no consistent association between the FDR ranking and 10 strength of association with expression. Similar to the tissue specificity analysis, BSmooth and 11 DSS DMRs with highest effect sizes exhibit comparable enrichment to dmrseq, with metilene 12 considerably lower, and the enrichment when including all DMRs often drops lower for 13 BSmooth, DSS, or metline than for dmrseq (Supplementary Figures S8 and S9). 14 15 5. Discussion 16 We have described dmrseq, a method useful for discovering and prioritizing DMRs from WGBS 17 data. The approach is based on rigorous statistical reasoning and is the first method that permits 18 accurate inference on DMRs that are found by scanning the genome. By developing a 19 transformation that results in summary statistics from candidate regions being exchangeable, we 20 are able to borrow strength across the genome to build a null distribution that permits inference 21 with a sample size as small as 2. We have demonstrated how the method clearly outperforms 22 The murine leukemia DMRs found by dmrseq are enriched for inverse associations with 7 DE genes, and this enrichment is stronger for DMRs with lower FDRs (Figure 4). Additionally, 8 enrichment is generally stronger than that of BSmooth, DSS, and metline. While metiline also 9 provides an FDR estimate, there is no consistent association between the FDR ranking and 10 strength of association with expression. 4.2 Human tissue and murine leukemia experimental data 6 11 The tissue-specific DMRs found by dmrseq are enriched for inverse associations with DE 12 genes, and this enrichment is stronger for DMRs with lower FDRs (Figure 4). Additionally, 13 enrichment of dmrseq DMRs is generally stronger than that of alternative methods. While 14 metiline also provides an FDR estimate, there is no consistent association between the FDR 15 ranking and strength of association with expression. DMRs identified by BSmooth and DSS 16 cannot be ranked by FDR and the default settings may not be ideal, so we also rank DMRs by 17 effect size (raw methylation difference) with optimized parameter settings (see Supplementary 18 materials Section 3.2). The BSmooth and DSS DMRs with highest effect sizes exhibit 19 comparable enrichment to dmrseq, with metilene considerably lower (Supplementary Figures S6 20 and S7). However, arbitrary cutoffs of effect size do not directly correspond to significance level, 21 and the enrichment when including all DMRs is highest for dmrseq (Figure 4). 22 23 23 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: ioRxiv preprint 4.2.2 DNMT3a loss in murine leukemia models 1 In the murine leukemia models, dmrseq finds the most DMRs in the comparison of AML and the 2 control (Table 5), which is also the comparison for which the most DE genes were identified (see 3 Supplementary Table S4). In contrast, DSS and metline both find the most DMRs in the 4 comparison with the fewest DE genes identified, and BSmooth identified similar numbers of 5 DMRs in each comparison, each with far more DMRs than the other methods. 6 4.2.2 DNMT3a loss in murine leukemia models 1 and MASON, C. E. (2012). methylKit: a comprehensive R package for the analysis of 20 genome-wide DNA methylation profiles. Genome Biol 13, R87. 21 ARYEE, M. J., JAFFE, A. E., CORRADA-BRAVO, H., LADD-ACOSTA, C., FEINBERG, A. P., HANSEN, 22 K. D. and IRIZARRY, R. A. (2014). Minfi: a flexible and comprehensive Bioconductor 23 package for the analysis of Infinium DNA methylation microarrays. Bioinformatics 30, 24 1363-1369. 25 currently used tools with several experimental data examples and Monte Carlo simulation. The 1 method is implemented as open source software in the form of an R package. 2 currently used tools with several experimental data examples and Monte Carlo simulation. The 1 method is implemented as open source software in the form of an R package. 2 4.2.2 DNMT3a loss in murine leukemia models 1 Similar to the tissue specificity analysis, BSmooth and 11 DSS DMRs with highest effect sizes exhibit comparable enrichment to dmrseq, with metilene 12 considerably lower, and the enrichment when including all DMRs often drops lower for 13 BSmooth, DSS, or metline than for dmrseq (Supplementary Figures S8 and S9). 14 We have described dmrseq, a method useful for discovering and prioritizing DMRs from WGBS data. The approach is based on rigorous statistical reasoning and is the first method that permits accurate inference on DMRs that are found by scanning the genome. By developing a transformation that results in summary statistics from candidate regions being exchangeable, we 20 are able to borrow strength across the genome to build a null distribution that permits inference 21 with a sample size as small as 2. We have demonstrated how the method clearly outperforms 22 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint currently used tools with several experimental data examples and Monte Carlo simulation. The 1 method is implemented as open source software in the form of an R package. 2 3 Supplementary Material 4 The reader is referred to the online Supplementary Materials for further details of data 5 acquisition and processing, additional methodological details, software implementation details, 6 and supplementary results. In addition, annotated R scripts for the simulation and case study 7 analyses are available in the GitHub repository https://github.com/kdkorthauer/dmrseqPaper, and 8 the R package dmrseq is available on GitHub at https://github.com/kdkorthauer/dmrseq. 9 10 Acknowledgement 11 Conflict of Interest: None declared 12 13 Funding 14 The work of all authors was partially by NIH R01 grants HG005220 and GM083084. 15 16 References ABRAHANTES, J. C. and AERTS, M. (2012). A solution to separation for clustered binary data. 17 Statistical Modelling 12, 3-27. 18 AKALIN, A., KORMAKSSON, M., LI, S., GARRETT-BAKELMAN, F. E., FIGUEROA, M. E., MELNICK, 19 A. Supplementary Material 4 The reader is referred to the online Supplementary Materials for further details of data 5 acquisition and processing, additional methodological details, software implementation details, 6 and supplementary results. In addition, annotated R scripts for the simulation and case study 7 analyses are available in the GitHub repository https://github.com/kdkorthauer/dmrseqPaper, and 8 the R package dmrseq is available on GitHub at https://github.com/kdkorthauer/dmrseq. 9 10 ARYEE, M. J., JAFFE, A. E., CORRADA-BRAVO, H., LADD-ACOSTA, C., FEINBERG, A. P., HANSEN, 22 K. D. and IRIZARRY, R. A. (2014). Minfi: a flexible and comprehensive Bioconductor 23 package for the analysis of Infinium DNA methylation microarrays. Bioinformatics 30, 24 1363-1369. 25 Acknowledgement 11 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint BENJAMINI, Y. and HOCHBERG, Y. (1995). Controlling the False Discovery Rate: A Practical and 1 Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society Series B 2 57, 289-300. 3 BENJAMINI, Y. and HOCHBERG, Y. (1995). Controlling the False Discovery Rate: a Practical and 4 Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society B 57, 5 289-300. 6 BENJAMINI, Y., TAYLOR, J. and IRIZARRY, R. A. (2016). Selection Corrected Statistical Inference 7 for Region Detection with High-througput Assays. bioRxiv. 8 BIRD, A. (2002). DNA methylation patterns and epigenetic memory. Genes & Development 16, 9 6-21. 10 CLEVELAND, W. S. (1979). Robust locally weighted regression and smoothing scatterplots. 11 Journal of the American Statistical Association 74, 829-836. 12 DOLZHENKO, E. and SMITH, A. D. (2014). Using beta-binomial regression for high-precision 13 differential methylation analysis in multifactor whole genome bisulfite sequencing 14 experiments. BMC Bioinformatics 15, 215. 15 DOLZHENKO, E. and SMITH, A. D. (2014). Using beta-binomial regression for high-precision 16 differential methylation analysis in multifactor whole-genome bisulfite sequencing 17 experiments. BMC Bioinformatics 15, 215. 18 GELMAN, A., JAKULIN, A., PITTAU, M. G. and SU, Y.-S. (2008). A weakly informative default 19 prior distribution for logistic and other regression models. The Annals of Applied 20 Statistics 2, 1360-1383. 21 HANSEN, K. D., LANGMEAD, B. and IRIZARRY, R. A. (2012). BSmooth: from whole genome 22 bisulfite sequencing reads to differentially methylated regions. Genome Biol 13, R83. 23 HE, Y., HARIHARAN, M., GORKIN, D. U., DICKEL, D. E., LUO, C., CASTANON, R. G., NERY, J. R., 24 LEE, A. Y., WILLIAMS, B. A., TROUT, D., AMRHEIN, H., FANG, R., CHEN, H., LI, B., 25 VISEL, A., PENNACCHIO, L. A., REN, B. and ECKER, J. R. (2017). Spatiotemporal DNA 26 Methylome Dynamics of the Developing Mammalian Fetus. bioRxiv. 27 HEBESTREIT, K., DUGAS, M. and KLEIN, H. U. (2013). Detection of significantly differentially 28 methylated regions in targeted bisulfite sequencing data. Bioinformatics 29, 1647-1653. 29 JAFFE, A. E., MURAKAMI, P., LEE, H., LEEK, J. T., FALLIN, M. D., FEINBERG, A. P. and IRIZARRY, 30 R. A. (2012). JONES, R. H. and BOADI-BOATENG, F. (1991). Unequally Spaced Longitudinal Data with AR(1) 33 Serial Correlation. Biometrics 47, 161-175. 34 Acknowledgement 11 Bump hunting to identify differentially methylated regions in epigenetic 31 epidemiology studies. Int J Epidemiol 41, 200-209. 32 JONES, R. H. and BOADI-BOATENG, F. (1991). Unequally Spaced Longitudinal Data with AR(1) 33 Serial Correlation. Biometrics 47, 161-175. 34 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint JUHLING, F., KRETZMER, H., BERNHART, S. H., OTTO, C., STADLER, P. F. and HOFFMANN, S. 1 (2016). metilene: fast and sensitive calling of differentially methylated regions from 2 bisulfite sequencing data. Genome Res 26, 256-262. 3 KHAMIS, A. M., LIOZNOVA, A. V., ARTEMOV, A. V., RAMENSKY, V., BAJIC, V. B. and 4 MEDVEDEVA, Y. A. (2017). CpG traffic lights are markers of regulatory regions in 5 humans. bioRxiv. 6 KRUEGER, F. and ANDREWS, S. R. (2011). Bismark: a flexible aligner and methylation caller for 7 Bisulfite-Seq applications. Bioinformatics 27, 1571-1572. 8 LEE, W. and MORRIS, J. S. (2016). Identification of Differentially Methylated Loci Using 9 Wavelet-Based Functional Mixed Models. Bioinformatics 32, 664-672. 10 LEEK, J. T., SCHARPF, R. B., BRAVO, H. C., SIMCHA, D., LANGMEAD, B., JOHNSON, W. E., 11 GEMAN, D., BAGGERLY, K. and IRIZARRY, R. A. (2010). Tackling the widespread and 12 critical impact of batch effects in high-throughput data. Nat Rev Genet 11, 733-739. 13 LOADER, C. (1999). Local Regression and Likelihood. New York: Springer. 4 LOVE, M. I., HUBER, W. and ANDERS, S. (2014). Moderated estimation of fold change and 15 dispersion for RNA-seq data with DESeq2. Genome Biol 15, 550. 16 LUN, A. T. and SMYTH, G. K. (2014). De novo detection of differentially bound regions for ChIP- 17 seq data using peaks and windows: controlling error rates correctly. Nucleic Acids Res 18 42, e95. 19 MARX, V. (2016). Genetics: profiling DNA methylation and beyond. Nat Methods 13, 119-122. 20 PACIS, A., TAILLEUX, L., MORIN, A. M., LAMBOURNE, J., MACISAAC, J. Acknowledgement 11 L., YOTOVA, V., 21 DUMAINE, A., DANCKAERT, A., LUCA, F., GRENIER, J. C., HANSEN, K. D., GICQUEL, B., 22 YU, M., PAI, A., HE, C., TUNG, J., PASTINEN, T., KOBOR, M. S., PIQUE-REGI, R., GILAD, 23 Y. and BARREIRO, L. B. (2015). Bacterial infection remodels the DNA methylation 24 landscape of human dendritic cells. Genome Res 25, 1801-1811. 25 PARK, Y., FIGUEROA, M. E., ROZEK, L. S. and SARTOR, M. A. (2014). MethylSig: a whole 26 genome DNA methylation analysis pipeline. Bioinformatics 30, 2414-2422. 27 PARK, Y. and WU, H. (2016). Differential methylation analysis for BS-seq data under general 28 experimental design. Bioinformatics 32, 1446-1453. 29 PINHEIRO, J., BATES, D., SARKAR, S. D. D. and R CORE TEAM (2017). nlme: Linear and 30 Nonlinear Mixed Effects Models. pp. https://CRAN.R-project.org/package=nlme. 31 ROBINSON, M. D., KAHRAMAN, A., LAW, C. W., LINDSAY, H., NOWICKA, M., WEBER, L. M. and 32 ZHOU, X. (2014). Statistical methods for detecting differentially methylated loci and 33 regions. Front Genet 5, 324. 34 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint SAITO, Y., TSUJI, J. and MITUYAMA, T. (2014). Bisulfighter: accurate detection of methylated 1 cytosines and differentially methylated regions. Nucleic Acids Res 42, e45. 2 SCHULTZ, M. D., HE, Y., WHITAKER, J. W., HARIHARAN, M., MUKAMEL, E. A., LEUNG, D., 3 RAJAGOPAL, N., NERY, J. R., URICH, M. A., CHEN, H., LIN, S., LIN, Y., JUNG, I., SCHMITT, 4 A. D., SELVARAJ, S., REN, B., SEJNOWSKI, T. J., WANG, W. and ECKER, J. R. (2015). 5 Human body epigenome maps reveal noncanonical DNA methylation variation. Nature 6 523, 212-216. 7 SHAFI, A., MITREA, C., NGUYEN, T. and DRAGHICI, S. (2017). A survey of the approaches for 8 identifying differential methylation using bisulfite sequencing data. Brief Bioinform. 9 SIEGMUND, D. O., ZHANG, N. R. and YAKIR, B. (2011). False discovery rate for scanning 10 statistics. Acknowledgement 11 Biometrika 98, 979-985. 11 SMITH, Z. D. and MEISSNER, A. (2013). DNA methylation: roles in mammalian development. Nat 12 Rev Genet 14, 204-220. 13 SUN, D., XI, Y., RODRIGUEZ, B., PARK, H. J., TONG, P., MEONG, M., GOODELL, M. A. and LI, W. 14 (2014). MOABS: model based analysis of bisulfite sequencing data. Genome Biol 15, 15 R38. 16 WEN, Y., CHEN, F., ZHANG, Q., ZHUANG, Y. and LI, Z. (2016). Detection of differentially 17 methylated regions in whole genome bisulfite sequencing data using local Getis-Ord 18 statistics. Bioinformatics 32, 3396-3404. 19 WU, H., XU, T., FENG, H., CHEN, L., LI, B., YAO, B., QIN, Z., JIN, P. and CONNEELY, K. N. 20 (2015). Detection of differentially methylated regions from whole-genome bisulfite 21 sequencing data without replicates. Nucleic Acids Res 43, e141. 22 YU, X. and SUN, S. (2016). HMM-DM: identifying differentially methylated regions using a 23 hidden Markov model. Stat Appl Genet Mol Biol 15, 69-81. 24 ZILLER, M. J., HANSEN, K. D., MEISSNER, A. and ARYEE, M. J. (2015). Coverage 25 recommendations for methylation analysis by whole-genome bisulfite sequencing. Nat 26 Methods 12, 230-232, 231 p following 232. 27 28 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint Table 2: Null comparison results for sample size 2 (N2) and sample size 3 (N3). Numbers of DMRs identified by dmrseq and metilene are shown at the 0.05 FDR level. Default settings were used for BSmooth and DSS. Tables It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted August 31, 2017. ; https://doi.org/10.1101/183210 doi: bioRxiv preprint Table 3: Simulated DMR results for sample size 2 (D2) and sample size 3 (D3). Numbers of DMRs identified by dmrseq and metilene are shown at the 0.05 FDR level. Default settings were used for BSmooth and DSS. True Positives (TPs) is the number of simulated DMRs that are overlapped by at least one identified DMR. False Positives (FPs) are DMRs that do not overlap any of the simulated DMRs Table 3: Simulated DMR results for sample size 2 (D2) and sample size 3 (D3). Numbers of DMRs identified by dmrseq and metilene are shown at the 0.05 FDR level. Default settings were used for BSmooth and DSS. True Positives (TPs) is the number of simulated DMRs that are overlapped by at least one identified DMR. False Positives (FPs) are DMRs that do not overlap any of the simulated DMRs. Table 3: Simulated DMR results for sample size 2 (D2) and sample size 3 (D3). Numbers of DMRs identified by dmrseq and metilene are shown at the 0.05 FDR level. Default settings were used for BSmooth and DSS. True Positives (TPs) is the number of simulated DMRs that are overlapped by at least one identified DMR. False Positives (FPs) are DMRs that do not overlap any of the simulated DMRs. Simulation Method DMRs TPs (unique) FPs D2 dmrseq 914 816 42 BSmooth 73,252 2,466 70,688 DSS 2,086 762 655 metilene 329 210 30 D3 dmrseq 1,620 1,455 78 BSmooth 72,764 2,646 69,999 DSS 2,858 1,257 763 metilene 652 441 27 Table 4: Tissue-specific DMR results. Number of DMRs found by dmrseq and metilene at FDR level 0.05, and BSmooth and DSS under default settings. Table 4: Tissue-specific DMR results. Number of DMRs found by dmrseq and metilene at FDR level 0.05, and BSmooth and DSS under default settings. Tables Table 1: Summary of datasets used. Summary measures include the number of samples per population (‘Samples’), the number of CpGs with at least one read in all samples in the population (‘CpGs Covered’), median number of reads mapping to each covered CpG (‘Median Coverage’), minimum and maximum number of reads mapping to each covered CpG (‘Coverage Range’). Since the number of CpGs and their coverage are identical in the null comparisons and DMR simulations, the entries for N2 and D2 are combined. Likewise for N3 and D3. Table 1: Summary of datasets used. Summary measures include the number of samples per population (‘Samples’), the number of CpGs with at least one read in all samples in the population (‘CpGs Covered’), median number of reads mapping to each covered CpG (‘Median Coverage’), minimum and maximum number of reads mapping to each covered CpG (‘Coverage Range’). Since the number of CpGs and their coverage are identical in the null comparisons and DMR simulations, the entries for N2 and D2 are combined. Likewise for N3 and D3. Dataset Populations Samples CpGs Covered Median Coverage Range Maximum Coverage Range Human Tissues Heart, Left Ventricle 2 27458696 59-71 453000- 1473499 Heart, Right Ventricle 2 27340755 27-59 554455- 779621 Sigmoid Colon 2 27477877 70-76 564656- 671429 Small Intestine 2 27344594 22-71 269326- 758025 Murine Leukemia ALL 2 17666741 5 2848074- 3274608 AML 2 18306783 6-8 2279583- 2491520 Control 2 18661620 7-9 3207310- 4909532 Simulated Simulations N2 & D2 2 22015096 9-10 200- 236 Simulations N3 & D3 3 21795211 9-10 200- 236 Table 2: Null comparison results for sample size 2 (N2) and sample size 3 (N3). Numbers of DMRs identified by dmrseq and metilene are shown at the 0 05 FDR level Default settings were used for BSmooth and DSS Table 2: Null comparison results for sample size 2 (N2) and sample size 3 (N3). Numbers of DMRs identified by dmrseq and metilene are shown at the 0.05 FDR level. Default settings were used for BSmooth and DSS. Null Comparison Method DMRs (FPs) N2 dmrseq 0 BSmooth 76,563 DSS 661 metilene 31 N3 dmrseq 0 BSmooth 76,319 DSS 770 metilene 27 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. Tables Tissue Comparison dmrseq BSmooth DSS metilene Left Ventricle vs Right Ventricle 0 88,443 6,312 24 Sigmoid Colon vs Small Intestine 14,695 75,968 51,744 949 Left Ventricle vs Small Intestine 33,740 76,078 153,217 6,344 Left Ventricle vs Sigmoid Colon 106,461 76,307 229,729 8,133 Right Ventricle vs Small Intestine 32,143 76,334 129,106 5,756 Right Ventricle vs Sigmoid Colon 73,431 76,643 196,998 7,692 Table 5: Murine Leukemia model DMR results. Number of DMRs found by dmrseq and metilene at FDR level 0.10, and BSmooth and DSS under default settings. Condition Comparison dmrseq BSmooth DSS metilene AML vs Control 16,465 43,818 21,256 3,004 ALL vs Control 8,855 51,723 16,478 3,182 AML vs ALL 9,253 50,004 23,582 3,360 fic DMR results. Number of DMRs found by dmrseq and metilene at FDR level 0.05, and BSmooth and i Table 4: Tissue-specific DMR results. Number of DMRs found by dmrseq and metilene at FDR level 0.05, and BSmooth and DSS under default settings Table 4: Tissue-specific DMR results. Number of DMRs found by dmrseq and metilene at FDR level 0.05, and BSmooth and DSS under default settings. Tissue Comparison dmrseq BSmooth DSS metilene Left Ventricle vs Right Ventricle 0 88,443 6,312 24 Sigmoid Colon vs Small Intestine 14,695 75,968 51,744 949 Left Ventricle vs Small Intestine 33,740 76,078 153,217 6,344 Left Ventricle vs Sigmoid Colon 106,461 76,307 229,729 8,133 Right Ventricle vs Small Intestine 32,143 76,334 129,106 5,756 Right Ventricle vs Sigmoid Colon 73,431 76,643 196,998 7,692 Table 5: Murine Leukemia model DMR results. Number of DMRs found by dmrseq and metilene at FDR level 0.10, and BSmooth and DSS under default settings. Table 5: Murine Leukemia model DMR results. Number of DMRs found by dmrseq and metilene at FDR level 0.10, and BSmooth and DSS under default settings. Condition Comparison dmrseq BSmooth DSS metilene AML vs Control 16,465 43,818 21,256 3,004 ALL vs Control 8,855 51,723 16,478 3,182 AML vs ALL 9,253 50,004 23,582 3,360 Figures Figures Figure 1: Illustration of why FDR at the loci level is not the same as FDR at the region level. This schematic shows a plot of genomic location versus methylation difference estimates at several neighboring loci. The individual CpGs (points) are shaded by whether they are a true or false positive. Regions are denoted by lines. The loci FDR is 𝑭𝑫𝑹𝒍𝒐𝒄𝒊= # 𝑭𝒂𝒍𝒔𝒆 𝑷𝒐𝒔𝒊𝒕𝒊𝒗𝒆 𝑳𝒐𝒄𝒊 𝑻𝒐𝒕𝒂𝒍 # 𝒐𝒇 𝑺𝒊𝒈𝒏𝒊𝒇𝒊𝒄𝒂𝒏𝒕 𝑳𝒐𝒄𝒊, which is equal to 0.25 in this example. Tables The region FDR is 𝑭𝑫𝑹𝒓𝒆𝒈𝒊𝒐𝒏= # 𝑭𝒂𝒍𝒔𝒆 𝑷𝒐𝒔𝒊𝒕𝒊𝒗𝒆 𝑹𝒆𝒈𝒊𝒐𝒏𝒔 𝑻𝒐𝒕𝒂𝒍 # 𝒐𝒇 𝑺𝒊𝒈𝒏𝒊𝒇𝒊𝒄𝒂𝒏𝒕 𝑹𝒆𝒈𝒊𝒐𝒏𝒔, which is equal to 0.50 in this example. Genomic loca+on Methyla+on Diﬀerence 0 True Posi+ve False Posi+ve True Nega+ve Loci Region Genomic loca+on Methyla+on Diﬀerence 0 True Posi+ve False Posi+ve True Nega+ve Loci Region Methyla+on Diﬀerence Genomic loca+on Figure 1: Illustration of why FDR at the loci level is not the same as FDR at the region level. This schematic shows a plot of genomic location versus methylation difference estimates at several neighboring loci. The individual CpGs (points) are shaded by whether they are a true or false positive. Regions are denoted by lines. The loci FDR is 𝑭𝑫𝑹𝒍𝒐𝒄𝒊= # 𝑭𝒂𝒍𝒔𝒆 𝑷𝒐𝒔𝒊𝒕𝒊𝒗𝒆 𝑳𝒐𝒄𝒊 𝑻𝒐𝒕𝒂𝒍 # 𝒐𝒇 𝑺𝒊𝒈𝒏𝒊𝒇𝒊𝒄𝒂𝒏𝒕 𝑳𝒐𝒄𝒊, which is equal to 0.25 in this example. The region FDR is 𝑭𝑫𝑹𝒓𝒆𝒈𝒊𝒐𝒏= # 𝑭𝒂𝒍𝒔𝒆 𝑷𝒐𝒔𝒊𝒕𝒊𝒗𝒆 𝑹𝒆𝒈𝒊𝒐𝒏𝒔 𝑻𝒐𝒕𝒂𝒍 # 𝒐𝒇 𝑺𝒊𝒈𝒏𝒊𝒇𝒊𝒄𝒂𝒏𝒕 𝑹𝒆𝒈𝒊𝒐𝒏𝒔, which is equal to 0.50 in this example. Figure 2: dmrseq provides accurate FDR control of regions. Specified versus observed region-level FDR level is plotted for two different sample size settings from simulated data for dmrseq. Note that region-level FDR cannot be specified for BSmooth or DSS, and results for metilene are shown in Supplementary Figure S3. 0.0 0.1 0.2 0.3 0.4 0.0 0.1 0.2 0.3 0.4 Specified FDR level Observed FDR level FDR control by dmrseq Simulation D2 D3 FDR control by dmrseq 0.0 0.1 0.2 0.3 0.4 0.0 0.1 0.2 0.3 0.4 Specified FDR level Observed FDR level FDR control by dmrseq Simulation D2 D3 Simulation D2 D3 Figure 2: dmrseq provides accurate FDR control of regions. Specified versus observed region-level FDR level is plotted for two different sample size settings from simulated data for dmrseq. Note that region-level FDR cannot be specified for BSmooth or DSS, and results for metilene are shown in Supplementary Figure S3. Figure 3: dmrseq is more powerful than other methods. FDR and power results for (A) Simulation D2 and (B) Simulation D3, with method denoted by color. dmrseq and metilene results are displayed for several different FDR cutoffs. Since region level FDR control is not possible for BSmooth and DSS, results using default settings are displayed. Power is calculated as the proportion of simulated DMRs overlapped by at least one identified DMR. FDR is calculated as the proportion of DMRs identified that do not overlap with any of the simulated DMRs. Tables Odds of inverse association between methylation difference of (A) tissue-specific DMRs and (B) murine leukemia DMRs with differential expression of nearby DE genes (log2 transformed) is displayed on the y-axis. For dmrseq and metilene, the x-axis represents the FDR threshold (square-root scaled) for which the odds calculation (cumulative) is performed. Since FDR cannot be specified for BSmooth or DSS, the odds are calculated over all DMRs identified and displayed as a horizontal line. Note that the comparison between Left and Right Ventricles is not shown, since no DE genes were identified. (B) Murine Leukemia Models 0 1 2 3 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) log2 odds AML vs Control 0 1 2 3 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) ALL vs Control 0 1 2 3 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) AML vs ALL Method BSmooth dmrseq DSS metilene (B) Murine Leukemia Models 0 1 2 3 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) log2 odds AML vs Control Method (B) Murine Leukemia Models 0 1 2 3 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) log2 odds AML vs Control 0 1 2 3 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) ALL vs Control 0 1 2 3 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) AML vs ALL Method BSmooth dmrseq DSS metilene 0 ) 0 1 2 3 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) ALL vs Control 0 1 2 3 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) AML vs ALL d BSmooth dmrseq DSS metilene 0 1 2 3 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) ALL vs Control BSmooth dmrseq DSS m 0 1 2 3 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) AML vs ALL metilene AML vs ALL 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) achieves stronger inverse association of methylation and differential expression at lower FDR Figure 4: dmrseq achieves stronger inverse association of methylation and differential expression at lower FDR thresholds. Tables 0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00 FDR Power (A) Simulation D2 0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00 FDR (B) Simulation D3 Method BSmooth dmrseq DSS metilene 0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00 FDR Power (A) Simulation D2 0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00 FDR (B) Simulation D3 Method BSmooth dmrseq DSS metilene (B) Simulation D3 (A) Simulation D2 (B) Figure 3: dmrseq is more powerful than other methods. FDR and power results for (A) Simulation D2 and (B) Simulation D3, with method denoted by color. dmrseq and metilene results are displayed for several different FDR cutoffs. Since region level FDR control is not possible for BSmooth and DSS, results using default settings are displayed. Power is calculated as the proportion of simulated DMRs overlapped by at least one identified DMR. FDR is calculated as the proportion of DMRs identified that do not overlap with any of the simulated DMRs. Figure 4: dmrseq achieves stronger inverse association of methylation and differential expression at lower FDR thresholds. Odds of inverse association between methylation difference of (A) tissue-specific DMRs and (B) murine leukemia DMRs with differential expression of nearby DE genes (log2 transformed) is displayed on the y-axis. For dmrseq and metilene, the x-axis represents the FDR threshold (square-root scaled) for which the odds calculation (cumulative) is performed. Since FDR cannot be specified for BSmooth or DSS, the odds are calculated over all DMRs identified and displayed as a horizontal line. Note that the comparison between Left and Right Ventricles is not shown, since no DE genes were identified. Tables (A) Roadmap Tissue Comparisons 1 2 3 4 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) log2 odds Right Ventricle vs Small Intestine 1 2 3 4 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) Right Ventricle vs Sigmoid Colon 1 2 3 4 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) Left Ventricle vs Small Intestine 1 2 3 4 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) log2 odds Left Ventricle vs Sigmoid Colon 1 2 3 4 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) Sigmoid Colon vs Small Intestine (B) Murine Leukemia Models 0 1 2 3 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) log2 odds AML vs Control 0 1 2 3 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) ALL vs Control 0 1 2 3 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) AML vs ALL Method BSmooth dmrseq DSS metilene (A) Roadmap Tissue Comparisons 1 2 3 4 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) log2 odds Right Ventricle vs Small Intestine 1 2 3 4 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) Right Ventricle vs Sigmoid Colon 1 2 3 4 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) Left Ventricle vs Small Intestine 1 2 3 4 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) log2 odds Left Ventricle vs Sigmoid Colon 1 2 3 4 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) Sigmoid Colon vs Small Intestine (A) Roadmap Tissue Comparisons (A) Roadmap Tissue Comparisons (A) Roadmap Tissue Comparison 1 2 3 4 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) log2 odds Right Ventricle vs Small Intestine 1 2 3 4 1 2 3 4 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) Left Ventricle vs Small Intestine 1 2 3 4 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) Right Ventricle vs Sigmoid Colon Left Ventricle vs Small Intestine Right Ventricle vs Small Intestine 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) FDR threshold (square root scaled) FDR threshold (square root scaled) FDR threshold (square root scaled) FDR threshold (square root scaled) 1 2 3 4 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) log2 odds Left Ventricle vs Sigmoid Colon 1 2 3 4 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) Sigmoid Colon vs Small Intestine 1 2 3 4 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) log2 odds Left Ventricle vs Sigmoid Colon Sigmoid Colon vs Small Intestine 1 2 3 4 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) Sigmoid Colon vs Small Intestine 0.01 0.10 0.25 0.50 0.75 1.00 FDR threshold (square root scaled) FDR threshold (square root scaled) Figure 4: dmrseq achieves stronger inverse association of methylation and differential expression at lower FDR thresholds. Tables Odds of inverse association between methylation difference of (A) tissue-specific DMRs and (B) murine leukemia DMRs with differential expression of nearby DE genes (log2 transformed) is displayed on the y-axis. For dmrseq and metilene, the x-axis represents the FDR threshold (square-root scaled) for which the odds calculation (cumulative) is performed. Since FDR cannot be specified for BSmooth or DSS, the odds are calculated over all DMRs identified and displayed as a horizontal line. Note that the comparison between Left and Right Ventricles is not shown, since no DE genes were identified. Methylation 0.2 0.5 0.8 Left Ventricle Small Intestine chr5: 135,027,359 − 135,030,207 (width = 2,849) dmrseq: qval=0.005 BSmooth: area rank=0.001, mean rank=0 DSS: area rank=0, mean rank=0.012 Metilene: qval=0 (A) gure 5: dmrseq ranks regions by statistical significance. Example regions from the human tissue and murine leukemia dies are displayed for three cases that illustrate the increased variability of regions that are highly ranked by area or mean ference statistics of BSmooth and DSS but not dmrseq. For each case, the q-value is shown for dmrseq and metiline, and the nk percentile by the area statistic and mean difference statistics are both shown for BSmooth and DSS (see Supplement Section for details). (A) All methods assign a consistently high rank. (B) dmrseq assigns a low rank, but the mean difference statistic BSmooth and DSS assign a high rank. (C) dmrseq assigns a low rank, but the area statistic of BSmooth and DSS assign a high nk. The condition comparison is indicated by the labels to the right of each plot. Tables Methylation 0.2 0.5 0.8 Left Ventricle Small Intestine chr5: 135,027,359 − 135,030,207 (width = 2,849) dmrseq: qval=0.005 BSmooth: area rank=0.001, mean rank=0 DSS: area rank=0, mean rank=0.012 Metilene: qval=0 Methylation 0.2 0.5 0.8 Right Ventricle Sigmoid Colon chr16: 85,551,503 − 85,552,644 (width = 1,142) dmrseq: qval=0.722 BSmooth: area rank=0.383, mean rank=0.029 DSS: area rank=0.118, mean rank=0.185 Metilene: qval=0.004 Methylation 0.2 0.5 0.8 AML ALL chr18: 39,489,445 − 39,490,331 (width = 887) dmrseq: qval=0.651 BSmooth: area rank=0.074, mean rank=0.946 DSS: area rank=0.009, mean rank=0.951 Metilene: qval=0.17 (A) (B) (C) 1 chr5: 135,027,359 − 135,030,207 (width = 2,849) (A) Methylation 0.2 0.5 0.8 Left Ventricle Small Intestine , , , , ( , ) dmrseq: qval=0.005 BSmooth: area rank=0.001, mean rank=0 DSS: area rank=0, mean rank=0.012 Metilene: qval=0 Methylation 0.2 0.5 0.8 Right Ventricle Sigmoid Colon chr16: 85,551,503 − 85,552,644 (width = 1,142) dmrseq: qval=0.722 BSmooth: area rank=0.383, mean rank=0.029 DSS: area rank=0.118, mean rank=0.185 Metilene: qval=0.004 (A) (B) DSS: area rank=0, mean rank=0.012 Metilene: qval=0 Methylation 0.2 0.5 0.8 Right Ventricle Sigmoid Colon chr16: 85,551,503 − 85,552,644 (width = 1,142) dmrseq: qval=0.722 BSmooth: area rank=0.383, mean rank=0.029 DSS: area rank=0.118, mean rank=0.185 Metilene: qval=0.004 (B) (B) Metilene: qval 0.004 Methylation 0.2 0.5 0.8 AML ALL chr18: 39,489,445 − 39,490,331 (width = 887) dmrseq: qval=0.651 BSmooth: area rank=0.074, mean rank=0.946 DSS: area rank=0.009, mean rank=0.951 Metilene: qval=0.17 (C) (C) seq ranks regions by statistical significance. Example regions from the human tissue and murine leukemia Figure 5: dmrseq ranks regions by statistical significance. Example regions from the human tissue and m Figure 5: dmrseq ranks regions by statistical significance. Example regions from the human tissue and murine leukemia studies are displayed for three cases that illustrate the increased variability of regions that are highly ranked by area or mean difference statistics of BSmooth and DSS but not dmrseq. For each case, the q-value is shown for dmrseq and metiline, and the rank percentile by the area statistic and mean difference statistics are both shown for BSmooth and DSS (see Supplement Section 2.8 for details). (A) All methods assign a consistently high rank. (B) dmrseq assigns a low rank, but the mean difference statistic of BSmooth and DSS assign a high rank. (C) dmrseq assigns a low rank, but the area statistic of BSmooth and DSS assign a high rank. The condition comparison is indicated by the labels to the right of each plot. Tables 1 Figure 5: dmrseq ranks regions by statistical significance. Example regions from the human tissue and murine leukemia studies are displayed for three cases that illustrate the increased variability of regions that are highly ranked by area or mean difference statistics of BSmooth and DSS but not dmrseq. For each case, the q-value is shown for dmrseq and metiline, and the rank percentile by the area statistic and mean difference statistics are both shown for BSmooth and DSS (see Supplement Section 2.8 for details). (A) All methods assign a consistently high rank. (B) dmrseq assigns a low rank, but the mean difference statistic of BSmooth and DSS assign a high rank. (C) dmrseq assigns a low rank, but the area statistic of BSmooth and DSS assign a high rank. The condition comparison is indicated by the labels to the right of each plot. 1
https://openalex.org/W2005802544	https://zenodo.org/record/2345917/files/article.pdf	unk	null	Colourless Crystals of Hæmoglobin	Nature	1,916	public-domain	4	©1916 Nature Publishing Group
https://openalex.org/W4221157815	https://dergipark.org.tr/tr/download/article-file/2276695	Latin	null	Optimization of welding job-shop scheduling problem under variable workstation constraint: an industrial application with Arena simulation based genetic algorithm	Mühendislik bilimleri dergisi/Mühendislik bilimleri dergisi	2,022	cc-by	8,288	Abstract İş atölyesi planlaması, 'emek yoğun proje tipi üretim' için zor bir konudur. Çünkü bu tür bir üretimde, gerçek işlem süreleri üretim bitene kadar tam olarak bilinmez ve bu işlem süreleri siparişin teknik özelliklerine göre değişir. İşlem sürelerini tahmin etmek için olasılık dağılımlarını kullanmak uygun bir yöntemdir. Bu makale, emek-yoğun proje tipi çalışan kaynak atölyesinin değişken iş istasyonu kısıtlamaları altında planlanması için endüstriyel bir uygulama sunmaktadır. Bu kısıt, ürünlerin boyuna bağlı olarak ortaya çıkan özel bir üretim şeklinin sonucudur. Amaç, bir grup bekleyen iş emrinin tamamlanma süresini en aza indirmektir. Genetik algoritma (GA) bu amaçla, atölyeye girmeyi bekleyen iş emirlerinin atölyeye giriş sırasını oluşturmak ve bunları 6 özdeş kaynak istasyonuna göndermek için kullanılır. Atölyenin dinamik koşulları, Arena simülasyon programı ile simüle edilir. Algoritmanın girdi verileri olarak stokastik işlem süreleri kullanılır. Kaynak iş istasyonu çizelgeleme için değişken iş istasyonu kısıtlaması altında stokastik işleme sürelerinin kullanılması daha önce araştırılmamıştır. Deneysel sonuçlara göre, GA ve Arena simülasyonu birlikte, değişken iş istasyonu kısıtlaması altında bu tür problemlerde bir grup işin toplam tamamlanma zamanını etkili bir şekilde azaltmaktadır. GA destekli Arena çizelgesi, bu sorun için GA kullanmadan önerilen çizelgeden daha iyi performans gösterir. Simülasyon sonuçları, bekleyen siparişlerin toplam üretim süresinin, GA kullanılmadan önerilen çizelgelerle karşılaştırıldığında yaklaşık % 9,25 oranında azaldığını göstermektedir. Job-shop scheduling is a difficult issue for 'labor-intensive project type manufacturing'. Because in this type of production, the actual processing times are not exactly known until the production is finished and these processing times vary depending on the order’s technical specifications. It is an appropriate method to use probability distributions to forecast the processing times. This paper provides an industrial application for the scheduling of a labor-intensive project type working welding job-shop under variable workstation constraints. This constraint is consequence of a special production type that is depending on the length of the products. The aim is minimizing the makespan of a group of waiting orders. Genetic algorithm (GA) is used for this purpose to establish the entry sequence of the job-shop's waiting orders and dispatching them to the 6 identical welding stations. The dynamic conditions of the job-shop are simulated by the Arena simulation program. Stochastic processing times are used as the input data of the algorithm. Using stochastic processing times under variable workstation constraint for welding job-shop scheduling is not investigated previously. Abstract According to the experimental results, GA and Arena simulation together effectively reduces the makespan in this type of problem under variable workstation constraint. The GA aided Arena schedule outperforms the schedules proposed without using GA for this problem. Simulation results indicate that the total manufacturing time of pending orders is nearly 9.25% reduced when compared with the schedules proposed without using GA. Keywords: Arena simulation, Genetic algorithm, Labor-intensive project type production, Makespan minimization, Variable workstation constraint, Welding shop scheduling problem. Anahtar kelimeler: Arena simülasyonu, Genetik algoritma, Emek- yoğun proje tipi üretim, Tamamlanma zamanı minimizasyonu, Değişken iş istasyonu kısıtı, Kaynak atölyesi çizelgeleme problemi. many assumptions are needed to be made to mathematically model the real-life problem, this causes the problem to diverge from real life. Stochastic scheduling on dynamic shop floors is more appropriate for scheduling real life problems. There are many factors that trigger stochastic operating times, such as technical order requirements, operator skills, job-shop bottlenecks, transporters, and so on. The appropriate operations are known but processing times are not exactly known in stochastic scheduling. Stochastic processing times make it harder for order scheduling. In addition to using stochastic operation times, because of the dynamic job-shop environment, the use of simulation provides more important results for real industrial plants. Revision/Düzeltme Tarihi: 13.04.2021 doi: 10.5505/pajes.2021.41948 Research Article/Araştırma Makalesi doi: 10.5505/pajes.2021.41948 Research Article/Araştırma Makalesi Revision/Düzeltme Tarihi: 13.04.2021 Received/Geliş Tarihi: 14.01.2021 Accepted/Kabul Tarihi: 14.04.2021 Received/Geliş Tarihi: 14.01.2021 Accepted/Kabul Tarihi: 14.04.2021 Received/Geliş Tarihi: 14.01.2021 Accepted/Kabul Tarihi: 14.04.2021 Corresponding author/Yazışılan Yazar 1Department of Industrial Engineering, Engineering Faculty, Balıkesir University, Balıkesir, Turkey. deniz@balikesir.edu.tr 1Department of Industrial Engineering, Engineering Faculty, Balıkesir University, Balıkesir, Turkey. deniz@balikesir.edu.tr 1Department of Industrial Engineering, Engineering Faculty, Balıkesir University, Balıkesir, Turkey. deniz@balikesir.edu.tr Pamukkale Univ Muh Bilim Derg, 28(1), 139-147, 2022 Pamukkale Univ Muh Bilim Derg, 28(1), 139-147, 2022 Kaynak atölyesi çizelgeleme probleminin değişken iş istasyonu kısıtlaması altında optimizasyonu: Arena simülasyonu tabanlı genetik algoritma ile endüstriyel bir uygulama Aslan Deniz KARAOGLAN1 1 Introduction [17] established a multi- objective optimization model for a JSSP and used NSGA to minimize some performance measurements such as the total cost and the total completion time. Tang et al. [18] considered limited starting time interval and tolerated time interval to conduct a MO-FJSSP. They also presented hybrid discrete PSO with simulated annealing (HDPSO-SA) for the optimization of this problem. Liao and Lin [19] studied on optimization of job- shop supply chain scheduling problem using PSO. Wang et al. [20] used hybrid GA PSO for inverse JSSP and performed The WSSP for JSSP using GA is considered in the current paper. In several industrial fields, including the production of mechanical manufacturing, WSSP can be applied. There are however, just a few studies based on WSSP that include the use of several machines. Rao et al. [27] studied on bi-objective WSSP (BWSSP) to minimize the machine interaction effects and the total tardiness. BWSSP is a concern with special flow-shop scheduling problem (FSSP). In this problem, on a certain stage, more than one machine (or workstation) can process one job. They modeled the problem as mixed integer programming model (MIPM) and then solve the model by NSGA with a restarted strategy. However WSSP for JSSP (which is the subject of the current study) is studied by Karaoglan et al. [23] and a novel problem is established with variable workstation constraint. They solved the problem using ABC algorithm coded by Matlab. Although they assumed that the processing times of the operations are known before starting the operation. This assumption results not able to fully reflect dynamic job-shop conditions. In the present paper, WSSP is considered for JSSP for probabilistic (stochastic) processing times. This is the first novelty of this research. The optimization is performed by using GA combined with Arena simulation. Using probabilistic processing times and Arena simulation provide to consider the dynamic job-shop conditions and get more applicable solutions. The contributions of this study to the literature are: (i) using variable workstation constraint, (ii) solving welding shop scheduling problem (WSSP) - which is studied by Rao et al. [27] and Karaoglan et al. [23]-under the stochastic processing times. They used mixed-integer programming model (MIPM). However, if the production line is labor-intensive then using integer programming causes not discussing the dynamic shop conditions. 1 Introduction It is critical for enterprises to supply customer requirements within due date and to reduce the labor cost for each customer order. Order scheduling is required for various purposes, for example makespan minimization, optimizing the use of resources, etc. Before starting the production, processing times for each job are known in deterministic scheduling and there are no process disruptions, except for a few production adjustments. These production systems can be widely computer-controlled, and the problem can be analytically solved by using linear programming. However, such types of solutions are far from actual production environments. Since so 139 Pamukkale Univ Muh Bilim Derg, 28(1), 139-147, 2022 A.D. Karaoglan Pamukkale Univ Muh Bilim Derg, 28(1), 139-147, 2022 A.D. Karaoglan The labor intensive project type job-shop has a dynamic production environment. Stochastic operation times and stochastic scheduling are triggered by a dynamic production environment. Heuristics are used to address that issue because of these problems are known as the NP-hard problem. Particle swarm, genetic algorithm, ant colony and so on have been commonly used meta-heuristics in the last decade to solve problems with job-shop scheduling. discrete event simulation. Zhang et al. [21] used PSO and ANN together for JSSP. They treated each particle in the swarm as a connection in the ANN to minimize the maximum makespan. Zhu et al. [22] used improved WOA for JSSP based on the quantum computing. They optimized maximum makespan, maximum machine load and total machine load. Karaoglan et al. [23] used artificial bee colony (ABC) algorithm to optimize the welding shop scheduling problem (WSSP) to minimize the makespan. They used deterministic processing times. Shi et al. [24] used fuzzy and immune GA for optimizing MO-FJSSP. They considered on minimizing energy consumption, maximum makespan and consumer dissatisfaction. Gu et al. [25] proposed a discrete genetic-grey wolf optimization (GA-GWO) algorithm to solve low-carbon JSSP. Vital-Soto [26] used hybridized bacterial foraging optimization (BFO) algorithm for the FJSSP. To minimize weighted lateness for the FJSSP with sequencing flexibility, they modeled the problem with mixed integer linear programming (MILP). The summary of the related literature is given in Table 1. Some of the papers selected - those closely relevant to the subject matter of this paper - are as follows: Jia et al. [1] proposed decode select string (DSS) decoding genetic algorithm (GA) for job-shop scheduling problem (JSSP). Azadeh et al. [2] studied on makespan minimization in stochastic JSSP. 1 Introduction They used artificial neural networks (ANN) and computer simulation together to select the optimum dispatching rule for each machine under a set of different constraints and rules. Huang et al. [3] studied on process sequence flexibility problem for JSS. They used GA to minimize the makespan. Aydemir and Koruca [4] developed a priority rule-based GA (PRGA) scheduling module to minimize the total completion time. Flexible job-shop scheduling (FJSS) is also a NP hard problem. Ba et al. [5] presented a novel mathematical model for a multi- resource FJSS (MR-FJSS). They used GA to minimize the makespan. Deng et al. [6] studied on bee evolutionary guiding non-dominated sorting GA (NSGA) for multi-objective FJSSP (MO-FJSSP). Ocaktan et al. [7] used GA and Arena simulation together to minimize the makespan of customized orders of a job-shop. Zhang et al. [8] also used GA for MO-FJSSP to minimize the longest makespan of the workpieces and machine loads (for each machine and total machine). The shortest processing time (SPT) and balanced use of machines is considered by them. Hu et al. [9] used improved cuckoo search algorithm (CSA) for JSSP. This algorithm is a hybrid of CSA and particle swarm optimization (PSO) algorithm. The results are compared with PSO and GA. The simulations are performed by using Matlab program. Jiang et al. [10] studied on energy-efficient JSS instead of conventional performance criteria such as production efficiency, makespan etc. To optimize this environmental metric, they used grey wolf optimization (GWO) algorithm. Also they proposed a double-searching mode for GWO to solve this problem. Jiang et al. [11] studied on the same subject by using whale optimization algorithm (WOA). Seng et al. [12] proposed a low-carbon scheduling model for MO and multi-speed (MS) FJSS and used NSGA for optimization. They considered energy consumption, low carbon emissions, and makespan as the performance criteria. Karaoglan et al. [13] studied on JSSP by considering the ergonomic constraints. They used combined GA and Arena simulation to minimize the makespan. Zhang et al. [14] used binary PSO to optimize the multi-technique, multi- response FJSSP. Zhong et al. [15] optimized dual resource constraint JSSP to optimize the makespan and total processing cost. They proposed branch population GA for optimization. Sel and Hamzadayi [16] used simulated annealing (SA) for JSSP and used Arena for simulations. Fu et al. Pamukkale Univ Muh Bilim Derg, 28(1), 139-147, 2022 A.D. Karaoglan Pamukkale Univ Muh Bilim Derg, 28(1), 139-147, 2022 A.D. Karaoglan Table 1. Summary of the literature review. Author Method Problem/Model (or constraints) Objective Jia et al. [1] DSS GA JSSP Makespan Azadeh et al. [2] ANN Stochastic JSSP Makespan Huang et al. [3] GA JSSP with process sequence flexibility Makespan Aydemir and Koruca [4] PRGA, Faborg-Sim JSSP Total completion time Ba et al. [5] GA MR-FJSSP Makespan Deng et al. [6] Bee evolutionary guiding - NSGA MO-FJSSP Total workload of all machines, makespan, workload of the most loaded machine Ocaktan et al. [7] GA, Arena simulation Stochastic JSSP Makespan Zhang et al. [8] Multi-population GA FJSSP Makespan, load of each machine, load of all machines Hu et al. [9] Improved CSA JSSP Makespan, Jiang et al. [10] GWO with double-searching mode JSSP Energy efficiency Jiang et al. [11] WOA JSSP Energy efficiency Seng et al. [12] NSGA Low-carbon scheduling model for MO and MS-FJSS Makespan, energy consumption Karaoglan et al. [13] GA, Arena simulation JSSP with ergonomic constraints Makespan Zhang et al. [14] Binary PSO Multi-technique, multi-response FJSSP Makespan, cost Zhong et al. [15] Branch population GA Dual resource constraint JSSP Makespan, total processing cost Sel and Hamzadayi [16] SA, Arena simulation Stochastic JSSP Flowtime, Tardiness Fu et al. [17] NSGA JSSP Total cost, total completion time Tang et al. [18] HDPSO-SA FJSSP, limited starting time interval and tolerated time interval (TTI) Total overtime of TTI, total tardiness, completion time Liao and Lin [19] PSO Job-shop supply chain scheduling Completion time Wang et al. [20] Hybrid GA-PSO, discrete event simulation Inverse JSSP Total weighted makespan Zhang et al. [21] PSO, ANN JSSP Makespan Zhu et al. [22] Improved WOA, quantum computing JSSP Makespan, maximum machine load, total machine load Karaoglan et al. [23] ABC WSSP with deterministic times Makespan Shi et al. [24] Fuzzy and immune GA MO-FJSSP Makespan, energy consumption, consumer dissatisfaction Gu et al. [25] GA-GWO Low-carbon JSSP Sum of energy consumption cost, completion time cost Vital-Soto [26] BFO FJSSP with sequencing flexibility Weighted lateness Rao et al. [27] NSGA BWSSP with deterministic times Total tardiness, total penalty Table 1. Summary of the literature review. The aim is to minimize the average makespan of a group of waiting orders by using GA aided Arena simulation. Pamukkale Univ Muh Bilim Derg, 28(1), 139-147, 2022 A.D. Karaoglan The second motivation is presenting the results of GA aided simulation based scheduling of the WSSP under the variable workstation constraint to the readers. perform the production of transformers' vessels. In transformer vessel manufacturing, vessel bottom pan, side walls and top cover are manufactured using st37 black sheet metal. Then, these parts are assembled by welding operation. The welding operations at the shop floor are performed under 4-tasks (task names are not specified for commercial confidentiality). The aim of this study is scheduling the orders at welding shop for makespan minimization. GA specifies the entry sequence of the waiting orders to the mechanical manufacturing (welding shop). ARENA has been used to bring the problem under consideration closer to the dynamic workshop conditions and provide fitness value calculations for GA. A brief overview of GA aided Arena simulation and the real industrial problem is given in the next section. Then in Section 3, the results and discussions are discussed. Finally, in Section 4, the conclusions are given. 4, the conclusions are given. 1 Introduction Because in the labor-intensive jobs, there are so many factors those have effect on the processing times of the same jobs processed in different shifts (such as the tempo factor of the workers, material searching, waiting for crane or forklift, and etc.). This is also the motivation of this research. Using stochastic processing times under variable workstation constraint for welding job-shop scheduling is not investigated previously and this is the novelty aspect of this research. Some researchers [24] used fuzzy processing times, however in this study GA aided Arena simulation is used and probability distributions of processing times are suitable for this software. 140 2.1 Problem definition The case study is being carried out in the mechanical workshop of a transformer manufacturer. The company's manufacturing operations for power transformers consist of five main phases. They are: 1) magnetic core manufacturing, 2) winding, 3) assembly of active parts, 4) mechanical manufacturing (welding shop), and 5) final assembly. The workshop is a labor-intensive project-type manufacturing process consisting of sequential processes including CNC sheet metal cutting, other semi-product preparation processes, lathe leveling, welding, etc. These operations are presented under the “Task” headings in Table 2. In companies that operate on the basis of project type labor-intensive production, the dynamic production environment causes stochastic times for operations. This study is carried out in mechanical manufacturing (welding shop) section. In the welding shop the primary purpose is to 141 141 Pamukkale Univ Muh Bilim Derg, 28(1), 139-147, 2022 A.D. Karaoglan Pamukkale Univ Muh Bilim Derg, 28(1), 139-147, 2022 A.D. Karaoglan Stochastic operation times make it harder for order scheduling. There are several explanations why running times are stochastic, such as technical order requirements, operator expertise, job-shop bottlenecks, and so on. Production times differ according to the characteristics of the materials that will be processed. To determine the distribution of the processing times, random samples were taken, then the probability distributions were fitted by using Arena input analyzer and chi- square goodness-of-fit test was performed to test the significance of these fitted probability distributions. Table 2 gives the determined probability distributions (which are fitted to the processing times observed from the workshop) those can be used for simulation. The chi-square goodness-of-fit test results under 95% confidence level are significant for the distributions given in the Table 2. In Table 2, 8 meters is cut- point for classifying the vessel length (1: vessel length≥8 meters, and 2: vessel length<8 meters). The values are in terms of hours. For a total of 20 orders, the data was collected from the company's current pending list (list of waiting orders), and provided in Table 3. Table 2. Task-based probability distributions for orders discussed in the example (hours). 2.1 Problem definition Welding order number (WON) Product type Model Vessel length type Task 1 2 3 4 1 1 1 1 Uniform (130, 140) Triangular (80, 90, 100) Normal (23.8, 5.2) Normal (811.9, 30.8) 2 1 2 1 Uniform (85, 95) Triangular (40, 50, 60) Normal (42.4, 7.2) Normal (542.8, 22.9) 3 1 3 1 Uniform (110, 120) Triangular (110, 120, 130) Normal (26.6, 5.4) Normal (960.2, 42.8) 4 1 4 1 Uniform (140, 150) Triangular (30, 40, 50) Normal (63.6, 8.6) Normal (371.4, 12.9) 5 1 5 1 Uniform (200, 210) Triangular (120, 130, 140) Normal (41.8, 6.3) Normal (1014.4, 50.7) 6 1 6 1 Uniform (120, 130) Triangular (110, 120, 130) Normal (50.8, 7.2) Normal (606.7, 25.8) 7 2 1 1 Uniform (125, 135) Triangular (40, 50, 60) Normal (16.3, 4.1) Normal (402.5, 18.3) 8 2 2 1 Uniform (230, 240) Triangular (10, 20, 30) Normal (24.5, 5.4) Normal (1667.7, 63.5) 9 3 1 1 Uniform (130, 140) Triangular (90, 100, 110) Normal (47.1, 6.2) Normal (1611.3, 59.4) 10 4 1 2 Uniform (315, 325) Triangular (200, 210, 220) Normal (89.8, 10.2) Normal (2016.4, 70.5) 11 4 2 1 Uniform (360, 370) Triangular (300, 310, 320) Normal (73.5, 5.4) Normal (1017.7, 48.2) 12 4 3 2 Uniform (330, 340) Triangular (110, 120, 130) Normal (65.3, 6.2) Normal (2312.9, 75.2) 13 5 1 1 Uniform (170, 180) Triangular (60, 70, 80) Normal (46.1, 4.3) Normal (676.1, 35.8) 14 6 1 1 Uniform (140, 150) Triangular (30, 40, 50) Normal (24.1, 3.2) Normal (269.7, 12.8) 15 6 2 1 Uniform (210, 220) Triangular (90, 100, 110) Normal (96.1, 8.4) Normal (1157.9, 55.7) 16 6 3 1 Uniform (100, 110) Triangular (40, 50, 60) Normal (51.2, 5.4) Normal (365.4, 9.8) 17 6 4 2 Uniform (330, 340) Triangular (70, 80, 90) Normal (73.9, 5.9) Normal (1230.1, 51.5) 18 6 5 1 Uniform (110, 120) Triangular (5, 8, 10) Normal (32.2, 3,5) Normal (210.9, 6.7) 19 6 6 1 Uniform (90, 100) Triangular (60, 70, 80) Normal (64.6, 4.9) Normal (495.1, 9.8) 20 6 7 1 Uniform (170, 180) Triangular (50, 60, 70) Normal (49.1, 5.1) Normal (510.2, 11.3) Table 3. Current waiting orders list. 2.1 Problem definition [27]): According to Table 3, only one vessel has to be manufactured in the first order, while 4 transformers have to be manufactured in the second order. This implies that there are 4 identical vessels that have to be manufactured. Similarly, the 8th order has 2 identical transformers, which means that this order has 2 identical vessels and so on. To minimize the makespan, GA aided Arena simulation was carried out to dispatch the orders to the welding stations. The problem can also be mathematically represented. The related equations are given below to give an idea to the readers; however the problem is modeled with Arena simulation instead of integer programming (for details see Rao et al. [27]): for scheduling. Input data is the stochastic order processing times that differ in compliance with the design requirements requested by the customers. In Table 3, also the rounded measured average processing times of 10 simulation runs for each order (that is generated in the simulations given in Section 3) is presented. This study's assumptions and constraints are listed below. General assumptions related with simulation based scheduling: - It is assumed that production is not interrupted until all scheduled jobs are produced, - It is assumed that production is not interrupted until all scheduled jobs are produced, Objective: Objective: - It is assumed that the workers in the workshop are identical, 𝑀𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑓= 𝐶𝑚𝑎𝑥𝑎𝑣𝑔 (1) (1) - Just one task can be processed on a workstation at the same time, Constraints: if 𝑙≥8 meters than 𝑍𝑖𝑗ℎ+ 𝑍𝑖𝑗ℎ+1 = 1; h = 1,2 or h = 4,5 (5) (5) if 𝑙≥8 meters than 𝑍𝑖𝑗ℎ+ 𝑍𝑖𝑗ℎ+1 = 1; h = 1,2 or h = 4,5 (5) Note that the l is the length of the transformer vessel. Which means 1 & 2, 2 & 3 welding stations, and 4 & 5, 5 & 6 welding stations can be merged to perform the welding operation of the vessels with those have more than 8 meters length. 2.1 Problem definition Welding Order Number (WON) Product Type Model Number of Vessels in the Order Vessel ID (VID) Length of the Vessel (meters) Number of Welding Stations Needed (S) Processing time (Arena Results: Mean of 10 Runs-Rounded) 1 1 1 1 1 4 1 1063 2 1 2 4 2, 3, 4, 5 2 1 709, 720, 729, 718 3 1 3 1 6 7 1 1207 4 1 4 1 7 3 1 625 5 1 5 1 8 7 1 1384 6 1 6 1 9 4 1 913 7 2 1 1 10 4 1 602 8 2 2 2 11, 12 7 1 1945, 2026 9 3 1 1 13 5 1 1980 10 4 1 3 14, 15, 16 11 2 2631, 2579, 2446 11 4 2 1 17 7 1 1762 12 4 3 2 18, 19 13 2 2828, 2829 13 5 1 2 20, 21 4 1 988, 969 14 6 1 1 22 2 1 467 15 6 2 2 23, 24 7 1 1575, 1608 16 6 3 2 25, 26 3 1 546, 539 17 6 4 2 27, 28 11 2 1736, 1724 18 6 5 1 29 2 1 377 19 6 6 2 30, 31 4 1 719, 721 20 6 7 2 32, 33 6 1 798, 731 Table 2. Task-based probability distributions for orders discussed in the example (hours). 142 142 Pamukkale Univ Muh Bilim Derg, 28(1), 139-147, 2022 A.D. Karaoglan The given 20 orders in this table is lead to manufacture of a total of 33 vessels (in other words, 20 separate customers demand 33 vessels) and in this case we used 6 identical welding stations According to Table 3, only one vessel has to be manufactured in the first order, while 4 transformers have to be manufactured in the second order. This implies that there are 4 identical vessels that have to be manufactured. Similarly, the 8th order has 2 identical transformers, which means that this order has 2 identical vessels and so on. To minimize the makespan, GA aided Arena simulation was carried out to dispatch the orders to the welding stations. The problem can also be mathematically represented. The related equations are given below to give an idea to the readers; however the problem is modeled with Arena simulation instead of integer programming (for details see Rao et al. Constraints: ∑𝑍𝑖𝑗ℎ ℎ = 𝑁𝑖𝑗 & 𝑁𝑖𝑗≥1; 𝑁𝑖𝑗≤𝑝𝑁𝑗; i = 1, … , n; j = 1, … , m (2) ∑ 𝑦𝑖𝑖′𝑗= 1; i’ = 1, … , n; j = 1, … , m 𝑛 𝑖=0 (3) ∑ 𝑦𝑖𝑖′𝑗= 1; i = 1, … , n; j = 1, … , m 𝑛+1 𝑖′=1 (4) ∑𝑍𝑖𝑗ℎ ℎ = 𝑁𝑖𝑗 & 𝑁𝑖𝑗≥1; 𝑁𝑖𝑗≤𝑝𝑁𝑗; i = 1, … , n; j = 1, … , m ( - Since each vessel is quite large and heavy, after placing the vessel once at the welding stations, even if a more urgent job comes, the urgent job is not queued before the current job list is completed. In other words, in order to perform another task, the process cannot be disrupted, (2) ∑ 𝑦𝑖𝑖′𝑗= 1; i’ = 1, … , n; j = 1, … , m 𝑛 𝑖=0 (3) ∑ 𝑦𝑖𝑖′𝑗= 1; i = 1, … , n; j = 1, … , m 𝑛+1 𝑖′=1 (4) - The factory operates in 3 shifts, each consisting of 8 hours. Each shift is assumed to have 1 hour lunch break and 2 rest breaks (30 minute each), (4) (4) Where i and 𝑖′are the index of the jobs, j is the index of stages, h is the index of the machines used for processing during stage j, and 𝑁𝑖𝑗 is the number of processing machines for ith job at stage j. If the machine h processed ith job at stage j then the 𝑍𝑖𝑗ℎ=1; and otherwise 𝑍𝑖𝑗ℎ=0. Similarly, 𝑦𝑖𝑖′𝑗= 1 if job i is processed before 𝑖′ at stage j, and 𝑦𝑖𝑖′𝑗= 0 otherwise. Equation (1) is the objective and represents the minimizing the average makespan, Equation (2) means that a job is processes by at least one machine and also it ensures that the number of processing machines cannot exceed the number of available machines. Equation (3) satisfies that at per stage, each job has just one pre-order job. And finally at each stage, Equation (4) ensures that just one follow-up job can be processed after each job. Constraints: According to the sample problem given in Figure 1, the constraint for variable workstation can be defined as: Where i and 𝑖′are the index of the jobs, j is the index of stages, h is the index of the machines used for processing during stage j, and 𝑁𝑖𝑗 is the number of processing machines for ith job at stage j. If the machine h processed ith job at stage j then the 𝑍𝑖𝑗ℎ=1; and otherwise 𝑍𝑖𝑗ℎ=0. Similarly, 𝑦𝑖𝑖′𝑗= 1 if job i is processed before 𝑖′ at stage j, and 𝑦𝑖𝑖′𝑗= 0 otherwise. Equation (1) is the objective and represents the minimizing the average makespan, Equation (2) means that a job is processes by at least one machine and also it ensures that the number of processing machines cannot exceed the number of available machines. Equation (3) satisfies that at per stage, each job has just one pre-order job. And finally at each stage, Equation (4) ensures that just one follow-up job can be processed after each job. According to the sample problem given in Figure 1, the constraint for variable workstation can be defined as: - Task 1, Task 2, Task 3 and Task 4 should be performed in successive order and one after the other, - The processing times are probabilistic (as presented in Table 2), - The number of the orders at the ‘pending order list’ is fixed at the beginning of the simulation, Variable workstation constraint (newly added novel constraint for WSSP): The number of welding stations varies in compliance with the order's technical requirements. If the length of the transformer vessel is 8 meters or more, two horizontally neighbor stations are combined by removing the mobile sound barrier between these stations and this vessel is assigned to both stations. In addition, the successive stations are merged to match the vessel to the station if the widths of the jobs are 8 meters and over. One order occupies two welding stations at the same time in this situation. Welding stations 1-2, 2-3, 4- 5, and 5-6 can be merged. This means, at the same time 6 small or 2 small and 2 large vessels can be operated at the same time. However 3 large vessels cannot be operated at the same time. Figure 1 represents the layout of the welding shop. 3 Results and discussions The input data for the 20 waiting orders was obtained from the company's current list of pending orders, and given in Table 3. The list of these 20 orders corresponds to the manufacture of 33 vessels in total. In the method, six welding stations operate. Constraints relating to the parameters of the technological design of the orders, number of manufacturing resources, resource capability, etc. influence the average makespan of the waiting orders in the competitive manufacturing environment, as studied in previous literature studies. The objective here is to minimize the average makespan of the orders described under the variable workstation constraint in Table 3. The multiplication of PopSize and mutation rate determines the number of chromosomes to apply mutation. To maximize diversity, the mutation operator is applied to selected chromosomes in two ways: swap and insert. A random number (rnd) between 0 and 1 is generated for each selected chromosome, and if rnd > 0.5, insert is applied; otherwise, swap is applied. For the crossover strategy; two point crossover is used [7]. The pseudocode for this GA aided simulation method is given in Figure 2 [7],[13]. A sample result of this code is given in Figure 3. The flow chart for the proposed GA aided simulation is presented in Figure 4. In this figure, each gene (customer order) is seen on the chromosome only once and Arena dispatches them to the welding stations respectively. ARENA assigns these orders sequentially (according to the sequence at the chromosome) and primarily to the lowest numbered free welding station. This chromosome represents the entry sequence of the orders to the Arena simulation model and determined by using GA. The Gantt chart presents the pure Arena simulation results without using GA is given in Figure 5. The simulations are run on the “Intel Core i5-2430M CPU 2.40GHz with 4 GB RAM PC. The orders are scheduled primarily to their order number. The average makespan is calculated as 11368 hours. Notice that the 2nd and 3rd welding stations are merged while orders 14, 16, 19, and 28 are processed. Owing to the width limit (width of these transformers≥8 meters), the 5th and 6th workstations are also merged during the processing of orders 15, 18, and 27. In other words, the 14th, 16th, 19th, and 28th orders are assigned at the same time to the 2nd and 3rd welding stations. this paper for the simulation [7],[13],[33]. The suggested GA is coded in the VBA environment, making it hassle-free to link to the Arena simulation program. In VBA, all GA unique elements are implemented. Arena is used to measure, by simulating a complex system, each chromosome’s fitness value (FV). The FV is the average makespan calculated by the ARENA for the selected order sequence. The system here corresponds to the welding shop's complex labor-intensive project-type manufacturing structure including different queues, operations, and capacity constraints. The GA-based Arena simulation method proposed is to minimize the average makespan. Average makespan is total completion time for all pending orders. Here, we use GA to tune the sequence of waiting orders to import them into the system. This algorithm starts with generating the initial population randomly, which includes the number of chromosomes in population size (PopSize). Each chromosome consists of a permutable sequence of orders ranging from 1 to the total number of orders, where the number of orders corresponds to each gene. Figure 4. Flow chart of the proposed GA aided simulation. Figure 4. Flow chart of the proposed GA aided simulation. 2.2 Simulation-based GA For the special types of problems (such presented in WSSP) where the search space is huge but the feasible solutions are limited and also if the discrete optimization is needed, GA is ideal (PSO, GWO, WOA and some other popular algorithms given in the literature review of this study are more suitable for continuous optimization). Also the variable workstation constraint causes additional complexity on the problem. Because of these reasons, GA has been used in this study instead of other population-based algorithms in the solution of this new type of WSSP problem [30],[31],[32]. g g p Figure 2. Pseudo Code used for scheduling by GA aided Arena simulation. Figure 3. The representation of a sample chromosome. Figure 4. Flow chart of the proposed GA aided simulation. Figure 2. Pseudo Code used for scheduling by GA aided Arena simulation. Simulating the complex systems allows the researchers to analyze and experiment the internal interactions. Numerical experiments are performed on a digital computer in a simulation job, and a general programming language or a special programming language for simulation may be used [33]. There are several simulation-purposed languages for performing a simulation experiment, such as Arena, Promodel, and etc. In order to integrate directly with other systems such as Microsoft Office, Arena uses ‘Visual Basic for Applications’ (VBA) editor. Arena 14 is used as the simulation software in Figure 2. Pseudo Code used for scheduling by GA aided Arena simulation. Figure 2. Pseudo Code used for scheduling by GA aided Arena simulation. Figure 3. The representation of a sample chromosome. Figure 3. The representation of a sample chromosome g p p Figure 4. Flow chart of the proposed GA aided simulation. this paper for the simulation [7],[13],[33]. 2.2 Simulation-based GA Figure 1. Welding shop layout. GA is run through the data to find the best customer order sequence that minimizes average makespan. Arena is run to measure processing times of the orders and also to calculate the makespan of operations depend on GA-generated order sequences. Please refer to [28],[29] for a detailed discussion on GA and also refer to [7],[13] for a detailed discussion on simulation based GA. JSSP is well-known as NP hard problem. In this very complex problem type; we search for the best solution because there is no known way to arrive at a final answer. Heuristic algorithms are frequently used in JSSP to find the best solution. However, performances of the heuristics are decreased when the inputs become more complex and varied. GA is well suited to solving JSSP because, unlike heuristic Figure 1. Welding shop layout. 143 Pamukkale Univ Muh Bilim Derg, 28(1), 139-147, 2022 A.D. Karaoglan Pamukkale Univ Muh Bilim Derg, 28(1), 139-147, 2022 A.D. Karaoglan there is no other constraint such as the welding station is operating at that moment, the assignment order is from 1 to 6). In the simulation model if an order with vessel length≥8 meters is assigned to a welding station, then the available neighbor welding station is also merged to operate this order. methods, they work on a population of solutions instead of a single solution. This population of solutions in production scheduling consists of several responses with varying and often conflicting goals. For the special types of problems (such presented in WSSP) where the search space is huge but the feasible solutions are limited and also if the discrete optimization is needed, GA is ideal (PSO, GWO, WOA and some other popular algorithms given in the literature review of this study are more suitable for continuous optimization). Also the variable workstation constraint causes additional complexity on the problem. Because of these reasons, GA has been used in this study instead of other population-based algorithms in the solution of this new type of WSSP problem [30],[31],[32]. methods, they work on a population of solutions instead of a single solution. This population of solutions in production scheduling consists of several responses with varying and often conflicting goals. 3 Results and discussions For orders with the numbers 15, 18, and 27, the same problem applies. The idle times of the stations 1st, 4th and 5th are very high, and there is no compliance with the workload balances When an order enters the arena simulation model, it is primarily assigned to a suitable welding station from 1 to 6 (if 144 144 Pamukkale Univ Muh Bilim Derg, 28(1), 139-147, 2022 A.D. Karaoglan between the welding stations. The total CPU time is calculated as 56 seconds for this first scenario. However, the completion time of the 28th order is noticeably increased. The total CPU time is calculated as 68 seconds for this second scenario. However, the completion time of the 28th order is noticeably increased. The total CPU time is calculated as 68 seconds for this second scenario. In the second scenario, to reduce the idle times of the welding stations, some stations are reserved for the width vessels and the average makespan of given orders is calculated as 11732 hours with Arena simulation. According to this scenario; as much as possible, the one with the lowest order number was processed first, but when it came time, idle machine time was tried not to be allowed. Therefore, in some cases, orders with larger order numbers are allowed to be processed first to prevent workstations from waiting idle. However, the order with the lowest order number in the queue continues to be assigned to welding stations (to the reserved stations) at the first opportunity. In the simulation the 1st and 2nd welding stations; and the 4th and 5th welding stations are reserved if the width vessels are in the queue. The 1st and 2nd welding stations are merged while orders 15, 18, and 27 are processed. The 4th and 5th welding stations are also merged during the processing of orders 14, 16, 19, and 28. The idle times of stations 1st, 4th and 5th are significantly decreased as compared to the results shown in Figure 5, but there are still idle times at stations 1st, 3rd and 4th. Also the workload balances between the welding stations still are not in accordance. The average makespan is calculated as 11732 hours. The total completion time is increased when it is compared with classical simulation presented in Figure 5. However the total idle times of the welding stations and the total tardiness for the orders are decreased. 6 Author contribution statement In this study, all of the studies such as creating the idea, making the design, obtaining resources and materials, collecting data, performing analyzes, literature review, writing and critical review were all carried out by Aslan Deniz KARAOĞLAN. 3 Results and discussions Karaoglan Pamukkale Univ Muh Bilim Derg, 28(1), 139-147, 2022 A.D. Karaoglan Figure 7. Gantt chart of the schedule with GA aided Arena. Figure 7. Gantt chart of the schedule with GA aided Arena. The GA aided Arena schedule outperforms the other schedules, because in this schedule the manufacturing of pending order list is completed much earlier when compared with other schedules. The total completion time is nearly 9.25% and 12.06% reduced when compared with the schedules presented in Figure 5 and Figure 6, respectively. Also total idle times of the welding stations are nearly zero (which is a very important managerial criterion) and unlike the other two approaches the workload of the workstations are balanced. This is very important for the manufacturing cost. Because the main cost items are the material cost, labor cost, and the overhead costs. The overhead costs are dispatched to the customer orders according to the labor times. So idle times and unbalanced workstations will lead to incorrect assignment of overhead cost to orders. 4 Conclusion The focus of this study was to determine the best possible waiting order schedule to reduce the average makespan under the conditions of the job-shop and some design requirements constraints. The novelty of this study is using variable workstations constraint under stochastic processing times to minimize average makespan of the orders in a labor-intensive and project type manufacturing. GA is used for this purpose. The average makepan is minimized after 100000 iterations for this problem as a result of assignments made under some constraints. It is observed that the algorithm's computational effectiveness is strong. According to the experimental results, it is noted that for the given order set, the GA-based Arena simulation decreased the average makespan by almost 9.25% (from 11368 to 10317 hours). Also idle times of the welding stations are prevented and the balance between the workstations is provided. Total idle times are nearly zero according to the GA-based Arena schedule. The same problem can be addressed in future studies by taking into account additional vessel design constraints and ergonomic constraints that influence processing times. Also as a future research, under variable workstation constraint fuzzy processing times can be considered instead of using probabilistic distribution and solved by using integer linear programming. In addition, a performance comparison can be presented between the results of integer linear programming and Arena simulation. The focus of this study was to determine the best possible waiting order schedule to reduce the average makespan under the conditions of the job-shop and some design requirements constraints. The novelty of this study is using variable workstations constraint under stochastic processing times to minimize average makespan of the orders in a labor-intensive and project type manufacturing. GA is used for this purpose. There is no conflict of interest with any person/institution in the article prepared. 5 Acknowledgements I'd like to express my gratitude to the editor and the anonymous reviewers, whose insightful comments helped to improve the organization of this paper. I would also like to thank Mr. Esen BİRLİK for his valuable help in revealing this paper. 7 Ethics committee approval and conflict of interest statement There is no need to obtain permission from the ethics committee for the article prepared. 3 Results and discussions The Gantt chart for this schedule is given in Figure 6. When the result presented in Figure 5 and Figure 6 are compared; the completion times for the orders (<8 meters) those have order numbers>17 are decreased seriously. Then GA aided Arena simulation is performed to reduce the makespan. The GA parameters are determined after some preliminary trials. Totally 105 trials are performed for this purpose. The levels of the parameters are determined as: population size (PopSize): [40:100] with 10 step size, crossover rate (cr): [0.4:0.6] with 0.1 step size, and mutation rate (mr): [0.1:0.5] with 0.1 step size. The best performing parameter combination between these trials is selected as the GA parameters. The GA run with maximum number of iterations = 100000, PopSize = 80, crossover rate (cr) = 0.5, and mutation rate (mr) = 0.4 parameters. The average makespan of given order is calculated as 10317 hours. The chromosome which gives the best FV (10317 hours) is 1-2-3-4-5-6-14-15-7-9-8-10- 11-12-18-16-13-17-19-27-20-21-22-23-28-24-25-26-33-32- 29-30-31. The Gantt chart for this schedule is given in Figure 7. The total CPU time is calculated as 36 minutes 28 seconds for this third scenario. According to the results given in Figure 7, the results presented in Figure 5 are improved and by using GA, the balance between the welding stations is provided. Also the dispatching is acceptable when the order numbers are considered. The 95% confidence interval for order sequence generated by GA is calculated as [10270.55 ; 10363.45]. This is quite a narrow confidence interval for this type of production. For the 95% confidence interval the z-critical value is 1.96 according to the statistical tables. This means standard error coefficient is 23.698. So it can be concluded that 10 runs may be enough to validate the results for the presented data for this problem. Figure 5. Gantt chart of the schedule with Arena without using GA (order number has priority). Figure 6. Gantt chart of the schedule with Arena without using GA (reserved stations). Figure 5. Gantt chart of the schedule with Arena without using GA (order number has priority). Figure 5. Gantt chart of the schedule with Arena without using GA (order number has priority). Figure 6. Gantt chart of the schedule with Arena without using GA (reserved stations). Figure 6. Gantt chart of the schedule with Arena without using GA (reserved stations). 145 Pamukkale Univ Muh Bilim Derg, 28(1), 139-147, 2022 A.D. 8 References [1] Jia Z, Lu X, Yang J, Jia D. “Research on job-shop scheduling problem based on genetic algorithm”. International Journal of Production Research, 49(12), 3585-3604, 2011. [2] Azadeh A, Negahban A, Moghaddam M. “A hybrid computer simulation-artificial neural network algorithm for optimisation of dispatching rule selection in stochastic job shop scheduling problems”. International Journal of Production Research, 50(2), 551-566, 2012. [3] Huang XW, Zhao XY, Ma XL. “An improved genetic algorithm for job-shop scheduling problem with process sequence flexibility”. International Journal of Simulation Modelling, 13(4), 510-522, 2014. [4] Aydemir E, Koruca HI. “A new production scheduling module using priority-rule based genetic algorithm”. International Journal of Simulation Modelling, 14(3), 450-462, 2015. [5] Ba L, Li Y, Yang MS, Gao XQ, Liu Y. “Modelling and simulation of a multi-resource flexible job-shop scheduling”. International Journal of Simulation Modelling, 15(1), 157-169, 2016. 146 Pamukkale Univ Muh Bilim Derg, 28(1), 139-147, 2022 A.D. Karaoglan [19] Liao J, Lin C. “Optimization and simulation of job-shop supply chain scheduling in manufacturing enterprises based on particle swarm optimization”. International Journal of Simulation Modelling, 18(1), 187-196, 2019. [6] Deng Q, Gong G, Gong X, Zhang L, Liu W, Ren Q. “A bee evolutionary guiding nondominated sorting genetic algorithm II for multiobjective flexible job-shop scheduling”. Computational Intelligence and Neuroscience, 2017. https://doi.org/10.1155/2017/5232518. [20] Wang Y, Yang O, Wang SN. “A solution to single-machine inverse job-shop scheduling problem”. International Journal of Simulation Modelling, 18(2), 335-343, 2019. [7] Ocaktan MAB, Kucukkoc I, Karaoglan AD, Cicibas A, Buyukozkan K. “Scheduling Customized Orders: A Case Study at BEST Transformers Company”. 6th International Conference on Mechanics and Industrial Engineering (ICMIE’17), Rome, Italy, 8-10 June, 2017. [21] Zhang Z, Guan ZL, Zhang J, Xie X. “A novel job-shop scheduling strategy based on particle swarm optimization and neural network”. International Journal of Simulation modeling, 18(4), 699-707, 2019. [8] Zhang W, Wen JB, Zhu YC, Hu Y. “Multi-objective scheduling simulation of flexible job-shop based on multi- population genetic algorithm”. International Journal of Simulation Modelling, 16(2), 313-321, 2017. [22] Zhu J, Shao ZH, Chen C. “An improved whale optimization algorithm for job-shop scheduling based on quantum computing”. International Journal of Simulation Modelling, 18(3), 521-530, 2019. [9] Hu HX, Lei WX, Gao X, Zhang Y. “job-shop scheduling problem based on improved cuckoo search algorithm”. International Journal of Simulation Modelling, 17(2), 337-346, 2018. [23] Karaoglan AD, Cetin E. Industrial Engineering in the Big Data Era. Editors: Calisir F, Cevikcan E, Akdag HC. 8 References Part I: Industrial Engineering, Artiﬁcial Bee Colony Algorithm for Labor Intensive Project Type Job Shop Scheduling Problem: A Case Study, 79-88, Cham, Switzerland, Springer, 2019. [10] Jiang T, Zhang C, Zhu H, Deng G. “Energy-efficient scheduling for a job shop using grey wolf optimization algorithm with double-searching mode”. Mathematical Problems in Engineering, 2018. https://doi.org/10.1155/2018/8574892 [24] Shi DL, Zhang BB, Li Y. “A multi-objective flexible job-shop scheduling model based on fuzzy theory and immune genetic algorithm”. International Journal of Simulation Modelling, 19(1), 123-133, 2020. g [11] Jiang T, Zhang C, Zhu H, Gu J, Deng G. “Energy-efficient scheduling for a job shop using an improved whale optimization algorithm”. Mathematics, 2018. https://doi org/10 3390/math6110220 [25] Gu J, Jiang T, Zhu H, Zhang C. “Low-carbon job shop scheduling problem with discrete genetic-grey wolf optimization algorithm”. Journal of Advanced Manufacturing Systems, 19(1), 1-14, 2020. [12] Seng DW, Li JW, Fang XJ, Zhang XF, Chen J. “low-carbon flexible job-shop scheduling based on improved nondominated sorting genetic algorithm-II”. International Journal of Simulation Modelling, 17(4), 712-723, 2018. [26] Vital-Soto A, Azab A, Mohammed FB. “Mathematical modeling and a hybridized bacterial foraging optimization algorithm for the flexible job-shop scheduling problem with sequencing flexibility”. Journal of Manufacturing Systems, 54, 74-93, 2020. [13] Karaoglan AD, Ocaktan MAB, Ocaktan DG, Oral A, Kundakci SS, Tuncer C. “Scheduling customized orders by considering the ergonomic constraints: a case study at Yemtar company”. 7th International Conference on Mechanics and Industrial Engineering (ICMIE’18), Madrid, Spain, 16-18 August, 2018. [27] Rao Y, Meng R, Zha J, Xu X. “Bi-objective mathematical model and improved algorithm for optimisation of welding shop scheduling problem”. International Journal of Production Research, 58(9), 2767-2783, 2020. [14] Zhang HP, Ye JH, Yang XP, Muruve NW, Wang JT. ”Modified binary particle swarm optimization algorithm in lot- splitting scheduling involving multiple techniques”. International Journal of Simulation Modelling, 17(3), 534-542, 2018. [28] Holland JH. Adaptation in Natural and Artificial Systems, Cambridge, Massachusetts, USA, MIT Press, 1975. [29] Goldberg DE. Genetic Algorithms in Search, Optimization and Machine Learning, Boston, MA, USA, Addison-Wesley Longman Publishing, 1989. [15] Zhong Q, Yang H, Tang T. “Optimization algorithm simulation for dual-resource constrained job-shop scheduling”. International Journal of Simulation Modelling, 17(1), 147-158, 2018. [30] Lim C, Sim E, “Production planning in manufacturing/remanufacturing environment using genetic algorithm”. Proceedings of the 7th Annual Conference on Genetic and Evolutionary Computation (GECCO’05), Washington, USA, 25-29 June, 2005. [16] Sel C, Hamzadayi A. 8 References "A simulated annealing approach based simulation-optimisation to the dynamic job-shop scheduling problem”. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi, 24(4), 665-674, 2018. [31] Wall M. A Genetic Algorithm for Resource-Constrained Scheduling. PhD Thesis, Massachusetts Institute of Technology, Cambridge-Massachusetts, USA, 1996. [17] Fu HC, Liu P. “A multi-objective optimization model based on non-dominated sorting genetic algorithm”. International Journal of Simulation Modelling, 18(3), 510-520, 2019. [32] Kachitvichyanukul V. "Comparison of three evolutionary algorithms: GA, PSO, and DE". Industrial Engineering & Management Systems, 11(3), 215-223, 2012. [18] Tang H, Chen R, Li Y, Peng Z, Guo S, Du Y. “Flexible job-shop scheduling with tolerated time interval and limited starting time interval based on hybrid discrete PSO-SA: An application from a casting workshop”. Applied Soft Computing, 78, 176-194, 2019. [33] Rosetti MD. Simulation Modeling and Arena. 2nd ed. New Jersey, Hoboken, USA, John Wiley & Sons Inc, 2010. 147
https://openalex.org/W1864382742	https://europepmc.org/articles/pmc2936519?pdf=render	English	null	Instantaneous Non-Linear Processing by Pulse-Coupled Threshold Units	PLOS computational biology/PLoS computational biology	2,010	cc-by	13,767	Introduction currents as low-pass filtered Gaussian white noise with non- vanishing temporal correlations [12–15]. Again, these results are strictly valid only if the synaptic amplitudes tend to zero and their rate of arrival goes to infinity. For finite incoming synaptic events which are excitatory only, the steady state solution can still be obtained analytically [16,17] and also the transient solution can efficiently be obtained by numerical solution of a population equation [18]. A different approach takes into account non-zero synaptic amplitudes to first calculate the free membrane potential distribution and then obtain the firing rate by solving the first passage time problem numerically [19]. This approach may be extendable to conductance based synapses [20]. Exact results for the steady state have so far only been presented for the case of exponentially distributed synaptic amplitudes [21]. Understanding the dynamics of single neurons, recurrent networks of neurons, and spike-timing dependent synaptic plasticity requires the quantification of how a single neuron transfers synaptic input into outgoing spiking activity. If the incoming activity has a slowly varying or constant rate, the membrane potential distribution of the neuron is quasi stationary and its steady state properties characterize how the input is mapped to the output rate. For fast transients in the input, time-dependent neural dynamics gains importance. The integrate-and-fire neuron model [1] can efficiently be simulated [2,3] and well approximates the properties of mammalian neurons [4–6] and more detailed models [7]. It captures the gross features of neural dynamics: The membrane potential is driven by synaptic impulses, each of which causes a small deflection that in the absence of further input relaxes back to a resting level. If the potential reaches a threshold, the neuron emits an action potential and the membrane potential is reset, mimicking the after-hyperpolarization. The spike threshold renders the model an extremely non-linear unit. However, if the synaptic input signal under consideration is small compared to the total synaptic barrage, a linear approx- imation captures the main characteristics of the evoked response. In this scenario all remaining inputs to the neuron are treated as background noise (see Figure 1A). Calculations of the linear response kernel in the diffusion limit suggested that the integrate- and-fire model acts as a low-pass filter [22]. Instantaneous Non-Linear Processing by Pulse-Coupled Threshold Units Moritz Helias1., Moritz Deger2., Stefan Rotter2,3, Markus Diesmann1,2,4 1 RIKEN Brain Science Institute, Wako City, Japan, 2 Bernstein Center for Computational Neuroscience, Freiburg, Germany, 3 Computational Neuroscience Lab, Faculty of Biology, Albert-Ludwig University, Freiburg, Germany, 4 Brain and Neural Systems Team, RIKEN Computational Science Research Program, Wako City, Japan 1 RIKEN Brain Science Institute, Wako City, Japan, 2 Bernstein Center for Computational Neuroscience, Freiburg, Germany, 3 Compu Biology, Albert-Ludwig University, Freiburg, Germany, 4 Brain and Neural Systems Team, RIKEN Computational Science Research P Abstract Contemporary theory of spiking neuronal networks is based on the linear response of the integrate-and-fire neuron model derived in the diffusion limit. We find that for non-zero synaptic weights, the response to transient inputs differs qualitatively from this approximation. The response is instantaneous rather than exhibiting low-pass characteristics, non- linearly dependent on the input amplitude, asymmetric for excitation and inhibition, and is promoted by a characteristic level of synaptic background noise. We show that at threshold the probability density of the potential drops to zero within the range of one synaptic weight and explain how this shapes the response. The novel mechanism is exhibited on the network level and is a generic property of pulse-coupled networks of threshold units. Citation: Helias M, Deger M, Rotter S, Diesmann M (2010) Instantaneous Non-Linear Processing by Pulse-Coupled Threshold Units. PLoS Comput Biol 6(9): e1000929. doi:10.1371/journal.pcbi.1000929 Editor: Karl J. Friston, University College London, United Kingdom Editor: Karl J. Friston, University College London, United Kingdom Received February 25, 2010; Accepted August 10, 2010; Published September 9, 2010 Received February 25, 2010; Accepted August 10, 2010; Published September 9, 2010 Copyright: 2010 Helias et al. This is an open-access article distributed under the terms of the Creative Commons Attributi unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. ias et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits tion, and reproduction in any medium, provided the original author and source are credited. Funding: This work was supported by BMBF Grant 01GQ0420 to BCCN Freiburg, http://www.bmbf.de/en/3063.php; EU Grant 15879 (FACETS), http://cordis. europa.eu/fetch?CALLER = PROJ_ICT&ACTION = D&CAT = PROJ&RCN = 75455; DIP F1.2, http://www.internationales-buero.de/de/787.php; Helmholtz Alliance on Systems Biology (Germany), http://www.helmholtz.de/en/joint_initiative_for_innovation_and_research/initiating_and_networking/helmholtz_alliances/ helmholtz_alliance_on_systems_biology/; Next-Generation Supercomputer Project of MEXT (Japan), http://www.nsc.riken.jp/project-eng.html. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. E-mail: helias@brain.riken.jp . These authors contributed equally to this work. PLoS Computational Biology \| www.ploscompbiol.org September 2010 \| Volume 6 \| Issue 9 \| e1000929 Author Summary Our work demonstrates a fast-firing response of nerve cells that remained unconsidered in network analysis, because it is inaccessible by the otherwise successful linear response theory. For the sake of analytic tractability, this theory assumes infinitesimally weak synaptic coupling. However, realistic synaptic impulses cause a measurable deflection of the membrane potential. Here we quantify the effect of this pulse-coupling on the firing rate and the membrane-potential distribution. We demonstrate how the postsynaptic potentials give rise to a fast, non-linear rate transient present for excitatory, but not for inhibitory, inputs. It is particularly pronounced in the presence of a characteristic level of synaptic background noise. We show that feed-forward inhibition enhances the fast response on the network level. This enables a mode of information processing based on short-lived activity transients. More- over, the non-linear neural response appears on a time scale that critically interacts with spike-timing dependent synaptic plasticity rules. Our results are derived for biologically realistic synaptic amplitudes, but also extend earlier work based on Gaussian white noise. The novel theoretical framework is generically applicable to any threshold unit governed by a stochastic differential equation driven by finite jumps. Therefore, our results are relevant for a wide range of biological, physical, and technical systems. In simulations of neuron models with realistic amplitudes for the postsynaptic potentials, we observed a systematic deviation of the output spike rate and the membrane potential distribution from the predictions by the Fokker-Planck theory modeling synaptic currents by Gaussian white noise. We excluded any artifacts of the numerics by employing a dedicated high accuracy integration algorithm [44,45]. The novel theory developed here explains these observations and lead us to the discovery of a new early component in the response of the neuron model which linear response theory fails to predict. In order to quantify our observations, we extend the existing Fokker-Planck theory [46] and hereby obtain the mean time at which the membrane potential first reaches the threshold; the mean first-passage time. The advantage of the Fokker-Planck approach over alternative techniques has been demonstrated [47]. For non-Gaussian noise, however, the treatment of appropriate boundary conditions for the membrane potential distribution is of utmost importance [48]. In the results section we develop the Fokker-Planck formalism to treat an absorbing boundary (the spiking threshold) in the presence of non-zero jumps (postsynaptic potentials). Author Summary For the special case of simulated systems propagated in time steps, an analog theory has recently been published by the same authors [49], which allows to assess artifacts introduced by time-discretization. has been demonstrated in the regime of low background noise [24]. The maximization of the input-output correlation at a finite amplitude of additional noise is called stochastic resonance and has been found experimentally in mechanoreceptors of crayfish [25], in the cercal sensory system of crickets [26], and in human muscle spindles [27]. The relevance and diversity of stochastic resonance in neurobiology was recently highlighted in a review article [28]. Our theory applied to the integrate-and-fire model with small but finite synaptic amplitudes [1], introduced in section ‘‘The leaky integrate-and-fire model’’, quantitatively explains the deviations of the classical theory for Gaussian white noise input. After reviewing the diffusion approximation of a general first order stochastic differential equation we derive a novel boundary condition in section ‘‘Diffusion with finite increments and absorbing boundary’’. We then demonstrate in section ‘‘Applica- tion to the leaky integrate-and-fire neuron’’ how the steady state properties of the model are influenced: the density just below threshold is increased and the firing rate is reduced, correcting the preexisting mean first-passage time solution [10] for the case of finite jumps. Turning to the dynamic properties, in section ‘‘Response to fast transients’’ we investigate the consequences for transient responses of the firing rate to a synaptic impulse. We find an instantaneous, non-linear response that is not captured by linear perturbation theory in the diffusion limit and that displays marked stochastic resonance. On the network level, we demon- strate in section ‘‘Dominance of the non-linear component on the network level’’ that the non-linear fast response becomes the most important component in case of feed-forward inhibition. In the discussion we consider the limitations of our approach, mention possible extensions and speculate about implications for neural processing and learning. Linear response theory enables the characterization of the recurrent dynamics in random networks by a phase diagram [22,29]. It also yields approximations for the transmission of correlated activity by pairs of neurons in feed-forward networks [30,31]. Furthermore, spike-timing dependent synaptic plasticity is Figure 1. Firing rate response n((t)) to a synaptic input. A The neuron receives excitatory and inhibitory background events (gray spikes) from many synapses. We focus on one such incoming synapse that carries a synaptic impulse at t0 (black spike). Introduction Here spectrum and amplitude of the synaptic background input are decisive for the transient properties of the integrate-and-fire model: in contrast to white noise, low-pass filtered synaptic noise leads to a fast response in the conserved linear term [12]. Linear response theory predicts an optimal level of noise that promotes the response [23]. In the framework of spike-response models, an immediate response depending on the temporal derivative of the postsynaptic potential The analytical treatment of the threshold process is hampered by the pulsed nature of the input. A frequently applied approximation treats synaptic inputs in the diffusion limit, in which postsynaptic potentials are vanishingly small while their rate of arrival is high. In this limit, the summed input can be replaced by a Gaussian white noise current, which enables the application of Fokker-Planck theory [8,9]. For this approximation the stationary membrane potential distribution and the firing rate are known exactly [8,10,11]. The important effect of synaptic filtering has been studied in this limit as well; modelling synaptic PLoS Computational Biology \| www.ploscompbiol.org September 2010 \| Volume 6 \| Issue 9 \| e1000929 September 2010 \| Volume 6 \| Issue 9 \| e1000929 1 Instantaneous Non-Linear Neural Processing sensitive to correlations between the incoming synaptic spike train and the firing of the neuron (see Figure 1), captured up to first order by the linear response kernel [32–38]. For neuron models with non-linear membrane potential dynamics, the linear response properties [39,40] and the time-dependent dynamics can be obtained numerically [41]. Afferent synchronized activity, as it occurs e.g. in primary sensory cortex [42], easily drives a neuron beyond the range of validity of the linear response. In order to understand transmission of correlated activity, the response of a neuron to fast transients with a multiple of a single synaptic amplitude [43] hence needs to be quantified. Author Summary B The firing rate of the neuron triggered on this event shows a deflection (black solid curve) from the base rate n0. If the synaptic efficacy obeys a spike-timing dependent learning rule, the synaptic weight changes by Ds according to the relative timing of the presynaptic spike and the action potentials emitted by the neuron (gray dotted curve indicates typical depen- dence). The time-averaged change in synaptic weight depends on the integrated pointwise product of both curves. Their relative position depends on the axonal and dendritic delays involved (neglected for simplicity in this schematic). d i 10 1371/j l bi 1000929 001 Figure 1. Firing rate response n((t)) to a synaptic input. A The neuron receives excitatory and inhibitory background events (gray spikes) from many synapses. We focus on one such incoming synapse that carries a synaptic impulse at t0 (black spike). B The firing rate of the neuron triggered on this event shows a deflection (black solid curve) from the base rate n0. If the synaptic efficacy obeys a spike-timing dependent learning rule, the synaptic weight changes by Ds according to the relative timing of the presynaptic spike and the action potentials emitted by the neuron (gray dotted curve indicates typical depen- dence). The time-averaged change in synaptic weight depends on the integrated pointwise product of both curves. Their relative position depends on the axonal and dendritic delays involved (neglected for simplicity in this schematic). Figure 1. Firing rate response n((t)) to a synaptic input. A The neuron receives excitatory and inhibitory background events (gray spikes) from many synapses. We focus on one such incoming synapse that carries a synaptic impulse at t0 (black spike). B The firing rate of the neuron triggered on this event shows a deflection (black solid curve) from the base rate n0. If the synaptic efficacy obeys a spike-timing dependent learning rule, the synaptic weight changes by Ds according to the relative timing of the presynaptic spike and the action potentials emitted by the neuron (gray dotted curve indicates typical depen- dence). The time-averaged change in synaptic weight depends on the integrated pointwise product of both curves. Their relative position depends on the axonal and dendritic delays involved (neglected for simplicity in this schematic). PLoS Computational Biology \| www.ploscompbiol.org September 2010 \| Volume 6 \| Issue 9 \| e1000929 Diffusion with finite increments and absorbing boundary The first and second infinites- imal moment evaluate to A1(x)~f (x)zm and A2~s2, where we introduced the shorthand m ~ def JznzzJ{n{ and s2 ~ def J2 znzz J2 {n{. The time evolution of the probability density p(x, t) is then governed by the Kramers-Moyal expansion, which we truncate after the second term to obtain the Fokker-Planck equation where f (y) captures the deterministic time evolution of the system (with f (y)!{y for the leaky integrate-and-fire neuron). We follow the notation in [46] and employ the Kramers-Moyal i i h h i fi i i l A ( ) li 1 expansion with the infinitesimal moments An(x, t)~ limh?0 h S(y(tzh){y(t))nDy(t)~xT n [ N. The first and second infinites- imal moment evaluate to A1(x)~f (x)zm and A2~s2, where we introduced the shorthand m ~ def JznzzJ{n{ and s2 ~ def J2 znzz J2 {n{. The time evolution of the probability density p(x, t) is then governed by the Kramers-Moyal expansion, which we truncate after the second term to obtain the Fokker-Planck equation w~½f (h)zp(h)znzPinst(Jz) ð4Þ ð4Þ L Lt p(x, t)~{ L Lx A1(x){ 1 2 L Lx A2 p(x, t) ~{ L Lx Sp(x, t), ð2Þ ð2Þ Pinst(s)~ ðh h{s p(x)dx, ð5Þ ð5Þ where S~f (x)zm{ s2 2 L Lx denotes the probability flux operator. with ½xz~fx for xw0; 0 elseg. To evaluate the integral in (5), for small Jz%h{SxT we expand q(x) into a Taylor series around h. This is where our main assumption enters: we assume that the stationary Fokker-Planck equation (3) for xvh is a sufficiently accurate characterization of the jump diffusion process. We solve this equation for q’(x)~{ 2 A2 z 2A1(x) A2 q(x) ~ def c1zd1(x)q(x): It with ½xz~fx for xw0; 0 elseg. To evaluate the integral in (5), for small Jz%h{SxT we expand q(x) into a Taylor series around h. This is where our main assumption enters: we assume that the stationary Fokker-Planck equation (3) for xvh is a sufficiently accurate characterization of the jump diffusion process. We solve this equation for q’(x)~{ 2 A2 z 2A1(x) A2 q(x) ~ def c1zd1(x)q(x): It In the presence of an absorbing boundary at h, we need to determine the resulting boundary condition for the stationary solution of (2). Without loss of generality, we assume the absorbing boundary at h to be the right end of the domain. Diffusion with finite increments and absorbing boundary Diffusion with finite increments and absorbing boundary The membrane potential of the model neuron follows a first order stochastic differential equation. Therefore, in this section we consider a general first order stochastic differential equation driven by point events. In order to distinguish the dimensionless quantities in this section from their counterparts in the leaky integrate-and-fire model, we denote the rates of the two incoming Poisson processes by nz (excitation) and n{ (inhibition). Each incoming event causes a finite jump Jk~Jzw0 (the excitatory synaptic weight) for an increasing event and Jk~J{v0 (the inhibitory synaptic weight) for a decreasing event. The stochastic differential equation takes the form q({?)~0 q(h)~0: Moreover, a finite probability flux demands the density to be a continuous function, because of the derivative in the flux operator S. In particular, the solution must be continuous at the point of flux insertion x0 (however, the first derivative is non-continuous at x0 due to the step function in the right hand side of (3)). Continuity especially implies a vanishing density at threshold h. Once the solution of (3) is found, the normalization condition 1~ Ð h {? p(x)dx determines the stationary flux w{1~ Ð h {? q(x)dx. _y~f (y)z X k Jkd(t{tk), ð1Þ Now we return to the problem of finite jumps. We proceed along the same lines as in the diffusion limit, seeking the stationary solution of the Fokker-Planck equation (2). We keep the boundary conditions at {? and at x0 as well as the normalization condition as before, but we need to find a new self-consistent condition at threshold h, because the density does not necessarily have to vanish if the rate of incoming jumps is finite. The main assumption of our work is that the steady state solution satisfies the stationary Fokker-Planck equation (3) based on the diffusion approximation within the interval ½{?, h), but not necessarily at the absorbing boundary h, where the solution might be non-continuous. To obtain the boundary condition, we note that the flux over the threshold has two contributions, the deterministic drift and the positive stochastic jumps crossing the boundary ð1Þ where f (y) captures the deterministic time evolution of the system (with f (y)!{y for the leaky integrate-and-fire neuron). We follow the notation in [46] and employ the Kramers-Moyal expansion with the infinitesimal moments An(x, t)~ limh?0 1 h S(y(tzh){y(t))nDy(t)~xT n [ N. Instantaneous Non-Linear Neural Processing The approach we take is to modify the existing Fokker-Planck theory in order to capture the major effects of the finite jumps. To this end, we derive a novel boundary condition at the firing threshold for the steady state membrane potential distribution of the neuron. We then solve the Fokker-Planck equation obtained from the standard diffusion approximation [8,10,11,22,23] given this new condition. Sq(x)~1x0ƒxƒh f (x)zm{ s2 2 L Lx q(x)~1x0ƒxƒh , ð3Þ ð3Þ with 1C~f1 if C, 0 elseg. with 1C~f1 if C, 0 elseg. First we consider the diffusion limit, in which the rate of incoming events diverges, while the amplitude of jumps goes to zero, such that mean m and fluctuations s remain constant. In this limit, the Kramers-Moyal expansion truncated after the second term becomes exact [51]. This route has been taken before by several authors [8,22,23], here we review these results to consistently present our extension of the theory. In the above limit equation (3) needs to be solved with the boundary conditions PLoS Computational Biology \| www.ploscompbiol.org Instantaneous Non-Linear Neural Processing Instantaneous Non-Linear Neural Processing inhibitory synaptic inputs, as they occur in balanced neural networks [50]. We aim to obtain the mean firing rate n0 and the steady state membrane potential distribution p(V). The input current I(t) is modeled by point events tk [ fincoming spikesg, drawn from homogeneous Poisson processes with rates ne and ni, respectively. The membrane potential is governed by the differential equation t dV dt (t)~{V(t)zRI(t). An excitatory spike causes a jump of the membrane potential by Jk~w, an inhibitory spike by Jk~{gw, so RI(t)~t P Jkd(t{tk)zRI0, where I0 is a constant background current. Whenever V reaches the threshold Vh, the neuron emits a spike and the membrane potential is reset to Vr, where it remains clamped for the absolute refractory time tr. The approach we take is to modify the existing Fokker-Planck theory in order to capture the major effects of the finite jumps. To this end, we derive a novel boundary condition at the firing threshold for the steady state membrane potential distribution of the neuron. We then solve the Fokker-Planck equation obtained from the standard diffusion approximation [8,10,11,22,23] given this new condition. boundary is reinserted into the system. For the example of an integrate-and-fire neuron, reinsertion takes place due to resetting the neuron to the same potential after each threshold crossing. This implies a constant flux w through the system between the point of insertion x0 and threshold h. Rescaling the density by this flux as q(x)~w{1p(x) results in the stationary Focker-Planck equation, which is a linear inhomogeneous differential equation of first order inhibitory synaptic inputs, as they occur in balanced neural networks [50]. We aim to obtain the mean firing rate n0 and the steady state membrane potential distribution p(V). The input current I(t) is modeled by point events tk [ fincoming spikesg, drawn from homogeneous Poisson processes with rates ne and ni, respectively. The membrane potential is governed by the diff i l i dV ( ) V( ) RI( ) A i ik differential equation t dV dt (t)~{V(t)zRI(t). An excitatory spike causes a jump of the membrane potential by Jk~w, an inhibitory spike by Jk~{gw, so RI(t)~t P Jkd(t{tk)zRI0, where I0 is a constant background current. Whenever V reaches the threshold Vh, the neuron emits a spike and the membrane potential is reset to Vr, where it remains clamped for the absolute refractory time tr. The leaky integrate-and-fire model Consider a leaky integrate-and-fire model [1] with membrane time constant t and resistance R receiving excitatory and doi:10.1371/journal.pcbi.1000929.g001 September 2010 \| Volume 6 \| Issue 9 \| e1000929 September 2010 \| Volume 6 \| Issue 9 \| e1000929 September 2010 \| Volume 6 \| Issue 9 \| e1000929 PLoS Computational Biology \| www.ploscompbiol.org 2 Application to the leaky integrate-and-fire neuron We now apply the theory developed in the previous section to the leaky integrate-and-fire neuron with finite postsynaptic potentials. Due to synaptic impulses, the membrane potential drifts towards m ~ def tw(ne{gni) and fluctuates with the diffusion constant s2 ~ def tw2(nezg2ni). This suggests to choose the natural units u~t=t for the time and y~(V{RI0{m)=s for the voltage to obtain the simple expressions A1(y)~{y for the drift- and A2~1 for the diffusion-term in the Fokker-Planck operator (2). The probability flux operator (2) is then given as S~{y{ 1 2 L Ly. In the same units the stationary probability density scaled by the flux reads q(y)~ s tn0 p(V(y)) where tn0 is the flux w corresponding to the firing rate in units of t. As q is already scaled by the flux, application of the flux operator S yields unity between reset yr and threshold yh and zero outside We now apply the theory developed in the previous section to the leaky integrate-and-fire neuron with finite postsynaptic potentials. Due to synaptic impulses, the membrane potential drifts towards m ~ def tw(ne{gni) and fluctuates with the diffusion constant s2 ~ def tw2(nezg2ni). This suggests to choose the natural units u~t=t for the time and y~(V{RI0{m)=s for the voltage to obtain the simple expressions A1(y)~{y for the drift- and A2~1 for the diffusion-term in the Fokker-Planck operator (2). 1 L cnz1~c’nzc1dn dnz1~d’nzd1dn, ð6Þ ð6Þ with the additional values c0(h)~0 and d0(h)~1, as q(0)(x)~q(x) denotes the function itself. Inserting the Taylor series into (5) and performing the integration results in The probability flux operator (2) is then given as S~{y{ 1 2 L Ly. In the same units the stationary probability density scaled by the flux reads q(y)~ s tn0 p(V(y)) where tn0 is the flux w corresponding to the firing rate in units of t. As q is already scaled by the flux, application of the flux operator S yields unity between reset yr and threshold yh and zero outside Pinst(s)~w X ? n~0 1 (nz1)! (cnzdnq)Dh({s)nz1, ð7Þ ð7Þ which is the probability mass moved across threshold by a perturbation of size s and hence also quantifies the instantaneous response of the system. After dividing (4) by w we solve for q(h) to obtain the Dirichlet boundary condition Sq(y)~1yrƒyƒyh {y{ 1 2 L Ly q(y)~1yrƒyƒyh: ð10Þ q(h)~ 1znz P? n~0 1 (nz1)! cn(h)({Jz)nz1 ½f (h)z{nz P? n~0 1 (nz1)! Application to the leaky integrate-and-fire neuron dn(h)({Jz)nz1 : ð8Þ ð10Þ ð8Þ The steady state solution of this stationary Fokker-Planck equation If Jz is small compared to the length scale on which the probability density function varies, the probability density near the threshold is well approximated by a Taylor polynomial of low degree; throughout this work, we truncate (7) and (12) at n~3. The boundary condition (8) is consistent with q(h)~0 in the diffusion limit, in which the rate of incoming jumps diverges, while their amplitude goes to zero, such that the first (m) and second moment (s) stay finite. This can be seen by scaling nz, n{!j, Jz, J{!1= ﬃﬃﬃj p , with j?? such that the mean m~nzJzzn{J{ is kept constant [51]. Inserting this limit in (8), we find If Jz is small compared to the length scale on which the probability density function varies, the probability density near the threshold is well approximated by a Taylor polynomial of low degree; throughout this work, we truncate (7) and (12) at n~3. The boundary condition (8) is consistent with q(h)~0 in the diffusion limit, in which the rate of incoming jumps diverges, while their amplitude goes to zero, such that the first (m) and second moment (s) stay finite. This can be seen by scaling nz, n{!j, Jz, J{!1= ﬃﬃﬃj p , with j?? such that the mean m~nzJzzn{J{ is kept constant [51]. Inserting this limit in (8), we find q(y)~Aqh(y)zqp(y) ð11Þ ð11Þ is a linear superposition of the homogeneous solution qh(y)~e{y2 and the particular solution qp(y)~2e{y2 Ð yh max (yr,y) eu2 du. The latter is chosen to be continuous at yr and to vanish at yh. Using the recurrence (6) for the coeffcients of the Taylor expansion of the membrane potential density, we obtain fcn(y)g~f0, {2, 4y, {8y2z8, . . .g and fdn(y)g~f1, {2y, 4y2{2, {8y3z12y, . . .g, where n starts from 0. The first important result of this section is the boundary value q(yh) of the density at the threshold following from (8) as is a linear superposition of the homogeneous solution qh(y)~e{y2 and the particular solution qp(y)~2e{y2 Ð yh max (yr,y) eu2 du. The latter is chosen to be continuous at yr and to vanish at yh. Using the recurrence (6) for the coeffcients of the Taylor expansion of the membrane potential density, we obtain fcn(y)g~f0, {2, 4y, {8y2z8, . . .g and fdn(y)g~f1, {2y, 4y2{2, {8y3z12y, . . PLoS Computational Biology \| www.ploscompbiol.org Diffusion with finite increments and absorbing boundary A stationary solution exists, if the probability flux exiting at the absorbing September 2010 \| Volume 6 \| Issue 9 \| e1000929 3 Instantaneous Non-Linear Neural Processing is easy to see by induction, that the function and all its higher derivatives q(n), n§0 can be written in the form q(n)(x)~cn(x)z dn(x)q(x), whose coefficients for n§1 obey the recurrence relation Application to the leaky integrate-and-fire neuron .g, where n starts from 0. The first important result of this section is the boundary value q(yh) of the density at the threshold following from (8) as lim j?? q(h)~ lim j?? 1z 1 2 c1(h)nzJ2 z ½f (h)zznzJz{nz 1 2 d1(h)J2 z ~0, ð9Þ q(yh)~ 1ztne P? n~0 1 (nz1)! cn(yh)({ w s )nz1 ½{ Vh{RI0 s z{tne P? n~0 1 (nz1)! dn(yh)({ w s )nz1 : ~ def A(t,ne,ni,w,g,I0)e{yh ð12Þ since c0~0, d0~1 and nzJn z vanishes for nw2, nzJ2 z is bounded and nzJz??. The general solution of the stationary Fokker-Planck equation (3) is a sum of a homogeneous solution qh that satisfies Sqh(x)~0 and a particular solution with Sqp(x)~1x0ƒxƒh. The homoge- neous solution is qh(x)~ exp ( 2 s2 ðx 0 f (u)zmdu), where we fixed the integration constant by chosing qh(0)~1. The particular solution can be obtained by variation of constants and we chose it to vanish at the threshold h as qp(x)~ 2 s2 ðh x qh(x{u)1x0ƒuƒh du. The complete solution is a linear combination, where the prefactor A~q(h)=qh(h) is determined by the boundary condition (8) in the case of finite jumps, or by A~0 for Gaussian white noise The constant A in (11) follows from A~q(yh)=qh(yh). The second result is the steady state firing rate n0 of the neuron. With n0tr being the fraction of neurons which are currently refractory, we obtain the rate from the normalization condition of the density 1~n0t Ð yh {? q(y)dyzn0tr as 1 n0 ~t ﬃﬃﬃp p ðyh yr ey2(erf(y)z1)dyz A 2 (erf(yh)z1) ztr p(x)~w Aqh(x)zqp(x) with w{1~ ðh {? Aqh(x)zqp(x)dx: p(x)~w Aqh(x)zqp(x) with with with with yh,r~ Vh,r{RI0{m s A~A(t, ne, ni, w, g, I0): ð13Þ ð13Þ yh,r~ Vh,r{RI0{m s w{1~ ðh {? Aqh(x)zqp(x)dx: The normalization condition determines the as yet unknown constant probability flux w through the system. The normalized steady state solution Figure 2A therefore has the PLoS Computational Biology \| www.ploscompbiol.org September 2010 \| Volume 6 \| Issue 9 \| e1000929 PLoS Computational Biology \| www.ploscompbiol.org 4 Instantaneous Non-Linear Neural Processing Figure 2. Finite synaptic potentials distort the stationary membrane potential density p((V)). A Black thin curve: direct simulation. Parameters t~20 ms, Vh~15 mV, Vr~0, I0~0, w~0:1 mV, g~4, tr~1 ms. Incoming spike rates ne~29800 Hz, ni~5950 Hz (corresponding to m~12 mV and s~5 mV). Histogram binned with DV~0:01 mV, entries connected by straight lines. Application to the leaky integrate-and-fire neuron Gray: novel approximation p(V) given by (14). B Magnification of A around spike threshold, simulated data displayed as dots. Light gray: solution in diffusion limit of [22]. C,D Density for supra-threshold current RI0~20 mV and incoming rates ne~95050 Hz, ni~22262:5 Hz (corre- sponding to m~12 mV and s~9:5 mV). Other parameters and gray code as in A,B. E Approximation of the density by a cubic polynomial near threshold. Solid light gray curve: analytical result p(V) given by (14), superimposed black thin curve: direct simulation. Dark gray solid curve: cubic polynomial approximating the density around Vh using the Taylor expansion (6). Parameters as in A. F Membrane potential distribution near threshold for synaptic amplitudes w~0:05 mV (black), w~0:1 mV (dark gray), w~0:2 mV (gray), w~0:5 mV (light gray). Other parameters as in A. which the probability density function varies. So the probability density near the threshold is well approximated by a Taylor polynomial of low degree; throughout this work, we truncate the series in (12) at n~3. A comparison of this approximation to the full solution is shown in Figure 2E. For small synaptic amplitudes (w~0:1 mV shown), below threshold and outside the reset region (Figure 2A,C) the approximation agrees with the simulation within its fluctuation. At the threshold (Figure 2B,D) our analytical solution assumes a finite value q(yh)w0 whereas the direct simulation only drops to zero on a very short voltage scale on the order of the synaptic amplitude. For larger synaptic weights (w§0:2 mV, see Figure 2F), the density obtained from direct simulation exhibits a modulation on the corresponding scale. The reason is the rectifying nature of the absorbing boundary: A positive fluctuation easily leads to a threshold crossing and absorption of the state in contrast to negative fluctuations. Effectively, this results in a net drift to lower voltages within the width of the jump distribution caused by synaptic input, visible as the depletion of density directly below the threshold and an accumulation further away, as observed in Figure 2F. The second term (proportional to A) appearing in (13) is a correction to the well known firing rate equation of the integrate-and-fire model driven by Gaussian white noise [10]. Figure 3 compares the firing rate predicted by the new theory to direct simulation and to the classical theory. The classical theory consistently overestimates the firing rate, while our theory yields better accuracy. Application to the leaky integrate-and-fire neuron Our correction resulting from the new boundary condition becomes visible at moderate firing rates when the density slightly below threshold is sufficiently high. At low mean firing rates, the truncation of the Kramers-Moyal expansion employed in the Fokker-Planck description may contribute comparably to the error. Our approximation captures the dependence on the synaptic amplitude correctly for synaptic amplitudes of up to wƒ0:2 mV (Figure 3B). The insets in Figure 3C,D show the relative error of the firing rate as a function of the noise amplitude. As expected, the error increases with the ratio of the synaptic effect w compared to the amplitude of the noise fluctuations s. For low noise s^2:5 mV, our theory reduces the relative error by a factor of 2 compared to the classical diffusion approximation. Figure 2. Finite synaptic potentials distort the stationary membrane potential density p((V)). A Black thin curve: direct simulation. Parameters t~20 ms, Vh~15 mV, Vr~0, I0~0, w~0:1 mV, g~4, tr~1 ms. Incoming spike rates ne~29800 Hz, ni~5950 Hz (corresponding to m~12 mV and s~5 mV). Histogram binned with DV~0:01 mV, entries connected by straight lines. Gray: novel approximation p(V) given by (14). B Magnification of A around spike threshold, simulated data displayed as dots. Light gray: solution in diffusion limit of [22]. C,D Density for supra-threshold current RI0~20 mV and incoming rates ne~95050 Hz, ni~22262:5 Hz (corre- sponding to m~12 mV and s~9:5 mV). Other parameters and gray code as in A,B. E Approximation of the density by a cubic polynomial near threshold. Solid light gray curve: analytical result p(V) given by (14), superimposed black thin curve: direct simulation. Dark gray solid curve: cubic polynomial approximating the density around Vh using the Taylor expansion (6). Parameters as in A. F Membrane potential distribution near threshold for synaptic amplitudes w~0:05 mV (black), w~0:1 mV (dark gray), w~0:2 mV (gray), w~0:5 mV (light gray). Other parameters as in A. PLoS Computational Biology \| www.ploscompbiol.org Response to fast transients Black: Analytical solution (13), light gray: firing rate in diffusion limit [10,29]. Inset shows relative error D~(ntheory{nsim)=nsim of theory compared to direct simulation as percentage, black dots: error of analytical expression (13), light gray dots: error of rate in diffusion limit. Parameters t~20 ms, Vh~15:0 mV, Vr~0, w~0:25 mV, g~4, tr~1 ms. ne (given on top abcissa) and ni chosen to realize the mean input m as given by the bottom abscissa and fluctuations s~5 mV. B Analytical firing rate compared to simulation depending on the size of synaptic jumps w for fixed s~5 mV and m~½7, 9:5, 12, 14 mV (from bottom to top). Light gray dashed line: firing rate in diffusion limit [10,29], other gray code as in A. C Analytical firing rate compared to simulation as a function of the fluctations s at fixed mean drive m~12 mV. Same gray code and other parameters as in A. D Analytical firing rate compared to simulation depending on the fluctation s at fixed m~12 mV for smaller synaptic amplitude w~0:1 mV. Same gray code and other parameters as in A. d i /j l bi p doi:10.1371/journal.pcbi.1000929.g003 time (5), which is zero for sv0. This instantaneous response has several interesting properties: For small s it can be approximated in terms of the value and the slope of the membrane potential dn0 dm from (13) and obtain dn0 dm from (13) and obtain dn0 dm ~{n2 0 t s ﬃﬃﬃp p ey2r (1zerf(yr)){q(yh) zerfc({yh) q(yh){q(yh{ w s ) erf(yh){erf(yh{ w s ) 0 B @ 1 C A 1 C A : ð16Þ Figure 5. Noise dependence of response to a d-current perturbation. A Integrated response of the firing rate (15)of the integrate-and-fire model as a function of synaptic background noise s. The upper traces are the responses to a positive perturbation of magnitude s~0:5 mV, the lower to a perturbation of s~{0:5 mV. Black crosses: direct simulation with background noise of finite synaptic weights w~0:1 mV. Black dots: direct simulation with Gaussian white background noise. Mid gray thin solid curve: analytical result using (16). Underlying thick light gray solid curve: analytical result for Gaussian white noise. B Instantaneous response depending on synaptic background noise s. The upper trace is the response to a positive impulse of weight s~0:5 mV, the lower one to a negative of s~{0:5 mV. Gray code as in A. Response to fast transients doi:10 1371/journal pcbi 1000929 g004 Figure 3. Correction of the firing rate. A Analytical firing rate compared to simulation in dependence of the mean drive m. Black: Analytical solution (13), light gray: firing rate in diffusion limit [10,29]. Inset shows relative error D~(ntheory{nsim)=nsim of theory compared to direct simulation as percentage, black dots: error of analytical expression (13), light gray dots: error of rate in diffusion limit. Parameters t~20 ms, Vh~15:0 mV, Vr~0, w~0:25 mV, g~4, tr~1 ms. ne (given on top abcissa) and ni chosen to realize the mean input m as given by the bottom abscissa and fluctuations s~5 mV. B Analytical firing rate compared to simulation depending on the size of synaptic jumps w for fixed s~5 mV and m~½7, 9:5, 12, 14 mV (from bottom to top). Light gray dashed line: firing rate in diffusion limit [10,29], other gray code as in A. C Analytical firing rate compared to simulation as a function of the fluctations s at fixed mean drive m~12 mV. Same gray code and other parameters as in A. D Analytical firing rate compared to simulation depending on the fluctation s at fixed m~12 mV for smaller synaptic amplitude w~0:1 mV. Same gray code and other parameters as in A. doi:10.1371/journal.pcbi.1000929.g003 Figure 4. Firing rate response to a d-current perturbation. A Black: response to an additionally injected current st R d(t) causing a voltage deflection of s~0:5 mV at t~0, gray: s~{0:5 mV. B Magnification of A. Due to binning of the histogram with bin size h~0:1 ms, the immediate response contributes to the time bin ({h, 0 in the case of a positive perturbation only. Black crosses: analytical peak responses Pinst h (7) for positive and negative perturbations. C Medium gray curve: instantaneous response Pinst (7) as a function of s for finite weights w~0:1 mV. Black dots: direct simulation. Light gray curve: diffusion limit of (7). Medium gray dots: direct simulation of diffusion limit with temporal resolution 10{4 ms. D Gray curve: integral response for finite weights (15). Black dots: direct simulation. Gray dots: direct simulation for Gaussian white noise background input. Simulated data averaged over 2:5:108 (s~0:1 mV) . . . 2:5:106 (s~1:0 mV) perturbation events. Other parameters as in Figure 2A. doi:10.1371/journal.pcbi.1000929.g004 Figure 3. Correction of the firing rate. A Analytical firing rate compared to simulation in dependence of the mean drive m. Response to fast transients A Analytical firing rate e of the mean drive m. Black: rate in diffusion limit [10,29]. m)=nsim of theory compared to ck dots: error of analytical r of rate in diffusion limit. Vr~0, w~0:25 mV, g~4, ni chosen to realize the mean a and fluctuations s~5 mV. B ation depending on the size of d m~½7, 9:5, 12, 14 mV (from e: firing rate in diffusion limit ytical firing rate compared to tions s at fixed mean drive arameters as in A. D Analytical ending on the fluctation s at plitude w~0:1 mV. Same gray Figure 4. Firing rate response to a d-current perturbation. A Black: response to an additionally injected current st R d(t) causing a voltage deflection of s~0:5 mV at t~0, gray: s~{0:5 mV. B Magnification of A. Due to binning of the histogram with bin size h~0:1 ms, the immediate response contributes to the time bin ({h, 0 in the case of a positive perturbation only. Black crosses: analytical peak responses Pinst h (7) for positive and negative perturbations. C Medium gray curve: instantaneous response Pinst (7) as a function of s for finite weights w~0:1 mV. Black dots: direct simulation. Light gray curve: diffusion limit of (7). Medium gray dots: direct simulation of diffusion limit with temporal resolution 10{4 ms. D Gray curve: integral response for finite weights (15). Black dots: direct simulation. Gray dots: direct simulation for Gaussian white noise background input. Simulated data averaged over 2:5:108 (s~0:1 mV) . . . 2:5:106 (s~1:0 mV) perturbation events. Other parameters as in Figure 2A. doi:10.1371/journal.pcbi.1000929.g004 Figure 4. Firing rate response to a d-current perturbation. A Black: response to an additionally injected current st R d(t) causing a voltage deflection of s~0:5 mV at t~0, gray: s~{0:5 mV. B Magnification of A. Due to binning of the histogram with bin size h~0:1 ms, the immediate response contributes to the time bin ({h, 0 in the case of a positive perturbation only. Black crosses: analytical peak responses Pinst h (7) for positive and negative perturbations. C Medium gray curve: instantaneous response Pinst (7) as a function of s for finite weights w~0:1 mV. Black dots: direct simulation. Light gray curve: diffusion limit of (7). Medium gray dots: direct simulation of diffusion limit with temporal resolution 10{4 ms. D Gray curve: integral response for finite weights (15). Black dots: direct simulation. Response to fast transients We now proceed to obtain the response of the firing rate n to an additional d-shaped input current ts R d(t). Such a current can be due to a single synaptic event or due to the synchronized arrival of several synaptic pulses. In the latter case, the effective amplitude of the summed inputs can easily exceed that of a single synapse. The fast current transient ts R d(t) causes a jump s of the membrane potential at t~0 and (2) suggests to treat the incident as a time dependent perturbation of the mean input m(t)~mzstd(t). First, we are interested in the integral response Pr(s) ~ def Ð ? 0 fs(t)dt of the excess firing rate fs(t)~ns(t){n0. Since the perturbation has a flat spectrum, up to linear order in s the spectrum of the excess rate is ^fs(z)~stH(z)zO(s2), where H(z) is the linear transfer function with respect to perturbing m at Laplace frequency z. In particular, Pr(s)~^fs(0). As H(0) is the DC susceptibility of the system, we can express it up to linear order as H(0)~ Ln0 Lm. Hence, doi:10.1371/journal.pcbi.1000929.g002 complete form complete form p(V)~ n0t s Ae{y2z2e{y2 ðyh max (yr,y) eu2 du : ð14Þ ð14Þ Figure 2B,D shows the steady state solution near the threshold obtained by direct simulation to agree much better with our analytical approximation than with the theory for Gaussian white noise input. Even for synaptic amplitudes (here w~0:1 mV) which are considerably smaller than the noise fluctuations (here s~5 mV), the effect is still well visible. The oscillatory deviations with periodicity w close to reset Vr~0 observable in Figure 2A are due to the higher occupation probability of voltages that are integer multiples of a synaptic jump away from reset. The modulation washes out due to coupling of adjacent voltages by the deterministic drift as one moves away from reset. The oscillations at lower frequencies apparent in Figure 2A are due to aliasing caused by the finite bin width of the histogram (0:01 mV). The synaptic weight is typically small compared to the length scale s on Pr(s)~ ð? 0 n(t){n0 dt~st dn0 dm zO(s2): ð15Þ ð15Þ We also take into account the dependence of A on m to calculate We also take into account the dependence of A on m to calculate September 2010 \| Volume 6 \| Issue 9 \| e1000929 PLoS Computational Biology \| www.ploscompbiol.org 5 Instantaneous Non-Linear Neural Processing rate. Response to fast transients Gray dots: direct simulation for Gaussian white noise background input. Simulated data averaged over 2:5:108 (s~0:1 mV) . . . 2:5:106 (s~1:0 mV) perturbation events. Other parameters as in Figure 2A. doi:10.1371/journal.pcbi.1000929.g004 Figure 3. Correction of the firing rate. A Analytical firing rate compared to simulation in dependence of the mean drive m. Black: Analytical solution (13), light gray: firing rate in diffusion limit [10,29]. Inset shows relative error D~(ntheory{nsim)=nsim of theory compared to direct simulation as percentage, black dots: error of analytical expression (13), light gray dots: error of rate in diffusion limit. Parameters t~20 ms, Vh~15:0 mV, Vr~0, w~0:25 mV, g~4, tr~1 ms. ne (given on top abcissa) and ni chosen to realize the mean input m as given by the bottom abscissa and fluctuations s~5 mV. B Analytical firing rate compared to simulation depending on the size of synaptic jumps w for fixed s~5 mV and m~½7, 9:5, 12, 14 mV (from bottom to top). Light gray dashed line: firing rate in diffusion limit [10,29], other gray code as in A. C Analytical firing rate compared to simulation as a function of the fluctations s at fixed mean drive m~12 mV. Same gray code and other parameters as in A. D Analytical firing rate compared to simulation depending on the fluctation s at fixed m~12 mV for smaller synaptic amplitude w~0:1 mV. Same gray code and other parameters as in A. Figure 4. Firing rate response to a d-current perturbation. A Black: response to an additionally injected current st R d(t) causing a voltage deflection of s~0:5 mV at t~0, gray: s~{0:5 mV. B Magnification of A. Due to binning of the histogram with bin size h~0:1 ms, the immediate response contributes to the time bin ({h, 0 in the case of a positive perturbation only. Black crosses: analytical peak responses Pinst h (7) for positive and negative perturbations. C Medium gray curve: instantaneous response Pinst (7) as a function of s for finite weights w~0:1 mV. Black dots: direct simulation. Light gray curve: diffusion limit of (7). Medium gray dots: direct simulation of diffusion limit with temporal resolution 10{4 ms. D Gray curve: integral response for finite weights (15). Black dots: direct simulation. Gray dots: direct simulation for Gaussian white noise background input. Simulated data averaged over 2:5:108 (s~0:1 mV) . . . 2:5:106 (s~1:0 mV) perturbation events. Other parameters as in Figure 2A. PLoS Computational Biology \| www.ploscompbiol.org Response to fast transients Mid gray solid curve: analytical peak response Pinst using (7) for noise of finite synaptic weights. Light gray solid curve: analytical peak response Pinst for Gaussian white background noise. Simulations averaged over N~1000 neurons for T~100 s. Incoming spike rates ne, ni chosen to realize m~12 mV and s on the abscissa for synaptic weights w~0:1 mV and g~4. doi:10.1371/journal.pcbi.1000929.g005 ð16Þ Figure 4D shows the integral response to be in good agreement with the linear approximation. This expression is consistent with the result in the diffusion limit w?0: Here the last term becomes Figure 4D shows the integral response to be in good agreement with the linear approximation. This expression is consistent with the result in the diffusion limit w?0: Here the last term becomes Figure 5. Noise dependence of response to a d-current perturbation. A Integrated response of the firing rate (15)of the integrate-and-fire model as a function of synaptic background noise s. The upper traces are the responses to a positive perturbation of magnitude s~0:5 mV, the lower to a perturbation of s~{0:5 mV. Black crosses: direct simulation with background noise of finite synaptic weights w~0:1 mV. Black dots: direct simulation with Gaussian white background noise. Mid gray thin solid curve: analytical result using (16). Underlying thick light gray solid curve: analytical result for Gaussian white noise. B Instantaneous response depending on synaptic background noise s. The upper trace is the response to a positive impulse of weight s~0:5 mV, the lower one to a negative of s~{0:5 mV. Gray code as in A. Mid gray solid curve: analytical peak response Pinst using (7) for noise of finite synaptic weights. Light gray solid curve: analytical peak response Pinst for Gaussian white background noise. Simulations averaged over N~1000 neurons for T~100 s. Incoming spike rates ne, ni chosen to realize m~12 mV and s on the abscissa for synaptic weights w~0:1 mV and g~4. doi:10.1371/journal.pcbi.1000929.g005 q(yh){q(yh{ w s ) erf(yh){erf(yh{ w s ) ~ Lq Ly (yh) 2ﬃﬃﬃp p e{y2 h ~{ ﬃﬃﬃp p ey2 h, where we used Lq Ly (yh)~{2, following from (10) with q(yh)~0. This results in dn0 dm ~n2 0 t s ﬃﬃﬃp p ey2 h(1zerf(yh)){ey2r (1zerf(yr)) , which can equivalently be obtained directly as the derivative of (13) with respect to m setting A~0. Response to fast transients Taking the limit w?0, however, does not change significantly the integral response compared to the case of finite synaptic amplitudes (Figure 4D, Figure 5A). q(yh){q(yh{ w s ) erf(yh){erf(yh{ w s ) ~ Lq Ly (yh) 2ﬃﬃﬃp p e{y2 h ~{ ﬃﬃﬃp p ey2 h, where we used Lq Ly (yh)~{2, following from (10) with q(yh)~0. This results in dn0 dm ~n2 0 t s ﬃﬃﬃp p ey2 h(1zerf(yh)){ey2r (1zerf(yr)) , which can equivalently be obtained directly as the derivative of (13) with respect to m setting A~0. Taking the limit w?0, however, does not change significantly the integral response compared to the case of finite synaptic amplitudes (Figure 4D, Figure 5A). q(yh){q(yh{ w s ) erf(yh){erf(yh{ w s ) ~ Lq Ly (yh) 2ﬃﬃﬃp p e{y2 h ~{ ﬃﬃﬃp p ey2 h, where we used Lq Ly q(yh){q(yh{ w s ) erf(yh){erf(yh{ w s ) ~ Lq Ly (yh) 2ﬃﬃﬃp p e{y2 h ~{ ﬃﬃﬃp p ey2 h, where we used Lq Ly (yh)~{2, following from (10) with q(yh)~0. This results in dn0 dm ~n2 0 t s ﬃﬃﬃp p ey2 h(1zerf(yh)){ey2r (1zerf(yr)) , which can equivalently be obtained directly as the derivative of (13) with respect to m setting A~0. Taking the limit w?0, however, does not change significantly the integral response compared to the case of finite synaptic amplitudes (Figure 4D, Figure 5A). p (yh)~{2, following from (10) with q(yh)~0. This results in dn0 dm ~n2 0 t s ﬃﬃﬃp p ey2 h(1zerf(yh)){ey2r (1zerf(yr)) , which can The instantaneous response of the firing rate to an impulse-like perturbation can be quantified without further approximation. The perturbation shifts the probability density by s so that neurons with V [ ½Vh{s, Vh] immediately fire. This results in the finite firing probability Pinst(s) of the single neuron within infinitesimal PLoS Computational Biology \| www.ploscompbiol.org September 2010 \| Volume 6 \| Issue 9 \| e1000929 6 Instantaneous Non-Linear Neural Processing Figure 6. Dominance of non-linearity at the network level. A Two identical populations of N~1000 neurons each, receive uncorre- lated background input (light gray spikes). At t~0 the neurons simultaneously receive an additional input of size s in the upper and {s in the lower population (symbolized by black single spike). B Pooled response of the populations normalized by the number of neurons. Black cross: analytical instantaneous response Pnet(s). C Magnification of B. Response to fast transients D Pnet(s) (black dots: direct simulation, gray curve: analytical result) as a function of s. Other parameters as in Figure 4A. doi:10.1371/journal.pcbi.1000929.g006 distribution below the threshold (using (7) for nƒ2), so it has a linear and a quadratic contribution in s. Figure 4A shows a typical response of the firing rate to a perturbation. The peak value for a positive perturbation agrees well with the analytical approximation (7) (Figure 4C). Even in the diffusion limit, replacing the background input by Gaussian white noise, the instantaneous response persists. Using the boundary condition q(yh)~0 our theory is applicable to this case as well. Since the density just below threshold is reduced, (5) yields a smaller instantaneous response (Figure 4C, Figure 5B) which for positive s still exhibits a quadratic, but no linear, dependence. The increasing and convex dependence of the response probability on the amplitude of the perturbation is a generic feature of neurons with subthreshold mean input that also persists in the case of finite synaptic rise time. In this regime, the membrane potential distribution has a mono-modal shape centered around the mean input, which is inherited from the underlying superposition of a large number of small synaptic impulses. The decay of the density towards the threshold is further enhanced by the probability flux over the threshold: a positive synaptic fluctuation easily leads to the emission of a spike and therefore to the absorption of the state at the threshold, depleting the density there. Consequently, the response probability Pinst(s) of the neuron is increasing and convex as long as the peak amplitude s of the postsynaptic potential is smaller than the distance of the peak of the density to the threshold. It is increasing and concave beyond this point. At present the integrate-and-fire model is the simplest analytically tractable model with this feature. Figure 6. Dominance of non-linearity at the network level. A Two identical populations of N~1000 neurons each, receive uncorre- lated background input (light gray spikes). At t~0 the neurons simultaneously receive an additional input of size s in the upper and {s in the lower population (symbolized by black single spike). B Pooled response of the populations normalized by the number of neurons. Black cross: analytical instantaneous response Pnet(s). C Magnification of B. D Pnet(s) (black dots: direct simulation, gray curve: analytical result) as a function of s. Other parameters as in Figure 4A. Discussion In this work we investigate the effect of small, but non-zero synaptic impulses on the steady state and response properties of the integrate-and-fire neuron model. We obtain a more accurate description of the firing rate and the membrane potential distribution in the steady state than provided by the classical approximation of Gaussian white noise input currents [10]. Technically this is achieved by a novel hybrid approach combining a diffusive description of the membrane potential dynamics far away from the spiking threshold with an explicit treatment of threshold crossings by synaptic transients. This allows us to obtain a boundary condition for the membrane potential density at threshold that captures the observed elevation of density. Our work demonstrates that in addition to synaptic filtering, the granularity of the noise due to finite non-zero amplitudes does affect the steady state and the transient response properties of the neuron. Here, we study the effect of granularity using the example of a simple neuron model with only one dynamic variable. The quantitatively similar increase of the density close to threshold observed if low-pass filtered Gaussian white noise is used as a model for the synaptic current has a different origin. It is due to the absence of a diffusion term in the dynamics of the membrane potential [12,13,15]. The analytical treatment of finite synaptic amplitudes further allows us to characterize the probability of spike Response to fast transients doi:10.1371/journal.pcbi.1000929.g006 that the dependence of peak height Pnet(s)~ 1 2 Pinst(s) on s still exhibits the supra-linearity. The quite exact cancellation of the response for tw0 originates from the symmetry of the response functions for positive and negative perturbations in this interval (shown in Figure 4A,B). The pooled firing rate of the network is the sum of the full responses: the instantaneous response at t~0 does not share the symmetry and hence does not cancel. This demonstrates that the result of linear perturbation theory is a good approximation for tw0 and that the instantaneous response at the single time point t~0 completes the characterization of the neuronal response. The integral response (15) as well as the instantaneous response (5) both exhibit stochastic resonance; an optimal level of synaptic background noise s enhances the transient. Figure 5A shows this noise level to be at about s~3 mV for the integral response. The responses to positive and negative perturbations are symmetric and the maximum is relatively broad. The instantaneous response in Figure 5B displays a pronounced peak at a similar value of s. This non-linear response only exists for positive perturbations; the response is zero for negative ones. Though the amplitude is reduced in the case of Gaussian white noise background, the behavior is qualitatively the same as for noise with finite jumps. Stochastic resonance has been reported for the linear response to sinusoidal periodic stimulation [23]. Also for non-periodic signals that are slow compared to the neuron’s dynamics an adiabatic approximation reveals stochastic resonance [52]. In contrast to the latter study, the rate transient observed in our work is the instantaneous response to a fast (Dirac d) synaptic current. PLoS Computational Biology \| www.ploscompbiol.org September 2010 \| Volume 6 \| Issue 9 \| e1000929 Dominance of the non-linear component at the network l l This is the reason why previous work based on linear response theory did not report on the existence of an instantaneous component when modulating the mean input and on the contrary characterized the nerve cell as a low-pass in this case [22,23]. Modulation of the noise amplitude, however, has been shown to cause an instantaneous response in linear approximation in the diffusion limit [23], confirmed experimentally in real neurons [55]. While linear response theory has proven extremely useful to understand recurrent neural networks [29], the categorization of the integrate-and-fire neuron’s response kernel as a low-pass is misleading, because it suggests the absence of an immediate response. Furthermore we find that in addition to the nature of the background noise, response properties also depend on its amplitude: a certain level of noise optimally promotes the spiking response. Hence noise facilitates the transmission of the input to the output of the neuron. This is stochastic resonance in the general sense of the term as recently suggested [28]. As noted in the introduction, stochastic resonance of the linear response kernel has previously been demonstrated for sinusoidal input currents and Gaussian white background noise [23]. Furthermore, also slow aperiodic transients are facilitated by stochastic resonance in the integrate-and-fire neuron [52]. We extend the known results in two respects. Firstly, we show that the linear response shows aperiodic stochastic resonance also for fast transients. Secondly, we demonstrate that the instantaneous non-linear response exhibits a qualitatively similar, but even more pronounced dependence on noise intensity. For realistically small synaptic amplitudes, the instantaneous non- linear response is typically small compared to the linear contribu- The instantaneous non-linear response is potentially a relevant mechanism for processing of transient signals by neurons. In auditory cortex, the irregular firing of neurons has been shown to be driven by simultaneous coactivation of several of their synaptic afferents [56]. The effective postsynaptic potential hence has the amplitude of multiple single synapses, which easily drives the spiking response into the supra-linear regime. The convex increase of firing probability is of advantage to obtain output spikes closely locked to the input. Furthermore, the non-linearity enables the neuron to perform non-trivial computations on the inputs [57]. In particular the memory capacity of networks in a categorization task can be increased by non-linear elements [58]. Dominance of the non-linear component at the network l l The circuit presented in section ‘‘Dominance of the non-linear component at the network level’’ establishes a quadratic input-output relationship for fast transient signals that may be useful for non-linear processing, analogous to the non-linear f-I curve (spike frequency as a function of input current) in the case of quasi-stationary rate-coded signals. Our finding of an immediate non-linear response has an implication on the intensely debated question how common input affects the correlation of the spiking activity of pairs of neurons [30,31,59]. The immediate response adds to the correlation at zero time lag, because it increases the probability of both neurons to simultaneously emit a spike. Due to the non-linearity of the mechanism, the immediate firing probability easily becomes the dominant contribution. Our theory yields a means to quantita- tively assess this contribution to firing synchrony. Synapses with spike timing dependent plasticity (STDP) [60] are sensitive to the input-spike triggered firing rate of the neuron. The fast response is relevant, because closely time-locked pre- and postsynaptic activity most effectively changes the synaptic weight. This is illustrated in Figure 1B. The direction of weight change depends on whether the fast response falls on the potentiating or the depotentiating part of the STDP curve, determined by the difference between dendritic and axonal synaptic delay [33]. Assuming that the causal fast response strengthens the synapse (Hebbian learning [61]), the supra-linearity combined with multiplicative spike-timing-dependent learning rules may add new fixed points for the synaptic weight and thus influence pattern formation in recurrent networks [62]. Previous work restricted the analysis of the interplay of neural dynamics and synaptic plasticity in feed-forward [32–34] as well as in recurrent networks [35–38] to the linear response of the neuron. Our framework extends the scope of analytical investigations of synaptic dynamics to the inherently non-linear response properties of neurons. The pronounced stochastic resonance of the individual neuron implies an optimal level of synaptic background noise that supports cooperativity among afferent synapses and hence also the sensitivity to correlations among them [34]. Measuring synaptic plasticity in the presence of network activity might elucidate how stochastic resonance influences cooperative synaptic learning. Postsynaptic potentials exhibit a finite rise time, whereas the membrane potential of the integrate-and-fire neuron model jumps at each incoming synaptic event. Although this is a simplification, the model reproduces experimental spike trains surprisingly well [6]. Dominance of the non-linear component at the network l l The infinitely fast rise of the postsynaptic potential in the integrate-and-fire model leads to the immediate emission of a spike with finite probability. For excitatory inputs, this probability depends supra-linearly on the amplitude of the synaptic impulse and it is zero for inhibitory impulses. The supra-linear increase for small positive impulse amplitudes relates to the fact that the membrane potential density decreases towards threshold: the probability to instantaneously emit a spike equals the integral of the density shifted over the threshold. The detailed shape of the density below threshold therefore determines the response properties. For Gaussian white noise synaptic back- ground, the model still displays an instantaneous response. However, since in this case the density vanishes at threshold, the response probability to lowest order grows quadratically in the amplitude of a synaptic impulse. This is the reason why previous work based on linear response theory did not report on the existence of an instantaneous component when modulating the mean input and on the contrary characterized the nerve cell as a low-pass in this case [22,23]. Modulation of the noise amplitude, however, has been shown to cause an instantaneous response in linear approximation in the diffusion limit [23], confirmed experimentally in real neurons [55]. While linear response theory has proven extremely useful to understand recurrent neural networks [29], the categorization of the integrate-and-fire neuron’s response kernel as a low-pass is misleading, because it suggests the absence of an immediate response. Furthermore we find that in addition to the nature of the background noise, response properties also depend on its amplitude: a certain level of noise optimally promotes the spiking response. Hence noise facilitates the transmission of the input to the output of the neuron. This is stochastic resonance in the general sense of the term as recently suggested [28]. As noted in the introduction, stochastic resonance of the linear response kernel has previously been demonstrated for sinusoidal input currents and Gaussian white background noise [23]. Furthermore, also slow aperiodic transients are facilitated by stochastic resonance in the integrate-and-fire tion. However, this changes at the network level in the presence of feed-forward inhibition: a synchronized pair of an excitatory and an inhibitory pulse evokes spiking responses in two distinct neural populations, whose linear contributions mutually cancel and only the non-linear immediate contribution remains. Hence the immediate response dominates even for small synaptic amplitudes. Dominance of the non-linear component at the network l l Due to the convex nature of the instantaneous response (Figure 4C) its relative contribution to the integral response increases with s. For realistic synaptic weights ƒ1 mV the contribution reaches ^30 percent. An example network in which the linear non-instantaneous response cancels completely and the instantaneous response becomes dominant is shown in Figure 6A. At t~0 two populations of neurons simultaneously receive a perturbation of size s and {s respectively. This activity may, for example, originate from a third pool of synchronous excitatory and inhibitory neurons. It may thus be interpreted as feed-forward inhibition. The linear contributions to the pooled firing rate response of the former two populations hence is zero. The instantaneous response, however, causes a very brief overshoot at t~0 (Figure 6B). Figure 6C reveals that the response returns to baseline within ^0:3 ms. Figure 6D shows PLoS Computational Biology \| www.ploscompbiol.org September 2010 \| Volume 6 \| Issue 9 \| e1000929 September 2010 \| Volume 6 \| Issue 9 \| e1000929 7 Instantaneous Non-Linear Neural Processing emission in response to synaptic inputs for neuron models with a single dynamical variable and renewal. Alternatively, this response can be obtained numerically from population descriptions [18, 39–41] or, for models with one or more dynamic variables and gradually changing inputs, in the framework of the refractory density approximation [15]. Here, we find that the response can be decomposed into a fast, non-linear and a slow linear contribution, as observed experimentally about a quarter of a century ago [53] in motor neurons of cat cortex in the presence of background noise. The existence of a fast contribution proportional to the temporal change of the membrane potential was predicted theoretically [54]. In the framework of the refractory density approach [15], the effective hazard function of an integrate-and-fire neuron also exhibits contributions to spike emission due to two distinct causes: the diffusive flow through the threshold and the movement of density towards the threshold. The latter contribution is propor- tional to the temporal change of the membrane potential and is corresponding to the instantaneous response reported here, but for the case of a gradually increasing membrane potential. Contem- porary theory of recurrent networks so far has neglected the transient non-linear component of the neural response, an experimentally observed feature [53] that is generic to threshold units in the presence of noise. September 2010 \| Volume 6 \| Issue 9 \| e1000929 PLoS Computational Biology \| www.ploscompbiol.org Dominance of the non-linear component at the network l l The presented approximate analytical results are illustrated and confirmed by direct simulation. emission in response to synaptic inputs for neuron models with a single dynamical variable and renewal. Alternatively, this response can be obtained numerically from population descriptions [18, 39–41] or, for models with one or more dynamic variables and gradually changing inputs, in the framework of the refractory density approximation [15]. Here, we find that the response can be decomposed into a fast, non-linear and a slow linear contribution, as observed experimentally about a quarter of a century ago [53] in motor neurons of cat cortex in the presence of background noise. The existence of a fast contribution proportional to the temporal change of the membrane potential was predicted theoretically [54]. In the framework of the refractory density approach [15], the effective hazard function of an integrate-and-fire neuron also exhibits contributions to spike emission due to two distinct causes: the diffusive flow through the threshold and the movement of density towards the threshold. The latter contribution is propor- tional to the temporal change of the membrane potential and is corresponding to the instantaneous response reported here, but for the case of a gradually increasing membrane potential. Contem- porary theory of recurrent networks so far has neglected the transient non-linear component of the neural response, an experimentally observed feature [53] that is generic to threshold units in the presence of noise. The infinitely fast rise of the postsynaptic potential in the integrate-and-fire model leads to the immediate emission of a spike with finite probability. For excitatory inputs, this probability depends supra-linearly on the amplitude of the synaptic impulse and it is zero for inhibitory impulses. The supra-linear increase for small positive impulse amplitudes relates to the fact that the membrane potential density decreases towards threshold: the probability to instantaneously emit a spike equals the integral of the density shifted over the threshold. The detailed shape of the density below threshold therefore determines the response properties. For Gaussian white noise synaptic back- ground, the model still displays an instantaneous response. However, since in this case the density vanishes at threshold, the response probability to lowest order grows quadratically in the amplitude of a synaptic impulse. Dominance of the non-linear component at the network l l For non-zero rise times, the instantaneous firing rate response reported here is spread out in time over the rising flank of the PLoS Computational Biology \| www.ploscompbiol.org September 2010 \| Volume 6 \| Issue 9 \| e1000929 September 2010 \| Volume 6 \| Issue 9 \| e1000929 8 Instantaneous Non-Linear Neural Processing regions below threshold and around the reset potential are outside the scope of our theory. The response properties considered in this work are entirely based on the assumption, that the dynamics has reached the steady state prior to arrival of the perturbing input. A valuable future extension of our work is to consider finite amplitude synaptic background noise and additional sinusoidal current injection. This would allow to quantify in a frequency resolved manner how the transfer properties of the model are influenced by finite-grained noise. Technically, the linear perturbation theory for the diffusion limit [22] would have to be combined with our boundary condition. Complications might arise from the fact that the boundary condition is now time- dependent if the mean drive reaches suprathreshold values in certain epochs. Our treatment of stochastic differential equations with finite jumps and absorbing boundaries is general, as long as the jumps are sufficiently small. We expect it to be applicable to other fluctuation driven dynamical systems in quantitative biology and physics. Potential areas include the diffusion of particles in domains with absorbing walls, chemical reactions with activation thresholds, circuit theory and solid state physics. postsynaptic potential and is proportional to the derivative of the membrane potential [43,54]. The asymmetry for excitatory and inhibitory synaptic events and the supralinear increase of the response probability with excitatory postsynaptic amplitude, however, are generic features that carry over to finite rise time if the neuron operates in the fluctuation driven regime. Comparing the non-linear and the linear response probability experimentally can serve as an indicator to decide on the importance of each contribution in real neurons. The integral linear response can be obtained from similar arguments as in section ‘‘Response to fast transients’’ as Pr(s)~ s Rm tLn0 LI with the slope of the f-I curve Ln0 LI and the membrane resistance Rm. Previous work has shown that the spike generation mechanism influences the transient properties of neurons [63,64]. Specifically, a soft threshold, as realized in the exponential integrate-and-fire neuron model [63] is more realistic than the hard threshold of the leaky integrate-and-fire model considered here. Acknowledgments The hybrid approach combining a diffusion approximation with an explicit treatment of finite jumps near the boundary allowed us to uncover hitherto unknown properties of the integrate-and-fire model by analytical means. The diffusion approximation, however, still limits our approach: for synaptic amplitudes w§0:2 mV moments of order higher than two, which are neglected by the Fokker-Planck equation, become relevant. A combination of our boundary condition with an assessment of higher moments [19,65] seems promising. Also, the oscillatory modulations of the probability density on a scale w%s in the We acknowledge fruitful discussions with Carl van Vreeswijk, Nicolas Brunel, Benjamin Lindner and Petr Lansky and are grateful to our colleagues in the NEST Initiative. 1. Stein RB (1965) A theoretical analysis of neuronal variability. Biophys J 5: 173–194. Dominance of the non-linear component at the network l l Future work needs to investigate how this affects the fast response. We expect qualitatively similar findings, because a positive synaptic impulse shifts membrane potential density into the basin of attraction for spike generation. This will then result in an increased spiking density in a finite time window following the synaptic event. Author Contributions Analyzed the data: M. Helias, M. Deger. Contributed reagents/materials/ analysis tools: M. Helias, M. Deger. Wrote the paper: M. Helias, M. Deger, S. Rotter, M. Diesmann. Performed the analytic analysis and simulations: M. Helias, M. Deger. Supervised the theoretical analysis and simulations: S. Rotter, M. Diesmann. References 1. Stein RB (1965) A theoretical analysis of neuronal variability. Biophys J 5: 173–194. 1. Stein RB (1965) A theoretical analysis of neuronal variability. Biophys J 5: 173–194. 16. Sirovich L, Omurtag A, Knight BW (2000) Dynamics of neuronal populations: The equilibrium solution. SIAM J Appl Math 60: 2009–2028. q J pp 17. Sirovich L (2003) Dynamics of neuronal populations: eigenfunction theory; some solvable cases. Network: Comput Neural Systems 14: 249–272. 2. Rotter S, Diesmann M (1999) Exact digital simulation of time-invariant linear systems with applications to neuronal modeling. Biol Cybern 81: 381–402. solvable cases. Network: Comput Neural Systems 14: 249–272. 18. Omurtag A, Knight BW, Sirovich L (2000) On the simulation of large populations of neurons. J Comput Neurosci 8: 51–63. 3. Gewaltig MO, Diesmann M (2007) NEST (Neural Simulation Tool). Scholarpedia J 2: 1430. populations of neurons. J Comput Neurosci 8: 51–63. 19. Hohn N, Burkitt AN (2001) Shot noise in the leaky integrate-and-fire neuron. Phys Rev E 63: 1–11. 4. Rauch A, La Camera G, Lu¨scher H, Senn W, Fusi S (2003) Neocortical pyramidal cells respond as integrate-and-fire neurons to in vivo like input currents. J Neurophysiol 90: 1598–1612. 20. Wolff L, Lindner B (2008) Method to calculate the moments of the membrane voltage in a model neuron driven by multiplicative filtered shot noise. Phys Rev E 77. p y 5. La Camera G, Rauch A, Luscher H, Senn W, Fusi S (2004) Minimal models of adapted neuronal response to in vivo-like input currents. Neural Comput 10: 2101–2124. 21. Jacobsen M, Jensen AT (2007) Exit times for a class of piecewise exponential markov processes with two-sided jumps. Stoch Proc Appl 117: 1330–1356. 6. Jovilet R, Kobayashi R, Rauch A, Naud R, Shinomoto S, et al. (2008) A benchmark test for a quantitative assessment of simple neuron models. J Neurosci Methods 169: 417–424. p j p pp 22. Brunel N, Hakim V (1999) Fast global oscillations in networks of integrate-and- fire neurons with low firing rates. Neural Comput 11: 1621–1671. 22. Brunel N, Hakim V (1999) Fast global oscillations in networks of fire neurons with low firing rates. Neural Comput 11: 1621–167 23. Lindner B, Schimansky-Geier L (2001) Transmission of noise coded versus additive signals through a neuronal ensemble. Phys Rev Lett 86: 2934–2937. 7. Izhikevich E (2004) Which model to use for cortical spiking neurons? IEEE Trans Neural Netw 5: 1063–1070. g g y 24. 2. Rotter S, Diesmann M (1999) Exact digital simulation of time-invariant linear systems with applications to neuronal modeling. Biol Cybern 81: 381–402. PLoS Computational Biology \| www.ploscompbiol.org References Gerstner W (2000) Population dynamics of spiking neurons: fast transien asynchronous states, and locking. Neural Comput 12: 43–89. 8. Ricciardi LM, Sacerdote L (1979) The Ornstein-Uhlenbeck process as a model for neuronal activity. Biol Cybern 35: 1–9. asynchronous states, and locking. Neural Comput 12: 43–89. 25. Douglass JK, Wilkens L, Pantazelou E, Moss F (1993) Noise enhancement of information transfer in crayfish mechanoreceptors by stochastic resonance. Nature 365: 337–340. 9. Risken H (1996) The Fokker-Planck Equation Springer Verlag Berlin Heidelberg. 10. Siegert AJ (1951) On the first passage time probability problem. Phys Rev 81: 617–623. 26. Levin JE, Miller JP (1996) Broadband neural encoding in the cricket cereal sensory system enhanced by stochastic resonance. Nature 380: 165–168. 11. Johannesma PIM (1968) Diffusion models for the stochastic activity of neurons. In: Caianiello ER, ed. Neural Networks: Proceedings of the School on Neural Networks Ravello, June 1967 Springer-Verlag Verlag. pp 116–144. 27. Cordo P, Inglis JT, Sabine V, Collins JJ, Merfeld DM, et al. (1996) Noise in human muscle spindles. Nature 383: 769–770. 28. McDonnell MD, Abbott D (2009) What is stochastic resonance? definitions, misconceptions, debates, and its relevance to biology. PLoS Comput Biol 5: e1000348. doi:10.1371/journal.pcbi.1000348. 12. Brunel N, Chance FS, Fourcaud N, Abbott LF (2001) Effects of synaptic noise and filtering on the frequency response of spiking neurons. Phys Rev Lett 86: 2186–2189. 29. Brunel N (2000) Dynamics of sparsely connected networks of excitatory and inhibitory spiking neurons. J Comput Neurosci 8: 183–208. 13. Fourcaud N, Brunel N (2002) Dynamics of the firing probability of noisy integrate-and-fire neurons. Neural Comput 14: 2057–2110. 30. Tetzlaff T, Buschermo¨hle M, Geisel T, Diesmann M (2003) The spread of rate and correlation in stationary cortical networks. Neurocomputing 52–54: 949–954. 14. Moreno-Bote R, Parga N (2006) Auto- and crosscorrelograms for the spike response of leaky integrate-and-fire neurons with slow synapses. Phys Rev Lett 96: 028101. 31. De la Rocha J, Doiron B, Shea-Brown E, Kresimir J, Reyes A (2007) Correlation between neural spike trains increases with firing rate. Nature 448: 802–807. 15. Chizhov AV, Graham LJ (2008) Efficient evaluation of neuron populations receiving colored-noise current based on a refractory density method. Phys Rev E 77: 011910. 32. Kempter R, Gerstner W, van Hemmen JL (1999) Hebbian learning and spiking neurons. Phys Rev E 59: 4498–4514. Instantaneous Non-Linear Neural Processing 33. Morrison A, Diesmann M, Gerstner W (2008) Phenomenological models of synaptic plasticity based on spike-timing. Biol Cybern 98: 459–478. 48. Hanggi P, Talkner P (1985) First-passage time problems for non-Markovian processes. Phys Rev A 32: 1934–1937. 34. Helias M, Rotter S, Gewaltig M, Diesmann M (2008) Structural plasticity controlled by calcium based correlation detection. Front Comput Neurosci 2: doi:10.3389/neuro.10.007.2008. 49. Helias M, Deger M, Diesmann M, Rotter S (2010) Equilibrium and response properties of the integrate-and-fire neuron in discrete time. Front Comput Neurosci 3: doi:10.3389/neuro.10.029.2009. 35. Gilson M, Burkitt AN, Grayden DB, Thomas DA, van Hemmen JL (2009) Emergence of network structure due to spike-timing-dependent plasticity in recurrent neuronal networks. I. Input selectivity - strengthening correlated input pathways. Biol Cybern 101: 81–102. 50. van Vreeswijk C, Sompolinsky H (1996) Chaos in neuronal networks with balanced excitatory and inhibitory activity. Science 274: 1724–1726. 51. Risken H (1989) The Fokker-Planck Equation Springer, 2nd edit 52. Collins JJ, Chow CC, Capela AC, Imhoff TT (1996) Aperiodic stochastic resonance. Phys Rev E 54: 5575–5584. p y y 36. Gilson M, Burkitt AN, Grayden DB, Thomas DA, van Hemmen JL (2009) Emergence of network structure due to spike-timing-dependent plasticity in recurrent neuronal networks. II. Input selectivity - symmetry breaking. Biol Cybern 101: 103–114. 53. Fetz E, Gustafsson B (1983) Relation between shapes of postsynaptic potentials and changes in firing probability of cat motoneurones. J Physiol (Lond) 341: 387–410. 37. Gilson M, Burkitt AN, Grayden DB, Thomas DA, van Hemmen JL (2009) Emergence of network structure due to spike-timing-dependent plasticity in recurrent neuronal networks III. Partially connected neurons driven by spontaneous activity. Biol Cybern 101: 411–426. 54. Knox CK, Poppele RE (1977) Correlation analysis of stimulus-evoked changes in excitability of spontaneously firing neurons. J Neurophysiol 40: 616–625. 55. Silberberg G, Bethge M, Markram H, Pawelzik K, Tsodyks M (2004) Dynamics of population codes in ensembles of neocortical neurons. J Neurophysiol 91: 704–709. p y y 38. Gilson M, Burkitt AN, Grayden DB, Thomas DA, van Hemmen JL (2009) Emergence of network structure due to spike-timing-dependent plasticity in recurrent neuronal networks IV. Structuring synaptic pathways among recurrent connections. Biol Cybern 101: 427–444. 56. DeWeese MR, Zador AM (2006) Non-gaussian membrane potential dynamics imply sparse, synchronous activity in auditory cortex. J Neurosci 26: 12206–12218. 39. Richardson MJE (2007) Firing-rate response of linear and nonlinear integrate- and-fire neurons to modulated current-based and conductance-based synaptic drive. Instantaneous Non-Linear Neural Processing Phys Rev E 76: 1–15. 57. Herz AVM, Gollisch T, Machens CK, Jaeger D (2006) Modeling single-neuron dynamics and computations: A balance of detail and abstraction. Science 314: 80–84. y 40. Richardson MJE (2008) Spike-train spectra and network response functions for non-linear integrate-and-fire neurons. Biol Cybern 99: 381–392. 58. Poirazi P, Mel BW (2001) Impact of active dendrites and structural plasticity on the memory capacity of neural tissue. Neuron 29: 779–796. non-linear integrate-and-fire neurons. Biol Cybern 99: 381–392. 41. De Kamps M (2003) A simple and stable numerical solution for the population density equation. Neural Comput 15: 2129–2146. 59. Renart A, De La Rocha J, Bartho P, Hollender L, Parga N, et al. (2010) The asynchronous state in cortical cicuits. Science 327: 587–590. density equation. Neural Comput 15: 2129–2146. 42. Poulet J, Petersen C (2008) Internal brain state regulates membrane potential synchrony in barrel cortex of behaving mice. Nature 454: 881–5. 60. Bi Gq, Poo Mm (1998) Synaptic modifications in cultured hippocampal neurons: Dependence on spike timing, synaptic strength, and postsynaptic cell type. J Neurosci 18: 10464–10472. synchrony in barrel cortex of behaving mice. Nature 454: 881–5 y y g 43. Goedeke S, Diesmann M (2008) The mechanism of synchronization in feed- forward neuronal networks. New J Phys 10: 015007. 61. Hebb DO (1949) The organization of behavior: A neuropsychological theory. New York: John Wiley & Sons. 44. Morrison A, Straube S, Plesser HE, Diesmann M (2007) Exact subthreshold integration with continuous spike times in discrete time neural network simulations. Neural Comput 19: 47–79. J y 62. Morrison A, Aertsen A, Diesmann M (2007) Spike-timing dependent plasticity in balanced random networks. Neural Comput 19: 1437–1467. balanced random networks. Neural Comput 19: 1437–1467. p 45. Hanuschkin A, Kunkel S, Helias M, Morrison A, Diesmann M (2010) A general and efficient method for incorporating precise spike times in globally time-driven simulations. Submitted. 63. Fourcaud-Trocme´ N, Hansel D, van Vreeswijk C, Brunel (2003) How spike generation mechanisms determine the neuronal response to fluctuating inputs. J Neurosci 23: 11628–11640. 46. Ricciardi LM, Di Crescenzo A, Giorno V, Nobile AG (1999) An outline of theoretical and algorithmic approaches to first passage time problems with applications to biological modeling. MathJaponica 50: 247–322. J 64. Naundorf B, Geisel T, Wolf F (2005) Action potential onset dynamics and the response speed of neuronal populations. J Comput Neurosci 18: 297–309. 65. References PLoS Computational Biology \| www.ploscompbiol.org September 2010 \| Volume 6 \| Issue 9 \| e1000929 September 2010 \| Volume 6 \| Issue 9 \| e1000929 9 PLoS Computational Biology \| www.ploscompbiol.org Instantaneous Non-Linear Neural Processing Kuhn A, Aertsen A, Rotter S (2003) Higher-order statistics of input ensembles and the response of simple model neurons. Neural Comput 15: 67–101. 47. Van den Broeck C, Ha¨nggi P (1984) Activation rates for nonlinear stochastic flows driven by non-gaussian noise. Phys Rev A 30: 2730–2736. PLoS Computational Biology \| www.ploscompbiol.org September 2010 \| Volume 6 \| Issue 9 \| e1000929 10
https://openalex.org/W4389407420	https://liksprava.com/index.php/journal/article/download/1074/977	Ukrainian	null	Взаємозв’язок показників церебрального кровотоку і рефлекторних проявів при цервікогенній вертебробазилярній недостатності мозкового кровообігу	Lìkarsʹka sprava/Lìkarsʹka sprava	2,013	cc-by	2,581	76 ISSN 0049-6804. «Лік. справа=Врачеб. дело». 2013, № 4 НА ДОПОМОГУ ПРАКТИЧНОМУ ЛІКАРЕВ 76 ISSN 0049-6804. «Лік. справа=Врачеб. дело». 2013, № 4 НА ДОПОМОГУ ПРАКТИЧНОМУ ЛІКАРЕВ УДК 616.831–005–008.64–08 Надійшла 12.12.2012 6 ISSN 0049-6804. «Лік. справа=Врачеб. дело». 2013, № 4 НА ДОПОМОГУ ПРАКТИЧНОМУ ЛІКАРЕВІ О. М. Козар-Гуріна. Взаємозв’язок показників церебрального кровотоку при ВБН О. М. Козар-Гуріна. Взаємозв’язок показників церебрального кровотоку при ВБН взаємозв’язок між рефлекторними проявами ВБН та показниками церебрального крово- току. Патологічний рефлекс Россолімо – Вендеровича у хворих свідчить про ранню стадію недостатності мозкового кровообігу у вертебробазилярному басейні та збере- женість ауторегуляції церебрального кровообігу. Симптом Хвостека ІІІ у хворих з ВБН, навпаки, вказує на поглиблення ВБН та зрив процесів ауторегуляції церебрального кро- вообігу. Ключові слова: цервікогенна вертебробазилярна недостатність мозкового кровообігу, рефлекторні прояви, показники церебрального кровотоку. Вступ. З допомогою сучасних діагностичних досліджень нервової системи не завжди можна виявити причину захворювання у зв’язку з тим, що на початкових стадіях недостатності мозкового кровообігу відсутні макроанатомічні зміни у тка- нині мозку [5]. Клінічний неврологічний огляд хворого дає можливість встано- вити діагноз на початкових стадіях вертебробазилярної недостатності (ВБН). У зв’язку з цим важливим є порівняння даних клініко-неврологічного обстежен- ня з показниками кровотоку по церебральних судинах, отриманими за допомогою транскраніальної допплерографії. У наукових працях відмічається неоднозначність даних, отриманих за допомогою допплерографії [1–3]: у деяких хворих з недо- статністю мозкового кровообігу швидкість кровотоку по церебральних судинах підвищується, у решти знижується. Це частково пов’язано з тим, що існують два басейни кровопостачання головного мозку – каротидний та вертебробазилярний і ці басейни кровопостачання взаємопов’язані, тому зміни кровопостачання в од- ному басейні призводять до змін кровопостачання в іншому [4]. В процесі до- слідження ми вирішили порівняти рефлекторні прояви у хворих з ВБН, які, на нашу думку, відображають кровопостачання головного мозку, з показниками кро- вотоку, отриманими при транскраніальній допплерографії (ТКД). у р р р р р р ф ( ) Мета дослідження – покращання діагностики ВБН шляхом виявлення взаємозв’язку між показниками церебрального кровотоку і рефлекторними про- явами. Матеріали і методи. У 100 хворих (27 чоловіків та 73 жінки) з ВБН, пов’язаною з патологією шийного відділу хребта, віком від 20 до 50 років проводили клініко- неврологічний огляд і залежно від проявів патологічних рефлексів хворих поді- лили на три групи. Під час транскраніальної допплерогафії (ТКД) церебральних судин отримали пікові систолічні швидкості кровотоку в артеріях вертебробази- лярного басейну та у вені Розенталя. Отримані клінічні та допплерографічні дані порівнювали за допомогою статистичних методів. Результати та їх обговорення. При клініко-неврологічному обстеженні хворих з цервікогенною ВБН виявлено такі рефлекторні прояви: м’язово-тонічний син- дром у м’язах задньої поверхні шиї; різниця у сухожилкових рефлексах на руках та ногах; патологічний рефлекс Россолімо – Вендеровича; симптом Хвостека ІІІ. М’язово-тонічний синдром виявлений у м’язах задньої поверхні шиї у всіх хворих. О. М. КОЗАР-ГУРІНА ВЗАЄМОЗВ’ЯЗОК ПОКАЗНИКІВ ЦЕРЕБРАЛЬНОГО КРОВОТОКУ І РЕФЛЕКТОРНИХ ПРОЯВІВ ПРИ ЦЕРВІКОГЕННІЙ ВЕРТЕБРО- БАЗИЛЯРНІЙ НЕДОСТАТНОСТІ МОЗКОВОГО КРОВООБІГУ Кафедра невропатології та нейрохірургії (зав. – проф. А. В. Паєнок) Львівського національ- ного медичного університету ім. Данила Галицького <kozar.gurina@gmail.com> У 100 хворих з цервікогенною вертебробазилярною недостатністю (ВБН) мозкового кровообігу проводили клініко-неврологічне обстеження та транскраніальну допплеро- графію судин вертебробазилярного басейну. Виявлені закономірності, що вказують на © Î. Ì. Êîçàð-Ãóðіíà, 2013 77 О. М. Козар-Гуріна. Взаємозв’язок показників церебрального кровотоку при ВБН Тонус у м’язах задньої поверхні шиї був асиметричним, підвищеним у більшості хворих зліва (55 %), а у меншої кількості справа (45 %). При цьому у чоловіків тонус м’язів задньої поверхні шиї частіше був підвищеним справа (52 %), у жінок – зліва (57,5 %). Зміни у сухожилкових рефлексах відмічали у 93 % хворих і тільки у 7 % хворих сухожилкові рефлекси були нормальними (рівномірними та симетрич- ними). б б У більшості випадків (45 %) сухожилкові рефлекси були посилені зліва як на руці, так і на нозі. У 31 % хворих сухожилкові рефлекси були посилені справа як на руці, так і на нозі. Перехресний тип сухожилкових рефлексів (по- силені на одній руці та на протилежній до руки нозі) ми спостерігали у 17 % хворих. Отже, у 76 % хворих з ВБН сухожилкові рефлекси були посилені за гемітипом. Під час дослідження співвідношення м’язово-тонічного синдрому в м’язах задньої поверхні шиї і різниці сухожилкових рефлексів у хворих виявлено певні ISSN 0049-6804. «Лік. справа=Врачеб. дело». 2013, № 4 НА ДОПОМОГУ ПРАКТИЧНОМУ ЛІКАРЕВІ 78 закономірності. У більшості хворих з ВБН (36 %) сторона локалізації підвищен- ня тонусу м’язів задньої поверхні шиї збігалась із стороною посилення сухожил- кових рефлексів. закономірності. У більшості хворих з ВБН (36 %) сторона локалізації підвищен- ня тонусу м’язів задньої поверхні шиї збігалась із стороною посилення сухожил- кових рефлексів. р ф В результаті неврологічного огляду хворих з ВБН виявлено також патологіч- ний рефлекс Россолімо – Вендеровича та/або симптом Хвостека ІІІ у 85 % хворих. З них рефлекс Россолімо – Вендеровича був виявлений у 45 хворих (І група), симптом Хвостека ІІІ відмічено у 21 хворого (ІІ група). У 19 хворих – і рефлекс Россолімо – Вендеровича і симптом Хвостека ІІІ (ІІІ група). Виявлені клінічні дані вказують на те, що рефлекс Россолімо – Вендеровича та симптом Хвостека ІІІ є об'єктивними клінічними діагностичними критеріями недостатності мозкового кровообігу у вертебробазилярному басейні, пов’язаної з патологією шийного від- ділу хребта. д у р Розподіл рефлекторних проявів, виявлених при неврологічному обстеженні хворих на цервікогенну ВБН, наведено у табл. 1. Таблиця 1. Розподіл рефлекторних проявів, виявлених при неврологічному обстеженні хворих з цервікогенною вертебробазилярною недостатністю Рефлекторні прояви Кількість хворих абс. од. % М’язово-тонічний синдром у м’язах шиї 100 100 Різниця у сухожилкових рефлексах на руках та ногах 93 93 Рефлекс Россолімо – Вендеровича та/або симптом Хвостека ІІІ 85 85 Таблиця 1. О. М. Козар-Гуріна. Взаємозв’язок показників церебрального кровотоку при ВБН Розподіл рефлекторних проявів, виявлених при неврологічному обстеженні хворих з цервікогенною вертебробазилярною недостатністю За допомогою ТКД обстежено 100 хворих з ВБН і визначено пікову систоліч- ну швидкість кровотоку (Vps) в артеріях вертебробазилярного басейну: у правій і лівій хребтових артеріях (ХА), в основній артерії (ОА) та у правій і лівій задніх мозкових артеріях (ЗМА), а також пікову систолічну швидкість кровотоку у вені Розенталя (ВР). ( ) Дані пікових систолічних (Vps) швидкостей кровотоку в артеріях вертеброба- зилярного басейну та у вені Розенталя наведено в табл. 2. Таблиця 2. Дані пікових систолічних швидкостей кровотоку в артеріях вертебробазилярного басейну та у вені Розенталя у хворих з цервікогенною вертебробазилярною недостатністю (M ± m) Судина Vps Коефіцієнт достовірності, Р1 ЗМА права 63,46 ± 1,25 < 0,001* ліва 66,21 ± 1,14 < 0,01* ОА 79,23 ± 1,42 < 0,001* ХА права 49,42 ± 2,11 < 0,001* ліва 52,47 ± 1,67 < 0,01* ВР 27,39 ± 1,18 < 0,001* * Статистично достовірні показники. Таблиця 2. Дані пікових систолічних швидкостей кровотоку в артеріях вертебробазилярного басейну та у вені Розенталя у хворих з цервікогенною вертебробазилярною недостатністю (M ± m) Для виявлення взаємозв’язку між показниками церебрального кровотоку і рефлекторними проявами у хворих з ВБН обрали систолічну швидкість крово- току в ОА. Це пов’язано з тим, що ОА непарна і це спрощує оцінку результатів обстеження. Однак отримані дані систолічних швидкостей кровотоку по ОА були неоднозначними. В одних пацієнтів швидкість кровотоку по ОА була підвищеною, в інших – зниженою. На наш погляд, це пов’язано із станом ауторегуляції мозко- вого кровотоку. Підвищення систолічної швидкості кровотоку в церебральних артеріях, зокрема в ОА, при ВБН може свідчити про спазм церебральних судин і О. М. Козар-Гуріна. Взаємозв’язок показників церебрального кровотоку при ВБН 79 бути проявом компенсаторних механізмів в умовах недостатності кровотоку. Зни- ження швидкості кровотоку в артеріях вертебробазилярного басейну, зокрема в ОА, на нашу думку, може свідчити про зрив компенсаторних механізмів цере- брального кровотоку при ВБН. У трьох групах хворих, сформованих залежно від проявів патологічних рефлексів, ми розрахували середню Vps в ОА. Отримані ре- зультати зображені на рис. 1. Рис. 1. Взаємозв’язок патологічних рефлексів та систолічної швидкості кровотоку в основній артерії у хворих з вертебробазилярною недостатністю: 1 – рефлекс Россолімо – Вендеровича; 2 – симптом Хвостека ІІІ; 3 – рефлекс Россолімо – Вендеровича + симптом Хвостека ІІІ Рис. 1. у р р р р 1 – симптом Хвостека ІІІ; 2 – рефлекс Россолімо – Вендеровича О. М. Козар-Гуріна. Взаємозв’язок показників церебрального кровотоку при ВБН Взаємозв’язок патологічних рефлексів та систолічної швидкості кровотоку в основній артерії у хворих з вертебробазилярною недостатністю: 1 – рефлекс Россолімо – Вендеровича; 2 – симптом Хвостека ІІІ; 3 – рефлекс Россолімо – Вендеровича + симптом Хвостека ІІІ На рис. 1 видно, що у хворих І групи з рефлексом Россолімо – Вендеровича середня Vps в ОА максимальна і становить (86,50 ± 1,82) см/с (Р < 0,01). У хворих ІІ групи із симптомом Хвостека ІІІ середня Vps в ОА найменша – (67,80 ± 1,68) см/с (Р < 0,01). У ІІІ групі хворих з поєднанням рефлексу Россолімо– Вендеровича і симптому Хвостека ІІІ середня Vps в ОА проміжна між середніми Vps в І та ІІ групах і становить (78,20 ± 1,75) см/с (Р < 0,01). ps Ми також вирішили дослідити зв’язок систолічної швидкості кровотоку у ВР і виявлених патологічних рефлексів. У нормі Vps у ВР дорівнює 12 см/с. Отрима- ні дані зображено на рис. 2, з якого видно, що у хворих з цервікогенною ВБН систолічна швидкість у ВР є підвищеною. Однак у тих хворих, у яких виявлений рефлекс Россолімо – Вендеровича, середня систолічна швидкість кровотоку (Vps) у ВР є вищою, ніж у тих хворих, у яких виявлений симптом Хвостека ІІІ. Рис. 2. Взаємозв’язок патологічних рефлексів та систолічної швидкості кровотоку (Vps) у вені Розенталя у хворих з вертебробазилярною недостатністю: 1 – симптом Хвостека ІІІ; 2 – рефлекс Россолімо – Вендеровича Рис. 2. Взаємозв’язок патологічних рефлексів та систолічної швидкості кровотоку (Vps) у вені Розенталя у хворих з вертебробазилярною недостатністю: 80 ISSN 0049-6804. «Лік. справа=Врачеб. дело». 2013, № 4 НА ДОПОМОГУ ПРАКТИЧНОМУ ЛІКАРЕВ Отже, показники церебрального кровотоку і рефлекторні прояви при ВБН є взаємопов’язаними. У хворих з рефлекторними проявами у вигляді патологічно- го рефлексу Россолімо – Вендеровича виявляється найбільша систолічна швид- кість кровотоку в основній артерії і вені Розенталя, що пов’язано з добре нала- штованими процесами ауторегуляції мозкового кровообігу. Значне підвищення швидкості кровотоку в основній артерії можна пояснити компенсаторним спазмом ОА у відповідь на недостатність церебрального кровотоку, а пов’язане з ним під- вищення швидкості кровотоку у ВР сприяє кращому відтоку крові з порожнини черепа. У хворих із симптомом Хвостека ІІІ систолічна швидкість кровотоку в ОА найнижча, що свідчить про погано налаштований стан ауторегуляції мозко- вого кровообігу. У хворих з рефлексом Россолімо – Вендеровича і симптомом Хвостека ІІІ систолічна швидкість кровотоку в ОА є проміжною, що свідчить про перехідний (від добре налаштованого до погано налаштованого) стан ауторегуля- ції мозкового кровообігу. С п и с о к л і т е р а т у р и Bornsteina Cerebral vasomotor reactivity of patients with acute ischemic stroke: Cortical versus subcorti- cal infarcts: An Israeli – Turkish collaborative study // J. of the Neurolog. Sciences. – 2007. – Vol. 257 (Issues 1–2). – P. 121–125. О. М. Козар-Гуріна. Взаємозв’язок показників церебрального кровотоку при ВБН р у Таким чином, рефлекс Россолімо – Вендеровича у хворих вказує на початко- ву стадію недостатності мозкового кровообігу у вертебробазилярному басейні та збереження механізмів ауторегуляції церебрального кровообігу, симптом Хвосте- ка ІІІ, навпаки, – на поглиблену стадію недостатності мозкового кровообігу у вертебробазилярному басейні та порушення механізмів ауторегуляції церебраль- ного кровообігу. р у Висновки. 1. При ВБН існує взаємозв’язок між рефлекторними проявами і показниками церебрального кровотоку. 2. Патологічний рефлекс Россолімо – Вен- деровича у хворих з цервікогенною ВБН пов’язаний з більш значним підвищенням швидкості кровотоку в основній артерії та вені Розенталя і вказує на ранню ста- дію недостатності мозкового кровообігу у вертебробазилярному басейні та збере- женість ауторегуляції церебрального кровообігу. 3. Симптом Хвостека ІІІ у хворих з цервікогенною ВБН пов’язаний з менш значним підвищенням швидкості крово- току по основній артерії та вені Розенталя і вказує на поглиблену стадію недо- статності мозкового кровообігу у вертебробазилярному басейні та на зрив ауто- регуляції церебрального кровообігу. С п и с о к л і т е р а т у р и 1. Бутко Д. Ю. Состояние церебральной гемодинамики и статокинетических функций у боль- ных с вертебрально-базилярной сосудистой недостаточностью // Журн. невропатологии и психиатрии им. С. С. Корсакова. – 2004. – Т. 104, Вып. 12. – С. 38–42. 1. Бутко Д. Ю. Состояние церебральной гемодинамики и статокинетических функций у боль- ных с вертебрально-базилярной сосудистой недостаточностью // Журн. невропатологии и психиатрии им. С. С. Корсакова. – 2004. – Т. 104, Вып. 12. – С. 38–42. 2. Задорожний А. М. Стан церебрального кровотоку у хворих з гіпоталамічним синдромом // Тез. наук.-практ. конф. «Здоров’я та довголіття». – Л., 2012. – С. 46–47. 3. Орос М. М. Використання доплерографії для визначення ступеня важкості дисциркулятор- ної енцефалографії // Укр. вісн. психоневрології. – 2002. – Т. 10, Вип. 3, № 32. – С. 20–21. Ф 5 3. Орос М. М. Використання доплерографії для визначення ступеня важкості дисциркулятор- ної енцефалографії // Укр. вісн. психоневрології. – 2002. – Т. 10, Вип. 3, № 32. – С. 20–21. 4. Физиология человека: В 4 т.: Пер с англ. / Под ред. Р. Шмидта и Г. Тевса. – М.: Мир, 1985. – Т 3 – 288 с 3. Орос М. М. Використання доплерографії для визначення ступеня важкості дисциркулятор- ної енцефалографії // Укр. вісн. психоневрології. – 2002. – Т. 10, Вип. 3, № 32. – С. 20–21. 4. Физиология человека: В 4 т.: Пер с англ. / Под ред. Р. Шмидта и Г. Тевса. – М.: Мир, 1985. – ц ф р ф // р р , , 4. Физиология человека: В 4 т.: Пер с англ. / Под ред. Р. Шмидта и Г. Тевса. – М.: Мир, 1985. – Т. 3. – 288 с. 5. Gura Alexander Y., Gücüyenerb Demet, Üzünerb Nevzat et al. Korczyna and Natan M. Bornsteina Cerebral vasomotor reactivity of patients with acute ischemic stroke: Cortical versus subcorti- cal infarcts: An Israeli – Turkish collaborative study // J. of the Neurolog. Sciences. – 2007. – Vol. 257 (Issues 1–2). – P. 121–125. 5. Gura Alexander Y., Gücüyenerb Demet, Üzünerb Nevzat et al. Korczyna and Natan M. Bornsteina Cerebral vasomotor reactivity of patients with acute ischemic stroke: Cortical versus subcorti- cal infarcts: An Israeli – Turkish collaborative study // J. of the Neurolog. Sciences. – 2007. – Vol. 257 (Issues 1–2). – P. 121–125. 5. Gura Alexander Y., Gücüyenerb Demet, Üzünerb Nevzat et al. Korczyna and Natan M. Е. М. Козар-Гурина (Львов) Е. М. Козар-Гурина (Львов) У 100 больных с цервикогенной вертебробазилярной недостаточностью (ВБН) мозгового кровообращения проводили клинико-неврологическое обследование и транскраниальную доп- плерографию сосудов вертебробазилярного бассейна. Обнаружены закономерности, указываю- щие на взаимосвязь между рефлекторными проявлениями ВБН и показателями церебрально- го кровотока. Патологический рефлекс Россолимо – Вендеровича у больных свидетельствует о ранней стадии недостаточности мозгового кровообращения в вертебробазилярном бассейне 81 О. М. Козар-Гуріна. Взаємозв’язок показників церебрального кровотоку при ВБН и сохранность ауторегуляции церебрального кровообращения, симптома Хвостека III у больных с ВБН – на углубление ВБН и срыв ауторегуляции церебрального кровообращения. и сохранность ауторегуляции церебрального кровообращения, симптома Хвостека III у больны с ВБН – на углубление ВБН и срыв ауторегуляции церебрального кровообращения. Ключевые слова: цервикогенная вертебробазилярная недостаточность мозгового крово- обращения, рефлекторные проявления, показатели церебрального кровотока. RELATIONSHIP OF INDICES OF CEREBRAL BLOOD FLOW AND REFLEX MANIFESTATIONS IN PATIENTS WITH CERVICOGENIC VERTEBROBASILAR INSUFFICIENCY О. М. Kozar-Gurina (Lviv, Ukraine) Department of neurology and neurosurgery, Lviv National Medical University Department of neurology and neurosurgery, Lviv National Medical University In 100 patients with cervicogenic vertebral basilar insufficiency of cerebral circulation was performed clinical and neurological examination, and transcranial Doppler arteries of vertebral basi- lar area. Found patterns that suggest the existence of links between reflex manifestations of vertebral basilar insufficiency and of cerebral blood flow. The presence of patients with abnormal reflex Ros- solimo – Venderovich may indicate an early stage of cerebrovascular insufficiency in the vertebral basilar area and the preservation of autoregulation of cerebral blood flow. The presence of patients with VBI Chvostek III contrary may indicate a deepening of vertebral basilar insufficiency and failure of autoregulation of cerebral blood flow. Key words: cervicogenic vertebrоbasilar insufficiency, reflex manifestations, indicators of ce- rebral circulation.
https://openalex.org/W2400712671	http://doisrpska.nub.rs/index.php/biomedicinskaistrazivanja/article/download/2003/1929	Croatian	null	Primjena cistatina C u određivanju jačine glomerularne filtracije kod oboljelih od tipa 2 dijabetes melitusa	Biomedicinska istraživanja	2,016	cc-by	3,690	BIOMEDICINSKA ISTRAŽIVANJA 2015;6(2):99-105 BIOMEDICINSKA ISTRAŽIVANJA 2015;6(2):99-105 UDK: 616.379-008.64:577.112 DOI: 10.7251/BII1502099R Originalni naučni rad Originalni naučni rad Adresa autora: Dr Ivona Risović, mr sc. Maksima Gorkog 15 78 000 Banja Luka ivona.risovic@gmail.com Primjena cistatina C u određivanju jačine glomerularne filtracije kod oboljelih od tipa 2 dijabetes melitusa Ivona Risović1, Snježana Popović-Pejičić2, Vlastimir Vlatković2 1Internacionalni dijaliza centar, Laktaši, Republika Srpska, Bosna i Hercegovina 2Univerzitetska bolnica-Klinički centar, Banja Luka, Republika Srpska, Bosna i Hercegovina Kratak sadržaj Uvod. Funkcija bubrega se u praksi najčešće procjenjuje izračunavanjem jačine glomerularne filtracije (JGF) na osnovu koncentracije kreatinina u serumu. Primjena relativno novijeg parametra, cistatina C pokazala se pouzdanom i osjetljivom u procjeni bubrežne funkcije. Cilj rada je bio da se kod oboljelih od dijabetes melitusa (DM) tipa 2 odredi JGF pomoću 24-časovnog klirensa kreatinina, primjenom jednačina iz serumskog kreatinina i cistatina C, te da se utvrdi stepen senzitivnosti i specifičnosti primjenjenih metoda. Metode. Šezdeset ispitanika, dobi od 40-70 godina starosti, podijeljeno je u dvije grupe od po 30 ispitanika: grupa oboljelih od DM tipa 2 i kontrolna grupa. Za određivanje JGF korišćen je 24-časovni klirens kreatinina, jednačine prema Modification of Diet in Renal Disease (MDRD) za kreatinin i Chronic Kidney Disease Epidemiology (CKD-EPI) za cistatin C. Rezultati. Prosječne vrijednosti serumskog cistatina C su bile statistički značajno više kod oboljelih od DM u poređenju sa kontrolnom grupom (p <0,001). Utvrđena je statistički visoko značajna pozitivna korelacija između JGF izračunate pomoću jednačina koje koriste koncentracije kreatinina i cistatina C u serumu (r =0,74, p< 0,001). Jednačina po CKD-EPI pokazala je veću senzitivnost i specifičnost kod svih ispitanika. Ova senzitivnost je bila izraženija kod oboljelih od DM i iznosila je za CKD-EPI 0,938, za MDRD 0,878, za 24-časovni klirens kreatinina 0,818, a specifičnost za CKD-EPI 0,878, za MDRD 0,867, a za 24-časovni klirens kreatinina 0,802. Zaključak. Koncentracija cistatina C u serumu se pokazala kao pouzdan, senzitivan i specifičan parametar za izračunavanje JGF, posebno osjetljiv kod osoba sa blagim smanjenjem funkcije bubrega. Ključne riječi: dijabetes melitus, cistatin C, kreatinin, jačina glomerularne filtracije 99 Biomedicinska istraživanja 2015;6(2):99-105 bubrežne insuficijencije [9,10]. Uvod i j j [ , ] Cilj rada je bio da se kod oboljelih od DM tipa 2 odredi JGF pomoću 24-časovnog klirensa kreatinina, primjenom jednačina iz serumske koncentracije cistatina C i kreatinina, te da se utvrdi stepen senzitivnosti i specifičnosti JGF određene tim metodama. Jačina glomerularne filtracije (JGF) se definiše kao volumen plazme koju glomeruli filtriraju u jedinici vremena [1,2]. Određuje se metodom klirensa supstanci. Klirens neke supstance predstavlja određeni volumen krvi koji se u jed- inici vremena očisti od te supstance. Promjene u JGF mogu biti prvi znak oštećenja bubrežne funkcije [2]. Prema vodiču Američke Nacion- alne fondacije za bubreg predložena je podjela hronične bubrežne bolesti u stadijume u zavis- nosti od JGF [3]. Idealan marker za procjenu JGF je supstanca koja se stvara u organizmu u konstantnoj količini, filtrira se u glomeruli- ma, ne sekretuje se u tubulima i ne vraća se u cirkulaciju [2,4]. Metode zlatnog standarda za određivanje JGF poput klirensa inulina ili primjene radioizotopa se rijetko primjenjuju u kliničkoj praksi [1,2]. U svakodnevnom radu JGF se određuje iz nivoa kreatinina u serumu i urinu metodom klirensa endogenog kreati- nina ili primjenom jednačina izvedenih iz serumske koncentracije kreatinina [2,3]. Kod oboljelih od tipa 2 dijabetes melitusa (DM) treba odmah po postavljanju dijagnoze bolesti da se izračuna JGF na osnovu koncentracije kreatinina u serumu, a potom najmanje jednom godišnje, nezavisno od stepena albuminurije. Za izračunavanje JGF preporučuje se primjena jednačina po Modification of Diet in Renal Disease (MDRD) ili Chronic Kidney Disease Epidemiology (CKD-EPI) [5]. Metode rada Studija je sprovedena na Odjeljenju en- dokrinologije Univerzitetske bolnice- Kliničkog centra u Banjoj Luci, u periodu od jula 2011. godine do februara 2012. godine. Studijom je obuhvaćeno 60 ispitanika u dobi od 40-70 godina starosti, koji su bili podijeljeni u dvije grupe od po 30 ispitanika: prva grupa su bili oboljeli od DM tip 2 i druga kontrolna grupa, zdravi ispitanici. Ispitivanje je sprovedeno uz poštovanje Helsinške deklaracije o medicin- skim istraživanjima i principima dobre naučne prakse. Saglasnost za protokol istraživanja odobrio je Etički komitet Univerzitetske bol- nice-Kliničkog centra u Banjoj Luci. Prosječno trajanje DM bilo je 10,16±8,74 godina. Zbog mogućeg uticaja na vrijednosti cistatina C u serumu iz ispitivanja su isključeni bolesni- ci sa akutnim dijabetičkim komplikacijama, poremećajem funkcije štitaste žlijezde (hipotir- eoidizam i hipertireoidizam), malnutricijom, srčanom dekompenzacijom, znacima zapal- jenskog procesa, te oni koji su u terapiji imali kortikosteroide [7,11]. ( ) [ ] Na vrijednosti kreatinina utiče dob, pol, rasa, ishrana i konstitucija. Zato se kreatinin ne smatra najpreciznijim parametrom za proc- jenu bubrežne funkcije [4,6]. Niskomolekularni proteini, poput cistatina C su pokazali veću pouzdanost u procjeni bubrežne funkcije, u odnosu na dosadašnje metode koje se baziraju na kreatininu. Cistatin C se filtrira u glomer- ulima, reapsorbuje u tubulima i ne vraća se u cirkulaciju [6,7]. Slično kao i za serumski kreatinin, postoji više jednačina za određivanje JGF iz serumske koncentracije cistatina C [8]. Ovaj parametar je pokazao dobru korelaciju sa metodama zlatnog standarda za određivanje JGF [1,4]. Cistatin C se do sada pokazao i kao osjetljiv parametar za procjenu početnog oštećenja bubrežne funkcije, što kod oboljelih od DM omogućava prevenciju i preduzimanje mjera u usporavanju ili zaustavljanju progresije hronične bubrežne bolesti i razvoju terminalne Sistolni i dijastolni krvni pritisak mjereni su aneroidnim sfingomanometrom firme Welch Allyn prije fizikalnog pregleda. Hipertenzija je kod oboljelih od DM tipa 2 definisana kao vrijednosti krvnog pritiska >140/80 mmHg, prema smjernicama Američke Dijabetološke Asocijacije iz 2014. godine ili ukoliko su ispi- tanici koristili antihipertenzivne lijekove [5]. Izračunavanje indeksa tjelesne mase (Body Mass Index, BMI) za procjenu i praćenje stanja uhranjenosti vršeno je prema Quetelet-ovoj formuli: BMI = tjelesna težina u kg/(tjelesna visina u m)2. Biohemijske analize krvi i urina rađene su u Zavodu za laboratorijsku dijagnostiku Univerzitetske bolnice-Kliničkog centra u Ban- joj Luci. Uzorci krvi i urina za laboratorijsku obradu uzimani su ujutru. Metode rada Uzimanjem anam- nestičkih podataka i uvidom u medicinsku 100 Procjena JGF cistatinom C u tipu 2 dijabetes melitusa dokumentaciju utvrđeni su dužina trajanja DM, prisustvo komorbiditeta i terapija koju su bolesnici koristili. Cistatin C u serumu je određivan imunoturbidimetrijskom metodom, primjenom Roche komercijalnih testova, na aparatu Cobas 6000, firme Roche. Referentne vrijednosti, prema upustvu proizvođača su 0,47 - 1,09 mg/l. Na aparatu AU 680, firme Olympus određivani su: kreatinin u serumu puferiranom kinetičkom Jaffe-ovom reakcijom bez depro- teinizacije; glikemija enzimatskom metodom sa heksokinazom i glikolizirani hemoglobin (HbA1c%) testom inhibicije lateks aglutinacije. Na osnovu koncentracije kreatinina u urinu i serumu i minutnog volumena urina izračunat je 24-časovni klirens kreatinina. Dobijena vri- jednost je korigovana prema DuBois formuli za površinu tijela TP (m²) = 0,20247 x tjelesna visina (m)0.725 x tjelesna težina (kg)0.425, uz pomoć elektronskog kalkulatora [12]. Uz pomoć elek- tronskog kalkulatora izračunata je JGF prema MDRD i CKD-EPI jednačini [13]. JGF(MDRD) (ml/min/1,73m2) = 186 x kreatinin−1.154(µmol/l) × godine−0.203 (x 0,742 ukoliko je osoba ženskog pola). JGF (CKD-EPI) (ml/min/1,73m2) = 76,7 × Cistatin C−1,19 [14]. i neparametarski Mann-Whitney test (ako je razlika u varijansama posmatranih obilježja statistički značajna). Za utvrđivanje stepena povezanosti (korelacije) posmatranih obi lježja korišćena je Pearson-ova parametarska korelacija. Stepen senzitivnosti i specifičnosti određivan je primjenom binarne logističke re- gresije, nakon koje su uporedno predstavljene ROC krivom. Prag značajnosti bila je vjero- vatnoća 0,05. Za statističku analizu, tabelarne i grafičke prikaze rezultata korišćeni su MS Office Word 2007 i MS Office Excel 2007, SPSS 16.0 for Windows. Rezultati U tabeli 1 prikazani su demografski, klinički i laboratorijski podaci o ispitanicima. Prosječne vrijednosti glikemije natašte, HbA1c i dijastol- nog krvnog pritiska u grupi oboljelih od DM su bile iznad preporučenih. Za regulaciju glikemi- je 15 bolesnika koristilo je insulin, 12 bolesnika oralne antihiperglikemike, a 3 bolesnika su imala kombinaciju oralne i insulinske terapije. Prosječne vrijednosti serumskog cistatina C su bile statistički značajno više kod oboljelih od DM u poređenju sa kontrolnom grupom (p <0,001), za razliku od prosječnih vrijednosti serumskog kreatinina i 24-časovnog klirensa kreatinina koje su bile u okviru referentnih vrijednosti. Prosječne vrijednosti JGF su bile smanjene kod oboljelih od DM. U grupi obolje- Za obradu i analizu podataka u radu prim- ijenjene su metode deskriptivne statistike. Za upoređivanje razlika između grupa korišćen je Χ 2 test kontigencije, za upoređivanje sred- njih vrednosti obilježja korišćen je Student-ov t-test za nezavisne uzorke (ako razlika varijansi posmatranih obilježja nije statistički značajna) Parametar Oboljeli od DM tipa 2 Kontrolna grupa Statistička značajnost Pol (žene / muškarci) 15/15 14/16 nz Dob, godine 60,1 ± 8.5 58,03 ± 6.23 nz TA sistolni, mmHg 132,16 ± 15.40 132,17 ±15.41 nz TA dijastolni, mmHg 81,17 ± 7.15 78,0 ± 6,36 nz BMI, kg /m2 28,53±4,24 23,36 ± 8,51 p <0,05 Glukoza, mmol/l 11,54 ± 4,31 5,26 ± 0,53 p<0,05 HbA1c, % 10,24 ± 2,47 – Kreatinin, µmol/l 91,10 ± 59,09 83,6 ± 9,71 nz Cistatin C, mg/l 1,21 ± 0,49 0,71 ± 0,77 p <0,001 24-časovni klirens kreatinina, ml/min/ 1,73m2 97,25 ± 23,16 93,86 ± 37,15 nz JGF po MDRD, ml/min/ 1.73m2 80,2 ± 23,86 95,96 ± 5,14 p <0,05 JGF po CKD-EPI, ml/min/ 1.73m2 83,56± 30,12 98,53 ± 15,04 p < 0,05 Tabela 1. Demografski, klinički i laboratorijski podaci o ispitanicima Rezultati prikazani kao srednja vrijednost± standardna devijacija (X± SD); nz – nije značajno; BMI- indeks tjelesne mase; TA- krvni pritisak; HbA1c-hemoglobin A1c; JGF- jačina glomerularne filtracije; MDRD - Modification of Diet in Renal Disease; CKD-EPI- Chronic Kidney Disease Epidemiology Tabela 1. Demografski, klinički i laboratorijski podaci o ispitanicima Rezultati prikazani kao srednja vrijednost± standardna devijacija (X± SD); nz – nije značajno; BMI- indeks tjelesne mase; TA- krvni pritisak; HbA1c-hemoglobin A1c; JGF- jačina glomerularne filtracije; MDRD - Modification of Diet in Renal Disease; CKD-EPI- Chronic Kidney Disease Epidemiology 101 Biomedicinska istraživanja 2015;6(2):99-105 lih od DM, kod 4 ispitanika, prema CKD-EPI, i 3 ispitanika prema MDRD jednačini, uočene su vrijednosti JGF<60 ml/min/1,73m2. Rezultati Utvrđena je statistički visoko značajna pozitivna korelaci- ja između JGF izračunate prema jednačinama iz serumske koncentracije kreatinina i cistatina C (r =0,74, p< 0,001). Cistatin C nije pokazao statistički značajnu korelaciju sa dobi, polom i BMI kod oboljelih od DM tipa 2. U ovoj grupi ispitanika vrijednosti serumskog cistatina C pokazale su statistički značajnu pozitivnu korelaciju sa sistolnim krvnim pritiskom (r= 0,46, p<0,05), trajanjem DM (r= 0,31, p<0,05) i vrijednostima HbA1c (r= 0,38, p<0,05). Veća pouzdanost cistatina C u procjeni JGF poka- zana je primjenom binarne logističke regresije na osnovu traženih parametara, te analizom ROC. krive. Dobijeni rezultati su pokazali da je CKD-EPI jednačina bila osjetljivija i specifičnija za procjenu JGF. Ovo je bilo naročito izraženo u grupi oboljelih od DM gdje je senzitivnost za CKD-EPI bila 0,938, za MDRD 0,878, a za klirens kreatinina 0,818. Specifičnost za CKD- EPI je bila 0,878, za MDRD 0,867, za klirens kreatinina 0,802. Ovo je prikazano grafički ROC krivom na kojoj se vidi da oblast koju pokriva CKD-EPI iznosi 0,973, MDRD 0,887, a 24-časovni klirens kreatinina 0,819 (Grafikon 1). Diskusija Pored cistatina C, određivan je serumski kreatinin, 24-časovni klirens kreatinina, proteinurija, te izračunavana JGF prema jednačinama za cis- tatin C i kreatinin. Period praćenja je bio pet godina. Retultati su pokazali da su cistatin C, ali i serumski kreatinin i proteinurija, dobri prediktori odbacivanja grafta [18]. i prosječne vrijednosti serumskog cistatina C, za razliku od prosječnih vrijednosti serumskog kreatinina i 24-časovnog klirensa kreatinina, koje su bile u okviru referentnih kod obje grupe ispitanika. Povišene vrijednosti cistatina C i normalne vrijednosti kreatinina u serumu, pokazuju da je cistatin C osjetljiv na promjene u “slijepom opsegu kreatinina“ [15,16]. Pozna- to je da se normalne vrijednosti kreatinina u serumu mogu održavati sve dok se bubrežna funkcija ne ošteti za polovinu od normalne. Pucci i saradnici [9] su u svom istraživanju pokazali prednost cistatina C, u odnosu na kreatinin, za procjenu početnog oštećenja bubrežne funkcije kod oboljelih od oba tipa DM. Cistatin C je u njihovom istraživanju bio povišen kada su vrijednosti JGF bile blizu ref- erentnih vrijednosti (90ml/min/1,73m2 ) ili u opsegu od 60 do 90 ml/min/1,73m2 . Stoga su zaključili, da cistatin C omogućava ranu detekciju, prevenciju i liječenje dijabetesne nefropatije. U našem istraživanju vrijednosti cistatina C su bile u značajnoj korelaciji sa glavnim faktorima rizika za oštećenje bubrežne funkcije u DM: vrijednostima sistolnog krvnog pritiska, HbA1c i dužinom trajanja DM. Ovu povezanost su utvrdili i drugi autori [9,15]. Cistatin C nije pokazao značajnu korelaciju sa dobi, polom i BMI kod oboljelih od DM, što mu daje prednost u odnosu na kreatinin. Bevc i saradnici [16] su u svom istraživanju pokazali da je cistatin C dobar pokazatelj JGF kod go- jaznih osoba i osoba sa DM tipa 2. Oni su zak- ljučili da on ne može u potpunosti da zamijeni metode „zlatnog standarda“ za izračunavanje JGF, ali može da pomogne u selekciji bolesnika za primjenu tih invazivnih metoda. Perkins i saradnici [17] su pratili vrijednosti JGF kod oboljelih od DM tokom četiri godine. Vrijed- nosti JGF izračunavane su pomoću jednačina iz serumskih vrijednosti cistatina C i kreatinin, te upoređivane sa klirensom jotalamata kao metodom „zlatnog standarda“ za određivanje JGF. Vrijednosti JGF iz serumskog cistatina C su imale bolju korelaciju sa jotalamatom i pratile su promjene JGF koje su se dešavale j g Određivanjem senzitivnosti i specifičnos- ti primjenjenih metoda u određivanju JGF utvrđeno je da je CKD-EPI jednačina imala najveću senzitivnost i specifičnost, koja je bila izraženija u grupi oboljelih od DM. Diskusija Određivanje JGF predstavlja jednu od najčešćih metoda za procjenu bubrežne funkcije, koja odražava stanje i funkciju nefrona [1-4]. S ob- zirom da DM predstavlja jedan od vodećih uzroka hronične bubrežne bolesti, preporuke su da se evaluacija bubrežne funkcije kod obol- jelih od tipa 2 DM uradi odmah po postavljanju dijagnoze bolesti [5]. Jedan od glavnih ciljeva procjene bubrežne funkcije kod oboljelih od DM jeste da se otkriju bolesnici sa oštećenjem bubrega u ranoj fazi, kako bi se preduzele mjere sa ciljem da se uspori dalja progresija bubrežnog oštećenja [6]. Ovo prvo ispitivanje treba da obuhvati određivanje nivoa albumi- na u urinu i vrijednosti kreatinina u serumu. Na osnovu vrijednosti serumskog kreatinina se izračunava JGF primjenom matematičkih jednačina. Kod oboljelih od DM preporučuje se primjena MDRD ili CKD-EPI jednačine za izračunavanje JGF [5]. Značaj određivanja JGF u ovoj populaciji je u tome što može da postoji smanjena JGF bez prisustva albuminurije [5]. U našem istraživanju JGF je određivana prim- jenom jednačina za procjenu JGF na osnovu koncentracije kreatinina i cistatina C. j Kod oboljelih od DM utvrđene su povišene Grafikon 1. ROC krive za jačinu glomerularne filtracije određenu pomoću CKD-EPI jednačine za cistatin (JGFCys), MDRD jednačine i 24-časovnog klirensa kreatinina (CCr). CKD-EPI jednačina je senzitivnija i specifičnija u određivanju JGF u odnosu na MDRD jednačinu i 24-časovni klirens kreatinina (CKD-EPI: ROC oblast: 0,973, standardna greška – SE: 0,076, p< 0,000; MDRD: ROC oblast: 0,887, SE: 0,067, p< 0,000; CCr: ROC oblast: 0,819, SE: 0,076, p< 0,003). CCr–24-časovni klirens kreatinina; JGFCys – JGF izračunata pomoću CKD-EPI jednačinini za cistatin C; MDRD – JGF izračunata pomoću MDRD jednačine k ROC k l l fil d đ CKD EPI d (JGFC 4-časovni klirens kreatinina; JGFCys – JGF izračunata pomoću CKD-EPI jednačinini za cistatin C; MDRD – JGF izračuna u MDRD jednačine Grafikon 1. ROC krive za jačinu glomerularne filtracije određenu pomoću CKD-EPI jednačine za cistatin (JGFCys), MDRD jednačine i 24-časovnog klirensa kreatinina (CCr). CKD-EPI jednačina je senzitivnija i specifičnija u određivanju JGF u odnosu na MDRD jednačinu i 24-časovni klirens kreatinina (CKD-EPI: ROC oblast: 0,973, standardna greška – SE: 0,076, p< 0,000; MDRD: ROC oblast: 0,887, SE: 0,067, p< 0,000; CCr: ROC oblast: 0,819, SE: 0,076, p< 0,003). p j 102 Procjena JGF cistatinom C u tipu 2 dijabetes melitusa tokom vremena [17]. Ležaić i saradnici [18] su istraživali da li cistatin C može da bude prediktor odbacivanja bubrežnog grafta. Diskusija Na prikazu ROC krive je oblast koju pokriva CKD-EPI iznosila 0,973, za MDRD 0,887, a za klirens kreatinina 0,819. El-Shafey i saradnici [19] su pokazali da je oblast koju pokriva ROC kriva bila najviša za jednačinu JGF iz cistatina C u odnosu na jednačinu iz kreatinina i 24-časovni kreatinin. Slične rezultate pokazao je MacIsaac sa saradnicima [20]. Oni su kod oboljelih od DM tipa 2 upoređivali jednačine JGF izračunate iz serumskog cistatina C i kreatinina, te dobi- jene vrijednosti JGF upoređivali sa referent- nom metodom, odnosno klirensom dietilen-tri- aminopentaacidna kiselina. Najveća oblast ROC krive bila je JGF iz cistatina C (97,9%). Ovi rezultati upućuju da je cistatin C osjetljiv i pouzdan parametar za izračunavanje JGF. Autori izjavljuju da nemaju sukob interesa. The authors declare no conflicts of interest. Literatura fel-Wagner B. Effect of corticosteroid therapy on serum cystatin C and β2-Microglobulin con- centrations. Clin Chem 2002;48:1123–6. 1. Woo KS, Choi JL, Kim BR, Kim JE, Han JY. Clinical usefulness of serum cystatin C as marker of renal function. Diabetes Metab J 2014;38(4):278–84. 12. Body Surface Area – Cornell University Avail- able from: http://www-users.med.cornell. edu/~spon/picu/calc/bsacalc.htm Accessed March 26, 2014 2. Hojs R, Bevc S, Ekart R, Gorenjak M, Puklavec M. Serum cystatin C as a endogenous marker of renal function in patients with mild to moder- ate impairment of kidney function. Neph Dial Transplant 2006;21:1855–62. 13. National Kidney Foundation. Calculator For Health Care Professionals. Available from: http://www.kidney.org/professionals/ KDOQI/gfr_calculator Accessed March 18, 2014 3. National Kidney Foundation. K/DOQI Clinical Practice Guidelines for Chronic Kidney Disease: Evaluation, Classification and Stratification. Am J Kidney Dis 2002;39(suppl 1):S1–S266. Avail- able from: https://www.kidney.org/sites/ default/files/docs/ckd_evaluation_classifica- tion_stratification.pdf Accessed June 15, 2015 14. National Kidney Foundation. Cystatin C. Avail- able from: https://www.kidney.org/atoz/ content/CystatinC Accessed March 18, 2014 15. Jeon YK, Kim MR, Huh JE, et al. Cystatin C as an early biomarker of nephropathy in pa- tients with type 2 diabetes. J Korean Med Sci 2011;26:258–63. 4. KDIGO 2012 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease. Kidney Int 2013;3(Suppl 1):S5–14. Available from: http://www.kdigo. org/clinical_practice_guidelines/pdf/CKD/ KDIGO_2012_CKD_GL.pdf Accessed June 15, 2015 16. Bevc S, Hojs R, Kart R, Završnik M, Gorenjak M, Puklavec L. Simple cystatin C formula for estimation of glomerular filtration rate in over- weight patients with diabetes mellitus type 2 and chronic kidney disease. Exp Diabetes Res 2012;2012:1–8. 5. American Diabetes Association. Standards of Medical Care in Diabetes-2014. Diabetes Care 2014;37(Suppl 1):S14–80. 17. Perkins BA, Nelson RG, Ostrander BE, et al. Detection of renal function decline in patients with diabetes and normal or elevated GFR by serial measurements of serum cystatin C con- centration: results of a 4-year follow-up study. J Am Soc Nephrol 2005;16:1404–12. 6. Walczak K, Rosniak-Bak K, Paradowski M, Moczulski D. Cystatin C as a marker of renal function in diabetic kidney disease. Diabet Dosw Klin 2009;9(2):69–72. 7. Brguljan PM, Cimerman N. Human Cystatin C. Turk J Biochem 2007;32(3):95–103. 18. Lezaic V, Dajak M, Radivojevic D, Ristic S, Marinkovic J. Cystatin C and serum creatinine as predictors of kidney graft outcome. Int Urol Nephrol 2014;46(7):1447–54. 8. Beauvieux M.C, Le Moigne F, Lasseur C, et al. New predictive equations improve monitoring of kidney function in patients with diabetes. Zaključak Određivanje JGF je obavezan dio procjene bubrežne funkcije kod oboljelih od DM. U ovom istraživanju cistatin C se pokazao kao osjetljiv, specifičan i pouzdan parametar u određivanju JGF. Zbog svoje relativno visoke cijene, cistatin C za sada ne spada u rutinske parametre za procjenu bubrežne funkcije nam- ijenjene širokoj upotrebi. 103 Biomedicinska istraživanja 2015;6(2):99-105 Literatura Diabetes Care 2007;30(8):1988-94. 19. El-Shafey M, El-Nagar GF, Selim MF, El-Sorogy HA, Sabry AA. Is serum cystatin C an accurate endogenous marker of glomerular filteration rate for detection of early renal impairment in patients with type 2 diabetes mellitus? Ren Fail 2009;31(5):355–9. 9. Pucci L, Triscornia S, Lucchesi D, et al. Cystatin C and estimates of renal function: searching for better measure of kidney function in diabetic patients. Clin Chem 2007;53:480–8. 20. MacIsaac RJ, Premaratne E, Jermus G. Esti- mating glomerular filtration rate in diabetes using serum cystatin C. Clin Biochem Rev 2011;32(2):61–7. 10. Fricker M, Wiesli P, Brandle M, Schwegler B, Schmid C. Impact of thyroid dysfunction on serum cystatin C. Kidney Int 2003;63:1944–7. 11. Bokenkamp A, Van Wil A E, Lentze M J, Stof- 104 Procjena JGF cistatinom C u tipu 2 dijabetes melitusa Ivona Risović1, Snježana Popović-Pejičić2, Vlastimir Vlatković2 Ivona Risović1, Snježana Popović-Pejičić2, Vlastimir Vlatković2 1 International Dialysis Center Laktaši, the Republic of Srpska, Bosnia and Herzegovina 2 University Hospital – Clinical Center , Banja Luka, the Republic of Srpska, Bosnia and Herzegovina 1 International Dialysis Center Laktaši, the Republic of Srpska, Bosnia and Herzegovina nternational Dialysis Center Laktaši, the Republic of Srpska, Bosnia and Herzegovina University Hospital – Clinical Center , Banja Luka, the Republic of Srpska, Bosnia and Herzego y p p g Hospital – Clinical Center , Banja Luka, the Republic of Srpska, Bosnia and Herzegovina Introduction. Kidney function is usually assessed by estimating the glomerular filtration rate (GFR) from serum creatinine. Application of a relatively newer parameter, cystatin C, has proven to be a reliable and sensitive method in estimating kidney function. The aim of this study is to estimate GFR in patients with type 2 diabetes mellitus (DM), using 24-hour creatinine clearance test and applying equations from serum creatinine and cystatin C, and to determine the sensitivity and specificity of these methods. Methods. Sixty respondents, aged 40-70 years, were divided into two groups of 30: the group of patients with type 2 DM and the control group. 24-hours creatinine clearance test was used to estimate GFR, according to the equation Modification of Diet in Renal Disease (MDRD) for creatinine and Chronic Kidney Disease Epidemiology (CKD-EPI) for cystatin C. Results. The mean values of cystatin C were significatly higher in patients with DM compared with the control group (p<0,001). Highly significant positive correlation was determined between GFR estimated by the equations of serum creatinine and cystatin C (r =0.74, p< 0.001). Equation by CKD-EPI has shown higher sensitivity and specificity in all patients. This sensitivity was more expressed in patients with type 2 DM and was for CKD-EPI 0.938 vs. 0.878 for MDRD vs. 0.818 for 24-hours creatinine clearance, and the specificity for CKD-EPI 0.878 vs. 0.867 for MDRD vs. 0.802 for 24-hour creatinine clearance. Conclusion. In this study, the application of serum cystatin C has proven to be a reliable, sensitive and specific parameter for GFR estimation, especially in patients with mild decrease in GFR. Keywords: diabetes mellitus, cystatin C, creatinine, glomerular filtration rate Prihvaćen – Accepted: 24/08/2015 Primljen – Received: 22/03/2015 105
https://openalex.org/W2335982431	https://www.jstage.jst.go.jp/article/hrl/9/4/9_61/_pdf	English	null	Parallel computing for high-resolution/large-scale flood simulation using the K supercomputer	Hydrological research letters	2,015	cc-by	6,613	Abstract: 2012. It was ranked first in a list of the Top 500 supercom- puter sites in June and November 2011 during its develop- ment stage. Since then, the computer has been considered one of the fastest computers in Japan from the perspective of LINear algebra PACKage (LINPACK). The CPU is SPARC64 VIIIfx (2.0 GHz) with 8 cores. This computer has a total of 82944 CPUs and thus, 663552 (= 82944 × 8) cores. This paper reports the implementation of high-performance computing using the K supercomputer in Kobe, Japan, for large-scale/high-resolution flood simulation. Supercomputer K was developed in 2012 by RIKEN and Fujitsu and ranked first in the list of Top 500 supercomputer sites in 2011 during its development stage. A two-dimensional inundation simu- lation model developed based on a shallow water equation using an existing numerical scheme was parallelized with the K supercomputer. Osaka and other cities along the Yodo River were chosen as application sites and the area discret- ized by 12824442 (= 3453 × 3714) nodes with a resolution of 10 m. The computational time for the five-hour flood sim- ulation was measured by changing the number of 8-core CPUs of the K supercomputer. As a result, computational time was decreased to 9.3 min by using 128 × 64 = 8192 8-core CPUs. The computational time was 1423.7 min for one 8-core CPU. Thus, the simulation speed increased by a factor of 153.2 with the use of the K supercomputer. ( ) This paper reports the application of the K supercomputer to flood inundation simulation of Osaka and other cities in its vicinity. The simulation area was approximately 34.5 km × 37.1 km, discretized at a resolution of 10 m, resulting in 12824442 (= 3453 × 3714) computational nodes. The dis- tributed rainfall-runoff/flood-inundation simulation model (DRR/FI, Kobayashi et al., 2011, 2012, 2014a, b; Kobayashi and Takara, 2012) in the C environment with a parallel com- puting capability was applied. The proposed parallelization method is basically universal, and can thus be applied to any standard parallel computer. However, script code was written in order to stage-in the K supercomputer, to run the model, and to stage-out with the simulation results. Moreover, hybrid parallelization was used, i.e. Abstract: 8 cores in each CPU were used exclusively for thread parallelization with Open Multi-Processing (OpenMP), while the CPUs of the K supercomputer were used for process parallelization (domain decomposition) with a Message Passing Interface (MPI). This strategy for core utilization differs from the pre- vious works outlined below. Hybrid parallelization is recom- mended in the use of the K supercomputer because of the memory use and data transfer reduction. In case the core is also used for process parallelization (domain decomposi- tion), the number of processes by MPI in each CPU increases; thus approaching limitations of both the memory and data transfer. Instructions related to the hybrid parallelization are included in the script code. KEYWORDS K supercomputer; large-scale/high-resolution; flood inundation simulation; Yodo River; MPI; OpenMP KEYWORDS K supercomputer; large-scale/high-resolution; flood inundation simulation; Yodo River; MPI; OpenMP Parallel computing for high-resolution/large-scale flood simulatio using the K supercomputer Kenichiro Kobayashi1, Dai Kitamura2, Kazuto Ando2 and Noriyuki Ohi3 1Research Center for Urban Safety and Security, Kobe University, Japan 2Japan Agency for Marine-Earth Science and Technology (JAMSTEC), Japan 3Fujitsu Systems East Ltd., Japan Correspondence to: Kenichiro Kobayashi, Research Center for Urban Safety and Security, Kobe University, 1-1 Rokkodai-machi, Nada-ku, Kobe 657-8501, Japan. E-mail: kkobayashi@phoenix.kobe-u.ac.jp Hydrological Research Letters 9(4), 61–68 (2015) Published online in J-STAGE (www.jstage.jst.go.jp/browse/hrl). DOI: 10.3178/hrl.9.61 Hydrological Research Letters 9(4), 61–68 (2015) Published online in J-STAGE (www.jstage.jst.go.jp/browse/hrl). DOI: 10.3178/hrl.9.61 ©2015, Japan Society of Hydrology and Water Resources. KEYWORDS K supercomputer; large-scale/high-resolution; flood inundation simulation; Yodo River; MPI; OpenMP Received 29 June, 2015 Accepted 2 October, 2015 Received 29 June, 2015 Accepted 2 October, 2015 ©2015, Japan Society of Hydrology and Water Resources. INTRODUCTION Parallelization is a method to enhance the computational speed of flood modelling. In fact, there are several approaches to shorten the CPU time required for flood modelling, such as changing the algorithm of the numerical model, enhanc- ing the capability of the computer hardware, and/or parallel computing. A recent trend in computer hardware develop- ment aims at increasing the number of cores in a CPU rather than the clock frequency of a CPU. Therefore, at present, there are more advantages in the development of parallel computing methodology with respect to the reduction of computational time. This study is one such attempt, i.e. a parallel computing application on a particularly large-scale/ high-resolution flood simulation using the K supercomputer in Kobe, Japan. The K supercomputer is a next-generation high-performance computer developed by the Institute of Physical and Chemical Research (RIKEN) and Fujitsu as a part of the High-Performance Computing Infrastructure (HPCI) Initiative led by the Japanese Ministry of Education, Culture, Sports, Science and Technology (MEXT). The shared use of the K supercomputer began in November Several previous studies have investigated the application of parallel computing to flood-inundation modelling. In gen- eral, some inundation models have attempted to solve the full two-dimensional (2D) shallow water equations (Hunter et al., 2008), while others have used diffusive wave simplifi- cation by neglecting the inertial terms of the momentum equation (Leandro et al., 2014; Yu, 2010) or applied the so-called storage cell model based on Manning’s equation for the flux calculation between cells (Bates and De Roo, 2000). ) Neal et al. (2010) presented the application of different (i.e. OpenMP, MPI, and accelerator card) parallelization techniques on the 2D inundation model LISMIN developed Received 29 June, 2015 Accepted 2 October, 2015 —61— K. KOBAYASHI ET AL. based on LISFLOOD-FP by Bates and De Roo (2000). LISMIN applies an adaptive time step. The LISMIN-MPI code was run on 1 to 100 Intel E5462 cores (2.8 GHz) with a computational domain of 200 × 384 cells (at a resolution of 2 m). As a result, the code realized an efficiency (Speed-up = Serial run time/Parallel run time, Efficiency = Speed-up/ Number of cores) of 70% by using 100 cores (2 rows per core) when the domain was fully wet. K SUPERCOMPUTER The K supercomputer was developed jointly by RIKEN and Fujitsu, and its shared use began in 2012. K in Japanese means 10 quadrillion. The computer can execute 10 trillion calculations per second (i.e. 10 PFLOPS) with LINPACK. This supercomputer was ranked first in a list of the Top 500 supercomputer sites in June and November 2011. The CPU of the K supercomputer is SPARC64 VIIIfx (2.0 GHz) with 8 cores. Its theoretical performance is 128 GFLOPS with 8 cores. The computer has a total of 82944 CPUs and thus, 663552 cores. The main memory is a DDR3-SDRAM-dual inline memory module (DIMM), and its main storage capac- ity is 16 GB per node. The theoretical bandwidth between the memory is 64 GB/s. Torus fusion (Tofu) is developed for interconnecting the 82944 nodes of the computer. The net- work topology is a six-dimensional (6D) mesh/torus. The theoretical link throughput is bidirectionally 5 GB/s. The specifications of the K-supercomputer are listed in Table II. , p y However, because of the simulation stability criterion, some previous studies reported that diffusive wave models are computationally less effective than dynamic wave mod- els with high-resolution meshes (Neal et al., 2012). Neal et al. (2012) reported that the time step in their diffusion wave model was proportional to Δx2; whereas, the inertial/full shallow water model was proportional to Δx (Note that Δx is the spatial grid size). Moreover, the time step in their diffu- sive model was further reduced (i.e., closer to zero) when the water surface slope was small—i.e. it approached zero because the time step was related to the surface slope. Thus, a linearization of the time-step equation was implemented at a low slope. For this reason, the diffusion wave model becomes less computationally efficient when the cell size is small. Bates et al. (2010) reported that the computational cost of the original LISFLOOD-FP was higher than that of the full 2D shallow water equation for relatively fine-resolution (1–10 m) grids; therefore, it was improved by introducing the inertial terms in the governing equation. Bates et al. (2010) concluded that the new inertial code would most likely be faster than either diffusive or full shallow water models at any given spatial resolution, although they also commented that at a very low surface friction a full shallow water model may give more accurate results.l INTRODUCTION However, the results deteriorated in the case of a point source in Neal et al.’s (2010) simulation where the computational domain was not large compared with the simulated inundated area from the point source. The fully wet domain refers to the entire com- putational domain when it is initially filled with water to a depth of 0.1 m. The point source case refers to the case when a hydrograph is provided for a point in the computational domain that was initially dry. Although the following exam- ple below was a point source case, the inundation was lim- ited to an area within the larger computational domain; thus, the computational speed as a point source case in the present paper could be faster than the speed under a fully wet case in the same computational domain. If this is true, the result is opposite to Neal et al. (2010). ferent computational settings, it is not possible to describe the relative merits of the studies accurately. Nevertheless, the points for consideration basically consist of (1) the type of the governing equation of the model, (2) the computa- tional cost at a fine resolution, (3) whether the model applies an adaptive time step, (4) the effectiveness when the number of nodes is significantly increased, and (5) whether the model is run on a workstation or a supercomputer. Table I shows a comparison of these points in the literature. With respect to (1), a full shallow water equation was used in this study. The models in the literature used either diffusion wave simplification or a cell storage model. With respect to (2), we set the model resolution to 10 m, which is not a superfine resolution, such as that of 2 m argued for in Neal et al. (2010). Nevertheless, the computational model domain is considerably larger, at approximately 34.5 km × 37.1 km, as compared to that considered by Bates and De Loo (2000), who targeted a 35-km stretch of the Meuse River in the Netherlands. We do not address the necessity of a super- fine-resolution model in practice. With respect to (3), the model proposed in this paper does not handle an adaptive time step, considering the coupling with a dynamic wave river model and a slot model for the sewage network. Table I also identifies the studies in which an adaptive time step was used. INTRODUCTION With respect to (4), the number of nodes was increased considerably to 12824442 (= 3453 × 3714) in the present study because, with respect to (5), a supercomputer was used. The number of nodes and the use of a supercom- puter in previous studies are also summarized in Table I. pp ( ) Leandro et al. (2014) presented the parallelized diffusive model (called P-DWave). The diffusive wave simplification neglected the inertial terms of the momentum equation and thus, generally led to faster CPU time. P-DWave can handle an adaptive time step. The model was developed in both Matlab and Fortran environments. Further, a speed-up of 1.7–5.2 times and 1.1–1.7 times for Matlab and Fortran, respectively, was realized depending on the domain size and the number of processors. The authors used a mesh consist- ing of a maximum of 900 × 900 cells (at a resolution of 178.8 m) by using an AMD OpteronTM processor 6276 with 12 cores (2.3 GHz). When using 12 cores, their model resulted in a speed-up of 5.1 and 1.5 times for Matlab and Fortran, respectively. FLOOD SIMULATION WITH K SUPERCOMPUTER FLOOD SIMULATION WITH K SUPERCOMPUTER Table I. Simulation conditions in the past literature Researcher Neal et al. (2010) Leandro et al. (2014) Bates et al. (2010) Bates et al. (2000) Yu et al. (2010) This study CPU Quad core Intel Harpertown E5462 processor (2.8 GHz) at the University of Bristol AMD OpteronTM Processor 6276 (2.3 GHz) at the RUHR University of Bochum (Not parallel) Pentium II PC (Not parallel) 20 computational nodes with 4 × 2 cores each (i.e. 160 processors; 1.6 GHz) at Loughborough University CPU: SPARC64 VIIIfx (2.0 GHz) of the K supercomputer Number of processors 100 cores are used 12 cores are used 64 cores are used 8-core 8192 CPUs are used Target area Urban area in Glasgow, UK Benchmark example Urban area in Glasgow, UK 35-km stretch of the River Meuse, NL River Wharfe, and Ouse in the UK Osaka area, Japan Number of cells 200 × 384 900 × 900 (max case) 500 × 200 108464 1000000 12824442 (= 3453 × 3714) Resolution 2 m resolution 178.8 m resolution 2 m resolution 25 m resolution 8 m resolution 10 m resolution Simulation duration Max 7200 s 1 h 120 min Until no water depth variation Max 300 h 5 h Type of time step Adaptive time step Fixed time step Adaptive time step Fixed time step Fixed time step Fixed time step Table I. Simulation conditions in the past literature Table II. Specifications of the K supercomputer (Fujitsu Journal, 2012) Hardware K supercomputer Node SPARC64VIIIx (2.0 GHz), 8 cores/CPU/node Theoretical performance: 16 GFlops/core Cache L1 command: 32 KB/core L1 data: 32 KB/core L2 (shared): 6 MB Memory Main storage capacity: 16 GB Theoretical memory bandwidth: 64 GB/s Interconnect Communication mechanism: Tofu Communication topology: 6D mesh/torus Theoretical communication bandwidth: bidirectionally 5 GB/s Software OS: Exclusive use OS (Linux base) Compiler Fujitsu C/C++ Compiler; K-1.2.0-14 version Table II. Specifications of the K supercomputer (Fujitsu Journal, 2012) Hardware K supercomputer Table II. Specifications of the K supercomputer (Fujitsu Journal, 2012) an explicit numerical scheme called the leapfrog method was used for temporal discretization (e.g. Inoue et al., 2000, Japan Society of Civil Engineers, 2001). TWO-DIMENSIONAL INUNDATION SIMULATION A 2D inundation simulation module of the DRR/FI model was used to exclusively study the effect of parallel comput- ing on a 2D inundation modelling using the K supercom- puter. The model was developed in the C environment. The following 2D shallow water equation was used with a struc- tured grid: Since the flood models in the papers were developed by several different researchers in different countries with dif- —62— FLOOD SIMULATION WITH K SUPERCOMPUTER 2 * * * * , 1/2 , 1/2 1/2, 1/2 1/2, 1/2 ( ) ( ) 2 n n i j i j i j i j M M u M u M t x             * * * * , 1 , ( ) ( ) i j i j v M v M y    1 1 1 1 1/2, 1/2 1/2, 1/2 1/2, 1/2 1/2, 1/2 ( )( ) 2 n n n n i j i j i j i j h h H H g x                   2 2 2 , 1/2 , 1/2 , 1/2 , 1/2 2 , 1/2 4/3 1 1 1/2, 1/2 1/2, 1/2 ( ) ( ) ( ) 2 ( ) / 2 n n n n i j i j i j i j i j n n i j i j M M u v gn h h                 (5) * 1, 1/2 * 1/2, 1/2 * * 3/2, 1/2 1/2, 1/2 1 2 0.5( ) i j i j i j i j M u h h              * , 1/2 * * 1/2, 1/2 1/2, 1/2 0.5( ) i j i j i j M h h           (6) * 1/2, * , * * 1/2, 1/2 1/2, 1/2 1 2 0.5( ) i j i j i j i j N v h h           * 1/2, * * 1/2, 1/2 1/2, 1/2 0.5( ) i j i j i j N h h           (7) * * * 1/2, 1/2 , 1/2 * * 1/2, 1/2 * * * 1/2, 1/2 1, 1/2 ( 0) ( ) ( 0) i j i j i j i j i j u M if u u M u M if u               (8) * * * , , 1/2 * * , * * * , , 1/2 ( 0) ( ) ( 0) i j i j i j i j i j v M if v v M v M if v        (9) * 2 * 1 1 ( ) / 2, ( ) / 2 n n n n M M M h h h        (10) In the simulation, no flux boundary conditions were imposed on the surrounding boundaries. FLOOD SIMULATION WITH K SUPERCOMPUTER Mass conservation in Equation (1) was discretized as follows: in Cell h M N q t x y A           (1) 2 2 2 1 3 M uM vM H u v gh gn u t x y x h              (2) 2 2 2 1 3 N uN vN H u v gh gn v t x y y h              (3) 3 1 2 2 1/2, 1/2 1/2, 1/2 1, 1/2 , 1/2 2 n n n n i j i j i j i j h h M M t x                 1 2 2 , 1/2, 1/2 1/2, 1 1/2, n n n in i j i j i j Cell q N N y A              (4) (3) (4) where h denotes the inland water depth; ACell (10 m × 10 m) represents the 2D grid cell area; M = uh and N = vh indicate the discharge fluxes in the x- and y-directions, respectively; u and v denote the velocities in the x- and y-directions, respectively; and H represents the water depth + land eleva- tion. A staggered grid was used for spatial discretization, and where i, j indicates the location of a cell in the x- and y- directions, respectively, and n denotes the time step. The momentum equation in the x-direction was discretized as follows: —63— K. KOBAYASHI ET AL. FLOOD SIMULATION WITH K SUPERCOMPUTER The land elevation of the entire simulation area is shown in Figure 1. The Yodo River indicated in blue in Figure 1 was eliminated from the simula- tion area. The red cross in Figure 1 is the location where the dike break occurs. A simulation result with qin = 1000 m3 s–1 from the cross after five hours is shown in Figure 2 for refer- ence. A one-dimensional dynamic wave river model was already parallelized. However, this method and results are not described here as the exclusive purpose of this paper is to discuss the effect of parallelizing a 2D inundation model by using the K supercomputer. In practice, it is difficult to predict the exact location of the dike break, and this is a hur- dle for real-time prediction unless the break location is known immediately after the dike break occurs. (5) FLOOD SIMULATION WITH K SUPERCOMPUTER FLOOD SIMULATION WITH K SUPERCOMPUTER Figure 4. Schematic representation of the data transfer using sleeves (grey and orange). (i), (ii), (iii), and (iv) correspond to the procedure described in the main text of the paper Figure 2. Simulated inundation depth with qin = 1000 m3 s–1 from the cross after five hours Figure 4. Schematic representation of the data transfer using sleeves (grey and orange). (i), (ii), (iii), and (iv) correspond to the procedure described in the main text of the paper Figure 2. Simulated inundation depth with qin = 1000 m3 s–1 from the cross after five hours In addition, the file output procedure was parallelized since the data conversion to text incurs a high cost in terms of computational speed. Previously, all simulation results were gathered in the master process, and then, only the mas- ter process was carried out for data conversion to text. Consequently, the computational burden was concentrated on the master process, while the other processes waited for the task to be completed; i.e. the master process was the sim- ulation bottleneck. To solve this problem, the code was mod- ified in such a way that each process carried out the text conversion, and then, the converted text was sent to the mas- ter process. As the file size is large when the number of the computational nodes is 12824442, this parallelization reduces the computational speed. Figure 3. Schematic representation of the domain decompo- sition. The thin line represents the original domain. The orig- inal domain is decomposed by the larger domain denoted with bold lines. The orange and grey colours indicate the sleeves for the data transfer Moreover, thread parallelization was implemented through- out the code when possible using OpenMP. PARALLELIZATION METHOD (6) The computational domain shown in Figure 1 was decom- posed for parallel computing as shown in Figure 3. In this figure the thin line denotes the original computational domain composed of 18 × 18 small cells. The original domain was decomposed along the bold lines, as shown in the figure, into domains of 3 × 3 cells as an example. The light blue colour shows a larger domain including the so-called sleeves, i.e. two adjacent peripheral nodes (orange and grey). The data transfer between the domains was car- ried out using these sleeves. The data transfer itself was real- ized using the MPI functions MPI_Isend and MPI_Irecv four times. (7) (8) (9) First, (i) the data in the vertical orange sleeve of domain A were transferred to the vertical orange sleeve of east domain B, as shown in Figure 4. At the same time, (ii) the vertical (10) Figure 1. Land elevation of the entire simulation area Figure 1. Land elevation of the entire simulation area —64— RESULTS The results of the parallel computing application are shown in Figures 5 and 6 and Tables III–V. Figure 5 shows the elapsed time for the five-hour simulation shown in Figure 2 against the number of CPUs of the K supercomputer. Figure 6 shows the speed-up and efficiency of the simulation. As already described, Speed-up = Serial run time/Parallel run time, Efficiency = Speed-up/Number of cores. However, in this study, the number of cores was equivalent to the number of CPUs in the calculation, because the 8 cores in each CPU are used exclusively for thread parallelization. Thus, this number cannot necessarily be compared with that in the past research discussed above. Vertical decomposition indicates that the computational domain was decomposed only in the x-direction. Likewise, horizontal decomposition indicates that the computational domain was decomposed only in the y-direction. Average decomposition means that the domain was decomposed in both x- and y-directions. The effects of the decomposition of the domain are presented in Tables Figure 3. Schematic representation of the domain decompo- sition. The thin line represents the original domain. The orig- inal domain is decomposed by the larger domain denoted with bold lines. The orange and grey colours indicate the sleeves for the data transfer grey sleeve of domain A receives the data of the vertical grey sleeve of east domain B. The same procedures were carried out for the north–south directions, i.e. (iii) the horizontal grey sleeve of domain A received the data from the horizon- tal grey sleeve of south domain C. Then, (iv) the data of the horizontal orange sleeve of domain A were sent to the hori- zontal orange sleeve of south domain C. In the north–south data transfer, the data of the four corners were renewed, despite the fact that they were already renewed once in the east–west data transfer. —65— K. KOBAYASHI ET AL. Figure 6. Speed-up and efficiency for the five-hour simula- tion against the number of nodes (CPUs) of the K supercom- puter Figure 6. Speed-up and efficiency for the five-hour simula- tion against the number of nodes (CPUs) of the K supercom Figure 6. Speed-up and efficiency for the five-hour simula- tion against the number of nodes (CPUs) of the K supercom- puter Figure 6. Speed-up and efficiency for the five-hour simula- tion against the number of nodes (CPUs) of the K supercom- puter Figure 5. RESULTS Elapsed time for the five-hour simulation against the number of nodes (CPUs) of the K supercomputer Table III. Elapsed time, speed-up, and efficiency for the five-hour simulation against the number of nodes (CPUs) of the K supercomputer. The domain was decomposed only in the x-direction Total number of domains 1 8 16 32 48 64 128 256 512 1024 Number of domains in the x-direction 1 8 16 32 48 64 128 256 512 1024 Number of domains in the y-direction 1 1 1 1 1 1 1 1 1 1 Elapsed time [min] 1423.6 279.9 143.1 75.2 51.5 41.2 28.7 18.5 14.0 13.8 Speed-up 5.1 9.9 18.9 27.6 34.6 49.5 77.1 101.7 102.8 Efficiency [%] 63.6 62.2 59.2 57.6 54.0 38.7 30.1 19.9 10.0 d time, speed-up, and efficiency for the five-hour simulation against the number of nodes (CPUs) of the r. The domain was decomposed only in the x-direction Table III. Elapsed time, speed-up, and efficiency for the five-hour simulation against the number of node K supercomputer. The domain was decomposed only in the x-direction 66 Table IV. Same as in Table III but the domain was decomposed only in y-direction Total number of domains 1 8 16 32 48 64 128 256 512 1024 Number of domains in the x-direction 1 1 1 1 1 1 1 1 1 1 Number of domains in the y-direction 1 8 16 32 48 64 128 256 512 1024 Elapsed time [min] 1423.6 213.2 121.8 71.0 51.8 42.6 27.6 20.3 15.9 14.5 Speed-up 6.7 11.7 20.1 27.5 33.4 51.6 70.2 89.4 98.5 Efficiency [%] 83.5 73.0 62.7 57.2 52.2 40.3 27.4 17.5 9.6 Table V. RESULTS Same as in Table III but the domain was decomposed in both x- and y-directions Total number of domains 8 8 16 32 32 48 48 64 128 128 256 512 512 1024 2048 2048 4096 8192 8192 Number of domains in the x-direction 2 4 4 4 8 6 8 8 16 8 16 32 16 32 64 32 64 128 64 Number of domains in the y-direction 4 2 4 8 4 8 6 8 8 16 16 16 32 32 32 64 64 64 128 Elapsed time [min] 205.6 219.4 118.5 69.3 79.2 52.2 55.7 44.7 26.5 27.5 17.9 13.3 13.6 11.2 10.0 10.1 9.5 9.3 9.4 Speed-up 6.9 6.5 12.0 20.6 18.0 27.3 25.5 31.8 53.7 51.7 79.4 106.9 104.5 127.6 142.4 140.6 150.1 153.2 151.3 Efficiency [%] 86.5 81.1 75.1 64.2 56.2 56.8 53.2 49.7 42.0 40.4 31.0 20.9 20.4 12.5 7.0 6.9 3.7 1.9 1.8 Table IV. Same as in Table III but the domain was decomposed only in y-direction Table IV. Same as in Table III but the domain was decomposed only in y-direction Total number of domains 1 8 16 32 48 64 128 256 512 1024 Number of domains in the x-direction 1 1 1 1 1 1 1 1 1 1 Number of domains in the y-direction 1 8 16 32 48 64 128 256 512 1024 Elapsed time [min] 1423.6 213.2 121.8 71.0 51.8 42.6 27.6 20.3 15.9 14.5 Speed-up 6.7 11.7 20.1 27.5 33.4 51.6 70.2 89.4 98.5 Efficiency [%] 83.5 73.0 62.7 57.2 52.2 40.3 27.4 17.5 9.6 Table V. CONCLUDING REMARKS AND FUTURE RESEARCH DIRECTIONS In this paper, we presented an example of the application of the K supercomputer to a large-scale/high-resolution flood-inundation simulation. The application site was the Osaka region along the Yodo River. The simulation area size was approximately 34.5 km × 37.1 km. The area was dis- cretized with a mesh resolution of 10 m, resulting in 12824442 (= 3453 × 3714) computational nodes. As a result, the computational speed was considerably enhanced. The fastest computational time among all the simulations was realized by a 128 × 64 = 8192 domain decomposition (CPUs) with 8 threads (cores) in each CPU. The computational time was 9.3 min for a five-hour simulation. Compared to the simulation with 1 × 1 domain (CPU) with 8 threads (cores) in each CPU, the speed was 153.2 times faster. This speed is sufficiently fast and is therefore suitable for real-time simu- lations as well. III–V. Table III shows the results for the case when the domain was decomposed only in the x-direction. Likewise, Table IV shows the results when the domain was decom- posed only in the y-direction. Table V shows the results obtained when the domain was decomposed in both x- and y-directions.i y From these figures and tables, we can summarize the fol- lowing: (1) The fastest computational time among all simulations was realized by a 128 × 64 = 8192 domain decomposi- tion (CPUs) with 8 threads (cores) in each CPU. The minimum computational time was 9.3 min. (2) The maximum computational time among all simula- tions was 1423.6 min for a 1 × 1 domain (CPU) with 8 threads (cores) in each CPU. Note that the computa- tional time with 1 CPU and 1 core is more than 24 h, which is the maximum time limit for the use of the K supercomputer; thus, the exact time taken could not be calculated. In the future, the simulation speed would be even faster if the original algorithm of the inundation model is further improved; however, we believe that the results reported in this paper will not be rendered inappropriate because of this improvement. In any case, the capability of the K supercom- puter has been proven. The K supercomputer power will become more prominent when the number of computational nodes increases. A total of 12824442 (= 3453 × 3714) com- putational nodes is not considered extremely high in the field of computational fluid mechanics. CONCLUDING REMARKS AND FUTURE RESEARCH DIRECTIONS On the other hand, atten- tion should be paid to the use of the model in practice. One way is to reduce the number of CPUs to a practical level so that the model can be used with a generic parallel computer. The computational efficiency calculated in this study sug- gests that the level of efficiency does not necessarily improve when more CPUs are used, probably because of the load imbalance of the simulation itself between the decomposed domains, which is the next problem to be solved. (3) The maximum/minimum computational time was 153.2. In other words, the simulation speed increased by a factor of 153.2.i (4) However, the efficiency obtained by using the 8-core 8192 CPUs was 1.9%.i (5) An efficiency of around 50% was realized when the total number of CPUs was 64 (i.e. 512 cores). The speed-ups in these cases were around 30%–35%. (5) An efficiency of around 50% was realized when the total number of CPUs was 64 (i.e. 512 cores). The speed-ups in these cases were around 30%–35%. p p 6) In general, the computational time decreased with an increase in the number of CPUs. Nevertheless, a linear speed-up was not realized. The main reason for this is most likely related to a load imbalance problem. The decomposed domain with the dry cells must wait for the calculation of the wet cells, i.e. for slower calcula- tion domains. Data are transferred at each time step, meaning that the load imbalance that accumulates after each step does not result in an ideal speed-up. Table VI shows the communication (data transfer) time with respect to five decomposition patterns (viz., 1 × 16, 16 × 1, 1 × 256, 256 × 1, and 16 × 16). It is clear from this table that the time needed for communication is neither very significant nor different with respect to the vertical and horizontal decomposition between 16 and 256 decompositions. This is because the required time for communication in sleeves does not change consider- ably during the one-way decomposition given that the amount of communication in the sleeve is almost the same. In contrast, there was a significant reduction in the communication time when the domain was decom- posed in both i and j directions, implying that there is less communication in a root-square order as the domain decomposes in two directions. FLOOD SIMULATION WITH K SUPERCOMPUTER FLOOD SIMULATION WITH K SUPERCOMPUTER Table VI. Time for communication (i.e. data transfer) in the simulation Number of CPUs 16 16 256 256 256 i-decomposition 1 16 1 256 16 j-decomposition 16 1 256 1 16 Communication (min) 1.8 1.6 1.8 2.3 0.5 Table VI. Time for communication (i.e. data transfer) in the simulation Table VI. Time for communication (i.e. data transfer) in the simulation pared to that required for the main loops (i.e. the calcu- lations), which are mostly already parallelized. Thus, an ideal speed-up was not realized probably because of the load imbalance. Number of CPUs 16 16 256 256 256 i-decomposition 1 16 1 256 16 j-decomposition 16 1 256 1 16 Communication (min) 1.8 1.6 1.8 2.3 0.5 CONCLUDING REMARKS AND FUTURE RESEARCH DIRECTIONS In any case, the time required for communication is negligible, as com- In any cases, an approach to enhance the simulation speed with a commercial desktop computer is worth pursuing. Nevertheless, the approach using supercomputers should be developed further for a bright future, since the evolution of the computer is often beyond our expectations. RESULTS Same as in Table III but the domain was decomposed in both x- and y-directions Total number of domains 8 8 16 32 32 48 48 64 128 128 256 512 512 1024 2048 2048 4096 8192 8192 Number of domains in the x-direction 2 4 4 4 8 6 8 8 16 8 16 32 16 32 64 32 64 128 64 Number of domains in the y-direction 4 2 4 8 4 8 6 8 8 16 16 16 32 32 32 64 64 64 128 Elapsed time [min] 205.6 219.4 118.5 69.3 79.2 52.2 55.7 44.7 26.5 27.5 17.9 13.3 13.6 11.2 10.0 10.1 9.5 9.3 9.4 Speed-up 6.9 6.5 12.0 20.6 18.0 27.3 25.5 31.8 53.7 51.7 79.4 106.9 104.5 127.6 142.4 140.6 150.1 153.2 151.3 Efficiency [%] 86.5 81.1 75.1 64.2 56.2 56.8 53.2 49.7 42.0 40.4 31.0 20.9 20.4 12.5 7.0 6.9 3.7 1.9 1.8 —66— ACKNOWLEDGEMENTS In this research, we used the computational resources of the K supercomputer provided by the RIKEN Advanced Institute for Computational Science through the HPCI System Research Project (Project ID: hp140220, PI: Dr. Kazuo Saito from MRI, Japan). Further, the first author was involved with the following projects: MEXT ‘Precise Impact Assessments on Climate Change’ with the Program for Risk Information on Climate Change (SOUSEI Program, PI: Prof. Eiichi Nakakita, Group Leader (GL): Prof. Yasuto Tachikawa, both from Kyoto University); MEXT Global —67— K. KOBAYASHI ET AL. COE programme, ‘Sustainability/Survivability Science for a Resilient Society Adaptable to Extreme Weather Conditions’ (PI: Prof. Kaoru Takara from Kyoto University); CECAR– Africa project ‘Enhancing Resilience to Climate and Ecosystem Changes in Semi-Arid Africa: An Integrated Approach’ (PI: Prof. Kazuhiko Takeuchi from the United Nations University, GL: Prof. Hirohiko Ishikawa from Kyoto University); and Annual Research Budget of RCUSS, Kobe University (PI: Prof. Satoru Oishi). Through the pro- grammes, the first author received considerable help in terms of both research guidance and budget, particularly for the purchase of the workstations used for the model develop- ment. Prof. Kazuya Inoue and Dr. Kenji Kawaike, both from Kyoto University; Dr. Hiromu Seko from MRI; Dr. Yuichiro Oku from the University of Hyogo; Dr. Shigenori Otsuka from RIKEN AICS; Dr. Tsutao Oizumi from JAMSTEC; Dr. Kosei Yamaguchi from Kyoto University and colleagues of Fujitsu Inc., Fujitsu Systems East Ltd.; and Dr. Hiromichi Tsumori from Sompo Japan Nipponkoa Risk Management Inc. provided us with useful insights. We would like to express our sincere gratitude to all of them. Japanese). Japan Society of Civil Engineers. 2001. Hydraulics Formula, Example 2.3. Japan Society of Civil Engineers, CD-ROM (in Japanese). Kobayashi K, Takara K. 2012. Development of a distributed rainfall-runoff/flood inundation simulation and economic risk assessment model. Journal of Flood Risk Management 6: 85–98. DOI: 10.1111/j.1753-318X.2012.01157.x. Kobayashi K, Takara K, Oku Y. 2011. Development of an inte- grated model for rainfall-runoff/flood inundation simulation and economic loss estimation focusing on Sayo town, Hyogo, Japan. Journal of Japan Society of Civil Engineers, Ser. B1 (Hydraulic Engineering) 67: I_949–I_954. DOI: 10.2208/ jscejhe.67.I_949 (in Japanese). Kobayashi K, Takara K, Sano H, Tsumori H, Sekii K. 2012. Development of a standardized distributed rainfall-runoff/ flood inundation model applicable to property insurance. Journal of Japan Society of Civil Engineers, Ser. B1 (Hydraulic Engineering) 68: I_1069–I_1074. DOI: 10.2208/ jscejhe.68.I_1069 (in Japanese). ACKNOWLEDGEMENTS Kobayashi K, Takara K, Sano H, Tsumori H, Sekii K. 2014a. A high-resolution large-scale flood hazard and economic risk model for the property loss insurance in Japan. Journal of Flood Risk Management. DOI: 10.1111/jfr3.12117. REFERENCES Kobayashi K, Oku Y, Nakakita E, Nakano M, Takara K. 2014b. Flood scale change on Yodogawa river catchment by Isewan Typhoon pseudo-global warming experiment. Journal of Japan Society of Civil Engineers, Ser. B1 (Hydraulic Engineering) 70: I_391–I_396. DOI: 10.2208/jscejhe.70.I_391 (in Japanese). Bates PD, De Roo APJ. 2000. A simple raster-based model for flood inundation simulation. Journal of Hydrology 236: 54–77. DOI: 10.1016/S0022-1694(00)00278-X. Bates PD, Horritt MS, Fewtrell TJ. 2010. A simple inertial for- mulation of the shallow water equations for efficient two- dimensional flood inundation modelling. Journal of Hydrology 387: 33–45. DOI: 10.1016/j.jhydrol.2010.03.027. Leandro J, Chen AS, Schumann A. 2014. A 2D parallel diffusive wave model for flood plain inundation with variable time step (P-DWave). Journal of Hydrology 517: 250–259. DOI: 10.1016/j.jhydrol.2014.05.020. Neal JC, Fewtrell TJ, Bates PD, Wright NG. 2010. A comparison of three parallelisation methods for 2D flood inundation models. Environmental Modelling and Software 25: 398–411. DOI:10.1016/j.envsoft.2009.11.007. Fujitsu Journal. 2012. http://www.fujitsu.com/jp/about/resources/ publications/magazine/backnumber/vol63-3.html. Last access September 25, 2015 (in Japanese with an English abstract). Hunter NM, Bates PD, Neelz S, Pender G, Villanueva I, Wright NG, Liang D, Falconer RA, Lin B, Waller S, Crossley AJ, Mason D. 2008. Benchmarking 2D hydraulic models for urban flood simulations. Proceedings of the Institution of Civil Engineers: Water Management 161: 13–30. DOI: 10.1680/wama.2008.161.1.13. Neal JC, Villanueva I, Write N, Willis T, Fewtrell T, Bates PD. 2012. How much physical complexity is needed to model flood inundation? Hydrological Processes 26: 2264–2282. DOI: 10.1002/hyp.8339. Yu D. 2010. Parallelization of a two-dimensional flood inundation model based on domain decomposition. Environmental Modelling & Software 25: 935–945. DOI:10.1016/j.envsoft. 2010.03.003. Inoue K, Toda K, Maeda O. 2000. Inundation model in the region of river network system and its application to Mekong Delta. Annual Journal of Hydraulic Engineering, Japan Society of Civil Engineers 44: 485–490. DOI: 10.2208/prohe.44.485 (in —68—
https://openalex.org/W2886942184	https://jyx.jyu.fi/bitstream/123456789/59123/1/games%20for%20enhancing%20basic%20reading%20and%20mathsskills%20main%20text.pdf	English	null	Games for enhancing basic reading and maths skills: A systematic review of educational game design in supporting learning by people with learning disabilities	British journal of educational technology	2,018	public-domain	7,037	This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details. Author(s): Lämsä, Joni; Hämäläinen, Raija; Aro, Mikko; Koskimaa, Raine; Äyrämö, Sanna- Mari Title: Games for enhancing basic reading and maths skills : A systematic review of educational game design in supporting learning by people with learning disabilities Title: Games for enhancing basic reading and maths skills : A systematic review of educational game design in supporting learning by people with learning disabilities Year: Version: Copyright: Rights: Rights url: Please cite the original version: In Copyright http://rightsstatements.org/page/InC/1.0/?language=en © 2018 British Educational Research Association Accepted version (Final draft) Lämsä, J., Hämäläinen, R., Aro, M., Koskimaa, R., & Äyrämö, S.-M. (2018). Games for enhancing basic reading and maths skills : A systematic review of educational game design in supporting learning by people with learning disabilities. British Journal of Educational Technology, 49(4), 596-607. https://doi.org/10.1111/bjet.12639 2018 Year: Version: Copyright: Rights: Rights url: In Copyright http://rightsstatements.org/page/InC/1.0/?language=en © 2018 British Educational Research Association Accepted version (Final draft) 2018 Please cite the original version: Lämsä, J., Hämäläinen, R., Aro, M., Koskimaa, R., & Äyrämö, S.-M. (2018). Games for enhancing basic reading and maths skills : A systematic review of educational game design in supporting learning by people with learning disabilities. British Journal of Educational Technology, 49(4), 596-607. https://doi.org/10.1111/bjet.12639 Games for enhancing basic reading and maths skills: A systematic review of educational game design in supporting learning by people with learning disabilities Joni Lämsäa, Raija Hämäläinena, Mikko Aroa, Raine Koskimaab and Sanna-Mari Äyrämöb aDepartment of Education, University of Jyvaskyla, Jyvaskyla, Finland; bDepartment of Music, Art and Culture Studies, University of Jyvaskyla, Jyvaskyla, Finland oni Lämsäa, Raija Hämäläinena, Mikko Aroa, Raine Koskimaab and Sanna-Mari Äyräm Joni Lämsäa*, Raija Hämäläinena, Mikko Aroa, Raine Koskimaab and Sanna-Mari Äyrämöb aDepartment of Education, University of Jyvaskyla, Jyvaskyla, Finland; bDepartment of Music, Art and Culture Studies, University of Jyvaskyla, Jyvaskyla, Finland Joni Lämsä Department of Education, University of Jyvaskyla tel: +358 40 8054271 e-mail: joni.lamsa@jyu.fi Mr Joni Lämsä is a Ph.D. student at the Faculty of Education and Psychology, University of Jyväskylä, Finland. His research interests include collaborative learning in technology- enhanced contexts. PhD Raija Hämäläinen works as a full Professor at the University of Jyväskylä. Her main research interest deal with technology-enhanced learning. Raija is the associate editor in Educational Research Review (2018–2022). PhD Mikko Aro is a professor of Special Education at University of Jyväskylä. His primary research interests relate to development of literacy and mathematical skills; dyslexia, dyscalculia and their comorbidity; and interventions of developmental reading and arithmetic disabilities. PhD Raine Koskimaa is a Professor of Contemporary Culture Studies at the University of Jyvaskyla, and the Vice Director of the Finnish Center of Excellence in Game Culture Studies. Koskimaa is the editor-in-chief in the Finnish Yearbook of Game Studies. Sanna-Mari Äyrämö is a Doctor of Philosophy in Digital Culture. Her interests include new digital games. Sanna-Mari Äyrämö is a Doctor of Philosophy in Digital Culture. Her interests include new digital games. Abstract The development of games for people with learning disabilities is one way to enhance the quality of learning and respond to the need for inclusive special educational support. Recently, game researchers have highlighted the need for paying more attention to identifying the game design choices that can strengthen learning. This article reviews recent studies in the field of games that aim at supporting people with difficulties in learning, particularly in basic reading and maths skills. We identify the major characteristics and learning outcomes of the reviewed studies, as well as key design principles that have been used in games for enhancing basic reading and maths skills. The results show that people with specific learning difficulties have positive improvements in the quality of learning. We also found specific gamification elements that have been used to promote the learning of basic reading and maths skills. However, we call for research, which would explicitly examine the effects of game design choices on learning. Currently, the studies that address learning disabilities do not specifically define which kind of games and game design the results refer to, while game design studies do not clarify how these games influence learning. Thus, there is a need to rethink previous empirical studies on game settings for people with learning difficulties via advancing the role of game design in empirical intervention studies. Keywords: dyscalculia, dyslexia, game design, games, learning disability Introduction There is a great deal of optimism surrounding the potential of games in supporting learning by people with learning disabilities. In the area of basic reading and maths skills, games have the potential to enhance children’s engagement with literacy and maths activities, foster skill reinforcement and empower learners’ perceptions of their reading progress (Koskimaa & Fenyvesi, 2015). Learning has cognitive, social and emotional dimensions in which players engage. The central challenges of game-based learning relate to getting children with learning disabilities excited about learning in game settings, maintaining their motivation to practice a skill that has been proven difficult to grasp and mastering a skill that requires considerable repetition. However, motivation is often considered self-evident in any game-like application. Some game researchers have even proposed that learning games may be poorly designed, and that game developers have little understanding of how to design these types of games (e.g., Haworth & Sedig, 2011). Thus, from the perspective of game design, there is a need to identify the mechanisms of a learning game design to strengthen the game features (e.g., good games are cohesive, are varied and have good user interaction; Äyrämö, 2017). We argue that utilising games to support learning by people with learning disabilities is a challenging process that should consider two perspectives: game design and learning disabilities. The current article draws on these two areas, seeking to bridge the gap between them and advance learning game development and research on learning disabilities. Next, we briefly define the framework of learning game studies and research on learning disabilities. Educational Games To understand how to design learning games, it is important to first define the term game, of which there is no universally accepted definition. We build on Juul’s (2005) ‘classic game model’, which defines a game as ‘a rule-based system with a variable outcome, where different outcomes are assigned different values, the player exerts effort to influence the outcome, the player feels emotionally attached to the outcome, and the consequences of the activity are negotiable’. Salen and Zimmerman (2004) reached a bit more condensed formulation, defining games as systems in which players engage in an artificial conflict and whose actions are defined by rules, all of which results in a quantifiable outcome. The socio-cultural theory of games emphasises that games are situated within game-playing communities, and the theory ‘focuses on learning as participation in social practice’ (Ramirez & Squire 2015). Qualities attached to non-educational games are not sufficient on their own to make the games educational. Educational games are often seen as a subcategory of serious games (Hainey, Connolly, Stansfield, & Boyle, 2011), which, according to Zyda (2005), are mental contests played with a computer that use ‘entertainment to further government or corporate training, education, health, public policy, and strategic communication objectives’. Instead of educational games as such, researchers’ interest increasingly has been on game-based learning and gamification of learning. Gamification refers to the practice of applying game elements or functionalities such as points, rewards, tasks, challenges, goals or immediate feedback for learning purposes. Gamification aims at fostering playful and gameful attitudes, which are considered beneficial for learning. Suits (1990) wrote about the player’s lusory attitude that makes the player accept the artificial rules and obstacles of a game and see them as enhancing the game’s fun, rather than merely being nuisances. Mäyrä (2012) identified free play, creative fun and non-instrumental leisure as the characteristics embodied in playfulness. Educational games are increasingly used to support learning by people with learning difficulties (e.g., for supporting job training of persons with developmental disabilities; Kwon & Lee, 2016). Game designers should take pedagogical principles and elements regarding accessibility into account so that disabled students are also able to achieve the desired learning outcomes (Hersh & Leporini, 2013). Hersh and Leporini (2012) argued that game categorisations (e.g., alternate reality games, simulations/microworlds, and role-playing games) link the diverse types of games to the type of learning they support. Educational Games Furthermore, according to Bedwell and colleagues (2012), the game’s characteristics that are relevant to learning can be classified into the following categories: 1) action language, 2) assessment, 3) conflict/challenge, 4) control, 5) environment, 6) game fiction, 7), human interaction, 8) immersion and 9) rules/goals. We also must consider that not all educational applications are called ‘games’ by their developers; one reason for not identifying something as a game is to avoid competition with commercial games. However, it is not the aim of the current article to dwell on the demarcations of what qualifies a game, and our materials include both products called games and platforms with explicit gamified quantities. Learning Disabilities The major challenges of a child’s first years of school relate to the acquisition of accurate and fluent mastery of written language, as well as basic calculation skills. Specific learning disabilities in literacy or arithmetic are the most common and extensively studied learning difficulties (e.g., Galuschka, Ise, Krick, & Schulte-Körne, 2014). From the point of view of supporting learners who are struggling with the mastery of these basic academic skills, educational games can have several benefits. First, games can support a student’s motivation for practicing the compromised skill for extended periods, which is usually required in the case of learning difficulties (Hersh, 2014). Second, the adaptation logic of the games can help tailor the practice of the subskill and skill level, making it the most beneficial for each individual learner’s stage of development (Saine, Lerkkanen, Ahonen, Tolvanen & Lyytinen, 2011). Third, games can provide immediate, supportive and corrective feedback of their performance, which has been shown to be an important element of educational interventions (Räsänen, Salminen, Wilson, Aunio, & Dehaene, 2009). Finally, game-based learning can also be cost- effective, here reducing the need for teacher involvement during the practice of skills. Dyslexia and dyscalculia are the most common reading and maths disabilities. Estimated prevalence rates of both developmental learning difficulties vary between 4–9% for dyslexia and 3–7% for dyscalculia (Landerl & Moll, 2010). Both learning disabilities are also associated with generally low educational outcomes and have been the focus of research aimed at gaining a better understanding of the cognitive underpinnings of these problems. Lyon and colleagues (2003) summarised developmental dyslexia as a specific learning disability characterised by difficulties in accurate and/or fluent word recognition and spelling and decoding problems. These difficulties are generally thought to result from a deficit in processing language at the phonological level. Although dyslexic readers often have problems with reading comprehension as well, their difficulties are considered as resulting from problems with word recognition. Thus, the most successful approaches for supporting learners with dyslexia have focused on language skills — more specifically on phonological processing, decoding and word recognition skills. Dyscalculia refers to a deficit in one or many content areas of mathematics. Learning Disabilities As Furlong and colleagues (2016) pointed out, most research thus far has focused on early problems in achieving basic mathematical skills, such as arithmetic reasoning (addition, subtraction, multiplication and division) or number-specific precursor skills (‘number sense’), for example, estimating quantities, understanding counting and transcoding between number words, digits and quantities. However, the nature of dyscalculia is multi- faceted, which has been a challenge for practitioners attempting to develop interventions (Dowker, 2005). We focus on games meant for enhancing basic reading and maths skills: we also take into consideration digital learning games that comply with the formal definitions of a game (player, conflict, rules and quantifiable outcome), as well as digital learning platforms and applications that have explicit gamified qualities. First, we investigate the major characteristic and learning outcomes of the studies in which the aim of the game is to enhance basic reading and maths skills. Second, we identify the key design principles that have been implemented in the game design. Finally, we evaluate the beneficial gamification elements that can support learning by people with learning disabilities. The following three research questions (RQ) are addressed: RQ 1: What are the major characteristics and learning outcomes of the studies on serious games for enhancing basic reading and maths skills? RQ 1: What are the major characteristics and learning outcomes of the studies on serious games for enhancing basic reading and maths skills? RQ 2: What kinds of key design principles can be identified in research that uses games for enhancing basic reading and maths skills? RQ 2: What kinds of key design principles can be identified in research that uses games for enhancing basic reading and maths skills? RQ 3: What kind of gamification elements are used to promote learning by people with learning difficulties? RQ 3: What kind of gamification elements are used to promote learning by people with learning difficulties? Method and materials We started by conducting a systematic review in which we searched the articles focusing on games for enhancing basic reading and maths skills. After that, we analysed the selected articles in three phases. Search protocol Our search protocol is based on recommendations from the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (Moher, Liberati, Tetzlaff, Altman, & the PRISMA Group, 2010). We applied the following four phases: identification, screening, eligibility and inclusion. An overview of the search protocol is presented in Figure 1 in which we present the number of studies and the inclusion criteria in each phase. We elaborate our search protocol in Table S1. Figure 1. An overview of the search protocol: the number of studies and the inclusion criteria in each phase. Figure 1. An overview of the search protocol: the number of studies and the inclusion criteria in each phase. Data analysis We developed a coding scheme to systematically analyse the studies. The articles were coded and analysed by the study’s authors. Our coding was done in three phases so that we could identify the (1) main characteristics and learning outcomes of the studies (RQ1), (2) key design principles of the games (RQ2) and (3) the gamification elements that refer to being beneficial for learning (RQ3). First, we focused on the main characteristics and the main learning outcomes of the studies of serious games and how they could enhance basic reading and maths skills. This coding included 1) whether the game focuses on reading or maths performance, 2) the number and age of participants, 3) the descriptions of control groups, 4) the form of pre- and post-tests, 5) the length of the gamified interventions/periods and 6) the learning outcomes. In the second phase, we adopted a set of variables (Järvelä, Ekman, Kivikangas, & Ravaja, 2014) to classify different types of games: to identify the key game design principles of the selected studies. As a result, we looked at six gamification elements of the educational games found by our review: 1) what is the context and fantasy of the game, 2) whether the game is played individually or collaboratively, 3) whether the playing is occasional or long term basis, 4) whether the game is adjustable and customisable, 5) whether the game utilises audio-visual aspects and 6) whether the game is adaptable. We also looked at four other properties: 7) whether the game is a learning or entertainment game, 8) whether the task type is cognitive or kinetic, 9) how the game’s achievements relate to learning and 10) how the game utilises log data. Additionally, we focused on the role of teachers during gamified activities. In the third phase, our aim was to evaluate what kind of gamification elements are beneficial in the context of supporting learning by people with particular learning difficulties. We conducted a thematic content analysis (Braun & Clarke, 2006) of the 20 empirical articles. In analysing the 20 research articles, we followed Aveyard’s (2010) idea of evaluating the differences and similarities in the different papers rather than simply summarising them. The aim was to evaluate studies from the perspective of game design and learning outcomes. In this phase, we excluded the articles that focused on piloting the game or conducting a study without learning outcomes. The major characteristics and the main learning outcomes of the studies The summary of the major characteristics of the reviewed studies and their measured learning outcomes are presented in Table S2. Most of the studies (14 articles) focused on games and reading disabilities (17 different games) while six presented a game with the aim of enhancing calculation skills (five different games). Most of the studies were small scale: 11 studies had less than 50 participants, three had between 50 and 100, and only two studies had 100–200 participants. Three of the studies did not report the number of participants and one study did not specify participant characteristics. Overall, 657 participants were included in the 20 papers, from which 12 of the studies were journals, five conferences and three book articles. The participants were 4–12-year-old children. In the reviewed studies, the aim of the games was to improve either the accuracy or speed of reading or calculation performance, or in the case of more targeted goals of practice, the respective subskills of performance. For example, Saine et al. (2011) found that playing GraphoGame had a positive influence on reading accuracy and fluency, as well as letter knowledge and spelling; a study on NumberRace (e.g., Wilson, Dehaene, Dubois, & Fayol, 2009) reported encouraging results of basic mathematical skills achievement. To evaluate the effects of these games, however, attention should be paid to their experimental designs. As shown in Table S2, the most popular research design consisted of pre- and immediate post- tests (n = 10) with a control group (n = 7). The control groups comprised both children with learning disabilities (n = 4, e.g., de Castro Bissaco, Panccioni, Rodrigues, & Domingues, 2014) and children without the need of extra support for early number or reading skills (n = 3, e.g., Salminen, Koponen, Räsänen, & Aro, 2015). The features of the control group, such as the correspondence to the experimental group and the support offered for the control group, are relevant for making conclusions concerning the effects of the game (Mohd Syah, Hamzaid, Murphy, & Lim, 2016). The length of the intervention studies varied from 1 to 28 weeks. Four intervention studies (e.g., Salminen et al., 2015) in learning-oriented publication channels reported the effect sizes of the learning outcomes (see Table S2). In addition to statistical significance, the effect sizes revealed the practical relevance of the findings. As already reported by Räsänen et al. Data analysis In total, we found seven studies (and seven different games) (underlined in Tables S2 and S3) with scientific evidence regarding the learning effects. We grounded our thematic analysis on the six gamification elements identified in the second phase (labelled as 1–6 also in Table S3). We evaluated if and how the studies used these elements to enhance learning. Finally, five authors engaged in a process of critical discussion of the evaluation patterns. Key design principles of the games that enhance basic reading and maths skills The general feature of the studies described above is that they validate the use of a game. However, despite this advantage, it is still difficult to pinpoint which design principles are influential for increasing students’ learning (see also Wilson et al., 2009). In Table S3, we illustrate the key design principles of the games that can be identified in the research on games for enhancing basic reading and maths skills. Furthermore, our evaluation sought indications on how game design is associated with learning. In the following subsection, we analyse further how gamification elements (context and fantasy; individual or collaborative playing; occasional or long-term playing; adjustability and customisability; audio-visual aspects; and adaptability; see Table S3) were used in the selected studies (underlined in Table S2). However, it should be noted that the gamification elements were not an explicit focus of these studies; rather, they were implicit indications. From a game design perspective, all the games were learning games, and almost all the games in our materials shared some qualities (see Table S3). In all the 22 games listed in Table S3 (reported in 20 articles), the task type was cognitive (with some kinaesthetic challenges), which is understandable because these games are meant to help children with cognitive disabilities. The games were intended to be played occasionally so that the games were easy to learn in short sessions. In all the studies, a common idea was that gamification supplemented teacher instruction, and the role of the teacher was not essential during gamified activities because most of the studies (n = 14) mentioned nothing about it. Some of the games include time limits (e.g., Hamilton, 2016), which may pose kinaesthetic challenges in that the player must physically react within the given time frame, but there is no additional spatial precision requirement (e.g., it is sufficient to click on a mouse key, no matter where the cursor is located, or the area to click is relatively large). All the games were played individually, but de Castro et al. (2014) illustrated that it is possible to add social functions to single-player games, such as chatting. As shown in Table S3, more than a third of the cases did not take advantage of adjustability or adaptability. Most cases employed audio-visual content, but in some cases, the games only had audio or graphics but in a modest way. The major characteristics and the main learning outcomes of the studies (2009), a comparison of effect sizes is challenging because they are calculated both by comparing the differences between groups (n = 3, e.g., computer-assisted intervention group vs. mainstream learners) and by comparing the differences within intervention groups (n = 1). Only Torgesen and colleagues (2010) followed up the study with a post-intervention to evaluate the stability of the positive effects observed. Instead of effect sizes, the game design studies (e.g., Wilson et al., 2009) typically reported whether there was a statistically significant difference between the pre- and post-test results (n = 6). Three of these studies (e.g., de Castro et al., 2014) did not report an interaction effect; they only compared pre- and post-test results separately for the experimental and control groups. Altogether, five studies (and five different games) reported at least some improvements in reading skills. Similarly, improvements in mathematical skills were found in five studies (four different games). In addition to intervention studies, Vasalou and colleagues (2017) conducted a qualitative study and reported the results of a thematic analysis on children’s verbal and non-verbal interactions during gamified activities. Finally, we found two studies reporting the preliminary findings of piloting a game and six studies that described the design and development processes of a game (see Table S2). The results regarding LexyLink were discovered in a study that did not fulfil our inclusion criteria (a more than 10-year-old study). Evaluation of beneficial gamification elements in supporting learning by people with learning disabilities Our evaluation aimed to determine whether and, if so, how the specific gamification elements (see Table S3) were used in the empirical studies (underlined in Table S2). Content-based adaptation and feedback systems (indicated by Räsänen et al., 2009; Saine et al., 2011; Salminen et al., 2015; Torgesen et al., 2010; Wilson et al., 2009) were mentioned as the key features of game design. For instance, Räsänen et al. (2009) and Salminen et al. (2015) focused on comparing the effect of two different games (GraphoGame-Math and Number Race) with students’ number skills performance. These studies compared the effect of different dyscalculia-oriented elements (the importance of the exact magnitudes and number symbols vs. approximate comparison processes). However, in both games, adaptation was used to increase variation in the practised context. Content-based adaptation in learning games is often seen as a way to adapt the challenge to an ideal level (to maintain the flow), and it is one key feature of game design that is beneficial for learning purposes (e.g., Salen & Zimmerman, 2004). Salminen et al. (2015) also indicated that a time pressure was used to trigger playing. In game design, challenges may require time-critical or time-independent performance (Karhulahti, 2013). Time-critical performance is often conducted in haste (Elverdam & Aarseth, 2007), and it puts a greater emphasis on motoric skills and reaction times. As in the educational games discussed here, the main task is a mental one, not kinaesthetic, so it is important to adjust the temporal pressure in such a way that the possible limitations in the student’s motoric skills are not interfering with performance. Adaptation and an individualised feedback system can also be seen as favouring the individual learning of children with learning disabilities (e.g., Salminen et al., 2015) because dyslexia and dyscalculia often require individually tailored interventions. When designing the context, fantasy and audio-visual aspects of the games, an essential viewpoint is to engage children in the gameplay. Mohd Syah et al. (2016), Torgesen et al. (2010) and de Castro et al. (2014) highlighted that colourful settings and fun characters (mentioned also by Salminen et al., 2015), possibly along with audio-visual effects, may be engaging elements. A scoring system (Mohd Syah et al., 2016) and immediate feedback (Saine et al., 2011) were also mentioned as engaging and motivating features of the games. Scoring and feedback are an integral part of game design (Schwartz & Bayliss, 2011). Key design principles of the games that enhance basic reading and maths skills Almost all the cases provided some sort of fictional world where the tasks take place (e.g., Vasalou et al., 2017), and often, the world included characters (e.g., de Castro et al., 2014). The game world was mainly nature oriented with animal characters. Discussion Several game-like tools exist (see Figure 1), and the advantages of games as potential tools for supporting learning by people with learning disabilities are generally supported. However, the current research seems to be divided between studies focusing on supporting learning by people with learning disabilities and studies describing the design of game-based technological applications, and there is little discussion between these two approaches. On the one hand, the articles originating from the tradition of gaming define the game content and report learners’ experiences of the game environments in detail, here only marginally focusing on either the learning outcomes or theoretical basis of the application. On the other hand, the reports more focused on the learning disability describe, often marginally, the specifics of the game design and mechanics, focusing instead on the learning processes or subskills that are supposedly practised in the game, as well as the practical outcomes from the point of view of the compromised skill. In game design studies, several advances for learning were found (e.g., positive experiences, as indicated by Gaggi et al., 2017), and 10 of the studies showed the positive effects on cognitive skills (see Table S2). However, the social and emotional dimensions of learning seem to be lacking. Thus, our recommendation for the use of educational games to support learning by people with learning disabilities is that future games should better utilise the social and emotional dimensions of learning. Currently, it seems that the games may not yet optimally match the needs of twenty-first-century learning environments, such as learning approaches that focus on social forms of learning (Kafai et al., 2017) or have a high enough level of technological development. Our concern is that if the gap between learning games and other games grows too wide, learning games may suffer because students with disabilities participate routinely in a wide variety of entertaining and engaging games, making the learning games less appealing. Optimally, games supporting learning by people with learning disabilities should be just as motivating as these non-educational technological contexts, with the added value that players will learn. Next, we will further discuss and provide recommendations for how the design principles and game design reflected in the current study could be better employed in learning games. First, currently, the studied games did not take full advantage of adjustability and adaptability. 1 If user-tracking data are transferred, collected or forwarded from a personal device to an external server for research or whatever purposes, ethical concerns may arise. The normal requirements concerning personal data security and informed consent apply in this case. Typically, however, quantitative and big data approaches are employed on such data, and all data is anonymised and there is no connection to any individual user. Evaluation of beneficial gamification elements in supporting learning by people with learning disabilities This kind of engagement is important because usually, engagement is specifically what is sought through gamification. On the other hand, there are different dimensions to engagement: behavioural, cognitive and emotional (Ge & Ifenthaler, 2017). Emotional engagement alone without consequent cognitive engagement does not necessarily contribute to learning. Immediate feedback is important in articulating how the elements being manipulated should be directly relevant for the skill being learned. Fun-increasing elements, on the other hand, do not seem to have a direct connection to learning success (Iten & Petko, 2016). None of the studied games required long-term playing; the learning effects were achieved by engaging in occasional, recurrent playing. We also evaluated how adjustability has been used to promote learning. Even though all the selected studies, save for Torgesen et al. (2010), indicated that the games were adjustable, their role in learning performance was not analysed or considered. If learners selected which skills they would practice and at what level, the teacher’s role should be to map the learning feedback with game achievements so that the game teaches the right lessons to its player. In the current review, it was not easy to detect the extent to which this was accomplished, but it seemed to be the case in roughly half of the studies. Discussion However, both adjustability (the possibility for the user to modify the appearance or some of the functionalities of the game) and adaptability (the game monitoring user performance1 to maintain the ideal level of challenge) are well-developed aspects in game design. For the future, we recommend using these aspects to address the need for individual tailoring. Second, our recommendation is that future applications support both adaptability for individual needs and social functions (only one of the reviewed studies took advantage of technology-mediated collaboration) in the same game context should be used to bring together two important aspects of inclusive education. Social aspects of game play could be used to heighten the sense of achievement by enabling easy sharing of play sessions or specific stages. Educational games could apply social functions, such as what is done in the game Monument Valley, an entertainment game with educational potential in geometry and proof construction and that includes a one-click option to share screenshots of the beautiful, and often surprising, imagery (Koskimaa & Fenyvesi, 2015). Third, occasional play sessions may often be practical (cf. Peirce & Wade, 2011) but not necessarily optimal for maintaining cognitive engagement for longer periods (‘light-attention mode of engagement’, Hinton & Hjorth, 2013), which is often needed because skill improvement for people with learning disabilities usually takes effort over an extended period of time. For example, Ronimus, Kujala, Tolvanen, and Lyytinen (2014) indicated that although novel tasks and fantasy elements in games may increase disabled children’s engagement in learning games, this effect may not be long-lasting if there are shortcomings in the game design. Games with longer storylines or wider fictional worlds, on the other hand, are impractical in that reaching a specific point may require a long sequence of events. To better meet the need for engagement, we recommend combining the flexibility of casual games and longer interest factor of bigger games, for example, by making a series of casual games, as employed in the studies of Gaggi et al. (2017) (see also Smith & Sanchez, 2010). Our final recommendation concerns future empirical studies on game settings for people with learning difficulties. Namely, our evaluation of the studies illuminated that the game-specific parameters related to, for example, stimulus presentation, adaptation algorithms or game mechanics, were not the explicit focus of the studies. Rather, the parameters seem to be fixed elements of the game. Conclusion The current review illustrates that along with the development of gaming technologies, game design has become a topical issue in terms of enhancing basic reading and maths skills. The present study illustrates that to develop the field further, serious attempts should be made to integrate the knowledge of learning disabilities and experimental intervention methodologies with explicit knowledge on gaming research (see also Hersh, 2014). There is a need to rethink the previous empirical studies on game settings for people with learning difficulties (see e.g. Hwang & Wu, 2012) through advancing the element that has thus far been lacking: the role of game design. With this type of integrative approach, one would not only be able to develop better and more efficient gaming applications for people with specific learning disabilities, but also better understand the effects of gaming parameters from the perspective of motivation and engagement in general. Discussion The unanswered question is what kind of role these properties play in students’ learning and gaming experiences and what kind of parameters best enhance students’ reading and maths performance. Therefore, we call for empirical intervention studies based on the given design guidelines and recommendations (see e.g., Hersh & Leporini, 2013) that would separately investigate the effect of gamification elements and/or design principles on the one hand and the effect of learning-disability-oriented properties in game design on the other (see e.g- Mohd Syah et al., 2016). This would enable the development of a research-based framework for designing or evaluating educational games that support learning by people with learning disabilities (cf. Hersh, 2014). This research was supported by the Academy of Finland (Project 258659, Project 276012, Project 277340) and the Finnish Cultural Foundation. Statements on open data, ethics and conflict of interest Statements on open data, ethics and conflict of interest The research ethics guidelines of the Finnish Advisory Board on Research Integrity (2012) were followed throughout this study. Data collection, analysis and publishing followed the modes of action endorsed by the research community: integrity, meticulousness and accuracy in conducting research. The review data are available to others. There is no conflict of interest. References Acknowledgements This research was supported by the Academy of Finland (Project 258659, Project 276012, Project 277340) and the Finnish Cultural Foundation. References Aveyard, H. (2010). Doing a literature review in health and social care: A practical guide (2nd ed.). Berkshire, UK: Open University Press. Bedwell, W., Pavlas, D., Heyne, K., Lazzara, E., & Salas, E. (2012). Toward a taxonomy linking game attributes to learning: An empirical study. Simulation & Gaming, 43(6), 729–760. Berg, S. (1988). Snowball sampling. In S. Kotz & N. L. Johnson (Eds.), Encyclopedia of statistical sciences, vol. 8 (pp. 528–532). Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101. De Castro, M. V., Bissaco, M. A. S., Panccioni, B. M., Rodrigues, S. C. M., & Domingues, A. M. (2014). Effect of a virtual environment on the development of mathematical skills in children with dyscalculia. PLOS ONE, 9(7), e103354. Dowker, A. (2005). Early identification and intervention for students with mathematics difficulties. Journal of Learning Disabilities, 38(4), 324–332. Elverdam, C., & Aarseth, E. (2007). Game classification and game design: Construction through critical analysis. Games and Culture, 2(3), 3–22. Furlong, M., McLoughlin, F., McGilloway, S., & Geary, D. (2016). Interventions to improve mathematical performance for children with mathematical learning difficulties (MLD). Cochrane Database of Systematic Reviews, 4. Gaggi, O., Palazzi, C. E., Ciman, M., Galiazzo, G., Franceschini, S., Ruffino, … Facoetti, A. (2017). Serious games for early identification of developmental dyslexia. Computers in Entertainment, 15(2), 4. Galuschka, K., Ise, E., Krick, K., & Schulte-Körne, G. (2014). Effectiveness of treatment approaches for children and adolescents with reading disabilities: A meta-analysis of randomized controlled trials. PloS one, 9(2), e89900. Ge, X., & Ifenthaler, D. (2017). Designing engaging educational games and assessing engagement in game-based learning. In R. Zheng & M. K. Gardner (Eds.), Handbook of research on serious games for educational applications (pp. 253–270). Hershey, Pennsylvania: IGI Global. Hamilton, H. (2016). Multisensory associative guided instruction components—Spelling. Journal of Educational Technology Systems, 45(2), 256–284. Hwang, G. J., & Wu, P. H. (2012). Advancements and trends in digital game‐based learning research: A review of publications in selected journals from 2001 to 2010. British Journal of Educational Technology, 43(1), E6–E10. Hainey, T., Connolly, T., Stansfield, M., & Boyle, L. (2011). The use of computer games in education: A review of the literature. In P. Felicia (Ed.) Handbook of research on improving learning and motivation through educational games: Multidisciplinary approaches (pp. 29–50). Hershey, PA: Information Science Reference. Haworth, R., & Sedig, K. (2011). The importance of design for educational games. References In A. Méndez-Vilas (Ed.), Education in a Technological World: communicating current and emerging research and technological efforts (pp. 518–522). Badajoz, Spain: Formatex. Hersh, M. (2014). Evaluation framework for ICT-based learning technologies for disabled people. Computers & Education, 78, 30–47. Hersh, M. A., & Leporini, B. (2013). An overview of accessibility and usability of educational games. In C. Gonzalez (Ed.), Student usability in educational software and games: Improving experiences (pp. 1–40). Hershey, PA: IGI Global. Hersh, M., & Leporini, B. (2012). Accessibility and usability of educational gaming environments for disabled students. Proceedings of Advanced Learning Technologies (ICALT), 2012 IEEE 12th International Conference (pp. 752–753). Rome, Italy. Hinton, S., & Hjorth, L. (2013). Understanding social media. London, UK: Sage. Iten, N., & Petko, D. (2016). Learning with serious games: Is fun playing the game a predictor of learning success? British Journal of Educational Technology, 47(1), 151–163. Juul, J. (2005). Half-real. Video games between real rules and fictional worlds. Cambridge, MA: The MIT Press. Järvelä, S., Ekman, I., Kivikangas J. M., & Ravaja, N. (2014). A practical guide to using digital games as an experiment stimulus. Transactions of the Digital Games Research Association, 1(2), 85–115. Karhulahti, V.-M. (2013). A kinesthetic theory of videogames: Time-critical challenge and aporetic rhematic. Game Studies, 13(1). Kafai, Y., Sawyer, C. I. R., Papert, S., Harel, S. C. I. I., Papert, S., & Duval, E. (2017). Technology and theories of learning. Technology Enhanced Learning: Research Themes, 17(1), 169. Koskimaa, R., & Fenyvesi, K. (2015). A mission impossible? Learning the logic of space with impossible figures in experience-based mathematics education. Opus et Educatio, 2(1), 70–84. Kwon, J., & Lee, Y. (2016). Serious games for the job training of persons with developmental disabilities. Computers & Education, 95, 328–339. Landerl, K., & Moll, K. (2010). Comorbidity of learning disorders: Prevalence and familial transmission. Journal of Child Psychology and Psychiatry, 51(3), 287–294. Lyon, G. R., Shaywitz, S. E., & Shaywitz, B. A. (2003). A definition of dyslexia. Annals of Dyslexia, 53(1), 1–14. Mohd Syah, N., Hamzaid, N.A., Murphy B. P., & Lim, E. (2016). Development of computer play pedagogy intervention for children with low conceptual understanding in basic mathematics operation using the dyscalculia feature approach. Interactive Learning Environments, 24(7), 1477–1496. Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., & The PRISMA Group. (2010). Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. International Journal of Surgery, 89, 336–341. Mäyrä, F. Äyrämö, S.-M. (2017). In order to enable meaningful playing: How to support player's learning through digital game narrative design. Jyväskylä, Finland: University of Jyväskylä. Jyväskylä Studies in Humanities, 307. References (2012). Playful mobile communication: Services supporting the culture of play. Interactions: Studies in Communication & Culture, 3(1), 55–70. Peirce, N., & Wade, V. (2011). Personalized learning for casual games: The ‘language trap’ online language learning game. In T. Connolly (Ed.), Leading issues in games-based learning (pp. 159–177). Reading, UK: Academic Publishing International Limited. Ramirez, D. & Squire, K. (2015). Gamification and learning. In S. P. Waltz & S. Deterding (Eds.), Gameful world: Approaches, issues, applications (629–652). Cambridge, MA: The MIT Press. Ronimus, M., Kujala, J., Tolvanen, A., & Lyytinen, H. (2014). Children's engagement during digital game-based learning of reading: The effects of time, rewards, and challenge. Computers & Education, 71, 237-246. Räsänen, P., Salminen, J., Wilson, A. J., Aunio, P., & Dehaene, S. (2009). Computer-assisted intervention for children with low numeracy skills. Cognitive Development, 24(4), 450– 472. Saine, N. L., Lerkkanen, M.-K., Ahonen, T., Tolvanen, A., & Lyytinen, H. (2011). Computer- assisted remedial reading intervention for school beginners at risk for reading disability. Child Development, 82(3), 1013–1028. Salen, K., & Zimmerman, E. (2004). Rules of play. Game design fundamentals. Cambridge, MA: The MIT Press. Salminen, J., Koponen, T., Räsänen, P., & Aro, M. (2015). Preventive support for kindergarteners most at-risk for mathematics difficulties: Computer-assisted intervention. Mathematical Thinking and Learning, 17(4), 273–295. Schwartz, D. I., & Bayliss, J. D. (2011). Unifying instructional and game design. In P. Felicia (Ed.), Handbook of research on improving learning and motivation through educational games: Multidisciplinary approaches (pp. 192–214). Hershey, PA: Information Science Reference. Smith, P. A., & Sanchez, A. (2010). Mini-games with major impacts. In J. Cannon-Bowers & C. Bowers (Eds.), Serious game design and development: Technologies for training and learning (pp. 1–12). Hershey, PA: Information Science Reference. uits, B. (1990). The grasshopper: Games, life and utopia. Boston, MA: David R. Godine Torgesen, J. K., Wagner, R. K., Rashotte, C. A., Herron, J., & Lindamood, P. (2010). Computer-assisted instruction to prevent early reading difficulties in students at risk for dyslexia: Outcomes from two instructional approaches. Annals of Dyslexia, 60(1), 40– 56. Vasalou, A., Khaled, R., Holmes, W., & Gooch, D. (2017). Digital games-based learning for children with dyslexia: A social constructivist perspective on engagement and learning during group game-play. Computers & Education, 114, 175–192. Wilson, A. J., Dehaene, S., Dubois, O., & Fayol, M. (2009). Effects of an adaptive game intervention on accessing number sense in low-socioeconomic-status kindergarten children. Mind, Brain, and Education, 3(4), 224–234. a, M. (2005). From visual simulation to virtual reality to games. Computer, 38(9), 25–32 References Zyda, M. (2005). From visual simulation to virtual reality to games. Computer, 38(9), 25–32. Äyrämö, S.-M. (2017). In order to enable meaningful playing: How to support player's learning through digital game narrative design. Jyväskylä, Finland: University of Jyväskylä. Jyväskylä Studies in Humanities, 307.
https://openalex.org/W1968549539	https://zenodo.org/records/2308968/files/article.pdf	German	null	Ueber das Paeoniaalkaloid	Zeitschrift für Experimentelle Pathologie und Therapie	1,916	public-domain	10,953	1) Hager's Handbuch der Pharmazeutischen Praxis. 1902. Bd. 2. S. 552. 2) Arch. d. •Apothekervereins im nSrdlichen Deutschland. ¢~ 1825. Bd. 13. S.188. 3) Paeonia Mcutan, aus dem pharmazeutischen Lahoratorium zu Tokio (Japan) yon Dr. G. Martin~ Arch. d. Pharmaz. 3. Reihe. 1878. Bd. 13. S. 335. 4) Beitr~ge zur Chemic der Paeonien. Untersuchungen aus dem pharmazeutischen Institut der Univorsit~t Dorpat. Mitgeteilt yon D ragen d orff. Arch. d. Pharmaz. 1879. 3. Reihe. Bd. 14. S. 412. Zeitgchrift f. exp. Pathologie u. Therapie. 18. Bd. 1. H. 1 Zeitgchrift f. exp. Pathologie u. Therapie. 18. Bd. 1. H. Aus dem Pharmakologisehen Institut der Universit/it (Vorstand: Prof. Dr. H. Kionka). Ueber das Paeoniaalkaloid. Yon Dr. reed. Arnold Holste, Assistenten des Instituts. (Hierzu Tafeln I und ]I und 2 Abbildaugen ira Text.) Unter den vielen zur Gattung Ranunculaeeae gehSrenden Paeonia- arten sind die pharmakologisch in Betracht kommenden folgendel): Paeonia officinalis L., heimiseh in Sfideuropa und als Gartenpflanze wohl- bekannt, deren Wurzeln, Bliiten und Samen in iilterer Zeit zu versehiedenen Zwecken, z. B. gegen Epilepsie, als R~uchermittel und zur FSrderung des Zahnens benutzt wurden. Paeonia albifiora Pall., welche am Himalaya, in Japan und Sibirien zu Hause ist, und deren Wurzel bei Frauenkrank- heiten Verwendung finder. Paeonia Moutan Sims., beheimatet in China und Japan, wird in diesen L/indern mit ihrer Wurzel und Rinde als Volks- mittel gegen nervSse Leiden verwendet. Paeonia obovata Maxim., mit natfirlichem Vorkommen auf Yesso, dercn Wurzel bei Magenaffektionen gebraucht wird. g Sehon im Jahre 1825 finder sieh in dem Archly der Pharmazie ein Refcrat 2) fiber die Analyse der Giehtrosen- oder Paeonienwurzel yon Morin, welche jedoch nach heutigen Begriffen als wenig brauchbar zu bezeichnen ist. Untersuehungen fiber die Wurzel der Paeonia Moutan sind yon Jagi 8) in dem pharmazeutischen Laboratorium zu Tokio ausgeffihrt. Derselbe stel]te aus tier yon japanisehen Aerzten h/iufig als Arzneimittel benutzten Moutanwurzel eine der Caprins~ure beziiglich ihrer prozentischen Zusammensetzung /ihnliehe, aber durch ihren hSheren Schmelzpunkt sich unterscheidende Fettsiiure dar. Sehr wertvolle Beitriige zur Chemic der Paeonien stammen yon Dragendorff.und Stahre ~) aus dem pharma- zeutischen Institute der Universit/it Dorpat. Aus dem Samen und der Wurzel der Paeonia peregrina wurden dargestellt eine Gerbs'/iure, Paeonia- fiuorescin, Paeoniabraun, ein indifferentes Paeoniaharz, eine Paeoniaharz- s~iure, ein Paeoniokristallin und schliesslich aueh ein Alkaloid, welches 1 1 Arnold Holste, 2 Dragendorff als durehaus verschieden yon den in Delphinium Staphi- sagria- und Akonitarten aufgefundenen Pflanzenbasen bezeiehnet, auf dessen weitere Untersuchung er aber verzichtet hat. Als Fortsetzung der eben angefiihrten Ver5ffentlichung wurde die quantitative Analyse des Paeoniasamens yon Stahre 1) und die Zusammensetzung der Wurzel der Peregrina yon Mandelin und Johannsone) ausgeffihrt. Aus der Wurzel der Paeonia Moutan stellte Will 8) einen in farblosen Nadeln kristallisierenden Kfrper her, welchen Nagai als ein aromatisches Keton. erkannte. In der ¥oraussetzung, class in tier Paeonia irgend tin wirksames Prinzip enthalten sein mfisse, weil verschiedene Teile derselben in fremden Litndern und in friiheren Jahren auch bei uns therapeutische Verwendung gefunden haben und eine nahe Verwandtschaft mit der Hydrastis cana- densis L. besteht, widmete ich reich dem pharmakologischen Studium dieser Pflanze. Zu dem Entsehlusse, an die Lfsung dieser schweren hufgabe heranzutreten, bin ich dutch Herrn Prof. Kionka ermutigt worden. 1) Beitr~go zur Chemio tier Paoonien. Untersuchungen aus dem pharmazeutischen Institut der UniversitEt Dorpat. Mitgoteilt yon D rag e n d orff, Arch. d. Pharmaz. 1879. 3. Roiho. Bd. 14. S. 531. 2) Ebenda. S. 535. 3) Arch. d. Pharmaz. 1886. 3. Reihe. Bd. 24. S. 806. 4) Gattermann~ Die Praxis dos organischon Chomikers. 1907. S. 42. 1) Beitr~go zur Chemio tier Paoonien. Untersuchungen aus dem pharmazeutischen Institut der UniversitEt Dorpat. Mitgoteilt yon D rag e n d orff, Arch. d. Pharmaz. 1879. 3. Roiho. Bd. 14. S. 531. Aus dem Pharmakologisehen Institut der Universit/it (Vorstand: Prof. Dr. H. Kionka). Ueber das Paeoniaalkaloid. Yon Dr. reed. Arnold Holste, Assistenten des Instituts. (Hierzu Tafeln I und ]I und 2 Abbildaugen ira Text.) Ausserdem sind die Mitteilungen i)ragendorff's fiber das yon ihm als blassgelber, amorpher, leicht wasserlfslicher Kfrper isolierte Paeoniaalkaloid sehr unbestimmt und ist nicht reeht zu verstehen, weshalb das n~there Studium dieses Alkaloids, haupts~ichlich hinsichtlich der physiologischen Wirksamkeit nicht durchgefiihrt worden ist. p y g g Zuniichst wandte ich mein Interesse der Paeonia albiflora zu, und zwar deren Wurzeln, welche ich vonder Firma Haage &Schmidt in Erfurt bezog. Die Wurzeln wurden mit Biirste und Wasser gereinigt, auf einem Reibeisen zerrieben und mit schwefels~.urehaltigem Wasser mehrere Male 1/ingere Stunden digeriert, wobei darauf geachtet wurde, (lass die Temperatur nicht fiber 50 o C stieg; darauf wurde durchgeseiht und mit dem Seihtuche ausgepresst; diese Ausziige vereinigt und auf dam Wasserbade bei mfgliehst niedriger Temperatur zur Sirupkonsistenz ein- geengt. Dieser Sirup wurde mit dem vierfachen Volumen absoluten Alkohols vermischt und unter hii.ufigem Umrfihren 24 Stunden stehen gelassen und dann abfiltriert. Das saute, alkoholische Filtrat wurde im geschiitzten Wasserbade eingeengt, mit destilliertem Wasser auf genau 100 ccm gebraeht und im Scheidetrichter ausgesehtittelt, und zwar in folgender Weise: 1. Die saure, w/issrige Flfissigkeit wurde aus- geschiittelt mit Petrol~tther (Siedepunkt 35--80o), Benzol (Siedepunkt 8l o) und sehliesslich mit Chloroform; 2. eventuell gelfstes Chloroform dureh Schiitteln mit etwas Petrol~ither eutzogen, sowie Ammoniak im Ueber- schuss zugefiigt und ebenfalls mit Petrol/ither, Benzol und Chloroform ausgeschiittelt. Nach der Vorschrift yon Gattermann 4) wurden den Ausschfittelungen kleine Mengen absoluten Alkohols tropfenweise zugefiigt, Ueber das Paooniaalkaloid. 3 um das Absetzen zu besehleunigen, was sich besonders bei den w/issrig- ammoniakalisehen LSsungen als notwendig erwies. Diese 6 Aussch~itte- lungen wurden im Vakuum bis zur Trockene abdestilliert, der R~ickstand in etwas Alkohol und Wasser au[genommen, auf dem geschi]tzten Wasserbade eingedampft und im Exsikkator aufbewahrt. Nach geschehener Gewichtsbestimmung wurden diese 6 RSekst/inde in Wasser und Alkohol ana 10 ecru, letzterer zum Zwecke der Konservierung, gelSst und zu den nachfolgenden Versuchen verwandt. g Mit diesen aus den wassrigen Extrakten der Wurzel von Paeonia albiflora gewonnenen Substanzen habe ich zun/ichst bei einer sehr grossen Menge yon FrSschen Vorversuche ausgefiihrt, um ausfindig zu maehen, ob es sieh um eine stark toxische Substanz handelt, oder nicht; die einzelnen Versuchstiere wurden ltingere Zeit, bis zu 48 Stunden beobachtet. Die auf- tretenden Erscheinungen waren nicht sehr charakteristisch, so dass man mit Sicherheit annehmcn konnte, es handele sich um eine nicht ausgesproehen giftige Substanz, namentlich hinsichtlich ihrer Einwirkung auf das Central- nervensystem und das Herz. 1) Boitr~igo zur Chemie der Paeonien. Untersuchungon aus dem pharmazeutischen Institut dor Univorsit~it Dorpat. Mitgoteilt yon Dragen dorff, Arch. d. Pharmaz. 1879. 3. Roihe. Bd. 14. S. 418. Aus dem Pharmakologisehen Institut der Universit/it (Vorstand: Prof. Dr. H. Kionka). Ueber das Paeoniaalkaloid. Yon Dr. reed. Arnold Holste, Assistenten des Instituts. (Hierzu Tafeln I und ]I und 2 Abbildaugen ira Text.) Bei Anwendung grosser Dosen zcigte aller- dings das Herz eine Verlangsamung seiner T/i.tigkeit und war stellenweise eine leichte Herabsetzung der Reflexe zu beobachten. Dagegen fand sich ziemlich h/iufig eine br/iunlich-rote Verf/~rbung der Haut der Versuchs- tiere, welche auf die Wirkung der von Dragendorff beschriebenen Farb- stoffe zurSekzuf~ihren ist, sowie zuweilen eine starke Sehleimabsonderung. Es handelte sich bei den zu diesen Vorversuchen benutzten Substanzen um keine reinen KSrper; dieselben enthielten vielmehr noch Pflanzen- kolloide und vor allem jenen eben erw/~hnten Farbstoff. Eine Zusammen- stellung der bei diesen Versuchen an FrSschen mit den erwahnten sechs LSsungen der Paeonia albiflora erhaltenen Resultate ergibt folgende Tat- sachen : 1. Saute LSsungen. 1. Saute LSsungen. g a) Petrol/ither: 1 ecru LSsung ~-0,015 g Substanz. 0,25 ccm =0,0038 g . . . Niehts. 0,35 ~ =0,0053 g . . . 0,5 , =0,0075 g . . . 0,75 ~ ~0,0113 g... Nichts. Stupor, Atmung und Herzaktion verlangsamt. Herabsetzung der Reflexe. b) Chloroform: I ccm L5sung ~ 0,025 g Subst~nz. 0,15 ccm ---- 0,00375 g . . . Starke Schleimabsonderung. 0,2 , =0,0050 g . . . Herabsetzung der Reflexe in ge- ringem Grade. 0,25 , = 0,00625 g . . . Verringerungder Reflexe, Verlang- samung der Herzaktion. 0,5 , =0,0125 g . . . Stupor, Atmung und Herzaktion verlangsamt. c) Benzol: 1 ccm LSsung=0,01 g Substanz. 0,5 ccm ----- 0,005 g .... Nichts. 1,5 , =0,015 g .... Herabsetzung der Reflexe. 1" 1" 4 Arnold Holste, 2. Alkalische LSsungen. 2. Alkalische LSsungen. a) I%trol/tther: 1 ccm LSsung = 0,005 g Substanz. 0,5 eem = 0,0025 g . . . Nichts. 1,5 , =0,0075g . . . Niehts. a) I%trol/tther: 1 ccm LSsung = 0,005 g Substan 0,5 eem = 0,0025 g . . . Nichts. 1,5 , =0,0075g . . . Niehts. b) Chloroform: 1 ccm LSsung---- 0,04 g Substanz. 0,5 ccm = 0,02 g .... Nichts. 1,5 , =0,06 g .... Leichte Herabsetzung der Reflexe. b) Chloroform: 1 ccm LSsung---- 0,04 g Substanz. 0,5 ccm = 0,02 g .... Nichts. 1,5 , =0,06 g .... Leichte Herabsetzung der Reflexe. c) Benzol: 1 ccm LSsung = 0,005 g Substanz. 025 ccm = 0,0025 g . . . Nicbts. 1,5 ~: =0,0075g . . . Nichts. Die alkoholisch-w/issrigen LSsungen der sauren Petrol/ither- und Chloro- formaussehiittelungen der Albifiorawurzel ergaben, wie aus der vorstehenden Zusammenfassung ersiehtlich ist, allein deutliche physiologisehe Reaktionen und sind infolgedessen einer n/theren chemischen Priifung unterworfen. Sie wurden n/imlich mit Tierkohlepulver geschiittelt, abfiltriert und der Filterriickstand gewaschen. Zu der durehfiltrierten und auf wenige Kubik- centimeter abgedampften Fliissigkeit wurdo Kaliumquecksilberjodid zu- gegeben; der auf dem Filter zuriickbleibende Nicderschlag mit H2S-Wasser mehrere Male ausgewaschen, das Filtrat eingeengt und mit AgSO4-LSsung versetzt; darauf mit Schwefelwasserstoffwasser das Silber entfernt und aus der LSsung des Alkaloidsulfates die Basis durch Barythydrat in Freiheit geseizt; schliesslich der iiberschiissige Baryt dureh C02 gef~illt und filtriert. Das Filtrat wurde bei niederer Temperatur im Vakuum zur Trockene abdestilliert und der Troekenriickstand in destilliertem Wasser aufgenommen. Die mit konzentrierter Salpeter- und Schwefel- siiure, Fr6hdes, Mandelins, Meekes und Marquis' Reagenz aus- gefiihrten Reaktionen fielen sowohl bei der sauren Petrol/ither-, wie bei der Chloroformausschiittelung der Albiflorawurzel negativ aus. Trotzdem ist aber doeh wohl das Vorhandensein eines alkaloidartigen, wirksamen Prinzips anzunehmen, welchem die oben mitgeteilten physiologischen Reaktionen zuzuschreiben sind. In analoger Weise vermoehte Dragen- dorff 1) bei dem aus dem Samen der Paeonia peregrina hergestellten Alkaloide keinerlei charakteristische Farbenreaktionen nachzuweisen. In der Annahme, dass das Alkaloid in der Officinalis in gr(isserer Menge als in der Albiflora vorhanden w/ire, unterwarf ich diese Art, welche ihren Beinamen deshalb tr~igt, well sie in Europa fiir Arznei- zweckc benutz~ worden ist, ausschliesslich meiner Untersuchung. Aueh hier schenkte ich mein Interesse zuniichst den Wurzeln, welche ebenfalls yon Haage & Schmidt in Erfurt bezogen wurden. Die Behandlung dieser Officinaliswurzeln war genau die gleiche, wie oben beider Albiflora be- Ueber das Paeoniaalkaloid. 5 sehrieben. 1) Kipponberger, Grundlagen fiir don Nachwois yon Giftstoffen. 1897. S. 45. 2) Dragondorff und Stahre, Boitr~ge zur Chemio der Paoonien. Arch. d. Pharmaz. 1879. 3. goiho. Bd. 14. S. 417. pp g g y 2) Dragondorff und Stahre, Boitr~ge zur Chemio der Paoonien. Arch. d. Pharmaz. 1879. 3. goiho. Bd. 14. S. 417. 1) Kipponberger, Grundlagen fiir don Nachwois yon Giftstoffen. 1897. S. 45. 2. Alkalische LSsungen. Die dort ebenfalls genannfen sechs Alkaloidreaktionen waren si~mtlich negativ bet der sauren und alkalischen Petrol/ither-, sowie bet der alkalischen Benzolausschiittelung, dagegen teilweise positiv bet der alkalischen und sauren Chloroform- und der sauren Benzolausschiittelung. Die mit diesen L~sungen angestellten zahlreiehen Frosehversuche ergaben mit geringen quantitativen Unterschieden ebenfalls die bet tier Albiflora beschriebenen Erscheinungen. g Da die in dem Vorhergehenden zur Anwendung gelangfe Methode~ welche einem gebr/iuehliehen, zur Ermittlung unbekannter organischer K~rper angegebenen Verfahren 1) angepasst ist, nicht zur exakten Iso- lierung des reinen Alkaloids zu fiihren schien, benutzte ich bet meinem weiteren Arbeiten nachstehend beschriebenen Weg, welcher auf dem yon Dragendorff und Stahre 2) im Archly der Pharmazie berichteten basiert~ aber in verschiedenen Punkten eine Ab/i.nderung durch mich erfahren hat. Alle nachfolgenden Untersuchungen nach diesem Arbeitsverfahren sind mit der Paeonia officinalis ausgefiihrt worden, und zwar in erster Linie mit dam Samen dieser Pflanze. Ehe ich abet zum Berichte der mit dem Samen erhaltenen Resultate iibergehe, will ich nicht unterlassen, aueh die Ergebnisse meiner Priifungen der Offieinaliswurzeln kurz zu berichten. Die auf dem Reibeisen geriebenen Wurzeln wurden mit einer aus- reichenden Monge destillierten Wassers angesetzt und auf dem Wasserbade bet nicht mehr als 50 o C digeriert. Diese Digestion wurde teilweise bis zu 4:3 Stunden fortgesetzt und mehrere Male wiederholt. Darauf wurde durch ein Koliertuch durchgeseiht und schliesslich mit Hilfe einer Pflanzen- presse ausgepresst. Die Gesamtmenge wurde bet nicht mehr als 50 o C auf dem Wasserbade his zur Sirupkonsistenz eingedampft und sodann mit dem doppelten Volumen 96proz. Alkohols versetzt. Diese alkoholisch- w/issrige Fliissigkeit liess man unter hiiufigem Umschiitteln in einer grossen Ballonflasche 24 Stunden lang stehen; darauf wurde abfiltriert und der Alkohol im Vakuum abdestilliert; dann reichlieh Natriumkarbonat pro Analyst in Substanz zugesetzt, um eine stark alkalische Reaktion zu erzielen; schliesslich mit Aether ausgeschiittelt. Nach Abgiessen dieser Aetherausschiittelung wurde der R~ickstand auf dem Wasserbade ein- geengt, sowie das dreifache Quantum absoluten Alkohols zugesetzt, filtriert und destilliert. Der wtissrige Destillationsriickstand wurde mit konzen- trierter Schwefels~iure anges~uert, der harzige Niedersehlag abfiltriert, sowie das Filtrat mit Kaliumquecksilberjodid versetzt; dieser Niederschlag abfiltriert und mehrere Male mit dem Destillate nachgewaschen, sodann mit H2S zerlegt und das Filtrat bet niederer Temperatur verdunstet. Es hinterblieb eine br/iunlich gef/trbte LSsung mit einer in Wasser unlSs- lichen braunen Haut, welehe abfiltriert wurde. Dies Filtrat wurde mit AgSO~ yore Jod befreit, dann mit H2S gefffllt, um den Ag-Ueberschuss zu entfernen, fittriert und eingeengt. 2. Alkalische LSsungen. Darauf Zusatz yon Barythydrat, 6 Arnold Holste~ um die Basis in Freiheit zu setzen; Ausf~illung des fibersch~issigen Baryts dureh CO~, Filtration und Einengung. Es resultierte naeh dem Verdampfen ein gelblich gef/irbter K5rper, weleher naeh vollkommner Troeknung im Exsikkator der Glaswand ausserordentlich fest anhaftete, so dass er mit einem Spatel abgesehabt werden musste. Naeh wenigen Minuten aber verwandelte sieh die Substanz beim Stehen an der Luft in eine klebrige Masse; der KSrper war also sehr hygroskopiseh. Man musste sieh infolgedessen dazu entschliessen, ihn in Wasser und etwas Alkohol, letzteren zu Konservierungszwecken, aufzunehmen. Die LSsung war gelb- brKunlich. Die mit dieser LSsung der Base der Offieinalis-Wurzel angestellten Alkaloid-Reaktionen ergaben folgende Resultate: 1. konzentrierte Salpeters/~ure. grasgrfin, 2. , Schwefelsi~ure briiunlich, 3. Fr~hde . . . . . . . . . britunlich, 4. Mandelin . . . . . . . . . negativ, 5. Mecke . . . . . . . . . braunrot, 6. Marquis . . . . . . . . . negativ. Da also das Vorhandensein eines Alkaloids sieher erwiesen war, ging ich dazu fiber, die gewonnene Substanz auf ihre Wirksamkeit am fiberlebenden Uterus nach der Kehrerschen 1) Methode zu priifen. Die- selbe beruht bekanntlich darauf 7 dass ein Uterushorn, z. B. yore Meer- sehweinehen, dem narkotisierten Tiere entnommen, in einem mit Ringer- seher Flfissigkeit gefiillten Beeherglase unter Hilfe eines Thermostaten auf KSrpertemperatur erhalten und dureh einen mit einem HKkehen versehenen Faden mit dem Schreibhebel verbunden wird, weleher die Bewegungen der Uterusmuskulatur auf einem Kymographion verzeichnet. Ohne an dem ])rinzipe dieser Methode etwas zu g.ndern, babe ieh die Apparatur einigen Modifikationen unterzogen. Aus nebenstehender Abb. 1 ergibt sich ohne weiteres, dass die Aufh/ingung des Aluminiumschreib- hebels eine besonders feine ist, sowie dass der Uterus am hinteren Hebelarme angreift, w/ihrend das /iquilibrierende Gewicht tier Hebelaehse mSgliehst nahe gerfiekt ist. Ein empfindlicher Thermoregulator sorgt fiir Temperaturkonstanz in dem mit Asbest umkleideten Bleehkasten. Ffir besonders wertvoll babe ieh es erachtet, den Kubikinhalt des Becherglases fiir die Ringerl5sung, in welche der Uterus gebracht wird, auf ein Mini- mum zu reduzieren, um infolge der auf diese Weise erzielten Yerringerung der Verdfinnung mit mSglichst wenig der zu priifenden Substanz arbeiten zu kSnnen. Allerdings ist dieser Verkleinerung des Kubikinhaltes dadureh eine Grenze gesetzt, dass ausser dem Glashaken, an welehem derUterus befestigt ist, auch ein Thermometer, sowie das ffir den zuzuffihrenden Sauerstoff bestimmte Glasrohr Raum finden mfissen. 1) K eh r er, Physiologische und pharmakologische Untersuchungen an den fiber- lebenden und lebendcn inncren Genitalien. Arch. f. Gyn. 1907. Bd. 81. S. 161. 2. Alkalische LSsungen. 14. Meersehweinehen, virginal. Uterus in tote. Zusatz yon 0,036 g Wurzelalkaloid zu 100 ccm Ringer. 2. Versuch veto 9.9. 14. Meersehweinehen, virginal. Uterus in tote. Zusat 0 036 W l lk l id 100 Ri 2. Versuch veto 9.9. 14. Meersehweinehen, virginal. Uterus in tote. Zusatz yon 0,036 g Wurzelalkaloid zu 100 ccm Ringer. yon 0,036 g Wurzelalkaloid zu 100 ccm Ringer. g g Beobachtungszeit: 380 Min. ~--- 6 Std. 20 Min. Spontanbewegungen: Ganz kleine Exkursionen his zu 3 ram. Zusatz yon je 0,009 g Wurzelalkaloid bei 55, 62, 75, 126 Min. Sofort beginnt eine deutliche VergrSsserung tier einzelnen Exkursionen des Schreibhebels, und zwar yon 1 em his zu 21/2 und 3 ore. Bei 350 Min. wurde die GiftlSsung durch reine RingerlSsung ersetzt; sogleich hSrto die oben besohriebene Wirkung auf und setzte die Zeichnun~ der anf~nglichen Normal- kurve mit Exkursionen his zu 3 mm wieder el,. Ausserdem wurde die Einwirkung dieses Wurzelalkaloids auf den Blutdruek und die Atmung am Kaninehen erprobt und festgestellt, dass keinerlei Einfiuss dureh die intravenSse Injektion auch betr/ichtlicher Mengen der Base zu erzieten war, wie aus nachstehendem Protokoll er- sichtlich ist. Ve r s u c h veto 11.9.14. Kaninchen. Blutdruck, htmung. ReGhte Carotis Blutdruck; Injektion in die linke Vena jugularis yon 2 real 0,0045 g und 2 real 0,009 g, insgesamt 0,027 g Wurzelakaloid, ohne den geringsten Einfiuss auf Herzt~itigkeit und Atmung. Das Hauptgewieht meiner Untersuchungen habe ieh auf die chemische und physiologisehe Priifung des Samens der Paeonia officinalis gelegt. Der Samen wurde anf~nglich wie die bisher verwandte Droge yon Haage & Schmidt in Erfurt, spitter yon Caesar & Loretz in Halle bezogen und yon der letzteren Firma zu feinem Pulver vermahlen. Das Samenmehl wurde mit destilliertem Wasser bei Zimmertemperatur unter h/iufigem Umschiitteln lange Stunden extrahiert und diese Operation naeh gesche- henem Abpressen mehrere Male wiederhol~. Das weitere Verfahren zur Herstellung des Samen-Alkaloids war genau dasselbe wie dasjenige, welches bei der Wurzel zur Anwendung gekommen und vorstehend be- schrieben ist. Beziiglich der Ausbeute gebe ich folgende Zahlen: 18. 7. 1914 823 g Samenpulver ergaben 1,9 g Alkaloidbase. 6. 10. 1914 960 g , , 0,51 g , 16. 11. 1914 3000 g , , 3,2 g , 9. 12. 1914 3000 g , ~ 2~4 g , Die Alkaloidbase des yon Haage & Schmidt bezogenen Samens der Paeonia offieinalis veto 18. 7. 1914 ergab folgende Reaktionen: 1. konzentrierte Salpeters/iure grasgriin, 2. , Schwefels/iure braun, 3. 2. Alkalische LSsungen. Im Gegensatze zu Kehrer babe ich nicht mit einem abgetrennten Horne des Uterus ge- Ueber das Paeoniaalkaloid. 7 arbeitet, sondern fiir besser gefunden, den Uterus naeh MSgliehkeit un- verletzt zu benutzen. Zu diesem Zwecke wurde die Scheide dieht unterhalb des /iusseren Muttermundes durchschnitten, der Uterus in toto herausgenommen, dureh die Portio zur Fixierung des Organs an dora Glashaken ein Seidenfaden geffihrt und ein feines Silberh/~kehen durch das /iusserste Ende des einen Hornes hindurehgestossen; ein an diesem H~kehen befestigter Faden fibertrug die Uterusbewegungen auf den Schreibhebel. Die soeben geschilderte Verwendung des Gesamtuterus scheint mir den physiologischen Verhg:ltnissen n~her zu kommen, als die Methode des abgesehnittenen Homes. Dass die Herausnahmo des Uterus naeh ausgefiihrter Laparotomie, sowie die Ueberfiihrung in die auf 37,50 C gehaltene RingerlSsung in denkbar kiirzester Zeit zu erfolgen hat, soll Abbildung 1. Abbildung 1. nebenbei bemerkt werden. Naeh dieser von mir modifizierten Methode wurden zun/~ehst die Spontanbewegungen des Meersehweinchen-Uterus in einer Normalkurve fixiert, darauf ein bestimmtes Quantum des Wurzel- alkaloides zu der Ringerl5sung zugesetzt und die eintretenden Bewegungs- /inderungen des Sehreibhebels mit Hilfe des Kymographions registriert. Es ergab sieh eine deutliche Steigerung der peristatisehen Bewegungen~ sowie des Gesamttonus des Uterus, und zwar von versehiedener Inten- sitg, t, weiehe in einem gewissen proportionalen Verh/iltnisse zu der ver- wandten Giftmenge stand. Ersetzte man die vergiftete RingerlSsung zweeks Kontrolle dureh unvergiftete, so h5rte die eben beschriebene charakteristisehe Wirkung auf und setzte die Zeiehnung der anfftngliehen 5Iormalkurve wieder ein. Einzelheiten ergeben sich aus naehstehenden Protokollen, we/che ich als Beispiele aus einer grSsseren Serie anfiihre. 8 Arnold Holst% 1.Versuch veto 17. 9. 14. Meerschweinehen. Nieht tr~iehtiges Muttertier. Uterus in tote. Offioinalis-Wurzelalkaloid: 0~018 g zu 100 ecru Ringer. g Beobachtungszeit: 470 Min. == 7 Std. 50 Min. g Spontanbewegungen mit Exkursionen yon 1/4 bis 4 cm. Bei 45 Min. Zusatz yen 0,018 g Wm'zelalkaloid, sofortiges Auftreten yon Exkursionen bis zu 7 era; bei 125 Min Zeiehnungen yon 6 cm g Spontanbewegungen mit Exkursionen yon 1/4 bis 4 cm. Bei 45 Min. Zusatz yen p g g y y 0,018 g Wm'zelalkaloid, sofortiges Auftreten yon Exkursionen bis zu 7 era; g g y bei 125 Min. Zeiehnungen yon 6 cm, g y , ,, 155 , , , 5 ,~ , 185 ,, , , 51/2,, , 320 ,, ,, , 81/2,, mit Sehwankungen bis zu 9 cm, ,, 440 ,, ,, ,, 9 ,, 470 ,, , ,, 8 , HShe. 2. Versuch veto 9.9. 2. Alkalische LSsungen. FrShde . . . . . . . . . braun 7 Ueber das PaooniaMkaloid. 9 4. Mandelin . . . . . . . . bdiunliGh ? 5. Mecke . . . . . . . . . braunrot, 6. Marquis . . . . . . . . . britunlich. Es zeigte sieh also eine fast genaue Uebereinstimmung mit der oben besehriebenen Reaktion des Wurzelalkaloids. Das am 16. 11. 1914 yon Haage & Sehmidt bezogene Samenquantum war sehr fettreich und liess sigh night zu so feinem Pulver verarbeiten, wie die friiheren Mengen. Die Reaktionen waren folgende: 1. konzentrierte Salpeters~iure . gelb, 2. , Sehwefelsiiure rotbraun, 3. FrShde . . . . . . . . . rotbraun, 4. Mandelin . . . . . . . . br~unlieh, 5. Mecke . . . . . . . . . br~unlich, 6. Marquis . . . . . . . . . briiunlich. Das aus der yon Caesar & Loretz direkt bezogenen Samenmenge vom 9. 12. 1914 hergestellte Alkaloid zeigte folgende Reaktionen: g g g 1. konzentrierte Salpetersiiure gelb, 2. , Schwefels~ture rotbraun, 3. FrShde . . . . . . . . schwarzbraun~ 4. Mandelin . . . . . . . braun, 5. Mecke . . . . . . . . braunrot, 6. Marquis . . . . . . . bdiunlich. Es ergibt sieh also, dass die Ausbeute an reiner Base Sehwankungen unterliegt, welehe sigh aus der verschiedenen Beschaffenheit und Pro- venicnz der Droge erkliiren; auch die Nuancen der Farbenreaktionen sind wohl zum grSssten Teile auf denselben Grund zuriickzufiihren, teilweise aber auch dadurch entstanden, dass die Alkaioidbase trotz vielfacher Bemfihungen noch night vollkommen rein yon mir dargestellt werden konnte, wie die nachfolgenden Verbrennungswerte der Substanzmenge yore 18. 7. 1914 ergeben: H : 4,72 pCt., C : 29,02 pCt., N ---- 0,72 pCt, Asche ~-- 19,61 pCt. p Wenn ich das Resultat meiner chemisehen Untersuehungen noch einmal kurz zusammenfasse, so babe ich festzustellen, dass ieh sowohl aus der Wurzel, wie aus dem Samen der Paeonia officinalis eine Alkaloid- base hergestellt habe. Da aber die Ausbeute aus dem Samen eine grSssere waq wurde schliesslich nur das Samenmehl als Ausgangspunkt benutzt. Es gelang, die Alkaloidbase in der Form eines hellgelben und amorphen Pulvers herzustellen, welches die oben besehriebenen Reaktionen aufwies. Dieser KSrper, dessen Zusammensetzung die vorstehenden Ver- brennungswerte crkennen lassen, erwies sich als stark hygroskopisch und rein wasserlSslich. Koino Einwirkung auf Blutdruck und Atmung. 2. Alkalische LSsungen. Ich war deshalb gezwungen, alle im folgenden be- sehriebenen Experimente mit einer witssrigen LSsung des Samen-Alkaloids (in den Protokollen mit S.-A. bezeichnet) auszufiihren, welcher zu Kon- servierungszwecken etwas Alkohol zugesetzt wurde. Die yon mir mit dem Samen-Alkaloid angestellten Experimcnte dienten zur Feststellung der Wirksamkeit auf den Blutdruck und die 10 Arnold Holste~ Atmung beim Kaninchen, sowie auf die Gerinnungsf~higkeit des Blutes, ferner der Einwirkung auf das iiberlebende Froschherz mit Hilfe des yon mir verbesserten 1) Williams'schen Apparates. Sodann studierte ich die durch das Samen-Alkaloid hervorgerufenen Ver~nderungen des Tonus der Nierenkapillaren und untersuehte den Einfluss auf den virginalen und graviden Uterus beim Meerschweinchen. In dieser Reihenfolge gebe ieh nachstehend die Ergebnisse meiner Experimente. g p Der Blutdruck der rechten Carotis der betreffenden Kaninchen wurde am Kymographion registriert und in die linke grosse Halsvene die A1- kaIoidlSsung injiziert, ohne dass irgend eine Einwirkung auf den Blut- druck oder die Atmung zu konstatieren gewesen w~tre. 1. Versuch yore 2.2.15. Kaninchon: Blutdruck, htmung. IntravenSsO,02OgS.-h. 12 Uhr 5 Min. Normalkurvo. 12 , 45 ,, injiziort 0,002 g S.-A. 12 ,, 49 ,, ,, 0,002 g S.-A. 12 ,, 54 ,, ,, 0,008 g S.-A. 1 ,, -- ,, ,, 0,008 g S.-A. 1 ,, 30 ,, Endo, Boobachtungszeit: 1 Std. 25 Min. 1. Versuch yore 2.2.15. Kaninchon: Blutdruck, htmung. IntravenSsO,02OgS.-h. 2. Vorsuch vomS. 2.15. Kaniachen: Blutdruck, htmung. IntravenSs0,040gS.-A. 12 Uhr 6 Min. Normalkurve. 12 ,~ 11 ,, injiziert 0,016 g S.-A. 12 ,, 24 ,, ,, 0,016 g S.-A. 5 ,, 20 ,, ,, 0,008g S.-A. 6 ,, -- ,, Ende. Beobachtungszeit: 5 Std. 54 Min. Ueber das Paeoniaalkaloid. Bei der Ausffihrung der Blutdruckexperimente stellte sich die iiber- raschende Tatsaehe heraus, dass sehr h/iui'ig nach Vornahme der intra- venSsen Injektion eine Gerinnung in der Arterienkanfile auftrat. Diese 5fter wiederkehrende Erscheinung rief in mir die Ueberzeugung wach, dass das Samenalkaloid eina Steigerung der Gerinnungsf~,higkeit des im lebenden Tiere kreisendan Blutas verursacht.. Um diese Annahme, welaha yon ausserordentliah grosset Bedeutung for die praktische Verwendbarkeit des Samenalkaloids sein wiirde, zu bestiitigen, sind die nachfolgend be- schriebenen Blutuntersuchungen nach der yon Sahultz x) angegebenen Methode ausgefiihrt worden. Dieselbe baruht darauf, dass in eine 10 cm lange Glaskapillare, welche aus 16 gleich grossen, aneinandergereihten Hohlperlen besteht, einige Trapfen Blutes eingezogen, darauf nach gleiahen Zeitintervallen die einzelnen, mit einer kleinen Feile umritzten Perlen abgebroehen und in Reagenzgl/iser mit ja 1 cem physiolagischer Koch- salzlSsnng gewarfen werden. Man beabaahtet den Moment, we das erste sehwaehe Gerinnsel in einer solehen Perle auftritt, markiert ferner die Zeit, wenn die H~lfta dar Parle yon geronnenam Blute angafiillt ist, und bezeiahnet als das Ende der Untersuchung den Augenblick, we ein kam- paktes Gerinnsel den Inhalt der Glasperle ausmacht, ahna dass siah in der KoehsalzlSsung noch Blut ausschiitteln liesse, dieselbe also vollst/indig klar bleibt. Meiner Ansicht nach beruht die technische Sehwierigkeit dieses Verfahrens, nachdem die gewiss nicht leiehte Frage der Hahl- perlenkapiilaren yon dem Autor auf das vorziigliehste gelSst worden ist, in der Beobachtung der hei der Entnahme des Blutes notwendigen Vorsichtsmassregeln. Um ganz cinwandfreie Resultate zu erzielen und die Einwirkung thrombokinatischer Fermente sicher zu vermeiden, babe ich das zu untersuchende Blut mit Hilfe einer Glaskaniile der Carotis entnommen und vor jeder solehen Entnahme eine neue Kaniile, welcha sorgfiiltig mit Aq. dest., Alkohol und Aether gereinigt und vollkommen troaken war, eingebunden, wail sonst wesentlich andera Werte erhalten werden, als die bei Beohaehtung dieser Vorschrift resultierenden. Ausser- dem empfehle ich, die in gleieher Weise vorbereitete Hohlperlenkapillare und Rekordspritze dutch Stiickehen sehr feinen Gummischlauahes mit der Arterienkaniile zu verbinden; auf diese Weise hat man es bei vorsiehtigem Oeffnen der Arterienklemme vallkommen in der Hand, den Bluteintritt in die Hohlperlenkapillare zu regulieren und jeden Lufteintritt zu vermeiden. Unter Anwendung dieser Sicherheitsmassregeln sind naehstehende Versuche an Kaninchen ausgefiihrt. Die Zeitintervalle zwisehen dem Abbreehen der einzelnen Perlen betragen genau 1 Minute und werden durch die rSmischen Zahlen der obersten horizontalen Reihe tier falgenden Protokolte wieder- gegeben. 1) Schultz, Eiue neuo Methode zur Bestimmung der Gerinnungsfii, higkeit des Blutes. Berliner klin. Wochensohr. 1910. Nr. 12. S. 527. 2. Vorsuch vomS. 2.15. Kaniachen: Blutdruck, htmung. IntravenSs0,040gS.-A. 12 ,, 24 ,, ,, 0,016 g S.-A. 6 ,, -- ,, Ende. Beobachtungszeit: 5 Std. 54 Min. Keine Einwirkung auf Blutdrack und Atmung. Keine Einwirkung auf Blutdrack und Atmung. 3. Vo rsuch vom 3.2.15. Kaninchon : Blutdruck~ htmung. IntravenSs0,064gS.-A 11 Uhr 53 Min. Normalkurve. 11 Uhr 53 Min. Normalkurve. 12 ,, l0 ,, injiziert 0,016 g S.-A. 12 ,, 17 ,, ,, 0,016g S.-A. 5 ,, 45 ,, ,, 0,016 g S.-A. 5 ,, 50 , ,, 0,016 g S.-A. 6 ,, 30 ,, Ende. Beobachtungszeit: 6 Std. 37 Min. Keino Einwirkung auf Blutdruok und Atmung. 4. Vers u ¢h yore 16. 2.15. Kaninchon: Blutdrack, Atmung. IntravenSsO,080gS.-h. 4. Vers u ¢h yore 16. 2.15. Kaninchon: Blutdrack, Atmung. IntravenSsO,080gS.-h. 11 Uhr 55 Min. Normalkurve. 12 ,, 7 ,, injiziert 0,016 g S.-A. 12 ,, 9 ,, ,, 0,016 g S.-A. 3 ,, -- ,, ,, 0,016 g S.-A. 3 ,, 4 ,, ,, 0,016 g S.-A. 3 , 17 ,, ,, 0,016, S.-A. 5 ,, 15 ,, Ende. Boobachtungszoit: 5 Std. 20 Min. K i Ei i k f Bl td k d At 11 Uhr 55 Min. Normalkurve. 11 Uhr 55 Min. Normalkurve. 12 ,, 7 ,, injiziert 0,016 g S.-A. 12 ,, 9 ,, ,, 0,016 g S.-A. 3 ,, -- ,, ,, 0,016 g S.-A. 5 ,, 15 ,, Ende. Boobachtungszoit: 5 Std. 20 Min. 5 ,, 15 ,, Ende. Boobachtungszoit: 5 Std. 20 Min. Koino Einwirkung auf Blutdruck und Atmung. 1) ttolste, ZurWertbostimmung yon Horzmittoln. DiosoZeitschr. 1914. Bd.15. S. 385. 1) ttolste, ZurWertbostimmung yon Horzmittoln. DiosoZeitschr. 1914. Bd.15. S. 385. 11 Ueber das Paeoniaalkaloid. Ueber das Paeoniaalkaloid. Es entsprechen demnach diese rSmischen Ziffern auch den Nummern der mit je 1 acre physialogisaher KochsalzlSsung geffilltan Reagenzgl/tser, in welchen die Ausschfittelung der abgebroahenen Hohl- l d i 12 Arnold Holsto, Datum 1915 I lI Zeitintervalle yon 1 Minute III IV V gI VII VIII IX X X[ XII Zeit naeh der Injektion in Minuten 4"5" I o do: o do. [ 0 do: do. do. I 21. 6. do. do. do. do. do. do. do. a) Normalversuehe. ++ ++ ++ ÷÷ ++ ++ ; +++ ++ ++ ++ ++ + ++ ++ ++ +++1 ++ +++1 ++ +++1 +++ ++ +++ b) Samenalkaloid: 0,115 g. -- t{aninehen: Gewieht 1400 g. 0 0 ; +++ -4-+ c) Same 0 O + ++ + ++ o + + +++ ++ +++ ++ ++ +++ ~+ +;+ ++ ++ ++ +++++ ++ ++ ++÷ alkaloid: 0,16 g. -- Kaninehen: Gewizht 2450 g. + ~± ++ ++ +++ ++ +++ ++ ÷+ +++ +++ 5 Min. 24 , 39 , 54 8~ 5 Min. 23 32 , 43 , 53 , 68 , 76 . 84 , Datum 1915 I lI Zeitintervalle yon 1 Minute III IV V gI VII VIII IX X X[ XII Zeit naeh der Injektion in Minuten 4"5" I o do: o do. [ 0 do: do. do. I 21. 6. do. do. do. do. do. do. do. a) Normalversuehe. ++ ++ ++ ÷÷ ++ ++ ; +++ ++ ++ ++ ++ + ++ ++ ++ +++1 ++ +++1 ++ +++1 +++ ++ +++ b) Samenalkaloid: 0,115 g. -- t{aninehen: Gewieht 1400 g. 0 0 ; +++ -4-+ c) Same 0 O + ++ + ++ o + + +++ ++ +++ ++ ++ +++ ~+ +;+ ++ ++ ++ +++++ ++ ++ ++÷ alkaloid: 0,16 g. -- Kaninehen: Gewizht 2450 g. + ~± ++ ++ +++ ++ +++ ++ ÷+ +++ +++ 5 Min. 24 , 39 , 54 8~ 5 Min. 23 32 , 43 , 53 , 68 , 76 . 84 , Zeitintervalle yon 1 Minute Zur Erkl/~rung der vorstehenden Tabellen teile ich mit, dass mit einer Null diejenigen Hohtperlen bezeichnet werden, aus denen sieh das gesamte Nut ohne Gerinnsel aussehiitteln litsst. Ein Kreuz bedeutet, dass zwar die grSsste Menge Nut des Perleninhaltes in die physiologische KochsalzlSsung des Reagenzglases noch iibergeht, aber doch schon feine Gerinnsel auftreten. 1) S. oben. 2) Kobert, Ueber die Beeinflussung der peripheren Gcf~sse durch pharmako- logische Agentien. Arch. f. experim. Path. u. Pharmak. 1886. Bd. 22. S. 81. 3) Pick, Uebor Beeinflussung der ausstrSmonden Blutmenge dutch die Gef~ss- weite ~ndernde Mittel. Ebenda. 1899. Bd. 42. S. 401. Ueber das Paeoniaalkaloid. Zwei Kreuze markieren den Moment, wo ann/ihernd die Htflfte in der Glasperle fest geronnen ist. Mit drei Kreuzen habe ich, ebenso wie Schultz, den Sehluss der Beobaehtung bezeichnet, wo die Glasperle von einem festen Gerinnsel ausgefiillt ist und die physio- logische KoehsalzlSsung beim Schiitteln voltst/indig klar bleibt. Der erste Versueh ist mit normalem Blute au.sgefiihrt, der zweite und dritte naehdem in die linke Vena jug. eines 1400 g bzw. 2450 g sehweren Kaninehens 0,115 g bzw. 0,16 g Samenalkaloid in wKssriger LSsung in- jiziert war. Die naeh dieser Injektion bis zur n/ichstfolgenden Blut- entnahme verstriehene Zeit in Minuten ist hinter jedem Versuche bemerkt. Aus dem Vergleiehe der drei Tabellen ergibt sieh unmittelba 5 dass das Samenalkaloid eine Besehleunigung der Blutgerinnung verursaeht. Die Steigerung der KoagulationsfXhigkeit des Blutes tritt bereits deuflieh 5 Minuten naeh der intravenSsen Injektion hervor, erfg.hrt im Laufe der n/iehsten Stunde eine betr/tehtliehe Zunahme, welehe in dem zweiten Versuehe bet 54 Minuten, in dem dritten bet 68 Minuten ihre HShe er- reicht. WS~hrend diese Maximalwirkung im zweiten Palle bet 84 Minuten eine leiehte Verringerung erffthrt, ist sie in dem dritten Versuehe bet 84 Minuten noeh vollstgmdig erhalten und damit der Beweis erbracht, Ueber das Paeoniaalkaloid. 13 dass diese Eigenschaft des Samenalkaloids sich unter Umsti~nden im ]ebenden Tiere 11/2 Stunden lang erhalten kann. Diese Tatsache, auf we\|che ich am Schlusse meiner Arbeit Bezug nehme, ist meines Er- achtens voa grosser Bedeutung fiir die Verwendbarkei~ des Samenalka]oids in der Praxis. Ich will nicht unerw/ihnt lassen, dass die obigen, durch die exakte Schultz'sehe Methode gewonnenen Resultate mit meinen ein- gangs gescbilderten Beobachtungen bei Gelegenheit der Blutdruckversuche vSllig iibereinstimmen. Um zu erweisen, ob das Samenalkaloid eine Einwirkung auf den Herzmuskel als solchen besitzt, habe ich eine Reihe yon Untersuehungen am iiberlebenden Froschherzen naeh der yon mir verbesserten Williams- schen Methode ausgeffihrt. Es ergab sieh, dass das Samenalkaloid keinerlei Einfiuss auf die T~itigkeit des Herzmuskels besitzt, indem es gelang, das isolierte Herz viele Stunden lang im vergifteten kfinstlichen Kreislaufe bei unveritnderter Funktion am Leben zu erhaiten. 1. Versuch yore 13.10. 14. Temporaria. Zusatz yon 0,0475 g S.-A. Koin Still- stand: 60 Minuten beobachtet. 1. Versuch yore 13.10. 14. Temporaria. Zusatz yon 0,0475 g S.-A. Koin Still- stand: 60 Minuten beobachtet. • 2. Versuch vom 13. 10. 14. Temporaria. Zusatz von 0,0713 g S.-A. Kein Still- stand: 60 Minuten boobaohtet. 3. Versuch yore 15. lO. 14. Tomporaria. Zusatz yon 0,095 g S.-A. Ueber das Paeoniaalkaloid. Kein Still- stand: 180 Minuten beobachtet. 4. Versuch yore 16. 10. 14. Tomporaria. Zusatz yon 0,095 g S.-A. Kein Still- stand: 405 Minuten boobachtot. Die zu diesen Herzversuehen verwandten Substanzmengen sind ausser- ordentlieh grosse im Vergleiehe zu denjenigen, deren Herzwirkung ieh in der oben genannten Arbeit 1) ermittelt babe. Die Quantitiit der dort be- stimmten Herzmittel schwankt zwisehen 0,2 und 4,0 mg, w~thrend bei den vorstehenden Versuchen viele Zentigramme des Samenalkaloids Ver- wendung gefunden haben, ohne dass selbst bei sehr langer Beobachtungs- zeit die geringste Wirkung zu konstatieren gewesen w~tre. Was nun die Beeinflussung der peripheren Gefftsse, insbesondere der Kapillaren dutch das Samena[kaloid anbelangt, so habe ich zuniichst einige Bemerkungen fiber die Technik, sowie fiber anderweitig sieh er- gebende Schwierigkeiten der Versuchsanordnung zu maehen. Kobert 2) hat die hufmerksamkeit bereits darauf gelenkt, dass bei Vornahme kfinstlieher Durchblutungen an Warmblfitern Organ und Blut yon dem- selben Tiere stammen und eine Steigerung der Bluttemperatur fiber 39,5 0 C vermieden werden muss. Ausserdem stellt er die Forderung auf, dass die durch das Gift verursaehte Aenderung der Ausfiussgesehwindig- keit beim nachherigen DurchstrSmen mit unvergiftetem Blute wenigstens teilweise wieder zur Norm zuriiekkehren muss. Die Methodik, welehe Pick 3) anwandte, bestand in der Benutzung einer graduierten Btirette, Arnold Holste~ 14 durch deren Heben und Senken an einem Stativ das Flfissigkeitsniveau konstant erhalten wurde. Ein sehr wichtiges Resultat der Piekschen Untersuchung ist die Tatsaehe, dass~ wenn die gef~sskontrahierende Wir- kung eines Mittels in einem bestimmten Bezirke erwiesen ist, es einen grossen Fehler bedeuten wiirde~ denselben Effekt in einem anderen Organe vorauszusetzen. Ein weiterer in Betracht zu ziehender Umstand fiir die Beurteilung der Kapillarwirkung eines chemischen KSrpers ist yon Mosso 1) festgestellt~ n/imlich~ dass beim DurchstrSmen eines exstirpierten Organes mit defibriniertem Blute die _A_.uslaufsgeschwindigkeit langsam geringer wird. Was nun die Frage nach der eigentlichen Technik der Durch- blutung anbetrifft, so sind versehiedene Methoden angegeben und eine Reihe yon Apparaten konstruiert worden, yon denen ich oben bereits einen angedeutet habe; ein solcher~ allerdings sehr komplizierter~ stammt yon Jacobj:). y j ) Fiir meinen speziellen Zweck habe ich mir nach dem Prinzip des konstanten Druckes den durch nebenstehende hbb. 2 illustrierten hpparat konstruiert. In einem mit empfindlichem Thermoregulator versehenen Thermostaten ist ein viereckiger~ diekwandiger Glaskasten so eingebaut, dass er mit seinem Boden etwa 2 cm yon der Grundfi/iche des um- gebenden Thermostaten entfernt bleibt 7 auf diese Weise also yon allen Seiten yon der W/irmfiiissigkeit umspflt werden kann. 1) Mosso~ Ueber die physiologische Wirkung des Cocains. Arch. f. exp. Path. u. Pharmak. 1887. Bd. 23. S. 190. 2) J acobj~ Apparat zur Durchblutung isolierter iiberlebender Organe. Ebenda. 1889. Bd. 26. S. 388. -- Ueber das Funktionsverm~igen der kfinstlich durchbluteten Niere. Ebenda. 1892. Bd. 29. S. 27. -- Ein Beitrag zur Technik der k~insttichen Durchblutung iiberlebender Organe. Ebenda. 1895. Bd. 36. S. 330. 1) Mosso~ Ueber die physiologische Wirkung des Cocains. Arch. f. exp. Path. u. Pharmak. 1887. Bd. 23. S. 190. Ueber das Paeoniaalkaloid. Durch ein korre- spondierend in der Wand des Glaskastens und des Thermostaten an- gebrachtes Loch fiihrt ein Glasrohr nach aussen, welches den inneren Glaskasten in tier HShe seines Bodens verl/isst. Dieses Glasrohr, welches mit den nStigen Gummidichtungen in den inneren und /iusseren Kasten eingefiigt ist~ dient als Abfiuss fiir das venSse Blut des zu untersuchenden Organes. Es ist ausserhalb des Thermostaten gebogen und liisst das ausfliessende~ venSse Blut auf ein Deckgl~ischen fallen, welches der Hebel eines yon mir konstruierten elektrischen Tropfenz/ihlers an seinem ~iusseren Ende tr~gt. Dieser genau /iquilibrierte Hebel besitzt an seinem vorderen Arme ausserdem ein feines Metallst/tbchen~ das in eine kleine: gl/~serne Quecksilberwanne eintaucht und einen elektrisehen Stromkrcis sehliesst, in welchen ein Despretzsches Signal eingeschaltet ist. Dieses registriert die fallenden Tropfen auf einem Kymographion. Unter dem beschriebenen Deekgl/ischen dient ein graduierter Glaszylinder zur Messung des dureh- gefiossenen Blutquantums. Ein an einem festen Stativ angebrachter, ]anger Metallstab tr/igt ein Bleehgeffiss, in welchem zwei gleich grosse, zylindrische Glasfiaschen so angebracht sind 7 dass sie mit einem rohr- artigen Fortsatz% gut abgedichtet~ den Boden verlassen und in einem doppelt durehbohrten Glashahn zusammenlaufen. Von diesem fiihrt ein kurzer Gummischlaueh zu einem den Verh/~ltnissen des beschriebenen Thermostaten entsprechend gekriimmten Glasrohre, welches an seinem in 15 Uebor das Paeoniaalkaloid. dem Glaskasten befindlichen Ende ein kurzes Stiick Gummisehlaueh triigt, um die Verbindung mit der gl~sernen Arterienkaniile des zu untersuehenden Organs herzustellen. Auf gleiche Weise wird die Verbindung zwischen der Venenkaniile und dem oben erkl/irten Ausfiussrohre bewerkstelligt. Mit Hilfe dieses yon mir konstruierten Apparates sind dig nachstehenden Versuche in folgender Weise ausgefiihrt. g g Als DurchstrSmungsfiiissigkeit benutzte ich das defibrinierte Blur desselben Tieres, welches ich, wie bei den yon mir verSffentlichten Herz- Abbildung 2. Abbildung 2. versuchcn, in dem Verh~iltnisse von 1 : 2 mit physiologischer Koehsalz- 15sung versetzte. Diese Blutmischung hat sich mir bei meinen Arkeiten am isolierten Herzen als die beste DurchstrSmungsfiiissigkeit erwiesen, und ich kann dieselbe auf Grund meiner zahlreichen Vorversuche aueh ffir den vorliegenden Zweck empfehlen. Es wurden nun je 250 ccm dieser Blutmischung in die zylindrischen Glasgef/isse geffllt und von diesen Quanten das eine mit dem zu untersuchenden Samenalkaloid versetzt. Der Glashahn gestattet die Umschaltung des vergifteten und unvergifteten Blutes naeh Belieben. In das umgebende Bleehgef/iss wurde warmes Wasser getan und so die beiden Blutquanten auf 37,5 o C unter Kontrolle versuchcn, in dem Verh~iltnisse von 1 : 2 mit physiologischer Koehsalz- 15sung versetzte. Ueber das Paeoniaalkaloid. Diese Blutmischung hat sich mir bei meinen Arkeiten am isolierten Herzen als die beste DurchstrSmungsfiiissigkeit erwiesen, und ich kann dieselbe auf Grund meiner zahlreichen Vorversuche aueh ffir den vorliegenden Zweck empfehlen. Es wurden nun je 250 ccm dieser Blutmischung in die zylindrischen Glasgef/isse geffllt und von diesen Quanten das eine mit dem zu untersuchenden Samenalkaloid versetzt. Der Glashahn gestattet die Umschaltung des vergifteten und unvergifteten Blutes naeh Belieben. In das umgebende Bleehgef/iss wurde warmes Wasser getan und so die beiden Blutquanten auf 37,5 o C unter Kontrolle 16 Arnold Holste~ eines eingefiihrten Thermometers w/ihrend der ganzen Versuchsdauer er- halten. Den Glaskasten fiillte man mit Ringerseher Fliissigkeit: deren konstante Temperatur yon 37fi°C ein empfindlieher Thermoregulator garantierte. Ein an einem Stativ angeklemmtes, winklig gebogenes und dutch einen Gummischlauah mit der Sauerstoflbomba verbundenes Glas- roAr liess mit Hilfa eines gut funktionierenden Drossalventils den Sauer- stoff in der Ringerfiiissigkeit in Perlen aufsteigan. Das aus dem Ausfluss- rohre kommende und in dam kleinen Masszylinder gemessene Blut wurde dureh Sanerstoff arterialisiert~ bevor es in des betreffende Druekgef/iss zuriiekgegossen wurde. Es ist zu bemerkan, dass alia Versuehe bei der gleiehen DruckhShe yon 90 cm (gemessen vom Boden des Glaskastens bis zum Fliissigkeitsniveau in der Druakfiasahe) ausgefiihrt worden sind, deren Optimum man durah geeignete Vorversnehe ermittelte. Alle naeh- stehenden Experimente sind an Sahweinenieren ausgefiihrt, welche in dam Schlachthause sofort dem getSteten Tiere entnommen und in kSrper- warmer RingerlSsung aufbewahrt, in das [nstitut gebracht wurden. Es kam natiirlieh darauf an, dass die Niare unverletzt in ihrer Kapsel er- halten blieb. Naeh Ligatur des Ureters band man in die Arterie und Vene eine passende Glaskaniile, welche vermittelst kleiner Schlauah- stiiekehen mit dem Zu- und Abflussrohre des Durchblutungsapparates vereinigt wurden. Eine geeignet gebogene Glashiirde hielt die Niere innerhalb der RingerlSsung des Glaskastens in zweekentspreehender Lage. Nach den Feststellungen yon Frey 1) antwortet die Niere auf eine sehwaehe Blutverdfinnung mit einer Tonussteigerung, auf eine starke mit einer Verengerung der Blutgefiisse; aussardem wird nach Mosso ~-) die Durchlaufsgeschwindigkeit in einem exstirpierten Organa bei Benutzung yon defibriniertem Blute allmiihlieh verringert. Um diesen experimantellen Tatsachan gereeht zu werden, haba ich zun/ichst eine Reiha von Normal- versuchen, yon denen ich naehstehend einen mitteile, ausgefiihrt, welche beweisen, dass bei Anwendung der yon mir oben angegebenen Blut- mischung die Tropfenzahl wi~hrend einar mahrstiindigen Beobaehtungszeit anniihernd konstant erhalten blei.ben kann. Versuoh yore 16. 3.15. Normalversuch. Schweineniere. Anfangll Uhr35Min., EndQ 7 Uhr. Yersuchsdauer 7 Std. 25 Min. 1) Frey, Die Reaktion der Niere auf Blutverdfinnung. Arch. f. d. ges. Physiol. Bd. 120. S. 136. 2) M b 1) Frey, Die Reaktion der Niere auf Blutverdfinnung. Arch. f. d. ges. Physiol. Bd. 120. S. 136. 2) Mosso: s. oben. 2) Mosso: s. oben. Ueber das Paeoniaalkaloid. 1 cm Abscissenl~.nge ~ 17 Sek. 11 Uhr 35 Min ......... 18 Tropfen 12 ,, 15 , ........ 14 ,, 3 ,: 20 ,, ........ 16 ,, 5 ~ 20 , ......... 16 . 6 , 25 ,, ........ 12 , 7 , -- . ........ 18 , Versuoh yore 16. 3.15. Normalversuch. Schweineniere. Anfangll Uhr35Min., EndQ 7 Uhr. Yersuchsdauer 7 Std. 25 Min. 1 cm Abscissenl~.nge ~ 17 Sek. Es folgt also unmittelbar, dass w/ihrend der ganzen, langen Dauer dieses Versuehes die Tropfenzahl kaine nennenswerta Aenderung erfahren hat, indem die tatsiichlieh vorhandenen Differenzen so klein sind, dass sie unbeaehtet bleiben kSnnen. In dem vorliegenden Falle stellte sich 17 Uober das Paooniaalkaloid. sogar die iiberraschende Tatsache heraus, dass die Tropfenzah[ beim Anfang und Ende des Versuches die gleiche und die Endziffer grSsser ist als die vier derselben vorangehenden Werte. Ich bin demnach berechtigt, die kapillarverengernde Wirkung eines Mittels anzunehmen, wenn eine wesentliche Verlangsamung der Tropfenfolge beim Durchblutungsversuch zu konstatieren ist, es andererseits abet gelingt, nach Umschaltung des unvergifteten Blutes eine Steigerung der Tropfenfrequenz zu bekommen, obwohl es in den meisten F~illen nicht mSglich sein wird, die Anfangs- geschwindigkeit vollkommen wieder zu erreichen. Urn den physiologischen Beweis einer solehen kapillarverengernden Wirkung vollst~ndig zu er- bringen, miisste als zweite Forderung aufgestellt werden, dass aueh dig Durehlaufsmenge wiihrend der genannten Einwirkung verringert wiirde und nach Umschaltung des guten, unvergiftGten Blutes ann~hernd zur anfftnglichen Quantitiit zuriickkehrte. Dieses zweite Postulat ist nieht bei jedem Versuehe erfiillt worden, indem hitufig nach Auswechslung der Gift- 15sung mit gutem Blute die zu erwartende Steigerung der Ausfiussmenge ausblieb. Zweifelsohne spielt die Dauer der Gifteinwirkung eine grosse Rolle; hiilt dieselbe zu lange Zeit an, so wird dadurch eine derartige Alteration dcr Kapillarwiinde hervorgeru{en, dass auch nach litngerem DurchstrSmen mit gutem 131ute eine Reparation night mehr mSg]ieh ist. Diese yon mir yon vornherein gemachte Voraussetzung fand ich bei einer Reihe yon orientierenden Experimenten best~itigt. Man darf also die Gift- 15sung nur relativ kurze Zeit einwirken lassen, urn die Riiekkehr zur Norm bis zu einer gewissen Grenze wenigstens wicder erzielen zu kSnnen. Ich mSchte an dieser Stelle auf eine Schwierigkeit aufmerksam machen~ welche sich bei Vornahme einer grossen Reihe solcher Untersuchungen herausgestellt hat, n~tmlich den Zeitpunkt genau zu markieren, wo die Giftwirkung in der Niere einsetzt, oder nach Umsehaltung des un- vergifteten Blutes aufhSrt. 1) Virchow's Arch. Bd. 121. S. 606. 2) Ebenda. Bd. 117. S. 288. 3) Kionk% Uober die Giftwirkang dor schwefligen S~ure und ihrer Salze. Habilitationsschr. 1896. S. 8. Zeitschrift f. exp. Pathologic u. Therapie. 18. Bd. 1. H. Zeitschrift f. exp. Pathologic u. Therapie. 18. Bd. 1. H. Ueber das Paeoniaalkaloid. Diese Frage ist sehr wichtig zur Best[mmung der Dauer der Giftwirkung und 1/tsst sieh nur mit Hilfe eines Indikators 15sen, weleher der Blutmischung beigemengt wird, ohne irgend welehen Einfluss auf die Niere zu besitzen, und eventuell durch eine exakte Farbenreaktion am Ausflussrohre naehzuweisen ist. Indem ieh mir die LSsung dieser Aufgabe far sp/itere Zeiten vorbehalte, mSehte ieh nieht unerw~ihnt lassen, dass sieh zu diesem Zweeke mSglieherweise das Indig- karmin eignen kSnnte, welches naeh Pilehnel), Silbermann °-) und Kionka 8) zur Methode der SelbstNrbung lebender Organe benutzt und als absolut ungiftig bezeiehnet wird. Um die soeben mitgeteilten Tatsaehen zu belegen, gebe ich aus einer A hl D hl f h l h i h it d S sogar die iiberraschende Tatsache heraus, dass die Tropfenzah[ beim Anfang und Ende des Versuches die gleiche und die Endziffer grSsser ist als die vier derselben vorangehenden Werte. Ich bin demnach berechtigt, die kapillarverengernde Wirkung eines Mittels anzunehmen, wenn eine wesentliche Verlangsamung der Tropfenfolge beim Durchblutungsversuch zu konstatieren ist, es andererseits abet gelingt, nach Umschaltung des unvergifteten Blutes eine Steigerung der Tropfenfrequenz zu bekommen, obwohl es in den meisten F~illen nicht mSglich sein wird, die Anfangs- geschwindigkeit vollkommen wieder zu erreichen. Urn den physiologischen Beweis einer solehen kapillarverengernden Wirkung vollst~ndig zu er- bringen, miisste als zweite Forderung aufgestellt werden, dass aueh dig Durehlaufsmenge wiihrend der genannten Einwirkung verringert wiirde und nach Umschaltung des guten, unvergiftGten Blutes ann~hernd zur anfftnglichen Quantitiit zuriickkehrte. Dieses zweite Postulat ist nieht bei jedem Versuehe erfiillt worden, indem hitufig nach Auswechslung der Gift- 15sung mit gutem Blute die zu erwartende Steigerung der Ausfiussmenge ausblieb. Zweifelsohne spielt die Dauer der Gifteinwirkung eine grosse Rolle; hiilt dieselbe zu lange Zeit an, so wird dadurch eine derartige Alteration dcr Kapillarwiinde hervorgeru{en, dass auch nach litngerem DurchstrSmen mit gutem 131ute eine Reparation night mehr mSg]ieh ist. Diese yon mir yon vornherein gemachte Voraussetzung fand ich bei einer Reihe yon orientierenden Experimenten best~itigt. Man darf also die Gift- 15sung nur relativ kurze Zeit einwirken lassen, urn die Riiekkehr zur Norm bis zu einer gewissen Grenze wenigstens wicder erzielen zu kSnnen. Ich mSchte an dieser Stelle auf eine Schwierigkeit aufmerksam machen~ welche sich bei Vornahme einer grossen Reihe solcher Untersuchungen herausgestellt hat, n~tmlich den Zeitpunkt genau zu markieren, wo die Giftwirkung in der Niere einsetzt, oder nach Umsehaltung des un- vergifteten Blutes aufhSrt. Ueber das Paeoniaalkaloid. Diese Frage ist sehr wichtig zur Best[mmung der Dauer der Giftwirkung und 1/tsst sieh nur mit Hilfe eines Indikators 15sen, weleher der Blutmischung beigemengt wird, ohne irgend welehen Einfluss auf die Niere zu besitzen, und eventuell durch eine exakte Farbenreaktion am Ausflussrohre naehzuweisen ist. Indem ieh mir die LSsung dieser Aufgabe far sp/itere Zeiten vorbehalte, mSehte ieh nieht unerw~ihnt lassen, dass sieh zu diesem Zweeke mSglieherweise das Indig- karmin eignen kSnnte, welches naeh Pilehnel), Silbermann °-) und Kionka 8) zur Methode der SelbstNrbung lebender Organe benutzt und als absolut ungiftig bezeiehnet wird. Um die soeben mitgeteilten Tatsaehen zu belegen, gebe ich aus einer grossen Anzahl yon Durehlaufsversuehen, welehe ieh mit dem Samen- alkaloid an Sehweinenieren vorgenommen habe, einige Beispiele. Die ersten fiinf Protokolle geniigen hinsiehtlich der Tropfenzahl den auf- gestellten 13edingungen, w/ihrend die beiden letzten den auseinander- 18 Arnold Holsto, gesetzten Forderungen sowohl bezSglich der Tropfenfrequenz wie der Durchlaufsmenge entsprechen. Zum Verst~.ndnis bemerke ich, dass die Zahl der Tropfen auf 1 cm Abscissenliinge~---17 Sek. ausgezAhlt worden is[. 1. Versuch yore 22.3.15. Schweinoniero: 0,192 g Samonalkaloid:250 ccm Blut- misohung. 1 cm Absoissenliinge = 17 SoL 1. Versuch yore 22.3.15. Schweinoniero: 0,192 g Samonalkaloid:250 ccm Blut- misohung. 1 cm Absoissenliinge = 17 SoL g g 11 Uhr 15 Min. Anfang, 7 Uhr 30 Miu. Endo. Beobachtungszoit: 8 Std. 15 Min. 11 Uhr 15 Min. Anfang, 7 Uhr 30 Miu. Endo. Beobachtungszoit: 8 Std. 15 Min. 11 Uhr 15 Min. 8 Tropfon. 11 Uhr 15 Min. 8 Tropfon. 12 ,, -- ,, Zusatz yon 0,192g S.-A. 12 ,, -- ,, Zusatz yon 0,192g S.-A. 3 ,, 22 ,, 4 Tropfon 4 ,, 30 , umgeschaltet, unvergiftotes Blut. 6 25 7 T f 6 ,, 25 ,, 7 Tropfen. 2. Vor s u c h yore 23.3.15. Schweineniere: 07256 g Samenalkaloid : 250 ccm Blut- mischung. 1 cm Abscissonliinge ~ 17 Sek. g g 9 Uhr 45 Min. Anfang, 7 Uhr 10 Min. Endo. Boobachtungszeit: 9 Std. 25 Min. 10 Uhr 50 Min. 17 Tropfon. p 11 , -- , Zusatz yon 0,256g S.-A. 11 , I5 , 5Tropfon. 12 ,, 45 ,, 2 Tropfen. 2 ,, 30 , 1 Tropfen. 2 ,, 34 , umgeschaltet~ unvergiftetos Blut. 3 ,, 50 , 4Tropfen. 4 , 30 ,, 5Tropfen. 3. Versuch yore 27.4. 15. Schweineniere: 0,23 g Samenalkaloid:950 ccm Blut- mischung. 1 cm hbscissenl~nge -~- 17 Sek. mischung. 1 cm hbscissenl~nge -~- 17 Sek. a) Tropfenzahh 12 Uhr 20 Min. 16 Tropfen. p 3 ,, 50 ,, Zusatz yon 0~23g S.-A. y 4 , 15 ,, 4Tropfen. 4 ,, 58 ,, 4Tropfen. 5 ,, 29 ,1 umgeschaltet, unvergiftetos Blut. 6 , 14 ,, 8Tropfon. gelaufeno Mengo in Kubikzentimotern pro Minute: b) Durch 12 Uhr 20 Min. bis 1 Ohr 16 Min. ~ 7 ccm 1 ,, 25 ,, Zusatz yon 0,115 g S.-A. 1 ,, 25 ,, Zusatz yon 0,115 g S.-A. 1 , 25 ,, bis 3 Uhr 45 Min. ---- 6 com. 3 ,, 50 ,, Zusatz yon 0,115 g S.-A. 3 ,, 45 ,, his 5 Uhr 28 Min. ----- 0,5 ccm. 5 ,, 29 ,, umgeschaltet, unvergiftetos Blut. 5 , 29 ,, his 7 Uhr 25 Min. = 0~5 com. 7. Versuoh veto 28.4. 15. Schweineniere: 0,23 g Samenalkaloid:250 corn Blut- misohung. 1 cm Abscissenl~inge : 17 Sek. g g 11 Uhr Anfang, 8 Uhr 2 Min. Ende. Beobaohtungszeit: 9 Std. 2 Min. Ueber das Paeoniaalkaloid. g g 12 Uhr 10 Min. hnfang, 6 Uhr 30 Min. Ende. Beobaohtungszeit: 6 Std. 20 Min. 12 Uhr 10 Min. hnfang, 6 Uhr 30 Min 12 Uhr 25 Min. 12 Tropfen. 12 , 45 , Zusatz yon 0,23g S.-A. 1 ,, 40 , 6Tropfen. 3 , 25 ,, umgeschaltet~ unvergiftetes Blur. 3 ,, 55 , 12 Tropfen. 4. Versuch yore 29.4. 15. Schweineniere: 0~23 g Samenalkaloid:250 ccm Blut- mischung. 1 cm Abscissenliinge ~ 17 Sek. 11 Uhr Anfang~ 6 Uhr 45 Min. Ende. Beobachtungszeit: 7 Std. 45 Min. 11 Uhr Anfang~ 6 Uhr 45 Min. Ende. B 11 Uhr -- Min. 8 Tropfen. 11 Uhr -- Min. 8 Tropfen. 12 , -- , Zusatz yon 0,23 g S.-A. 12 , -- , Zusatz yon 0,23 g S.-A. 1 , 20 ,, 6Tropfen. 2 , -- 7, umgeschaltet, unvergiftetes Blur. 2 ,, 25 ,, 7 Tropfen. 5. Yersuch vom 30. 4. 15. Schweineniere: 0,23 g Samenalkaloid:250 ocm Blut- 5. Yersuch vom 30. 4. 15. Schweineniere: 0,23 g Samenalkaloid:250 ocm Blut- mischung. 1 cm hbscissenliinge : 17 Sek. mischung. 1 cm hbscissenliinge : 17 Sek. mischung. 1 cm hbscissenliinge : 17 Sek. 10 Uhr 30 Min. Anfang, 6 Uhr 50 Min. Ende. Beobachtungszeit: 3 Std. 20 Min. 10 Uhr 30 Min. 17 Tropfen. 12 , 10 ,, Zusatz yon 0,23 g S.-A. 3 , 55 , 3Tropfen. 4 , 55 ,, umgeschaltet, unvergiftetes Blur. 5 ,, 30 , 16 Tropfen. 6. Ver su c h yore 26. 4. 15. Schweineniere: 0,23 g Samenalkaloid:250 ccm Blut- 6. Ver su c h yore 26. 4. 15. Schweineniere: 0,23 g Samenalkaloid:250 ccm Blut- mischung l cm hbscissenlii ngo ~---17 Sek 6. Ver su c h yore 26. 4. 15. Schweineniere: 0,23 g Samenalkaloid:250 ccm Blut- mischung, l cm hbscissenlii, ngo ~--- 17 Sek. mischung, l cm hbscissenlii, ngo ~--- 17 Sek. mischung, l cm hbscissenlii, ngo ~--- 17 Sek. 12 Uhr 20 Min. Anfang, 7 Uhr 25 Min. Ende. Beobachtungszeit: 7 Std. 5 Min. 19 Ueber das Paeoniaalkaloid. a) Tropfenzahl: a) Tropfenzahl: 11 Uhr -- Min. 16 Tropfen. 11 Uhr -- Min. 16 Tropfen. !2 ,, 14 ,, Zusatz yon 0,23 g S.-A. 12 ,, 30 ,, 10 Tropfon. 12 ,, 30 ,, 10 Tropfon. 3 , 15 , umgesohaltet, unvergiftetes Blur. 3 ,, 40 ,, 23 Tropfen. 5 , 20 ,, umgesohaltet, vergiftetes Blut. 6 ,, 15 ,, 17 Tropfon, unterbrochen durch freio Intervallo yon regelm~issig 23 und 24 Sek. Dauer. 7 ,, -- ,, umgeschaltet, unvergiftetes Blut. 7 ,, 40 ,, bis 8 Uhr 2 Min. 30 Tropfen. Bostehenbleiben der Periodiciti~t freier Intervalle, welohe abet auf 1t, 12, 13~ 14 Sek. abgokiirzt sind. b) Durchgelaufeno ~enge in Kubikzontimetern pro Minute: 3 Uhr 15 Min. bis 5 Uhr 20 Min. = 3,2 com. 5 ,, 20 ,, umgeschaltet, vergiftetes Blur. 5 ,, 20 ,, his 7Uhr 15Min.=2,3ccm. 7 ,, 15 , umgeschaltet, unvergiftetes Blut. 7 , 15 ,, bis 8Uhr 2Min.~2~3ccm. b) Durchgelaufeno ~enge in Kubikzontimetern pro Minute: 3 Uhr 15 Min. bis 5 Uhr 20 Min. = 3,2 com. 5 ,, 20 ,, umgeschaltet, vergiftetes Blur. 5 ,, 20 ,, his 7Uhr 15Min.=2,3ccm. 7 ,, 15 , umgeschaltet, unvergiftetes Blut. 7 , 15 ,, bis 8Uhr 2Min.~2~3ccm. b) Durchgelaufeno ~enge in Kubikzontimetern pro Minute: 3 Uhr 15 Min. bis 5 Uhr 20 Min. = 3,2 com. 5 ,, 20 ,, umgeschaltet, vergiftetes Blur. 5 ,, 20 ,, his 7Uhr 15Min.=2,3ccm. 7 ,, 15 , umgeschaltet, unvergiftetes Blut. 7 , 15 ,, bis 8Uhr 2Min.~2~3ccm. Wiihrend die sechs ersten F~illo ohne weiteres verstii.ndlich sind, habe ich bez/iglich des siebenten folgende Auseinandersetzungen zu maqhen. bIach der zweiten Einschaltung des vergiffeten Blutes zeigte sich eine Periodizitiit der Kurve, indem naeh einor Reihe yon Tropfen ein Intervall regelm~ssig wiederkehrte, in welehem kein Tropfen fiel und infolgedessen das Despretz'sche Signal im Ruhezustando verblieb. Es wurden 18 soleher Interval[e, also Tropfenpausen, mit dem Chronoskop gemessen und fest- gestellt, dass die Dauer dieser Pausen zwisehen 22 und 24 Sekunden schwankte, also fast genau gleich gross war. Nach Umschaltung des unvergifteten, guten Blutes blieb diese Periodieititt zwar bestehen, aber es wurden die Abschnitte der aufeinander folgenden Tropfen grSsser und 2* 20 Arnold Holste, im Gegensatze dazu die Pausen kiirzer, indem die tropfenfreien Zeiten zwischen 11 und 14 Sekunden betrugen. Die Nierensektion ergab eine tadellose Durchblutung. a) Tropfenzahl: Die Erkl/irung dieser Erscheinung beruht darauf, dass die dureh zahlreiehe Versuehe bewiesene kapillarverengernde Wirkung des Samenalkaloids zu einem fempor/iren Spasmus gefiihrt hat, dessen Dauer durch die Einwirkung des guten, unvergifteten Blutes zwar auf die H'/ilfte verringert, abet w/ihrend der Beobachtungszeit nicht ganz auf- gehoben werden konnte. g Die Untersuehung des Samenalkaloids hinsichtlich seiner Wirksamkeit auf den Meersehweinchenuterus wurde nach genau derselben Methode aus- gefiihrt, welche ich eingangs beim Wurzelalkaloid besehrieben habe. Mit dem virginalen Uterus wurden 9, mit dem Uterus yon Muttertieren, welche nicht tritchtig waren, aber einmal gcworfen hatten, 4, und mit gravidem Uterus 5, im ganzen also 18 einwandfreie Versuehe angestellt. Im nach- stehenden gebe ich einige Beispiele. A. Uterus im virginalen Zustande. 1. Versuoh vom 7. I. 15. Uterus in tote. Samenalkaloid 0,032 g:60 ccm Ringer. 10 Uhr vorm. Anfang, 7 Uhr 10 Min. abends Ende. Beobaohtungszeit: 9 Std. 10 Min. Die anf/inglichen Spontanbewegungen zeigon Exkursionen yon 2a/2--41/2 em L~inge. Naoh dem Zusatze yon 0,032 g S.-A. hSren sofort diese st~rkeren, peristaltischen Be- wegungen auf; start dessen erhebt sich die Kurvo 6 em fiher die Normallinie und bleibt bis zum Abbruche des Versuches auf dieser HShe als feingezackte Linie mit hSehstens 1 mm grossen Exkursionen (Tetanus). 1. Versuoh vom 7. I. 15. Uterus in tote. Samenalkaloid 0,032 g:60 ccm Ringer. 10 Uhr vorm. Anfang, 7 Uhr 10 Min. abends Ende. Beobaohtungszeit: 9 Std. 10 Min. Die anf/inglichen Spontanbewegungen zeigon Exkursionen yon 2a/2--41/2 em L~inge. Naoh dem Zusatze yon 0,032 g S.-A. hSren sofort diese st~rkeren, peristaltischen Be- wegungen auf; start dessen erhebt sich die Kurvo 6 em fiher die Normallinie und bleibt bis zum Abbruche des Versuches auf dieser HShe als feingezackte Linie mit hSehstens 1 mm grossen Exkursionen (Tetanus). 2. Versuch vom 8. 1.15. Uterus in tote. Samenalkaloid 0,016 g : 70 ecru Ringer. 12 Uhr 50 Min. mittags Anfang, 6 Uhr 35 Min. abends Ende. Beobaohtungszoit: 5 Std. 45 Min. 2. Versuch vom 8. 1.15. Uterus in tote. Samenalkaloid 0,016 g : 70 ecru Ringer. 12 Uhr 50 Min. mittags Anfang, 6 Uhr 35 Min. abends Ende. Beobaohtungszoit: 5 Std. 45 Min. 2. Versuch vom 8. 1.15. Uterus in tote. Samenalkaloid 0,016 g : 70 ecru Ringer. 12 Uhr 50 Min. mittags Anfang, 6 Uhr 35 Min. abends Ende. Beobaohtungszoit: 5 Std. 45 Min. 12 Uhr 50 Min. bis 2 Uhr 5 Min. Spontanbewegungen. B. Uterus yon nieht tr~chtigen Muttortieron. 1.Versueh veto 28.9.14. Uterus in tote. Samenalkaloid 0,190 g : 60 ecru Ringer. 11 Uhr Anfang, 6 Uhr Ende. Beobaehtungszeit: 7 Std. 11 Uhr Spontanbewegungen. 1 Uhr 55 Min. Zusatz yon 0,190 g S.-A. Die an- f~nglioh maximal 1/2 cm grossen Exkursionen stiegen 3 Uhr 25 Min. his zu 3 em L~inge~ a) Tropfenzahl: we die HShe der Kurve fiber tier Normallinie 8 on] betrug. Bis 6 Uhr stieg die Kurve auf 101/2 ore. 2. Vers u oh veto 19.1.15. Uterus in tote. Saraenalkaloid 0,020 g : 60 corn Ringer. ll Uhr 15 Min. vorm. Anfang, 7 Uhr 30 Min. abends Ende. Beobaohtungszeit: 8 Std. 15 Min. 11 Uhr 15 Min. Spontanbewegungen. 12 Uhr 15 Min. KurvenhShe 91/2 ore; nach Zusatz Steigerung auf 13 era, 4 Uhr 21 Min. auf 15 cm fiber die Normallinie; Exkursionen des Schreibhebels 6 ore. 3. Versuoh vora 22. I. 15. Uterus in tote. Samenalkaloid 0,040 g:60 corn Ringer. ll Uhr 15 Min. vorm. Anfang, 7 Uhr 30 ]Vlin. Ende. Beobaohtungszeit: 8 Std. 15 Min. ll Uhr 15 Min. Spontanbewegungen. Maximale HShe tier Kurve fiber der Normal- linie 11 cm; naoh Zusatz 12 Uhr 35 Min. 131/2 cm; diese H5he bleibt mit kleinen Schwankungen bis zum hbbruche des Versuches bestehen. Die Exkursionen schwanken zwischen 0,1 cra 2 Uhr (Tetanus) und 1 era bis zum Ende. Bei dieser Serie von Versuchen wurde zum Schluss die vergiftete RingerlSsung mit dem gleichen Quantum unvergiftcter ausgewechselt; der Erfolg war, dass die anftingliohe Normalzeichnung wiederkehrte. Bei dieser Serie von Versuchen wurde zum Schluss die vergiftete RingerlSsung mit dem gleichen Quantum unvergiftcter ausgewechselt; der Erfolg war, dass die anftingliohe Normalzeichnung wiederkehrte. C. Uteras yon trtichtigen Tieren. 1. Versuch vora 29.9. 14. Uterus in tote. Rechtes Horn gravid, linkes Horn nioht gravid, schreibt. Saraenalkaloid 0090 g : 60 ccm Ringer. 1. Versuch vora 29.9. 14. Uterus in tote. Rechtes Horn gravid, linkes Horn nioht gravid, schreibt. Saraenalkaloid 0090 g : 60 ccm Ringer. g , g g ll Uhr Anfang, 6 Ubr 40 Min. Ende. Beobaohtungszeit: 7 Std. 40 Min. ll Uhr Anfang, 6 Ubr 40 Min. Ende. Beobaohtungszeit: 7 Std. 40 11 Uhr Kurve 5 era fiber der Norraallini% Exkursionen bis zu 1/2 era. 12 Uhr Zusatz von 0,190g S.-A.; sofortiges Ansteigen der Kurve maximal bis 81/2 era bei 2 Uhr 25 Min. ;von 2 Uhr 30 Min. his zura Ende auf 91/2 om mit kleinen Sohwankungen, sowie Vergr5sserung der Exkursionen bis zu 4 ore. 11 Uhr Kurve 5 era fiber der Norraallini% Exkursionen bis zu 1/2 era. 12 Uhr Zusatz von 0,190g S.-A.; sofortiges Ansteigen der Kurve maximal bis 81/2 era bei 2 Uhr 25 Min. ;von 2 Uhr 30 Min. a) Tropfenzahl: Exkursionen bis zu 1 em L~ingo; die Kurve bewegt sich 5--71/2 cm fiber der Normallinie. ~aeh dem Zusatze yon 0,016 g S.A. erhebt sich die Kurvo auf 111/2, 13, 14 cm um 4 Uhr 25 Min. naehm, und zeigt bei Abbrueh des Versuohes noch eine HShe yon ll era. Die Exkursionen dieses Absehnittes zeigen Sohwankungen bis zu 11/2 cm L~ngo. 12 Uhr 50 Min. bis 2 Uhr 5 Min. Spontanbewegungen. Exkursionen bis zu 1 em L~ingo; die Kurve bewegt sich 5--71/2 cm fiber der Normallinie. ~aeh dem Zusatze yon 0,016 g S.A. erhebt sich die Kurvo auf 111/2, 13, 14 cm um 4 Uhr 25 Min. naehm, und zeigt bei Abbrueh des Versuohes noch eine HShe yon ll era. Die Exkursionen dieses Absehnittes zeigen Sohwankungen bis zu 11/2 cm L~ngo. 3. Ve r s u e h veto 1"2.1.15. Uterus in tote. Samenalkaloid 0,008 g : 80 cem Ringer. 12 Uhr mittags Anfang, 7 Uhr abends Ende. Beobachtungszeit: 7 Std. , g g 12 Uhr mittags Anfang, 7 Uhr abends Ende. Beobachtungszeit: 7 Std. Die Spontanbewegungen zeigen Exkursionon his zu 21/2 cm. Naeh dora Zusatze yon 0,008 g S.-A. vergrSssern sieh diese Zeichnungen bis zu 71/2 cm. Die HShe der Kurve betr~gt anf~nglich maximal 9 cm und steigt his gegen Endo des Vorsuehes auf 121/2 em fiber die Normallinie. 4. Ve rs u eh veto 15.1.15. Uterus in tote. Samenalkaloid 0,0176 g : 60 ccm Ringer. Anfang am 15. I. 15. ll Uhr 45 Min. vorm., Ende am 16.1.15. 5 Uhr 45 Min. nachm. Beobachtungszeit: 30 Std. 11 Uhr 45 Min. Spontanbewegungen. Kurve 7 cm fiber der Normallinie, senkt sioh gegen 1 Uhr his zu 21/3 cm. Nach dem Zusatze des S.-A. steigt die Kurvo his zu 121/2 cm HSho und zeigt bis 8 Uhr abends Sehwankungen zwischen 101/2 und 4 cm. Die Maximalerhebung am zweiten Tage 12 Uhr mittags betrug 12 cm. B. Uterus yon nieht tr~chtigen Muttortieron. 1.Versueh veto 28.9.14. Uterus in tote. Samenalkaloid 0,190 g : 60 ecru Ringer. 11 Uhr Anfang, 6 Uhr Ende. Beobaehtungszeit: 7 Std. 11 Uhr Spontanbewegungen. 1 Uhr 55 Min. Zusatz yon 0,190 g S.-A. Die an- f~nglioh maximal 1/2 cm grossen Exkursionen stiegen 3 Uhr 25 Min. his zu 3 em L~inge~ 21 Ueber das Paooniaalkaloid. we die HShe der Kurve fiber tier Normallinie 8 on] betrug. Bis 6 Uhr stieg die Kurve auf 101/2 ore. a) Tropfenzahl: his zura Ende auf 91/2 om mit kleinen Sohwankungen, sowie Vergr5sserung der Exkursionen bis zu 4 ore. 2. Versuch vora 30. 9. t4. Uterus in tote. Recbtes Horn gravid, linkes Horn nicht gravid, sohreibt. Samenalkaloid 0,285 g : 60 ccra Ringer. 2. Versuch vora 30. 9. t4. Uterus in tote. Recbtes Horn gravid, linkes Horn nicht gravid, sohreibt. Samenalkaloid 0,285 g : 60 ccra Ringer. 11 Uhr vorra. Anfang, 6 Uhr 15 Min. abends Ende. Beobachtungszeit: 7 Std. 15 Min. 11 Uhr KurvenhShe 61/2 cm, 12 Uhr 10 Min. Zusatz yon 0,285 g S.-A. Erhebung maximal bei 2 Uhr 15 Min. 10 cm. Exkursionen im Anfange yon 2 tara, naoh Zusatz bis zu 1 era. 11 Uhr KurvenhShe 61/2 cm, 12 Uhr 10 Min. Zusatz yon 0,285 g S.-A. Erhebung maximal bei 2 Uhr 15 Min. 10 cm. Exkursionen im Anfange yon 2 tara, naoh Zusatz bis zu 1 era. 3. Versuch veto 25.1. 15. Uterus in tote. Ein Horn gravid, das andere Horn nicht gravid, schreibt. Samenalkaloid 0,024 g : 60 eora Ringer. 3. Versuch veto 25.1. 15. Uterus in tote. Ein Horn gravid, das andere Horn nicht gravid, schreibt. Samenalkaloid 0,024 g : 60 eora Ringer. 11 Uhr 15Min. vorm. Anfang, 8Ubrl5Min. abends Ende. Boobachtungszeit: 9 Std. g, g 12 Uhr KurvenhShe 41/2 era; Zusatz yon 0,024 g S.-A. Steigerung dor t(urve maximal 3 Uhr 15 Min. auf I01/2 era, 4 Ubr 25 Min. auf 11 era fiber die Normallinie; annghernd gleiehe HShe his zura Sehluss. Anf~ngliohe Exkursionen 1 ram, naeh Zusatz bis 1 era. 12 Uhr KurvenhShe 41/2 era; Zusatz yon 0,024 g S.-A. Steigerung dor t(urve maximal 3 Uhr 15 Min. auf I01/2 era, 4 Ubr 25 Min. auf 11 era fiber die Normallinie; annghernd gleiehe HShe his zura Sehluss. Anf~ngliohe Exkursionen 1 ram, naeh Zusatz bis 1 era. Durch diese Protokolle wird der Beweis erbracht, class das Samen- alkaloid den Meerschweinchen-Uterus im virginalen und graviden Zustande, sowie denjenigen nieht tr~chtiger Muttertiere in dem Sinne einer energischen Steigerung des Tonus und einer Kr/fftigung der einzelnen peristaltischen Bewegungen beeinflusst. Vergleiehende Untersuehungen, welche ieh in grSsserer Anzahl mit Seeale-Pr~paraten des Handels yon anerkannt sehr hohem Gehalte an wirksamen Prinzipien angestellt babe, ergaben als Resultat, dass das Samenalkaloid seine Wirkung in ganz gleicher, aber etwas sehw~cherer Weise entfaltet. 22 Arnold Holsto, Uobor das Paeoniaalkaloid. 1) Sioho obon. a) Tropfenzahl: Meine Experimente erbringen also den Beweis, class das Samen- alkaloid, abgesehen yon der zuletzt beschriebenen Utoruswirkung, die Nierenkapillaren verengert und die Gerinnungsffihigkeit des Blutes im lebenden KSrper steigert, ohne eine nachweisbare Einwirkung auf den Blutdruck oder das Herz des Warmbtiiters selber zu besitzen. Deshalb fiihle ich mica auf Grund meiner Feststellungen berechtigt, das Samen- alkaloid oder dessen Salze zur klinischen Verwendung zu empfehlen. Die in Betracht kommenden Indikationen diirften Uterusblutungen aus ver- schiedenen Ursachen sein, welche nicht mit schweren anatomischen Ver- /inderungen verkniipft sind; ferner Blutungen aus anderen 0rganen, ins- besondero Nieren- und Lungenblutungen. Nach Pick's 1) Untersuchungen daft man allerdings nicht die fiir die Nierenkapillaren yon mir erwiesene Tatsache bei den Lungenkapillaren ebenfalls als vorhallden annehmen, abet selbst bei dem eventuellen Fehlen der vasokonstriktorischen Eigen- schaft des Paeoniaalkaloids in dem letzteren 0rgane wiirde dot Einfiuss dieser Substanz auf die Blutgerinnung die Verwendung auch gegen Lungcnblutungen aussichtsvoll erscheinen lassen. Gerade bei dieser Indikation ist es sear wertvoll, dass das Samenalka]oid Herz und Blut- druck absolut nicht beeinfiusst, andererseits a~r die Gerinnungsfiihigkeit des Blutes in betr/ichtlicher Weise und langdauernd erhtiht. ~ ~ s ~ ~ J J ~ jjt jzj~~ ~ S ~ J / ~ ~ r j j~/j J j S I k J q J 4 k J q J 4 k J q J 4 ~ ~ s ~ ~ J J ~ jjt jzj~~ ~ S ~ J / ~ ~ r j j~/j J j S I k J q J 4 J 4 J 4 a ~oto, virg;~al a]o 0.'308 9 r ~[ ,""~/ ¢I I P~ / I ' ,Jr,., ~' / 'r i 4 ~" [ I ', ' ~ ~' ! '~ '~" I ~ ~, ", t '~''i I '\ ufver~uch
https://openalex.org/W4387320440	https://jurnal.faperta-unras.ac.id/index.php/JS/article/download/112/121	Indonesian	null	PENYUSUNAN LAPORAN KEUANGAN BERDASARKAN SAK EMKM PADA TOKO BUSA DAN KAIN SOFA AL RAZAAK AIR BANG	Jurnal Saintifix/Jurnal Saintifik	2,021	cc-by-sa	3,167	PREPARATION OF FINANCIAL STATEMENTS BASED ON SAK EMKM IN FOAM AND SOFA FABRIC SHOPS AL RAZAAK AIR BANG Tuti Hermelinda1 Upi Niarti2 Paddery3 1,2,3 Program Studi Akuntansi, Politeknik Raflesia Korespondensi e-mail : tutihermelinda.polraf@gmail.com ARTICLE HISTORY Tujuan penenlitian ini adalah menerapkan penyusunan laporan keuangan berdasarkan SAK EMKM pada Toko Busa dan Kain Sofa Al Razaak dan mengetahui kendala kendala apa saja yang dihadapi dalam pembuatan laporan keuangan Toko Busa dan Kain Sofa Al Razaak dalam menerapkan SAK EMKM. Analisis dalam membuat laporan keuangan ini menggunakan teknik penelitian deskriptif dengan pendekatan kualitatif. Teknik analisis data deskriptif merupakan teknik analisis yang dipakai untuk menganalisis data dengan mendeskripsikan atau menggambarkan data-data yang sudah dikumpulkan seadanya tanpa ada maksud membuat generalisasi dari hasil penelitian. Berdasarkan hasil penyusunan laporan keuangan pada Toko Busa Dan Kain Sofa Al Razaak,Penyusunan laporan keuangan berdasarkan SAK EMKM pada Toko Busa Dan Kain Sofa Al Razaak diawali dengan pengumpulan informasi kegiatan dalam usaha melalui wawancara, dilanjutkan pengamatan langsung terhadap buku harian yang berisi catatan transaksi pembelian dan penjualan. Dilanjutkan dengan penyusunan laporan keuangan. Setelah dilakukan penyusunan laporan keuangan dapat di ambil kesimpulan bahwa Toko Al Razaak memproleh penjualan berkisar Rp. 89.650.000 Dan memperoleh laba sebelum PPH berkisar Rp. 25.004.000 Jadi toko busa dan kain al razaak memperoleh ratusan juta rupiah pertahun. KEYWORDS Financial Statements, SAK EMKM ABSTRACT The purpose of this research is to apply the preparation of financial reports based on SAK EMKM on Al Razaak Foam and Fabric Sofa Shop and find out what obstacles are encountered in preparing the financial reports of Al Razaak Foam and Fabric Shop in implementing SAK EMKM. The analysis in making these financial reports uses descriptive research techniques with a qualitative approach. Descriptive data analysis technique is an analytical technique used to analyze data by describing or describing data that has been collected soberly without any intention of making generalizations from the research results. Based on the results of compiling financial reports at the Al Razaak Sofa Foam and Fabric Shop, the preparation of financial statements based on SAK EMKM at the Al Razaak Sofa Foam and Fabric Shop begins with collecting information on business activities through interviews, followed by direct observation of the diary which contains records of purchase transactions and sale. Followed by the preparation of financial reports. After compiling the financial statements, it can be concluded that the Al Razaak Store earns sales of around Rp. 89,650,000 and earned a profit before PPH of around Rp. 25,004,000 So the foam and fabric shop al razaak earns hundreds of millions of rupiah per year. ABSTRAK ARTICLE HISTORY Received [20 April 2021] Revised [29 April 2021] Accepted [16 May 2021] Laporan keuangan sangat membantu dalam mengontrol kegiatan perusahaan. Fungsi laporan keuangan yaitu memberikan informasi mengenai kondisi keuangan perusahaan dalam kondisi baik ataupun buruk. Jika laporan keuangan tidak dikelola dengan benar, maka akan berpengaruh terhadap keputusan yang akan diambil pihak manajemen. Laporan keuangan ialah paparan data yang mengungkapkan atau mendeskripsikan keadaan keuangan entitas yang bisa dijadikan gambaran kapasitas atau kinerja keuangan entitas. Laporan keuangan bertujuan untuk menyajikan infromasi keuangan entitas yang berguna untuk hampir sebagian besar pemakai informasi dalam pemungutan keputusan ekonomi. Dalam PSAK 1 (2015:1.3) bagian dari keseluruhan laporan keuangan yaitu laporan posisi keuangan pada akhir periode, laporan laba rugi dan penghasilan menyeluruh selama periode, laporan perubahan ekuitas selama periode, laporan arus kas selama periode, catatan atas laporan keuangan yang berisi ringkasan kebijakan akuntansi yang signifikan dan informasi lainnya, dan laporan keuangan pada awal periode terdekat sebelumnya ketika entitas mengimplementasikan suatu kebijakan akuntansi secara peninjauan kembali atau membuat penyajian kembali pos-pos laporan keuangan, atau ketika entitas mereklasifikasi pos-pos dalam laporan keuangannya. UMKM (Ningtyas, 2018) merupakan upaya produktif milik orang atau badan usaha perorangan, bukan cabang perusahaan yang dimiliki usaha yang memenuhi tolak ukur usaha kecil atau menengah. (D S d Ak i K 2016) d l h d d l k ik k i k d l h l i i ik g p y g y g g (Dewan Standar Akuntansi Keuangan, 2016) adalah standar dalam praktik akutansi keuangan, dalam hal ini mikro, kecil dan menengah yang serupa dengan kekayaan usaha yang memiliki limit, yang dipunyai seorang atau sekerumun orang dalam menjalankan sebuah upaya produktif. g j p y p Laporan Keuangan berdasarkan SAK-EMKM mencakup laporan posisi keuangan, laporan laba rugi dan catatan atas laporan keuangan. Dalam SAK EMKM dijelaskan bahwa entitas harus mengidentifikasi secara jelas laporan keuangan dan menunjukkan informasi berikut, yaitu : 1) Nama entitas yang menyusun laporan keuangan. 2) Tanggal akhir periode laporan keuangan 3) Rupiah sebagai mata uang penyajian. 4) Pembulatan angka digunakan dalam penyajian laporan keuangan. Pada laporan posisi keuangan (Dewan Standar Akuntansi Keuangan, 2016) memuat aset, liabilitas dan ekuitas pada entitas. Aset yang dimaksud yaitu sumber daya yang dipunyai entitas dan yang dinantikan manfaat ekonominya. Sedangkan liabilitas yaitu kewajiban entitas yang berasal dari dampak kejadian masa lalu dengan cara penyelesaiannya membawa dampak pada arus keluar dari aset entitas. PENDAHULUAN p perusahaan baik skala kecil maupun besar pasti memerlukan sebuah laporan keuangan. Latar belakang setiap perusahaan baik skala kecil maupun besar pasti memerlukan sebuah laporan k JURNAL SAINTIFIK (Multi Science Journal), Vol. 19 No. 2 May 2021 page: 65 – 70\| 65 JURNAL SAINTIFIK (Multi Science Journal), Vol. 19 No. 2 May 2021 page: 65 – 70\| 66 Liabilitas atau kewajiban dapat berupa kewajiban hukum seperti kontrak mengikat dan kewajiban konstruktif yang timbul dari praktik masa lalu dan harapan yang kuat dari pihak entitas akan melakukan kewajiban tersebut. Lalu ekuitas yaitu hak atas aset entitas sesudah dikurangi seluruh kewajibannya. y g j y Pada laporan laba rugi (Dewan Standar Akuntansi Keuangan, 2016) memuat pendapatan, beban-beban dan beban pajak. Pendapatan adalah penghasilan yang datang dari operasional entitas yang normal. Sedangkan beban ialah penurunan manfaat ekonomi dalam bentuk arus keluar. Beban pajak penghasilan yang digunakan berdasarkan ketentuan perpajakan yang berlaku. Di Indonesia, pajak penghasilan untuk UMKM berlaku yaitu Pajak Penghasilan Final dengan pengenaan pada omzet perjualan sebesar 0,5%. Dan pada catatan atas laporan keuangan (Dewan Standar Akuntansi Keuangan, 2016) memuat kebijakan akuntansi yang diterapkan, ikthisar akuntansi dan perincian akun lainya. Catatan atas laporan keuangan mengatur prinsip yang melandasi informasi yang disajikan dan didalamnya harus saling merujuk silang pada informasi yang bersangkutan dalam catatan atas laporan keuangan. Tujuan laporan keuangan berdasarkan SAK EMKM yaitu untuk menyediakan informasi posisi keuangan dan kinerja suatu entitas yang bermanfaat bagi para pengguna laporan keuangan dalam pengambilan keputusan ekonomi. Pengguna laporan keuangan meliputi penyedia sumber daya untuk entitas, seperti kreditur maupun investor. Selain itu, laporan keuangan juga bertujuan untuk menunjukkan pertanggungjawaban manajemen atas sumber daya yang telah dipercayakan kepadanya (IAI, 2016: 3). y y g p y p y Laporan keuangan tidak hanya mengenai hasil tetapi juga mengenai proses penyusunannya. Laporan keuangan menggunakan Standar Akuntansi Keuangan (SAK) yang berbeda-beda untuk setiap tingkatan bisnis. Standar Akuntansi Keuangan Entitas Mikro, Kecil dan Menengah (SAK EMKM) adalah standar untuk Usaha Mikro, Kecil dan Menengah (UMKM). Toko Busa dan Kain Sofa Al Razaak adalah salah satu UMKM yang ada di Curup. Toko Busa dan Kain Sofa Al Razaak adalah usaha dagang yang menjual berbagai macam kebutuhan pembuatan sofa. Toko ini berdiri pada tahun 2019 dan masih terus berkembang sampai sekarang. Toko ini juga melayani pembelian dengan jumlah banyak atau partai tetapi bisa juga satuan atau eceran. Toko ini berlokasi di jalan Stadion Air Bang Kec.Curup Tengah. Berdasarkan pengamatan awal yang dilakukan penulis, pihak toko memiliki laporan keuangan. Namun belum sesuai dengan Standar Akuntansi. Pihak toko melakukan pencatatan terkait transaksi yang terjadi dalam aktivitas usahanya sehari-hari. Pada umumnya, ada dua transaksi yang terjadi yaitu transaksi penerimaan kas dan pengeluaran kas. Setelah ditelusuri lebih dalam, ternyata pihak toko hanya melakukan pencatatan terhadap transaksi penjualan. Bukti transaksi berupa nota penjualan sudah tersedia JURNAL SAINTIFIK (Multi Science Journal), Vol. 19 No. 2 May 2021 page: 65 – 70\| 66 METODE PENELITIAN Penelitian ini menggunakan teknik penelitian deskriptif dengan pendekatan kualitatif. Teknik analisis data deskriptif merupakan teknik analisis yang dipakai untuk menganalisis data dengan mendeskripsikan atau menggambarkan data-data yang sudah dikumpulkan seadanya tanpa ada maksud membuat generalisasi dari hasil penelitian. Penelitian kualitatif disini pada pengolahan data laporan keuangan sehingga data dapat menjadi sebuah laporan keuangan yang bermanfaat pada UMKM, SAK EMKM pada penelitian ini digunakan sebagai alat analisis untuk peneliti. SAK EMKM dijadikan standar akuntansi keuangan untuk digunakan dalam penyusunan laporan Keuangan pada usaha kecil menengah yang kebanyakan belum menerapkan standar dalam pencatatan keuangan Diawali dengan mempelajari SAK EMKM lalu dianalisis supaya mengetahui bagaimana penyusunan laporan sesuai dengan SAK EMKM selanjutnya melakukan perbandingan pada unsur laporan Keuangan sesuai SAK EMKM. Langkah terakhir yaitu menerapkan SAK EMKM sebagai pedoman untuk menyusun laporan keuangan. JURNAL SAINTIFIK (Multi Science Journal), Vol. 19 No. 2 May 2021 page: 65 – 70\| 67 Neraca saldo Setelah posting akun di buku besar selanjutnya membuat neraca saldo Toko Busa dan Kain sofa Al Razaak periode oktober – desember 2021, adapun data yang diperoleh adalah sebagai berikut: Tabel 1. Neraca Saldo Periode Oktober – desember 2021 (Dalam Rupiah) TOKO AL RAZAAK NERACA SALDO UNTUK PERIODE YANG BERAKHIRAN 31 DESEMBER 2021 NAMA AKUN DEBIT KREDIT Kas 120,194,000 Persediaan Barang Dagang 85,099,000 Perlengkapan Toko 500,000 Sewa Di Bayar Dimuka 15,000,000 Peralatan Toko 1,000,000 Kendaraan 60,000,000 Akm Peny Peralatan 437,493 Akm Peny Kendaraan 3,000,000 Utang Dagang 0 Modal Tn. Dimas 253,225,006 Penjualan 89,650,000 Hpp 59,651,000 Beban Gaji Dan Upah 3,000,000 Beban Listrik Dan Air 306,000 Beban Penyusutan Peralatan Toko 62,499 Beban Penyusutan Kendaraan 1,500,000 TOTAL 346,312,499 346,312,499 Tabel 1. Neraca Saldo Periode Oktober – desember 2021 (Dalam Rupiah) JURNAL SAINTIFIK (Multi Science Journal), Vol. 19 No. 2 May 2021 page: 65 – 70\| 67 p g Tabel 2. Laporan Laba Rugi Periode Oktober – desember 2021 (Dalam Rupiah) TOKO AL RAZAAK LAPORAN LABA RUGI UNTUK PERIODE YANG BERAKHIR PADA 31 DESEMBER 2021 Pendapatan Penjualan 89,650,000 HPP 59,651,000 JUMLAH PENDAPATAN 29,999,000 Beban Beban gaji dan upah 3,000,000 beban penyusutan peralatan toko 62,499 beban penyusutan kendaraan 1,500,000 beban listrik dan air 306,000 JUMLAH BEBAN 4,868,499 LABA SEBELUM PAJAK PENGHASILAN 25,130,501 Beban Pajak Penghasilan 125,653 LABA SETELAH PAJAK PENGHASILAN 25,004,848 TOKO AL RAZAAK LAPORAN LABA RUGI UNTUK PERIODE YANG BERAKHIR PADA 31 DESEMBER 2021 Pendapatan Penjualan 89,650,000 HPP 59,651,000 JUMLAH PENDAPATAN 29,999,000 Beban Beban gaji dan upah 3,000,000 beban penyusutan peralatan toko 62,499 beban penyusutan kendaraan 1,500,000 beban listrik dan air 306,000 JUMLAH BEBAN 4,868,499 LABA SEBELUM PAJAK PENGHASILAN 25,130,501 Beban Pajak Penghasilan 125,653 LABA SETELAH PAJAK PENGHASILAN 25,004,848 Sumber data diolah 2021 TOKO AL RAZAAK CATATAN ATAS LAPORAN KEUANGAN UNTUK PERIODE DESEMBER 2021 1. UMUM Usaha Toko Busa Dan Kain Al Razaak beralamat di jl. Stadion Air Bang usaha ini berdiri pada tahun 2019 termasuk golongan usaha kecil dengan laba per Oktober - Desember 2021 Rp. 25,004,848 2. IKHTISAR KEBIJAKAN AKUNTANSI PENTING a. Pernyataan kepatuhan Melalui laporan keuangan disusun menggunakan Standar Akuntansi Keuangan Entitas Mikro, Kecil, dan Menengah. b. Dasar Penyusunan Dasar penyusunan laporan keuangan merupakan biaya historis dan menggunakan asumsi dasar akrual. Untuk mata uang penyajian yang digunakan untuk penyusunan laporan keuangan adalah Rupiah. c. Aset Tetap Aset tetap dicatat sebesar biaya perolehannya jika aset tersebut dimiliki secara hukum oleh entitas. Aset tetap disusutkan menggunakan metode garis lurus dengan taksiran umur ekonomis sebagai berikut : Peralatan = 4 tahun Sumber data diolah 2021 1. UMUM g g g p p 2. IKHTISAR KEBIJAKAN AKUNTANSI PENTING a. Pernyataan kepatuhan Melalui laporan keuangan disusun menggunakan Standar Akuntansi Keuangan Entitas Mikro, Kecil, dan Menengah. b. Dasar Penyusunan Dasar penyusunan laporan keuangan merupakan biaya historis dan menggunakan asumsi dasar akrual. Untuk mata uang penyajian yang digunakan untuk penyusunan laporan keuangan adalah Rupiah. c. Aset Tetap Aset tetap dicatat sebesar biaya perolehannya jika aset tersebut dimiliki secara hukum oleh entitas. Aset tetap disusutkan menggunakan metode garis lurus dengan taksiran umur ekonomis sebagai berikut : Peralatan = 4 tahun b. Dasar Penyusunan Dasar penyusunan laporan keuangan merupakan biaya historis dan menggunakan asumsi dasar akrual. Untuk mata uang penyajian yang digunakan untuk penyusunan laporan keuangan adalah Rupiah. c. Aset Tetap Aset tetap dicatat sebesar biaya perolehannya jika aset tersebut dimiliki secara hukum oleh entitas. Aset tetap disusutkan menggunakan metode garis lurus dengan taksiran umur ekonomis sebagai berikut : Peralatan = 4 tahun JURNAL SAINTIFIK (Multi Science Journal), Vol. 19 No. 2 May 2021 page: 65 – 70\| 68 3. PENJELASAN AKUN-AKUN PADA LAPORAN POSISI KEUANGAN J a. Kas Kas sebesar Rp. 120,194,000 merupakan kas per 31 desember 2021 yang ada pada Usaha toko busa dan kain al Razaak. b. Persediaan Pesediaan terdiri dari barang siap dijual Rp. 85,099,000 c. Peralatan Peralatan diperoleh sebesar Rp. 1.000.000 sebagai berikut : Gunting khusus Rp. 500.000 Palu Rp. 100.000 Tang biasa Rp. 100.000 Gunting biasa Rp. 300.000 Total Rp. 1.000.000 Kas sebesar Rp. 120,194,000 merupakan kas per 31 desember 2021 yang ada pada Usaha toko busa dan kain al Razaak. b. Persediaan d. Kendaraan Kendaraan diperoleh sebesar Rp. 60.000.000 merupakan kendaraan yang dimiliki Usaha toko busa dan Kendaraan diperoleh sebesar Rp. 60.000.000 merupakan kendaraan yang dimiliki Usaha toko busa dan kain al Razaak yaitu terdiri dari 1 buah mobil. 1) Modal kain al Razaak yaitu terdiri dari 1 buah mobil. 1) Modal Modal sebesar Rp. 253,225,006 berasal dari modal milik pribadi. kain al Razaak yaitu terdiri dari 1 buah mobil. 1) Modal Modal sebesar Rp. 253,225,006 berasal dari modal milik pribadi. Kendaraan = 10 tahun d. Pengakuan Pendapatan dan Beban d. Pengakuan Pendapatan dan Beban Pendapatan penjualan dan beban akan diakui ketika terjadi pemasukan dan pengeluaran kas. P j k g p Pendapatan penjualan dan beban akan diakui ketika terjadi pemasukan dan pengeluaran kas. e. Pajak Pajak penghasilan mengikuti ketentutan perpajakan yang berlaku di Indonesia. Tarif PPh final UMKM yang ditetapkan oleh pemerintah sesuai dengan Peraturan Pemerintah (PP) Nomor 23 Tahun 2018 adalah sebesar 0,5%. 4. PENJELASAN AKUN-AKUN PADA LABA RUGI a. Pendapatan Penjualan Pendapatan penjualan diakui ketika tagihan diterbitkan atau pengiriman dilakukan kepada pelanggan. Beban diakui saat terjadi. b. Pajak Penghasilan Pajak penghasilan mengikuti ketentuan perpajakan yang berlaku di Indonesia. c. Beban usaha dan beban lain Beban yang terdiri dari penyusutan kendaraan dan peralatan. Pendapatan penjualan diakui ketika tagihan diterbitkan atau pengiriman dilakukan kepada pelanggan eban diakui saat terjadi. Pajak penghasilan mengikuti ketentuan perpajakan yang berlaku di Indonesia. Beban yang terdiri dari penyusutan kendaraan dan peralatan. Beban yang terdiri dari penyusutan kendaraan dan peralatan. JURNAL SAINTIFIK (Multi Science Journal), Vol. 19 No. 2 May 2021 page: 65 – 70\| 69 KESIMPULAN Toko busa dan kain sofa al razaak belum menerapkan penyusunan laporan berdasarkan SAK EMKKM. Penyusunan laporan keuangan berdasarkan SAK EMKM pada Toko Busa Dan Kain Sofa Al Razaak diawali dengan pengumpulan informasi kegiatan dalam usaha melalui wawancara, dilanjutkan pengamatan langsung terhadap buku harian yang berisi catatan transaksi pembelian dan penjualan. Dilanjutkan dengan penyusunan laporan keuangan. Setelah dilakukan penyusunan laporan keuangan dapat di ambil kesimpulan bahwa Toko Al Razaak memproleh penjualan berkisar Rp. 126.845.000 Dan memperoleh laba sebelum PPH berkisar Rp. 105.595.000. Jadi toko busa dan kain al razaak memperoleh ratusan juta rupiah pertahun. Berdasarkan penelitian yang dilakukan, kendala kendala yang dihadapi pada saat melakukan penyusunan laporan keuangan pada Toko Busa Dan Kain Al Razaak diantaranya yaitu Toko Al Razaak hanya melakukan pencatatan transaksi keuangan sebatas pada pemasukan dan pengeluaran saja, sehingga penerapan akuntansi berdasarkan SAK EMKM pada usaha ini masih sangat lemah. Pencatatan dan penyususunan laporan menjadi kendala tesendiri bagi pemilik, hal ini disebabkan karna toko Al Razaak ini lebih berfokus pada kegiatan penjualan sehingga pencatatan dan pelaporan seringkali terabaikan.Berdasarkan kesimpulan yang telah dijelaskan diatas, adapun saran yang dapat diberikan yaitu sebagai berikut : a. Kiranya pemilik toko al razaak memberikan perhatian terhadap aspek keuangan usahanya dan meningkatkan pemahaman akan pentingnya laporan keuangan. b. Ditingkatkan serta disiplin dalam melakukan pencatatan keuangan serta mendokumentasikan bukti-bukti transaksi. Apabila memungkinkan melakukan pembelajaran yang lebih untuk melakukan pencatatan keuangan ( laporan keuangan JURNAL SAINTIFIK (Multi Science Journal), Vol. 19 No. 2 May 2021 page: 65 – 70\| 69 DAFTAR PUSTAKA Abdillah, W. 2018. Metode Penelitian Terpadu Sistem Informasi Pemodelan Teoritis, Pengukuran dan Pengujian Statistis. (R. I. Utami, Ed.). Yogyakarta: Penerbit Andi. gy Alimul, Aziz Hidayat. 2001. Menyusun Instrumen Penelitian dan Uji Validitas Reliabilitas. Surabay Books Publish. Ari Nuvitasari1, N. C. 2019. Implementasi SAK EMKMSebagai Dasar Penyusunan Laporan Keuangan Usaha Mikro Kecil dan Menengah (UMKM). nternational Journal of Social Science and Business. Volume 3, Number 3, Tahun 2019, pp. 341-347, 342-345. , , pp , Aufar. 2014. Definisi UMKM menurut Kementrian Koperasi dan UMKM dalam Aufar (2014:8). Afriansyah, B. (2019). Penyusunan Laporan Pertanggung Jawaban (LPJ) Bendahara Pengeluaran p Pelayanan Perbendaharaan Negara Curup. Jurnal Ilmiah Raflesia Akuntansi, 5(1), 18-23. y g Afriansyah, B., Niarti, U., & Hermelinda, T. (2021). ANALISIS IMPLEMENTASI PENYUSUNAN LAPORAN KEUANGAN PADA UMKM BERDASARKAN STANDAR AKUNTANSI KEUANGAN ENTITAS MIKRO, KECIL DAN MENENGAH (SAK EMKM). Jurnal Saintifik, 19(1), 25-30. Afriansyah, B. (2021). PENYUSUNAN LAPORAN KEUANGAN PADA PIZZA ZAZA DAN FERA KUE MENGGUNAKAN APLIKASI UKM. Jurnal Ilmiah Raflesia Akuntansi, 7(1), 40-57. DSAK-IAI. 2016, September 23). draft_ed_sak_emkm_kompilasi. Diambil kembali dari http://iaiglobal Eri Kristanto, 2011. Penerapan Standar Akuntansi Keuangan Untuk Entitas Tanpa Akuntabilitas Publik (SAK – ETAP) Pada UMKM Pengrajin Rotan Di Desa Tranggan Kecamaran Gatak, Kabupaten Sukoharjo, Jurusan Akuntansi Fakultas Keguruan Dan Ilmu Pendidikan Universitas Sebelas Maret. Ghozali, Imam. 2011. Aplikasi Analisis Multivariat Dengan Program SPSS. Semarang : Badan Universitas Diponegoro Hamdani. 2019. Mengenal Usaha Mikro Kecil dan Menengah (UMKM) Lebih Dekat. Surabaya : Uwais Inspirasi Indonesia. Hardita, N. 2020. pengertian umkm menurut para ahli dan undang undang yang harus dipahami sebagai pebisnis. Diambil kembali dari https://www.diadona.id/career. Hermelinda, T. (2018). PENGARUH KOMPETENSI SUMBER DAYA MANUSIA TEKNOLOGI INFORMASI DAN PENATAUSAHAAN KEUANGAN DAERAH TERHADAP KUALITAS LAPORAN KEUANGAN PEMERINTAH DAERAH (Studi Empiris Pada Pemerintah Daerah Kabupaten Rejang Lebong). JAZ: Jurnal Akuntansi Unihaz, 1(2), 18-30. nda, T. (2019). JOB ORDER COSTING: METODE & IMPLEMENTASI. Jurnal Ilmiah Raflesia kuntansi, 5(2), 26-35. Hermelinda, T., Meriana, M., & Afriansyah, B. (2021). Pengaruh Implementasi Sistem Pengendalian Mutu Internal dan Budaya MutuTerhadap Peningkatan Mutu Pendidikan Tinggi. JAZ: Jurnal Akuntansi Unihaz, 3(2), 183-195. ( ) Hermelinda, T., Niarti, U., & Natalia, N. (2021). ANALISIS PENERAPAN SISTEM AKUNTANSI PENERIMAAN DAN PENGELUARAN KAS PADA PT. LANCAR ABADI SEKAWAN CURUP. Jurnal Saintifik, 19(2), 49-54. Ikatan Akuntansi Indonesia. 2011. Tujuan Laporan Keuangan. .Jakarta : IAI Ikatan Akuntan Indonesia. 2012. Standar Akuntansi Keuangan. Jakarta: PT Salemba Empat. Ikatan Akuntansi Indonesia. 2011. Tujuan Laporan Keuangan. .Jakarta : IAI Ikatan Akuntan Indonesia. 2012. Standar Akuntansi Keuangan. JURNAL SAINTIFIK (Multi Science Journal), Vol. 19 No. 2 May 2021 page: 65 – 70\| 70 DAFTAR PUSTAKA Jakarta: PT Salemba Empat. tandar Akuntansi Keuangan. Jakarta: PT Salemba E Meriana, M. (2018). Pengaruh Tingkat Suku Bunga terhadap Tingkat Pertumbuhan Nasabah pada BTPN Cabang Kepahiang. Jurnal Ilmiah Raflesia Akuntansi, 4(2), 1-11. Meriana, M., Paddery, P., & Natalia, N. (2021). ANALISIS BIAYA PRODUKSI, TINGKAT PENJUALAN, DAN LABA SAAT DAN SEBELUM PANDEMI COVID-19 PADA UMKM KOPI BUBUK CAP JEMPOL AIR BANG CURUP TENGAH. Jurnal Saintifik, 19(1), 31-36. Niarti, U. (2018). Analisis Perbandingan Laba-Rugi pada CV. Maju Jaya Abadi (MJA). Jurnal Ilmiah Raflesia Akuntansi, 4(2), 6-9. Paddery, P. (2018). THE EFFECT OF HUMAN RESOURCE THE EFFECT OF HUMAN RESOURCE CAPACITY, UTILIZATION OF INFORMATION TECHNOLOGY AND INTERNAL CONTROL ACCOUNTING OF RELIABILITY LOCAL GOVERNMENT FINANCIAL REPORTING (Studi Pada Pemerintah Kabupaten Rejang Lebong). Jurnal Ilmiah Raflesia Akuntansi, 4(2), 63-85. JURNAL SAINTIFIK (Multi Science Journal), Vol. 19 No. 2 May 2021 page: 65 – 70\| 70
https://openalex.org/W4243450541	https://pure.eur.nl/ws/files/47807668/REPUB_99403-OA.pdf	English	null	Longitudinal assessment of bleomycin-induced lung fibrosis by Micro-CT correlates with histological evaluation in mice	Multidisciplinary respiratory medicine	2,017	cc-by	7,878	ORIGINAL RESEARCH ARTICLE ORIGINAL RESEARCH ARTICLE Abstract Background: The intratracheal instillation of bleomycin in mice induces early damage to alveolar epithelial cells and development of inflammation followed by fibrotic tissue changes and represents the most widely used model of pulmonary fibrosis to investigate human IPF. p y g Histopathology is the gold standard for assessing lung fibrosis in rodents, however it precludes repeated and longitudinal measurements of disease progression and does not provide information on spatial and temporal distribution of tissue damage. g Here we investigated the use of the Micro-CT technique to allow the evaluation of disease onset and progression at different time-points in the mouse bleomycin model of lung fibrosis. Micro-CT was throughout coupled with histological analysis for the validation of the imaging results. Methods: In bleomycin-instilled and control mice, airways and lung morphology changes were assessed and reconstructed at baseline, 7, 14 and 21 days post-treatment based on Micro-CT images. Ashcroft score, percentage of collagen content and percentage of alveolar air area were detected on lung slides processed by histology and subsequently compared with Micro-CT parameters. Results: Extent (%) of fibrosis measured by Micro-CT correlated with Ashcroft score, the percentage of collagen content and the percentage of alveolar air area (r2 = 0.91; 0.77; 0.94, respectively). Distal airway radius also correlated with the Ashcroft score, the collagen content and alveolar air area percentage (r2 = 0.89; 0.78; 0.98, respectively). Conclusions: Micro-CT data were in good agreement with histological read-outs as micro-CT was able to quantify effectively and non-invasively disease progression longitudinally and to reduce the variability and number of animals used to assess the damage. This suggests that this technique is a powerful tool for understanding experimental pulmonary fibrosis and that its use could translate into a more efficient drug discovery process, also helping to fill the gap between preclinical setting and clinical practice. Keywords: Bleomycin, Lung fibrosis, Micro-CT, Mice * Correspondence: fb.stellari@chiesi.com †Equal contributors 1Chiesi S.p.A., Pre-Clinical R & D, Parma, Italy 7Chiesi Farmaceutici, Pharmacology & Toxicology Department Corporate Pre-Clinical R & D, Largo Belloli, 11/A, Parma 43122, Italy Full list of author information is available at the end of the article © The Author(s). Longitudinal assessment of bleomycin- induced lung fibrosis by Micro-CT correlates with histological evaluation in mice Francesca Ruscitti1†, Francesca Ravanetti2†, Jeroen Essers3, Yanto Ridwan3, Sasha Belenkov4, Wim Vos5, Francisca Ferreira5, Alex KleinJan6, Paula van Heijningen3, Cedric Van Holsbeke5, Antonio Cacchioli2, Gino Villetti1 and Franco Fabio Stellari1,7* Ruscitti et al. Multidisciplinary Respiratory Medicine (2017) 12:8 DOI 10.1186/s40248-017-0089-0 Ruscitti et al. Multidisciplinary Respiratory Medicine (2017) 12:8 DOI 10.1186/s40248-017-0089-0 Open Access * Correspondence: fb.stellari@chiesi.com †Equal contributors 1Chiesi S.p.A., Pre-Clinical R & D, Parma, Italy 7Chiesi Farmaceutici, Pharmacology & Toxicology Department Corporate Pre-Clinical R & D, Largo Belloli, 11/A, Parma 43122, Italy Full list of author information is available at the end of the article Background about the temporal and spatial distribution of the fi- brotic lesion over the course of the study. Moreover, the tendency for a spontaneous resolution of the fibrotic le- sion, its patchy distribution with some areas of the lung unaffected, and possible sampling errors with regard to the selection and the number of fields to be analyzed, are additional shortcomings that hamper successful translation of the results to the clinic. Idiopathic pulmonary fibrosis (IPF) is a chronic, progres- sively fatal lung disease characterized by fibroblast pro- liferation and extracellular matrix remodeling. IPF etiology is unknown and possesses the worst prognosis among all interstitial lung diseases, with a median sur- vival of 2–4 years after diagnosis [1]. The pathogenesis of IPF is not fully understood. It has been hypothesized that multiple cycles of lung injuries lead to destruction of epithelial alveolar cells that in turn cause the migration, proliferation and activation of mesenchymal cells and the exaggerated accumulation of fibroblasts and myofibroblasts. This induces an excessive collagen depos- ition within the lung interstitium and alveolar space, mirroring abnormal wound repair [1, 2]. In humans, IPF manifests histopathologically as usual interstitial pneumo- nia (UIP) and as subpleural and basal predominant reticula- tion with honeycombing on high-resolution computed tomography (HRCT) of the chest [3]. Recently, non-invasive Magnetic Resonance (MRI), Micro- Computer Tomography (Micro-CT) and optical techniques have been applied to longitudinal monitoring of airway remodeling and inflammation in murine models [13–17]. In particular, lung Micro-CT providing high air-tissue contrast is being increasingly used in pulmonary re- search for the investigation of lung fibrosis in experi- mental animals. Initial evidence suggests that by using this imaging modality it is possible to quantify early pathological changes associated with pulmonary fibrosis that correlate to histopathological findings [18–20]. Bleomycin is a glycosylated linear non-ribosomal pep- tide antibiotic produced by the bacterium Streptomyces verticillus. Bleomycin has potent tumor killing proper- ties, which have given it an important role in cancer chemotherapy for curable diseases such as germinative tumors and Hodgkin’s lymphoma. However, the major limitation of bleomycin therapy is pulmonary toxicity which can be life threatening in up to 10% of the pa- tients receiving the drug [4–6]. Background Indeed, serial Micro-CT based analysis as compared with classical histological approaches results in quantita- tive datasets that should allow for longitudinal assess- ment, comparison among different treatment groups including the effect of therapeutic intervention and de- tailed tomographic information documenting the extent of the disease in the individual animal. Macro-CT and Micro-CT based functional respiratory imaging (FRI) has demonstrated to be a powerful tool to look at disease progression in IPF. Using FRI, different studies have been performed in both human subjects and animal models to assess the response to drug and intervention therapy [21–23], to assist in better diagno- sis [24] and to study regional variations in lung structure and function [25]. Based on these evidences, bleomycin is the most widely used drug for the induction of experimental pul- monary fibrosis. The bleomycin model has many advan- tages: it is easy to perform, widely accessible and highly reproducible [7, 8]. Whilst lung bleomycin administration is considered the easiest and the most reproducible way to induce consist- ent lung fibrosis in rodents, concerns remain with regard to the temporal aspects of the model and the translation of the results to the clinic. The bleomycin model has two distinct phases: the inflammatory stage that develops within the first 2 weeks after the injury, which then sub- sides to the fibrotic phase. It is now evident that targeting mechanisms occurring during the inflammatory phase are not likely to translate into a clinical benefit and that in order to determine new drugs efficacy their testing must be performed during the fibrotic phase [9]. In this study, we intended to quantify longitudinally lung fibrosis progression and airway radius changes by using a semi-automated algorithm at different time- points after bleomycin administration in mice starting from baseline and at days 7, 14 and 21. Furthermore, complementary histological measurements were per- formed at all time-points to validate the outcome of the serial Micro-CT investigation. Abstract 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Page 2 of 10 Page 2 of 10 Ruscitti et al. Multidisciplinary Respiratory Medicine (2017) 12:8 Experimental set up The conventional assessment tools used in the bleo- mycin model [9] are terminal and primarily based on labor-intensive histology and quantitative assessment of biochemical biomarkers (i.e. hydroxyproline levels and lung collagen content) [10–12]. These parameters are known to be highly variable among animals and repre- sent only a snapshot at a given time-point of a far more complex and dynamic biological process. Neither histo- pathology nor biochemical analysis provide information Sixty mice were randomly assigned to receive either sa- line or bleomycin (dose) and were double instilled intra- tracheally on day 0 and 4. Lungs were harvested in 5 animals per group for histological analysis at 7, 14 and 21 days after the first treatment. Parallel groups of 5 ani- mals for each treatment arm were imaged with micro- CT at the same time-points and the lungs harvested soon after the scanning to assess the correlation between micro-CT with histological data. Page 3 of 10 Ruscitti et al. Multidisciplinary Respiratory Medicine (2017) 12:8 Page 3 of 10 g p g Micro-CT data were analyzed by an independent company (FLUIDDA NV, Belgium) g p g Micro-CT data were analyzed by an independent company (FLUIDDA NV, Belgium) Bleomycin administration Micro-CT scan data was converted into 3D models of airways, lung lobes and fibrosis using Mimics 15 (Materialise, Leuven, Belgium) a previously validated software package (Food and Drug Administration, K073468; Conformité Européenne certificate, BE 05/ 1191.CE.01). Airways and lungs were obtained based on thresholding segmentation. Lungs were split into lobes by identification of the fissure lines from the CT scan. We identified lobes as: Right Cranial Lobe (RCrL), Right Middle Lobe (RMdL), Right Caudal Lobe (RCdL), Right Accessory Lobe (RAcL) and Left Lung (LL) (Fig. 5b). For intratracheal administrations, 50 μl/mouse either saline or bleomycin sulphate from Streptomyces Verticillus (Sigma B2434) (1 mg/Kg) dissolved in saline were instilled into the tracheal lumen through a cannula under isoflurane (2.5% in oxygen) anaesthesia. The pro- cedure was performed using a small laryngoscope (Penn-Century Inc., Philadelphia, U.S.A.) to visualise the epiglottis and ensure a good positioning of the cannula into the tracheal lumen. Bleomycin or saline were injected at day 0 and at day 4 (Fig. 1). y g g Airway models were split into trachea and airways leading to each previously defined lobe (Fig. 5a). Airway radius was given by the average airway lumen radius per zone. The FRI fibrosis parameter was obtained based on thresholding segmentation and consequent intersection with the lobe masks in order to obtain regional informa- tion. Since blood is not a natural contrast on CT images, if no contrast agent is added when performing the scan, no distinction can be made between blood vessels and fibrosis. For that reason, the FRI fibrosis parameter is often a combination of blood vessels and fibrosis. Experimental animals projection radiographs at 30 frames per second. The images were acquired with the X-ray tube set to 90 kVp, 160 μA and projection radiographs were taken throughout the 360° gantry rotation for a total scan time of 4.5 min. The entire set of projection radiographs was retrospectively gated using proprietary intrinsic image-based gating and then reconstructed at functional residual capacity (FRC) by a modified Feldkamp cone-beam filtered back-projection algorithm. The reconstructed volume was 20 × 20 × 20 mm and the voxel size was 40 × 40 × 40 μm. C57Bl/6 female mice, 8–9 weeks supplied by San Pietro NatisoneHorst, The Netherlands (UD), were used. Prior to use, animals were acclimatized for at least 7 days to the local vivarium conditions (room temperature: 20– 24 °C; relative humidity: 40–70%; 12-h light–dark cycle), having free access to standard rodent chow and softened tap water. All animal experiments described herein were approved by the intramural animal-welfare committee for animal experimentation of Chiesi Farmaceutici and ERASMUS MC under protocol number: EMC 3349 (138-14-07) comply with the European Directive 2010/ 63 UE, Italian D.Lgs 26/2014 and the revised “Guide for the Care and Use of Laboratory Animals” [10]. Collagen content In the bleomycin-treated groups, the Ashcroft score was significantly increased for the whole duration of the study compared with the saline groups (Fig. 2e). Single fibrotic masses were evident starting from day 7 (Score 4) (Fig. 2b) and evolved into confluent conglomerates of substitutive collagen at day 14 (Score 4.83) (Fig. 2c). At day 21, the fibrotic score remained stable (Score 4.76) (Fig. 2d), and no changes in morphological distribution and appearance of fibrosis pattern were revealed. The saline-treated group, as expected, showed normal lung architecture (Score <1) at all points of observation with no prominent inflammation or fibrosis masses in the parenchyma (Fig. 2a and e). The degree of fibrosis was quantified by the NIS-AR image analysis software (Nikon). As ROIs three 10× magnification 1.1 × 0.9 mm, randomly selected were considered per each slide. By standardization of image contrast, brightness and colour threshold settings, the image analysis program was configured to detect areas of green-stained collagen within each ROI. The collagen content is expressed as a percent- age of collagen deposition area (μm2) referred to the lung total area (μm2) within the ROI. Bronchi and blood vessels has been removed from the ROI area. Ashcroft score h l l Masson’s trichrome staining was used to stage the sever- ity of the fibrosis by Ashcroft score and to quantify colla- gen content percentage and alveolar air area fraction percentage (Fig. 2e, f and g). Representative images of histological sections stained with Masson’s trichrome showing the fibrosis progression of saline and bleomycin- treated mice at the designated time-points are shown in Fig. 2a, b, c, and d. Morphological changes in lung sections were graded semi-quantitatively according to the scale defined by Ashcroft [26] modified by H bner et al. [27]. All the parenchyma in the lung sections were assessed by a sys- tem of 0–8 score. In every field (20× magnification microscopic field), the predominant degree of fibrosis was recorded as that occupying more than half of the field area. The final score was expressed as a mean of in- dividual scores observed on all microscopic fields. Histopathological examination revealed that bleomycin induced a fibrotic pattern characterized by a patchy dis- tribution of the fibrotic foci in the lung parenchyma that lead to a marked geographic heterogeneity in the distri- bution of fibrosis within the same lobe. Data analysis Mice were sacrificed and lungs were removed and inflated with a cannula through the trachea by gentle infusion with 0.6 ml of 10% neutral-buffered formalin. The lungs were then placed in a vial containing 10% formalin and fixed for 24 h. For histological assessment, the whole lungs were dehydrated in graded ethanol series, clarified in xylene and paraffin embedded. Data are expressed as the mean and the standard error of the mean (s.e.m.). Statistical analysis was performed using one-way ANOVA followed by Dunnett’s t test for multiple comparison (PRISM Statistical software v 4.0.3). p < 0.05, p < 0.01 was considered a level of statistical significance. Results Sections of 5 μm thick were cut with a rotary micro- tome (Reichert-Jung 1150/Autocut) in dorsal plane. The sections were Masson’s trichrome stained according to the manufacturer’s specifications (Histo-Line Laborator- ies). For analysis the whole-slide imaging were acquired by the NanoZoomer S-60 Digital slide scanner (Hama- matsu). Fibrotic lung injury was assessed morphologic- ally by semi-quantitative and quantitative parameters as follow. Two independent pathologists with experience in animal models of lung fibrosis performed blind histo- logical analysis of the specimens/slides. Development of pulmonary fibrosis induced by intratra- cheal administration of bleomycin in mice is often un- predictable. There is a high degree of variability among individual animals in the extent of fibrosis and the fibro- sis often tends to resolve spontaneously if inhaled bleo- mycin is administered as single shot. This represents a real limitation for pharmacological studies due to the re- stricted time window available for testing new drugs as therapeutic intervention. To overcome these issues, in the present study we have used a double instillation of bleomycin for inducing a more robust and reproducible lung fibrosis as detailed in the experimental set up and described in Fig. 1. Micro-CT scanning After mice were anaesthetized with isoflurane, the mouse lung imaging was performed with a Quantum FX Micro- CT scanner (PerkinElmer, Inc., Waltham, MA) using a cardio-respiratory gated technique as reported previously (PMID: 26967592). The X-ray system of the scanner uses a microfocus tube with a focal spot size of 5 μm at 4 W and produces X-rays in a cone-beam geometric formation. The detection system of the scanner is composed of an amorph- ous silicon digital X-ray plat panel which can acquire Fig. 1 Experimental timeline for bleomycin-induced lung mouse fibrosis. Sixty C57Bl/6 female were randomly assigned to receive either saline or bleomycin (dose) and were double instilled intratracheally on day 0 and 4. Lungs were harvested in 5 animals per group for histological analysis at 7, 14 and 21 days after the first treatment. Parallel groups of 5 animals for each treatment arm were imaged with micro-CT at the same time-points and the lungs harvested soon after the scanning to assess the correlation between micro-CT with histological data Fig. 1 Experimental timeline for bleomycin-induced lung mouse fibrosis. Sixty C57Bl/6 female were randomly assigned to receive either saline or bleomycin (dose) and were double instilled intratracheally on day 0 and 4. Lungs were harvested in 5 animals per group for histological analysis at 7, 14 and 21 days after the first treatment. Parallel groups of 5 animals for each treatment arm were imaged with micro-CT at the same time-points and the lungs harvested soon after the scanning to assess the correlation between micro-CT with histological data Ruscitti et al. Multidisciplinary Respiratory Medicine (2017) 12:8 Page 4 of 10 Page 4 of 10 Page 4 of 10 Ruscitti et al. Multidisciplinary Respiratory Medicine (2017) 12:8 Alveolar air area p < 0.05; *p < 0.01 Lung tissue with normal architecture or characterized by mild fibrotic changes, such as alveolar septum thick- ening, partly enlarged and rarefied alveoli corresponding to an Ashcroft score from 0 to 3 (Fig. 4b and c). The Micro-CT technology was able to visualize and detect the lung parenchyma showing a very well overlapping with histology (Fig. 4a, b and c). the other two parameters measured, the collagen content percentage was significant increased over the time in bleomycin-treated mice compared with saline group (Fig. 2g). Indeed, the areas of fibrosis that stained posi- tively for collagen were increased at day 7, peaking at day 14 and remained stable till the last point of observation. In Fig. 3, Micro-CT scan images from a 21 days saline- treated mouse and the same bleomycin-treated mice at baseline, 7, 14 and 21 days are shown, highlighting the capability of Micro-CT to visualize longitudinally the pro- gressive anatomical changes of the lung architecture, well visible if compare to the saline-treated mice projections. gy g Single fibrotic masses or confluent conglomerates of substitutive collagen and lesions corresponding to an Ashcroft score of 4–5 (Fig. 4e and f) appeared as darker grey areas in the Micro-CT images, Fig. 4d. In higher density fibrotic areas, classified with an Ashcroft score above 6 (Fig. 4e and g), confluent fibrotic masses were present preventing air access to the alveoli. These re- gions appeared as black areas in the Micro-CT pictures (Fig. 4d), due to the missing air contrast. In order to quantify the airway radius, airway models were split into a central and a distal part, that anatomically is the intra- pulmonary airway tract. Representative pictures of seg- mented airways and lung lobes from the Micro-CT scan data are shown in Fig. 5a and b, respectively. To compare the 2-dimensional histological analysis with the Micro-CT images, the lungs were embedded in paraffin in order to maintain their body position and cut along the dorsal plane. Once the serial sections were cut and stained, the corresponding Micro-CT images were identified based on the bronchial tree and the paren- chyma structure. For each observation time-point a sat- isfactory structural matching between the histology slides and the Micro-CT projections was identified. Alveolar air area As an indirect parameter of fibrosis the alveolar air area was quantified. By mean of the same software and ROIs applied for the fibrosis fraction, the air area within the ROIs was detected using a white threshold. The Air Area is expressed as a percentage of area (μm2) occupied by air referred to the lung total area (μm2) within the ROI. Bronchi and blood vessels has been removed from the ROI area. Histomorphometric analysis of the alveolar air area in control animals revealed that this parameter was un- changed at all time-points, representing around 70% of total lung area. On the contrary, a significant decrease in the alveolar air area percentage was observed in bleo- mycin injured mice compared to control groups over the course of the study (Fig. 2f). Fully in agreement with Ruscitti et al. Multidisciplinary Respiratory Medicine (2017) 12:8 Page 5 of 10 Fig. 2 Quantitative histological analysis of lung fibrosis progression in bleomycin mouse model. Representative images of Trichromic Masson-staining of mice lung sections: of saline and bleomycin-treated mice over time. a saline, (b, c and d). Bleomycin treated mice at 7–14 and 21 respectively used for the quantification (10×). Bars: 200 μm. Time course of disease progression of saline and bleomicyn-treated mice at 7, 14 and 21 days: e Ashcroft’s score criteria, f Alveolar Air area percentage and g Collagen content percentage. The experiment was repeated three times and each point presents the mean ± s.e.m of 10 animals, for total of 60 mice per experimental session. Changes were compared to the saline group using one-way ANOVA followed by Dunnett’s t test for multiple comparison. p < 0.05; *p < 0.01 Fig. 2 Quantitative histological analysis of lung fibrosis progression in bleomycin mouse model. Representative images of Trichromic Masson-staining of mice lung sections: of saline and bleomycin-treated mice over time. a saline, (b, c and d). Bleomycin treated mice at 7–14 and 21 respectively used for the quantification (10×). Bars: 200 μm. Time course of disease progression of saline and bleomicyn-treated mice at 7, 14 and 21 days: e Ashcroft’s score criteria, f Alveolar Air area percentage and g Collagen content percentage. The experiment was repeated three times and each point presents the mean ± s.e.m of 10 animals, for total of 60 mice per experimental session. Changes were compared to the saline group using one-way ANOVA followed by Dunnett’s t test for multiple comparison. Alveolar air area A representative matching between Micro-CT images and lung stained slides obtained from the same individual mice treated either with saline or with bleomycin at 14 days is depicted in Fig. 4a, b, d, and e. The quantification of the airway radius and the per- centage of fibrosis from the Micro-CT scan data of saline and bleomycin-treated mice was performed at baseline, 7, 14 and 21 days post-treatment (Fig. 5c Ruscitti et al. Multidisciplinary Respiratory Medicine (2017) 12:8 Page 6 of 10 Fig. 3 Serial Micro-CT imaging during bleomycin-induced lung fibrosis. 3D Micro-CT imaging of progressive pulmonary fibrosis in a representative saline and bleomycin-treated mouse over time. Mice were scanned weekly, from baseline to 21 days after induction Fig. 3 Serial Micro-CT imaging during bleomycin-induced lung fibrosis. 3D Micro-CT imaging of progressive pulmonary fibrosis in a representative saline and bleomycin-treated mouse over time. Mice were scanned weekly, from baseline to 21 days after induction Fig. 3 Serial Micro-CT imaging during bleomycin-induced lung fibrosis. 3D Micro-CT imaging of progressive pulmonary fibrosis in a representative saline and bleomycin-treated mouse over time. Mice were scanned weekly, from baseline to 21 days after induction Fig. 4 Micro-CT and histology view-matching. a, d Micro-CT dorsal planes of saline and bleomycin-treated mouse at 14 days fitted with the appropriate histological sections, b, e (10×) magnification of normal lung parenchyma of saline (c), moderately fibrotic (f) and heavily fibrotic lung parenchyma of bleomycin-treated mice (g). The squares indicate the position and the matching among histology at low magnification and the Micro-CT. The magnification of the green and black squares (c, f) indicate normal and moderately fibrotic parenchyma detected by histology and Micro-CT, the red one (g) indicates the heavily fibrotic lung that appeared as black in the Micro-CT view. The respective Ashcroft scores are represented in the right side of the figure Fig. 4 Micro-CT and histology view-matching. a, d Micro-CT dorsal planes of saline and bleomycin-treated mouse at 14 days fitted with the appropriate histological sections, b, e (10×) magnification of normal lung parenchyma of saline (c), moderately fibrotic (f) and heavily fibrotic lung parenchyma of bleomycin-treated mice (g). The squares indicate the position and the matching among histology at low magnification and the Micro-CT. Alveolar air area The magnification of the green and black squares (c, f) indicate normal and moderately fibrotic parenchyma detected by histology and Micro-CT, the red one (g) indicates the heavily fibrotic lung that appeared as black in the Micro-CT view. The respective Ashcroft scores are represented in the right side of the figure Ruscitti et al. Multidisciplinary Respiratory Medicine (2017) 12:8 Page 7 of 10 Fig. 5 Segmentation and quantification of airways and lung lobes on Micro-CT. Airway models were split into a central and a distal part, that anatomically is the intrapulmonary airway tract, leading to each defined lobe (a). Airway radius was evaluated with Micro-CT in the distal part of the airway tree of untreated and treated mice with bleomycin at 7, 14 and 21 day (c). Lungs were split into lobes by identification of the fissure lines from the CT scan. Lobes were identified as: Right Cranial Lobe (RCrL), Right Middle Lobe (RMdL), Right Caudal Lobe (RCdL), Right Accessory Lobe (RAcL) and Left Lung (LL) (b). Total lung fibrosis was evaluate on Micro-CT of treated mice with saline or bleomycin at 7, 14 and 21 day (d). Each point presents the mean ± s.e.m of 5 animals, for a total of 30 mice. Changes were compared to the saline group using one-way ANOVA followed by Dunnett’s t test for multiple comparison. p < 0.05; *p < 0.01 Fig. 5 Segmentation and quantification of airways and lung lobes on Micro-CT. Airway models were split into a central and a distal part, that anatomically is the intrapulmonary airway tract, leading to each defined lobe (a). Airway radius was evaluated with Micro-CT in the distal part of the airway tree of untreated and treated mice with bleomycin at 7, 14 and 21 day (c). Lungs were split into lobes by identification of the fissure lines from the CT scan. Lobes were identified as: Right Cranial Lobe (RCrL), Right Middle Lobe (RMdL), Right Caudal Lobe (RCdL), Right Accessory Lobe (RAcL) and Left Lung (LL) (b). Total lung fibrosis was evaluate on Micro-CT of treated mice with saline or bleomycin at 7, 14 and 21 day (d). Each point presents the mean ± s.e.m of 5 animals, for a tota of 30 mice. Changes were compared to the saline group using one-way ANOVA followed by Dunnett’s t test for multiple comparison. p < 0.05; **p < 0.01 and d). Alveolar air area A statistical significant increase in fibrosis and airway radius was found for all the time points of observation between the saline and bleomycin-treated mice (Fig. 5c and d), in agreement with histological findings as described in Fig. 2e, f, and g. distal airway radius, and the three histological measure- ments: Ashcroft score, alveolar air area and collagen content percentage. Distal airway radius expressed in mm was plotted against Ashcroft score (Fig. 6a), colla- gen content percentage (Fig. 6b) and alveolar air area percentage (Fig. 6c), for the mean of each group at the different time points. Linear regression analysis showed g g g The plots in Fig. 6 show the correlation between the two Micro-CT FRI parameters evaluated, fibrosis and Fig. 6 Correlation between Micro-CT and histological parameters. Airway radius (mm) correlation with: Ashcroft score (a), Collagen Content % (b) and Alveolar Air Area% (c) Fibrosis (%) correlation with: Ashcroft score, (d), Collagen content % (e) and Alveolar Air area % (b). Each single dot represents the mean ± s.e.m of 5 animals for a total of 30 mice Fig. 6 Correlation between Micro-CT and histological parameters. Airway radius (mm) correlation with: Ashcroft score (a), Collagen Content % (b) and Alveolar Air Area% (c) Fibrosis (%) correlation with: Ashcroft score, (d), Collagen content % (e) and Alveolar Air area % (b). Each single dot represents the mean ± s.e.m of 5 animals for a total of 30 mice Ruscitti et al. Multidisciplinary Respiratory Medicine (2017) 12:8 Page 8 of 10 Page 8 of 10 that airway radius positively correlated with the classical Ashcroft scoring (r2 = 0.91; p = 0.003), (Fig. 6a) and with collagen content percentage (r2 = 0.77; p = 0.022), (Fig. 6b) but negatively with the alveolar air area per- centage as expected (r2 = 0.94; p = 0.001), (Fig. 6c). evaluation of Ashcroft score, and histomorphometric measurements of collagen content percentage and alveo- lar air area fraction. Moreover, mice need to be sacrificed at each time point in order to have a sufficient number for statistical power, precluding the possibility to monitor the disease progression in the same subject. When a mouse is sacri- ficed and disappears from the experimental setting, a missing information is generated in the biological sys- tem. The missing information cannot be retrieved any- more and performing new studies may increase biological variability. Alveolar air area Micro-CT seems to be an ideal technology for quantifying pulmonary disease progres- sion lung parenchyma, remodeling and the bronchial tree changes. Micro-CT approach assures that the entire lung is assessed in each animal, while the conventional tools are limited [31]. g g The same correlation was performed using the FRI fibrosis parameter. Fibrosis percentage was plotted against: Ashcroft score (Fig. 6d), collagen content per- centage (Fig. 6e) and alveolar air area percentage (Fig. 6f) for the mean of each group at the different time points. Linear regression analysis showed that fibrosis percent- age positively correlated with the classical Ashcroft scor- ing (r2 = 0.89; p = 0.005) (Fig. 6d) and with collagen content percentage (r2 = 0.78; p = 0.019) (Fig. 6e), but negatively correlated with the alveolar air area percent- age as expected (r2 = 0.98; p < 0.001) (Fig. 6f). These findings were consistent and demonstrate that FRI parameters from Micro-CT can serve as a reliable indication of the fibrotic changes occurring in the lung parenchyma. In this study Micro-CT was used to quantify longitudin- ally disease progression by evaluating two parameters: air- way radius in the distal part of the bronchial tree (Fig. 5a) and fibrosis percentage in the lung parenchyma (Fig. 5b). However, the Micro-CT revealed a limitation on the fibro- sis percentage detection at early time points. (Fig. 5d). It is bleomycin treatment induced an inflammatory phase [32] and there is a balance between inflammation and fibrosis at day 7 resulting with slightly overestimation of fibrosis percentage parameter measured by Micro-CT compare to histology. The Micro-CT projections of the lungs present- ing mild or moderate degree of fibrosis areas with an Ash- croft score (0–4) found a perfect match with histology (Fig. 4a, b, c, and f). The fibrosis pattern is characterized by a patchy distribution of the fibrotic lesions containing some areas with a high degree of fibrosis with an Ashcroft score (5- > 6), Fig. 4d, e and g. These zones in Micro-CT images appeared darker till almost become black corre- sponding to an increase in severity of fibrosis, due to the missing contrast from the alveolar air space, limiting the possibility to overlap Micro-CT images with histology. Despite some technical limitations, the correlation plots (Fig. 6) clearly highlight convincing thresholds for each histological parameter and Micro-CT allowed the quantifi- cation of distal airway radius and fibrosis (Fig. 5c and d). Discussion Despite recent advances in therapeutics available for IPF, there is a continued need to find more effective therap- ies to prevent or reverse lung fibrosis [28]. Moreover, the only treatment proven effective in prolonging sur- vival is lung transplantation with a post-transplantation 5-year survival for IPF patients of approximately 44% [29]. The development of novel therapies relies on how well experimental animal models are able to translate findings to clinical outcomes but there is still much de- bate around the reliability of current animal models for pulmonary fibrosis [28]. Single bleomycin administration in mice produces fibrotic lesions with an high degree of variability among individual animals and, moreover, the fibrosis tends to resolve spontaneously. In order to overcome some of the intrinsic animal modelling limitations, we have set up a more robust and reproducible lung fibrosis mouse model that requires a double instillation of bleomycin as detailed in Fig. 1. In vivo imaging in small animal models is a powerful tool to visualize disease progression, to assess their topo- graphical distribution and to qualitatively evaluate the severity in a non-invasive manner. This approach may help to fill the gap between preclinical setting and clin- ical practice [30]. An increase of the distal airway radius parameter could be explained by the obliteration of the alveoli and the con- sequent airflow limitation. This could lead to the airway tree dilation as a compensatory phenomenon, which ex- plains the very good correlation found with Ashcroft score and alveolar air area percentage (Fig. 6a and c) and even though weaker, a significant correlation between airway radius and collagen content (%) was also obtained. In the present work we monitored the lung fibrosis in bleomycin-induced mouse model by using Micro-CT technology coupled to classical histology. The major limitation of the conventional assessments of fibrosis re- lies on invasive ex vivo measurements. These method- ologies produce solid and reproducible data, but could be biased from sampling methods. In fact, only a small fraction of the total lung is accounted for the histological FRI fibrosis parameter increased over the time points of observation. Additionally, the amazing correlation found with alveolar air area percentage (r2 = 0.98), (Fig. 6f) and with Ashcroft score and collagen content demonstrated Ruscitti et al. Ethics approval and consent to participate Ethics approval and consent to participate All animal experiments described herein were approved by the intramural animal-welfare committee for animal experimentation of Chiesi Farmaceutici and ERASMUS MC under protocol number: EMC 3349 (138-14-07) comply with the European Directive 2010/63 UE, Italian D.Lgs 26/2014 and the revised “Guide for the Care and Use of Laboratory Animals”. Nowadays is well known that drug effects are crucially dependent on timing of compound administration and that the testing of anti-fibrotic agents must be performed during a specific fibrotic phase of the animal model [33]. This limitation can be overcome using Micro-CT scan: all the mice could be screened before entering in the drug treatment and divided according to the severity of the dis- ease. The huge advantage is that each mouse considered as a single subject could be the control of itself; and an in- terim analysis of drug efficacy could be performed in order to guide and plan in advance the future experiments. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Author details 1 1Chiesi S.p.A., Pre-Clinical R & D, Parma, Italy. 2Dipartimento di Scienze Medico Veterinarie, Università di Parma, Parma, Italy. 3Department of Molecular Genetics, Vascular Surgery, and Radiation Oncology, Erasmus MC, Rotterdam, The Netherlands. 4PerkinElmer, Inc, Waltham, MA, USA. 5Fluidda NV, Kontich, Belgium. 6Department of Pulmonary Medicine Erasmus MC, Rotterdam, The Netherlands. 7Chiesi Farmaceutici, Pharmacology & Toxicology Department Corporate Pre-Clinical R & D, Largo Belloli, 11/A, Parma 43122, Italy. Authors’ contributions C i d d i FR Conception and design, FR, JE, FFS. Laboratory testing, FR, FR, JE, YA, SB, PVH, FFS. Data collection FR, FR, JE, YA, SB, PVH, FFS. Data analysis and interpretation: FR, FR, JE, YA, SB, WV, FF, CVH, AC, GV, FFS. Drafting of manuscript: FR, FR, JE, SB, WV, FF, AK, FFS. All authors read and approved the final manuscript. Our approach with Micro-CT in future drug efficacy studies would be to obtain baseline evaluation of each ani- mal before fibrosis is induced, then confirmation of fibrosis before the treatment is initiated and finally evaluation of potential treatment effects within the therapeutic window. Received: 24 December 2016 Accepted: 10 March 2017 Received: 24 December 2016 Accepted: 10 March 2017 Acknowledgements The authors would like to thank Daniela Pompilio for technical help Availability of data and materials Data sharing is not applicable to this article. Abbreviations l FRI: Functional respiratory imaging; IPF: Idiopathic pulmonary fibrosis; LL: Left Lung; Micro CT: Micro-Computer Tomography; RAcL: Right Accessory Lobe; RCdL: Right Caudal Lobe; RCrL: Right Cranial Lobe; RMdL: Right Middle Lobe; ROI: Region of interest FRI: Functional respiratory imaging; IPF: Idiopathic pulmonary fibrosis; LL: Left Lung; Micro CT: Micro-Computer Tomography; RAcL: Right Accessory Lobe; RCdL: Right Caudal Lobe; RCrL: Right Cranial Lobe; RMdL: Right Middle Lobe; ROI: Region of interest Moreover, the capability of Micro-CT technology to quantify the fibrosis and the airway radius changes is more likely to reflect directly in the pathologic and also therapeutic changes in the lungs. Discussion Multidisciplinary Respiratory Medicine (2017) 12:8 Page 9 of 10 Page 9 of 10 Page 9 of 10 that lung parameters from Micro-CT serve as a reliable indicator of the typical lung changes in IPF pathology, involving: airspace fibrosis, alveolar collapse and decrease in the gas-exchange tissue surface. Funding Funding This work was supported by Institutional funds from Chiesi Farmaceutici S.p.A. Even though the Micro-CT cannot be used as a stand- alone technology at the moment, it seems to be comple- mentary to histology. Micro-CT can provide the 3D model while histology provides more accurate informa- tion on the cellular and molecular level. Competing interests Competing interests The authors declare that they have no competing interests. We believe that this approach represents a smart tool to evaluate drug efficacy and eventually will prove to be superior compared to terminal assessment for drug discovery [31]. Conclusions Drug discovery process is very expensive and unproductive, [34–36] and there is a huge medical need for treating IPF patients. The in vivo study, that recapitulates the pathogen- esis, the lesions as well as the effect of a drug on the disease outcome, represents the main obstacle to be overcome during the last steps of development of such molecules. Toxicology Department Corporate Pre-Clinical R & D, Largo Belloli, 11/A, Parma 43122, Italy. Received: 24 December 2016 Accepted: 10 March 2017 References W TA Altered lung function relates to inflammation in an acute LPS mouse model. Pulm Pharmacol Ther. 2012;25(5):399–406. 32. Stellari FF, Ruscitti F, Pompilio D, Ravanetti F, Tebaldi G, Macchi F, et al. Heterologous Matrix Metalloproteinase Gene Promoter Activity Allows In Vivo Real-time Imaging of Bleomycin-Induced Lung Fibrosis in Transiently Transgenized Mice. Front Immunol. 2017;8:199. 12. Balloy V, Deveaux A, Lebeaux D, Tabary O, le Rouzic P, Ghigo JM, et al. Azithromycin analogue CSY0073 attenuates lung inflammation induced by LPS challenge. Br J Pharmacol. 2014;171(7):1783–94. 33. Peng R, Sridhar S, Tyagi G, Phillips JE, Garrido R, Harris P, et al. Bleomycin induces molecular changes directly relevant to idiopathic pulmonary fibrosis: a model for “active” disease. PLoS One. 2013;8(4):e59348. ll b h d 13. Stellari F, Bergamini G, Sandri A, Donofrio G, Sorio C, Ruscitti F, et al. In vivo imaging of the lung inflammatory response to pseudomonas aeruginosa and its modulation by azithromycin. J Transl Med. 2015;13:251. 34. Pammolli F, Magazzini L, Riccaboni M. The productivity crisis in pharmaceutical R & D. Nat Rev Drug Discov. 2011;10(6):428–38. 14. Bianchi A, Ozier A, Ousova O, Raffard G, Crémillieux Y. Ultrashort-TE MRI longitudinal study and characterization of a chronic model of asthma in mice: inflammation and bronchial remodeling assessment. NMR Biomed. 2013;26(11):1451–9. 35. Cook D, Brown D, Alexander R, March R, Morgan P, Satterthwaite G, et al. Lessons learned from the fate of AstraZeneca’s drug pipeline: a five- dimensional framework. Nat Rev Drug Discov. 2014;13(6):419–31. 36. Paul SM, Mytelka DS, Dunwiddie CT, Persinger CC, Munos BH, Lindborg SR, et al. How to improve R & D productivity: the pharmaceutical industry’s grand challenge. Nat Rev Drug Discov. 2010;9(3):203–14. 15. Lederlin M, Ozier A, Dournes G, Ousova O, Girodet PO, Begueret H, et al. In vivo micro-CT assessment of airway remodeling in a flexible OVA-sensitized murine model of asthma. PLoS One. 2012;7(10):e48493. 16. Stellari F, Sala A, Ruscitti F, Carnini C, Mirandola P, Vitale M, et al. Monitoring inflammation and airway remodeling by fluorescence molecular tomography in a chronic asthma model. J Transl Med. 2015;13:336. 17. Ma X, Prakash J, Ruscitti F, Glasl S, Stellari FF, Villetti G, et al. Assessment of asthmatic inflammation using hybrid fluorescence molecular tomography-x- ray computed tomography. J Biomed Opt. 2016;21(1):15009. 18. Jin GY, Bok SM, Han YM, Chung MJ, Yoon KH, Kim SR, et al. Effectiveness of rosiglitazone on bleomycin-induced lung fibrosis: assessed by micro- computed tomography and pathologic scores. References W TA 1. Wynn TA. Integrating mechanisms of pulmonary fibrosis. J Exp Med. 2011; 208(7):1339–50. This combination of better animal models, Micro-CT tech- nology together with histology data allowed a more accurate interpretation of the data obtained in vivo. The possibility to analyze more parameters from the Micro-CT is fundamental in order to fully describe and better understand dynamic pro- cesses during IPF disease onset, progression and therapy. 2. Wynn TA, Ramalingam TR. Mechanisms of fibrosis: therapeutic translation for fibrotic disease. Nat Med. 2012;18(7):1028–40. 3. Martin MD, Chung JH, Kanne JP. Idiopathic pulmonary fibrosis. J Thorac Imaging. 2016;31(3):127–39. 4. Nygaard K, Smith-Erichsen N, Hatlevoll R, Refsum SB. Pulmonary complications after bleomycin, irradiation and surgery for esophageal cancer. Cancer. 1978;41(1):17–22. Taken together, our findings underline the importance of the possibility to accurately quantify specific parameters to follow disease progression in a non-invasive way. We strongly believe that Micro-CT applied to the fibrosis research field could allow longitudinally studies reducing the variability and number of animals used and may, therefore, provide a novel tool for understanding pulmon- ary fibrosis onset, progression and therapy. 5. Balikian JP, Jochelson MS, Bauer KA, Skarkin AT, Garnick MB, Canellos GP, et al. Pulmonary complications of chemotherapy regimens containing bleomycin. AJR Am J Roentgenol. 1982;139(3):455–61. 6. Bargon G. Drug-induced changes in the lungs. [in german]. Rontgenblatter. 1984;37(7):261–5. 7. Antije M, Carlos R-LJ, Lingqiao W, Jack G, Martin K. Models of pulmonary fibrosis. Drug Discov Today Dis Model. 2006;3:243–9. 8. Scotton CJ, Chambers RC. Bleomycin revisited: towards a more representative model of IPF?. Am J Physiol Lung Cell Mol Physiol. 2010; 299(4):L439–41. Page 10 of 10 Page 10 of 10 Page 10 of 10 Ruscitti et al. Multidisciplinary Respiratory Medicine (2017) 12:8 9. Moore BB, Hogaboam CM. Murine models of pulmonary fibrosis. Am J Physiol Lung Cell Mol Physiol. 2008;294(2):L152–60. 30. De Langhe E, Vande Velde G, Hostens J, Himmelreich U, Nemery B, Luyten FP, et al. Quantification of lung fibrosis and emphysema in mice using automated micro-computed tomography. PLoS One. 2012;7(8):e43123. 10. N. R. C. Institute of Laboratory Animal Resources Commission on Life Sciences. Guide for the care and use of laboratory animals. Washington D.C.: National Academy Press; 1996. 31. Ask K, Labiris R, Farkas L, Moeller A, Froese A, Farncombe T, et al. Comparison between conventional and “clinical” assessment of experimental lung fibrosis. J Transl Med. 2008;6:16. 11. Håkansson HF, Smailagic A, Brunmark C, Miller-Larsson A, Lal H. 30. De Langhe E, Vande Velde G, Hostens J, Himmelreich U, Nemery B, Luyten FP, et al. Quantification of lung fibrosis and emphysema in mice using automated micro-computed tomography. PLoS One. 2012;7(8):e43123. 31. Ask K, Labiris R, Farkas L, Moeller A, Froese A, Farncombe T, et al. Comparison between conventional and “clinical” assessment of experimental lung fibrosis. J Transl Med. 2008;6:16. 32. Stellari FF, Ruscitti F, Pompilio D, Ravanetti F, Tebaldi G, Macchi F, et al. Heterologous Matrix Metalloproteinase Gene Promoter Activity Allows In Vivo Real-time Imaging of Bleomycin-Induced Lung Fibrosis in Transiently Transgenized Mice. Front Immunol. 2017;8:199. 33. Peng R, Sridhar S, Tyagi G, Phillips JE, Garrido R, Harris P, et al. Bleomycin induces molecular changes directly relevant to idiopathic pulmonary fibrosis: a model for “active” disease. PLoS One. 2013;8(4):e59348. 34. Pammolli F, Magazzini L, Riccaboni M. The productivity crisis in pharmaceutical R & D. Nat Rev Drug Discov. 2011;10(6):428–38. 35. Cook D, Brown D, Alexander R, March R, Morgan P, Satterthwaite G, et al. Lessons learned from the fate of AstraZeneca’s drug pipeline: a five- dimensional framework. Nat Rev Drug Discov. 2014;13(6):419–31. 36. Paul SM, Mytelka DS, Dunwiddie CT, Persinger CC, Munos BH, Lindborg SR, et al. How to improve R & D productivity: the pharmaceutical industry’s grand challenge. Nat Rev Drug Discov. 2010;9(3):203–14. References W TA Eur J Radiol. 2012;81(8): 1901–6. 19. Lee HJ, Goo JM, Kim NR, Kim MA, Chung DH, Son KR, et al. Semiquantitative measurement of murine bleomycin-induced lung fibrosis in in vivo and postmortem conditions using microcomputed tomography: correlation with pathologic scores–initial results. Invest Radiol. 2008;43(6):453–60. 20. Vande Velde G, Poelmans J, De Langhe E, Hillen A, Vanoirbeek J, Himmelreich U, et al. Longitudinal micro-CT provides biomarkers of lung disease that can be used to assess the effect of therapy in preclinical mouse models, and reveal compensatory changes in lung volume. Dis Model Mech. 2016;9(1):91–8. 21. Vos W, De Backer J, Poli G, De Volder A, Ghys L, Van Holsbeke C, et al. Novel functional imaging of changes in small airways of patients treated with extrafine beclomethasone/formoterol. Respiration. 2013;86(5):393–401. 22. De Backer J, Vos W, Vinchurkar S, Van Holsbeke C, Poli G, Claes R, et al. The effects of extrafine beclometasone/formoterol (BDP/F) on lung function, dyspnea, hyperinflation, and airway geometry in COPD patients: novel insight using functional respiratory imaging. J Aerosol Med Pulm Drug Deliv. 2015;28(2):88–99. 23. Schepens T, Cammu G, Maes S, Desmedt B, Vos W, and Deseure K. Functional respiratory imaging after neostigmine- or sugammadex- enhanced recovery from neuromuscular blockade in the anaesthetised rat: a randomised controlled pilot study. Braz J Anesthesiol (English Edition). 2016. available online http://dx.doi.org/10.1016/j.bjane.2015.11.004. 24. Van Holsbeke CS, Leemans G, Vos WG, De Backer JW, Vinchurkar SC, Geldof M, et al. Functional respiratory imaging as a tool to personalize respiratory treatment in patients with unilateral diaphragmatic paralysis. Respir Care. 2014;59(9):e127–31. 32. Stellari FF, Ruscitti F, Pompilio D, Ravanetti F, Tebaldi G, Macchi F, et al. Heterologous Matrix Metalloproteinase Gene Promoter Activity Allows In Vivo Real-time Imaging of Bleomycin-Induced Lung Fibrosis in Transiently Transgenized Mice. Front Immunol. 2017;8:199. Submit your next manuscript to BioMed Central and we will help you at every step: Submit your next manuscript to BioMed Central and we will help you at every step: 25. De Backer JW, Vos WG, Burnell P, Verhulst SL, Salmon P, De Clerck N, et al. Study of the variability in upper and lower airway morphology in Sprague– Dawley rats using modern micro-CT scan-based segmentation techniques. Anat Rec (Hoboken). 2009;292(5):720–7. • We accept pre-submission inquiries • Our selector tool helps you to ﬁnd the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit and we will help you at every step: 26. Ashcroft T, Simpson JM, Timbrell V. Simple method of estimating severity of pulmonary fibrosis on a numerical scale. J Clin Pathol. 1988;41(4):467–70. 27. Hübner RH, Gitter W, El Mokhtari NE, Mathiak M, Both M, Bolte H, et al. Standardized quantification of pulmonary fibrosis in histological samples. Biotechniques. 2008;44(4):507–11. 514–7. 28. Organ L, Bacci B, Koumoundouros E, Barcham G, Milne M, Kimpton W, et al. Structural and functional correlations in a large animal model of bleomycin- induced pulmonary fibrosis. BMC Pulm Med. 2015;15:81. 29. King TE, Pardo A, Selman M. Idiopathic pulmonary fibrosis. Lancet. 2011; 378(9807):1949–61. 29. King TE, Pardo A, Selman M. Idiopathic pulmonary fibrosis. Lancet. 2011; 378(9807):1949–61.
https://openalex.org/W2920993432	https://scholarworks.sfasu.edu/cgi/viewcontent.cgi?article=2003&context=ita	English	null	The Distribution of Passenger Pigeon on Caddo Sites in the Trans-Mississippi South	Index of Texas archaeology	2,014	cc-by	2,033	The Distribution of Passenger Pigeon on Caddo Sites in the Trans- The Distribution of Passenger Pigeon on Caddo Sites in the Trans- Mississippi South Mississippi South Timothy K. Perttula Heritage Research Center, Stephen F. Austin State University Article 30 Article 30 Volume 2014 Volume 2014 Volume 2014 Laura Nightengale Davaid H. Jurney Follow this and additional works at: https://scholarworks.sfasu.edu/ita Follow this and additional works at: https://scholarworks.sfasu.edu/ita Follow this and additional works at: https://scholarworks.sfasu.edu/ita Part of the American Material Culture Commons, Archaeological Anthropology Commons, Environmental Studies Commons, Other American Studies Commons, Other Arts and Humanities Commons, Other History of Art, Architecture, and Archaeology Commons, and the United States History Commons Part of the American Material Culture Commons, Archaeological Anthropology Commons, Environmental Studies Commons, Other American Studies Commons, Other Arts and Humanities Commons, Other History of Art, Architecture, and Archaeology Commons, and the United States History Commons Tell us how this article helped you. This Article is brought to you for free and open access by the Center for Regional Heritage Research at SFA ScholarWorks. It has been accepted for inclusion in Index of Texas Archaeology: Open Access Gray Literature from the Lone Star State by an authorized editor of SFA ScholarWorks. For more information, please contact cdsscholarworks@sfasu.edu. This article is available in Index of Texas Archaeology: Open Access Gray Literature from the Lone Star State: https://scholarworks.sfasu.edu/ita/vol2014/iss1/30 The Distribution of Passenger Pigeon on Caddo Sites in the Trans-Mississippi The Distribution of Passenger Pigeon on Caddo Sites in the Trans-Mississippi South South Creative Commons License Creative Commons License Creative Commons License Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License INTRODUCTION Following up on the discovery of 10 passenger pigeon elements from one bird in a Caddo burial feature (Burial 52) at the Mitchell site (41BW4) on the Red River in Bowie County, we have documented the distribu- tion of passenger pigeon on Caddo sites in the 7rans-Mississippi 6outh 7o date, we have identiÀed 10 Caddo sites dating between ca. A.D. 1160-1710 with passenger pigeon bones, along with two Woodland period sites (generally predating ca. A.D. 800-900) in the region. These sites range as far east as the Saline River basin in Southwest Arkansas, as far west as the George C. Davis (41CE19) and Spike (41DT16) sites in East Texas, as far north as the Spiro site (34Lf40) in eastern Oklahoma, and as far south as the McLelland site (16BO236) in Northwest Louisiana, primarily situated in the Pineywoods (Figure 1). As Jackson (2005) and Jackson et al. (201268) have noted, the passenger pigeons found on Caddo sites appear to reÁect a ´southward and westward extension of their range during the Late Woodland period (ca. A.D. 700-1000) that continues into the Caddo era (ca. A.D. 1000-1500), likely reaching East Texas sometime shortly after A.D. 1000.” The Distribution of Passenger Pigeon on Caddo Sites in the Trans-Mississippi South Timothy K. Perttula, Laura Nightengale, and David H. Jurney Journal of Northeast Texas Archaeology, Volume 47, 2014 WOODLAND AND CADDO SITES WITH PASSENGER PIGEON BONES Although generally rare in ancestral Caddo faunal assemblages, bones from passenger pigeons, an extinct species (Figure 2), were Tuite common in Late Caddo period contexts at the +ughes Mound site (3SA11) in the Ouachita River basin (Mary Beth Trubitt, December 2013 personal communication). Passenger pi- geon bones are also present in both ca. A.D. 1450-1500 and A.D. 1650-1700 deposits at the Hardman site (3CL418) in the Ouachita River basin (Early 1993:147, 150). Additionally, passenger pigeon bones (n=14) have been found at the Late Caddo McCurtain phase village at the Roden site (34Mc215) along the Red River (Perino 1981). There is one passenger pigeon left humerus from Zone H in the platform mound at the Hatchel site (41BW3) (Darrell Creel, December 2013 personal communication; April 24, 1975 letter from Lyndon L. Hargrave on Àle at TARL), less than a mile north of the Mitchell site. At the Mitchell site itself, the passenger pigeon bones were found placed inside a Pease Brushed-Incised jar in Burial 52; this individual also had 100+ turtle shell fragments found beneath the left arm and hand of the deceased (a 13-17 year old individual of indeterminate sex). Bones from a single passenger pigeon have been recovered at the ca. A.D. 1650-1710 McLelland site on the Red River (Kelley 1997:103). At the Crenshaw site (3MI6) on the Red River in southwest Arkansas, passenger pigeon bones (n=75) are abundant on the Áoor in Feature 6, the antler temple (Jackson et al. 2012:Table 3-1), and in the mid- den surrounding the antler temple. The antler temple deposits date at 2 sigma to between A.D. 1161-1254, calibrated (John Samuelson, December 2013 personal communication). The bones of at least 34 passenger pigeons (n=300 elements) have been recovered in roughly contemporaneous contexts at the George C. Davis site, in a large trash pit in one of the village areas (Thurmond and Kleinschmidt 1979:92). Passenger pigeon bones (n=2) were present in Burial 90 in the Great Mortuary at the Spiro site in eastern Oklahoma (Brown 1996:706), along with 10 turkey wing fan bones. This burial features dates from ca. A.D. 1350-1450, during the Spiro IV interval. Two passenger pigeon bones were recovered from Feature 3 at the Tyson site (41SY92), a large pit in a mid-14th to mid-15th century Caddo house context (Tom Middlebrook, December 2013 personal communication). WOODLAND AND CADDO SITES WITH PASSENGER PIGEON BONES Journal of Northeast Texas Archaeology, Volume 47, 2014 38 Journal of Northeast Texas Archaeology 47 (2014) Figure 1. Sites in the Trans-Mississippi South with passenger pigeon bones. Figure 1. Sites in the Trans-Mississippi South with passenger pigeon bones. The earliest example of passenger pigeon use in East Texas is a passenger pigeon faunal element that has been recovered in Woodland period contexts (ca. 2500-1200 years B.P.) at the Spike site (41DT16). This midden site is in the upper Sulphur River basin (Fields et al. 1993). Another passenger pigeon element has been found in a Woodland period context (ca. A.D. 800-900) at the Sliding Slab Shelter (3SB29) in the northern Ouachita Mountains in Arkansas (Harden 1981:132 and V.10). Although not a Caddo site, passenger pigeon bones were relatively common in 8th-10th century midden deposits in Mound S at the Toltec site (3LN42) in the central Arkansas River valley (Kelly 2012:141), and other Woodland and Mississippian period sites in Northeast Arkansas, including Zebree, Knappenberger, and Ridge House and elsewhere in the state (see Jackson 2005:Table 11.1). Passenger pigeon bones are present throughout the 4 m of deposits at Saltpeter Cave (3NW29), in the Arkansas Ozarks. Concentrations of passenger pigeon bones are noted within Woodland period deposits at the site, which also contain at least Journal of Northeast Texas Archaeology 47 (2014) 39 Figure 2. Early 20th century photograph of a passenger pigeon. Figure 2. Early 20th century photograph of a passenger pigeon. two burials. Passenger pigeon remains extend below the ceramic-bearing deposits into non-ceramic strata. In total, 135 elements, representing 19 minimum individuals, were recovered from Saltpeter Cave. Finally, a single passenger pigeon bone has been recovered in a refuse pit at the Poverty Point site in Northeast Louisiana (http://www.passengerpigeon.org/states/Louisiana.html), and at the famous Mississippian site of Cahokia in southern Illinois (Parmalee 1957). According to James and Neal (1986:198), passenger pigeons did not nest in Arkansas, but they would have been locally abundant and present in roosting areas in winter months along the Red River (cf. Kelly 2012:141). There are also 19th century accounts of roosts in the Ft. Smith, Arkansas area, and several “pigeon roost” place names are known in the Ouachita Mountains. ACKNOWLEDGMENTS We thank colleagues who responded to our reTuest for information about passenger pigeon bones: Dar- rell Creel, Ann M. Early, Ross C. Fields, Paul S. Gardner, Joel Greenberg, Ed Jackson, Thomas Labedz, Terrance J. Martin, Tom Middlebrook, Stephen P. Rogers, John Samuelson, and Mary Beth Trubitt. Thanks also to Lance Trask for preparing Figure 1. CONTEXT OF THE PASSENGER PIGEON BONES Passenger pigeon bones are not abundant in Late Woodland and ancestral Caddo faunal assemblages in the Trans-Mississippi South, and were not apparently an important food resource to local Caddo peoples. In most cases, the passenger pigeon bones have been found in mound contexts, in burials, probably as funerary offerings, in feasting contexts (at the Crenshaw, George C. Davis, and Tyson sites), and then lastly in habita- tion debris. Jackson et al. (2012:68) suggest that the passenger pigeon bones tend to occur in ritual and/or elite contexts on Caddo sites, as they do on contemporaneous Mississippian period sites in the Southeastern United States (see Jackson and Scott 1995), where they were a new and novel food item. 40 Journal of Northeast Texas Archaeology 47 (2014) REFERENCES CITED Brown, J. A. 1996 The Spiro Ceremonial Center. The Archaeology of Arkansas Valley Caddoan Culture in Eastern Okla- homa. 2 Vols. Memoir No. 29. Museum of Anthropology, University of Michigan, Ann Arbor. Early, A. M. (editor) Thurmond, J. P., and U. Kleinschmidt 1979 Report on the Fall 1978 Investigations at the George C. Davis Site, Caddoan Mounds State Historic Site, Cherokee County, Texas. Texas Archeological Research Laboratory, The University of Texas at Austin. Perino, G. 1981 Archeological Investigations at the Roden Site (MC-215), McCurtain County, Oklahoma. Potsherd Press No. 1. Museum of the Red River, Idabel. Journal of Northeast Texas Archaeology 47 (2014) 41 Early, A. M. (editor) 1993 Caddoan Saltmakers in the Ouachita Valley: The Hardman Site. Research Series No. 43. Arkansas Ar- cheological Survey, Fayetteville. Fields, R. C., E. F. Gadus, L. W. Klement, C. B. Bousman, and J. B. McLerran Fields, R. C., E. F. Gadus, L. W. Klement, C. B. Bousman, and J. B. McLerran 1993 Excavations at the Tick, Spike, Johns Creek, and Peerless Bottoms Sites, Cooper Lake Project, Delta & Hopkins Counties, Texas. Report of Investigations No. 91. Prewitt and Associates, Inc., Austin. Fields, R. C., E. F. Gadus, L. W. Klement, C. B. Bousman, and J. B. McLerran 1993 Excavations at the Tick, Spike, Johns Creek, and Peerless Bottoms Sites, Cooper Lake Project, Delta & Hopkins Counties, Texas. Report of Investigations No. 91. Prewitt and Associates, Inc., Austin. Harden, P. L. 1981 Excavations at Sliding Slab Shelter: Hunting and Gathering in the Ouachitas. Environmental Assess- ments, Inc., Pauls Valley. Jackson, H. E. 2005 “Darkening the Skies in Their Flight”: A Zooarchaeological Accounting of Passenger Pigeons in the Prehistoric Southeast. In Engaged Archaeology: Research Essays on North American Archaeology, Eth- nobotany, and Museology, edited by M. Hegmon and B. S. Eiselt, pp. 174-199. Anthropological Papers No. 94. Museum of Anthropology, University of Michigan, Ann Arbor. Jackson, H. E. and S. L. Scott 1995 The Faunal Record of the Southeastern Elite: The Implications of Economy, Social Relations, and Ideol- ogy. Southeastern Archaeology 14:103-119. Jackson, H. E., S. L. Scott, and F. F. Schambach 2012 At the House of the Priest: Faunal Remains from the Crenshaw Site (3MI6), Southwest Arkansas. In The Archaeology of the Caddo, edited by T. K. Perttula and C. P. Walker, pp. 47-85. University of Nebraska Press, Lincoln. s, D. A. and J. C. Neal Arkansas Birds: Their Distribution and Abundance. University of Arkansas Press, Fayetteville James, D. A. and J. C. Neal James, D. A. and J. C. Neal 1986 Arkansas Birds: Their Distribution and Abundance. University of Arkansas Press, Fayetteville. Kelley, D. B. (editor) 1997 Two Caddoan Farmsteads in the Red River Valley. Research Series No. 51. Arkansas Archeological Survey, Fayetteville. Kelly, L. S. 2012 Faunal Remains from Mound S, Toltec Mounds Site. In Toltec Mounds: Archeology of the Mound-and- Plaza Complex, by M. A. Rolingson, pp. 123-151. Research Series No. 65. Arkansas Archeological Survey, Fayetteville. 1957 Vertebrate remains from the Cahokia site, Illinois. Transactions of the Illinois Academy of Sciences 50:235-242. Early, A. M. (editor) Journal of Northeast Texas Archaeology 47 (2014) 41 Perino, G. 1981 Archeological Investigations at the Roden Site (MC-215), McCurtain County, Oklahoma. Potsherd Press No. 1. Museum of the Red River, Idabel. Thurmond, J. P., and U. Kleinschmidt 1979 Report on the Fall 1978 Investigations at the George C. Davis Site, Caddoan Mounds State Historic Site, Cherokee County, Texas. Texas Archeological Research Laboratory, The University of Texas at Austin.
https://openalex.org/W3000856589	https://ejurnal.its.ac.id/index.php/teknik/article/download/48493/5750	Indonesian	null	Perencanaan Perbaikan Tanah dan Perkuatan Tanggul Lepas Pantai Teluk Jakarta	Jurnal Teknik/Jurnal Teknik ITS	2,020	cc-by-sa	4,532	Perencanaan Perbaikan Tanah dan Perkuatan Tanggul Lepas Pantai Teluk Jakarta Apriana Hanggara Dewi, Yudhi Lastiasih, dan Herman Wahyudi Jurusan Teknik Sipil, Fakultas Teknik Sipil, Lingkungan, dan Kebumian, Institut Teknologi Sepuluh Nopember (ITS) e-mail: yudhi.lastiasih@gmail.com Abstrak-Jakarta merupakan Ibukota Indonesia. Jakarta terletak di pantai utara Pulau Jawa. Berada pada perairan laut Jawa menyebabkan Teluk Jakarta menjadi tempat bermuaranya sungai-sungai yang membelah kota Jakarta. Wilayah Jakarta mengalami penurunan tanah antara 10 cm sampai 180 cm, tergantung pada letak wilayahnya, sehingga mengancam kawasan Jakarta tergenang secara permanen. Salah satu solusi untuk menangani ancaman tergenangnya kawasan Jakarta adalah akan dibuat tanggul laut. Sehingga dilakukan penyelidikan tanah disepanjang Teluk Jakarta dimana susunan dan karakteristik lapisan tanahnya akan dikelompokan atas 15 (lima belas) ruas sungai dengan parameter kedalaman diukur dari MTS (Muka Tanah Setempat) dan pantai. Hasil dari penyelidikan tanah tersebut menyatakan bahwa sebagian besar kondisi tanah dasar dari Teluk Jakarta merupakan tanah lunak dengan nilai N-SPT < 10 dengan ketebalan yang bervariasi. Berdasarkan kondisi tanah seperti yang dijelaskan di atas, untuk mendapatkan tanggul yang stabil serta aman terhadap longsor dan juga gelombang air laut, maka pada Tugas Akhir ini perlu direncanakan dan dibandingkan tanggul dengan menggunakan sistem lereng dengan armour, secant pile dan SUPW (Sistem Urug dengan Perkuatan Wadah). Sedangkan untuk mengatasi permasalahan yang mungkin terjadi pada tanah dasar lunak yang harus menerima beban urugan, direncanakan perbaikan tanah tanah dengan metode pre-loading dan juga PVD (Prefabricated Vertical Drain). Dari hasil perencanaan PVD didapatkan hasil jarak pemasangan PVD masing-masing zona adalah antara 0,5 m-1,0 m dengan menggunakan pola pemasangan segitiga. Sedangkan untuk perkuatan lereng dengan armour, kemiringan lereng adalah 1:2 dengan tebal lapisan primer 1,2 m, lapisan sekunder 0,6 m dan berm 1,2 m. Besar angka keamanan untuk masing-masing zona adalah antara 1,5-1,96.. Selanjutnya perkuatan tanggul dengan secantpile, besar diameter secantpile masing-masing zona antara 1,6 m -2,0 m. Angka keamanan dari perkuatan secantpile adalah antara 5,74-8,7. Selanjutnya perkuatan SUPW, lebar pemasangan SUPW untuk masing-masing zona antara 11 m – 24,2 m. Angka keamanan yang didapatkan dari perkuatan SUPW adalah antara 1,54 – 1,9. Gambar 1. Peta penurunan muka tanah wilayah DKI Jakarta Gambar 1. Peta penurunan muka tanah wilayah DKI Jakarta Adanya ancaman tenggelamnya wilayah Jakarta maka dilakukan penanganan yaitu proyek bernama “Strategi Pengaman Pantai Jakarta” dengan nama lain NCICD (National Capital Integrated Coastal Development). Proyek tersebut merencanakan beberapa jenis solusi mulai jangka pendek sampai jangka panjang. Salah satu solusinya adalah akan dibuat tanggul laut. Berdasarkan hasil penyelidikan tanah disepanjang Teluk Jakarta dimana susunan dan karakteristik lapisan tanahnya akan dikelompokan atas 15 (lima belas) ruas sungai dengan parameter kedalaman diukur dari MTS (Muka Tanah Setempat) dan pantai. I. PENDAHULUAN J AKARTA merupakan Ibukota Indonesia. Jakarta terletak di pantai utara Pulau Jawa. Berada perairan laut Jawa, Teluk Jakarta merupakan tempat bermuaranya sungai-sungai yang membelah kota Jakarta. Pada wilayah pesisir pantai Jakarta mengalami penurunan tanah, menyebabkan kawasan Jakarta sedikit demi sedikit tenggelam. Besar penurunan tanah yang terjadi di walayah Jakarta adalah antara 10 cm sampai 180 cm, tergantung pada letak wilayahnya (seperti terlihat pada Gambar 1). Apabila tidak dilakukan tindakan apapun, maka kawasan tersebut terancam tergenang secara permanen. J Berdasarkan kondisi yang ada, maka dapat disimpulkan rumusan masalah yang perlu diperhatikan adalah: 1. Zoning data tanah berdasarkan N-SPT 2. Perencanaan perbaikan tanah dasar. 3. Perencanaan perkuatan tanggul lepas pantai. C102 C102 JURNAL TEKNIK ITS Vol. 8, No. 2, (2019) ISSN: 2337-3539 (2301-9271 Print) Perencanaan Perbaikan Tanah dan Perkuatan Tanggul Lepas Pantai Teluk Jakarta Apriana Hanggara Dewi, Yudhi Lastiasih, dan Herman Wahyudi Jurusan Teknik Sipil, Fakultas Teknik Sipil, Lingkungan, dan Kebumian, Institut Teknologi Sepuluh Nopember (ITS) e-mail: yudhi.lastiasih@gmail.com Dari hasil penyeledikan tersebut diperoleh berbagai jenis kondisi tanah dengan dominasi tanah merupakan soft clay. Berdasarkan kondisi tanah seperti yang dijelaskan di atas, untuk mendapatkan tanggul yang stabil serta aman terhadap longsor dan juga gelombang air laut, maka pada Tugas Akhir ini perlu direncanakan dan dibandingkan tanggul dengan menggunakan sistem lereng dengan armour, secant pile dan SUPW (Sistem Urug dengan Perkuatan Wadah). Sedangkan untuk mengatasi permasalahan yang mungkin terjadi pada tanah dasar lunak yang harus menerima beban urugan, direncanakan perbaikan tanah tanah dengan metode pre- loading dan juga PVD (Prefabricated Vertical Drain. Kata Kunci: Teluk Jakarta, Secant pile, SUPW. III. ANALISA DATA TANAH ' i i i h Si q E   =      ' i i i h Si q E   =      ' i i i h Si q E   =      (2) 2 2 0,5 370 / 6408,3 / m Si kN m kN m   =     = 0,0289 m ' i i i h Si q E   =      (2) (2) Pada perencanaan tugas akhir ini menggunakan data tanah berupa data SPT dengan jumlah data sebanyak 60 data. Untuk mempermudah analisa data tanah, maka dilakukan zoning berdasarkan jenis tanah dan besar nilai N-SPT. Hasil dari zoning, data tanah terbagi menjadi 6 zona. Pada zona 1,2, dan 4, didominasi oleh tanah lempung dengan konsistensi dari very soft sampai hard. Sedangkan untuk zona 3,5, dan 6, terdapat lapisan tanah pasir dengan konsistensi loose sampai dense, meskipun tetap didominasi oleh tanah lempung dengan konsistensi very soft sampau hard. 2 2 0,5 370 / 6408,3 / m Si kN m kN m   =     = 0,0289 m Setelah diketahui besar penurunan tanah, maka bisa dihitung didapatkan besar Hinitial timbunan. Tinggi Hinitial dapat dihitung dengan menggunakan persamaan berikut[1]: ( ) i i w initial timb q Sc H   + = (3) = 21,658 (3) (3) = 21,658 JURNAL TEKNIK ITS Vol. 8, No. 2, (2019) ISSN: 2337-3539 (2301-9271 Print) C103 IV. HASIL DAN PEMBAHASAN Setelah didapatkan Sc dan Hinitial, maka dapat dibuat grafik hubungan antara Hfinal dengan Hinitial seperti Gambar 2 , dan grafik hubungan Hfinal dengan Sc seperti Gambar 3. E. KESIMPULAN Kesimpulan diambil dari alternatif perencanaan perkuatan tanggul menggunakan metode sistem lereng dengan armour, secant pile, dan SUPW (Sistem Urug dengan Perkuatan Wadah) dipilih yang paling stabil. 0 0 . ' . log 1 1 ' c Cs H pc Cc H po p S e po e pc   +  = +   + +   (1) (1) Sc = 0,137 m Sc = 0,137 m Selanjutnya besar penurunan immediate tanah akibat timbunan[2]: B. PENGUMPULAN DAN ANALISA DATA Gambar 3. Hubungan Hfinal dengan Sc Data-data yang dipakai dalam perencanaan ini adalah data sekunder berupa data layout lokasi, data pengujian tanah SPT, dan data bathymetry. Gambar 3. Hubungan Hfinal dengan Sc • Perencanaan Perkuatan Tanah Perkuatan tanah dilakukan dengan membandingkan 3 (tiga) alternatif perkuatan yaitu perkuatan sistem lereng dengan armour, perkuatan secantpile, serta perkuatan SUPW (Sistem Urug dengan Perkuatan Wadah). Cek: Po’ + ΔP ≥ Pc’ 1,046 kN/m2+ 370,000 kN/m2 ≥ 25,546 kN/m2 371,046 ≥ 25,546 kN/m2 Sehingga besar penurunan konsolidasi tanah akibat timbunan: C. ZONIFIKASI DATA TANAH Dari beberapa data tanah dengan titik pengambilan di ruas ruas sungai pada Teluk Jakarta, dilakukan zonifikasi berdasarkan besar nilai N-SPT. A. Studi literatur Gambar 2. Hubungan Hfinal dengan Hinitial Gambar 3. Hubungan Hfinal dengan Sc Gambar 2. Hubungan Hfinal dengan Hinitial Studi literatur bertujuan untuk mempermudah penyelesaian tugas akhir ini. Studi literatur didapatkan dari diktat kuliah, buku-buku serta jurnal. Beberapa bahan referensi yang digunakan dalam mengerjakan tugas akhir ini adalah teori pemampatan, teori preloading, teori waktu konsolidasi, teori PVD (Prefabricated Vertikal Drain), teori perencanaan sistem lereng dengan armour, teori perencanaan secantpile, teori perencanaan SUPW (Sistem Urug dengan Perkuatan Wadah). Gambar 2. Hubungan Hfinal dengan Hinitial II. METODOLOGI Metodologi yang digunakan dalam penyelesaian tugas akhir ini adalah sebagai berikut: D. PERENCANAAN PERBAIKAN TANAH DAN PERKUATAN TIMBUNAN • Perencanaan Perbaikan Tanah • Perencanaan Perbaikan Tanah Rencanaan perbaikan tanah dengan menggunakan metode pre-loading. Apabila dalam dalam proses pre- loading waktu yang dibutuhkan tanah untuk pemampatan lebih dari 6 bulan, maka dilakukan upaya untuk mempercepat waktu pemampatan dengan menggunakan PVD (Prefabricated Vertical Drain). Rencanaan perbaikan tanah dengan menggunakan metode pre-loading. Apabila dalam dalam proses pre- loading waktu yang dibutuhkan tanah untuk pemampatan lebih dari 6 bulan, maka dilakukan upaya untuk mempercepat waktu pemampatan dengan menggunakan PVD (Prefabricated Vertical Drain). settlement) dan penurunan akibat penurunan penurunan konsolidasi (consolidation settlement). Penurunan terjadi akibat adanya tegangan-tegangan pada tanah. Tegangan yang terjadi antara lain tegangan overburden (Po’), tegangan prakonsolidasi (Pc’), dan distribusi tegangan (ΔP)[1]. p ( ) g g ( Berikut merupakan contoh perhitungan penurunan konsolidasi tanah pada zona 1: B. PVD Berdasarkan pada analisa data tanah, didapatkan hasil tanah didominasi oleh tanah lempung, sehingga pemampatan konsolidasi yang terjadi lambat. Pemasangan Prefabricated Vertical Drain (PVD) digunakan untuk mempercepat waktu penurunan tanah. Perencanaan dilakukan dengan menggunakan PVD dengan pola pemasangan segitiga dan pola pemasangan segiempat. Data spesifikasi PVD yang digunakan adalah: a = 100 mm b = 5 mm B ik k h hi PVD [1] a = 100 mm b = 5 mm a = 100 mm b = 5 mm Gambar 5. Garfik hubungan waktu konsolidasi dengan derajat konsolidasi untuk pola pemasangan segiempat Berikut merupakan contoh perhitungan PVD [1]: Cv = 0,000104 cm2/s = 0,0063 m2/minggu Ch = 3 x Cv = 3 x 0,0063 m2/minggu = 0,0188 m2/minggu Di k Berdasarkan hasil analisa didapatkan kedalaman dan jarak pemasangan tiap zona dapat dilihat pada Tabel 2. Dengan ketinggian pelaksanaan antara 21 m hingga 23 m, apabila hanya dilakukan perbaikan tanah saja maka timbunan tidak stabil sehingga perlu direncanaan perkuatan timbunan. Direncanakan: Direncanakan: Direncanakan: F(n)= 2 2 2 2 3 1 ln( ) 1 4 4 n n n n      − −       −       (4) = 1,351 Uh = 2 8 2 ( ) 1 1 t Ch D F n e       −       = 0,18 (5) Tv = ( ) 2 dr t Cv H  = 0,000030 (6) Ūv = 2 Tv          = 0,00616 (7) U rata-rata = 1 (1 )(1 ) 100% U Uh Uv   = − − −    (8) = [1 – (1-0,183)(1-0,00616)] x 100% = 18,909 (4) JURNAL TEKNIK ITS Vol. 8, No. 2, (2019) ISSN: 2337-3539 (2301-9271 Print) JURNAL TEKNIK ITS Vol. 8, No. 2, (2019) ISSN: 2337-3539 (2301-9271 Print) JURNAL TEKNIK ITS Vol. 8, No. 2, (2019) ISSN: 2337-3539 (2301-9271 Print) C104 Tabel 2 Besar Sc dan Hinitial pada setiap Zona Zona Sc (m) Hinitial (m) Waktu Pemampatan (tahun) 1 3,283 22,918 545,092 2 2,400 22,035 362,890 3 1,692 21,327 43,141 4 2,736 22,371 225,664 5 3,366 23,001 583,415 6 1,041 21,176 29,983 Tabel 1 Pemasangan PVD pada setiap Zona Zona Pola pemasangan Jarak antar PVD Kedalaman PVD (m) (m) (m) 1 Segitiga 0,6 14,5 2 0,6 12,5 3 1,0 2,5 4 0,9 14,5 5 0,5 12,5 6 0,5 2,5 Tabel 2 Besar Sc dan Hinitial pada setiap Zona Gambar 4. Garfik hubungan waktu konsolidasi dengan derajat konsolidasi untuk pola pemasangan segitiga Tabel 1 Pemasangan PVD pada setiap Zona Zona Pola pemasangan Jarak antar PVD Kedalaman PVD (m) (m) (m) 1 Segitiga 0,6 14,5 2 0,6 12,5 3 1,0 2,5 4 0,9 14,5 5 0,5 12,5 6 0,5 2,5 Tabel 1 Pemasangan PVD pada setiap Zona Tabel 1 Pemasangan PVD pada setiap Zona Gambar 4. Garfik hubungan waktu konsolidasi dengan derajat konsolidasi untuk pola pemasangan segitiga Gambar 4. Garfik hubungan waktu konsolidasi dengan derajat konsolidasi untuk pola pemasangan segitiga pada zona lain. Hasil besar settlement dan Hinitial dari masing- masing zona dapat dilihat pada Tabel 1. Gambar 5. Garfik hubungan waktu konsolidasi dengan derajat konsolidasi untuk pola pemasangan segiempat Karena pemampatan tanah membutuhkan waktu yang lama, untuk mempercepat waktu pemampatan maka dipasang PVD. A. Settlement Penambahan beban pada permukaan tanah menyebabkan lapisan tanah di bawahnya mengalami pemampatan. Penurunan yang diperhitungkan pada perencanaan ini terdiri dari penurunan akibat penurunan segera (immediate Setelah dibuat grafik, maka akan didapatkan persamaan untuk mencari nilai Hinitial dan Sc. Dilakukan yang sama D. Secant pile p t = 1/3 t W n k         = 0,927 m (14) (14) , , 2) Perhitungan Lapisan Pelindung Pertama Berat lapisan dihitung dengan menggunakan rumus sebagai berikut[3]: 3 3 ( 1) cot D H W K Sr    =  −  (11) = 111,834 kg Diameter batu D = 1/3 W        = 0,223 m (12) Lebar mercu B = 1/3 p W n k         = 0,739 m (13) Po’ = 21,970 kN/m2 ΔP = 203,746 kN/m2 Besar settlement pada lapisan tanah 1 dihitung dengan cara 1 Be [1]: (12) (12) 0 0 . ' . log 1 1 ' c Cs H pc Cc H po p S e po e pc   +  = +   + +   (17) Sc = 1,55 m (17) (17) Sc = 1,55 m B = 1/3 p W n k         = 0,739 m (13) Untuk zona lainnya dihitung dengan cara yang sama dengan hasil terlihat pada Tabel 3. (13) ≈ 2 meter (untuk keamanan dan maintance) ≈ 2 meter (untuk keamanan dan maintance) Tebal Lapisan ≈ 2 meter (untuk keamanan dan maintance) Tebal Lapisan C. Perkuatan Lereng dengan sistem armour Berikut merupakan contoh perhitungan perkuatan armour pada Zona 1 : (5) 1) Menentukan Elevasi Bangunan Elevasi muka air = MHWL + Sw + SLR 1) Menentukan Elevasi Bangunan Elevasi muka air = MHWL + Sw + SLR 1) Menentukan Elevasi Bangunan Elevasi muka air = MHWL + Sw + SLR dimana: MHWL = Elevasi rata-rata muka air = 17, 5 m Sw = Wave Set Up p dimana besar Sw dihitung dengan cara g g 2 0,19 1 2,82 H Sw H gT   = −     (9) (9) = 0,112 m SLR = Sea Level Rise = 0,9946 m = 0,112 m SLR = Sea Level Rise = 0,9946 m Sehingga elevasi muka air : Maka, setelah diketahui besar Urata-rata, bisa dibuat grafik hubungan waktu pemampatan dengan derajat konsolidasi, pada Gambar 5 grafik untuk PVD pola pemasangan segitiga dan Gambar 6 grafik untuk PVD pola pemasangan segiempat. r = MHWL + Sw + SLR = 17,5 + 0,112 + 0,9946 = 18,607 m ≈ 18,700 m Elevasi muka air = JURNAL TEKNIK ITS Vol. 8, No. 2, (2019) ISSN: 2337-3539 (2301-9271 Print) JURNAL TEKNIK ITS Vol. 8, No. 2, (2019) ISSN: 2337-3539 (2301-9271 Print) C105 = 415 /40 = 10,375 kN/m’ q ult = CNc + ½γ B Nγ (15) = (24,516)(5,14) + ½ (4,185) (40)(0) = 126,0102 kN/m’ SF = q ult / qmaks = 126,0102 / 10,375 = 12,146 > 3 (Memenuhi) 4) Settlement Armor Po’ = 21,970 kN/m2 ΔP = 203,746 kN/m2 B ttl t d l i t h 1 dihit d Tabel 4. Gaya Aktif Tanah Gaya Lengan (m) Momen Pa1 8,130 1,250 10,162 Pa2 15,667 1,667 26,111 Pa3 276,240 8,750 2,417,104 Pa4 361,010 11,667 4,211,778 Pa5 82,083 18,750 1,539,051 Pa6 9,679 19,167 185,518 Pa7 609,395 22,500 13,711,391 Pa8 40,562 23,333 946,437 Pa9 138,104Do 25+0,5Do 3453Do + 69,052 Do2 Pa10 1,622Do2 25+0,67Do 40,562 Do2 + 1,082 Do3 ∑Ma 23047,552+3452,6Do+109,613Do2+1, 082Do3 Tabel 5. Gaya Pasif Tanah Gaya Lengan (m) Momen Pp1 Pp2 Pp3 Pp4 Pp5 Pp6 346,199 46,572 357,595 89,676 107,390Do 3,587Do2 18,750 19,167 22,500 23,333 25+0,5Do 25+0,67Do 6491,236 892,634 8045,888 2092,446 2685Do + 52,695 Do2 89,676 Do2 + 2,391 Do3 ∑Ma 17522,204+2684,7Do+143,3 71Do2+2,391Do3 Tabel 4. Gaya Aktif Tanah Elevasi puncak = EL.rencana + Run up + Freeboard dimana : Tabel 3. C. Perkuatan Lereng dengan sistem armour Rekapitulasi Desain Armour tiap Zona Zona Kemiringan Tebal 1 Tebal 2 Berm SF overall (m) (m) (m) 1 1:2 1,2 0,6 1,2 1,51 2 1:2 1,2 0,6 1,2 1,9 3 1:2 1,2 0,6 1,2 1,53 4 1:2 1,2 0,6 1,2 1,96 5 1:2 1,2 0,6 1,2 1,52 6 1:2 1,2 0,6 1,2 1,34 Tabel 3. Rekapitulasi Desain Armour tiap Zona Tabel 3. Rekapitulasi Desain Armour tiap Zona Tabel 4. Gaya Aktif Tanah Elevasi puncak = EL.rencana + Run up + Freeboard dimana : EL.rencana = 18,700 m Run up = kenaikan gelombang EL.rencana = 18,700 m EL.rencana = 18,700 m Run up = kenaikan gelombang Run up = kenaikan gelombang Berikut merupakan contoh perhitungan untuk menentukan bilangan Ir [3]: H = 0,720 m T = 4,3 detik Lo = 1,56 T2 = 1,56 (4,32) = 28,844 tg θ = 0,5 ( ) 0,5 r tg I H Lo  = ( ) 0,5 0,5 0,720 28,844 = = 3,165 (10) Ru/H = 1,3 Ru = H x 1,3 = 0,720 x 1,3 = 0,936 m ≈ 1,00 m Sehingga tinggi elevasi adalah: Sehingga tinggi elevasi adalah: (15) Elevasi puncak Elevasi puncak Kontrol Overall stability Berdasarkan permodelan dengan menggunakan program bantu didapatkan nilai safety faktor sebesar 5,97 > 1,5 sehingga stabilitasnya dapat dikatakan aman dan stabil. Hasil dari desain perkuatan secantpile dapat dilihat pada Tabel 6. Tinggi timbunan pelaksana dan besar penurunan tanag tiap zona adalah sebagai berikut: Berikut merupakan data perencaan : Berikut merupakan data perencaan : - Zona 4 Hi = 22,371 m Sc = 1,637 m Sc = 2,736 m Zona 5 Hi = 23,001 m Sc = 3,366 m V. KESIMPULAN Dari 60 data tanah berupa SPT, dilakukan zoning dengan berdasarkan nilai SPT. Hasil dari zoning yang dilakukan didapatkan 6 zona. Untuk zona 1,2 dan 4 didominasi tanah lempung dengan konsistensi very soft sampai hard. Untuk zona 3,5, dan 6 terdapat lapisan tanah pasir dengan konsistensi very loose sampai dense, namun tetap didominasi tanah lempung dengan konsistensi very soft sampai hard. ∑ ∑M = -1,310 Do3-33,758 Do2+767,854 Do+5525,348 dilakukan penurunan, sehingga persamaan menjadi: ∑M = -1,310 Do3-33,758 Do2+767,854 Do+5525,348 dilakukan penurunan, sehingga persamaan menjadi: ∑M = -1,310 Do3-33,758 Do2+767,854 Do+5525,348 dilakukan penurunan, sehingga persamaan menjadi: M = -3,929 Do2 – 67,515 Do + 767,854 Momen maksimum adalah: Do1= 7,817 m →dipilih Do2= - 53,479 m Do1= 7,817 m →dipilih Do2= - 53,479 m Sehingga besar momen maksimum adalah: Sehingga besar momen maksimum adalah: E. SUPW (Sistem Urug dengan Perkuatan Wadah) g - Zona 1 - Zona 2 Hi = 22,918 m Hi = 22,035 m Sc = 3,283 m Sc = 2,400 m - Zona 3 - Zona 4 Hi = 21,691 m Hi = 22,371 m Sc = 1,637 m Sc = 2,736 m - Zona 5 - Zona 6 Hi = 23,001 m Hi = 21,176 m Sc = 3,366 m Sc = 1,041 m Zona 1 - Zona 2 Berikut merupakan data perencaan : Lebar panel = 2,200 m Tinggi panel = 1,500 m fc’ = K-350 Tulangan = 5D25 γtanah isi = 14,000 kN/m3 Berikut merupakan data perencaan : 2) Perencanaan Tulangan 2) Perencanaan Tulangan Dengan program bantu, dicek pemasangan tulangan 40D36 dengan besar ρ = 2,00 % < 6%. Sc = 1,51 m Sc = 1,51 m Sehingga besar settlement total: Sehingga besar settlement total: g ρ Kapasitas momen yang sanggup ditahan oleh secantpile sebesar 9212 kNm < Mmaks = 8839,282 kNm Sc = Sc+Si= 0,428 + 0,0026 = 0,431 m Sc = Sc+Si= 0,428 + 0,0026 = 0,431 m Hasil dari perencanaan perkuatan SUPW dapat dilihat pada Tabel 7. 3) Stabilitas Secantpile - Stabilitas Geser ∑Pp = 21861,979 kN ∑Pa = 2449,339 kN SF = ∑Pp/∑Pa = 21861,979 kN /2449,339 kN = 8,93 > 1,5 - Kontrol Overall stability 3) Stabilitas Secantpile - Stabilitas Geser ∑Pp = 21861,979 kN ∑Pa = 2449,339 kN SF = ∑Pp/∑Pa = 21861,979 kN /2449,339 kN = 8,93 > 1,5 - Kontrol Overall stability 3) Menentukan besar settlement Besar settlement pada lapisan tanah 1 dihitung dengan cara 3) Menentukan besar settlement Besar settlement pada lapisan tanah 1 dihitung dengan cara Mmaks= -1,310 Do3-33,758 Do2+767,854 Do+5525,348 = 8839,282 kNm 3) Menentukan besar settlement Besar settlement pada lapisan tanah 1 dihitung dengan cara [1]: [1]: [1]: = 8839,282 kNm 0 0 . ' . log 1 1 ' c Cs H pc Cc H po p S e po e pc   +  = +   + +   (20) (20) 1) Mencari kedalaman penanam secantpile Mencari kedalaman penanam secantpile dengan cara mencari sigma momen yang terjadi akibat gaya aktif dan pasif tekanan tanah dan juga tekanan air. Setelah diketahui persamaan momennya maka akan didapatkan panjang secantpile yang masuk kedalam tanah[5]. Dari perhitungan ≈ 1 m ≈ 1 m Dengan cara yang sama dilakukan perhitungan untuk lapisan kedua dan lapisan berm. 3) Stabilitas Armor 3) Stabilitas Armor Berdasarkan SNI 8460 Pasal 7.5.5 Tahun 2017[4], SF rencana diambila 1,5.Setelah dilakukan permodelan menggunakan program bantu, didapatkan nilai safety faktor sebesar 1,51 > 1,5 sehingga stabilitasnya dapat dikatakan aman dan stabil untuk menahan sliding atau tergelincirnya armour. Tabel 6. Rekapitulasi Desain Secantpile tiap Zona Zona D (mm) Tul lentur Panjang pemasangan (m) SF Geser SF overall Displace- ment (mm) 1 2 3 4 5 6 2000 1600 1800 1600 2000 2000 45D57 40D36 52D36 42D36 46D57 40D57 61 55 57 53 61 60 5,44 8,93 7,85 9,7 5,25 5,05 5,74 5,97 8,7 8,17 6,58 6,9 25,2 21,3 21,6 19,3 25,3 24,6 Tabel 6. Rekapitulasi Desain Secantpile tiap Zona Sedangkan untuk besar daya dukung tanah dihitung dengan cara berikut[5]: [ ] ∑V = 415 kN/m2 A = 40 m q maks = ∑V / A (14) ∑V = 415 kN/m2 A = 40 m q maks = ∑V / A (14) ∑V = 415 kN/m2 A = 40 m (14) q maks = ∑V / A JURNAL TEKNIK ITS Vol. 8, No. 2, (2019) ISSN: 2337-3539 (2301-9271 Print) C106 gaya aktif da pasif, maka akan diperoleh besar momen terjadi, yaitu: gaya aktif da pasif, maka akan diperoleh besar momen terjadi, yaitu: Tabel 7. Rekapitulasi Desain SUPW tiap Zona Zona Lebar (m) SF overall SF terhadap bearing capacity Sc (m) 1 2 3 4 5 6 24,2 13,2 13,2 11 24,5 22 1,57 1,66 1,65 1,9 1,54 1,63 3,48 28,29 4,34 27,16 7,2 3,13 1,561 0,656 0,475 0,641 1,498 0,585 Tabel 7. 1) Mencari kedalaman penanam secantpile Rekapitulasi Desain SUPW tiap Zona yaitu: ∑M = Ma-Mp = [1,082Do3+109,613 Do2+ 3452,6 Do+23047,552] – [2,391 Do3+143,371 Do2+ 2684,7 Do+17522,204] ∑M = -1,310 Do3-33,758 Do2+767,854 Do+5525,348 dengan cara coba-coba, didapatkan nilai Do Do1= 18,43544 →dipilih Do2= -38,2186 Do3= -5,98632 SF rencana = 1,5 Do = 1,5 x 18,43544 = 27,6531 m ≈ 28 m Panjang secantpile yang tertanam kedalam tanah : 2,5 + 5 + 28 = 35,5 m Panjang total dari secantpile = 20 m + 35,5 m = 55,5 m Setelah diketahui kedalaman dari secantpile, selanjutnya mencari momen maksimum yang terjadi. Momen maksimum t j di di i d l k k t h d y ∑M = Ma-Mp = [1,082Do3+109,613 Do2+ 3452,6 Do+23047,552] – [2,391 Do3+143,371 Do2+ 2684,7 Do+17522,204] Zon 1 2 ∑M = -1,310 Do -33,758 Do +767,854 D dengan cara coba-coba, didapatkan nilai Do dengan cara coba-coba, didapatkan nilai Do Hasil dari permodelan pada program bantu Geofive didapatkan hasil sebagai berikut: SF rencana = 1,5 Lebar = 11 x2,2 = 24,2 m SF = 1,57 > 1,5 (Memenuhi) Lebar = 11 x2,2 = 24,2 m SF = 1,57 > 1,5 (Memenuhi) Lebar = 11 x2,2 = 24,2 m SF = 1,57 > 1,5 (Memenuhi) Panjang secantpile yang tertanam kedalam tanah : Panjang secantpile yang tertanam kedalam tanah : j g p y g 2,5 + 5 + 28 = 35,5 m 2) Menentukan Bearing Capacity dengan menggunakan cara berikut[5]: Panjang total dari secantpile = 20 m + 35,5 m = 55,5 cara berikut[5]: q maks = ∑V / A (18) = 339,835 kN/m2/24,2 = 24,274 kN/m’ q ult = 1,3 CNc + q Nq + 0,4 γ B Nγ (19) = 1,3 (42,947)(5,14) + (43,939 x 1) +0,4 (4,185) (24,2) (0) = 339,835 kN/m’ SF = q ult / qmaks = 339,835 / 24,274 = 3,485 > 3 (Memenuhi) Setelah diketahui kedalaman dari secantpile, selanjutnya mencari momen maksimum yang terjadi. Momen maksimum yang terjadi dicari dengan melakukan penurunan terhadap ∑M. (18) 1) Menentukan lebar SUPW Sc = 3,366 m Dengan menggunakan program bantu, dilakukan permodelan untuk mencari lebar SUPW untuk menahan timbunan agar stabil. Pola pemasangan PVD yang dipilih adalah pola pemasangan segitiga. Jarak pemasangan PVD untuk Zona 1 dan 2 adalah 0,6 m; Zona 3 adalam 1 m; Zona 4 adalah 0,9 m; sedangkan untuk Zona 5 dan 6 jarak pemasangannya SF rencana = 1,5 (SNI 8460 Pasal 7.5.5. Tahun 2017)[4] SF rencana = 1,5 (SNI 8460 Pasal 7.5.5. Tahun 2017)[4] JURNAL TEKNIK ITS Vol. 8, No. 2, (2019) ISSN: 2337-3539 (2301-9271 Print) C107 adalah 0,5 m. PVD dipasang hingga kedalaman tanah compressible. 11 m atau 5 panel, sedangkan untuk zona 6 adalah 22 m atau 10 panel. Dengan lebar tersebut, dengan menggunakan program bantu didapatkan nilai stabilitasnya adalah antara 1,54 sampai 1,9. Dan juga nilai SF terhadap bearing capacity antara 9,09 sampai dengan 22,46. Hasil perencanaan perkuatan lereng dengan sistem armour dengan kemiringan 1:2 didapatkan hasil tebal primary layer setebal 1,2 m, secondary layer setebal 0,6 m, dan berm setebal 1,2 m. Setelah di cek dengan menggunakan program bantu, didapatkan besar stabilitasnya untuk zona 1, zona 3, zona 5, dan zona 6 adalah sebesar 1,5 sampai 1,53. Sedangkan untuk zona 2 dan zona 4 adalah 1,9 dan 1,96. Selain itu, dicek besar kapasitas daya dukung tanah adalah antara 11,5 sampai 62,9. Sehingga dapat disimpulkan, dari hasil perencanaan perkuatan lereng dengan sistem armor, perkuatan secantpile dan perkuatan SUPW apabila dipilih perkuatan yang ditinjau dari segi kekuatan dan stabilitasnya maka dipilih perkuatan dengan menggunakan secantpile. Perkuatan Secantpile didapatkan hasil kedalaman tanam secantpile adalah antara 33 m sampai dengan 41 m. Pada zona 1, zona 5, dan zona 6 menggunakan diameter 2 m, untuk zona 2 dan zona 4 menggunakan diameter 1,6 m, sedangkan untuk zona 3 menggunakan 1,8 m. Setelah dicek dengan program bantu didapatkan hasil SF antara 5,74 sampai 8,7 . Untuk besar displacement antara 19,3 mm sampai 25,3 mm. DAFTAR PUSTAKA [1] Mochtar, Noor Endah.2012.Modul Ajar Metode Perbaikan Tanah. S b I tit t T k l i S l h N b [1] Mochtar, Noor Endah.2012.Modul Ajar Metode Perbaikan Tanah. Surabaya: Institut Teknologi Sepuluh Nopember. [1] Mochtar, Noor Endah.2012.Modul Ajar Metode Perbaikan Tanah. Surabaya: Institut Teknologi Sepuluh Nopember. [2] Wahyudi, Herman. 1997. Teknik Reklamasi. Surabaya. Jurusan Teknik Sipil. [2] Wahyudi, Herman. 1997. Teknik Reklamasi. Surabaya. Jurusan Teknik Sipil. [3] Keputusan Menteri Pekerjaan Umum Nomor 498/KPTS/M/2005. Perencanaan Jeti Tipe Rubble Mound untuk Penanggulangan Penutupan Muara Sungai oleh Sedimen [3] Keputusan Menteri Pekerjaan Umum Nomor 498/KPTS/M/2005. Perencanaan Jeti Tipe Rubble Mound untuk Penanggulangan Penutupan Muara Sungai oleh Sedimen Hasil perencanaan perkuatan timbunan dengan menggunakan perkuatan SUPW dengan ukuran panel SUPW adalah 2,2 m x 1,5 m. Untuk zona 1 dan zona 5 dibutuhkan lebar 24,2 m atau 11 panel. Untuk zona 2 dan zona 3 dibutuhkan lebar 13,2 m atau 6 panel. Untuk zona 4 adalah [4] SNI 2847-2013. Tata Cara Perencanaan Struktur Beton Untuk Bangunan Gedung. [5] Das, Braja M. (translated by Mochtar N.E, and Mochtar I.B.). 1994. Mekanika Tanah (Prinsip-prinsip Rekayasa Geoteknik) Jilid II. Jakarta: Erlangga [5] Das, Braja M. (translated by Mochtar N.E, and Mochtar I.B.). 1994. Mekanika Tanah (Prinsip-prinsip Rekayasa Geoteknik) Jilid II. Jakarta: Erlangga
W3173260374.txt	https://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf	en		Diagenetic Controls on the Formation of the Anarraaq Clastic-Dominated Zn-Pb-Ag Deposit, Red Dog District, Alaska	Economic geology and the bulletin of the Society of Economic Geologists	2,021	cc-by	15,129	Economic Geology, v. 116, no. 8, pp. 1803–1824 Diagenetic Controls on the Formation of the Anarraaq Clastic-Dominated Zn-Pb-Ag Deposit, Red Dog District, Alaska Merilie A. Reynolds,1,† Sarah A. Gleeson,2,3 Robert A. Creaser,4 Betsy A. Friedlander,5 Jenny C. Haywood,5 Danny Hnatyshin,6,7 Jim McCusker,5 and John W.F. Waldron4 1Northwest Territories Geological Survey, Government of Northwest Territories, Yellowknife, Canada 2GFZ 3Institute of Geological Sciences, Freie Universität Berlin, Malteserstrasse 74-100, Berlin 12249, Germany 4Department 5Teck German Research Centre for Geosciences, 14473 Potsdam, Germany of Earth and Atmospheric Sciences, University of Alberta, Edmonton, Alberta T6G 2E3, Canada Resources Limited, Suite 3300, Bentall 5, 550 Burrard Street, Vancouver, British Columbia V6C 0B3, Canada 6Irish Centre for Research in Applied Geosciences (iCRAG), University College Dublin, Belfield, Dublin 4, Ireland 7UCD School of Earth Sciences, University College Dublin, Belfield, Dublin 4, Ireland Abstract The Anarraaq clastic-dominated (CD) Zn-Pb-Ag deposit (Red Dog district, Alaska, USA) has an inferred mineral resource of 19.4 Mt at 14.4% Zn, 4.2% Pb, and 73 g/t Ag and is spatially associated with a separate ~1 Gt barite body. This study presents new cross sections and petrographic evidence from the Anarraaq area. The barite body, previously shown to have formed in a shallow subsurface environment akin to a methane cold seep, contains multiple generations of barite with locally abundant calcite masses, which are discordant to sedimentary laminae, and is underlain by an interval of massive pyrite containing abundant framboids and radiolarians. Calcite and pyrite are interpreted to have formed by methane-driven diagenetic alteration of host sediment at the sulfate-methane transition (SMT). The sulfide deposit contains two zones of Zn-Pb mineralization bounded by faults of unknown displacement. The dominant hydrothermal minerals are marcasite, pyrite, sphalerite, quartz, and galena. The presence of hydrothermal pseudomorphs after barite, early pyrite resembling diagenetic pyrite associated with the barite body, and hydrothermal quartz and sphalerite filling voids formed by dissolution of carbonate all suggest that host sediment composition and origin was similar to that of the barite body prior to hydrothermal mineralization. Rhenium-osmium isochron ages of Ikalukrok mudstone (339.1 ± 8.3 Ma), diagenetic pyrite (333.0 ± 7.4 Ma), and hydrothermal pyrite (334.4 ± 5.3 Ma) at Anarraaq are all within uncertainty of one another and of an existing isochron age (~338 Ma) for the Main deposit in the Red Dog district. This indicates that the Anarraaq deposit formed soon after sedimentation and that hydrothermal activity was approximately synchronous in the district. The initial Os composition of the Anarraaq isochrons (0.375 ± 0.019–0.432 ± 0.025) is consistent with contemporaneous seawater, indicating that a mantle source was not involved in the hydrothermal system. This study highlights the underappreciated but important role of early, methane-driven diagenetic processes in the paragenesis of some CD deposits and has important implications for mineral exploration. Introduction The Red Dog district in northwestern Alaska contains several world-class clastic-dominated (CD) Zn-Pb deposits (Fig. 1; Blevings et al., 2013). CD Zn-Pb deposits are a subtype of sediment-hosted massive sulfide (SHMS) deposits that are hosted in carbonaceous mudstone (Leach et al., 2010); this type of deposit has traditionally been referred to as sedimentary exhalative (SEDEX; Goodfellow et al., 1993). Within the Red Dog district, the Anarraaq sulfide deposit contains an inferred mineral resource of 19.4 Mt at 14.4% Zn, 4.2% Pb, and 73.4 ppm Ag (Krolak et al., 2017). Barite and Fe sulfide (pyrite and marcasite) are common components of the Red Dog CD deposits. Importantly, barite produces a strong gravity anomaly that has proven to be a powerful tool in exploration of the district (Blevings et al., 2013). However, barite and Zn-Pb-Fe sulfides also occur independent of each other in the district, and Fe sulfide is the dominant component in some †Corresponding author: e-mail, Merilie_Reynolds@gov.nt.ca low-grade Zn-Pb prospects in the Red Dog district (Fig. 1; Kelley and Jennings, 2004; Blevings et al., 2013). The genetic controls on the varying relationships of these components are not well understood. The Anarraaq area of the Red Dog district is an ideal place to investigate the relationships between diagenetic and hydrothermal processes because it contains a sulfide-only deposit and a barite-only body (Kelley et al., 2004b), and recent drilling allows good geologic control. The Anarraaq barite body could contain as much as 1 Gt of barite and is separated from the sulfide deposit below by 20–60 m of barren host rock (Kelley et al., 2004b). Kelley et al. (2004b, p. 1586) reported only rare barite in the Anarraaq sulfide deposit or in laterally equivalent strata, concluding that it was “unlikely that the sulfides replaced barite.” They instead interpret sulfides to have replaced sedimentary carbonate layers within the mudstone (Kelley et al., 2004b). In this study, we present new cross sections and petrographic data from the Anarraaq barite body and the Anarraaq sulfide deposit and document the distribution of the major © 2021 Gold Open Access: This paper is published under the terms of the CC-BY-NC license. ISSN 0361-0128; doi:10.5382/econgeo.4849; 22 p. Digital appendices are available in the online Supplements section. 1803 Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest Submitted: February 18, 2019 / Accepted: February 8, 2021 1804 REYNOLDS ET AL. 163o00’00’’ Red Dog Alaska Ak Anarraaq (Fig. 3) SL A Q 4km M 68o05’00’’ Mine Area P N Late Jurassic-Cretaceous Okpikruak Fm Structural framework for Devonian-Jurassic sedimentary rocks of the Arctic Alaska terrane: Endicott Mountains subterrane Endicott Mountains Allochthon De Long Mountains subterrane I.P. Creek sheet (not pictured) Key Creek sheet Red Dog sheet hosts Zn-Pb ore Wolverine Creek sheet Projected to surface: Zn-Pb deposit Barite body Sulfide prospect Barite occurrence Geologic symbols normal fault thrust fault fault, unknown sense stratigraphic contact structural vergence Fig. 1. Regional map of thrust sheets in the Red Dog district. Modified after figures 2, 3, and 14 in Blevings et al. (2013). Mine area deposits include Main (M), Aqqaluk (A), Paalaaq (P), and Qanaiyaq (Q). Other deposits include Anarraaq, Aktigiruk (Ak), and Su Lik (SL). Several additional prospects and barite occurrences are also highlighted. paragenetic components on the deposit scale. We also present the first paragenetic scheme for the deposit that includes both sulfide and gangue minerals and show that hydrothermal sulfide mineralization did replace diagenetic barite. Finally, rhenium-osmium isotope geochemistry is used to constrain the age of the Anarraaq sulfide deposit relative to host sediment, diagenetic alteration, and other Red Dog deposits. Regional Geology The Red Dog district is in the Arctic Alaska terrane, which comprises Devonian to Jurassic sedimentary rocks deposited on a passive continental margin (Moore et al., 1986, 1994). The area underwent significant shortening during the Late Jurassic to Cretaceous Brookian orogeny, when the terrane collided with an intraoceanic island arc (Moore et al., 1986, 1994). District geology is characterized by thrusts, folds, and very low grade metamorphism (De Vera et al., 2004). The region has been divided into multiple fault-bounded and age-equivalent allochthons and structural plates (referred herein as sheets; Fig. 1; Young, 2004; Blevings et al., 2013). All known CD deposits in the Red Dog district are hosted in the Red Dog thrust sheet of the Endicott Mountains allochthon (Figs. 1, 2; Blevings et al., 2013). In the Anarraaq area, structural vergence is to the northeast (Fig. 1; De Vera, 2005), and thrust faults are offset by steeply dipping tear faults that accommodate variable lateral displacement (Fig. 3; Blevings et al., 2013). The Red Dog CD deposits are hosted in the Ikalukrok unit, which makes up the upper part of the Mississippian Kuna Formation and is the deep-water facies of the Lisburne Group (Dumoulin et al., 2004). The Kuna Formation was deposited in a sediment-starved, extensional basin flanked by carbonate platforms (Young, 2004). The Ikalukrok unit primarily comprises black biosiliceous mudstone and chert with locally abundant intercalations of carbonate and barite (Dumoulin et al., 2004; Johnson et al., 2004) and cut by igneous intrusions linked to extensional and transtensional deformation (Young, 2004). Rocks of the Ikalukrok unit are characterized by high concentrations of organic carbon, silicon, and phosphate, with abundant radiolarians (Dumoulin et al., 2004, 2014; Slack et al., 2004). Upwelling of nutrient-rich waters played an important role in the regional depositional regime (Dumoulin et al., 2014) and may have given rise to a fluctuating oxygen minimum zone that resulted in variable redox conditions (Reynolds et al., 2015). Carbonate layers in the Ikalukrok can be up to 40 m thick and up to 80 m in cumulative thickness and are generally thought to originate from the erosion of adjacent carbonate platforms (Dumoulin et al., 2004). The abundance of calcareous mass flow deposits in the Ikalukrok increases from south to north and is relatively high at Anarraaq (Dumoulin et al., 2004). The model for Red Dog CD Zn-Pb mineralization calls for a saline brine (14–19 wt % NaCl equiv) that originated from evaporated seawater and was heated to 100°–200°C as it circulated in the subsurface (Leach et al., 2004). The source of metals is inferred to be the fluvial-deltaic sandstones and conglomerates of the Devonian Endicott Group below the Kuna Formation (Young, 2004). The hydrothermal fluid ascended to the surface via extensional faults; some sulfides were deposited in shallow muds or replaced barite (e.g., mine area deposits; Kelley et al., 2004a), and some replaced calcareous mass flow deposits within the Ikalukrok unit (e.g., Anarraaq deposit, Kelley et al., 2004b). Hydrothermal pyrite in the Main deposit was dated at 338.3 ± 5.8 Ma (Re-Os), which is coeval, within uncertainty, to the biostratigraphic age of the upper Kuna Formation (Morelli et al., 2004). Anarraaq Kelley et al. (2004b) provide an excellent descriptive overview of the Anarraaq area. In brief, the Anarraaq sulfide deposit is a lenticular body located ~650 m below the surface. The highest grades (20–30% Zn) occur in the southeastern part of the deposit, which is inferred to have formed near a Carboniferous extensional fault that acted as the primary conduit for hydrothermal fluids, although no such fault has been identified. The northwestern part of the deposit is characterized by >60% pyrite and marcasite with minor amounts of base metal sulfide. To explain the size and shape of the sulfide body and Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest 1805 ANARRAAQ DEPOSIT, RED DOG DISTRICT, ALASKA Ipewik, Tingmerkpuk Cr os s (F -se ig c t . 6 io n ) Etivluk Group Siksik. Otuk Fm Penn. Triassic to Perm. Jurassic Late Jurassic to Cretaceous Okpikruak Formation Red Dog thrust sheet stratigraphy Mississippian Lisburne Group Kuna Fm ec ls na ) di . 5 itu Fig ng ( Lo Ikalukrok unit: hosts all known occurrences of Red Dog-style mineralization. 50 0 meters major detachment fault mudstone muddy and sandy limestone sandstone chert limestone, dolostone calcareous mudstone and limestone olistoliths of varied composition: chert, limestone / dolostone, conglomerate, etc. n Kivalina unit tio 100 Surface geology Thrust fault Tear fault Shapes projected to surface Barite body High-grade Zn-Pb mineralization Massive sulfide Drill core collars and traces: Mineralized interval re-logged Entire core re-logged Core photos, log available Log available Other cores (not in sections) North 200m Fig. 2. Generalized stratigraphy of the Red Dog thrust sheet, which hosts all known clastic-dominated Zn-Pb mineralization in the Red Dog district. Modified after figure 4 in Blevings et al. (2013). Fig. 3. Plan map of the Anarraaq area. Modified after figure 14 in Blevings et al. (2013). The high-grade Zn-Pb mineralization shape indicates grade thickness ≥140 Zn + Pb% × m (S.K. Blevings, pers. commun., 2019). some sulfide textures, Kelley et al. (2004b) suggested that the mineralizing fluids preferentially replaced calcareous mass flows deposited in a submarine channel. The sulfide body is underlain by a basal fault and may be separated from overlying strata by a thrust fault. The strata overlying the sulfide deposit contains a 9- to 67-m lower interval of mudstone with disseminated sulfides that extend up to 20 m above the sulfide deposit, 10–20 m of mudstone interbedded with calcareous mass flow deposits (i.e., lithic turbidite and calcareous radiolarite), 10–20 m of black mudstone, and 10–20 m of chert and calcareous radiolarite. Above that, the Anarraaq barite body ranges in thickness from 65 to 140 m; some parts may have been structurally thickened. The barite body consists of a lower zone of massive, laminated barite, the basal part of which is variably calcareous, and an upper zone of nodular barite in black chert and mudstone. The barite body is overlain by younger strata with multiple repeating thrust sheets verging northeast. sulfate-bearing bottom waters or shallow pore waters (Ayuso et al., 2004; Johnson et al., 2004, 2009). This type of environment is analogous to cold methane seeps (Suess, 2014). Baritic cold methane seeps generally form on continental margins with abundant biogenic productivity in the overlying water column (Torres et al., 2003). In such settings, biologically mediated barite precipitation in the water column results in relatively high concentrations of biogenic barite in organic-rich, and commonly opal-rich, sediment (Bishop, 1988; Paytan et al., 1993). Degradation of organic matter during burial creates strongly reducing, methane-rich pore waters that dissolve the biogenic barite (Brumsack, 1986; Torres et al., 1996). Where these fluids migrate into sulfate-bearing seawater or shallow pore waters, barite reprecipitates (Brumsack, 1986; Torres et al., 1996, 2003). Because barium is transported in methane-bearing fluids, barite precipitation is closely linked to an important biological process: sulfate driven-anaerobic oxidation of methane (SDAOM), which proceeds by reaction 1 (Barnes and Goldberg, 1976; Reeburgh, 1976): Methane-Driven Diagenetic Processes Barite in the Red Dog district is thought to have formed when diagenetic fluids enriched in methane and barium vented into CH4 + SO42– → HCO3– + HS– + H2O. Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest (1) 1806 REYNOLDS ET AL. This reaction is mediated by a consortium of sulfate-reducing bacteria and syntrophic anaerobic methanotrophic archea (Hoehler et al., 1994; Knittel and Boetius, 2009). Sulfate-driven anaerobic oxidation of methane commonly takes place below the sediment surface in a zone called the sulfate-methane transition (SMT; Fig. 4; Iversen and Jørgensen, 1985). There, upward-diffusing methane and downward-diffusing seawater sulfate are depleted by SD-AOM (Iversen and Jørgensen, 1985). Diagenetic barite precipitates just above the SMT (Dickens, 2001), where the absence of methane and the presence of sulfate result in a decrease in the solubility of barium. SD-AOM is also associated with the precipitation of authigenic carbonate (e.g., Paull et al., 1992) and pyrite (e.g., Lin, Q., et al., 2016; Lin, Z., et al., 2016). However, at Anarraaq, authigenic carbonate associated with the barite bed has not been identified previously (Johnson et al., 2004). At Red Dog, Kelley et al. (2004a) proposed that thermogenic reductive dissolution of early barite provided an important source of S for later hydrothermal sulfide mineralization in the mine area deposits (Qanaiyaq, Main, Aqqaluk, and Paalaaq; Fig. 1). Furthermore, Johnson et al. (2004) recognized the link between barite precipitation and SD-AOM as a possible source of reduced sulfide (HS–) for formation of Zn-Pb mineralization. The barite body at Anarraaq is not overprinted by Zn-Pb mineralization and, therefore, provides insight into how methane-driven diagenetic processes altered the host sediment before the hydrothermal system developed. As such, it is an ideal study site to assess the role that diagenetic processes may play in the development of the hydrothermal system. Methods Relogging and thin-section petrography The petrographic observations reported in this study are drawn from the detailed relogging of 13 drill cores along two sections through the Anarraaq area (Fig. 3), >500 samples of cut core, and >80 thin sections (locations shown on Figs. 5, 6). Cores were painted with a dilute HCl solution containing alizarin red S and potassium ferricyanide to allow rapid identification of carbonate minerals (Hitzman, 1999). In addition to detailed hand sample and thin-section petrographic descriptions, the scanning electron microscope (SEM) at the University of Alberta was used to confirm mineralogy and microscopic textural relationships with dispersive spectros- copy (EDS) and backscattered electron (BSE) imaging. Spot checks using micro-Raman spectroscopy were used to verify visual identification of pyrite and marcasite. Additional lithological and textural details were drawn from original logs and high-resolution photographs of fresh core. Re-Os sampling and analytical methods Three samples from drill core were used for Re-Os analyses (Fig. 5A). Sample 923-2123 is 30 cm of core composed of black mudstone with disseminated pyrite and minor quartzcalcite veinlets (App. Fig. A1A) and located approximately 12 m above the Anarraaq sulfide deposit. Veinlets and macroscopic accumulations of pyrite were avoided during sampling of mudstone. Sample 1716-2520.8 is 20 cm of core that consists of visibly unmineralized mudstone with diagenetic pyrite nodules from the Anarraaq sulfide deposit (App. Fig. A1B). The pyrite nodules are crosscut by minor quartz veinlets, which were excluded during sampling. Sample 1723-2104.5 is 20 cm of core that comprises hydrothermal pyrite enclosed in calcite (App. Fig. A1C). The Re-Os analyses were carried out at the Crustal Re-Os Geochronology Laboratory of the Department of Earth and Atmospheric Sciences, University of Alberta, Canada. Each sample was cut from original core material using a rock saw and then polished to remove any residue. Mudstone samples were then powdered in an agate mill, whereas pyrite samples were manually crushed using ceramics to a size of approximately 1–5 mm and handpicked under a stereoscope to remove any impurities. Several hundred milligrams of sample were then added to a Carius tube with a known amount of 185Re and 190Os spike solution and 8 mL of CrO3-H2SO4 (for mudstones) or 2 mL of 10 N HCl and 6 mL 16 N HNO3 (for pyrite). The Carius tubes were sealed and heated to 220°C for 48 (pyrite) to 72 h (mudstone) to ensure samplespike equilibration. Additional details of the chemical procedures for sample digestion and purification of Re and Os are provided by Selby and Creaser (2003) and Hnatyshin et al. (2016). The abundance and isotopic composition of Re and Os were determined by isotope dilution-negative thermal ionization mass spectrometry (ID-NTIMS) with a ThermoScientific Triton NTIMS under conditions comparable to those of Hnatyshin et al. (2016) to ensure a stable ion beam. Data were reduced according to the method of Kendall et al. (2004), seafloor Barite Pyrite Sulfate-driven anaerobic oxidation of methane (SD-AOM) CH4 + SO42- HCO3- + HS- + H2O Carbonate SMT Seawater SO42- Mineral Precipitation and Dissolution Diagenetic CH4 + Ba2+ Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest Fig. 4. The sulfate methane transition (SMT) forms where downward-diffusing seawater sulfate and upward-diffusing methane are depleted via sulfatedriven anaerobic oxidation of methane. This reaction can result in the precipitation of carbonate and pyrite. Barite is soluble in the methanic pore waters below the SMT but not above. 1807 ANARRAAQ DEPOSIT, RED DOG DISTRICT, ALASKA A NW SE ? ? ? Mudstone 1720 1160 1719 923 1716 1723 50 m 1714 Diagenetic pyrite 930 50 m 806 Hydrothermal pyrite 1712 Undifferentiated Siksikpuk Barite body - black chert with barite nodules Barite body - massive / laminated barite Intrusion Undifferentiated Ikalukrok Ikalukrok mudstone - intensely sheared with local Zn-Pb mineralization Sulfide mineralization - Upper zone Sulfide mineralization - Lower zone Intensely sheared Ikalukrok mudstone with minor clasts of other units Undifferentiated Red Dog and Wolverine Creek thrust sheets Fault (direction unknown) Fault (vergence direction) Fault (lower confidence) Thin section(s) Re-Os isotope sample Intersection with other section Drill core trace In-out of section Calcareous/ limestone interval Pyritic interval Calcareous turbidite interval B 1720 928 1719 806 50 m 923 1714 1716 1723 50 m 1712 C 928 1720 1719 923 1714 1723 1716 1712 50 m 806 50 m Fig. 5. (A) Longitudinal section through the central part of the Anarraaq sulfide deposit, oriented perpendicular to the direction of transport during Brookian shortening. Inset shows (B) spatial distribution of diagenetic components and (C) hydrothermal mineralization styles. See additional legend items in Figure 6. Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest 1808 A REYNOLDS ET AL. SW NE 1139 811 1185 50 m 1732 816 809 1714 1717 1713 50 m 808 1718 Dominant diagenetic components Barite (altered to quartz and sulfides) in black mudstone Poorly preserved host Fe sulfide layers, nodules >20% in black mudstone Fe sulfides layers, nodules <20% in black mudstone Barite identified in single sample Layers of tightly packed framboids identified in single sample Dominant mineralized textures Upper zone Widespread massive sulfide Laminated sphalerite Lower zone Preferential replacement of barite by Zn-Pb mineralization Widespread Zn-Pb mineralized breccia Zn-Pb mineralized breccia interspersed with intervals of mudstone Sphalerite veining B 811 1732 50 m ? ? 1714 809 1717 1713 1185 808 50 m 1718 C 811 1185 50 m 1732 1714 809 1717 1713 808 50 m 1718 Fig. 6. (A) Cross section through the central part of the Anarraaq sulfide deposit, oriented parallel to the inferred direction of transport during Brookian shortening (to the northeast). Inset shows spatial distribution of (B) diagenetic components and (C) hydrothermal mineralization styles. Additional legend items in Figure 5. Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest ANARRAAQ DEPOSIT, RED DOG DISTRICT, ALASKA whereby 2σ uncertainties for 187Re/188Os and 187Os/188Os are determined by numerical error propagation and implementation of an error correlation (ρ) between 187Re/188Os and 187Os/188Os. Blank corrections for mudstone data were 0.3 ± 0.1 pg for Os at a 187Os-188Os ratio of 0.20 ± 0.05 and 15 ± 3 pg for Re. For pyrite, the Os blank correction is 0.047 ± 0.012 pg at a 187Os-188Os ratio of 0.183 ± 0.056, the Re correction being 1.1 ± 0.6 pg. Age calculations of Re-Os isotope data were performed using Isoplot version 3.00 (Ludwig, 2001) and the 187Re decay constant of Smoliar et al. (1996). Results Cross sections Figures 5 and 6 show new cross sections constructed by correlating logs parallel to, and transverse to, the inferred thrust transport direction. Two discrete zones of sulfide mineralization can be distinguished: the Lower zone and the Upper zone (Figs. 5, 6). The mineralized zones are characterized by distinctive styles of Zn-Pb mineralization (see below for more details) and are bounded by intervals of sheared mudstone that contain local sulfide-bearing fragments. The boundaries of these shear zones are marked as faults in Figures 5 and 6, but the magnitude of displacement along them is poorly constrained. The interval of unmineralized Ikalukrok between the barite body and the sulfide deposit ranges from <50 m in the center of the deposit to >100 m in the northeast (Fig. 6A). The calcareous turbidite is broadly continuous across both sections. Whereas the turbidite unit is typically ~20 m thick, it can be as thin as a 2-m interval of calcareous sandstone or ~10 m of mudstone with minor micrite layers <10 cm in the central and southeastern parts of the area. Carbonate in the lower part of the barite body and in the underlying mudstone is discontinuous. The combined maximum thickness of the calcareous beds encountered in this study is 50 m in core 1723. At least one thrust fault cuts upsection through the Anarraaq barite body. Evidence for fault-related folding in the hanging wall includes changes in unit thickness and bedding oriented at a low angle to the core axis. Additional structures may be responsible for other variations in stratigraphic thickness of the massive and nodular barite units. The petrographic features of the barite body and sulfide deposit are described in the following section. Hand specimen descriptions, petrography, and paragenesis The Anarraaq barite body and sulfide deposit contain darkcolored, fine-grained, noncalcareous rocks with variable fissility and hardness. In this study, these rocks are referred to as mudstone. If conchoidal fracture is observed in hand specimen or microcrystallinity in thin section, the term “chert” is used instead, but these distinguishing features are difficult to identify in small domains of mudstone within zones of massive barite/sulfide or where abundant organic material obscures the texture in thin section. More detailed descriptions of the sedimentary rocks of the Ikalukrok can be found in Dumoulin et al. (2004), Kelley et al. (2004a), and Slack et al. (2004). Barite body: The barite body is well described by Kelley et al. (2004b); we note some additional detailed observations. In thin section, samples of massive, laminated barite typically 1809 containing chert with variable organic content and disseminated to patchy barite (Fig. 7B, C). The grain size of the barite is ≤30 µm and the crystal boundaries are highly irregular to embayed, suggesting dissolution and remobilization (Fig. 7F). Calcareous intervals are not strongly laminated but still commonly contain some chert (Fig. 7D, E). Locally the boundaries between calcite-rich and chert-rich domains crosscut remnant sedimentary laminae (Fig. 7E). Calcite also contains distinctive barite laths ~20 µm in length that appear to have been preserved from dissolution (Fig. 7G). Radiolarians are commonly preserved as spheres of chalcedony or coarser crystalline barite (~50 µm) or as spheres of calcite in the calcareous zones (Fig. 7). Massive cryptocrystalline barite clearly crosscuts cherty and calcareous baritic zones (Fig. 7A), suggesting it formed by replacing the latter. Veins of coarsely crystalline barite are ubiquitous in the Anarraaq barite body. In some places, these veins are clearly associated with a local increase in the abundance of disseminated barite in the surrounding host rock (Fig. 7B). Barite veins commonly contain patches of pyrobitumen enclosing euhedral barite crystals (Figs. 7H) and rare calcite. In several drill cores, an interval of bedded to nodular pyrite (≤6 m thick) occurs just below the massive barite (Figs. 5, 6). Pyrite beds consist of abundant framboids and spheroids, which are enclosed by a later generation of massive finecrystalline pyrite (Fig. 8A-D). Abundant siliceous skeletal material is preserved in the massive pyrite, including local radiolarians that are essentially intact (Fig. 8B, C). A minor later generation of cubic pyrite is associated with crosscutting quartz veinlets. Sulfide deposit—Lower zone: Host strata of the Lower zone contain diagenetic barite and pyrite, lesser marcasite, and rare apatite laths that predate Zn-Pb mineralization (Fig. 9). Evidence for barite is present in the form of pseudomorphs after discrete laths (generally ≤200 μm but locally >1 cm) and nodules and layers composed of laths that coalesce into amorphous masses (Fig. 10A, B). Although most barite is completely replaced by quartz (Fig. 10C) or sphalerite (Fig. 10D, E), rare barite has been detected by EDS (Fig. 10E). Diagenetic Fe sulfide occurs disseminated in mudstone and as macroscopic layers, laminae, and nodules (e.g., Fig. 8I). In hand sample, early Fe sulfides are very finely crystalline and relatively dull in luster. Nodules are several millimeters to 10 cm in diameter and range from rounded to subangular, tabular to irregular. Layers are commonly 1 cm thick or less. Nodules and layers locally coalesce such that up to a meter of core may be massive Fe sulfide. Pyrite occurs as framboids (Fig. 8F) or spheroids (Fig. 8G), euhedral crystals that locally overgrow framboids (Fig. 8F), and fine crystalline masses (Fig. 8F-H) that are locally intergrown with slightly coarser marcasite. A distinctive characteristic of the early Fe sulfide is the abundance of siliceous inclusions, many of which are clearly biogenic in origin. The best-preserved microfossils are radiolarians (Fig. 8G, H) and the morphology of less complete skeletal fragments is mostly consistent with this identification. The host strata in the Lower zone are best preserved in the three cores in the northwestern half of the longitudinal section (cores 1723, 1716, and 1712; Fig. 5A) where overprinting by hydrothermal mineralization is the least destructive of preexisting textures. These cores show a consistent stratigraphic Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest 1810 REYNOLDS ET AL. Fig. 7. Massive, laminated barite textures in the Anarraaq barite body. (A) Photograph of core in which massive cryptocrystalline barite crosscuts domains of chert ± calcite with disseminated barite. Black line with arrows indicates orientation of remnant laminae apparent in thin section. (B, C) Plane-polarized light photomicrographs of chert with variable organic content and disseminated barite. In B, barite vein has halo of disseminated barite. (D) Plane-polarized photomicrograph of chert intermixed with calcite; both contain disseminated barite. In B-D, spheres of chalcedony, barite, and calcite form after radiolarians. (E) Plane-polarized photomicrograph of the boundary between chert-rich and calcite-rich baritic domains (dashed line), which is highly irregular and discordant to remnant mudstone laminae. (F) Reflected-light photomicrograph shows disseminated barite in chert has highly irregular to embayed crystal boundaries, suggesting dissolution and remobilization. (G) Backscattered electron image shows disseminated barite laths in calcite, suggesting these crystals were protected from dissolution. (H) Plane-polarized photomicrograph of massive cryptocrystalline barite crosscut by coarsely crystalline barite ± pyrobitumen vein. Samples 809-1913.1 (A, F), 809-1893.2 (B, D, E, H), 809-1852.7 (C). Abbreviations: Brt = barite, Cal = calcite, Cdn = chalcedony, Cht = chert, Dis = disseminated, OM = organic material, Pbt = pyrobitumen, Rad = radiolarian. Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest ANARRAAQ DEPOSIT, RED DOG DISTRICT, ALASKA Fig. 8. (A) Photograph of core shows bedded to nodular pyrite that locally underlies massive barite of the barite body. (B, C) Reflected-light photomicrographs from near A show pyrite framboids and spheres enclosed by a later generation of massive fine-crystalline pyrite. Abundant siliceous radiolarians are preserved as fossil hash (B) and layers of intact skeletons (C). (D) Plane-polarized photomicrograph from near A shows chert interbeds contain abundant radiolarians preserved as chalcedony spheres. (E) Reflected-light photomicrograph shows coalescing colloform texture common in diagenetic pyrite of the Upper zone. Reflected-light photomicrographs of diagenetic pyrite in the Lower zone show a pyrite framboid (F) and sphere (G) overgrown by euhedral (F) and massive (F-H) pyrite. Massive pyrite contains siliceous radiolarian skeletal material. (I) Photograph of core showing laminated pyrite in mudstone crosscut by sphalerite veinlets and late calcite-quartz vein. (J) Plane-polarized photomicrograph of sphalerite from I shows crystals contain color zonation despite pale, uniform color in hand specimen. (K) Reflected-light photomicrograph shows cubic pyrite in quartz vein, which crosscuts early Fe sulfide generations. Samples 809 1972-1976ft (A), 809-1980.3 (B), 809-1971.8 (C), 809-1963.7 (D), 1719-2154.8 (E), 923-2404.8 (F), 1713-2352.5A (G, H), 1723-2197.6 (I, J), 923-2204.6 (K). Core photos oriented with up hole toward top of page. Abbreviations: Cal = calcite, Cdn = chalcedony, Cht = chert, Mdst = mudstone, Mrc = marcasite, Py = pyrite, Qz = quartz, Rad = radiolarian, Sp = sphalerite. Numbers indicate paragenetic phases in Figure 9. Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest 1811 1812 MINERAL Barite 1 (Carbonate) Pyrite 1-2 Marcasite 1 Apatite Marcasite 2 Pyrite 3 Sphalerite Quartz 1 Pyrobitumen 1 Calcite 1 Galena Stibnite Quartz 2 Dolomite 1 Quartz 3 Calcite 2 Pyrite 4 Dolomite 2 Barite 2 Pyrobitumen 2 REYNOLDS ET AL. DIAGENETIC HYDROTHERMAL LATE-STAGE Fig. 9. Paragenesis of the Upper and Lower zones in the Anarraaq sulfide deposit. Carbonate is in parentheses to indicate that its presence is inferred. Grayed minerals observed only in the Lower zone. Bolded minerals are most abundant. distribution of diagenetic components (Fig. 5B) including an upper interval of mudstone with (pseudomorphs after) barite, a middle interval of very poorly preserved host rock (i.e., of unknown composition) with local intervals or clasts of black mudstone and diagenetic Fe sulfide, and a basal interval of mudstone with abundant diagenetic Fe sulfide layers and nodules. Although the host rock is less well preserved in the southeastern, higher-grade part of the deposit, rare evidence of barite occurs at a similar stratigraphic level in cores 1714 and 923 (Fig. 4B). In the southwest-northeast cross section, intervals of silicified barite in black mudstone occur only in cores 1713 and 1717 (Fig. 6B). The hydrothermal mineral assemblage of the Lower zone is dominated by marcasite, pyrite, sphalerite, quartz, and galena (Fig. 9). Hydrothermal pyrite and marcasite do not contain biosiliceous inclusions and are coarser and brighter in luster than diagenetic Fe sulfide (Fig. 11F). Some marcasite crystals have been partially replaced by pyrite (Fig. 11F). The Fe sulfide is commonly intermixed with sphalerite at a fine scale, but it also occurs as macroscopic veins, cavity rims (Fig. 11A), vein rims, and rare stalactites (Fig. 11G). Hydrothermal Fe sulfide predates some sphalerite and vice versa, commonly within the same hand sample. Sphalerite in the Lower zone is highly variable in color (Fig. 11A, B, D), and color zonation within crystals is ubiquitous (Figs. 8J, 10C). Even sphalerite that appears uniform in hand sample shows color zonation in thin section (Fig. 8I, J). Sphalerite occurs disseminated in mudstone (Fig. 11I), as veins (Fig. 11I), as pseudomorphs after barite (Fig. 10C, D), and in complex mineralized breccias (Fig. 11A, B). Sphalerite veins crosscut mudstone and diagenetic Fe sulfide and barite (Figs. 10A, 11I). Veins are locally associated with disseminated sphalerite in mudstone, which may be abundant enough to obscure the host lithology (Fig. 11I; dashed circles). Sphalerite preferentially replaces barite rather than the surrounding mudstone (Fig. 10A). Zoned, euhedral sphalerite crystal outlines suggest that barite was entirely dissolved before sphalerite grew in the resulting cavities (Fig. 10C). The higher-grade parts of the deposit are characterized by mineralized breccias that contain a complex mixture of sulfides, quartz, calcite, mudstone, and diagenetic Fe sulfide (Fig. 11A, B). In some places, mineralized breccias are stratiform between undisturbed layers of mudstone and diagenetic Fe sulfides (Fig. 11H). Breccias contain irregular cavities up to 10 cm across that are lined with hydrothermal Fe sulfide (Fig. 11A), euhedral calcite (Fig. 11J), or sphalerite (Fig. 11D). Cavities contain internal sediment predominantly composed of sphalerite and quartz (Fig. 11A, D). Sphalerite forms snow-on-roof textures (Fig. 11D) and is intergrown with quartz (Fig. 11C). Quartz is euhedral or microcrystalline and contains organic matter including rounded pyrobitumen bodies inferred to have formed from hydrocarbon bubbles trapped at the time of hydrothermal mineralization (Fig. 11C). Internal sediment with minor sphalerite and abundant quartz is very dark gray and can easily be mistaken for Ikalukrok mudstone (Fig. 11J). Multiple cavities over several meters of core contain identical internal stratigraphic sequences, suggesting interconnection at the time of internal sedimentation and confirming lithification occurred before the main stage of hydrothermal mineralization. Galena mainly occurs in the zones of higher-grade Zn mineralization. In some places, galena and sphalerite appear to be coeval at the hand sample scale (Fig. 11D) but thin sections reveal that galena clearly crosscuts sphalerite (Fig. 11E) and hydrothermal Fe sulfides. Locally, galena contains rare stibnite or is intergrown with rhombic dolomite crystals (Fig. 11E) or anhedral, inclusion-rich quartz. In the Lower zone, the hydrothermal mineralization decreases in intensity from southeast to northwest (Figs. 5C, 6C); in the southeast, hydrothermal sulfides completely replace the host rock, whereas in the northwest, significant amounts of host rock are preserved. Common textures in the southeast include multiple generations of sphalerite, abundant ruby-red and brown sphalerite, abundant galena, mineralized breccias, and cavities with internal sediment. In the northwest, the uppermost part of the Lower zone is characterized by hydrothermal veining and preferential replacement of barite over surrounding mudstone, the middle to upper part shows intervals of near-complete host-rock replacement, and the base is characterized by relatively minor sphalerite and Fe sulfide (Fig. 10I). Late-stage quartz and calcite occur in the center of veins (Fig. 10A) and at the stratigraphic top of some cavities rimmed with hydrothermal sulfides and generally crosscut all earlier minerals (Figs. 8I, 11H). Rare local dolomite and barite also occur with late calcite. Pyrobitumen locally cements space between calcite crystals. Cubic pyrite commonly occurs at the edges of calcite and quartz veins that crosscut earlier generations of Fe sulfide (Fig. 8K). Late-stage minerals may represent a final episode of Mississippian hydrothermal mineralization, Brookian (Mesozoic) tectonism, or both. Sulfide deposit—Upper zone: The Upper zone is like the Lower zone in many ways (Fig. 9); only major differences are highlighted here. Quartz ± disseminated sphalerite pseudomorphs after barite laths are relatively rare and occur in the Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest ANARRAAQ DEPOSIT, RED DOG DISTRICT, ALASKA 1813 Fig. 10. Lower zone barite of the Anarraaq sulfide deposit. (A) Photograph of drill core with up hole toward right of page. Sphalerite vein crosscuts interbedded mudstone and silicified barite. Sphalerite preferentially replaces barite rather than the surrounding mudstone. Plane-polarized light photomicrographs show (B) silicified barite textures range from distinctive laths to amorphous masses. (C) Sphalerite pseudomorph after barite lath is composed of zoned crystals that indicate nucleation at the edge of the former barite crystal and inward growth into open space, suggesting barite was completely dissolved prior to sphalerite precipitation. (D) Distinctive lath textures are absent in some sphalerite after barite. (E) Backscattered electron image of barite and sphalerite pseudomorphs after barite laths. Samples 1717-2317 (A-D) and 923-2251.2 (E). Abbreviations: Brt = barite, Cal = calcite, Mdst = mudstone, Mrc = marcasite, Py = pyrite, Sp = sphalerite. upper half of the zone (Figs. 5B, 6B). Diagenetic Fe sulfide nodules are generally smaller in diameter (<1 cm; Fig. 12A) than those in the Lower zone. On a microscopic scale, diagenetic Fe sulfide overgrowths may consist of coalescing colloform features (Fig. 8E) but also contain abundant siliceous microfossil fragments, as in the Lower zone. Most of the Upper zone is characterized by abundant sphalerite with local preservation of black mudstone and diagenetic pyrite. The basal part of the zone is dominantly mudstone with abundant Fe sulfide layers and nodules. Rarely, layers of tightly packed framboidal, spheroidal, or euhedral pyrite occur at both the top and bottom of the main zone of Zn-Pb mineralization (Figs. 5B, 6B). Sphalerite in the Upper zone is distinctive in that it displays only subtle color variations in hand sample (cream, pale tan, pale pink; Fig. 12C) that are controlled by the relative concentrations of sphalerite, quartz, and mudstone (Fig. 12D-F). No color zonation is displayed within sphalerite crystals (Fig. 12F). In most places, very fine sphalerite is disseminated in microcrystalline quartz, but rare euhedral quartz crystals contain sphalerite inclusions (Fig. 12E, G) and are enclosed by massive sphalerite (Fig. 12E). Although these textures are like the internal sediment of the Lower zone, no welldefined cavities are observed in the Upper zone. Sphalerite also occurs as laminae in mudstone (Fig. 12A), typically in the basal part of the zone (Fig. 5C). Hydrothermal pyrite and marcasite overlap sphalerite precipitation and have a range of textures similar to that in the Lower zone but are generally finer grained; no stalactites have been observed. Galena Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest 1814 REYNOLDS ET AL. postdates sphalerite but also forms subhedral crystals up to several millimeters across in mudstone (Fig. 12H). Overall, veins of hydrothermal minerals are much less common relative to the Lower zone. The Upper zone is smaller and less laterally extensive and there is no lateral trend in intensity of base metal mineralization, although the thickest interval of high-grade Zn-Pb mineralization is in the southwesternmost drill core 809 (Fig. 6C). Rhenium-osmium geochronology Results from Re-Os analyses of mudstone and pyrite samples are reported in Table 1. The mudstone contains the highest concentrations of Re and Os (19.4–40.5 ppb and 888–1,709 ppt, respectively), and diagenetic pyrite has the lowest concentrations (0.47–1.30 ppb and 16.57–29.35 ppt). Hydrothermal pyrite contains intermediate values of Re (2.86–4.80 ppb) and Os (22.30–28.96 ppt). The hydrothermal pyrite is the most radiogenic with 187Os/188Os of 6.79–12.00; mudstone and diagenetic pyrite samples are characterized by 187Os/188Os of less than 2. Regression of seven mudstone analyses yields an age of 339.1 ± 8.3 Ma (mean square of weighted deviates [MSWD] = 19) with an initial Os ratio of 0.375 ± 0.019 (Fig. 13A). Analyses of six diagenetic pyrites yield an isochron with an age of 333.0 ± 7.4 Ma (MSWD = 1.08) and an initial Os ratio of 0.432 ± 0.025 (Fig. 13A). Results from three hydrothermal pyrite analyses are reported, and although these are not sufficient to build a meaningful isochron, the data points lie very close to the diagenetic pyrite isochron. Regression of all pyrite analyses together yields an age of 334.4 ± 5.3 Ma (MSWD = 3.7) and initial Os ratio of 0.422 ± 0.081 (Fig. 13B), similar, within the resolution of the analytical results, to the diagenetic-only regression. Furthermore, all pyrite analyses from this study lie very close to the isochron reported by Morelli et al. (2004) for hydrothermal pyrite from the Main deposit. When the analyses from both studies are regressed together, they yield an age of 336.0 ± 3.9 Ma (MSWD = 29) with an initial Os ratio of 0.34 ± 0.11 (Fig. 13C), which is within the uncertainty of the results of Morelli et al. (2004; 338.3 ± 5.8 Ma, initial Os of 0.20 ± 0.21) and of this study. Discussion Diagenetic and hydrothermal processes Anarraaq barite body: The Anarraaq barite body formed when diagenetic fluid bearing methane and barium vented into sulfate-bearing pore water or marine bottom waters (Johnson et al., 2004, 2009). In agreement with the previous studies (Johnson et al., 2004; Kelley et al., 2004b), there is abundant textural evidence that barite precipitated below the sediment surface, including sedimentary fabrics that are crosscut by barite (Fig. 7A) and veins that are interpreted to be feeders for disseminated barite (Fig. 7B). The location of the SMT and associated diagenetic barite precipitation relative to the seafloor is primarily controlled by the upward flux of methane, the seawater sulfate concentration, and the rate of sedimentation (Jørgensen et al., 2004; Arning et al., 2015). If the SMT advances upward toward preexisting diagenetic baritic strata, barite may undergo dissolution and reprecipitation (Snyder et al., 2007). Kelley at al. (2004b) interpreted the calcareous components of the Anarraaq barite body to be mass flow deposits overprinted by methane-transported diagenetic barite; however, here for the first time we present petrographic data that indicate that at least some of calcite in the barite body is authigenic: it crosscuts sedimentary laminae (Fig. 7E) and postdates some barite generations (Fig. 7G). In the massive, laminated part of the barite body, well-formed barite crystals only occur enclosed in calcite, suggesting that early formation of authigenic calcite preserved some barite from dissolution. The morphology of these crystals is similar to that of samples collected from modern methane seeps in the northern Gulf of Mexico (Feng and Roberts, 2011). The abundance and clustered nature of framboids in the nodular to layered pyrite that locally underlies the Anarraaq barite body (Fig. 8C) are characteristic of framboids formed at modern SMTs (Lin, Q., et al., 2016). The pyrite overgrowths are certainly secondary and are similar to overgrowths reported in modern SMT sediment (Lin, Q., et al., 2016; Lin, Z., et al., 2016) and the preservation of abundant siliceous radiolarian skeletons suggests pyrite formed relatively early during burial. Fig. 11. Mineralized textures from the Lower zone of the Anarraaq sulfide deposit. (A) Photograph of core containing mineralized breccia and cavity (dashed black line) rimmed in Fe sulfide and filled with internal sediment. Dashed white line marks bedding surface between layers of cryptocrystalline and euhedral quartz internal sediment. (B) Plane-polarized light photomicrograph of mineralized breccia in A, which contains clasts of mudstone and sphalerite enclosed in a complex mixture of quartz, sphalerite, and organic material. (C) Plane-polarized light photomicrograph of internal sediment in A. Dashed line marks bedding surface between cryptocrystalline and euhedral quartz, both of which are intergrown with sphalerite. Note subrounded pyrobitumen occurrence, which is interpreted to have been a hydrocarbon bubble trapped at the time of hydrothermal mineralization. (D) Photograph of core shows cavity rimmed in sphalerite, galena, and local Fe sulfide and filled with internal sediment primarily composed of quartz and sphalerite. Note snow-on-roof sphalerite textures. (E) Reflected-light photomicrograph shows galena crosscuts sphalerite leaving highly irregularly shaped sphalerite islands in galena. Euhedral dolomite is intergrown with galena. (F) Reflected-light photomicrograph shows hydrothermal euhedral marcasite crystal partially replaced by pyrite (dashed line marks boundary). Note the overall complexity of pyrite and marcasite textures. (G) Photograph of core shows Fe sulfide stalactites enclosed in calcite. (H) Photograph of core shows undisturbed layers of mudstone and diagenetic Fe sulfides overlying stratiform mineralized breccia. (I) Photograph of core shows Ikalukrok mudstone, which contains diagenetic Fe sulfide, sphalerite-galena veins, and disseminated sphalerite. Disseminated sphalerite locally obscures host lithology (dashed-line circle). (J) Photograph of core shows euhedral calcite crystals enclosed by internal sediment with subtle laminated texture (parallel to dashed line). In this sample, internal sediment is very dark gray because it is composed of quartz >> sphalerite and could easily be mistaken for Ikalukrok mudstone. Samples 1713-2495 (A-C), 1714-2471.2 (D), 923-2385.3 (E), 923-2376.6 (F), 1716-2303.5 (G), 1716-2347 (H), 1713-2536 (I), 1716-2344.4 (J). All core photos are oriented with up hole toward top of page. Abbreviations: Cal = calcite, Dol = dolomite, Gn = galena, IS = internal sediment, Mdst = mudstone, Mrc = marcasite, OM = organic material, Pbt = pyrobitumen, Py = pyrite, Qz = quartz, Sp = sphalerite. Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest ANARRAAQ DEPOSIT, RED DOG DISTRICT, ALASKA The bisulfide (HS–) produced by SD-AOM at the SMT produces pyrite when it reacts with dissolved iron. Some dissolved iron originates within SMT sediment, but excess HS– can also accumulate and diffuse beyond the SMT. Stratiform pyrite has 1815 been observed to form below the SMT where excess HS– encounters upward-diffusing Fe2+ sourced from deeper sediment (Jørgensen et al., 2004). Thus, we interpret the pyrite (Py1, Py2) to form from methane-driven diagenetic processes. Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest 1816 REYNOLDS ET AL. Fig. 12. Mineralized textures from the Upper zone of the Anarraaq sulfide deposit. (A) Photograph of core showing laminated sphalerite in Ikalukrok mudstone with diagenetic Fe sulfide and silicified barite nodules. Note sharp transition to stratabound massive sulfide at bottom of sample. (B) Plane-polarized light photomicrograph of silicified barite in A shows textures ranging from laths to amorphous mass. Elsewhere in the sample, pseudomorphs after barite also contain disseminated sphalerite and Fe sulfide. (C) Photograph of core shows subtle variations in sphalerite color. Plane-polarized light photomicrographs from the same sample show (D) pink domains contain a mixture of mudstone, quartz, and sphalerite. The boundaries between mudstone and quartz-sphalerite are irregular and somewhat gradational. (E) Tan domains contain a mix of quartz and sphalerite. This sample displays rare euhedral quartz crystals with sphalerite inclusions, enclosed by massive sphalerite. (F) Creamcolored domains contain massive sphalerite with no color zonation. (G) Backscattered electron image shows disseminated sphalerite along a growth zone in euhedral quartz. (H) Photograph of core showing galena crystals in mudstone. Samples 1719-2154 (A, B), 809-2332.7 (C-F), 923-2202.3 (G), 809-2396.7 (H). Core photos oriented with up hole toward top of page. Abbreviations: Gn = galena, Mdst = mudstone, Qz = quartz, Sp = sphalerite. The deposit-scale stratigraphic relationship between barite, authigenic carbonate, and diagenetic pyrite in the Anarraaq barite body is consistent with what is expected in SMT zones (e.g., Lash, 2015; Fig. 4), and diagenetic calcite after barite is reported in some locations (e.g., Dean and Schreiber, 1978). However, the thickness of the Anarraaq barite body (~50–100 m) is much greater than intervals of barite enrichment reported in modern SMT zones (from <1 cm to 3 m; e.g., Brumsack, 1986; Torres et al., 1996; Snyder et al., 2007) or even the thickest reported modern methane seep barite deposits (~10 m at Sea of Okhotsk and at the California Borderlands; Greinert et al., 2002; Gwiazda et al., 2019). The Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest 1817 ANARRAAQ DEPOSIT, RED DOG DISTRICT, ALASKA Table 1. Re-Os Isotope Results for Anarraaq Sample name 923-2123-A1 923-2123-A2 923-2123-B-2 923-2123-D-2 923-2123-G-2 923-2123-M 923-2123-N 1716-2520.8 A 1716-2520.8 Y 1716-2520.8 C+3 1716-2520.8 C+4 1716-2520.8 D+2 1716-2520.8 X 1723-2104.5 A-3 1723-2104.5 D+2 1723-2104.5 D+3 Sample type Shale Shale Shale Shale Shale Shale Shale Pyrite-diagenetic (Py1-2) Pyrite-diagenetic (Py1-2) Pyrite-diagenetic (Py1-2) Pyrite-diagenetic (Py1-2) Pyrite-diagenetic (Py1-2) Pyrite-diagenetic (Py1-2) Pyrite-hydrothermal (Py3) Pyrite-hydrothermal (Py3) Pyrite-hydrothermal (Py3) Re (ppb) 23.1 24.4 21.8 24.1 19.4 34.9 40.5 1.30 0.47 0.80 0.61 0.85 0.52 4.80 4.36 2.86 Os (ppt) 1,253 1,126 888 1,218 1,025 1,709 1,021 28.98 19.21 23.51 29.35 28.89 16.57 28.96 26.63 22.30 Anarraaq barite body also lacks the macrofauna and towers typically associated with seep deposits or any truly laminated textures that indicate deposition on the seafloor (Johnson et al., 2004; this study). The rock record contains other thick barite beds that appear to be diagenetic. For example, the 1.4 1.2 187Os/188Os 187Os/188Os Anarraaq Pre-ore Pyrite (blue) 1.6 Anarraaq Black Shale (black) Model33Solution Solution ((±95%-conf.) on77points points Model 95%-conf.) on Age = 339.1 Ma ± 8.3 Ma Age8.3 = 339.1 187 188 InitialInitial 187/188=0.375 0.019 Os/ Os = 0.375 ± 0.019 MSWDMSWD = 19, Probability = 0.000 = 19, Probability = 0.000 Initial 187/188 variation =0.012 (2σ) 187 188 1.0 Initial 60 100 140 Os/ 180 220 260 10 Rho 0.4 0.4 0.4 0.3 0.4 0.3 0.4 0.54 0.52 0.44 0.54 0.60 0.49 0.58 0.82 0.58 6 Model33Solution Solution (±95%-conf.) on 9 points points Model ( 95%-conf.) on Age = 334.4 Ma ± 5.3 Ma Age5.3 = 334.4 187 InitialInitial 187/188=0.422 0.081 Os/188Os = 0.422 ± 0.081 MSWD = 3.7,=Probability = 0.000 MSWD 3.7, Probability σ = 0.000 Initial 187/188 variation =0.15 (2σ) 187 188 A 0 0 187Re/188Os Initial 400 800 1200 Os/ Os variation = 0.15 (2σ) 1600 187Re/188Os 2000 2400 C 30 187Os/188Os ±2σ 0.002 0.003 0.003 0.002 0.002 0.002 0.004 0.01 0.03 0.01 0.03 0.04 0.01 0.16 0.08 0.08 8 2 300 0.925 1.046 1.135 0.970 0.944 1.001 1.675 1.91 1.20 1.51 1.06 1.34 1.40 12.00 10.94 6.79 Anarraaq Ore Stage Pyrite (red) with Pre-ore Pyrite 4 Os variation = 0.12 (2σ) 187Os/188Os B 12 MSWD = 1.08, Probability = 0.37 0.8 97.98 116.90 133.99 105.93 100.69 109.48 229.63 265.42 135.51 192.63 112.60 165.15 175.60 2,034.22 1,901.08 1,153.74 14 Model ( 95%-conf.) on 6 points points Model 11 Solution Solution (±95%-conf.) Age = 333.0 Ma ± 7.4 Ma Age7.4 = 333.0 Initial 187/188=0.432 187 Initial Os/188Os =0.025 0.432 ± 0.025 MSWD = 1.08, Probability = 0.37 1.8 ±2σ 0.3 0.4 0.5 0.4 0.4 0.4 0.8 1.52 1.99 1.31 1.91 3.11 1.48 17.84 12.76 9.03 Late Devonian barite occurrences found in the Selwyn basin, Canada, are 100–150 m thick and have isotopic (S, Sr) compositions that suggest they formed in a diagenetic environment (Fernandes et al., 2017). At times of steady-state diagenesis, the formation of barite units requires either high methane A 2.0 187Re/188Os Main Deposit Ore Stage Pyrite (black) with all Anarraaq Pyrite 20 B Model 95%-conf.) on Model33Solution Solution ((±95%-conf.) on1919points points Age = 336.0Age 3.9 = Ma 336.0 ± 3.9 Ma Initial 187/188=0.34 Initial 187Os/1880.11 Os = 0.34 ± 0.11 MSWD = 29, Probability = 0.000 MSWD = 29, Probability = 0.000 Initial 187/188 variation =0.31 (2σ) 10 Initial 187Os/188Os variation = 0.31 (2σ) 0 0 2000 4000 187Re/188Os 6000 Fig. 13. Re-Os isochron diagrams for (A) diagenetic pyrite in the Anarraaq sulfide deposit and overlying mudstone, (B) diagenetic and hydrothermal pyrite in the Anarraaq sulfide deposit, and (C) hydrothermal pyrite from the Anarraaq and Main deposits. Data from the Main deposit is from Morelli et al. (2004). MSWD = mean square of weighted deviates. Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest 1818 REYNOLDS ET AL. fluxes (and/or sulphate fluxes) or lower fluxes over long time periods (see Johnson et al., 2004, who calculate 50,000 y to form 50 m of barite). However, upwelling environments on continental margins are areas of nonsteady-state diagenetic processes, in which key variables (sedimentation rate, organic matter deposition, methane flux) vary on timescales from days to thousands of years (Kasten et al., 2003). Changes in the sedimentation rates can change the depth of the SMT and allow for recycling and reprecipitation of authigenic minerals; these processes are not well quantified in modern systems except in a few examples (Wehrmann et al., 2013) but nonsteady-state diagenetic processes may allow for thicker beds of authigenic minerals to be formed over longer time periods. In addition, to maximize these processes, the host sediment must be highly permeable. Both the Selwyn basin and Kuna basin barite units are hosted in radiolarian-rich, biosiliceous sediments; radiolarite can maintain porosities of up to 70% at 500 m (Isaacs, 1982). This may have allowed for the accumulation of abundant diagenetic components even in the subsurface (Magnall et al., 2015, 2020a). Anarraaq sulfide deposit: The diagenetic components of the host rock in the Anarraaq sulfide deposit have striking compositional, textural, and stratigraphic similarities to the Anarraaq barite body. Barite and layered to nodular pyrite occur in both the Upper zone and Lower zone, and diagenetic Fe sulfide textures include framboids and overgrowths with well-preserved radiolarian skeletal material. We suggest, therefore, that the sulfide zones overprinted preexisting SMTs in the subsurface, similar to the one preserved in the barite body. This contrasts previous interpretations that layered to nodular pyrite associated with Zn-Pb mineralization in the Red Dog district is hydrothermal (Kelley et al., 2004b; Blevings et al., 2013). The poorly preserved intervals of host rock indicate that a large amount of material underwent extensive dissolution during hydrothermal mineralization. Almost all the diagenetic barite in the Anarraaq sulfide deposit was replaced by quartz and sulfide minerals (Figs. 10, 12A, B). In the Lower zone, the colloform sphalerite pseudomorphs after barite indicate open-space growth (e.g., Fig. 10C), and we interpret the barite crystals to have been completely dissolved prior to sphalerite precipitation. For this to be possible, the mudstone surrounding the barite crystal must have been well lithified to prevent collapse into the open space of the dissolved crystal. The presence of large, mineralized cavities (Fig. 11A, D) and stratabound mineralized breccias (Fig. 11H) also suggests a fully cemented host rock. It seems that barite dissolution alone is unlikely to explain the large cavities found in the Lower zone. Of the original constituents of the host rock, the dissolution of siliceous mudstone or pyrite to form cavities up to 10 cm across seems relatively unlikely, although we would expect the solubility of a biosiliceous (opaline) mudstone to be somewhat higher than a clastic (quartz- and clay-rich) mudstone. Abundant phosphorite beds have been reported in other parts of the Kuna Formation (Dumoulin et al., 2014), but only minor local apatite nodules have been observed in the Ikalukrok unit in the Red Dog district (Slack et al., 2004). Therefore, we agree with Kelley et al. (2004b) that carbonate is the most likely precursor to the cavities. Whereas Kelley et al. (2004b) concluded that this carbonate was sedimentary in origin, we suggest that dissolution of authigenic carbonate that formed in association with diagenetic barite may have also contributed to the porosity and permeability exploited by the hydrothermal system. The Upper zone contains euhedral quartz enclosed in sphalerite (Fig. 12E) that is texturally similar to internal sediment filling cavities of the Lower zone (Fig. 11C) but lacks clearly defined cavity boundaries (Fig. 12C). In combination with the presence of irregular and gradational boundaries between mudstone and quartz-sphalerite domains (Fig. 12C, D) and the relative rarity of veining and angular mudstone clasts in the Upper zone, this suggests that Upper zone sediment may have been less completely cemented than the Lower zone at the time of Zn-Pb mineralization. Original stratigraphic relationship of the Anarraaq barite body and sulfide deposit The displacements on the thrust faults at Anarraaq are poorly constrained. If the displacements were small in magnitude, the present-day configuration of the Lower zone, Upper zone, and barite body could reflect stratigraphic relationships close to the original. If the displacements were more significant, the two zones of Zn-Pb mineralization and the barite body may have formed lateral to each other and were subsequently structurally superimposed. Low-angle faults appear to cut both up and down section in the inferred direction of transport (Fig. 6), indicating that Brookian deformation was more complex than the simple stacking of thrust imbricates in a duplex system, as suggested by De Vera et al. (2004). Several explanations are possible. First, the lenticular nature of the Upper and Lower zones may reflect large variations in strata thickness due to differential compaction, causing thrust faults to locally cut down section. Second, the direction of tectonic transport may not have been parallel to the cross section shown in Figure 6, and the ramps shown in the section may therefore be lateral or oblique ramps, rather than frontal ramps as implied by the vergence shown by Blevings et al. (2013). Third, thrusts may have been emplaced out of sequence during Brookian deformation, possibly because of the laterally heterogeneous mechanical properties of the massive sulfide bodies. More work is required to resolve this issue. Baritic strata of the Lower zone correlate well among cores 1723, 1716, 1712, 1717, and 1713, with a trend of gradual thickening toward the most intensely mineralized part of the deposit (Figs. 5B, 6B). In contrast to this trend, cores 1714 and 808 show a thinning or absence of barite and a dramatic thinning in the overall thickness of the Lower zone (Fig. 6B). This may reflect the original geometry of the diagenetic barite or later structural modification. Although the main thickness of the Upper zone is in the northwestern part of the sulfide deposit, intervals of Upper zone-style mineralization farther northeast are characterized by pervasive shearing that clearly postdates mineralization. This geometry could be partly inherited from a heterogeneous original distribution of mineralized units; however, the evidence for deformation strongly suggests that fragments of Upper zone material were transported along a fault. Re-Os constraints on the age and Os sources for Anarraaq The Re-Os isochron ages for the black mudstone just above the Anarraaq sulfide deposit (339.1 ± 8.3 Ma) and the diage- Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest 1819 ANARRAAQ DEPOSIT, RED DOG DISTRICT, ALASKA netic pyrite from within the Anarraaq sulfide deposit (333.0 ± 7.4 Ma) are within error of the Re-Os isochron age of hydrothermal pyrite in the Main deposit in the mine area (338.3 ± 5.8 Ma; Morelli et al., 2004). Although the three data points for hydrothermal pyrite at Anarraaq are not sufficient to create a meaningful isochron, the fact that they plot very close to the diagenetic pyrite isochron suggests that the Re-Os systematics of hydrothermal pyrite are not grossly different from those of diagenetic pyrite. Overall, the data are consistent with contemporaneous Zn-Pb mineralization at Anarraaq and the Main deposit. Although the contact between the Anarraaq sulfide deposit and the overlying nonmineralized sediment is a thrust fault, the overlapping Re-Os isotope ages are compatible with the idea that sediment deposition and hydrothermal mineralization were relatively close in age. As in the Main deposit in the mine area, the textural evidence at Anarraaq indicates that Zn-Pb mineralization was replacive, but the absolute timescale of replacement cannot be resolved with the current ReOs isochron ages. Initial Os composition The global marine Os record is not well constrained during the Carboniferous, but there are several published Re-Os isochrons for Late Devonian mudstones in western Canada and one for a Lower-Middle Pennsylvanian mudstone in the eastern United States (Fig. 14). These data indicate that marine 187Os/188Os was between 0.37 and 0.64 in the Late De- 1.2 vonian (~360 Ma; Creaser et al., 2002; Selby and Creaser, 2003, 2005) and no higher than 0.72 in the Lower to Middle Pennsylvanian (323 ± 7.8 Ma; Geboy et al., 2015). The initial 187Os/188Os value for the Anarraaq mudstone isochron (0.375 ± 0.019) is broadly consistent with these data and may be representative of global marine Os composition. Slack et al. (2015) proposed that Middle Mississippian (~335 Ma) marine 187Os/188Os was much higher (~1.08) and that Ikalukrok mudstones with lower initial 187Os/188Os record the local/intermittent influence of a metalliferous hydrothermal fluid venting into the seawater. This interpretation is based on single sample analyses and calculated initial 187Os/188Os values rather than full isochrons, which makes it difficult to assess the effects of possible postdepositional disturbance of the Re-Os system. To increase the global marine 187Os/188Os composition from 0.64 in the Late Devonian to 1.08 in the Middle Mississippian would require a large increase in radiogenic Os input via continental weathering—that is, an increase on par with what has been observed in more recent geologic history during the rise of the Himalayan mountains (Turekian and Pegram, 1997; Peucker-Ehrenbrink and Ravizza, 2000). The marine 87Sr/86Sr ratio is a proxy for global rates of continental weathering and is relatively well constrained over Earth history (Burke et al., 1982; Veizer et al., 1999). Mississippian marine 87Sr/86Sr was equal to, or lower than, Late Devonian values (Denison et al., 1994). It therefore seems unlikely that Mississippian global marine 187Os/188Os would have been significantly higher than Late Devonian values, although we can- Proposed Middle Mississippian seawater (~335 Ma) Slack et al. (2015) Modern Seawater Initial 187Os/188Os 1.0 0.8 Lower-Middle Pennsylvanian mudstone (323 ± 7.8 Ma) Eastern United States 0.6 Late Devonian mudstone (~360 Ma) Western Canada A 0.4 0.2 B C C Anarraaq Diagenetic Pyrite Anarraaq Mudstone (339.1 ± 8.3) Mantle Anarraaq Diagenetic & Hydrothermal Pyrite Red Dog Main Deposit hydrothermal pyrite 0 Fig. 14. Initial Os isotope composition (187Os/188Os) of diagenetic and hydrothermal pyrite from the Anarraaq sulfide deposit and overlying Ikalukrok mudstone. Also plotted are data from hydrothermal pyrite in the Red Dog Main deposit (Morelli et al., 2004), several Late Devonian mudstones (A: Selby and Creaser, 2005; B: Creaser et al., 2002; C: Selby and Creaser, 2003), and a Lower-Middle Pennsylvanian mudstone (Geboy et al., 2015). The Os compositions of modern seawater and the mantle reservoirs are from Peuckler-Ehrenbrink and Ravizza (2000). Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest 1820 REYNOLDS ET AL. not rule out local effects, especially if Ikalukrok mudstone was deposited in a restricted basin. We suggest that the variability in calculated initial 187Os/188Os values reported by Slack et al. (2015) was caused by postdepositional disturbance of the Re-Os system or the local contribution of a more radiogenic clastic component. The initial 187Os/188Os value of the Anarraaq diagenetic pyrite isochron (0.432 ± 0.025) is only slightly higher than that of the mudstone (0.375 ± 0.019). We interpret diagenetic pyrite to have formed at an SMT during early diagenesis upon mixing of pore waters originally derived from Mississippian seawater. The similar initial 187Os/188Os values of diagenetic pyrite and mudstone are consistent with this interpretation. The initial 187Os/188Os values reported in this study are within error of the value reported by Morelli et al. (2004) for hydrothermal pyrite in the Main deposit (0.2 ± 0.21; Fig. 14). However, our results are more precise and overlap only the highest part of the 2σ confidence range of Morelli et al. (2004). Because of the large uncertainty on the initial 187Os/188Os value, Morelli et al. (2004) considered both mantle and relatively young (<410 Ma) crustal reservoirs as possible sources of Os in the hydrothermal fluid. In view of the similarity between the initial 187Os/188Os of the mudstone isochron and the hydrothermal pyrite (regressed with diagenetic pyrite) at Anarraaq, influence from a mantle Os reservoir on the hydrothermal fluid composition is not required. In fact, the data from this study are compatible with the model of Leach et al. (2004), whereby hydrothermal fluids formed from a basinal brine evolved from evaporated seawater of Mississippian or Devonian age that circulated through the Devonian Endicott Group sedimentary rocks. Implications for the genesis of, and exploration for, CD deposits Evidence for early barite overprinted by zinc mineralization has been reported in the Main, Aqqaluk, Paalaaq, and Qanaiyaq deposits in the mine area (Kelley et al., 2004a; Blevings et al., 2013). As the Anarraaq sulfide deposit illustrates, barite can be poorly preserved and very difficult to identify: the deposit was discovered in 1999 (Kelley et al., 2004b), but only in this study do we show that barite was a significant component of the host rock, which has been almost totally replaced in the sulfide deposit. So, although several additional base metal deposits and prospects in the Red Dog district (e.g., Aktigiruk, Su Lik) reportedly contain no barite (Blevings et al., 2013), it is possible that barite was present initially but was then later replaced by hydrothermal minerals. Furthermore, we have suggested that at Anarraaq diagenetic pyrite and carbonate are also products of SD-AOM and are genetically linked to barite formation. Therefore, the presence of authigenic carbonate or laminated pyrite even in the absence of barite may reflect methane-driven diagenetic alteration, rather than hydrothermal processes. Johnson et al. (2004) speculated that SD-AOM may have been the key chemical pathway for producing bisulfide for massive sulfide deposits at Red Dog, especially in deposits like Anarraaq where little to no diagenetic barite had been identified. The recognition that SD-AOM was an important diagenetic process at Anarraaq (Johnson et al., 2004; this study) is consistent with recent findings in other CD districts in the Ca- nadian Cordillera. Magnall et al. (2016, 2020a, b) suggest that the CD Zn-Pb mineralization in the Tom and Jason deposits of Macmillan Pass (Yukon) overprints diagenetic barite and pyrite that formed at the SMT. Barite dissolution released sulfate and allows for sulfate cycling (Magnall et al., 2020b). Johnson et al. (2018) found that diagenetic pyrite and carbonate in the host strata of CD Zn-Pb mineralization at Howards Pass (Yukon-Northwest Territories) were also influenced by SD-AOM in the subsurface. All of these CD deposits, including those at Red Dog, are hosted in organic-rich sediment deposited on continental margins where upwelling of nutrient-rich deep ocean water resulted in abundant biological productivity in the water column (Dumoulin et al., 2014; Magnall et al., 2015, 2018, 2020b; Slack et al., 2017). At Howards Pass, Johnson et al. (2018) linked an increase in sedimentary carbon content to an increase in production of biogenic methane and influence of SD-AOM on diagenesis. It should be noted, however, that although SD-AOM produces sulfide, other reduction mechanisms such as bacterial and thermochemical sulfate reduction remain important processes and will also contribute reduced S to these large hydrothermal systems. Magnall et al. (2020b) showed that the Ba released from barite dissolution at MacMillan Pass forms a halo of Ba enrichment in the host sediment around the sulfide bodies. In methane seeps, barium from dissolved barite can reprecipitate in younger strata (Brumsack, 1986; Torres et al., 1996, 2003). Barite is common in the younger Siksikpuk and Otuk Formations in the Red Dog plate (Young, 2004). Distinguishing relative Ba enrichment in the Ikalukrok or overlying units due to barite dissolution in zones of Zn-Pb mineralization could provide a useful vector for exploration. However, since the Anarraaq mineralized zones are fault bounded and the younger strata contain multiple repeating thrust sheets, it is not possible to test this hypothesis without a detailed structural reconstruction. At Red Dog, the spatial coincidence between barite and overprinting Zn-Pb mineralization suggests a link between the mechanisms that control the location of methane-driven and hydrothermal alteration. One possibility is that early diagenetic and later hydrothermal fluids used similar pathways to flow into the shallow subsurface (Fig. 15). Diagenesis of biosiliceous mudstones has been linked with the development of nontectonic, intraformational faulting and pore fluid expulsion (Cartwright, 1994; Volpi et al., 2003; Davies et al., 2008), and this could have been the mechanism that controlled the location of large-scale SMT diagenesis. In the Red Dog district, hydrothermal fluids are thought to have flowed up extensional faults (Moore et al., 1986); these faults could have exploited preexisting faults or zones of weakness that served as flow pathways for diagenetic fluids bearing methane and barium. This interpretation would be consistent with the occurrence of some barite bodies that were not overprinted by hydrothermal fluids, as occur in the Red Dog district. This is important, as diagenetic barite creates gravity anomalies. Although these anomalies indicate the presence of prospective host rock, they are not directly linked to the existence of Zn-Pb mineralization. According to the subclassification system of Cooke et al. (2000), stratiform CD Zn-Pb deposits are formed by either oxidized or reduced hydrothermal fluids. In this scheme, bar- Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest 1821 ANARRAAQ DEPOSIT, RED DOG DISTRICT, ALASKA A) Host sediment diagenesis B) Hydrothermal mineralization seawater seawater seafloor SMT SO42- Ikalukrok unit CH4 + Ba2+ Ikalukrok unit Fault Diagenetic barite Authigenic carbonate Diagenetic pyrite metalliferous hydrothermal fluid Fault Preferential replacement of barite by Zn-Pb mineralization Widespread Zn-Pb mineralized breccia Zn-Pb mineralized breccia interspersed with intervals of mudstone Sphalerite veining Fig. 15. Deposit model for the Anarraaq sulfide deposit. (A) Methanic and barium-rich diagenetic fluid interacts with seawater sulfate in pore waters to form diagenetic barite, carbonate, and pyrite. At least some dissolution of barite may take place as part of methane-driven diagenetic alteration. (B) Hydrothermal fluid dissolves authigenic and sedimentary carbonate, and possibly more barite, further increasing the porosity and permeability. ite solubility is used as a boundary condition in the modeling of the chemistry of the hydrothermal fluid. This leads to the interpretation that these deposits formed from reduced fluids. However, as barite predates hydrothermal mineralization at Anarraaq (this study) and the Main deposits (Kelley et al., 2004a), its occurrence is independent of the composition of the hydrothermal fluid, and these fluids do not need to be reduced. At Macmillan Pass, Magnall et al. (2020b) show that hot (>300°C), carbonate-rich hydrothermal fluids resulted in the dissolution of barite and precipitation of barium carbonate minerals. Barium carbonates are not observed in the Anarraaq sulfide deposit (Kelley et al., 2004b; this study), which suggests a different mechanism drove barite dissolution. Barite is also soluble in hot (>200°C), strongly reducing fluids (Hanor, 2000). However, fluid inclusion microthermometry from the Main deposit suggests that hydrothermal fluids were 110°–180°C (Leach et al., 2004), and in this temperature range, barite has retrograde solubility (Bowers et al., 1984). An alternative explanation is that increasing temperatures associated with the onset of local hydrothermal activity resulted in the acceleration of diagenetic processes—a process that has been reported in modern methane seep settings (Guaymas basin; Einsele et al., 1980; Berndt et al., 2016) and inferred at Macmillan Pass (Magnall et al., 2020a). In the Anarraaq sulfide deposit, hydrothermal acceleration of diagenetic processes could explain the extensive dissolution of diagenetic barite and the formation of large volumes of diagenetic pyrite and potentially authigenic carbonate. Furthermore, pore water methane and/or the bisulfide produced by SD-AOM could have contributed to a trap for hydrothermal metals. Based on logging and petrographic observations, we suggest that the dominant sulfide component of the northwestern, lower-grade part of the Anarraaq sulfide deposit is diagenetic pyrite. Previous workers (Kelley et al., 2004b; Schardt et al., 2008) have considered this zone to represent a hydrothermal pyrite halo. Instead, we consider the distal hydrothermal signature in this part of the deposit to be characterized by relatively minor sphalerite and Fe sulfide veining (Fig. 8I). Further study of the chemistry of these different pyrite generations will be necessary to assess if the chemical and isotopic composition of the diagenetic pyrite can be distinguished from the hydrothermal pyrite, as has been shown in other deposits (Large et al., 2009; Gadd et al., 2016; Magnall et al., 2020c). In summary, understanding the mechanisms of formation of hydrothermal ore deposits that form in the shallow subsurface of marine sediment requires an understanding of key diagenetic processes. At Red Dog, diagenetic processes created physical and chemical traps for later Zn-Pb mineralization. These processes must be considered when developing an exploration program, and extra care must be taken to distinguish diagenetic and hydrothermal components during the interpretation of sedimentological, geochemical, and geophysical data. Conclusions The Anarraaq barite body contains barite, carbonate, and pyrite, which are interpreted to form in the subsurface by methane-driven diagenetic processes (i.e., at a paleo-SMT). New cross sections show that the Anarraaq sulfide deposit contains Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest 1822 REYNOLDS ET AL. two distinct zones of Zn-Pb mineralization (Upper and Lower), each of which contains evidence for a similar assemblage of diagenetic minerals. This suggests that hydrothermal mineralization overprinted paleo-SMTs. The presence of 10-cm quartz- and sulfide-infilled cavities in the Lower zone suggests that the host rock was mostly lithified before the onset of hydrothermal mineralization. Dissolution of barite and sedimentary and authigenic carbonate in the subsurface was an important permeability-generating process at Anarraaq. The cycle of barite precipitation and dissolution linked to the upward migration of an SMT over time may have introduced additional porosity and permeability to the host sediment prior to hydrothermal mineralization. Hydrothermal acceleration of diagenetic processes could explain the extensive dissolution of barite and the large volume of diagenetic pyrite and potentially authigenic carbonate present in the sulfide deposit. Furthermore, pore water methane and/or the bisulfide produced by SD-AOM could have contributed as a trap for metals transported by hydrothermal fluids. The Re-Os isochron ages of Ikalukrok mudstone, diagenetic pyrite, and hydrothermal pyrite (isochron constructed in combination with diagenetic pyrite data) at Anarraaq are all within error of the Re-Os isochron age of the Main deposit in the Red Dog mine area (Morelli et al., 2004). This indicates that hydrothermal mineralizing events were broadly synchronous on a regional scale. The initial Os composition of the Anarraaq isochrons is consistent with expected values for Devonian and Mississippian seawater; it is unlikely that a mantle source had a significant influence on hydrothermal fluids. Barite; pyrite layers, laminae, and nodules; textures suggestive of carbonate replacement; or some combination of these are reported in all known Red Dog deposits and prospects. This suggests that methane-driven diagenetic alteration was an important precondition for CD Zn-Pb mineralization in the Red Dog district. This idea is consistent with recent reports that SD-AOM was an important diagenetic process in sediments hosting CD deposits in Howard’s Pass and MacMillan Pass. We propose that the size, mineralogy, and pore water composition of paleo-SMTs may have played a key role in ground preparation for forming large CD deposits. Acknowledgments This work was funded by Teck Resources Limited and by a Collaborative Research and Development grant and a Vanier Canada Graduate Scholarship from the Natural Sciences and Engineering Research Council of Canada. Special thanks to Pilar Lecumberri-Sanchez at the University of Alberta and Mo Snyder at Acadia University, and to the geologists active in the Red Dog district for technical assistance and geologic discussions. We acknowledge Gema Olivo for her early review of this manuscript and Elizabeth Turner, Karen Kelley, and Craig Johnson for their thorough reviews of this paper. REFERENCES Arning, E.T., Gaucher, E.C., van Berk, W., and Schulz, H.M., 2015, Hydrogeochemical models locating sulfate-methane transition zone in marine sediments overlying black shales: A new tool to locate biogenic methane?: Marine and Petroleum Geology, v. 59, p. 563–574, doi: 10.1016/j. marpetgeo.2014.10.004. Ayuso, R.A., Kelley, K.D., Leach, D.L., Young, L.E., Slack, J.F., Wandlass, G., Lyon, A.M., and Dillingham, J.L., 2004, Origin of the Red Dog Zn-Pb-Ag deposits, Brooks Range, Alaska: Evidence from regional Pb and Sr isotope sources: Economic Geology, v. 99, p. 1533–1553, doi: 10.2113/ gsecongeo.9.7.1533. Barnes, R.O., and Goldberg, E.D., 1976, Methane production and consumption in anoxic marine sediments: Geology, May, p. 297–300, doi: 10.1130/0091-7613(1976)42.0.CO;2. Berndt, C., Hensen, C., Mortera-Gutierrez, C., Sarkar, S., Geilert, S., Schmidt, M., Liebetrau, V., Kipfer, R., Scholz, F., Doll, M., and Muff, S., 2016, Rifting under steam: How rift magmatism triggers methane venting from sedimentary basins: Geology, v. 44, no. 9, p. 767–770. Bishop, J.K.B., 1988, The barite-opal-organic carbon association in oceanic particulate matter: Nature, v. 332, p. 341–343, doi: 10.1038/332341a0. Blevings, S., Kraft, J., Stemler, J., and Krolak, T., 2013, An overview of the structure, stratigraphy, and Zn-Pb-Ag deposits of the Red Dog district, northwestern Alaska: Society of Economic Geologists, Special Publication 17, p. 361–387. Bowers, T.S., Jackson, K.J., and Helgeson, H.C., 1984, Equilibrium activity diagrams: Berlin, Springer-Verlag, p. 1–290. Brumsack, H.J., 1986, The inorganic geochemistry of Cretaceous black shales (DSDP Leg 41) in comparison to modern upwelling sediments from the Gulf of California: Geological Society of London, Special Publication 21, p. 447–462, doi: 10.1144/GSL.SP.1986.021.01.30. Burke, W.H., Denison, R.E., Hetherington, E.A., Koepnick, R.B., Nelson, H.F., and Otto, J.B., 1982, Variation of seawater Sr87Sr86 throughout Phanerozoic time: Geology, v. 10, p. 516–519, doi: 10.1130/0091-7613(1982)10<516:VOSSTP>2.0.CO. Cartwright, J.A., 1994, Episodic basin-wide fluid expulsion from geopressured shale sequences in the North Sea basin: Geology, v. 22, p. 447–450, doi: 10.1130/0091-7613(1994)022<0447:EBWFEF>2.3.CO;2. Cooke, D., Bull, S., Large, R., and McGoldrick, P., 2000, The importance of oxidized brines for the formation of Australian Proterozoic stratiform sediment-hosted Pb-Zn (Sedex) deposits: Economic Geology, v. 95, p. 1–18, doi: 10.2113/gsecongeo.95.1.1. Creaser, R.A., Sannigrahi, P., Chacko, T., and Selby, D., 2002, Further evaluation of the Re-Os geochronometer in organic-rich sedimentary rocks: A test of hydrocarbon maturation effects in the Exshaw Formation, Western Canada sedimentary basin: Geochimica et Cosmochimica Acta, v. 66, p. 3441–3452, doi: 10.1016/S0016-7037(02)00939-0. Davies, R.J., Goulty, N.R., and Meadows, D., 2008, Fluid flow due to the advance of basin-scale silica reaction zones: Bulletin of the Geological Society of America, v. 120, p. 195–206, doi: 10.1130/B26099.1. Dean, W.E., and Schreiber, B.C., 1978, Authigenic barite: Leg 41 Deep Sea Drilling Project, v. 41, p. 915–931. Denison, R.E., Koepnick, R.B., Burke, W.H., Hetherington, E.A., and Fletcher, A., 1994, Construction of the Mississippian, Pennsylvanian and Permian seawater 87Sr/86Sr curve: Chemical Geology: Isotope Geoscience Section, v. 112, p. 145–167. De Vera, J., 2005, Structure of the Red Dog district, Western Brooks Range, Alaska: Ph.D. thesis, London, University of London, 633 p., www.arlis.org/ docs/vol1/D/773627462.pdf. De Vera, J., McClay, K., and King, A., 2004, Structure of the Red Dog district, Western Brooks Range, Alaska: Economic Geology, v. 99, p. 1415– 1434, doi: 10.2113/gsecongeo.99.7.1415. Dickens, G.R., 2001, Sulphate profiles and barium fronts in sediments on the Blake Ridge: Present and past methane fluxes through a large gas hydrate reservoir: Geochimica et Cosmochimica Acta, v. 65, p. 529–543. Dumoulin, J.A., Harris, A.G., Blome, C.D., and Young, L.E., 2004, Depositional settings, correlation, and age of Carboniferous rocks in the western Brooks Range, Alaska: Economic Geology, v. 99, p. 1355–1384. Dumoulin, J.A., Johnson, C.A., Slack, J.F., Bird, K.J., Whalen, M.T., Moore, T.E., Harris, A.G., and Sullivan, P.B.O., 2014, Carbonate margin, slope, and basin facies of the Lisburne Group (Carboniferous-Permian) in northern Alaska: Society of Sedimentary Geologists (SEPM), Special Publication 105, p. 211–236, doi: 10.2110/sepmsp.105.02. Einsele, G., Gieskes, J.M., Curray, J., Moore, D.M., Aguayo, E., Aubry, M.-P., Fornari, D., Guerrero, J., Kastner, M., Kelts, K., Lyle, M., Matoba, Y., Molina-Cruz, A., Niemitz, J., et al., 1980, Intrusion of basaltic sills into highly porous sediments, and resulting hydrothermal activity: Nature, v. 283, p. 441–445. Feng, D., and Roberts, H.H., 2011, Geochemical characteristics of the barite deposits at cold seeps from the northern Gulf of Mexico continental slope: Earth and Planetary Science Letters, v. 309, no. 1–2, p. 89–99. Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest ANARRAAQ DEPOSIT, RED DOG DISTRICT, ALASKA Fernandes, N.A., Gleeson, S.A., Magnall, J.M., Creaser, R.A., Martel, E., Fischer, B.J., and Sharp, R., 2017, The origin of Late Devonian (Frasnian) stratiform and stratabound mudstone-hosted barite in the Selwyn basin, Northwest Territories, Canada: Marine and Petroleum Geology, v. 85, p. 1–15. Gadd, M.G., Layton-Matthews, D., Peter, J.M., and Paradis, S.J., 2016, The world-class Howard’s Pass SEDEX Zn-Pb district, Selwyn basin, Yukon. Part I: Trace element compositions of pyrite record input of hydrothermal, diagenetic, and metamorphic fluids to mineralization: Mineralium Deposita, v. 51, no. 3, p. 319–342. Geboy, N.J., Tripathy, G.R., Ruppert, L.F., Eble, C.F., Blake, B.M., Hannah, J.L., and Stein, H.J., 2015, Re-Os age for the Lower-Middle Pennsylvanian boundary and comparison with associated palynoflora: International Journal of Coal Geology, v. 140, p. 23–30, doi: 10.1016/j.coal.2015.01.002. Goodfellow, W.D., Lydon, J.W., and Turner, R.J., 1993, Geology and genesis of stratiform sediment-hosted (SEDEX) zinc-lead-silver sulphide deposits: Geological Association of Canada, Special Paper 40, p. 201–252. Greinert, J., Bollwerk, S.M., Derkachev, A., Bohrmann, G., and Suess, E., 2002, Massive barite deposits and carbonate mineralization in the Derugin basin, Sea of Okhotsk: Precipitation processes at cold seep sites: Earth and Planetary Science Letters, v. 203, no. 1, p. 165–180. Gwiazda, R.H., Paull, C.K., Caress, D., Preston, C.M., Lundsten, E., and Anderson, K., 2019, The extent of fault-associated modern authigenic barite deposits offshore northern Baja California revealed by high resolution mapping: Frontiers in Marine Science, v. 6, p. 460. Hanor, J.S., 2000, Barite-celestine geochemistry and environments of formation: Reviews in Mineralogy and Geochemistry, v. 40, p. 193–275. Hitzman, M.W., 1999, Routine staining of drill core to determine carbonate mineralogy and distinguish carbonate alteration textures: Mineralium Deposita, v. 34, no. 8, p. 794–798. Hnatyshin, D., Kontak, D.J., Turner, E.C., Creaser, R.A., Morden, R., and Stern, R.A., 2016, Geochronologic (ReOs) and fluid-chemical constraints on the formation of the Mesoproterozoic-hosted Nanisivik ZnPb deposit, Nunavut, Canada: Evidence for early diagenetic, low-temperature conditions of formation: Ore Geology Reviews, v. 79, p. 189–217, doi: 10.1016/ j.oregeorev.2016.05.017. Hoehler, T.M., Alperin, M.J., Albert, D.B., and Martens, C.S., 1994, Field and laboratory studies of methane oxidation in an anoxic marine sediment: Evidence for a methanogen‐sulfate reducer consortium: Global Biogeochemical Cycles, v. 8, p. 451–463, doi: 10.1029/94GB01800. Isaacs, C.M., 1982, Influence of rock composition on kinetics of silica phase changes in the Monterey Formation, Santa Barbara area, California: Geology, v. 10, no. 6, p. 304–308. Iversen, N., and Jørgensen, B.B., 1985, Anaerobic methane oxidation rates at the sulphate-methane transition in marine sediments from Kattegat and Skagerrak (Denmark): Limnology and Oceanography, v. 30, p. 944–955. Johnson, C.A., Kelley, K.D., and Leach, D.L., 2004, Sulfur and oxygen isotopes in barite deposits of the western Brooks Range, Alaska, and implications for the origin of the Red Dog massive sulfide deposits: Economic Geology, v. 99, p. 1435–1448, doi: 10.2113/gsecongeo.99.7.1435. Johnson, C.A., Emsbo, P., Poole, F.G., and Rye, R.O., 2009, Sulfur- and oxygen-isotopes in sediment-hosted stratiform barite deposits: Geochimica et Cosmochimica Acta, v. 73, p. 133–147, doi: 10.1016/j.gca.2008.10.011. Johnson, C.A., Slack, J.F., Dumoulin, J.A., Kelley, K.D., and Falck, H., 2018, Sulfur isotopes of host strata for Howards Pass (Yukon-Northwest Territories) Zn-Pb deposits implicate anaerobic oxidation of methane, not basin stagnation: Geology, v. 46, p. 619–622, doi: 10.1130/G40274.1. Jørgensen, B.B., Böttcher, M.E., Lüschen, H., Neretin, L.N., and Volkov, I.I., 2004, Anaerobic methane oxidation and a deep H2S sink generate isotopically heavy sulfides in Black Sea sediments: Geochimica et Cosmochimica Acta, v. 68, p. 2095–2118, doi: 10.1016/j.gca.2003.07.017. Kasten, S., Zabel, M., Heuer, V., and Hensen, C., 2003, Processes and signals of nonsteady-state diagenesis in deep-sea sediments and their pore waters, in Wefer, G., ed., The south Atlantic in the late Quaternary: Berlin, Heidelberg, Springer, p. 431–459. Kelley, K.D., and Jennings, S., 2004, A special issue devoted to barite and ZnPb-Ag deposits in the Red Dog district, western Brooks Range, northern Alaska: Economic Geology, v. 99, no. 7, p. 1267–1280. Kelley, K.D., Leach, D.L., Johnson, C.A., Clark, J.L., Fayek, M., Slack, J.F., Anderson, V.M., Ayuso, R.A., and Ridley, W.I., 2004a, Textural, compositional, and sulfur isotope variations of sulfide minerals in the Red Dog Zn-Pb-Ag deposits, Brooks Range, Alaska: Implications for ore formation: Economic Geology, v. 99, p. 1509–1532, doi: 10.2113/gsecongeo.99.7.1509. 1823 Kelley, K.D., Dumoulin, J.A., and Jennings, S., 2004b, The Anarraaq Zn-PbAg and barite deposit, northern Alaska: Evidence for replacement of carbonate by barite and sulfides: Economic Geology, v. 99, p. 1577–1591, doi: 10.2113/gsecongeo.99.7.1577. Kendall, B., Creaser, R.A., Ross, G.M., and Selby, D., 2004, Constraints on the timing of Marinoan “Snowball Earth” glaciation by 187Re-187Os dating of a Neoproterozoic, post-glacial black shale in western Canada: Earth and Planetary Science Letters, v. 222, p. 729–740. Knittel, K., and Boetius, A., 2009, Anaerobic oxidation of methane: Progress with an unknown process: Annual Review of Microbiology, v. 63, p. 311–334, doi: 10.1146/annurev.micro.61.080706.093130. Krolak, T., Palmer, K., Lacouture, B., and Paley, N., 2017, Red Dog mine, Alaska, USA: Teck Resources Limited, NI 43-101 Technical Report, 143 p. Large, R.R., Danyushevsky, L., Hollit, C., Maslennikov, V., Meffre, S., Gilbert, S., Bull, S., Scott, R., Emsbo, P., Thomas, H., Singh, B., and Foster, J., 2009, Gold and trace element zonation in pyrite using a laser imaging technique: Implications for the timing of gold in orogenic and Carlin-style sediment-hosted deposits: Economic Geology, v. 104, no. 5, p. 635–668. Lash, G.G., 2015, Pyritization induced by anaerobic oxidation of methane (AOM): An example from the Upper Devonian shale succession, western New York, USA: Marine and Petroleum Geology, v. 68, p. 520–535, doi: 10.1016/j.marpetgeo.2015.10.002. Leach, D.L., Marsh, E., Emsbo, P., Rombach, C.S., Kelley, K.D., and Anthony, M., 2004, Nature of hydrothermal fluids at the shale-hosted Red Dog Zn-Pb-Ag deposits, Brooks Range, Alaska: Economic Geology, v. 99, p. 1449–1480, doi: 10.2113/gsecongeo.99.7.1449. Leach, D.L., Bradley, D.C., Huston, D., Pisarevsky, S.A., Taylor, R.D., and Gardoll, J., 2010, Sediment-hosted lead-zinc deposits in Earth history: Economic Geology, v. 105, p. 593–625. Lin, Q., Wang, J., Algeo, T.J., Sun, F., and Lin, R., 2016, Enhanced framboidal pyrite formation related to anaerobic oxidation of methane in the sulfate-methane transition zone of the northern South China Sea: Marine Geology, v. 379, p. 100–108. Lin, Z., Sun, X., Peckmann, J., Lu, Y., Xu, L., Strauss, H., Zhou, H., Gong, J., Lu, H., and Teichert, B.M.A., 2016, How sulfate-driven anaerobic oxidation of methane affects the sulfur isotopic composition of pyrite: A SIMS study from the South China Sea: Chemical Geology, v. 440, p. 26–41, doi: 10.1016/j.chemgeo.2016.07.007. Ludwig, K., 2001, Isoplot 3.00, a geochronologic toolkit for Microsoft Excel: Berkeley Geochronological Center, Special Publication 4, p. 71. Magnall, J.M., Gleeson, S.A., and Paradis, S., 2015, The importance of siliceous radiolarian-bearing mudstones in the formation of sediment-hosted Zn-Pb: Ba mineralization in the Selwyn basin, Yukon, Canada: Economic Geology, v. 110, p. 2139–2146. Magnall, J.M., Gleeson, S.A., Stern, R.A., Newton, R.J., Poulton, S.W., and Paradis, S., 2016, Open system sulphate reduction in a diagenetic environment—isotopic analysis of barite (δ34S and δ18O) and pyrite (δ34S) from the Tom and Jason Late Devonian Zn-Pb-Ba deposits, Selwyn basin, Canada: Geochimica et Cosmochimica Acta, doi: 10.1016/j.gca.2016.02.015. Magnall, J.M., Gleeson, S.A., Poulton, S.W., Gordon, G.W., and Paradis, S., 2018, Links between seawater paleoredox and the formation of sedimenthosted massive sulphide (SHMS) deposits—Fe speciation and Mo isotope constraints from Late Devonian mudstones: Chemical Geology, v. 490, p. 45–60, doi: 10.1016/j.chemgeo.2018.05.005. Magnall, J.M., Gleeson, S.A., and Paradis, S., 2020a, A new subseafloor replacement model for the MacMillan Pass clastic-dominant Zn-Pb ± Ba deposits (Yukon, Canada): Economic Geology, v. 115, p. 953–959. Magnall, J.M., Gleeson, S.A., Creaser, R.A., Paradis, S., Glodny, J., and Kyle, J.R., 2020b, The mineralogical evolution of the clastic dominant (CD)-type Zn-Pb ± Ba deposits at Macmillan Pass (Yukon, Canada)—tracing subseafloor barite replacement in the layered mineralization: Economic Geology, v. 115, p. 961–979. Magnall, J.M., Gleeson, S.A., Hayward, N., and Rocholl, A., 2020c, Massive sulfide Zn deposits in the Proterozoic did not require euxinia: Geochemical Perspectives Letters, v. 13, p. 19–24. Moore, D.W., Young, L.E., Modene, J.S., and Plahuta, J.T., 1986, Geologic setting and genesis of the Red Dog zinc-lead-silver deposit, western Brooks Range, Alaska: Economic Geology, v. 81, p. 1696–1727, doi: 10.2113/ gsecongeo.81.7.1696. Moore, T.E., Wallace, W., Bird, K., Karl, S., Mull, C., and Dillon, J., 1994, Geology of northern Alaska, in Plafker, G., and Berg, H., eds., The geology of Alaska: Geological Society of America, Decade of North America Geology, v. G-1, p. 49–140. Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest 1824 REYNOLDS ET AL. Morelli, R.M., Creaser, R.A., Selby, D., Kelley, K.D., Leach, D.L., and King, A.R., 2004, Re-Os sulfide geochronology of the Red Dog sediment-hosted Zn-Pb-Ag deposit, Brooks Range, Alaska: Economic Geology, v. 99, p. 1569–1576, doi: 10.2113/gsecongeo.99.7.1569. Paull, C.K., Chanton, J.P., Neumann, A.C., Coston, J.A., Martens, C.S., and Showers, W., 1992, Indicators of methane-derived carbonates and chemosynthetic organic carbon deposits: Examples from the Florida escarpment: Palaios, v. 7, p. 361–375, doi: 10.2307/3514822. Paytan, A., Kastner, M., Martin, E.E., MacDougall, J.D., and Herbert, T., 1993, Marine barite as a monitor of seawater strontium isotope composition: Nature, v. 366, p. 445–449, doi: 10.1038/366445a0. Peucker-Ehrenbrink, B., and Ravizza, G., 2000, The marine osmium isotope record: Terra Nova, v. 12, p. 205–219, doi: 10.1046/j.1365-3121.2000.00295.x. Reeburgh, W.S., 1976, Methane consumption in Cariaco trench waters and sediments: Earth and Planetary Science Letters, v. 28, p. 337–344, doi: 10.1016/0012-821X(76)90195-3. Reynolds, M.A., Gingras, M.K., Gleeson, S.A., and Stemler, J.U., 2015, More than a trace of oxygen: Ichnological constraints on the formation of the giant Zn-Pb-Ag ± Ba deposits, Red Dog district, Alaska: Geology, v. 43, doi: 10.1130/G36954.1. Schardt, C., Garven, G., Kelley, K.D., and Leach, D.L., 2008, Reactive flow models of the Anarraaq Zn-Pb-Ag deposit, Red Dog district, Alaska: Mineralium Deposita, v. 43, p. 735–757, doi: 10.1007/s00126-008-0193-3. Selby, D., and Creaser, R.A., 2003, Re-Os geochronology of organic rich sediments: An evaluation of organic matter analysis methods: Chemical Geology, v. 200, p. 225–240, doi: 10.1016/S0009-2541(03)00199-2. ——2005, Direct radiometric dating of the Devonian-Mississippian timescale boundary using the Re-Os black shale geochronometer: Geology, v. 33, p. 545–548, doi: 10.1130/G21324.1. Slack, J.F., Dumoulin, J.A., Schmidt, J.M., Young, L.E., and Rombach, C.S., 2004, Paleozoic sedimentary rocks in the Red Dog Zn-Pb-Ag district and vicinity, western Brooks Range, Alaska: Provenance, deposition, and metallogenic significance: Economic Geology, v. 99, p. 1385–1414, doi: 10.2113/ gsecongeo.99.7.1385. Slack, J.F., Selby, D., and Dumoulin, J.A., 2015, Hydrothermal, biogenic, and seawater components in metalliferous black shales of the Brooks Range, Alaska: Synsedimentary metal enrichment in a carbonate ramp setting: Economic Geology, v. 110, p. 653–675, doi: 10.2113/econgeo.110.3.653. Slack, J.F., Falck, H., Kelley, K.D., and Xue, G.G., 2017, Geochemistry of host rocks in the Howards Pass district, Yukon-Northwest Territories, Canada: Implications for sedimentary environments of Zn-Pb and phosphate mineralization: Mineralium Deposita, v. 52, p. 565–593. Smoliar, M.I., Walker, R.J., and Morgan, J.W., 1996, Re-Os ages of group IIA, IIIA, IVA, and IVB iron meteorites: Science, v. 271, p. 1099–1102. Snyder, G.T., Dickens, G.R., and Castellini, D.G., 2007, Labile barite contents and dissolved barium concentrations on Blake Ridge: New perspectives on barium cycling above gas hydrate systems: Journal of Geochemical Exploration, v. 95, p. 48–65, doi: 10.1016/j.gexplo.2007.06.001. Suess, E., 2014, Marine cold seeps and their manifestations: Geological control, biogeochemical criteria and environmental conditions: International Journal of Earth Sciences, v. 103, p. 1889–1916, doi: 10.1007/ s00531-014-1010-0. Torres, M.E., Brumsack, H.J., Bohrmann, G., and Emeis, K.C., 1996, Barite fronts in continental margin sediments: A new look at barium remobilization in the zone of sulfate reduction and formation of heavy barites in diagenetic fronts: Chemical Geology, v. 127, p. 125–139, doi: 10.1016/0009-2541(95)00090-9. Torres, M.E., Bohrmann, G., Dubé, T.E., and Poole, F.G., 2003, Formation of modern and Paleozoic stratiform barite at cold methane seeps on continental margins: Geology, v. 31, p. 897, doi: 10.1130/G19652.1. Turekian, K.K., and Pegram, W.J., 1997, Os isotope record in a Cenozoic deep-sea core: Its relation to global tectonics and climate, in Ruddiman, W.F., ed., Tectonic uplift and climate change: Boston, Springer, p. 383–397, doi: 10.1007/978-1-4615-5935-1_17. Veizer, J., Ala, D., Azmy, K., Bruckschen, P., Buhl, D., Bruhn, F., Garden, G.A.F., Diener, A., Ebneth, S., Godderis, Y., Jasper, T., Korte, C., Pawellek, F., Podlaha, O.G., et al., 1999, 87Sr/86Sr, δ13C and δ18O evolution of Phanerozoic seawater: Chemical Geology, v. 161, p. 59–88, doi: 10.1016/ S0009-2541(99)00081-9. Volpi, V., Camerlenghi, A., Hillenbrand, C.-D., Rebescon, M., and Ivaldiz, R., 2003, Effects of biogenic silica on sediment compaction and slope stability on the Pacific margin of the Antarctic Peninsula: Basin Research, v. 15, p. 339–363, doi: 10.1046/j.1365-2117.2003.00210.x. Wehrmann, L.M., Arndt, S., März, C., Ferdelman, T.G., and Brunner, B., 2013, The evolution of early diagenetic signals in Bering Sea subseafloor sediments in response to varying organic carbon deposition over the last 4.3 Ma: Geochimica et Cosmochimica Acta, v. 109, p. 175–196, doi: 10.1016/j. gca.2013.01.025. Young, L.E., 2004, A geologic framework for mineralization in the Western Brooks Range, Alaska: Economic Geology, v. 99, p. 1281–1306, doi: 10.2113/gsecongeo.99.7.1281. Merilie Reynolds is a mineral deposits geologist at the Northwest Territories Geological Survey (Canada). She received a B.A. degree in geology from Smith College in 2008 and a Ph.D. degree in earth and atmospheric sciences from the University of Alberta in 2019. Prior to starting graduate school, she worked for Barrick Gold Exploration in the Carlin gold camps of northern Nevada. Downloaded from http://pubs.geoscienceworld.org/segweb/economicgeology/article-pdf/116/8/1803/5428899/4849_reynolds_et_al.pdf by guest
https://openalex.org/W4380364472	https://www.researchsquare.com/article/rs-2999671/latest.pdf	English	null	The association between number of oocytes retrieved and cumulative live birth rate in different female age strata	Research Square (Research Square)	2,023	cc-by	5,795	The association between number of oocytes retrieved and cumulative live birth rate in different female age strata Peixin Wang Zhejiang University Chenqiong Zhao Zhejiang University Wen Xu Zhejiang University Xiaoying Jin Sir Run Run Shaw Hospital Songying Zhang Sir Run Run Shaw Hospital Haiyan Zhu (  haiyan@zju.edu.cn ) Sir Run Run Shaw Hospital Conclusion: Age and number of oocytes retrieved are two independent factors affecting the CLBR. The discrepancy of the minimum number of oocytes retrieved for patients with different ages to achieve the ideal CLBR is instructive for clinical practice. The practice of controlling the stimulation dose is feasible for patients ≤35 years who can achieve ideal CLBR once the number of oocytes obtained is more than 6 while additional stimulation cycles and accumulation of embryos is necessary for elderly group especially those ≥38 years old who need to retrieve more than 14 oocytes to obtain higher live birth rate. Purpose: To evaluate the association between the number of oocytes retrieved and cumulative live birth rate (CLBR) in different female age strata. Results: The group ≥36 years exhibited lower cumulative pregnancy rates (CPRs) and cumulative live birth rates (CLBRs), which are proportional to the number of oocytes retrieved but opposite to increasing age. Multivariate logistic regression analysis revealed that the age and number of oocytes retrieved remainsignificant independent predictive factors. (P<0.001) Article Keywords: Cumulative live birth rate, live birth rate, number of oocytes retrieved, IVF/ICSI outcome. License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Version of Record: A version of this preprint was published at Scientific Reports on September 4th, 2023. See the published version at https://doi.org/10.1038/s41598-023-41842-7. Page 1/15 Methods: 17,931 women undergoing their first IVF/ICSI-ET cycle in the Sir Run Run Shaw Hospital of Zhejiang University were grouped by age (A: ≤35 years; B: ≥36 years) as well as the number of oocytes retrieved (a: ≤5; b:6-9; c:10-14; d: ≥15). Multivariate regression analysis was performed to assess the OR of CLBR for the variable ‘age’ and ‘number of oocytes retrieved’. 1. Introduction The purpose of this research was to reveal the relationship between the number of oocytes retrieved and the cumulative live birth rate (CLBR) in different female age strata after categorizing patients by age. 1. Introduction Controlled ovarian stimulation is an important part of assisted reproductive technologies(ARTs) [1, 2] With the substantial advancement of ART in both clinical and laboratory, the number of freeze-thaw embryo transfer cycle have increased along with the pregnancy rate. [3, 4] Clinical or ongoing pregnancy rate, embryo implantation rate and live birth rate per transfer have consuetudinary been used to evaluate the success rate of ART while the cumulative live birth rate (CLBR) appears to be a better one currently. The CLBR, which mainly refers to the first live birth after using all embryos for an integrated cycle (including fresh and subsequent freeze-thaw cycles), varies synchronously with the number of oocytes retrieved. [5, 6]. Previous studies have manifested that CLBR can even reach as high as 70% once the oocytes retrieved are more than 25 and with no obvious plateau. [7] Despite the benefit of ART in achieving adequate multiple follicles and providing considerable embryos for selection, the optimal number of oocytes need Page 2/15 Page 2/15 to retrieve and corresponding probability of live birth in elderly women after each ovulation induction therapy have not been determined. As the two-child policy fully issued in China since 2015, the demand for ART treatment has been growing steadily among women of advanced reproductive age. [8] An additional 90 million couples in China intend to have a second child according to official estimation, of whom 60% are reported to be older than 35 years, and 50% will be at least 40 years old. [8] Female age becomes the main challenge impacting the success rate of ART. Women ≥ 35 years old were verified to present a significant decline in the pregnancy rates and live birth rates. [9–11] As is known to all, ovarian function and oocyte quality are all inextricably linked with female age, with the group ≥ 35 years of age certainly experiencing less chance to achieve high number of oocytes retrieved as well as higher risk of developing aneuploidy oocytes and embryos. [12–14] Therefore, challenges that often encountered in clinical practice are as follows: How many oocytes are sufficient to achieve a live birth for women of advanced reproductive age? How many IVF/ICSI cycles are required for the elderly women to increase the probability of live birth? 2.1．Characteristics of basic clinical data In general, there are totally 18670 embryo transfer cycles in the aged ≤35 years group and 4498 in the aged ≥36 years group, Day3 embryos were transferred in the majority of embryo transfer cycles. We can observe that the group (age ≥36 years of age) had a significantly lower pregnancy rate 39.9% vs. 61.0%; OR: 2.351; 95% CI: 2.199-2.512; P<0.001) and implantation rate (27.5% vs. 44.8%, OR: 2.143; 95% CI: 2.031-2.261; P<0.001)compared with the group (age ≤ 35 years of age), which accords with a lower live birth rate (19.6% vs. 37.2%; OR: 2.429; 95% CI: 2.288-2.577; P<0.001). However, no significant differences in live birth weight and preterm birth rate were found between the two groups. 2.1．Characteristics of basic clinical data Characteristics of basic clinical data 2.2．Clinical outcomes of fresh cycle and freeze-thaw cycle Page 4/15 2.2．Clinical outcomes of fresh cycle and freeze-thaw cycle As illustrated in Table 2, the fertilization rate of IVF were higher in the group (women≥36 years of age) (P<0.01) while the number of oocytes retrieved and good quality embryos rate remained lower compared to other group (women ≤35 years of age). In general, there are totally 18670 embryo transfer cycles in the aged ≤35 years group and 4498 in the aged ≥36 years group, Day3 embryos were transferred in the majority of embryo transfer cycles. We can observe that the group (age ≥36 years of age) had a significantly lower pregnancy rate 39.9% vs. 61.0%; OR: 2.351; 95% CI: 2.199-2.512; P<0.001) and implantation rate (27.5% vs. 44.8%, OR: 2.143; 95% CI: 2.031-2.261; P<0.001)compared with the group (age ≤ 35 years of age), which accords with a lower live birth rate (19.6% vs. 37.2%; OR: 2.429; 95% CI: 2.288-2.577; P<0.001). However, no significant differences in live birth weight and preterm birth rate were found between the two groups. As illustrated in Table 2, the fertilization rate of IVF were higher in the group (women≥36 years of age) (P<0.01) while the number of oocytes retrieved and good quality embryos rate remained lower compared to other group (women ≤35 years of age). In general, there are totally 18670 embryo transfer cycles in the aged ≤35 years group and 4498 in the aged ≥36 years group, Day3 embryos were transferred in the majority of embryo transfer cycles. We can observe that the group (age ≥36 years of age) had a significantly lower pregnancy rate 39.9% vs. 61.0%; OR: 2.351; 95% CI: 2.199-2.512; P<0.001) and implantation rate (27.5% vs. 44.8%, OR: 2.143; 95% CI: 2.031-2.261; P<0.001)compared with the group (age ≤ 35 years of age), which accords with a lower live birth rate (19.6% vs. 37.2%; OR: 2.429; 95% CI: 2.288-2.577; P<0.001). However, no significant differences in live birth weight and preterm birth rate were found between the two groups. As illustrated in Table 2, the fertilization rate of IVF were higher in the group (women≥36 years of age) (P<0.01) while the number of oocytes retrieved and good quality embryos rate remained lower compared to other group (women ≤35 years of age). 2.1．Characteristics of basic clinical data The baseline data of 17931 women undergoing their first IVF/ICSI cycle were presented in Table1. Comparisons which mainly between patients ≤35 years of age and ≥36 years of age revealed significant discrepancies in levels of day3 basal FSH， LH and E2, duration of infertility and infertility factors. The group (age≥36 years of age) had significant longer duration of infertility (5.0 ± 4.0 vs 3.3 ± 2.3, p<0.001) but with higher basal FSH (8.4 ± 3.4 vs 7.1 ± 2.6, p<0.001) and E2(43.0 ± 27.4 vs 41.0 ± 28.2, p<0.001) levels. In addition, the ovarian stimulation characteristics reported in Table 2 showed that stimulation protocol, duration of stimulation or insemination type all varied strikingly in two age groups. Page 3/15 Page 3/15 Table 1. Characteristics of basic clinical data ≤35 years of age ≥36 years of age statistic P value OR (95% CI) No. of stimulation cycle, n (%) 14567 (81.2%) 3364 (18.8%) 　　　 Age ( mean± SD) 29.9 ± 3.0 38.8 ± 2.6 -172.26 <0.001 D3 FSH level (IU/L) ( mean ± SD) 7.1 ± 2.6 8.4 ± 3.4 -18.841 <0.001 D3 E2 level (ng/L) ( mean ± SD) 41.0 ± 28.2 43.0 ± 27.4 -3.597 <0.001 D3 LH level (IU/L) ( mean ± SD) 4.8 ± 4.5 4.3 ± 4.7 4.982 <0.001 Infertility factor, n (%) 384.92 <0.001 Tubal factor 9384 (64.4%) 2260 (67.2%) 0.884 (0.817- 0.958) Endometriosis 675 (4.6%) 102 (3.0%) Ovulatory disorder 630 (4.3%) 49 (1.5%) Diminished ovarian reserve 532 (3.7%) 358 (10.6%) Male factor 2463 (16.9%) 418 (12.4%) Unexplained 883 (6.1%) 177 (5.3%) Duration of infertility (years) (mean ± SD) 3.3 ± 2.3 5.0 ± 4.0 -22.371 <0.001 　 Stimulation protocol, n (%) 1759.592 <0.001 Long protocol 8295 (56.9%) 881 (26.2%) 3.727 (3.429- 4.052) Short protocol 1625 (11.2%) 764 (22.7%) Ultralong protocol 1380 (9.5%) 241 (7.2%) Micro-stimulation protocol 771 (5.3%) 751 (22.3%) Others 2496 (17.1%) 727 (21.6%) Duration of stimulation (day) (mean ± SD) 9.8 ± 2.0 8.7 ± 2.7 21.851 <0.001 2 2 Cli i l f f h l d f h l Table 1. 2.3．The cumulative pregnancy rate and live birth rate. The cumulative pregnancy rate (CPR) and cumulative live birth rate (CLBR) after one complete IVF/ICSI cycle including fresh and subsequent frozen-thaw cycles from first oocyte retrieval, which were significantly lower in the aged ≥36 years group compared with the aged ≤35 years group (CPR: 45.0% vs. 74.9%; OR: 3.639; 95% CI: 3.367-3.932; P<0.001; CLBR: 34.4% vs. 68.3%; OR: 4.125 95% CI: 3.811-4.466; ) The cumulative pregnancy rate (CPR) and cumulative live birth rate (CLBR) after one complete IVF/ICSI cycle including fresh and subsequent frozen-thaw cycles from first oocyte retrieval, which were significantly lower in the aged ≥36 years group compared with the aged ≤35 years group (CPR: 45.0% vs. 74.9%; OR: 3.639; 95% CI: 3.367-3.932; P<0.001; CLBR: 34.4% vs. 68.3%; OR: 4.125 95% CI: 3.811-4.466; P<0.001; respectively),as shown in Table3. Moreover, statistical analysis was performed in different age strata after grouped according to the number of oocytes retrieved. As expected, the CLBR, taking into account fresh and frozen cycles, increased strikingly parallel with the number of oocytes retrieved in specific age strata but decreased with increasing age for a given number of retrieved oocytes (Fig.1 a,b,c). Page 5/15 Table2. Clinical outcomes of fresh cycle and freeze-thaw cycle ≤35 years of age ≥36 years of age statistic P value OR (95% CI) No.Stimulation cycle, n (%) 14567 (81.2%) 3364 (18.8%) No. of oocytes retrieved (mean ± SD) 10.5 ± 6.5 5.8 ± 4.6 48.902 <0.001 Use of ICSI, n (%) 4937 (33.9%) 1403 (41.7%) 73.019 <0.001 0.717 (0.664- 0.774) Fertilization rate (%) IVF 70154/101362 (69.2%) 8631/12061 (71.6%) 28.058 <0.001 0.893 (0.857- 0.931) ICSI 35589/46928 (75.8%) 5300/6807 (77.9%) 13.382 <0.001 0.892 (0.840- 0.949) Good quality embryo rate (%) 49109/103765 (47.3%) 6727/13704 (49.1%) 15.048 <0.001 0.932 (0.899- 0.966) No. of embryo transfer cycle 18670 4498 D3 embryo transfer, n (%) 15950 (85.4%) 3909 (86.9%) 6.395 0.011 0.884 (0.803- 0.973) Blastocyst transfer, n (%) 2719 (14.6%) 589 (13.1%) Thawing cycle, n 16543 4019 Thawing cycle transfer, n (%) 16369/16543 (98.9%) 3975/4019 (98.9%) 0.057 0.811 1.041 (0.747- 1.452) Good quality embryo transfer rate (%) 22398/33788 (66.3%) 4910/7981 (61.5%) 64.861 <0.001 1.230 (1.169- 1.294) Avg. No. of Blastocyst transfer, (mean ± SD) 1.6 ± 0.5 1.6 ± 0.5 0.564 0.571 Avg. 2.3．The cumulative pregnancy rate and live birth rate. No.D3 embryos, (mean ± SD) 1.9 ± 0.4 1.8 ± 0.5 5.029 <0.001 Endometial thickness (mm) , (mean ± SD) 9.4 ± 2.7 9.3 ±2.6 2.527 0.012 Per.transfer pregnancy rate, n (%) 11380/18670 (61 0%) 1795/4498 (39 9%) 654.601 <0.001 2.351 (2 199 Table2. Clinical outcomes of fresh cycle and freeze-thaw cycle 2.512) ) Per.transfer embryonic implement rate, n (%) 15136/33788 (44.8%) 2192/7981 (27.5%) 798.897 <0.001 2.143 (2.031- 2.261) Per.transfer embryonic live birth rate , n (%) 12582/33788 (37.2%) 1567/7981 (19.6%) 893.191 <0.001 2.429 (2.288- 2.577) Multiple birth rate (%) 2703/10040 (26.9%) 232/1249 (18.6%) 40.231 <0.001 1.615 (1.391- 1.875) Male infant rate (%) 5998/12582 (47.7%) 762/1567 (48.6%) 0.511 0.475 0.962 (0.866- 1.069) Wt.single pregnancy (g), (mean ± SD) 329.8 ± 525.6 3267.5 ± 538.3 1.543 0.123 Wt. gemellary pregnancy (g), (mean ± SD) 2444.5 ± 445.0 2455.1 ± 475.2 -0.34 0.734 <37 weeks preterm birth, n (%) 1427/10040 (14.2%) 172/1249 (13.8%) 0.179 0.673 1.037 (0.875- 1.230) <34 weeks preterm birth, n (%) 375/10040 (3.7%) 42/1249 (3.4%) 0.433 0.511 1.115 (0.806- 1.542) Page 7/15 Table3. Comparions with the cumulative pregnancy and live birth rate 　 ≤35 years of age (n=14567) ≥36 years of age (n=3364) statistic P value OR (95% CI) Cumulative pregnancy rate hCG(+) (≥50 IU/L), n(%) 11734/14567 (80.6%) 1840/3364 (54.7%) 993.183 <0.001 3.431 (3.169- 3.714) Clinical pregnancy, n(%) 10905/14567 (74.9%) 1514/3364 (45.0%) 1144.116 <0.001 3.639 (3.367- 3.932) Cumulative live birth, n (%) 9957/14567 (68.3%) 1156/3364 (34.4%) 1339.757 <0.001 4.125 (3.811- 4.466) Fresh embryo transfer, n(%) 1108/9957 (11.1%) 141/1156 (12.2%) 1.187 0.279 0.901 (0.748- 1.087) Frozen-thawed embryo transfer, n(%) 8849/9957 (88.9%) 1015/1156 (87.8%) 1.187 0.279 1.109 (0.920- 1.337) 2.4．Multivariate logistic regression analysis for the cumulative live birth rate. To control possible confounding variables that might have changed over the study period, a multivariate logistic regression analysis was applied to estimate the OR with SE and 95% CI of the CLBR for variable “age” and “number of oocytes retrieved”. As shown in Table 4, after adjustment for relevant confounders including duration of stimulation, duration of infertility, endometrial thickness, insemination type, fertilization rate, and good quality embryo rate, the age and number of oocytes retrieved remained the independent predictive factors (P<0.001). The OR of CLBR dropped from 3.695(95%CI：2.960-4.611) in the aged ≤35 years group to 3.164(95%CI：2.563-3.906) and 2.230 (95%CI：1.785-2.786) in the aged 36-37 years and 38-39 years group respectively. 2.3．The cumulative pregnancy rate and live birth rate. In addition, the fertilization rate, good embryo quality rate and insemination method also have prominent effect on CLBR. Table3. Comparions with the cumulative pregnancy and live birth rate 　 ≤35 years of age (n=14567) ≥36 years of age (n=3364) statistic P value OR (95% CI) Cumulative pregnancy rate hCG(+) (≥50 IU/L), n(%) 11734/14567 (80.6%) 1840/3364 (54.7%) 993.183 <0.001 3.431 (3.169- 3.714) Clinical pregnancy, n(%) 10905/14567 (74.9%) 1514/3364 (45.0%) 1144.116 <0.001 3.639 (3.367- 3.932) Cumulative live birth, n (%) 9957/14567 (68.3%) 1156/3364 (34.4%) 1339.757 <0.001 4.125 (3.811- 4.466) Fresh embryo transfer, n(%) 1108/9957 (11.1%) 141/1156 (12.2%) 1.187 0.279 0.901 (0.748- 1.087) Frozen-thawed embryo transfer, n(%) 8849/9957 (88.9%) 1015/1156 (87.8%) 1.187 0.279 1.109 (0.920- 1.337) 2.4．Multivariate logistic regression analysis for the cumulative live birth rate. To control possible confounding variables that might have changed over the study period a multivariate . Comparions with the cumulative pregnancy and live birth rate 2.4．Multivariate logistic regression analysis for the cumulative live birth rate. 2.4．Multivariate logistic regression analysis for the cumulative live birth rate. To control possible confounding variables that might have changed over the study period, a multivariate logistic regression analysis was applied to estimate the OR with SE and 95% CI of the CLBR for variable “age” and “number of oocytes retrieved”. As shown in Table 4, after adjustment for relevant confounders including duration of stimulation, duration of infertility, endometrial thickness, insemination type, fertilization rate, and good quality embryo rate, the age and number of oocytes retrieved remained the independent predictive factors (P<0.001). The OR of CLBR dropped from 3.695(95%CI：2.960-4.611) in the aged ≤35 years group to 3.164(95%CI：2.563-3.906) and 2.230 (95%CI：1.785-2.786) in the aged 36-37 years and 38-39 years group respectively. In addition, the fertilization rate, good embryo quality rate and insemination method also have prominent effect on CLBR. Page 8/15 Table4. Multivariate logistic regression analysis for the cumulative live birth rate. Cumulative live birth rate OR SE 95% CI P value Age <0.001 ≤ 35 years of age 3.695 0.113 2.960-4.611 36-37 years of age 3.164 0.108 2.563-3.906 38-39 years of age 2.23 0.114 1.785-2.786 ≥ 40 years of age 1 No.of oocytes retrieved 1.146 0.004 1.138-1.155 <0.001 Fertilization rate, 3.72 0.084 3.720-3.154 <0.001 Good embryo quality rate 2.713 0.056 2.432-3.027 <0.001 Insemination method IVF 1.095 0.036 1.020-1.174 0.012 ICSI 1 3. Discussion Results of this study indicated that the CLBR was proportional to the number of retrieved oocytes, and this steady trend was evident in all female age strata. Notably, female age has a negative effect on the relationship between the number of retrieved oocytes and CLBR, with lower CLBR accompanied by increasing age for a given number of retrieved oocytes. Furthermore, older women exhibited a significantly lower LBR and CLBR in their fresh or frozen embryo transfer cycles compared with younger women. The CLBR, which decreased strikingly from 68.3% in the group ≤ 35 years of age to 34.4% in the group ≥ 36 years of age, can even drop to less than 20% in the group ≥ 40 years of age. Our research is completely consistent with two recent studies investigating the effect of retrieved oocytes numbers as well as frozen embryo transfer on the CLBR. [1, 7] Polyzos et al. confirmed that the CLBR could reach as high as 70% once the number of retrieved oocytes was more than 25. However, according to our analysis, the CLBR closing to 80% is possible as long as the retrieved oocytes numbers were more than 25, which can be explained by the discrepancies that CLBs are mainly from freeze-thaw embryo transfers in our study, accounting for 88.9% in the group ≤ 35 years of age and 87.8% in the group ≥ 36 years of age. Since vitrification was introduced into our reproductive center, 97% of the survival rate in embryo cryopreservation, nearly 60% of the pregnancy rate in freezing-thawing embryo transfer, 35% of the embryo implantation rate and 29% of the live birth rate per embryo were achieved among approximately 6,000 freezing-thawing cycles every year.[15] Page 9/15 Page 9/15 In the past decades, researches have been proposed to explore the optimal number of oocytes need to retrieve during controlled ovarian stimulation although verdicts are controversial. Rubio C and Verberg M F, et al demonstrated that ovarian hyperreactivity may not only increase the risk of chromosomal abnormalities in embryos but reduce the quality of embryos and the rate of high-quality embryos [16, 17] However, other prospective studies concluded that the live birth rate (LBR) is impervious to the ovarian hyperreactivity despite its negative role in fertility rate and the CLBR accompanied by the number of available embryos remains to increase, which can also be reflected in our investigation. 3. Discussion The most significant limitation of this study lies in its retrospective design, which is associated with inherent bias in the research itself that affects our outcomes. In addition, the long study period and clinical operational errors may inevitably be confounding factors influencing experimental results. 3. Discussion Based on our analysis, the fertilization rate and high-quality embryo rate of IVF/ICSI in the group ≤ 35 years of age were significantly lower than those ≥ 36 years of age while the clinical pregnancy rate, embryo implantation rate, live birth rate and CLBR were higher. Consequently, for elderly patients, embryos accumulation by multiple ovulation induction can raise the rate of normal karyotype embryo, which then increase the chance of a live birth accordingly. As is well known, the number of oocytes retrieved and the age of patients make up the most important independent factors affecting the CLBR. Consistent with other studies, we confirmed the independent property of them after controlling confounding factors including “fertilization rate”, “high-quality embryo rate” and “fertilization type”. The correlation analysis between the number of oocytes retrieved and the CLBR in each age stratum revealed that the number of oocytes retrieved had potential predictive value in the CLBR. For patients ≤ 35 years of age, the CLBR can reach more than 60% as long as the number of oocytes retrieved is between 6–8; but patients aged 36–37 years of age needed 9–13 oocytes to achieve a CLBR aboving 60%. What's worse, for patients aged 38–39 years of age, the CLBR of more than 60% can be accomplished only when the oocytes numbers are greater than 14 while the CLBR in patients ≥ 40 years of age can hardly reach 60% even if the number is more than 14. The most significant limitation of this study lies in its retrospective design, which is associated with inherent bias in the research itself that affects our outcomes. In addition, the long study period and clinical operational errors may inevitably be confounding factors influencing experimental results. Although we tried our best to minimize selection bias and control for knowable confounders through strict inclusion criteria, other undiscovered confounders cannot be completely eliminated. Moreover, the protocol of ovulation induction in fresh embryo transfer cycles can hardly be reconcile which varied in different groups, the older group mainly adopted the long scheme while the younger tend to receive the short scheme, leading to the discrepancy in the number of oocytes retrieved, and thus affecting the CLBR. Furthermore, the endometrial preparation pattern of freezing-thawing cycles is also inconsistent in the whole population. However, it is worth clarifying that no matter which preparation has an equivalent effect on the clinical outcome of freeze-thawing embryos [18]. 5.1. Study Population 17931 women undergoing their first IVF/ICSI cycle in the Centre for Reproductive Medicine of the Sir Run Run Shaw Hospital of Zhejiang University between January 2013 and December 2017 were retrospectively analyzed in the database. This study was approved by the institutional review board of our hospital. All methods were performed in accordance with the relevant guidelines and regulations. All patients gave written informed consent. Patients’ eligibility criteria: first IVF/ICSI cycle; live birth obtained from fresh or frozen embryo transplantation; no live birth but used all embryos. Patients’ exclusion criteria: non-first IVF/ICSI cycle; zero oocyte retrieved; cycles with preimplantation genetic test; cycles with medical freezing oocytes; no live birth but still have frozen embryos. Patients’ exclusion criteria: non-first IVF/ICSI cycle; zero oocyte retrieved; cycles with preimplantation genetic test; cycles with medical freezing oocytes; no live birth but still have frozen embryos. For the age of patients, women were categorized in two groups: Group A (to 1. For the age of patients, women were categorized in two groups: Group A (total 14567 cycles): women ≤35 years of age; Group B (total 3364 cycles): women ≥36 years of age. Group B were further divided into into three age strata：36-37 years (1361 cycles), 38-39 years (895 cycles) and ≥40 years(1108 cycles). 2. For the number of retrieved oocytes, women were categorized into four groups: a ≤5, b: 6-9; c: 10-14; d: ≥15. 4. Conclusion The age and number of oocytes retrieved are two independent vital factors affecting cumulative live birth rate. Patients ≤ 35 years old can achieve ideal cumulative live birth rates as long as the number of oocytes obtained is more than 6, indicating that practice of controlling the stimulation dose to ensure the Page 10/15 Page 10/15 live birth rate and reduce the occurrence of OHSS is feasible for them. However, the elderly patients, especially those ≥ 38 years old, need to retrieve more than 14 oocytes to obtain higher live birth rate. Therefore, it is of great necessity to receive additional stimulation cycles and accumulation of embryos live birth rate and reduce the occurrence of OHSS is feasible for them. However, the elderly patients, especially those ≥ 38 years old, need to retrieve more than 14 oocytes to obtain higher live birth rate. Therefore, it is of great necessity to receive additional stimulation cycles and accumulation of embryos. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. 5.2. Treatment protocols (IVF/ICSI-ET) All eligible patients were treated with recombinant and/or urinary gonadotrophins (rFSH/hMG) in ovarian stimulation protocols followed by IVF or ICSI. The choice of ovarian stimulation protocols and gonadotrophin dose depends on age, ovarian reverse function and number of basal follicles. Follicular development was monitored by ultrasound scanning, and final oocyte maturation was induced by injection of human chorionic gonadotrophin (hCG) when at least three follicles of 18 mm in diameter were observed. Oocytes retrieval was performed 36 h after hCG administration under the guidance of ultrasound. Fresh embryo transfer was carried out under ultrasound guidance three or five days after oocyte retrieval. For luteal phase support, intramuscular injection progesterone (60 mg qd) combined with oral progesterone (20 mg qd) was started the day after oocyte retrieval, reduced to intramuscular injection Page 11/15 progesterone (40 mg qd) combined with oral progesterone (20 mg qd) after a positive beta-hCG and until 8 weeks of pregnancy. 5.5. Main clinical outcome measures The cumulative live birth rate (CLBR), which referred to the delivery of a live born in the fresh or in the subsequent frozen-thawed cycles in relation to age category, served as the primary treatment reference indicators. Only the first delivery was considered in the analysis for the CLBR. The clinical pregnancy rate, implantation rate, live birth rate, multiple birth rate and preterm birth rate were also analyzed in the two age-groups. 5.4. Frozen-thawed embryo transfer cycle Embryos were evaluated for morphological thaw survival immediately after warming. Embryos with at least 50 % of their cells remain intact were considered as survival and able to be transferred. Three main types of clinical protocols were used for endometrial preparation: the natural cycle, hormone replacement cycle, or human menopausal gonadotropin (HMG)-stimulated cycles. [15] The type of preparation for the FET cycle was based on the menstrual cycle pattern of the patient. 5.6. Statistical analysis Patients’ baseline characteristics and ovarian stimulation characteristics of the IVF/ICSI treatment were analyzed by age strata. Data analysis were performed using Statistical Package for Social Sciences version 22.0. Continuous data were compared using the Student’s t-test and presented as the mean value ± SD. Categorical data were analyzed with the Chi-Square test, described by number of cases including numerator and denominator, percentages, odds ratio (OR) and confidence interval (CI). Multivariate logistic regression analysis was performed to evaluate the association of the cumulative live birth rate with variable ‘age’ and ‘number of oocytes retrieved’ , presented as an OR with SE and 95% CI. The Hosmer-Lemeshow goodness-of-fit test assessed the goodness-of-fit of the regression models. Statistical significance was defined as a two-tailed p-value of 0.05. 5.3. Selection criteria for embryo transfer and freezing Embryo quality was assessed according to the number of blastomeres and percentage of fragmentation as previously reported. An embryo with 6-12 blastomeres and grades 1 and 2 was defined as good quality [15]. Supernumerary Day 3embryos with at least four blastomeres and ≤35 % fragmentation were selected for cryopreservation with vitrification protocols and transfer. Acknowledgements： Acknowledgements： Page 12/15 This work was supported by grants from Natural Science Program of Zhejiang (LY22H040007). We acknowledge current and former members of the Key Laboratory of Reproductive DysfunctionManagement of Zhejiang Province，Sir Run Run Shaw Hospital, Zhejiang University School of Medicine for technical assistance. Author Contributions HZ and SZ and PW designed the study. HZ and PW drafted the article. PW,WX, CZ and XJ performed the statistical analysis. All authors contributed to the article and approved the submitted version. Competing interests statement: The author(s) declare no competing interests. The author(s) declare no competing interests. Consent to participate: The patients/participants provided their written informed consent to participate in this study. Article type: clinical article Article type: clinical article Data Availability Statement The original contributions presented in the study are included in the article. Further inquiries can be directed to the corresponding authors. Ethics declarations The study was approved by the Ethics Committee in Run Run Shaw Hospital of Zhejiang University. Consent to participate: The study was approved by the Ethics Committee in Run Run Shaw Hospital of Zhejiang University. Consent to participate: References 1. Law Y J, Zhang N, Venetis C A, et al. The number of oocytes associated with maximum cumulative live birth rates per aspiration depends on female age: a population study of 221 221 treatment cycles[J]. Hum Reprod,2019,34(9):1778–1787. 2. Fauser B C, Devroey P, Macklon N S. Multiple birth resulting from ovarian stimulation for subfertility treatment[J]. Lancet,2005,365(9473):1807–1816. 3. Roque M, Lattes K, Serra S, et al. Fresh embryo transfer versus frozen embryo transfer in in vitro fertilization cycles: a systematic review and meta-analysis[J]. Fertil Steril,2013,99(1):156–162. 4. Toftager M, Bogstad J, Lossl K, et al. Cumulative live birth rates after one ART cycle including all subsequent frozen-thaw cycles in 1050 women: secondary outcome of an RCT comparing GnRH- antagonist and GnRH-agonist protocols[J]. Hum Reprod,2017,32(3):556–567. 5. Chambers G M, Paul R C, Harris K, et al. Assisted reproductive technology in Australia and New Zealand: cumulative live birth rates as measures of success[J]. Med J Aust,2017,207(3):114–118. 5. Chambers G M, Paul R C, Harris K, et al. Assisted reproductive technology in Australia and New Zealand: cumulative live birth rates as measures of success[J]. Med J Aust,2017,207(3):114–118. Page 13/15 Page 13/15 6. Drakopoulos P, Errazuriz J, Santos-Ribeiro S, et al. Cumulative live birth rates in in-vitro fertilization[J]. Minerva Ginecol,2019,71(3):207–210. 7. Polyzos N P, Drakopoulos P, Parra J, et al. Cumulative live birth rates according to the number of oocytes retrieved after the first ovarian stimulation for in vitro fertilization/intracytoplasmic sperm injection: a multicenter multinational analysis including approximately 15,000 women[J]. Fertil Steril,2018,110(4):661–670. 7. Polyzos N P, Drakopoulos P, Parra J, et al. Cumulative live birth rates according to the number of oocytes retrieved after the first ovarian stimulation for in vitro fertilization/intracytoplasmic sperm injection: a multicenter multinational analysis including approximately 15,000 women[J]. Fertil Steril,2018,110(4):661–670. 8. Cheng P J, Duan T. China's new two-child policy: maternity care in the new multiparous era[J]. BJOG,2016,123 Suppl 3:7–9. 8. Cheng P J, Duan T. China's new two-child policy: maternity care in the new multiparous era[J]. BJOG,2016,123 Suppl 3:7–9. 9. Zhang M, Bu T, Tian H, et al. Use of Cumulative Live Birth Rate per Total Number of Embryos to Calculate the Success of IVF in Consecutive IVF Cycles in Women Aged >/=35 Years[J]. Biomed Res Int,2019,2019:6159793. 9. Zhang M, Bu T, Tian H, et al. References Use of Cumulative Live Birth Rate per Total Number of Embryos to Calculate the Success of IVF in Consecutive IVF Cycles in Women Aged >/=35 Years[J]. Biomed Res Int,2019,2019:6159793. 10. Mcpherson N O, Zander-Fox D, Vincent A D, et al. Combined advanced parental age has an additive negative effect on live birth rates-data from 4057 first IVF/ICSI cycles[J]. J Assist Reprod Genet,2018,35(2):279–287. 10. Mcpherson N O, Zander-Fox D, Vincent A D, et al. Combined advanced parental age has an additive negative effect on live birth rates-data from 4057 first IVF/ICSI cycles[J]. J Assist Reprod Genet,2018,35(2):279–287. 11. Devesa M, Tur R, Rodriguez I, et al. Cumulative live birth rates and number of oocytes retrieved in women of advanced age. A single centre analysis including 4500 women >/=38 years old[J]. Hum Reprod,2018,33(11):2010–2017. 11. Devesa M, Tur R, Rodriguez I, et al. Cumulative live birth rates and number of oocytes retrieved in women of advanced age. A single centre analysis including 4500 women >/=38 years old[J]. Hum Reprod,2018,33(11):2010–2017. 12. Munne S, Alikani M, Tomkin G, et al. Reprint of: Embryo morphology, developmental rates, and maternal age are correlated with chromosome abnormalities[J]. Fertil Steril,2019,112(4 Suppl1):e71- e80. 12. Munne S, Alikani M, Tomkin G, et al. Reprint of: Embryo morphology, developmental rates, and maternal age are correlated with chromosome abnormalities[J]. Fertil Steril,2019,112(4 Suppl1):e71- e80. 13. Capalbo A, Hoffmann E R, Cimadomo D, et al. Human female meiosis revised: new insights into the mechanisms of chromosome segregation and aneuploidies from advanced genomics and time- lapse imaging[J]. Hum Reprod Update,2017,23(6):706–722. 13. Capalbo A, Hoffmann E R, Cimadomo D, et al. Human female meiosis revised: new insights into the mechanisms of chromosome segregation and aneuploidies from advanced genomics and time- lapse imaging[J]. Hum Reprod Update,2017,23(6):706–722. 14. Munne S, Chen S, Colls P, et al. Maternal age, morphology, development and chromosome abnormalities in over 6000 cleavage-stage embryos[J]. Reprod Biomed Online,2007,14(5):628–634. 14. Munne S, Chen S, Colls P, et al. Maternal age, morphology, development and chromosome abnormalities in over 6000 cleavage-stage embryos[J]. Reprod Biomed Online,2007,14(5):628–634. 15. Zhu H Y, Xue Y M, Yang L Y, et al. Slow freezing should not be totally substituted by vitrification when applied to day 3 embryo cryopreservation: an analysis of 5613 frozen cycles[J]. J Assist Reprod Genet,2015,32(9):1371–1377. 15. Zhu H Y, Xue Y M, Yang L Y, et al. References Slow freezing should not be totally substituted by vitrification when applied to day 3 embryo cryopreservation: an analysis of 5613 frozen cycles[J]. J Assist Reprod Genet,2015,32(9):1371–1377. 16. Rubio C, Mercader A, Alama P, et al. Prospective cohort study in high responder oocyte donors using two hormonal stimulation protocols: impact on embryo aneuploidy and development[J]. Hum Reprod,2010,25(9):2290–2297. 16. Rubio C, Mercader A, Alama P, et al. Prospective cohort study in high responder oocyte donors using two hormonal stimulation protocols: impact on embryo aneuploidy and development[J]. Hum Reprod,2010,25(9):2290–2297. 17. Verberg M F, Eijkemans M J, Macklon N S, et al. The clinical significance of the retrieval of a low number of oocytes following mild ovarian stimulation for IVF: a meta-analysis[J]. Hum Reprod Update,2009,15(1):5–12. 17. Verberg M F, Eijkemans M J, Macklon N S, et al. The clinical significance of the retrieval of a low number of oocytes following mild ovarian stimulation for IVF: a meta-analysis[J]. Hum Reprod Update,2009,15(1):5–12. 18. Drakopoulos P, Blockeel C, Stoop D, et al. Conventional ovarian stimulation and single embryo transfer for IVF/ICSI. How many oocytes do we need to maximize cumulative live birth rates after utilization of all fresh and frozen embryos?[J]. Hum Reprod,2016,31(2):370–376. 18. Drakopoulos P, Blockeel C, Stoop D, et al. Conventional ovarian stimulation and single embryo transfer for IVF/ICSI. How many oocytes do we need to maximize cumulative live birth rates after utilization of all fresh and frozen embryos?[J]. Hum Reprod,2016,31(2):370–376. Page 14/15 Page 14/15 Figures Figure 1 a: The association between CLBR and the number of oocytes b.: The association between CLBR and the age c: The association between CLBR and the number of oocytes in different female age strata Figure 1 a: The association between CLBR and the number of oocytes b.: The association between CLBR and the age c: The association between CLBR and the number of oocytes in different female age strata c: The association between CLBR and the number of oocytes in different female age strata Page 15/15
https://openalex.org/W2757178207	https://www.frontiersin.org/articles/10.3389/fendo.2017.00254/pdf	English	null	Autoimmune Thyroid Diseases in Patients Treated with Alemtuzumab for Multiple Sclerosis: An Example of Selective Anti-TSH-Receptor Immune Response	Frontiers in endocrinology	2,017	cc-by	6,020	Review published: 28 September 2017 doi: 10.3389/fendo.2017.00254 published: 28 September 2017 doi: 10.3389/fendo.2017.00254 Mario Rotondi 1, Martina Molteni 1, Paola Leporati1, Valentina Capelli 2, Michele Marinò3 and Luca Chiovato1* Mario Rotondi 1, Martina Molteni 1, Paola Leporati1, Valentina Capelli 2, Michele Marinò3 and Luca Chiovato1* 1 Unit of Internal Medicine and Endocrinology, Laboratory for Endocrine Disruptors, ICS-Maugeri IRCCS, University of Pavia, Pavia, Italy, 2 Department of Internal Medicine and Therapeutics, and Department of Medical and Surgical Sciences, University of Pavia, Pavia, Italy, 3 Endocrinology Unit I, Department of Clinical and Experimental Medicine, University of Pisa, University Hospital of Pisa, Pisa, Italy Keywords: Graves’ disease, Alemtuzumab, multiple sclerosis, autoimmune thyroid disease, reconstitution syndrome Edited by: Cesidio Giuliani, Università degli Studi “G. d’Annunzio” Chieti – Pescara, Italy Alemtuzumab, a humanized anti-CD52 monoclonal antibody, is approved for the treatment of active relapsing-remitting multiple sclerosis (MS). Alemtuzumab induces a rapid and prolonged depletion of lymphocytes from the circulation, which results in a profound immuno-suppression status followed by an immune reconstitution phase. Secondary to reconstitution autoimmune diseases represent the most common side effect of Alemtuzumab treatment. Among them, Graves’ disease (GD) is the most frequent one with an estimated prevalence ranging from 16.7 to 41.0% of MS patients receiving Alemtuzumab. Thyrotropin (TSH) receptor (R)-reactive B cells are typically observed in GD and eventually present this autoantigen to T-cells, which, in turn, secrete several pro-inflammatory cytokines and chemokines. Given that reconstitution autoimmunity is more frequently characterized by autoantibody-mediated diseases rather than by destructive Th1-mediated disorders, it is not surprising that GD is the most commonly reported side effect of Alemtuzumab treatment in patients with MS. On the other hand, immune reconstitution GD was not observed in a large series of patients with rheumatoid arthritis treated with Alemtuzumab. This negative finding supports the view that patients with MS are intrinsically more at risk for developing Alemtuzumab-related thyroid dysfunctions and in particular of GD. From a clinical point of view, Alemtuzumab-induced GD is characterized by a surprisingly high rate of remission, both spontaneous and after antithyroid drugs, as well as by a spontaneous shift to hypothyroidism, which is supposed to result from a change from stimulating to blocking TSH-receptor antibodies. These immune and clinical peculiarities support the concept that antithyroid drugs should be the first-line treatment in Alemtuzumab- induced Graves’ hyperthyroidism. Reviewed by: Motoyasu Saji, The Ohio State University Columbus, United States Amanda Katherine Huber, University of Michigan, United States Correspondence: Luca Chiovato luca.chiovato@icsmaugeri.it Correspondence: Luca Chiovato luca.chiovato@icsmaugeri.it Specialty section: This article was submitted to Thyroid Endocrinology, a section of the journal Frontiers in Endocrinology Received: 26 July 2017 Accepted: 15 September 2017 Published: 28 September 2017 Received: 26 July 2017 Accepted: 15 September 2017 Published: 28 September 2017 Received: 26 July 2017 Accepted: 15 September 2017 Published: 28 September 2017 MS AND THYROID DISEASES Multiple sclerosis is a human chronic inflammatory disease of the central nervous system supposed to be a Th1/Th17 type cell-mediated autoimmune disorder (16, 17). Studies aimed at evaluating whether there is an increased prevalence of autoim- mune thyroid diseases (AITDs) in patients with MS as compared with healthy controls reported conflicting results. While early studies found an increased prevalence of AITD in patients with MS, more recent surveys reported rates which are consistent with the AITD prevalence in the general population (18–21). According to more recent views, an increased prevalence of AITD would be observed in family members of patients with MS (22). This is a rather intriguing and yet to be a fully elucidated observation (23, 24). The acute immuno-suppressive effect of Alemtuzumab is followed by the homeostatic reconstitution of immune cells. Typically, monocytes and B cells recover first, followed by CD4+ T cells. Changes in lymphocyte subsets result in an increased number of T regulatory (Treg) cells and of memory T and B lymphocytes; an increased production of anti-inflammatory cytokines also occurs (5). These events produce a profound rebal- ance of the immune system (6, 7). Circulating lymphocytes disappear within a few minutes after the administration of Alentuzumab. B cells recover within 3 months and a dominance of mature naïve cells (CD19+ CD23+ CD27−) over the memory B cells occurs. CD4+ T cell counts are restored after 35 months, while CD8+ T cell counts are restored after 20 months. The faster recovery of the latter subset of T cells might be related to the development of autoim- mune diseases (8). For at least 9 months after the administration of Alemtuzumab, most circulating T cells are represented by effector memory CD4+ and CD8+ cells. Baker et al. recently described the kinetics of lymphocyte subset reconstitution after Alemtuzumab (9). After depletion, B cells repopulated much more rapidly than T cells in general and Treg in particular (9). In this scenario, the reconstitution of B cells without adequate regulatory control by T cells may explain the high prevalence of post-Alentuzumab autoimmunity (9, 10). In addition to studies aimed at evaluating the prevalence of AITD in naive patients with MS, the occurrence of AITD, as a side effect of immunomodulatory treatments for MS, was extensively reported (25–27). At present, we know that treat- ment with interferon-β (IFN-β) increases the risk for worsening and/or de novo appearance of both thyroid autoimmunity and dysfunction. Citation: Rotondi M, Molteni M, Leporati P, Capelli V, Marinò M and Chiovato L (2017) Autoimmune Thyroid Diseases in Patients Treated with Alemtuzumab for Multiple Sclerosis: An Example of Selective Anti-TSH-Receptor Immune Response. Front. Endocrinol. 8:254. doi: 10.3389/fendo.2017.00254 September 2017 \| Volume 8 \| Article 254 1 Frontiers in Endocrinology \| www.frontiersin.org Alemtuzumab Therapy and Graves’ Disease Rotondi et al. MS AND THYROID DISEASES On the other hand, thyroid side effects were not observed following glatiramer acetate (GA) therapy, even in large series of patients with MS who were longitudinally fol- lowed for more than 10 years (11, 28, 29).i Specifically designed head-to-head clinical studies demon- strated a similar efficacy of IFN-β and GA, as assessed by their ability to prevent clinical relapses and disease progression in patients with MS (12). However, more effective pharmacologic agents are now available. Among them, Alemtuzumab is cur- rently regarded as an effective second-line treatment in patients with highly active RRMS (1, 2, 13). p y Alentuzumab-induced lymphocytopenia is followed by the homeostatic growth of T cells, which is stimulated by the T cell receptor–self peptide complex. The process results in the appear- ance of an oligoclonal cell population, which tends to autoreactiv- ity. New T cell populations have typical aspects of memory T cells, such as lower dependency to co-stimulation, need for lower antigen doses than naïve cells, and faster secrection of inflam- matory cytokines when re-stimulated (6–8). The above described immune derangements lead to a reduced self-tolerance. In most patients, the proliferation of regulatory lymphocytes is unable to prevent autoimmune deseases, possibly because T cells undergo a faster homeostatic growth, which increases their resistance to regulation (8). Patients who developed autoimmunity after Alentuzumab treatment also show high basal levels of IL-21, a cytokine which increases the growth of auto-reactive T cells. In general, the cytokine expression is skewed to the Th2 profile, in agreement with the high B cell counts (11, 12).f ALEMTUZUMAB AS AN IMMUNOMODULATING DRUG The above described immunomodulating actions of Alem tuzumab are responsible for its favorable effects in patients with RRMS (13), but also explain the high prevalence of Alemtuzumab-induced autoimmunity. The latter event received great concern and clinical trials aimed at evaluating potential preventive measures were designed (CAM-THY) (8, 14, 15). Alemtuzumab is a humanized monoclonal antibody that has been approved for the treatment of active relapsing-remitting (RR) multiple sclerosis (MS) (1, 2). As a main pharmacologic action, Alemtuzumab targets the cell-surface antigen CD52. CD52 is a cell-surface glycoprotein with a still poorly understood function. CD52 is expressed on the surface of more than 95% T and B cells, of monocytes and of some dendritic cells, and, although to a lesser extent, even on natural killer cells and other leukocytes (3). The binding of Alemtuzumab to lymphocytes induces cellular lysis lead- ing to their rapid and prolonged depletion from the circulation (4).hf Frontiers in Endocrinology \| www.frontiersin.org TSH-RECEPTOR (TSH-R) AS A MAJOR AUTOANTIGEN IN GRAVES’ DISEASE Graves’ disease (GD), also referred to as toxic diffuse goiter, is commonly regarded as an autoimmune organ-specific disease (30–32). The presence of extra-thyroid manifestations, such as Graves’ orbitopathy and pretibial myxedema, apparently contradicts this classification, which is, however, justified by the TSH-R being a common antigen shared by the thyroid gland and by extra-thyroid tissues (31). The TSH-R, a G-protein coupled receptor with seven transmembrane-spanning domains and a large extracellular portion, is expressed primarily on the surface of thyroid follicular cells, but it is also present in adipocytes, fibroblast, bone cells, and other sites including the heart (33). TSH-R Antiboides (TRAb) encompass stimulating, blocking, and neutral antibodies. In patiens with GD, TRAb mainly have Innate immunity is not affected, and no clinically relevant infection appears after Alentuzumab treatment. This can be due to the maintenance/growth of memory T cells. September 2017 \| Volume 8 \| Article 254 2 Alemtuzumab Therapy and Graves’ Disease Rotondi et al. thyroid-stimulating activity (TSAb), which results in hyper- thyroidism and goiter formation, TSAb bind only the naturally conformed TSH-R and induce cyclic AMP generation, thyroid cell proliferation, and thyroid hormone synthesis and secretion (34, 35). More rarely, and less functionally dominant, TRAb with thyroid blocking activity have been described in patients with GD (36). those receiving IFN-β-1a. In particular, 34% of patients treated with Alemtuzumab developed thyroid dysfunctions (39% receiv- ing 12 mg and 29% receiving 24 mg) as compared with a 6.5% rate in those treated with IFN-β-1a. As shown in Table 1, in the Alemtuzumab treatment group, GD was the most prevalent condition, being experienced by nearly 23% of patients (42). Hypothyroidism was observed in 7.4% patients and destructive thyroiditis with thyrotoxicosis in 4.2% of patients (42). Immune system abnormalities in GD are represented by TSH-R-reactive B cells, which escape deletion and eventually pre- sent this thyroid autoantigen to T cells. When activated, T cells secrete several pro-inflammatory cytokines and chemokines (37, 38). Hence both B cells and T cells play a central role in perpetuating the autoimmune cascade in GD (11, 37). y y p Some clinical peculiarities of these patients are worth noting. The first episode of thyroid dysfunction was observed starting from the first year after Alemtuzumab administration. Afterward, the episodes’ prevalence progressively increased each year for the first 3 years (from 4.6 to 16.1%) with a subsequent decrease in the following 4 years (from 11.3 to 5.9%). AITDs AND ALEMTUZUMAB Among several autoimmune conditions, which have been reported to occur following Alemtuzumab treatment, the present review will focus on GD, the most frequently observed one. The occurrence of GD in Alemtuzumab-treated patients with secondary progressive MS was first described in 1999 (39). In this early report, a third of MS patients (9/27) receiving the anti-CD52 monoclonal Ab developed GD, with circulat- ing TRAb and hyperthyroidism (39). Besides being the first description, the study by Coles et al. is of great interest, as it clearly shows that GD (and its humoral marker, TRAb) is the most prevalent form of AITD occurring in MS patients treated with Alemtuzumab (39). By contrast, in MS patients, treated with other immunomodulatory therapies (i.e., IFN-β), the most prevalent side effect is represented by euthyroid or hypothyroid autoimmune thyroiditis (26, 27, 40). Thus, a first crucial dif- ference between IFN-β and Alemtuzumab was already evident: they elicited different autoimmune reactions driving the onset of Hashimoto’s thyroiditis and GD, respectively.fi Table 1 \| Thyroid dysfunction during Alemtuzumab treatment in patients with multiple sclerosis (MS). Patients and dysfunctions (%) Patients with thyroid dysfunction in the 216 patients cohort 34 • Graves’ hyperthyroidism • Hypothyroidism • Destructive thyroiditis 22.4 7.4 4.2 Type of dysfunction in the 73 affected patients • Graves’ hyperthyroidism • Hypothyroidism • Destructive thryoiditis • Unknown 65.8 20.5 10.3 1.4 Number of events of thyroid dysfunction between the 102 recorded events • Single • Multiple 70.0 30.0 Type of event among the 102 recorded events • Graves’ hyperthyroidism ⚬ ⚬Overt ⚬ ⚬Subclinical • Hypothyroidism • Destructive thryoiditis • Unknown • Graves’ orbitophaty 58.8 81.7 18.3 29.4 9.8 2.0 6.0 Results of the CAMMS223 study as categorized by Daniels et al. (42). Table 1 \| Thyroid dysfunction during Alemtuzumab treatment in patients with multiple sclerosis (MS). Subsequent studies, investigating the efficacy and safety of Alemtuzumab therapy in patients with RRMS, confirmed GD as the main autoimmune sequela of this immunomodulatory treatment. In 2008, Coles et al. published a phase 2 clinical trial (CAMMS223) in which 334 patients with RRMS were randomized to receive either IFN-β-1a three times/week or annual cycles of Alemtuzumab (either 12 or 24 mg/day) for 3 years (41). TSH-RECEPTOR (TSH-R) AS A MAJOR AUTOANTIGEN IN GRAVES’ DISEASE There was a higher than expected prevalence (52.7%) of patients in whom Graves’ hyperthyroidism (either overt or subclinical) was spontaneously reverted to hypothyroidism (either overt or subclinical) (Table 2). An unusual frequency of patients converting from hyperthyroidism to hypothyroidism and vice versa was also observed. Importantly, the conversion from hyperthyroidism to hypothyroidism was accompanied by the occurrence of TRAb in nearly 77% of patients (Table 3). This observation strongly suggests that the conversion from hyperthyroidism to hypothyroidism was likely due to a shift in the ratio between stimulating and blocking TRAbs. In the routine endocrine practice of sporadic GD, the transition from hyperthyroidism to hypothyroidism is a unusual event, which is mainly observed several years after a successful course of antithyroid drugs and is rarely accompanied by TRAb positiv- ity (43). Taken together, these data indicate that the immune reconstitution occurring after Alemtuzumab treatment is Frontiers in Endocrinology \| www.frontiersin.org AITDs AND ALEMTUZUMAB Patients and TPOAb (%) Patients with postive TPOAb at baseline 8.0 Patients with negative TPOAb at baseline who develop thyroid dysfunction 31.0 Patients with postive TPOAb at baseline who develop thyroid dysfunction 69.0 • Graves’ hyperthyroidism • Hypothyroidism • Destructive thyroiditis • Euthyroidism 31.2 25.0 12.5 31.2 Positive TPOAb at the time of thyroid dysfunction 69.8 • Positive TPOAb at baseline • De novo positive TPOAb • Persistently positive TPOAb 15.1 54.8 30.1 De novo positive TPOAb without thyroid dysfunction 10.5 Data elaborated from Daniels et al. (42). Table 2 \| Treatment and outcome of Graves’ disease developing after Alemtuzumab therapy. (%) Therapy of overt hyperthyroidism in Graves’ disease • Antithyroid drugs alone • Antithyroid drugs + radiometabolic therapy (131-I) • Radiometabolic therapy alone (131-I) • Surgery (total thyroidectomy) 40.1 12.2 6.1 4.0 Outcome of overt hyperthyroidism in Graves’ disease • Spontaneous resolution with: – Euthyroidism – Hypothyroidism 36.7 20.4 16.3 Therapy of subclinical hyperthyroidism • Antithyroid drugs alone 36.4 Outcome of subclinical hyperthyroidism in Graves’ disease 63.6 • Spontaneous resolution • Subclinical hypothyroidism • Overt hypothyroidism • Unknown 18.2 18.2 18.2 9.1 Data elaborated from Daniels et al. (42). Table 5 \| Studies reporting Alemtuzumab-related thyroid dysfunction. Reference No. of patients No. of patients who developed thyroid dysfunctions Thyroid function Coles et al. (39) 27 9/27 (33%) Hyperthyroidism: 9 (33%)a Coles et al. (41) 216a 49/216 (22.7%) Hyperthyroidism: 32 (14.8%) Hypothyroidism: 15 (6.9%) Coles et al. (2) 596 100/596 (16.7%) Hyperthyroidism: 28 (4.7%) Hypothyroidism: 31 (5.2%) Cohen et al. (1) 376 68/376 (18%) Hyperthyroidism: 28 (7%) Hypothyroidism: 18 (5%) Tuohy et al. (45) 87 35/87 (41%) Hyperthyroidism: 22 (25.3%) Hypothyroidism: 12 (13.8%) aOne patient developed hypothyroidism before shifting to hyperthyroidism. le 5 \| Studies reporting Alemtuzumab-related thyroid dysfunction Table 3 \| TSH-receptor antibodies (TRAb) and thyroid dysfunction during Alemtuzumab treatment. Patients and TRAb (%) Positive TRAb at baseline 0.0 De novo positive TRAb 38.0 Thyroid dysfunction and TRAb status • Positive TRAb 70.0 Hyperthyroidism and TRAb status • Positive TRAb 84.7 Graves’ disease and TRAb status • Positive TRAb 100 Hypothyroidism and TRAb status • Positive TRAb 76.7 Data elaborated from Daniels et al. (42). Table 3 \| TSH-receptor antibodies (TRAb) and thyroid dysfunction during Alemtuzumab treatment. treatment (Table 2), was also higher (78%) than what commonly observed in the sporadic form of the disease (32). mainly humoral, being directed to the TSH-R as a major autoantigen. AITDs AND ALEMTUZUMAB Among other (rare) autoimmune side effects, such as trombocytopenic purpura and glomerulonephritis due to autoantibodies binding the glomerular basal membrane, they reported a significantly higher rate of thyroid autoimmunity in patients treated with Alemtuzumab as opposed to those receiv- ing IFN-β-1a (22.7 versus 2.8%, respectively) (41). In 2012, two phase-3 trials also reported the occurrence of mild to moderate thyroid dysfunction in nearly 18% of RRMS patients treated with Alemtuzumab (1, 2).h The early series of patients included in the CAMMS223 study (2) was further investigated with the specific aim of evaluating thyroid side effects of Alemtuzumab therapy (42). Daniels et al. prolonged the surveillance period of these patients up to a median time of 57.3 months and a maximum of 80.6 months (42). They confirmed that thyroid dysfunctions more frequently occurred in patients treated with Alemtuzumab as compared to September 2017 \| Volume 8 \| Article 254 Frontiers in Endocrinology \| www.frontiersin.org 3 Alemtuzumab Therapy and Graves’ Disease Rotondi et al. Table 3 \| TSH-receptor antibodies (TRAb) and thyroid dysfunction during Alemtuzumab treatment. Patients and TRAb (%) Positive TRAb at baseline 0.0 De novo positive TRAb 38.0 Thyroid dysfunction and TRAb status • Positive TRAb 70.0 Hyperthyroidism and TRAb status • Positive TRAb 84.7 Graves’ disease and TRAb status • Positive TRAb 100 Hypothyroidism and TRAb status • Positive TRAb 76.7 Data elaborated from Daniels et al. (42). Table 2 \| Treatment and outcome of Graves’ disease developing after Alemtuzumab therapy. (%) Therapy of overt hyperthyroidism in Graves’ disease • Antithyroid drugs alone • Antithyroid drugs + radiometabolic therapy (131-I) • Radiometabolic therapy alone (131-I) • Surgery (total thyroidectomy) 40.1 12.2 6.1 4.0 Outcome of overt hyperthyroidism in Graves’ disease • Spontaneous resolution with: – Euthyroidism – Hypothyroidism 36.7 20.4 16.3 Therapy of subclinical hyperthyroidism • Antithyroid drugs alone 36.4 Outcome of subclinical hyperthyroidism in Graves’ disease 63.6 • Spontaneous resolution • Subclinical hypothyroidism • Overt hypothyroidism • Unknown 18.2 18.2 18.2 9.1 Data elaborated from Daniels et al. (42). mainly humoral, being directed to the TSH-R as a major autoantigen. Risk factors for the development of Alemtuzumab-induced GD were a family history of thyroid diseases, female sex, younger age, smoking habit, lower administered dose of the monoclonal Ab, and pretretment postivity for thyroid peroxidase (TPO) antibody (Ab) However TPOAb had a minor relevance as a Table 4 \| TPOAb and thyroid dysfunction during Alemtuzumab treatment. Frontiers in Endocrinology \| www.frontiersin.org AITDs AND ALEMTUZUMAB p In a futher observational cohort study, Tuohy et al. (45) re-evaluated 87 patients with RRMS who had been treated with Alemtuzumab in investigator-led studies in Cambridge from 1999 to 2012. This series included 67 patients of the CAMMS224 trial (18) and 20 of the SM3 trial (46). Among the 86 patients who completed the study, 35 (41%) developed a thyroid dys- function, which was diagnosed as Graves’ hyperthyroidism in 22 (63%) and as hypothyroidism with positive tests for TPOAb in 12 (34%) of them. The main limitation of this study is that TRAb were not measured. At the present, large series studies aimed at evaluating the occurrence of thyroid dysfunctions in RRMS patients treated with Alemtuzumab reported a prevalence rang- ing from 16.7 to 41% (Table 5). The remaining published studies on the occurrence of GD in Alemtuzumab-treated patients mainly consist of single case or small series reports, which are summarized in Table 6. In the majority of these reports, patients developing AITD were taking Alemtuzumab for a RRMS (47–50). However, reconstitution GD was also described in patients receiving Alemtuzumab for other clinical conditions. The development of Graves’ hyperthyroidism was described by Risk factors for the development of Alemtuzumab-induced GD were a family history of thyroid diseases, female sex, younger age, smoking habit, lower administered dose of the monoclonal Ab, and pretretment postivity for thyroid peroxidase (TPO) antibody (Ab). However, TPOAb had a minor relevance as a risk factor, due to the low frequency of pre-teratment positive results for this autoimmune marker (Table 4). Daniels et al. found that only 16/206 (8%) patients were positive for TPOAb at baseline. Among them, the prevalence of thyroid dysfunction after Alentuzumab treatment was 69%, a much higher rate than the 31% one observed in patients who were TPOAb negative at baseline (44). However, the majority (85%) of patients developing a thyroid disorder were negative for TPOAb before Alentuzumab treatment. Therefore, regardless of the pretreatment TPOAb sta- tus, patients may develop a thyroid disfunction and should have thyroid function tests performed periodically (41).h The remission rate of Alemtuzumab-induced Graves’ hyperthyroidism, either spontaneous or after antithyroid drug September 2017 \| Volume 8 \| Article 254 4 Alemtuzumab Therapy and Graves’ Disease Rotondi et al. Table 6 \| Case reports of patients developing autoimmune thyroid diseases following Alemtuzumab administration. Reference Case(s) Disease Thyroid function Thyroid Ab Treatment Kirk et al. AITDs AND ALEMTUZUMAB (52) 1 (F) Kidney transplantation Hyperthyroidism TSH-RAb+ Carbimazole TgAb n/a TPOAb n/a Aranha et al. (47) 1 (F) Multiple sclerosis (MS) Hyperthyroidism TSH-RAb+ TgAb+ TPOAb+ Carbimazole, thyroidectomy 2 (F) MS Hyperthyroidism TSH-RAb+ TgAb− TPOAb− Carbimazole, thyroidectomy 3 (F) MS Hyperthyroidism, then hypothyroidism TSH-RAb+ TgAb n/a TPOAb n/a Carbimazole, then levothyroxine 4 (F) MS Hyperthyroidism TSH-RAb+ TgAb+ TPOAb+ Carbimazole Tsourdi et al. (48) 1 (M) MS Hyperthyroidism TSH-RAb+ TgAb+ TPOAb+ Thiamazole 2 (F) MS Hyperthyroidism TSH-RAb+ TgAb+ TPOAb+ Thiamazole, thyroidectomy 3 (F) MS Hyperthyroidism TSH-RAb+ TgAb+ TPOAb– Thiamazole, thyroidectomy 4 (M) MS Hyperthyroidism TSH-RAb+ TgAb− TPOAb+ Thiamazole, thyroidectomy 5 (F) MS Mild hyperthyroidism TSH-RAb+ TgAb+ TPOAb+ No therapy Williams et al. (53) 1 (M) Hematopoietic cell transplantation Hypothyroidism TSH-RAb n/a TgAb+ TPOAb+ Levothyroxine 2 (M) Hematopoietic cell transplantation Hyperthyroidism TSH-RAb− TgAb+ TPOAb+ Methimazole 3 (M) Hematopoietic cell transplantation Hyperthyroidism TSH-RAb− TgAb+ TPOAb+ Methimazole Mahzari et al. (49) 1 (M) MS Hyperthyroidism, then hypothyroidism TSH-RAb n/a TgAb n/a TPOAb+ Propylthiouracil, then levothyroxine 2 (F) MS Hyperthyroidism, then hypothyroidism TSH-RAb n/a TgAb n/a TPOAb+ No therapy, then levothyroxine 3 (F) MS Hyperthyroidism TSH-RAb n/a TgAb n/a TPOAb+ Methimazole 4 (F) MS Mild hyperthyroidism, then hypothyroidism TSH-RAb n/a TgAb n/a TPOAb+ No therapy, then levothyroxine Obermann et al. (50) 1 (M) MS Hyperthyroidism TSH-RAb+ TgAb+ TPOAb+ Carbimazole 2 (M) MS Subclinical hypothyroidism TSH-RAb− TgAb+ TPOAb+ No therapy Table 6 \| Case reports of patients developing autoimmune thyroid diseases following Alemtuzumab administration. September 2017 \| Volume 8 \| Article 254 Frontiers in Endocrinology \| www.frontiersin.org Alemtuzumab Therapy and Graves’ Disease Rotondi et al. GD is the most commonly reported side effects of Alemtuzumab treatment.h Walsh et al. in 11% of patients treated with Alemtuzumab for vasculitis (51). Other reports include (i) a young kidney trans- plant recipient who developed GD 4 years after Alemtuzumab treatment (52) and (ii) three pediatric cases of thyroid autoim- mune diseases in patients receiving Alemtuzumab after hemat- opoietic cell transplantation (53). The reason why, as compared with patients bearing other clini- cal conditions, those with MS carry a higher risk for the develop- ment of GD after Alemtuzumab treatment remains unknown. Genetic factors and/or specific clinical aspects of MS, such as the cytokine/chemokine milieu and/or the RR clinical course, might play a role, but there is still no definite proof at this regard. AUTHOR CONTRIBUTIONS Reconstitution GD may occur during the recovery phase of Alemtuzumab-induced CD52 cells depletion. Because reconsti- tution autoimmunity is more frequently related to autoantibody- mediated diseases rather than to destructive, Th1-mediated disorders (i.e., Hashimoto’s thyroiditis), it is not surprising that LC, MR, MaM, PL, VC, and MiM designed the study and reviewed the literature on thyroid side effects of Alemtuzumab. LC and MR wrote the manuscript. All the authors revised the paper and approved the final edition. AITDs AND ALEMTUZUMAB At variance with all the above described studies, the deve lopment of immune reconstition GD was not observed in a large series of patients with rheumatoid arthritis treated with Alemtuzumab (54). These negative findings support the view that patients with MS bear a higher intrinsic risk for the devel- opment of Alemtuzumab-related thyroid dysfunctions. Further support to the above statement stems from the notion that while Alemtuzumab is increasingly prescribed in chronic lymphocytic leukemia, no case of Graves’ hyperthyroidism has been reported in these patients. From a clinical point of view, there are peculiar aspects of Alemtuzumab-induced GD. First, hyperthyroid patients have an unusualy high rate of spontaneous shift to hypothyroidism. This shift is supposed to result from a change from stimulating to block- ing TRAb. Second, the remission rate of Graves’ hyperthyroidism, both spontaneous and after antithyroid drugs, is unexpectedly high, suggesting a less aggressive disease (14). This observation implies that antithyroid drugs should be the first-line treatment in patients with Alemtuzumab-induced Graves’ hyperthyroidism. REFERENCES primary cultures of thyrocytes: a clue for understanding the different risks of thyroid dysfunctions in patients with multiple sclerosis treated with either of the two drugs. J Neuroimmunol (2011) 234(1–2):161–4. doi:10.1016/j. jneuroim.2011.01.013 f primary cultures of thyrocytes: a clue for understanding the different risks of thyroid dysfunctions in patients with multiple sclerosis treated with either of the two drugs. J Neuroimmunol (2011) 234(1–2):161–4. doi:10.1016/j. jneuroim.2011.01.013 f 1. Cohen JA, Coles AJ, Arnold DL, Confavreux C, Fox EJ, Hartung HP, et al. Alemtuzumab versus interferon beta 1a as first-line treatment for patients with relapsing-remitting multiple sclerosis: a randomised controlled phase 3 trial. Lancet (2012) 380(9856):1819–28. doi:10.1016/S0140-6736(12)61769-3 12. Mikol DD, Barkhof F, Chang P, Coyle PK, Jeffery DR, Schwid SR, et al. Com parison of subcutaneous interferon beta-1a with glatiramer acetate in patients with relapsing multiple sclerosis (the REbif vs Glatiramer Acetate in Relapsing MS Disease [REGARD] study): a multicentre, randomised, parallel, open-label trial. Lancet Neurol (2008) 7(10):903–14. doi:10.1016/ S1474-4422(08)70200-X 2. Coles AJ, Twyman CL, Arnold DL, Cohen JA, Confavreux C, Fox EJ, et al. Alemtuzumab for patients with relapsing multiple sclerosis after disease-modifying therapy: a randomised controlled phase 3 trial. Lancet (2012) 380(9856):1829–39. doi:10.1016/S0140-6736(12)61768-1 3. Rao SP, Sancho J, Campos-Rivera J, Boutin PM, Severy PB, Weeden T, et al. Human peripheral blood mononuclear cells exhibit heterogeneous CD52 expression levels and show differential sensitivity to alemtuzumab mediated cytolysis. PLoS One (2012) 7(6):e39416. doi:10.1371/journal.pone.0039416 3. Rao SP, Sancho J, Campos-Rivera J, Boutin PM, Severy PB, Weeden T, et al. Human peripheral blood mononuclear cells exhibit heterogeneous CD52 expression levels and show differential sensitivity to alemtuzumab mediated cytolysis. PLoS One (2012) 7(6):e39416. doi:10.1371/journal.pone.0039416 i 13. Coles AJ, Fox E, Vladic A, Gazda SK, Brinar V, Selmaj KW, et al. Alemtuzumab more effective than interferon β-1a at 5-year follow-up of CAMMS223 clinical trial. Neurology (2012) 78(14):1069–78. doi:10.1212/ WNL.0b013e31824e8ee7 4. Serafini B, Rosicarelli B, Magliozzi R, Stigliano E, Aloisi F. Detection of ectopic B-cell follicles with germinal centers in the meninges of patients with secondary progressive multiple sclerosis. Brain Pathol (2004) 14(2):164–74. doi:10.1111/j.1750-3639.2004.tb00049.x 14. Weetman AP. Graves’ disease following immune reconstitution or immuno- modulatory treatment: should we manage it any differently? Clin Endocrinol (Oxf) (2014) 80(5):629–32. doi:10.1111/cen.12427 15. ClinicalTrials.gov. Keratinocyte Growth Factor to Prevent Autoimmunity After Alemtuzumab Treatment of Multiple Sclerosis. (2017). NCT01712945. Available from: https://clinicaltrials.gov/ct2/show/NCT01712945 5. Wiendl H, Kieseier B. Multiple sclerosis: reprogramming the immune reper- toire with alemtuzumab in MS. REFERENCES J Immunol (2010) 185(1):763–8. doi:10.4049/jimmunol.1000422 28. Rotondi M, Bergamaschi R, Chiovato L. Disease modifying therapies in multiple sclerosis: could a baseline thyroid check-up drive the therapeutic choice between interferon-β and glatiramer acetate? Mult Scler (2014) 20(14): 1918–9. doi:10.1177/1352458514533387 47. Aranha AA, Amer S, Reda ES, Broadley SA, Davoren PM. Autoimmune thyroid disease in the use of alemtuzumab for multiple sclerosis: a review. Endocr Pract (2013) 19(5):821–8. doi:10.4158/EP13020.RA b 48. Tsourdi E, Gruber M, Rauner M, Blankenburg J, Ziemssen T, Hofbauer LC. Graves’ disease after treatment with alemtuzumab for multiple sclerosis. Hormones (Athens) (2015) 14(1):148–53. doi:10.14310/horm.2002.1501 29. Frisullo G, Calabrese M, Tortorella C, Paolicelli D, Ragonese P, Annovazzi P, et al. Thyroid autoimmunity and dysfunction in multiple sclerosis patients during long-term treatment with interferon beta or glatiramer acetate: an Italian multicenter study. Mult Scler (2014) 20(9):1265–8. doi:10.1177/ 1352458514521311 49. Mahzari M, Arnaout A, Freedman MS. Alemtuzumab induced thyroid disease in multiple sclerosis: a review and approach to management. Can J Neurol Sci (2015) 42(5):284–91. doi:10.1017/cjn.2015.48 30. Leporati P, Groppelli G, Zerbini F, Rotondi M, Chiovato L. Etiopathogenesis of Basedow’s disease. Trends and current aspects. Nuklearmedizin (2015) 54(5): 204–10. doi:10.3413/Nukmed-0739-15-04 50. Obermann M, Ruck T, Pfeuffer S, Baum J, Wiendl H, Meuth SG. Simultaneous early-onset immune thrombocytopenia and autoimmune thyroid disease following alemtuzumab treatment in relapsing-remitting multiple sclerosis. Mult Scler (2016) 22(9):1235–41. doi:10.1177/1352458516638558 31. Marinò M, Latrofa F, Menconi F, Chiovato L, Vitti P. Role of genetic and non-genetic factors in the etiology of Graves’ disease. J Endocrinol Invest (2015) 38(3):283–94. doi:10.1007/s40618-014-0214-2 51. Walsh M, Chaudhry A, Jayne D. Long-term follow-up of relapsing/refractory anti-neutrophil cytoplasm antibody associated vasculitis treated with the lymphocyte depleting antibody alemtuzumab (CAMPATH-1H). Ann Rheum Dis (2008) 67(9):1322–7. doi:10.1136/ard.2007.081661 32. Bartalena L, Chiovato L, Vitti P. Management of hyperthyroidism due to Graves’ disease: frequently asked questions and answers (if any). J Endocrinol Invest (2016) 39(10):1105–14. doi:10.1007/s40618-016-0505-x 52. Kirk AD, Hale DA, Swanson SJ, Mannon RB. Autoimmune thyroid disease after renal transplantation using depletional induction with alemtuzumab. Am J Transplant (2006) 6(5 Pt 1):1084–5. doi:10.1111/j.1600-6143.2006.01258.x 33. Bahn RS, Dutton CM, Natt N, Joba W, Spitzweg C, Heufelder AE. Thyrotropin receptor expression in Graves’ orbital adipose/connective tissues: potential autoantigen in Graves’ ophthalmopathy. J Clin Endocrinol Metab (1998) 83(3):998–1002. doi:10.1210/jcem.83.3.4676 53. Williams KM, Dietzen D, Hassoun AA, Fennoy I, Bhatia M. Autoimmune thyroid disease following alemtuzumab therapy and hematopoietic cell trans- plantation in pediatric patients with sickle cell disease. REFERENCES Nat Rev Neurol (2013) 9(3):125–6. doi:10.1038/ nrneurol.2013.2 h 16. Batocchi AP, Rotondi M, Caggiula M, Frisullo G, Odoardi F, Nociti V, et al. Leptin as a marker of multiple sclerosis activity in patients treated with interferon-beta. J Neuroimmunol (2003) 139(1–2):150–4. doi:10.1016/S0165- 5728(03)00154-1 6. Milo R. Therapeutic strategies targeting B-cells in multiple sclerosis. Auto immun Rev (2016) 15(7):714–8. doi:10.1016/j.autrev.2016.03.006 7. Jones JL, Coles AJ. Mode of action and clinical studies with alemtuzumab. Exp Neurol (2014) 262(Pt A):37–43. doi:10.1016/j.expneurol.2014.04.018 7. Jones JL, Coles AJ. Mode of action and clinical studies with alemtuzumab. Exp Neurol (2014) 262(Pt A):37–43. doi:10.1016/j.expneurol.2014.04.018 17. Fletcher JM, Lalor SJ, Sweeney CM, Tubridy N, Mills KH. T cells in multiple sclerosis and experimental autoimmune encephalomyelitis. Clin Exp Immu nol (2010) 162(1):1–11. doi:10.1111/j.1365-2249.2010.04143.x 8. Weetman A. Immune reconstitution syndrome and the thyroid. Best Pract Res Clin Endocrinol Metab (2009) 23(6):693–702. doi:10.1016/j.beem.2009.07.003 9. Baker D, Herrod SS, Alvarez-Gonzalez C, Giovannoni G, Schmierer K. Interpreting lymphocyte reconstitution data from the pivotal phase 3 trials of alemtuzumab. JAMA Neurol (2017) 74(8):961–9. doi:10.1001/jamaneurol. 2017.0676 18. Durelli L, Oggero A, Verdun E, Isoardo GL, Barbero P, Bergamasco B, et al. Thyroid function and anti-thyroid antibodies in MS patients screened for interferon treatment. A multicenter study. J Neurol Sci (2001) 193(1):17–22. doi:10.1016/S0022-510X(01)00637-2 10. Steinman L. Induction of new autoimmune diseases after alemtuzumab therapy for multiple sclerosis learning from adversity. JAMA Neurol (2017) 74(8):907–8. doi:10.1001/jamaneurol.2017.0325 19. Marrie RA, Yu BN, Leung S, Elliott L, Warren S, Wolfson C, et al. The incidence and prevalence of fibromyalgia are higher in multiple sclerosis than the general population: a population-based study. Mult Scler Relat Disord (2012) 1(4):162–7. doi:10.1016/j.msard.2012. 06.001 11. Rotondi M, Stufano F, Lagonigro MS, La Manna L, Zerbini F, Ghilotti S, et al. Interferon-β but not glatiramer acetate stimulates CXCL10 secretion in September 2017 \| Volume 8 \| Article 254 Frontiers in Endocrinology \| www.frontiersin.org 6 Alemtuzumab Therapy and Graves’ Disease Rotondi et al. 20. Niederwieser G, Buchinger W, Bonelli RM, Berghold A, Reisecker F, Költringer P, et al. Prevalence of autoimmune thyroiditis and non-immune thyroid disease in multiple sclerosis. J Neurol (2003) 250(6):672–5. doi:10.1007/ s00415-003-1053-9 39. Coles AJ, Wing M, Smith S, Coraddu F, Greer S, Taylor C, et al. Pulsed monoclonal antibody treatment and autoimmune thyroid disease in multiple sclerosis. Lancet (1999) 354(9191):1691–5. doi:10.1016/S0140-6736(99) 02429-0 21. Sloka JS, Phillips PW, Stefanelli M, Joyce C. Co-occurrence of autoimmune thyroid disease in a multiple sclerosis cohort. REFERENCES J Autoimmune Dis (2005) 2:9. doi:10.1186/1740-2557-2-9 40. Rotondi M, Mazziotti G, Biondi B, Manganella G, Del Buono AD, Montella P, et al. Long-term treatment with interferon-beta therapy for multiple sclerosis and occurrence of Graves’ disease. J Endocrinol Invest (2000) 23(5):321–4. doi:10.1007/BF03343730 22. Broadley SA, Deans J, Sawcer SJ, Clayton D, Compston DA. Autoimmune dis- ease in first-degree relatives of patients with multiple sclerosis. A UK survey. Brain (2000) 123(Pt 6):1102–11. doi:10.1093/brain/123.6.1102 41. Coles AJ, Compston DA, Selmaj KW, Lake SL, Moran S, Margolin DH, et al. Alemtuzumab vs. interferon beta-1a in early multiple sclerosis. N Engl J Med (2008) 359(17):1786–801. doi:10.1056/NEJMoa0802670 23. Jenkins RC, Weetman AP. Disease associations with autoimmune thyroid disease. Thyroid (2002) 12(11):977–88. doi:10.1089/105072502320908312 42. Daniels GH, Vladic A, Brinar V, Zavalishin I, Valente W, Oyuela P, et al. Alemtuzumab-related thyroid dysfunction in a phase 2 trial of patients with relapsing-remitting multiple sclerosis. J Clin Endocrinol Metab (2014) 99(1):80–9. doi:10.1210/jc.2013-2201 h 24. Weetman AP. Diseases associated with thyroid autoimmunity: explanations for the expanding spectrum. Clin Endocrinol (Oxf) (2011) 74(4):411–8. doi:10.1111/j.1365-2265.2010.03855.x 25. Caraccio N, Dardano A, Manfredonia F, Manca L, Pasquali L, Iudice A, et al.Long-term follow-up of 106 multiple sclerosis patients undergoing interferon-beta 1a or 1b therapy: predictive factors of thyroid disease develop- ment and duration. J Clin Endocrinol Metab (2005) 90(7):4133–7. doi:10.1210/ jc.2004-2326 43. Wood LC, Ingbar SH. Hypothyroidism as a late sequela in patient with Graves’ disease treated with antithyroid agents. J Clin Invest (1979) 64(5):1429–36. doi:10.1172/JCI109601 44. Topliss DJ. Clinical update in aspects of the management of autoimmune thyroid diseases. Endocrinol Metab (Seoul) (2016) 31(4):493–9. doi:10.3803/ EnM.2016.31.4.493 26. Monzani F, Caraccio N, Meucci G, Lombardo F, Moscato G, Casolaro A, et al. Effect of 1-year treatment with interferon-beta1b on thyroid function and autoimmunity in patients with multiple sclerosis. Eur J Endocrinol (1999) 141(4):325–31. doi:10.1530/eje.0.1410325 45. Tuohy O, Costelloe L, Hill-Cawthorne G, Bjornson I, Harding K, Robertson N, et al. Alemtuzumab treatment of multiple sclerosis: long-term safety and efficacy. J Neurol Neurosurg Psychiatry (2015) 86(2):208–15. doi:10.1136/ jnnp-2014-307721 i 27. Rotondi M, Oliviero A, Profice P, Mone CM, Biondi B, Del Buono A, et al. Occurrence of thyroid autoimmunity and dysfunction throughout a nine- month follow-up in patients undergoing interferon-beta therapy for multiple sclerosis. J Endocrinol Invest (1998) 21(11):748–52. doi:10.1007/BF03348040 46. Somerfield J, Hill-Cawthorne GA, Lin A, Zandi MS, McCarthy C, Jones JL, et al. A novel strategy to reduce the immunogenicity of biological therapies. REFERENCES Pediatr Blood Cancer (2014) 61(12):2307–9. doi:10.1002/pbc.25102 34. Morshed SA, Davies TF. Graves’ disease mechanisms: the role of stimulating, blocking, and cleavage region TSH receptor antibodies. Horm Metab Res (2015) 47(10):727–34. doi:10.1055/s-0035-1559633 54. Lorenzi AR, Clarke AM, Wooldridge T, Waldmann H, Hale G, Symmons D, et al. Morbidity and mortality in rheumatoid arthritis patients with prolonged therapy-induced lymphopenia: twelve-year outcomes. Arthritis Rheum (2008) 58(2):370–5. doi:10.1002/art.23122 35. Inaba H, De Groot LJ, Akamizu T. Thyrotropin receptor epitope and human leukocyte antigen in Graves’ disease. Front Endocrinol (2016) 7:120. doi:10.3389/fendo.2016.00120 36. Chiovato L, Fiore E, Vitti P, Rocchi R, Rago T, Dokic D, et al. Outcome of thyroid function in Graves’ patients treated with radioiodine: role of thyroid- stimulating and thyrotropin-blocking antibodies and of radioiodine-induced thyroid damage. J Clin Endocrinol Metab (1998) 83(1):40–6. doi:10.1210/jcem. 83.1.4492 Conflict of Interest Statement: The authors declare that the research was con- ducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. 37. Rotondi M, Chiovato L, Romagnani S, Serio M, Romagnani P. Role of chemo- kines in endocrine autoimmune diseases. Endocr Rev (2007) 28(5):492–520. doi:10.1210/er.2006-0044 Copyright © 2017 Rotondi, Molteni, Leporati, Capelli, Marinò and Chiovato. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. 38. Rotondi M, Chiovato L. The chemokine system as a therapeutic target in autoimmune thyroid diseases: a focus on the interferon-γ inducible chemo kines and their receptor. Curr Pharm Des (2011) 17(29):3202–16. doi:10.2174/ 138161211798157559 September 2017 \| Volume 8 \| Article 254 Frontiers in Endocrinology \| www.frontiersin.org 7
https://openalex.org/W4387514859	http://www.ijml.org/vol13/IJML-V13N4-1147.pdf	English	null	Detection of DDoS Attacks Using SHAP-Based Feature Reduction	International journal of machine learning	2,023	cc-by	7,304	International Journal of Machine Learning, Vol. 13, No. 4, October 2023 International Journal of Machine Learning, Vol. 13, No. 4, October 2023 C Cynthia and Gopal Krishna Kamath are with the Department of Electrical and Electronics Engineering at BITS-Pilani Hyderabad Campus, Hyderabad, Telangana, India. Email: Manuscript received February 10, 2023; revised March 13, 2023; accepted April 8, 2023. ( ) Correspondence: p20210415@hyderabad.bits-pilani.ac.in (C.C.) gopal.kamath@hyderabad.bits-pilani.ac.in (G.K.K.) Debayani Ghosh is with the Department of Electronics and Communication Engineering at Thapar Institute of Engineering and Technology, Patiala, Punjab, India. Email: debayani.ghosh@thapar.edu (D.G.) Datagram Protocol (UDP). Datagram Protocol (UDP). Abstract—Machine learning techniques are widely used to protect cyberspace against malicious attacks. In this paper, we propose a machine learning-based intrusion detection system to alleviate Distributed Denial-of-Service (DDoS) attacks, which is one of the most prevalent attacks that disrupt the normal traffic of the targeted network. The model prediction is interpreted using the SHapley Additive exPlanations (SHAP) technique, which also provides the most essential features with the highest Shapley values. For the proposed model, the CICIDS2017 dataset from Kaggle is used for training the classification algorithms. The top features selected by the SHAP technique are used for training a Conditional Tabular Generative Adversarial Networks (CTGAN) for synthetic data generation. The CTGAN-generated data are then used to train prediction models such as Support Vector Classifier (SVC), Random Forest (RF), and Naïve Bayes (NB). The performance of the model is characterized using a confusion matrix. The experiment results prove that the attack detection rate is significantly improved after applying the SHAP feature selection technique. Machine Learning is the burgeoning technology used for data analysis and making predictions without human intervention. It is the task of imparting knowledge and intelligence to the machine to find insightful information from the data. There are a few steps involved in training the machine before deploying it to an environment: Data collection, data pre-processing (feature reduction/feature selection), choosing a model, training the model, model evaluation, parameter tuning, and making predictions. Machine learning techniques are used in various fields, and one among them is cyber-security. After training the model with adequate data, it can be used to detect network traffic abnormalities and prevent various cyber-attacks. For instance, to predict the spikes in network usage, the authors in [6] have used machine learning techniques. It helps to improve cyber-security by training the machine learning model with different cyber-attacks. Researchers have identified different machine learning algorithms for classifying and predicting cyber-attacks. To start with the data pre-processing, the following steps are involved: encoding, standardization, and finally, dimensionality reduction using Principle Component Analysis (PCA) [7]. In the paper [8], the researchers have used feature transformation-based dimensionality reduction for feature clustering using a Gaussian traffic attribute pattern. After data pre-processing, the classification algorithms like Random Forest (RF), XGBoost, and Keras Sequential algorithms explain the results of any machine learning model model [9]. Index Terms—DDoS, SHAP, IDS, machine learning, CTGAN Manuscript received February 10, 2023; revised March 13, 2023; accepted April 8, 2023. C Cynthia and Gopal Krishna Kamath are with the Department of Electrical and Electronics Engineering at BITS-Pilani Hyderabad Campus, Hyderabad, Telangana, India. Email: gopal.kamath@hyderabad.bits-pilani.ac.in (G.K.K.) Debayani Ghosh is with the Department of Electronics and Communication Engineering at Thapar Institute of Engineering and Technology, Patiala, Punjab, India. Email: debayani.ghosh@thapar.edu (D.G.) C d 20210415@h d b d bit il i i (C C ) Detection of DDoS Attacks Using SHAP-Based Feature Reduction C Cynthia, Debayani Ghosh, and Gopal Krishna Kamath I. INTRODUCTION The selected features can be used to train any generative model, which will then be used to generate synthetic data. 2) 2) Synthetic Data Generation - Features with significant importance are selected by SHAP and given as input to Conditional Tabular Generative Adversarial Networks (CTGAN) for synthetic data generation. For better prediction of DDoS attacks, we use a reduced number of features. We can infer that the model can generate the data with the most significant features. The output of this phase is the synthetic data with the same distribution as the original data, which is very efficient in predicting the abnormality of the network. The final stage is a prediction which uses this synthetic data for classification. 2) Synthetic Data Generation - Features with significant importance are selected by SHAP and given as input to Conditional Tabular Generative Adversarial Networks (CTGAN) for synthetic data generation. For better prediction of DDoS attacks, we use a reduced number of features. We can infer that the model can generate the data with the most significant features. The output of this phase is the synthetic data with the same distribution as the original data, which is very efficient in predicting the abnormality of the network. The final stage is a prediction which uses this synthetic data for classification. Generative adversarial networks (GANs) are generative machine learning modeling techniques that learn from the original data to generate plausible synthetic data, with the same distribution as that of the natural data. When natural data pertaining to the requisite process is insufficient, we can use this model to generate synthetic data, which would aid in better analysis of the data. GANs operate by training both the Generator and the Discriminator, which are its main components, to generate synthetic data. Further, two different data are used for training the networks: one is real, and the other is noise. The Discriminator should learn the real data, and the Generator should process the noise. The main objective is that the Discriminator should classify the real from fake data, whereas the Generator tries to fool the discriminator [22]. This, in turn, forms a feedback loop, thereby finally obtaining the synthetic or adversarial data via the Generator. In GAN, the primary task is to identify the distribution between the real and synthetic data; if it does not overlap, it results in a vanishing gradient problem. I. INTRODUCTION 4, October 2023 The following provides a summary of the crucial steps involved in our work: The following provides a summary of the crucial steps involved in our work: For these machine learning and deep learning models to be deployed successfully, the network traffic data must be appropriately captured, and important features are extracted. These features assist in classifying whether the network is safe or under attack. The most impactful features are sufficient in analyzing the network, thereby also resulting in feature reduction. Feature selection or feature extraction techniques can aid in identifying the most influencing attributes. 1) Feature Selection - There are two conventional methods for reducing the features: Feature Extraction and Feature Selection. Feature Extraction is the process of reducing the number of features without affecting the original information. The redundant data are removed using different combinations and transformations of the original dataset. In contrast, feature selection ranks features' importance and helps discard features with least importance. It better explains how our model performs accurate predictions with those selected features. We use SHAP for feature selection because of its prominent properties like local accuracy, missingness, and consistency. It assures a fair distribution of contribution among each of the features. SHAP gives feature importance at a global level by adding the absolute value of the SHAP for each data point. This provides us with the flexibility to select features with high importance for further processing. These features assist in classifying whether the network is safe or under attack. The most impactful features are sufficient in analyzing the network, thereby also resulting in feature reduction. Feature selection or feature extraction techniques can aid in identifying the most influencing attributes. SHapley Additive exPlanations (SHAP) [17] is one of the feature selection methods which provides feature importance. It is a coalition game-theoretic method to distribute the defray among the features justly. The SHAP explanation model gives the significance of the features, which supports interpretation and accuracy for model prediction [18]. Integration of machine learning methods and SHAP gives a better prediction using fewer number of features [19]. The dimensionality reduction techniques convert the high dimensional dataset to a low dimensional dataset by preserving the structures [20]. The operationalizations of SHAP values can be studied using different axiomatic approaches, and there are various techniques such as Baseline SHAP, Integrated Gradients, and Conditional Expectation Shapley [21]. I. INTRODUCTION The Internet is becoming increasingly pervasive since it is the primary mode for data sharing. This has led to its usage increasing rapidly. However, the sharing of files through the Internet is vulnerable to various malicious attacks. An organization or a disgruntled individual could create a system to perpetrate these cyber-attacks. Owing to such ease of perpetrating cyber-attacks, there are several cyber-attacks in existence today, including Malware, Phishing, Man-in-the-Middle attacks, Denial of Service (DoS) attacks, Distributed Denial of Service (DDoS) attacks and SQL Injection, to name a few [1, 2]. The DoS attack is exclusively meant to shut down a system or a network, thereby making it unavailable for legitimate users [3–5]. The DDoS attack is a form of DoS attack wherein different sources target a specific network/server to deprive legitimate users its access. Primarily, the DDoS attack is carried out at the application layer of a network protocol stack; however, certain types of DDoS attacks can also utilise transport layer protocols such as the Transmission Control Protocol (TCP) or the User There are a few other machine learning algorithms such as K_Nearest_Neighbours (KNN), Support Vector Machine (SVM), Naïve Bayes (NB), Decision Tree (DT), Random Forest (RF), and Logistic Regression (LR) algorithms used for the Intrusion Detection Systems (IDS) for IoT devices [10–12]. Deep learning is derived from machine learning, and it consists of different levels of algorithms which are based on very complicated neural networks that mimic the human brain. Deep Learning approaches such as Convolution Neural Networks (CNN), Recurrent Neural Networks (RNN), and Fully Connected Layers can also be used to extract features and detect DDoS attacks [13]. In [14], hybrid detection-based methods, Stacked Auto Encoder and CNN, and feature selection, are used to detect DDoS attacks better. In [15], the authors find that the network intrusion detection systems use Sparse Auto-Encoders and Auto-Encoders modules along with GINI feature selection. DDoS attacks in Software Defined Networks can be detected using other feature selection methods, such as Information Gain (IG) and RF to analyze the most relevant features of these attacks [16]. *Correspondence: p20210415@hyderabad.bits-pilani.ac.in (C.C.) 173 doi: 10.18178/ijml.2023.13.4.1147 International Journal of Machine Learning, Vol. 13, No. I. INTRODUCTION Wasserstein Generative Adversarial Networks (WGAN) overcomes this major issue by replacing the discriminator with a critic and calculating the loss function based on the difference between real and synthetic data. This Wasserstein loss function is based on the Earth Mover's Distance. The WGAN can be used to detect the mutation of attacks which is known as a polymorphic attack [23]. 3) 3) Prediction - Our model's final phase predicts whether the network is benign or under a DDoS attack. The trained model has been saved and used for further classifications. We have used three different classifiers; namely, SVC, , and NB. The remnant part of this paper has been organized as follows. Section II is an elaborate discussion of the related works. Section III elucidates the proposed methodology. In addition, the conduction of experiments with the existing CICIDS2017 dataset and the synthetic data generated by CTGAN. The later part includes the performance comparison with the current baselines in Section IV. Finally, the broad conclusion gives the proper scope for future research in Section V. III. PROPOSED FRAMEWORK In this section, we provide a detailed description of our proposed model. The steps followed to detect a DDoS attack are shown in Fig. 1. Captured network traffic data contains several features such as time stamp, IP address of the source and IP address of the destination, among others. These features can be used to identify abnormality in the network. DDoS attacks are among the cyber-attacks that make the server busy with the bombardment of simultaneous fake requests. The proposed model detects DDoS attacks with a reduced number of features. Sambangi et al. [8] use feature transformation based dimensionality reduction to detect low- and high-rate network attacks. They proposed a model to find the traffic similarity function between two networks to classify and detect the attacks. Our work proposes a DDoS attack detection framework that includes SHAP for Feature Selection and CTGAN model for synthetic data generation. The SHAP provides the most significant features that help the prediction. The remaining features are discarded. These features are used to train a CTGAN, which will then generate synthetic data. This synthetic data is used to obtain better performance for predictions of DDoS attacks via feedback of CTGAN. Finally, the classifiers are trained with the labeled original data, and then the model can be used for testing the synthetic data. Thus, the prediction accuracy has been improved with the reduced features. The following sub-sections provide comprehensive explanations of each block in our proposed model. Chen et al. [22] proposed an attack-agnostic defense model against poisoning attacks. A poisoning attack is an attack on the training data that leads to misclassification. The learning of data augmentation on the models and the predictions are shown to be better using GAN. This paper proposed an architecture that constitutes Synthetic Data Generation, Mimic Model Construction, and Poisoned Data Recognition. The training of the model using De-Pois to detect four different attacks; the attacks are Targeted Clean Label Poisoning Attack (TCL- Attack), Poisoning Attack with GAN (PGAN-attack), Label Flipping Attack (LF-attack), and Regression Attack (R-attack). Fig. 1. Proposed model. Chauhan et al. [23] proposed a model for detecting polymorphic attacks. Polymorphic malware is an attack that consistently changes its identifiable feature to evade the detection system. Since the attack feature profile continuously changes, the defense system should adopt Incremental Learning. This paper focuses on generating DDoS attacks using adversarial network techniques such as WGAN. II. RELATED WORK In this section, we briefly review pertinent research findings on different cyber-attacks in which the researchers used various machine learning algorithms. In the context of network security, the data is typically in the form of tables, as opposed to traditional time-series data in signal processing fields. Conditional Tabular Generative Adversarial Networks (CTGAN) is used to generate the tabular data without any vanishing gradient problem [24]. The CTGAN can effectively process both categorical and continuous data. The statistical properties between real and synthetic data are measured quantitatively using WGAN loss with gradient penalty [25]. Ismail et al. [2] used the Australian Centre for Cyber Security (ACCS) dataset containing DDoS attacks features. Therein, they used machine learning approaches with data pre-processing, including standardization and normalization. The researchers used RF and XGBoost Classifier algorithms and the evaluation of the result using a confusion matrix. Parvinder Singh Saini et al. [6] focus on the detection of DDoS attacks by the machine learning tool WEKA. Validation of this approach on various types of attacks like 174 International Journal of Machine Learning, Vol. 13, No. 4, October 2023 Normal, UDP-Flood, Smurf, SID DOS, and HTTP-Flood DDoS attacks. The J48 classifier is shown to outperform other classifying algorithms like MLP, RF, and NB. Accuracy is the metric for identifying the best performance of different algorithms. Normal, UDP-Flood, Smurf, SID DOS, and HTTP-Flood DDoS attacks. The J48 classifier is shown to outperform other classifying algorithms like MLP, RF, and NB. Accuracy is the metric for identifying the best performance of different algorithms. and RF. Xu et al. [24] designed a model for tabular data using Conditional GAN, which outperforms the Bayesian methods on most real datasets. Zilong Zhao et al. [25] have designed a model for encoding mixed categorical and continuous variables along with the missing values. The metrics used for measuring statistical similarity between the real and synthetic data are Jensen-Shannon divergence, Wasserstein distance, and Difference in pair-wise correlation. Fryer et al. [17] explain the SHapley Additive exPlanations (SHAP) and SHapley Additive Global importance (SAGE). They elaborate on the feature selection using SHAP. Wilson E.Marcilio-Jr and Danilo M.Eler. [20] proposed a method for dimensionality reduction using SHAP. Meenakshi et al. [13] proposed a model for DDoS attack detection, and the researchers used deep learning approaches like Recurrent Neural Networks and Convolutional Neural Networks. Lundberg and Lee [18] proposed a unified approach for interpreting the prediction for complex models using SHAP. The model has improved the computational performance and gives better consistency. Benedek Rozem-berczki et al. [19] designed a framework for the explainable machine-learning model using SHAP. Sudugala et al. [12] experimented on three different datasets: CIS-DoS, CISIDS2017, and CSE-CIC-IDS2018. The authors used four machine learning algorithms: SVM, NB, Simple Linear Regression, and DT to detect DDoS attacks. Alenezi and Ludwig [9] proposed different classification models to detect various types of cyber-attacks by using ensemble techniques. Researchers worked on two other cyber-security datasets: malicious URLs and Android malware. Explainable AI uses SHAP to identify the most significant features. TreeShap, KernelShap, and DeepShap are three SHAP methods used for extracting better feature contributions. The Machine Learning algorithms used for further analysis are Random Forest Classifier (RFC), XGBoost Classification, and the Keras Sequential algorithm. The criteria used by RFC is Gini which measures the quality of a split. The Random Forest with the TreeShap Explanation and XGBoost classification with the KernelShap Explanation classifies the attacks from the normal traffic. III. PROPOSED FRAMEWORK SHAP identifies the function feature of an attack. There are three IDS techniques to identify the various attacks: Signature-Based Detection (Knowledge-Based), Anomaly Based Detection (Behaviour Based), and Stateful Protocol Analysis (Specification Based). The classifiers for the Signature - Based detection systems are SVM, NB, KNN, DT, A. SHAP for Feature Selection ∑ where υ(s) is function matches every coalition where υ(s) is function matches every coalition where υ(s) is function matches every coalition s if ∑ then C(F) is lossless among the features elseif = 0 then Feature i has no contribution to the model elseif = then Two features has equal role on the model elseif (C,K) = (C) + (K) then Two subgames (C,i1),(K,i1) have equal contributions with i1 end if end for end for if ∑ then C(F) is lossless among the features Two features has equal role on the model elseif (C,K) = (C) + (K) then Two subgames (C i1) (K i1) have equal SHAP assigns an importance value to each feature that represents the effect on the model prediction of including that feature. The model has been trained with the presence and the absence of that feature . Predictions from the two models are compared on the current input , where is the values of input features in the set . The additive feature attribution methods are explained in the later part of this section [18]. contributions with i1 end if end for end for contributions with i1 contributions with i1 end if end for end for Let be the original prediction model and be the explanation model. The prediction based on every single input focuses on local methods. ∑ [ ] (3) (3) ∑ (1) (1) B. Synthetic Data Generation Using CTGAN B. Synthetic Data Generation Using CTGAN where is the number of input features, and . Conditional Tabular GAN is used for synthetic data generation with the selected number of features. The architecture of the model is shown in Fig. 2. CTGAN is a GAN-based method to model tabular data distribution and mode-specific normalization to account for the non-Gaussian and multimodal nature of distribution. Additional information, such as class labels, can be added for training the model. CTGANs model is trained by a zero-sum minimax game, where the critic (known as discriminator for regular GANs) tries to maximize the objective, and the generator tries to minimize it. The generator in a CTGAN is trained with a vector sampled from a standard multivariate normal distribution. The critic eventually obtains a deterministic transformation that maps the standard multivariate normal distribution to the data distribution. Local Accuracy is given by Lundberg as in (1). The explanation model matches the original model when , where represents the model output. (2) (2) Missingness constraints are represented as in (2), where 0 has no attributed impact. Furthermore, the next property is consistency. Let ( ) and denote setting the coordinate to zero, . For any two models and , if for all inputs , then . Fig. 2. Architecture of CTGAN. Fig. 2. Architecture of CTGAN. The only possible explanation model that satisfies all three properties (missingness, consistency, and local accuracy), is given by [18] as represented in (3). Algorithm 1 Feature Selection INPUT : Labelled dataset by x with i features. Consider a cooperative game with M features (numbered from 1 to M), and let F be the set of features. Assume S ⊆ F represents a coalition. FEATURE SELECTION : Set of Features F ={1,....,M} COALITION, S ⊆ F, where F is grand coalition Train and Test Feature Vector Sets are where and Assumptions: Let P be the permutation and C be the A. SHAP for Feature Selection Shapley value-based explanations of machine learning models are used for finding the contribution of each feature to the model prediction and providing the feature importance for the entire feature in the dataset. Machine Learning and Game theory work together in such a way that the model's input features are similar to players in a game while the model function is similar to the rules of the game. Transferable Utility is one of the assumptions in many cooperative games, where the individual players do not 175 International Journal of Machine Learning, Vol. 13, No. 4, October 2023 characteristic function, C : 2F → receive any payoff. Instead, the coalitions get standard payoffs. We integrate the other features using conditional expected value formulation to evaluate an existing model when only a subset of features is a part of the model. characteristic function, C : 2F → characteristic function, C : 2F → for i = {1,....,n} Perform permutations for all the features P(i) for each P(i) compute coalition ∑ where υ(s) is function matches every coalition s if ∑ then C(F) is lossless among the features elseif = 0 then Feature i has no contribution to the model elseif = then Two features has equal role on the model elseif (C,K) = (C) + (K) then Two subgames (C,i1),(K,i1) have equal contributions with i1 end if end for end for for i = {1,....,n} Perform permutations for all the features P(i) for each P(i) compute coalition for each P(i) compute coalition The partial dependence plot is a global method showing the marginal effect of one or two features on the predicted outcome. It portrays the relationship between a target and a feature and helps verify whether they are dependent linearly. One of the main properties of SHAP values is that they sum up the difference between the game outcome in presence of all features and the game outcome when no features are present. In our machine learning model, this means that SHAP values of all the input features will always sum up the difference between the expected output and the predicted output of the current model. C. Predictive Classifiers Generator - Minimize the objective function to fool the Critic. STEP 1: Generate mask vector , for 1 ≤ j ≤ n STEP 2: Create a condition vector Condnj, for 1 ≤ j ≤ m STEP 3: Generate fake data ̂ STEP 4: Extract real data sample , for STEP 5: Cost Function = ∑ ( \| ∑ ( ̂ \| STEP 6: Gradient Penalty = ∑ ( ̂ \| ) The SHAP summary plot combines the feature importance and its effects on the model prediction. The SHAP values explain each feature’s contribution toward the model’s predicted output. Algorithm 2 Synthetic Data Generation CTGAN INPUT : Labelled data with SHAP selected features Assumptions : Dtrain for Training Data, G for Conditional Generator, C for Critic, M for Mask Vector, Condnj for Condition Vector Critic - Maximize the objective function to classify real and fake. Generator - Minimize the objective function to fool the Critic. STEP 1: Generate mask vector , for 1 ≤ j ≤ n STEP 2: Create a condition vector Condnj, for 1 ≤ j ≤ m STEP 3: Generate fake data ̂ STEP 4: Extract real data sample , for STEP 5: Cost Function = ∑ ( \| ∑ ( ̂ \| STEP 6: Gradient Penalty = ∑ ( ̂ \| ) The feature importance measures are classified as local feature importance and global feature importance. Individual feature contribution to the predicted output for a specific data point is known as Local feature importance. In contrast, the average SHAP value for each feature across entire data points is given by Global feature importance. In the plot, the vertical line indicates global importance, and the horizontal line represents local importance. STEP 1: Generate mask vector , for 1 ≤ j ≤ n STEP 1: Generate mask vector , for 1 ≤ j ≤ n STEP 2: Create a condition vector Condnj, for 1 ≤ j ≤ m STEP 3: Generate fake data ̂ STEP 4: Extract real data sample STEP 3: Generate fake data ̂ STEP 4: Extract real data sample The CICIDS 2017 dataset contains 60 features, of which the SHAP selects the 20 most impactful features. As shown in Fig. 4, features towards the top end (represented in darker shades of red) contribute more towards the output, while the features towards the bottom end (represented by darker shades of blue) contribute less. Algorithm 1 Feature Selection Fig. 2. Architecture of CTGAN. The generator learns the re-sampled distribution, which is different from the real data distribution. The main goal is to 176 International Journal of Machine Learning, Vol. 13, No. 4, October 2023 IV. EXPERIMENTAL RESULTS re-sample efficiently such that all categories from discrete attributes are evenly sampled during the training process and to recover the real data during the test. Conditional GAN has been widely used to generate a particular class of data. The critic maximizes the objective function, whereas the generator minimizes the objection function. In our proposed method, we used the CICIDS2017 Kaggle dataset. This dataset captures the network traffic fluctuations for five days a week. The dataset comprises the following attacks: DoS, BruteForce, Web Attack, Bot, Portscan, Infiltration, and DDoS. We have used the data generated for Friday afternoon working hours in which 97,718 are benign, and are DDoS. There is a total of 79 features captured in this dataset. C. Predictive Classifiers The prediction phase uses the synthetic data with selected features and different machine learning algorithms to classify benign and DDoS attacks. SVC is one of the most efficient supervised learning algorithms for classification and regression problems. The main objective of this algorithm is to create the best-fit line that divides -dimensional spaces into different classes. This best fit is also known as a hyperplane. The number of dimensions depends on the total number of features in the dataset. For linear SVM, one straight line is sufficient to divide the space into classes, whereas for non-linear data, new dimensions have to be added to the existing one. The data points closer to hyperplanes influence the plane's orientation. There are four significant stages in our method as shown in Fig. 3. Fig. 3. Proposed workflow. is a classifier that combines multiple decision trees to create and classify the data. Merging numerous decision trees gives a better prediction. The classification type prefers voting where maximum voting wins. This classifier provides better regression accuracy and considers the average of all the outputs. RF is a type of bootstrap aggregation. It is an ensemble machine learning algorithm, which is also known as the bagging method. Bootstrap aggregation is the best method to reduce the variance, which causes the model to overfit. Step 1: Data Preprocessing is the process of removing the duplicated data and null spaces. The dataset should be balanced for the training to not overfit the model. The label encoding is to code benign as 0 and DDoS as 1. NB is a classifier based on Bayes' Theorem. From the given Bayes' equation (4), A is referred to as a hypothesis and B as evidence. It is named as Naïve because the presence of one particular feature does not affect the other. Step 2: After data cleaning, the most important features must be considered for synthetic data generation. The feature selection has been made using SHAP, where the model has been given with a train-test spit of , respectively. SHAP performs permutations to find the marginal contributions of each feature. (4) (4) (4) Algorithm 2 Synthetic Data Generation CTGAN INPUT : Labelled data with SHAP selected features Assumptions : Dtrain for Training Data, G for Conditional Generator, C for Critic, M for Mask Vector, Condnj for Condition Vector Critic - Maximize the objective function to classify real and fake. C. Predictive Classifiers Hence, for the CICIDS 2017 dataset that we make use of, feature 37 (mean packet length) is the most significant, whereas feature 5 (total length of the forward packet) has the least significance. , for STEP 5: Cost Function = ∑ ( \| ∑ ( ̂ \| , for STEP 5: Cost Function = ∑ ( \| ∑ ( ̂ \| STEP 6: Gradient Penalty = ∑ ( ̂ \| ) 177 International Journal of Machine Learning, Vol. 13, No. 4, October 2023 Fig. 4. SHAP feature Importance. Fig. 4. SHAP feature Importance. TABLE II: LIST OF USED ABBREVIATIONS AND IT'S DEFINITION SYMBOL DEFINITION True Positive (TP) It is the number of true DDoS attacks. True Negative (TN) It is the number of true legitimate traffic: the benign is recognized as legitimate. False Positive (FP) It is the number of false legitimate traffic: the benign is misidentified as attacks False Negative (FN) It is the number of DDoS attacks that cannot be recognized as an attack TABLE II: LIST OF USED ABBREVIATIONS AND IT'S DEFINITION SYMBOL DEFINITION Accuracy is the ratio of correctly classified values to the total number of values. It captures the model's overall performance, and is given by (5). (5) (5) Precision is the ratio of true positives to the total number of values identified as positive, and is written as (6). Fig. 4. SHAP feature Importance. (6) (6) Step 3: The dataset with reduced features is given for CTGAN to generate synthetic data. CTGAN has been used to create the tabular data for further analysis. There are no mode collapse and vanishing gradient issues in CTGAN. The synthetic data generation by CTGAN before and after feature selection is computed and the results are compared in the Table I. Recall is the ratio of correctly classified values to the total number of elements that belong to the positive class. It is also referred to as Detection Rate. True Positive Rate is the measure of true samples which are identified correctly, and is also referred to as Sensitivity. Mathematically, this is captured as (7). C. Predictive Classifiers Distributed Denial of Service Attack Detection using Deep Learning Approaches [13] CNN and RNN Stacked LSTM and CNN 0.9955 and 0.96 3.A Comparison of Various Machine Learning Algorithms in a Distributed Denial of Service Intrusion [11] max,min,me an and std deviation SVM Linear, RF, NB 0.6325, 0.9997, 0.9977 4. A Deep Learning Approach for DDoS Attack Detection Using Supervised Learning [14] Knowledge Graph Stacked Auto Encoder (SAE) and CNN 0.9997 5. A Flow Based Anomaly Detection Approach with Feature Selection Method Against DDoS Attacks in SDN [16] RFR and IG SDN 0.9995 6. Proposed Model SHAP SVM, RFC, NB 0.9947,0.999 7,0.9803 TABLE V: PREDICTIVE CLASSIFIERS COMPARISON BETWEEN WITH AND WITHOUT SHAPLEY FOR PRECISION CLASSIFIER PRECISION Regressor and Information Gain used in [16] to extract the best features from the existing dataset provides an accuracy of 99.95%. Table VII summarizes the comparison of the result of our proposed model with similar works in the literature. The proposed model demonstrates a significant improvement in accuracy, as indicated by the experimental results. The following lines compare the state of the art of proposed model with other references' state of the art: 1) Compared to the results presented in [8], our proposed model shows a substantial increase in accuracy by 10%. 1) Compared to the results presented in [8], our proposed model shows a substantial increase in accuracy by 10%. 2) The performance of our proposed model surpasses that of the model in [13] by 4%. 2) The performance of our proposed model surpasses that of the model in [13] by 4%. 3) In comparison to the results presented in [11], our proposed model exhibits comparable accuracy percentages, as shown in Table V for the RF classifier. 4) Our proposed model also shows similar accuracy percentage as that presented in [14], based on the results presented in the table. 5) The method presented in [16] shows a negligible improvement in accuracy by 0.02%, in comparison to our proposed model. Overall, the results suggest that our proposed model is quite effective and performs better than some existing models while being comparable to others. Furthermore, the implementation of SHAP feature selection in the proposed model has resulted in a noteworthy reduction in computational cost. Specifically, the training time for the CTGAN model using all features was observed to be 11.06 s/iterations, while the execution time was reduced to 5.9 s/iterations by utilizing the SHAP-selected features. C. Predictive Classifiers TABLE I: COMPARISON OF ERROR FOR DIFFERENT MODELS MODEL PREDICTION (RMSE) 100 PERCENT FEATURE UTILIZATION (no Shapley) 0.07260 50 PERCENT FEATURE UTILIZATION (Shapley) 0.04321 75 PERCENT FEATURE UTILIZATION (Shapley) 0.0484 TABLE I: COMPARISON OF ERROR FOR DIFFERENT MODELS (7) (7) And lastly, F1 Score is the harmonic mean of precision and recall as given by (8). (8) (8) (8) Step 4: The SHAP selected feature dataset has been used for training the predictive classifiers such as the SVC, RF, and NB. The confusion matrix gives the performance metric for all the classifiers. TABLE III: PREDICTIVE CLASSIFIERS COMPARISON BETWEEN WITH AND WITHOUT SHAPLEY FOR ACCURACY CLASSIFIER ACCURACY WITHOUT SHAP WITH SHAP (50 percent WITH SHAP (75 percent SVC 0.9933 0.9884 0.9947 RF 0.999 0.9996 0.9997 NB 0.9474 0.9759 0.9803 TABLE IV: PREDICTIVE CLASSIFIERS COMPARISON BETWEEN WITH AND WITHOUT SHAPLEY FOR F1-SCORE CLASSIFIER F1 SCORE WITHOUT SHAP WITH SHAP (50 percent WITH SHAP (75 percent SVC 0.9941 0.9899 0.9953 RF 0.9991 0.9996 0.9997 NB 0.9555 0.9791 0.9829 The Confusion Matrix is the map between actual and predicted values with the following evaluation parameters: Accuracy, F1 Score, Precision, and Recall. The abbreviations used to define these quantities, along with their definitions, are provided in Table II. Table III, Table IV, Table V, and Table VI portray the various performance metrics for our proposed model. Accuracy, F1 Score, Precision, and Recall are computed with SHAP (using reduced number of features) and without SHAP (using all 79 features) datasets. The Accuracy measure before and after applying SHAP is 99.9% for Random Forest Classifier. Naïve Bayes performs well after feature reduction using SHAP, where the measure is 98%. 178 International Journal of Machine Learning, Vol. 13, No. 4, October 2023 TABLE V: PREDICTIVE CLASSIFIERS COMPARISON BETWEEN WITH AND WITHOUT SHAPLEY FOR PRECISION CLASSIFIER PRECISION WITHOUT SHAP WITH SHAP (50 percent WITH SHAP (75 percent SVC 0.9904 0.9818 0.9919 RF 0.9998 1.0 1.0 NB 0.9161 0.9598 0.967 TABLE VI: PREDICTIVE CLASSIFIERS COMPARISON BETWEEN WITH AND WITHOUT SHAPLEY FOR RECALL CLASSIFIER RECALL WITHOUT SHAP WITH SHAP (50 percent WITH SHAP (75 percent SVC 0.9978 0.99803 0.9986 RF 0.9984 0.9992 0.9994 NB 0.9984 0.9992 0.9992 TABLE VII: PERFORMANCE COMPARISON OF PROPOSED MODEL WITH OTHER LITERATURES Reference Papers Feature Selection Classifier Accuracy 1. A Feature Similarity Machine Learning Model for DDoS Attack Detection in Modern Network Environments for Industry 4.0 [8] Swathi Naïve Bayes 0.9091 2. C. Predictive Classifiers These results indicate that the proposed DDoS attack detection framework is efficient and performs well with a reduced set of features obtained through SHAP feature selection. REFERENCES [17] D. Fryer, I. Strümke, and H. Nguyen, “Shapley values for feature selection: The good, the bad, and the axioms,” Feb. 2021. [1] K. Shaukat, S. Luo, V. Varadharajan, I. A. Hameed, and M. Xu, “A survey on machine learning techniques for cyber security in the last decade,” IEEE Access, vol. 8, pp. 222310–222354, 2020. doi: 10.1109/ACCESS.2020.3041951 [18] S. M. Lundberg, P. G. Allen, and S.-I. Lee, “A unified approach to interpreting model predictions,” Advances in Neural Information Processing Systems, 2017. [19] B. Rozemberczki, L. Watson, P. Bayer, H.-T. Yang, O. Kiss, S. Nilsson, and R. Sarkar, “The shapley value in machine learning,” arXiv preprint arXiv:2202.05594, 2022. [2] Ismail, M. I. Mohamand, H. Hussain, A. A. Khan, U. Ullah, M. Zakarya, A. Ahmed, M. Raza, I. U. Rahman, and M. Haleem, “A machine learning-based classification and prediction technique for DDoS attacks,” IEEE Access, vol. 10, pp. 21443–21454, 2022. doi: 10.1109/ACCESS.2022.3152577 [20] W. E. M. Júnior and D. M. Eler, “Explaining dimensionality reduction results using Shapley values,” Mar. 2021. [21] M. Sundararajan and A. Najmi, “The many shapley values for model explanation,” in Proc. International Conference on Machine Learning, 92699278, 21 Nov. 2020. [3] I. H. Sarker, A. S. M. Kayes, S. Badsha, H. Alqahtani, P. Watters, and A. Ng, “Cybersecurity data science: an overview from machine learning perspective,” J Big Data, vol. 7, no. 1, Dec. 2020. doi: 10.1186/s40537020-00318-5 [22] J. Chen, X. Zhang, R. Zhang, C. Wang, and L. Liu, “De-Pois: An Attackagnostic defense against data poisoning attacks,” IEEE Transactions on Information Forensics and Security, 2021. https://doi.org/10.48550/arXiv.2105.03592 [4] M. Wazid, A. K. Das, V. Chamola, and Y. Park, “Uniting cyber security and machine learning: Advantages, challenges and future research,” ICT Express, vol. 8, no. 3. pp. 313-321, Sep. 01, 2022. doi: 10.1016/j.icte.2022.04.007 [23] R. Chauhan, U. Sabeel, A. Izaddoost, and S. Shah Heydari, “Polymorphic adversarial cyberattacks using WGAN,” Journal of Cybersecurity and Privacy, vol. 1, no. 4, pp. 767–792, Dec. 2021. [5] B. B. Gupta, R. C. Joshi, and M. Misra, “Prediction of number of zombies in a DDoS attack using polynomial regression model,” Journal of Advances in Information Technology, vol. 2, no. 1, Feb. 2011. doi: 10.4304/jait.2.1.57-62 Cybersecurity and Privacy, vol. 1, no. 4, pp. 767–792, De [24] L. Xu, M. Skoularidou, A. Cuesta-Infante, and K. Veeramachaneni, “Modeling tabular data using conditional GAN,” Advances in Neural Information Processing Systems, 2019. [6] P. S. Saini, S. Behal, and S. International Journal of Machine Learning, Vol. 13, No. 4, October 2023 [9] R. Alenezi and S. A. Ludwig, “Explainability of cybersecurity threats data using SHAP,” in Proc. 2021 IEEE Symposium Series on Computational Intelligence (SSCI), IEEE, 2021, pp. 01–10. FUNDING [14] H. Tekleselassie, “A deep learning approach for DDoS attack detection using supervised learning,” in Proc. MATEC Web of Conferences, vol. 348, p. 01012, 2021. doi: 10.1051/matecconf/202134801012 This research is funded, in part, by BITS-Pilani Hyderabad Campus under the RIG scheme with grant number BITS/GAU/RIG/2022/H0753, and, in part, by Thapar Institute of Engineering and Technology under the grant number TU/DORSP/07/02/2022. C Cynthia's doctoral research is supported by BITS-Pilani Hyderabad Campus under the "Institute Fellowship" scheme. [15] C. Zhang, Y. Chen, Y. Meng, F. Ruan, R. Chen, Y. Li, and Y. Yang, “A novel framework design of network intrusion detection based on machine learning techniques,” Security and Communication Networks, vol. 2021, Article ID 6610675. https://doi.org/10.1155/2021/6610675 g q y vol. 2021, Article ID 6610675. https://doi.org/10.1155/2021/6610675 [16] M. S. El Sayed, N.-A. Le-Khac, M. A. Azer, and A. D. Jurcut, “A flow based anomaly detection approach with feature selection method against ddos attacks in sdns,” IEEE Transactions on Cognitive Communications and Networking, vol. 8, no. 4, pp. 1862–1880, 2022. mmunications and Networking, vol. 8, no. 4, pp. 1862–1880, 2022. AUTHOR CONTRIBUTIONS Ms. C Cynthia formulated the problem, proposed the methodology presented in this manuscript and implemented the proposed methodology via coding. Dr. Debayani Ghosh and Dr. Gopal Krishna Kamath helped formalize the problem and helped Ms. C Cynthia, in equal measure, solve the problem via the use of SHAP technique. All authors were involved in writing and revision of the manuscript, with Ms. C Cynthia taking the lead by preparing the preliminary draft and generating all the figures/tables presented in this manuscript. All authors have approved the final version of the manuscript. [10] R. J. Alzahrani and A. Alzahrani, “Security analysis of ddos attacks using machine learning algorithms in networks traffic,” Electronics (Switzerland), vol. 10, no. 23, Dec. 2021. doi: 10.3390/electronics10232919 [11] S. H. Kok, A. Abdullah, M. Supramaniam, T. R. Pillai, I. Abaker, and T. Hashem, “A comparison of various machine learning algorithms in a distributed denial of service intrusion,” 2019. [12] A. U. Sudugala, W. H. Chanuka, A. M. N. Eshan, U. C. S. Bandara, and K. Y. Abeywardena, “WANHEDA: A machine learning based DDoS detection system,” in Proc. ICAC 2020 - 2nd International Conference on Advancements in Computing, Dec. 2020, pp. 380–385. doi: 10.1109/ICAC51239.2020.9357130 [13] Meenakshi, K. Kumar, and S. Behal, “Distributed denial of service attack detection using deep learning approaches,” in Proc. 2021 8th International Conference on Computing for Sustainable Global Development (INDIACom), IEEE, 2021, pp. 491–495. V. CONCLUSION In this paper, we have proposed a framework to detect DDoS attacks with a reduced set of features. We have selected the CICIDS2017 dataset from the Kaggle repository, which contains information about DDoS attacks. In the proposed detection model, we have used the SHAP feature selection technique to reduce the number of features to work with. This reduces the computational cost and improves the accuracy. A feature's contribution towards model prediction and its importance are identified using the SHAP values. Synthetic data with the selected (reduced) features are generated using a CTGAN. Predictive classifiers are then used to detect the DDoS attack from the synthetic data. Our model has focused on the pattern of the specific attack, which is otherwise known as signature-based detection. This work can be extended in two ways: (i) To train the model for different cyber-attacks and (ii) To build anomaly-based intrusion detection systems. It is imperative to develop new techniques for better detection systems and thus also protect against any malicious attacks. We now provide results of related works. In [8], the feature transformation-based dimensionality reduction is performed with their novel approach, named Swasthi, and with the classifier Naïve Bayes, the accuracy percentage is 90.91%. In addition, [13] uses RNN and CNN models with an accuracy measure of 99.55% and 96%, respectively. Furthermore, the authors in [11] use a few feature selection methods using min, max, mean, and standard deviation with the predictive classifiers SVM Linear, Random Forest, and Naïve Bayes, resulting in accuracy percentages 63.25%, 99.97%, and 99.77%, respectively. In [14], authors compute a knowledge graph for better feature extraction, and the classifier integrates Stacked Auto-Encoder and CNN, which gives a better accuracy measure of 99.97%. The Random Forest CONFLICT OF INTEREST The authors declare no conflict of interest. 179 International Journal of Machine Learning, Vol. 13, No. 4, October 2023 REFERENCES Bhatia, “Detection of DDoS attacks using machine learning algorithms,” presented at 7th International Conference on Computing for Sustainable Global Development (INDIACom), 12-14 March 2020, New Delhi, India. [25] Z. Zhao, A. Kunar, H. van der Scheer, R. Birke, and L. Y. Chen, “CTAB-GAN: Effective table data synthesizing,” in Proc. Asian Conference on Machine Learning, pp. 97–112, 2021. ( ) [7] S. A. Abbas and M. S. Almhanna, “Distributed denial of service attacks detection system by machine learning based on dimensionality reduction,” Journal of Physics: Conference Series, Mar. 2021, vol. 1804, no. 1. doi: 10.1088/1742-6596/1804/1/012136 Copyright © 2023 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0). [8] S. Sambangi, L. Gondi, and S. Aljawarneh, “A feature similarity machine learning model for DDoS attack detection in modern network environments for Industry 4.0,” Computers and Electrical Engineering, vol. 100, May 2022. doi: 10.1016/j.compeleceng.2022.107955 180
https://openalex.org/W2970407827	https://www.frontiersin.org/articles/10.3389/fphar.2019.01107/pdf	English	null	Metabolomic Analysis of Platelets of Patients With Aspirin Non-Response	Frontiers in pharmacology	2,019	cc-by	7,540	Keywords: aspirin non-response, metabolomics, PFA-100, glycine, platelet Metabolomic Analysis of Platelets of Patients With Aspirin Non-Response Jiun-Yang Chiang 1,2, Sheng-Han Lee 3, Yen-Ching Chen 4, Cho-Kai Wu 1, Jing-Yuan Chuang 5, Shyh-Chyi Lo 6, Huei-Ming Yeh 7, Shih-Fan Sherri Yeh 8, Cheng-An Hsu 9, Bin-Bin Lin 9, Pi-Chu Chang 9, Chih-Hsin Chang 9, Hao-Jan Liang 3, Fu-Tien Chiang 1,10, Ching-Yu Lin 3* and Jyh-Ming Jimmy Juang 1* 1 Cardiovascular Center and Division of Cardiology, Department of Internal Medicine, National Taiwan University Hospital and National Taiwan University College of Medicine, Taipei, Taiwan, 2 Graduate Institute of Clinical Medicine, National Taiwan University College of Medicine, Taipei, Taiwan, 3 Institute of Environmental Health, College of Public Health, National Taiwan University, Taipei, Taiwan, 4 College of Public Health, Institute of Epidemiology and Preventive Medicine, National Taiwan University, Taipei, Taiwan, 5 Department of Medical Laboratory Science and Biotechnology, China Medical University, Taichung, Taiwan, 6 Department of Laboratory Medicine, National Taiwan University Hospital, Taipei, Taiwan, 7 Department of Anesthesiology, National Taiwan University Hospital and National Taiwan University College of Medicine, Taipei, Taiwan, 8 Department of Environmental and Occupational Medicine, National Taiwan University Hospital and National Taiwan University College of Medicine, Taipei, Taiwan, 9 Division of Haematology, Department of Laboratory Medicine, National Taiwan University Hospital, Taipei, Taiwan 10 Cardiovascular Center, Fu-Jen Catholic University Hospital, New Taipei, Taiwan ORIGINAL RESEARCH doi: 10.3389/fphar.2019.01107 published: 10 October 2019 Metabolomic Analysis of Platelets of Patients With Aspirin Non-Response Jiun-Yang Chiang 1,2, Sheng-Han Lee 3, Yen-Ching Chen 4, Cho-Kai Wu 1, Jing-Yuan Chuang 5, Shyh-Chyi Lo 6, Huei-Ming Yeh 7, Shih-Fan Sherri Yeh 8, Cheng-An Hsu 9, Bin-Bin Lin 9, Pi-Chu Chang 9, Chih-Hsin Chang 9, Hao-Jan Liang 3, Fu-Tien Chiang 1,10, Ching-Yu Lin 3* and Jyh-Ming Jimmy Juang 1* Edited by: Changhua Wang, Wuhan University, China Reviewed by: Jun-Song Wang, Nanjing University of Science and Technology, China Wimal Pathmasiri, University of North Carolina at Chapel Hill, United States Baharudin Ibrahim, Universiti Sains Malaysia, Malaysia Claudio Luchinat, University of Florence, Italy Background: Aspirin is the most commonly used antiplatelet agent for the prevention of cardiovascular diseases. However, a certain proportion of patients do not respond to aspirin therapy. The mechanisms of aspirin non-response remain unknown. The unique metabolomes in platelets of patients with coronary artery disease (CAD) with aspirin non- response may be one of the causes of aspirin resistance. Background: Aspirin is the most commonly used antiplatelet agent for the prevention of cardiovascular diseases. However, a certain proportion of patients do not respond to aspirin therapy. The mechanisms of aspirin non-response remain unknown. The unique metabolomes in platelets of patients with coronary artery disease (CAD) with aspirin non- response may be one of the causes of aspirin resistance. Correspondence: Ching-Yu Lin chingyulin@ntu.edu.tw Jyh-Ming Jimmy Juang jjmjuang@ntuh.gov.tw Materials and Methods: We enrolled 29 patients with CAD who were aspirin non- responders, defined as a study subject who were taking aspirin with a platelet aggregation time less than 193 s by PFA-100, and 31 age- and sex-matched patients with CAD who were responders. All subjects had been taking 100 mg of aspirin per day for more than 1 month. Hydrophilic metabolites from the platelet samples were extracted and analyzed by nuclear magnetic resonance (NMR). Both 1D 1H and 2D J-resolved NMR spectra were obtained followed by spectral processing and multivariate statistical analysis, such as partial least squares discriminant analysis (PLS-DA). Specialty section: This article was submitted to Cardiovascular and Smooth Muscle Pharmacology, a section of the journal Frontiers in Pharmacology Results: Eleven metabolites were identified. The PLS-DA model could not distinguish aspirin non-responders from responders. Those with low serum glycine level had significantly shorter platelet aggregation time (mean, 175.0 s) compared with those with high serum glycine level (259.5 s). However, this association became non-significant after correction for multiple tests. Received: 13 November 2018 Accepted: 28 August 2019 Published: 10 October 2019 INTRODUCTION treatment was associated with more ischemic events in patients with myocardial infarction (Reny et al., 2008). One recent reviewed article showed a trend toward more death and major adverse cardiovascular events in PFA-100-defined aspirin non-responders. The cutoffs of PFA-100 range from 150 to 193 s in different studies (Dretzke et al., 2015). However, research on normal reference of PFA-100 from Asian population is relatively limited. Cardiovascular diseases are the most common cause of mortality and morbidity worldwide. According to the American Heart Association’s updated report in 2018, cardiovascular mortality ranked first in all kinds of deaths and produce immense health and economic burdens (Wu et al., 2010; Benjamin et al., 2018). In Taiwan, cardiovascular disease ranked second in the top 10 leading causes of death in 2011. Atherothrombosis, characterized by atherosclerotic lesion disruption with thrombotic complications, is the main cause of acute cardiovascular events. Accumulation of platelets at the site of atherosclerotic lesion disruption is the first step in the formation of atherothrombosis; these elements are responsible for the formation of pathogenic thrombi in patients with atherothrombotic diseases, such as coronary artery disease (CAD), ischemic stroke/transient ischemic attack (TIA), and peripheral artery disease (PAD). Metabolomic studies provide a rapid and high throughput screening tool for understanding the relationships between metabolic profiles and biological effects in responding to stimulation of genetic modulation, drugs, toxicants, and various kinds of disease. The low molecular weight metabolites such as sugar, amino acids, lipids, intermediates in metabolism, and degradation products of exogenous compounds are analyzed in the field of metabolomics. Like genomics and proteomics, metabolomics can be examined in different levels of biological systems such as cells (Ellis et al., 2011; Wu et al., 2017), tissues (Viant et al., 2006; Lindon et al., 2007), biofluids (Tiziani et al., 2009), and whole organisms. Nuclear magnetic resonance (NMR) is a common analytical technique applied in metabolomics; it requires less sample preparation and provides rapid analysis and high reproducibility of metabolic profiles. Similar to other omics, metabolomics can also be applied in several research fields such as pharmacology (Bollard et al., 2005), toxicology (Azmi et al., 2005; Viant et al., 2006), and disease diagnosis (Brindle et al., 2002; Gowda et al., 2008). Aspirin is by far the most commonly used antiplatelet agent for the prevention of cardiovascular diseases. INTRODUCTION The Antithrombotic Trialists’ Collaboration documented a 22% reduction in death and serious ischemic vascular events with the use of antiplatelet therapy when compared with a placebo in their most recent meta-analysis of 287 randomized trials, comprising more than 200,000 patients (Antithrombotic Trialists' Collaboration, 2002). Although aspirin therapy is effective, 8% to 45% of patients do not respond to aspirin therapy as determined by different laboratory tests; these aspirin non-responders are at increased risk of thrombotic events (Eikelboom et al., 2002; Gum et al., 2003; Chen et al., 2004; Krasopoulos et al., 2008; Foussas et al., 2009; Frelinger et al., 2009; Li et al., 2014). Stejskal et al. reported that patients with acute coronary syndrome and aspirin non- response had an 88% incidence of recurrent cardiovascular events compared with 47% in aspirin responders (Stejskal et al., 2006). Gum et al. showed that aspirin non-response was significantly associated with an increased risk of major cardiovascular adverse event (24% vs 10%, hazard ratio 3.12) (Gum et al., 2003). Several single nucleotide polymorphisms, miRNAs, increased inflammatory activity, drug noncompliance, and metabolic syndromes have been proposed as contributors to residual platelet activity under aspirin treatment (Wurtz et al., 2012; Weng et al., 2013; Du et al., 2016). For example, genetic polymorphisms in platelet glycoprotein and P2Y receptors are linked to an attenuated effect of aspirin, and increased risk of thromboembolic events (Szczeklik et al., 2000; Li et al., 2007). Most studies of platelet of aspirin non-responder focused on upstream biological processes, but few have addressed the downstream omics such as metabolomics. Research on platelet’s metabolites has revealed that certain metabolites were associated with platelet’s recovery after storage and its survival (Zimring et al., 2016), proving the concept that metabolomics may be a reflection of certain phenotype of platelet. By utilizing metabolomic analysis, changes in the small molecules of platelets in aspirin non-responders may be detected, providing insights into the mechanism of non-response, and potentially discovering new biomarkers for this population (Johnson et al., 2016).i Platelets are circulating cell fragments that play a pivotal role in thrombosis. Although the platelet contains no nucleus, selective ligands for platelet membrane receptors activate their nuclear receptors and regulate platelet aggregation and activation. Abbreviations: AMI, acute myocardial infarction; TIA, transient ischemic attack; PAD, peripheral artery disease; PFA, platelet function assay; NMR, nuclear magnetic resonance; miRNA, micro ribonucleic acid; CAD, coronary artery disease; Col/EPI, collagen/epinephrine; Col/ADP, collagen/adenosine diphosphate; JRES, J-resolved spectroscopy; PCA, principal component analysis; PLA-DA, partial least squares discriminant analysis; ATP, adenosine triphosphate; ADP, adenosine diphosphate; AMP, adenosine monophosphate; DM, diabetes mellitus; SNP, single nucleotide polymorphism. Citation: Chiang J-Y, Lee S-H, Chen Y-C, Wu C-K, Chuang J-Y, Lo S-C, Yeh H-M, Yeh S-FS, Hsu C-A, Lin B-B, Chang P-C, Chang C-H, Liang H-J, Chiang F-T, Lin C-Y and Juang J-MJ (2019) Metabolomic Analysis of Platelets of Patients With Aspirin Non-Response. Front. Pharmacol. 10:1107. doi: 10.3389/fphar.2019.01107 Conclusions: The hydrophilic metabolic profile of platelets was not different between aspirin non-responders and responders. An association between lower glycine levels and higher platelet activity in patients younger than 65 years suggests an important role of glycine in the pathophysiology of aspirin non-response. October 2019 \| Volume 10 \| Article 1107 1 Frontiers in Pharmacology \| www.frontiersin.org Platelets in Aspirin Non-Response Chiang et al. INTRODUCTION A number of nuclear receptors have been found recently to be present in human platelets, including receptors for sex steroids, and glucocorticoids, along with peroxisome proliferator-activated receptors (PPARs) and retinoid X receptors (RXRs) (Bishop-Bailey, 2009), implying that biological reaction and metabolic activity in human platelet are active and complex.h We prospectively enrolled patients with CAD confirmed by coronary angiography or multidetector computed tomography (MDCT) who took aspirin daily and we then compared platelet metabolome between patients with CAD with and without aspirin non-response. There are a number of available tests for assessing platelet function. PFA-100 is one of the most commonly used modalities worldwide. Shortened PFA-100 closure time under aspirin NMR Spectroscopy Blood samples were collected via radial or femoral artery using 18-gauge needles on the enroll day at the catheterization laboratory, and processed within 1 h after collection. Fresh blood (10 ml) anticoagulated with ethylenediaminetetraacetic acid (EDTA) was collected for baseline hemoglobin and platelet count, and 20 ml whole blood was collected in evacuated tubes containing 3.2% sodium citrate. The whole blood mixed with sodium citrate was centrifuged at 200 g for 20 min to obtain platelet-rich plasma (PRP). The platelets were prepared from PRP by centrifugation (900g, 10 min) and washed with Tyrode’s buffer ((137 mM NaCl, 2.65 mM KCl, 12 mM NaHCO3, 0.43 mM NaH2PO4, 2 mM CaCl2, 1 mM MgCl2, 5 mM glucose, 5 mM HEPES, pH 7.35) containing 0.5 μM PGI2 and 0.2 unit/ml apyrase. Washed platelets were then re-suspended in Tyrode’s buffer and adjusted to 3 × 108/ml for platelet aggregation (Chang et al., 2015). Hydrophilic platelet metabolites were resuspended in a 600-µl D2O sodium phosphate buffer (0.1 M, pH 7.4) containing 0.25 mM 3-trimethylsilyl-2,2,3,3-d4-propionate (TSP) as an internal standard. NMR spectra were acquired from 500.13 MHz in Bruker Avance-500 spectrometers equipped with a 5-mm QNP CryoProbe (Bruker, BioSpin) at the Core Facility for Protein Structural Analysis supported by the National Core Facility Program for Biotechnology in Academia Sinica, Taiwan. g gy Both 1H and JRES NMR spectra were acquired at 300 K. 1H NMR spectra were acquired from the pulse sequence (relaxation delay: -90°-t-90°-tm-90°-acquired free induction decay) with presaturation of water resonance to suppress the water signal. For each sample, 32 k data points were collected from 128 scans in a 20-ppm spectral width with a 2.0-s relaxation delay and an acquisition time of 1.63 s. JRES NMR spectra were acquired from the pulse sequence (relaxation delay-90°-t1-180°-t1-acquired free induction delay, with water suppression during relaxation delay). Sixteen k data points in F2 frequency axis and 40 data points in F1 frequency axis were collected in eight scans with a 2.0-s relaxation delay. The spectral widths of the F2 and the F1 axis were set 6000 and 65 Hz, respectively. Assessment of Response to Aspirin Intake Blood collected in 3.2% sodium citrate was tested by PFA-100 (Siemen Healthcare Diagnostics, Marburg, Germany). PFA- 100 tests were performed within 1 h of blood sampling. Whole blood was loaded into prefabricated proprietary cartridges containing a combination of collagen and epinephrine as platelet agonists. Sample Preparation for Metabolomic Study Extraction of low molecular weight metabolites from the platelets was conducted as described in previous research (Lin et al., 2007). Briefly, metabolites were extracted by resuspending the platelets collected from fresh blood anticoagulated with EDTA (average 280106 platelet/ml) and precipitating platelet proteins in a total of 1.16 ml methanol-chloroform-water (2/2/1.8) solvent (Bligh and Dyer, 1959). After vortexing and centrifugation (4°C, 1,000g, 15 min), 0.6 ml hydrophilic metabolic extracts were taken and dried out with a speed vacuum and then stored at −80°C until NMR analysis. Subjectsh This cross‐sectional case-control study included a total of 60 Taiwanese cardiovascular patients aged 39 to 85 years who had October 2019 \| Volume 10 \| Article 1107 Frontiers in Pharmacology \| www.frontiersin.org 2 Platelets in Aspirin Non-Response Chiang et al. elective coronary angiography for CAD or valvular heart disease at National Taiwan University Hospital between June 2011 and February 2013. All enrolled subjects had been taking 100 mg of aspirin per day for more than 1 month. The study complies with the declaration of Helsinki, and the study protocol was approved by the institutional research ethics committee (IRB number: 201106077RC). All patients gave written informed consent. Each patient filled out a self‐report questionnaire and supplied blood samples for identification of aspirin users and further metabolomics research. A self‐report questionnaire was conducted to collect information on demographics, platelets count, disease history (i.e., hypertension, diabetes and fatty liver), lifestyle (i.e., smoking), and general biological parameters (i.e., hemoglobin, aspartate aminotransferase, creatinine, total cholesterol, triglycerides, sodium and potassium). Aspirin non-responder was defined as a study subject whose platelet aggregation time result was less than 193 s by PFA-100 test. Details of PFA-100 test are provided in the following section. For each aspirin non-responder, a responder, with PFA-100 result more than 193 s, was chosen matching on age, sex, diabetes mellitus, hypertension, dyslipidemia, smoking status, level of hemoglobin, platelet, renal function, liver function, and lipid profile as these factors are potential confounders of platelet function. individuals served as internal controls to define the normal range of platelet aggregation for Taiwanese people. The results showed 86 to 193 s for closure of Col/EPI cartridge, 61 to 109 s for Col/ADP cartridge. This cutoff value was similar to prior studies in Caucasian population (Gum et al., 2001; Jilma and Fuchs, 2001; Angiolillo et al., 2006; Simpson et al., 2014). The distribution of clotting times of aspirin responders and non-responders in our study was shown in Supplementary Figure 1. Participants with a clotting time greater than or equal to 300 s, the upper detection limit of PFA-100 assay, were grouped together. NMR Spectroscopy Platelet function analyzer closure time (the time for cessation of flow caused by formation of a platelet plug) was recorded in seconds. The maximum time allowed for closure was 300 s. Because no normal ranges had been established in Asian populations, we conducted a study on 100 healthy voluntary controls (50 men and 50 women; mean age = 42.1 ± 10 years) who were not taking any medications; these Statistical Analyses and Metabolite Identificationh i The processed p-JRES data were uploaded to Metaboanalyst 2.0 (http://www.metaboanalyst.ca) (Xia et al., 2012) for principal component analysis (PCA) and partial least‐squares discriminant analysis (PLS‐DA) (Hastie et al., 2009; Dormann J Elith et al., 2013). These two analyses evaluated the similarities among samples and identified unique metabolites corresponding to specific clusters. PCA, an unsupervised method, was used to reduce the complexity of metabolomics data matrix without additional group information. The PCA score plot provided the visual performance of the metabolome variation for each sample. In contrast, PLS‐DA, a supervised method, linked the group information and NMR dataset to determine the variance between the aspirin-responders and aspirin non-responders. The PLS‐DA score plot was used to illustrate if the metabolome of different aspirin-response types could be separated. The loading plots of PCA and PLS-DA were employed to identify the possible metabolites corresponding with different aspirin groups. Cross‐validated predictive capability (Q2) in 10‐fold internal cross‐validation and statistic value p in permutation test were conducted to evaluate the robustness and predictability of the PLS‐DA models (Westerhuis et al., 2008). Wilcoxon signed-rank tests (continuous variables) and Pearson’s χ2 tests (categorical variables) were utilized to compare the basic demographic data between aspirin responders and non-responders. Since age status was regarded as a major factor for clinical study, a stratified analysis was conducted to assess the association between metabolites and aspirin according to the referring value (clotting time = 193 s) or the median of the clotting time. The possible relationship between the significant changed metabolite and clotting time was assessed by the spearman’s rank correlation test. All analyses were administered by using SAS 9.2 (SAS Institute, Cary, NC, USA). All statistical tests were two‐sided, and the value of significance was set at p < 0.05.h TABLE 1 \| Characteristics of the study population. Metabolic Variation of Patients by Multivariate Statistical Analysis A representative NMR spectrum of hydrophilic platelet metabolites from one sample is shown in Supplementary Figure 2. A total of 11 metabolites were identified on the p-JRES spectum, including valine, lactate, alanine, glutamate, succinate, taurine, myo-inositol, glycine, ATP, ADP, and AMP. The PCA model from the analysis of platelet metabolome showed no clear separation between aspirin responders and non-responders before or after stratification by age. In the PLS-DA model, components were calculated to maximize the covariance between class assignment and the platelet metabolome. Each spot on the score plot denotes one sample. Aspirin responders and non-responders were slightly separated along both components (component 1 accounted for 5.7% of the total variance, and component 2 for 9.3%) in the PLS-DA scores plot (Figure 1). The predictive capability: Q2 of 10‐ fold cross‐validation is 0.04; 1000 random permutation test is p = 0.99 (Table 2). Variable influence on projection of each metabolite was shown in Supplementary Table 2. After stratification by age (<65 vs ≥65), the metabolic profile was analyzed again, and the PLS-DA score plots revealed separated clusters between aspirin responders and non-responders. However, the cross‐validation was not ideal in those younger than 65 years (Q2 = −0.17; ppermutation = 0.99) and those 65 years or older (Q2 = −0.26; ppermutation = 0.990). RESULTS spectra were zero-filled to 8 k data points in F2, and up to 128 increments in F1 and processed with 0.3 Hz exponential line- broadening in both dimensions prior to Fourier transformation. The spectra were tilted (45°), symmetrized by F1, skyline- projected first and then manually phased, and the baseline was corrected and calibrated as 1H NMR analysis. Characteristics of the Study Population A total 31 aspirin responders and 29 non-responders were enrolled in the study. There is no significant difference in mean age, distribution of sex, body mass index, serum level of hemoglobin, platelet, total cholesterol, triglyceride, and percentage of patients with hypertension and diabetes mellitus (DM) between aspirin responders and non-responders (Table 1). y The 1H and the J-resolved spectroscopy projections (p-JRES) data were further processed by custom-written ProMetab software (version ProMetab_v3_3) (Viant, 2003) located in the MATLAB platform (Version 7.8; The MathWorks, USA). Spectra were divided into 1,850 chemical shift bins by using a bin width of 0.005 ppm (2.5 Hz) between 0.2 and 10.0 ppm, and then integrated spectral intensity for each bin. The regions of internal standard and water resonance were excluded before being normalized by total spectral area. The binned data were adjusted by generalized log transformation and Pareto scaling prior to multivariate analysis (Supplementary Table 1) (van den Berg et al., 2006). NMR Spectral Processingi Sp g 1H NMR spectral data were zero-filled to 64 k points and processed with 0.3 Hz exponential line-broadening before Fourier transformation. The spectra were phased, and the baseline was corrected and calibrated by TSP (δ = 0.00 ppm) by Topspin software (Version 2.1; Bruker, Billerica, MA, USA). JRES October 2019 \| Volume 10 \| Article 1107 Frontiers in Pharmacology \| www.frontiersin.org 3 Platelets in Aspirin Non-Response Chiang et al. Frontiers in Pharmacology \| www.frontiersin.org Statistical Analyses and Metabolite Identificationh Variables Aspirin responder (n = 31) Aspirin non- responder (n = 29) P Age (years) 63.5 ± 8.08 62.3 ± 12.83 0.66 Male (%) 23 (74.2%) 24 (82.8%) 0.42 Body weight (kg) 67.8 ± 13.48 74.6 ± 16.64 0.12 Body mass index (kg/m2) 26.0 ± 4.17 27.0 ± 4.80 0.34 Hemoglobin (g/dl) 13.5 ± 1.99 14.5 ± 1.90 0.06 Platelet (k/dl) 234.7 ± 45.54 232.2 ± 73.77 0.11 Aspartate aminotransferase (μl) 23.3 ± 9.23 25.7 ± 13.66 0.93 Creatinine (mg/dl) 1.2 ± 1.09 1.1 ± 0.38 0.58 Total cholesterol (mg/dl) 180.3 ± 30.35 174.5 ± 37.27 0.31 Triglycerides (mg/dl) 135.3 ± 79.60 138.3 ± 71.28 1.00 Sodium (mmol/µl) 139.6 ± 2.64 139.7 ± 2.35 0.75 Potassium (mmol/µl) 4.3 ± 0.41 4.3 ± 0.36 0.49 n (%) Hypertension 20 (64.5) 19 (65.5) 0.94 Diabetes mellitus 18 (58.1) 12 (41.4) 0.20 Lipid 16 (51.6) 16 (55.2) 0.78 Ever smoker 5 (16.7) 8 (27.6) 0.37 Continuous variable is shown as mean ± SD. OR, odds ratio; SD, standard deviation. TABLE 1 \| Characteristics of the study population. i The platelet metabolites from the 1H NMR spectra were identified by using Chenomx NMR software suite (Professional Edition, Version 5.1; Chenomx, Inc.). October 2019 \| Volume 10 \| Article 1107 Frontiers in Pharmacology \| www.frontiersin.org 4 Chiang et al. Platelets in Aspirin Non-Response FIGURE 1 \| PLS-DA score plots from the analysis of p-JRES spectra using platelet samples. (A) Aspirin responder: n = 31; Aspirin non-responder: n = 29. (B) Age < 65 years. aspirin responder: n = 16; aspirin non-responder: n = 17. (C) Age ≥ 65 years. aspirin responder: n = 15; aspirin non-responder: n = 12. PLS-DA, partial least-squares discriminant analysis. FIGURE 1 \| PLS-DA score plots from the analysis of p-JRES spectra using platelet samples. (A) Aspirin responder: n = 31; Aspirin non-responder: n = 29. (B) Age < 65 years. aspirin responder: n = 16; aspirin non-responder: n = 17. (C) Age ≥ 65 years. aspirin responder: n = 15; aspirin non-responder: n = 12. PLS-DA, partial least-squares discriminant analysis. TABLE 2 \| PLS-DA models and permutation tests for distinguishing between aspirin responders and non-responders by age groups. Population Aspirin responder (n) Aspirin non-responder (n) PLS-DA ppermutation b Componentsa Q2 Overall 31 29 4 0.04 0.99 Age, < 65 years 16 17 4 −0.17 0.99 Age, ≥ 65 years 15 12 3 −0.26 0.99 PLS-DA, partial least squares-discriminant analysis. Statistical Analyses and Metabolite Identificationh Q2 = predictive capability of PLS-DA model. aThe number of components based on Q2 indicates the best classifier of PLS-DA using a 10-fold cross-validation method. b1,000 random permutations performed. Association Between Candidate Metabolite and Mean Clotting Time When each metabolite was dichotomized into high and low levels by its median value, the clotting time was similar for all candidate metabolites. Because age is an important confounding factor, we further stratified this analysis by age groups (<65 years vs ≥65 years). We found that the clotting time was significantly different between high- and low-glycine groups (259.5 s vs 175.0 s, p = 0.007, Association Between Candidate Metabolite and Mean Clotting Time When each metabolite was dichotomized into high and low lev Association Between Candidate DISCUSSION This is the first study focusing on hydrophilic metabolites of platelet in aspirin non-responders by NMR-based metabolomics. Although our study showed that the metabolomic profiles were not different between aspirin responders and non-responders, a lower level of platelet glycine may be associated with shorter clotting time for patients younger than 65 years. By analyzing A comprehensive intracellular metabolic profile of platelets has been investigated (Paglia et al., 2014; Paglia et al., 2015). By Ultra performance liquid chromatography - tandem mass spectrometer, Paglia et al. identified 96 metabolites within platelets. Comparison between fresh and stored platelet concentrate using pathway analysis further revealed the most affected pathways during storage. These pathways included pentose phosphate pathway and nucleotide sugar metabolism. In another case control study, different platelet metabolism profiles between patients with and without Alzheimer’s disease were identified. A prediction model was constructed based on these differences (Oberacher et al., 2017). The results of these studies suggested that different platelet metabolic profile might offer new biomarkers in different diseases.h TABLE 3 \| The mean clotting time for high and low (median as the cutoff) level of metabolite by age group (< 65 and ≥ 65 years). Metabolite level# Clotting time (second), mean (SD) Age, < 65 years Age, ≥ 65 years Valine High 224.0 (84.01) 211.9 (76.72) Low 202.2 (84.33) 210.8 (95.81) Lactate High 202.6 (84.27) 211.1 (80.15) Low 230.1 (82.30) 211.6 (89.74) Alanine High 223.6 (80.09) 210.9 (86.46) Low 197.8 (89.62) 211.8 (86.38) Glutamate High 218.4 (86.77) 203.4 (81.38) Low 207.4 (82.17) 216.9 (89.17) Succinate High 214.3 (89.43) 196.2 (85.53) Low 212.8 (81.49) 233.5 (82.38) Taurine High 213.9 (86.72) 225.8 (82.50) Low 212.9 (82.94) 202.9 (87.37) Myo-inositol High 227.6 (80.19) 210.3 (82.76) Low 198.3 (86.70) 212.3 (89.17) Glycine High 259.5 (68.33) 230.6 (76.57) Low 175.0 (76.57)* 190.8 (91.17) ATP High 205.1 (86.04) 211.6 (89.24) Low 224.7 (81.89) 211.3 (84.48) ADP High 238.7 (78.91) 210.5 (82.58) Low 183.1 (81.25) 212.1 (89.30) AMP High 231.5 (84.66) 225.9 (80.33) Low 194.2 (80.56) 199.8 (89.10) iff Thomas et al. conducted a systematic review of literature by collecting research on platelet proteomics and fluxomics (Thomas et al., 2014). They constructed a cell scale, mass, and charge balanced model of platelet metabolism, iAT-PLT-636. ssoc a o e ee Ca d da e Metabolites and Aspirin Responseh When each metabolite was dichotomized into high and low levels by its median value, the clotting time was similar for all candidate metabolites. Because age is an important confounding factor, we further stratified this analysis by age groups (<65 years vs ≥65 years). We found that the clotting time was significantly different between high- and low-glycine groups (259.5 s vs 175.0 s, p = 0.007, The levels of metabolites were similar between aspirin responders and non-responders in overall patients and in age-stratified patients (<65 years vs ≥65 years). There was no association between each candidate metabolite and aspirin response status after adjusting for confounders, such as sex and other comorbidities. October 2019 \| Volume 10 \| Article 1107 Frontiers in Pharmacology \| www.frontiersin.org 5 Platelets in Aspirin Non-Response Chiang et al. Table 3). The glycine level was significantly positive correlated with the clotting time in this age group (Supplementary Table 3). However, this association became non-significant after correction for multiple tests. The median value of glycine among patients younger than 65 years was 0.002696. The sensitivity, specificity, positive predictive value, and negative predictive value of glycine level below 0.002696 to predict aspirin non-response was 0.77, 0.69, 0.72, and 0.73, respectively (Supplementary Table 4). the results of PFA-100 from 100 healthy Han Chinese in Taiwan, we defined the normal range of closure time for collagen/ epinephrine and collagen/ADP cartridges.i Investigation of platelet metabolomic profiles in aspirin non- responders is of great value, because the metabolic profile is closer to a disease phenotype than other upstream omics (Worley and Powers, 2013). Pharmacogenetic studies have demonstrated an association between gene SNPs and aspirin non-response, but these associations are controversial (Maguire et al., 2008). Certain miRNA is associated with aspirin non-response, but the exact change in metabolic pathway is unclear (Kok et al., 2016). By examining metabolic alteration, key metabolites can be identified, and proposed causes of aspirin non-response may be validated.i CT, clotting time; ATP, adenosine triphosphate; ADP, adenosine diphosphate; AMP, adenosine monophosphate. #High and low groups of each metabolite were determined by the median level of corresponding metabolite. The significant p values are marked by (p < 0.05). DISCUSSION The metabolism of platelet in patients with aspirin resistance was analyzed with this model, and the researchers predicted increased glyceraldehyde 3-phosphate dehydrogenase and triose phosphate isomerase, and decreased 1,6-bisphosphate aldolase and reactive oxygen species via glutathione related pathways. As a result, flux through the pentose phosphate pathway and through glycolysis pathway increased, as did purine metabolism increased. Glycine takes part in the above three pathways. Glycine could affect platelet membrane potential and hence aggregation capacity by binding to glycine-gated chloride channels. Animal studies showed that dietary glycine significantly increased bleeding time of rats and decreased the amplitude of platelet aggregation (Schemmer et al., 2013). This evidence may suggest a potentially important role of glycine in the physiology of platelet activity. Our findings are consistent with this hypothesis by showing that lower glycine may be associated with higher platelet activity under aspirin treatment (Supplementary Figure 3). On the other hand, the correlation between glycine levels and aspirin non- response among patients older than 65 years is not significant. This implies a different mechanism of aspirin non-response in patients of different ages. In elderly patients, comorbidities, such as diabetes mellitus or metabolic syndromes, are commonly presented, and this may contribute to aspirin non-response (Wurtz et al., 2014; Du et al., 2016). Among our study population, the percentage of DM among patients older than 65 years is higher than those younger than 65 years. The presence of these comorbidities may attenuate the impact of glycine. Patients with sufficient platelet glycine might still be aspirin non-responsive given their background condition. October 2019 \| Volume 10 \| Article 1107 Frontiers in Pharmacology \| www.frontiersin.org 6 Chiang et al. Platelets in Aspirin Non-Response committee (IRB number: 201106077RC). All patients gave written informed consent. This finding has important clinical implications since lower glycine might be a biomarker of aspirin non-response in younger patients, but not in the elderly.h There are some limitations in our study. Our study focused on hydrophilic metabolites of platelets, and glycine is the only metabolite that may be different between aspirin responders and non-responders in patients younger than 65 years. The hypothesized pathway change of aspirin non-response cannot be constructed. In addition, the small sample size potentially caused the statistical analysis to be underpowered. Since there is no research focused on this topic, our study might serve as a pilot study. FUNDING f An association between lower glycine levels and higher platelet activity in patients younger than 65 years may suggest an important role of glycine in the pathophysiology of aspirin non-response. The study was supported by the Ministry of Science and Technology (MST 102-2628-B-002-048-MY3 and MST 105-2628-B-002- 026-MY3) in Taiwan. DISCUSSION Further research on hydrophobic metabolites or circulating metabolites, or with a different isotope, such as 13C, may provide more insights into aspirin resistance. PERSPECTIVES SECTION By metabolomic analysis, changes in the small molecules of platelets in aspirin non-responders may be detected, providing insights into the mechanism of aspirin non-response, and potentially discovering new biomarkers for aspirin non-responders. ETHICS STATEMENT The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fphar.2019.01107/ full#supplementary-material The study complies with the declaration of Helsinki and the study protocol was approved by the institutional research ethics ACKNOWLEDGMENTS The authors thank the Core Facility for Protein Structural Analysis supported by National Core Facility Program for Biotechnology for NMR technical assistance and the staff of the 6th core laboratory of National Taiwan University Hospital. p p (i) In patients younger than 65 years, a lower glycine level was associated with shorter clotting times g (ii) Low glycine level may be a clinical biomarker for aspirin non-response in patients younger than 65 years. CONCLUSION In this pilot study, the hydrophilic metabolomic profile of platelets was not different between aspirin non-responders and responders. An association between lower glycine levels and higher platelet activity in patients younger than 65 years may suggest an important role of glycine in the pathophysiology of aspirin non-response. In this pilot study, the hydrophilic metabolomic profile of platelets was not different between aspirin non-responders and responders. AUTHOR CONTRIBUTIONS J-YChi, S-HL, J-MJ, and C-YL had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of data analysis. Study concept and design: J-MJ, C-YL, Y-CC, S-HL, C-KW, H-MY, S-FY, J-YChu, S-CL, and F-TC. Acquisition of data: C-AH, B-BL, P-CC, C-HC, J-YChu, S-CL, and H-JL. Expert statistical consultation: S-HL, Y-CC, and C-YL. Analysis and interpretation of data: J-YChi, S-HL, Y-CC, C-YL, J-MJ, J-YChu, S-CL, and F-TC. Critical revision of the manuscript for important intellectual content: J-YChi, S-HL, C-YL, J-MJ, J-YChu, S-CL, and F-TC. REFERENCES doi: 10.18632/oncotarget.5811 suppresses podoplanin-induced platelet aggregation and cancer metastasis. Oncotarget 6 (40), 42733–42748. doi: 10.18632/oncotarget.5811 Li, J., Song, M., Jian, Z., Guo, W., Chen, G., Jiang, G., et al. (2014). Laboratory aspirin resistance and the risk of major adverse cardiovascular events in patients with coronary heart disease on confirmed aspirin adherence. J. Atheroscler Thromb 21 (3), 239–247. doi: 10.5551/jat.19521 Chen, W. H., Lee, P. Y., Ng, W., Tse, H. F., and Lau, C. P. (2004). Aspirin resistance is associated with a high incidence of myonecrosis after non-urgent percutaneous coronary intervention despite clopidogrel pretreatment. J. Am. Coll. Cardiol. 43 (6), 1122–1126. doi: 10.1016/j.jacc.2003.12.034 Lin, C., Wu, H., Tjeerdema, R., and Viant, M. (2007). Evaluation of metabolite extraction strategies from tissue samples using NMR metabolomics. Metabolomics 3 (1), 55–67. doi: 10.1007/s11306-006-0043-1 Dormann J Elith, C. F., Bacher, S., Buchmann, C., Carl, G., Carré, G., García Marquéz, J. R., et al. (2013). Collinearity: a review of methods to deal with it and a simulation study evaluating their performance. Ecography 36, 027–046. doi: 10.1111/j.1600-0587.2012.07348.x Lindon, J. C., Holmes, E., and Nicholson, J. K. (2007). Metabonomics in pharmaceutical R&D. Febs J. 274 (5), 1140–1151. doi: 10.1111/j.1742- 4658.2007.05673.x Dretzke, J., Riley, R. D., Lordkipanidze, M., Jowett, S., O’Donnell, J., Ensor, J., et al. (2015). The prognostic utility of tests of platelet function for the detection of ‘aspirin resistance’ in patients with established cardiovascular or cerebrovascular disease: a systematic review and economic evaluation. Health Technol. Assess. 19 (37), 1–366. doi: 10.3310/hta19370 Maguire, J. M., Thakkinstian, A., Sturm, J., Levi, C., Lincz, L., Parsons, M., et al. (2008). Polymorphisms in platelet glycoprotein 1balpha and factor VII and risk of ischemic stroke: a meta-analysis. Stroke 39 (6), 1710–1716. doi: 10.1161/ STROKEAHA.107.507228 Oberacher, H., Arnhard, K., Linhart, C., Diwo, A., Marksteiner, J., and Humpel, C. (2017). Targeted metabolomic analysis of soluble lysates from platelets of patients with mild cognitive impairment and alzheimer’s disease compared to healthy controls: is PC aeC40:4 a promising diagnostic tool? J. Alzheimers Dis. 57 (2), 493–504. doi: 10.3233/JAD-160172 Du, G., Lin, Q., and Wang, J. (2016). A brief review on the mechanisms of aspirin resistance. Int. J. Cardiol. 220, 21–26. doi: 10.1016/j.ijcard.2016.06.104 Eikelboom, J. W., Hirsh, J., Weitz, J. I., Johnston, M., Yi, Q., and Yusuf, S. (2002). Aspirin-resistant thromboxane biosynthesis and the risk of myocardial infarction, stroke, or cardiovascular death in patients at high risk for cardiovascular events. Circulation 105 (14), 1650–1655. doi: 10.1161/01. REFERENCES CIR.0000013777.21160.07 Paglia, G., Sigurjonsson, O. E., Rolfsson, O., Valgeirsdottir, S., Hansen, M. B., Brynjolfsson, S., et al. (2014). Comprehensive metabolomic study of platelets reveals the expression of discrete metabolic phenotypes during storage. Transfusion 54 (11), 2911–2923. doi: 10.1111/trf.12710 Ellis, J. K., Athersuch, T. J., Cavill, R., Radford, R., Slattery, C., Jennings, P., et al. (2011). Metabolic response to low-level toxicant exposure in a novel renal tubule epithelial cell system. Mol. Biosyst. 7 (1), 247–257. doi: 10.1039/ C0MB00146E Paglia, G., Sigurjonsson, O. E., Rolfsson, O., Hansen, M. B., Brynjolfsson, S., Gudmundsson, S., et al. (2015). Metabolomic analysis of platelets during storage: a comparison between apheresis- and buffy coat-derived platelet concentrates. Transfusion 55 (2), 301–313. doi: 10.1111/trf.12834 Foussas, S. G., Zairis, M. N., Tsirimpis, V. G., Makrygiannis, S. S., Patsourakos, N. G., Adamopoulou, E. N., et al. (2009). The impact of aspirin resistance on the long-term cardiovascular mortality in patients with non-ST segment elevation acute coronary syndromes. Clin. Cardiol. 32 (3), 142–147. doi: 10.1002/clc.20293 Reny, J. L., De Moerloose, P., Dauzat, M., and Fontana, P. (2008). Use of the PFA-100 closure time to predict cardiovascular events in aspirin-treated cardiovascular patients: a systematic review and meta-analysis. J. Thromb Haemost. 6 (3), 444– 450. doi: 10.1111/j.1538-7836.2008.02897.x Schemmer, P., Zhong, Z., Galli, U., Wheeler, M. D., Xiangli, L., Bradford, B. U., et al. (2013). Glycine reduces platelet aggregation. Amino Acids 44 (3), 925– 931. doi: 10.1007/s00726-012-1422-8 Frelinger, A. L., 3rd, Li, Y., Linden, M. D., Barnard, M. R., Fox, M. L., Christie, D. J., et al. (2009). Association of cyclooxygenase-1-dependent and -independent platelet function assays with adverse clinical outcomes in aspirin-treated patients presenting for cardiac catheterization. Circulation 120 (25), 2586– 2596. doi: 10.1161/CIRCULATIONAHA.109.900589 Simpson, S. H., Abdelmoneim, A. S., Omran, D., and Featherstone, T. R. (2014). Prevalence of high on-treatment platelet reactivity in diabetic patients treated with aspirin. Am. J. Med. 127 (1), 95.e1–95.e9. doi: 10.1016/j.amjmed.2013.09.019 Gowda, G. A., Zhang, S., Gu, H., Asiago, V., Shanaiah, N., and Raftery, D. (2008). Metabolomics-based methods for early disease diagnostics. Expert Rev. Mol. Diagn. 8 (5), 617–633. doi: 10.1586/14737159.8.5.617 Stejskal, D., Vaclavik, J., Lacnak, B., and Proskova, J. (2006). Aspirin resistance measured by cationic propyl gallate platelet aggregometry and recurrent cardiovascular events during 4 years of follow-up. Eur. J. Intern. Med. 17 (5), 349–354. doi: 10.1016/j.ejim.2006.01.006 Gum, P. A., Kottke-Marchant, K., Poggio, E. D., Gurm, H., Welsh, P. A., Brooks, L., et al. (2001). REFERENCES Profile and prevalence of aspirin resistance in patients with cardiovascular disease. Am. J. Cardiol. 88 (3), 230–235. doi: 10.1016/ S0002-9149(01)01631-9 Szczeklik, A., Undas, A., Sanak, M., Frolow, M., and Wegrzyn, W. (2000). Relationship between bleeding time, aspirin and the PlA1/A2 polymorphism of platelet glycoprotein IIIa. Br. J. Haematol. 110 (4), 965–967. doi: 10.1046/j.1365-2141.2000.02267.xh Gum, P. A., Kottke-Marchant, K., Welsh, P. A., White, J., and Topol, E. J. (2003). A prospective, blinded determination of the natural history of aspirin resistance among stable patients with cardiovascular disease. J. Am. Coll. Cardiol. 41 (6), 961–965. doi: 10.1016/S0735-1097(02)03014-0 Thomas, A., Rahmanian, S., Bordbar, A., Palsson, B. O., and Jamshidi, N. (2014). Network reconstruction of platelet metabolism identifies metabolic signature for aspirin resistance. Sci. Rep. 4, 3925. doi: 10.1038/srep03925 Hastie, T., Tibshirani, R., and Friedman, J. (2009). The Elements of Statistical Learning. Switzerland: Springer. doi: 10.1007/978-0-387-84858-7 Tiziani, S., Lopes, V., and Gunther, U. L. (2009). Early stage diagnosis of oral cancer using 1H NMR-based metabolomics. Neoplasia 11 (3), 269–276. 4p following. doi: 10.1593/neo.81396 Jilma, B., and Fuchs, I. (2001). Detecting aspirin resistance with the platelet function analyzer (PFA-100). Am. J. Cardiol. 88 (11), 1348–1349. doi: 10.1016/ S0002-9149(01)02143-9 van den Berg, R. A., Hoefsloot, H. C. J., Westerhuis, J. A., Smilde, A. K., and van der Werf, M. J. (2006). Centering, scaling, and transformations: improving the biological information content of metabolomics data. BMC Genomics 7. doi: 10.1186/1471-2164-7-142 Johnson, C. H., Ivanisevic, J., and Siuzdak, G. (2016). Metabolomics: beyond biomarkers and towards mechanisms. Nat. Rev. Mol. Cell Biol. 17 (7), 451–459. doi: 10.1038/nrm.2016.25 Viant, M. R. (2003). Improved methods for the acquisition and interpretation of NMR metabolomic data. Biochem. Biophys. Res. Commun. 310 (3), 943–948. doi: 10.1016/j.bbrc.2003.09.092 Kok, M. G., Mandolini, C., Moerland, P. D., de Ronde, M. W., Sondermeijer, B. M., Halliani, A., et al. (2016). Low miR-19b-1-5p expression in isolated platelets after aspirin use is related to aspirin insensitivity. Int. J. Cardiol. 203, 262–263. doi: 10.1016/j.ijcard.2015.10.098 Viant, M. R., Pincetich, C. A., and Tjeerdema, R. S. (2006). Metabolic effects of dinoseb, diazinon and esfenvalerate in eyed eggs and alevins of Chinook salmon (Oncorhynchus tshawytscha) determined by 1H NMR metabolomics. Aquat. Toxicol. 77 (4), 359–371. doi: 10.1016/j.aquatox.2006.01.009h Krasopoulos, G., Brister, S. J., Beattie, W. S., and Buchanan, M. R. (2008). Aspirin “resistance” and risk of cardiovascular morbidity: systematic review and meta- analysis. BMJ 336 (7637), 195–198. doi: 10.1136/bmj.39430.529549.BE Li, Q., Chen, B. REFERENCES report from the american heart association. Circulation 137 (12), e67–e492. doi: 10.1161/CIR.0000000000000558 report from the american heart association. Circulation 137 (12), e67–e492. doi: 10.1161/CIR.0000000000000558 Angiolillo, D. J., Fernandez-Ortiz, A., Bernardo, E., Ramirez, C., Sabate, M., Jimenez-Quevedo, P., et al. (2006). Influence of aspirin resistance on platelet function profiles in patients on long-term aspirin and clopidogrel after percutaneous coronary intervention. Am. J. Cardiol. 97 (1), 38–43. doi: 10.1016/j.amjcard.2005.07.106 Bishop-Bailey, D. (2009). The platelet as a model system for the acute actions of nuclear receptors. Steroids 75 (8–9), 570–575. doi: 10.1016/j.steroids.2009.09.005 Bligh, E. G., and Dyer, W. J. (1959). A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37 (8), 911–917. doi: 10.1139/ Bligh, E. G., and Dyer, W. J. (1959). A rapid method of total lipid extraction Bligh, E. G., and Dyer, W. J. (1959). A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37 (8), 911–917. doi: 10.1139/ y59-099 Antithrombotic Trialists' Collaboration. (2002). Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients. BMJ 324 (7329), 71–86. doi: 10.1136/bmj.324.7329.71fi Bollard, M. E., Keun, H. C., Beckonert, O., Ebbels, T. M., Antti, H., Nicholls, A. W., et al. (2005). Comparative metabonomics of differential hydrazine toxicity in the rat and mouse. Toxicol. Appl. Pharmacol. 204 (2), 135–151. doi: 10.1016/j. taap.2004.06.031 Azmi, J., Connelly, J., Holmes, E., Nicholson, J. K., Shore, R. F., and Griffin, J. L. (2005). Characterization of the biochemical effects of 1-nitronaphthalene in rats using global metabolic profiling by NMR spectroscopy and pattern recognition. Biomarkers 10 (6), 401–416. doi: 10.1080/13547500500309259 Brindle, J. T., Antti, H., Holmes, E., Tranter, G., Nicholson, J. K., Bethell, H. W., et al. (2002). Rapid and noninvasive diagnosis of the presence and severity of coronary heart disease using 1H-NMR-based metabonomics. Nat. Med. 8 (12), 1439–1444. doi: 10.1038/nm1202-802 Chang, Y. W., Hsieh, P. W., Chang, Y. T., Lu, M. H., Huang, T. F., Chong, K. Y., et al. (2015). Identification of a novel platelet antagonist that binds to CLEC-2 and Benjamin, E. J., Virani, S. S., Callaway, C. W., Chamberlain, A. M., Chang, A. R., Cheng, S., et al. (2018). Heart disease and stroke statistics-2018 update: a October 2019 \| Volume 10 \| Article 1107 Frontiers in Pharmacology \| www.frontiersin.org Chiang et al. Platelets in Aspirin Non-Response suppresses podoplanin-induced platelet aggregation and cancer metastasis. Oncotarget 6 (40), 42733–42748. REFERENCES L., Ozdemir, V., Ji, W., Mao, Y. M., Wang, L. C., et al. (2007). Frequency of genetic polymorphisms of COX1, GPIIIa and P2Y1 in a Chinese population and association with attenuated response to aspirin. Pharmacogenomics 8 (6), 577–586. doi: 10.2217/14622416.8.6.577 Weng, Z., Li, X., Li, Y., Lin, J., Peng, F., and Niu, W. (2013). The association of four common polymorphisms from four candidate genes (COX-1, COX-2, ITGA2B, ITGA2) with aspirin insensitivity: a meta-analysis. PLoS One 8 (11), e78093. doi: 10.1371/journal.pone.0078093 October 2019 \| Volume 10 \| Article 1107 Frontiers in Pharmacology \| www.frontiersin.org Platelets in Aspirin Non-Response Chiang et al. Westerhuis, J. A., Hoefsloot, H. C. J., Smit, S., Vis, D. J., Smilde, A. K., Velzen, E. J. J., et al. (2008). Assessment of PLSDA cross validation. Metabolomics 4 (1), 81–89. doi: 10.1007/s11306-007-0099-6 Xia, J., Mandal, R., Sinelnikov, I. V., Broadhurst, D., and Wishart, D. S. (2012). MetaboAnalyst 2.0-a comprehensive server for metabolomic data analysis. Nucleic Acids Res. 40 (W1), W127–WW33. doi: 10.1093/nar/gks374 Zimring, J. C., Slichter, S., Odem-Davis, K., Felcyn, J. R., Kapp, L. M., Bell, L. N., et al. (2016). Metabolites in stored platelets associated with platelet recoveries and survivals. Transfusion 56 (8), 1974–1983. doi: 10.1111/trf.13631 Worley, B., and Powers, R. (2013). Multivariate analysis in metabolomics. Curr. Metabolomics 1 (1), 92–107. doi: 10.2174/2213235X130108f Wu, C. K., Lin, J. W., Caffrey, J. L., Chang, M. H., Hwang, J. J., Lin, Y. S. (2010). Cystatin C and long-term mortality among subjects with normal creatinine-based estimated glomerular filtration rates: NHANES III (Third National Health and Nutrition Examination Survey). J. Am. Coll. Cardiol. 56 (23), 1930–1936. Conflict of Interest: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Wu, C. K., Tsai, H. Y., Su, M. M., Wu, Y. F., Hwang, J. J., Lin, J. L., et al. (2017). Evolutional change in epicardial fat and its correlation with myocardial diffuse fibrosis in heart failure patients. J. Clin. Lipidol. 11 (6), 1421–1431. Copyright © 2019 Chiang, Lee, Chen, Wu, Chuang, Lo, Yeh, Yeh, Hsu, Lin, Chang, Chang, Liang, Chiang, Lin and Juang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). October 2019 \| Volume 10 \| Article 1107 REFERENCES The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. i Wurtz, M., Kristensen, S. D., Hvas, A. M., and Grove, E. L. (2012). Pharmacogenetics of the antiplatelet effect of aspirin. Curr. Pharm. Des. 18 (33), 5294–5308. doi: 10.2174/138161212803251907t Wurtz, M., Hvas, A. M., Jensen, L. O., Kaltoft, A. K., Tilsted, H. H., Kristensen, S. D., et al. (2014). 24-hour antiplatelet effect of aspirin in patients with previous definite stent thrombosis. Int. J. Cardiol. 175 (2), 274–279. doi: 10.1016/j. ijcard.2014.05.013 October 2019 \| Volume 10 \| Article 1107 Frontiers in Pharmacology \| www.frontiersin.org 9
https://openalex.org/W2727286713	https://www.frontiersin.org/articles/10.3389/fpsyg.2017.01106/pdf	English	null	An Examination of the Four-Part Theory of the Chinese Self: The Differentiation and Relative Importance of the Different Types of Social-Oriented Self	Frontiers in psychology	2,017	cc-by	14,180	HYPOTHESIS AND THEORY published: 30 June 2017 doi: 10.3389/fpsyg.2017.01106 An Examination of the Four-Part Theory of the Chinese Self: The Differentiation and Relative Importance of the Different Types of Social-Oriented Self Chien-Ru Sun* Department of Psychology, National Chengchi University, Taipei, Taiwan Chien-Ru Sun* Department of Psychology, National Chengchi University, Taipei, Taiwan Chien Ru Sun Department of Psychology, National Chengchi University, Taipei, Taiwan Department of Psychology, National Chengchi University, Taipei, Taiwan Because culture has a deep and far-reaching inﬂuence, individuals who grew up within different cultures tend to develop different basic self-constructions. With respect to the Chinese under the inﬂuence of Chinese culture, Yang proposed the concepts of individual-oriented self and social-oriented self. He argued that, besides the individual-oriented self, the social-oriented self of the Chinese contains three types of self: the relationship-oriented self, the familistic (group)-oriented self, and the other-oriented self. The theory proposed that the Chinese self is appropriately covered only through this four-part theory of the Chinese self. However, this remains to be tested; whether these three types of sub-level “selves” can be effectively triggered, along with their relative importance. This study examines the four-part theory of the Chinese self. Through photo priming, Experiment 1 shows that the three types of social-oriented self are differentiated from each other and can be individually triggered. In Experiment 2, the importance of the three types of self was investigated, adopting the concept of limited self-regulation resources to design scenarios. The participants were asked to make counterarguments about the notion of each of the three types of self, with performance in the subsequent task serving as the main dependent variable. In Experiment 3, the relative importance of the three types of self was examined by investigating the choices made by individuals within the context of conﬂict under the three orientations of the social-oriented self. Overall, results of the experiments showed that the Chinese have a four-part self with the importance of the other-oriented self as the most remarkable. Edited by: Edited by: George Christopoulos, Nanyang Technological University, Singapore Edited by: George Christopoulos, Nanyang Technological University, Singapore Reviewed by: Angela Leung, Singapore Management University, Singapore Xiaoxiao Liu, Xiamen University, China Reviewed by: Angela Leung, Singapore Management University, Singapore Xiaoxiao Liu, Xiamen University, China Correspondence: Chien-Ru Sun chienru@nccu.edu.tw Specialty section: This article was submitted to Cultural Psychology, a section of the journal Frontiers in Psychology Received: 31 January 2017 Accepted: 15 June 2017 Published: 30 June 2017 Keywords: the four-part theory, Chinese self, social-oriented self, individual-oriented self, other-oriented self, the familistic (group)-oriented self, relationship-oriented self Citation: Subsequent cross-cultural studies have also noted that the impact of culture is both deep and far-reaching and that individuals who developed in diﬀerent cultures have long been molded into diﬀerent shapes. For example, Markus and Kitayama (1991, 2010) argued that the Western concept of self is a concept of independent self that emphasizes the self’s independence and uniqueness, believing that individuals must discover and demonstrate inherent personal qualities to distinguish from others. In contrast, individuals under the inﬂuence of Eastern cultures primarily aspire to an interdependent self that is both ﬂexible and subject to change, stressing coordination between an individual and the social environment through appropriate personal behavior. Individuals with an interdependent self tend to seek a harmonious relationship with others and hope to help others achieve their goals. In addition to analysis and research at the theoretical level, Sun and Wang’s (2005) experiment found that among Chinese, the social-oriented self was the main source of positive self- evaluation among individuals because when the social-oriented self is threatened, the aﬃrmation of important relations (e.g., with parents) that also belong to the relationship-oriented self can restore the balance of an individual’s self, however, self- aﬃrmation originating from the independence-oriented self seems irrelevant to restoring the balance of the social-oriented self. Sun (2004) also employed the research paradigm of “false uniqueness bias” and the “self-handicapping paradigm” and demonstrated that the Chinese did not show obvious modesty or self-eﬀacing tendencies, and the tendency toward self- enhancement demonstrated in the social-oriented self was more remarkable than that demonstrated in the individual-oriented self. Lastly, Kurman (2001) showed that the Chinese manifest self-enhancement in their communal traits. The results of these studies largely show that among the Chinese, the social-oriented self is more important than the individual-oriented self. Markus and Kitayama (1991, 2010) argued that these two diﬀerent self-construals—i.e., two diﬀerent basic self-schemas— would cause changes in how an individual assesses, organizes, and regulates his/her own experience and behavior. Furthermore, Singelis (1994) argued that although the “self” as conceived by Westerners is dominated by the independent self, while the “self” as conceived by Asians is dominated by the interdependent self, it is not necessarily true that the independent self and the interdependent self cannot coexist. Citation: Sun C-R (2017) An Examination of the Four-Part Theory of the Chinese Self: The Differentiation and Relative Importance of the Different Types of Social-Oriented Self. Front. Psychol. 8:1106. doi: 10.3389/fpsyg.2017.01106 Sun C-R (2017) An Examination of the Four-Part Theory of the Chinese Self: The Differentiation and Relative Importance of the Different Types of Social-Oriented Self. Front. Psychol. 8:1106. doi: 10.3389/fpsyg.2017.01106 When investigating the diﬀerences in individuals’ psychology or behavior and the interpersonal interaction process, most studies have made generalized comparisons and engaged in discussions from the cultural perspective (Triandis, 1995). Individuals inﬂuenced by an individualistic culture mostly focus on the realization of personal goals, the protection of personal interests, and the individual’s independence and privacy, believing that individuals ought to seek self-suﬃciency and June 2017 \| Volume 8 \| Article 1106 Frontiers in Psychology \| www.frontiersin.org The Four-Part Theory of the Chinese Self Sun avoid relying on others. Typical representatives of such cultures include those of some American and European countries such as the United States, Britain, and Canada. In contrast, the impact of a collectivist culture is very diﬀerent: in such cultures, individuals emphasize their responsibilities to the group or others and believe that people need to support each other. Within a collective culture, helping either others or the group to achieve is even more important than self-actualization (Dion and Dion, 1993; Triandis, 1995). Examples of countries in which a collectivist culture prevails are Latin American and Asian countries such as Venezuela, China, and Peru. emphasizing that this type of self which attaches importance in order to maintain harmonious interpersonal relationships, accountability, and responsibility and requires appropriate personal behaviors such that individuals position themselves according to their relationship with others. Yang (1993) also stressed that these two types of “self” can co-exist; in essence, within the Chinese culture, the importance of the social-oriented self cannot be ignored. This theory was supported later with the studies made by Lu et al. (2008) and Lu (2008), which they argued that the Chinese have a bicultural self, wherein that both the individual-oriented self and the social- oriented self are important. Later on, Yang et al. (2010) further attempted to understand and construct the bicultural self of the Chinese from the perspective of an individual’s development stages and have gained some initial support for their notion. Citation: Many subsequent studies have agreed with this view; For example, Kühnen and Oyserman (2002) demonstrated that they could eﬀectively trigger diﬀerent types of self in an individual and that when an individual’s interdependent self was triggered, the individual tended to observe the surrounding context. However, when the individual’s independent self was triggered, he/she was inclined to focus on the task itself. YANG’S FOUR-PART THEORY OF THE CHINESE SELF Does the distinction between the individual-oriented self and the social-oriented self suggest that complete understanding of the Chinese self has been achieved? The answer is no. In 2004, based on his previously proposed theory of the relation between the individual-oriented self-and the social-oriented self, Yang further established “the four-part theory of the Chinese self,” emphasizing that the social-oriented self contains three diﬀerent types of self [relationship-oriented self, familistic (group)-oriented self, and other-oriented self], and therefore, the four-part theory can completely cover the Chinese self (Yang, 2004). THE CHINESE SELF The so-called familistic (group)-oriented self refers to the interaction relations of an individual with his or her clan and family both within and without the clan (or family), whose interaction history and connotation is based on Chinese familism. Yang also emphasizes that this type of familistic (group)-oriented interaction can be either subject to generalization or transferred to groups outside the family (e.g., work or business organizations); accordingly, family orientation also represents group orientation. Yang also proposed the so-called “other” orientation, referring to interaction relations with non-speciﬁc others under certain circumstances in which “non-speciﬁc others” are deﬁned as a large number of anonymous “generalized others” with unknown faces. For example, within the sayings “too shameful to meet the clansmen” and “I wonder what others would think”, the “clansmen” and “others” are both “non-speciﬁc others” as deﬁned by Yang. At such a moment, the individual self is related to these generalized others and is called the other-oriented self. Yang (2004) believes that within the context of Chinese culture, “the relationship between two persons under the equality context, the relationship between two persons under authoritarian context, the communal context of family and the generalized context of others are the most core interactive modalities in the daily life of Chinese people and make up the main parts of life in the Chinese society. The characteristics, connotations and operation principles are diﬀerent for the aforementioned four modalities, and the modalities have their respective interactive ways and develop into four corresponding orientations over time, i.e., relationship orientation, authoritarian orientation, familistic (group) orientation and “other” orientation” (p. 22). Yang also believes that in terms of an individual’s self-operation, Chinese people’s four major social life modalities can be regarded as major interaction modalities in which the I-self assumes various functions and implements various acts, whereas individuals living in the modalities blend in from young to old and are naturally capable of skillfully and eﬃciently operating in appropriate social interaction modes after undergoing the processes of socialization, diﬀerentiation, and even automation. Yang (2004) also stresses both that the contexts under which diﬀerent types of self form are diﬀerent and that the individual’s needs/motives vary. Consequently, the ultimate self- evaluation aspired by the individual will also be diﬀerent. The individual-oriented self is decontextualized, attaches importance to autonomy and independence and looks forward to making achievements through the individual’s ability and performance. THE CHINESE SELF The relationship-oriented self mainly uses relationships between two persons as the context, attaching importance to each person’s roles and hoping to gain the other’s recognition and acceptance. The familistic (group)-oriented self uses family (group) as the context in which the individual expects to play an appropriate role and be accepted by the family (group) to meet his or her identity-related needs. The other-oriented self cares about non- speciﬁc others and hopes that the individual’s words and deeds can satisfy public expectations and be honored, earning a face before the public. p Therefore, Yang (2004) stresses that for Chinese people, “in the process of self-development, the individual’s performance and interaction eﬀectiveness in the four modalities are the main objectives or object for the Chinese I-self to observe, inspect, perceive, think, reﬂect, judge, evaluate, plan, organize, control, manipulate, adjust, and correct, thereby giving rise to the four types of Me-self” (p. 21). For reasons of brevity, Yang combines the relationship orientation and authoritarian orientation into one that retains the name “relationship orientation.” In other words, the social-oriented self can be subdivided into three types of self—i.e., the relationship-oriented self, the familistic (group)- oriented self, and the other-oriented self—which together with the individual-oriented self comprise the four-part Chinese self. Yang argues that these four types of self represent Chinese people’s methods of interacting with others in the four modalities, which are mutually distinctive. Yang further compares the similarities and diﬀerences among the four types of self with respect to the following 15 psychological aspects: (1) Dominant trend of adaptation; (2) Target of interaction; (3) Contextualization; (4) Role involvement; (5) Object of identiﬁcation; (6) Type of identity; (7) Sense of responsibility; (8) Mode of self-consistency; (9) Primary motivation; (10) Essential Indeed, this concept of the multi-part self is not unique to Yang: Greenwald and Breckler (1985), Sedikides and Brewer (2001), and Cross et al. (2003) have also proposed the idea of three types of self. The three types of self that they propose include the individual or private self, the relational self and the collective self. The individual or private self refers to characteristics unique to the individual, the relational self refers to the self-shared by the individual with important others and the collective self is the part of self that spans a group. THE CHINESE SELF What is the self-construction of the Chinese under the inﬂuence of the Confucian culture? The connotations of the individual- and social-oriented self as proposed by Yang (1993, 1995) largely correspond to the concepts of independent self and interdependent self, respectively. Yang (1993) further deﬁned the individual-oriented self as a combination of a tendency toward high personal autonomy and low homonymy that emphasizes an individual’s personal achievement, performance, uniqueness and autonomy. At the meantime, Yang deﬁned the social-oriented self as one that combines a tendency toward high homonymy (with the surrounding environment) and low personal autonomy, In previous studies on self, Western psychologists have regarded the subjective self (i.e., I-self) as an observer, an information handler or a cognition constructor while regarding the objective self (i.e., Me-self) as either the object to be observed/perceived or the object of cognition construction (e.g., Harter, 1999). This notion stresses that the I-self possesses perception, cognition, and evaluation functions, whereas the June 2017 \| Volume 8 \| Article 1106 Frontiers in Psychology \| www.frontiersin.org 2 The Four-Part Theory of the Chinese Self Sun Me-self is only the object or target that receives these functions. On the one hand, Yang (2004) believes that this notion is narrow- minded; on the other hand, he emphasizes that “for people living in countries with starkly diﬀerent historical, social, and cultural background, the principles, the ways and the contents adopted by the I-self when assuming the functions or playing the roles can be vastly diﬀerent, so the resultant Me-self can also be very diﬀerent” (p. 21). Yang further stresses that Western society has primarily been inﬂuenced by Christian culture, whereas Chinese society has been deeply inﬂuenced by Confucian culture. The two societies have drastically diﬀerent historical, social and cultural aspects, resulting in signiﬁcant diﬀerences between Westerners and Chinese with respect to the I-self and the Me- self. aﬀection; (11) Target of emotional attachment; (12) Type of self- actualization; (13) Type of self-concept; (14) Type of self-esteem; and (15) Type of happiness (Yang, 2004, see Table 1). yp pp g Overall, Yang’s so-called individual-oriented self refers to the demonstration and play of the characteristics and features that an individual uniquely possesses, whereas the relationship- oriented self emphasizes interaction relations in interpersonal modalities such as horizontal modality (e.g., husband and wife) and vertical modality (e.g., parents and children). June 2017 \| Volume 8 \| Article 1106 THE CHINESE SELF Yang argues that these discussions of self are inadequate because distinctions of selves are deﬁned only based on the number of people (one person, two or more persons) with which the self is involved: they neither are truly rooted in the local culture nor consider June 2017 \| Volume 8 \| Article 1106 Frontiers in Psychology \| www.frontiersin.org 3 The Four-Part Theory of the Chinese Self Sun TABLE 1 \| Summary of a systematic conceptual comparison of individual- and social-oriented selves in terms of 15 major aspects. Social-oriented Self Aspect Individual-oriented self Relationship-oriented self Familistic(group)-oriented self Other-oriented self 1. Dominant trend of adaptation Autonomous Homonomous (union with another person) Homonomous (union with one’s family or some other membership group) Homonomous (union with the non-speciﬁc others) 2. Target of interaction One’s own self or person A related person One’s family or some other membership group Non-speciﬁc others (the generalized other) 3. Contextualization Decontextualization Relationship-contextualized Family (group)-contextualized Non-speciﬁc-others-contextualized 4. Role involvement Moderate social role involvement High social role involvement in a relationship High social role involvement in one’s family or some other membership Moderate social role involvement in the non-speciﬁc others 5. Object of identiﬁcation Identiﬁcation with oneself (reﬂexive identiﬁcation) Identiﬁcation with a related person (relational identiﬁcation) Identiﬁcation with one’s family or some other membership group (familial or group identiﬁcation) Diffuse social identiﬁcation 6. Type of identity Personal identity Relational identity Familial (group) identity Diffuse social identity 7. Sense of responsibility Personal responsibility Reciprocal responsibility for the role partner in a speciﬁc relationship Familistic responsibility for one’s family or some other membership group One-sided, self-imposed responsibility for the non-speciﬁc others 8. Mode of self-consistency Self-consistency across situations and time (self-centered self-consistency) Self-consistency with respect to speciﬁc dyadic relationships (relationship-speciﬁc self-consistency) Self-consistency with respect to one’s family or some other speciﬁc social group (group-speciﬁc self-consistency) Self-consistency with respect to the non-speciﬁc others (other-oriented self-consistency) 9. Primary motivation Agency needs (e.g., needs for self-reliance, autonomy, independence, self-acceptance, personal achievement, personal efﬁcacy, personal superiority, individual-oriented self-esteem, and self-enhancement) Relational needs (e.g., needs for dependence, interdependence, relational acceptance, relational sharing, relational face, relationship-oriented self-esteem, and self-enhancement) Belongingness needs (e.g., needs for self extension, a larger self, group acceptance, group protection, group efﬁcacy, group achievement, group glory, group face, familistic(group)-oriented sell-esteem, and self-enhancement) Face-related needs (e.g., needs for public face, reputation, respect from the non-speciﬁc others, other-oriented self-esteem, and self-enhancement) 10. June 2017 \| Volume 8 \| Article 1106 Frontiers in Psychology \| www.frontiersin.org THE CHINESE SELF Essential affection Narcissistic affects (e.g., self-love, self-respect, self-envy, self-glory, hedonic pleasure, personal happiness, feelings of individual-oriented self-esteem, and self-enhancement) Role-speciﬁc dyadic affects (e.g., ﬁlial love, marital love, brotherly love, affects between teacher and pupil, feelings of gaining and losing relational face, feelings of relationship-oriented self-esteem and self-enhancement, feeling of relational shame) Familistic (group)-oriented affects (e.g., familial love, group glory, feelings of belongingness, unity, and security, feelings of gaining and losing group face, feelings of familistic(group)-oriented self-esteem and self-enhancement, feeling of group shame) Other-oriented affects (e.g., feelings of gaining and losing face, reputation, and respect from others, feelings of other-oriented self-esteem and self-enhancement, feeling of public shame) 11. Target of Emotional attachment Attachment to one’s self Attachment to another person as the role partner in a speciﬁc relationship Attachment to one’s family or some other membership group Attachment to the non-speciﬁc others 12. Kind of self-actualization Individual-oriented self-actualization Relationship-oriented self-actualization Familistic (group)-oriented self-actualization Other-oriented self-actualization 13. Kind of self-concept Individual-oriented self-concept Relationship-oriented self-concept Familistic (group)-oriented self-concept Other-oriented self-concept 14. Kind of self-esteem Individual-oriented self-esteem Relationship-oriented self-esteem Familistic (group)-oriented self-esteem Other-oriented self-esteem 15. Kind of well-being Individual-oriented well-being Relationship-oriented well-being Familistic (group)-oriented well-being Other-oriented well-being Frontiers in Psychology \| www.frontiersin.org 4 June 2017 \| Vo June 2017 \| Volume 8 \| Article 1106 4 The Four-Part Theory of the Chinese Self Sun the characteristics of the culture pertaining to the self being discussed. Chinese people deeply and properly but also distinguishing between the Chinese self and the Western self. However, except for the two previous studies authored by Yang (Weng and Yang, 2003; Lu and Yang, 2005), other studies focus exclusively on the diﬀerences between the individual-oriented self and the social-oriented self (e.g., Sun and Wang, 2005), failing to distinguish among the three types of social-oriented self—i.e., the relationship-oriented self, the familistic (group)-oriented self, and the other-oriented self. Yang’s four-part theory of the Chinese self is founded on Chinese culture and supported by results of some preliminary studies; for example, Lu and Yang (2005) note that the achievement to which Chinese people aspire is rather diﬀerent from that of the Westerners and should be divided into individual-oriented self-actualization and social-oriented self- actualization. THE CHINESE SELF In developing the scale of individual-oriented self- actualization and social-oriented self-actualization, Yang and Lu (2005) attempt to include items on diﬀerent orientations of the self [including the relationship-oriented self, the family (group)-oriented self and the other-oriented self], ﬁnding that the essential connotations of Chinese people’s self-actualization had three ﬁelds, i.e., “to become oneself completely,” “repay the family with personal achievements,” and “expending personal well-being to serve the community.” The ﬁrst of these ﬁelds belonged to the individual-oriented self, whereas the second and third ﬁelds clearly belonged to the self-actualization of the social- oriented self. Although these results do not fully echo the concept of the four-part Chinese self, a step forward has indeed been taken. In this study, we wanted to test (both experimentally and directly) diﬀerent types of self. In particular, we wanted to distinguish among diﬀerent types of self under the social- oriented self, to investigate (1) how to eﬀectively trigger diﬀerent types of self, especially to distinguish the relationship-oriented self, the familistic (group)-oriented and the other-oriented self; and (2) the relative importance of the three types of self—i.e., the relationship-oriented self, the familistic (group)-oriented, and the other-oriented self. “Non-speciﬁc others” involved in the other-oriented self are particularly noteworthy. The “non- speciﬁc others” proposed by Yang (1995, 2004) are a large number of anonymous and generalized other persons, whereas what is intended by the “other-oriented self” is that the individual’s words and deeds satisfy the expectation of these non-speciﬁc others, be honored and earn a face in public. So-called “non- speciﬁc others” hint at a larger number of people that must observe the same cultural or social norms as the individual. In other words, when we say, “I do not know what others would think,” those “others” should be others who share our cultural or social norms. In this study, we will therefore use this concept as the basis of discussion and our research design. Attempts related to the four-part theory of the Chinese self have been made in developing self-esteem scales; Weng and Yang (2003) attempted to conduct a conceptual analysis and scale development for social-oriented self-esteem and individual- oriented self-esteem in which the social-oriented self-esteem is divided into relationship-oriented self-esteem, familistic (group)- oriented self-esteem, and other-oriented self-esteem, resulting in four types of self-esteem when individual-oriented self- esteem is added. THE CHINESE SELF Based on the four-part Chinese self, Weng and Yang (2003) developed pre-test questionnaires on “The multi-part self-esteem scale for Chinese people” to investigate undergraduates from both Taiwan and Mainland China. The data obtained generated six oblique factors: “personal ability and independence,” “physical health and appearance,” “interpersonal relations and popularity,” “emotions and interactions of family members,” “family background and economy,” and “social identity and care.” The ﬁrst and second factors pertain to individual-oriented self-esteem (especially the ﬁrst factor), the third factor belongs to relationship-oriented self-esteem, the fourth and ﬁfth factors belong to family-oriented self-esteem, and the sixth factor belongs to other-oriented self-esteem. They integrate six subscales of self-esteem into the formal “Multi-part self-esteem scale of Chinese people” (Weng and Yang, 2003; Yang, 2004). In addition, the relationship-oriented self proposed by Yang was originally deﬁned as an intimate dyadic relationship such as the relationship between husband and wife, parent and child, etc. In this theory, the primary motive of satisfaction that is expected from the relationship-oriented self is that of mutual support, mutual reception, and sharing. Because we believe this deﬁnition is more appropriate when referring to equal-status relationships, we ﬁrst deﬁne the relationship-oriented self as the self in the relationship between spouses. However, according to Yang’s theory, relationships such as that between parent and child can also be viewed as involving the relationship-oriented self and shows some overlap with the familistic (group)-oriented self when an attempt is made to diﬀerentiate the two. Therefore, our basic idea is to use the distinctive “spouse/lover” relationship for discussing relationship-oriented self. If such a relationship is indistinct from that implicated by the familistic (group)-oriented self, then the four-part concept of Chinese self needs to be modiﬁed. Frontiers in Psychology \| www.frontiersin.org June 2017 \| Volume 8 \| Article 1106 Photos triggering diﬀerent types of self Photos triggering diﬀerent types of self In this experiment, ﬁve types of photos—one-person photos, two-person photos, family photos, social-activities photos, and natural-scenery photos—10 photos per type, were used. Pictures of an individual who is reﬂecting, at work, or participating in an activity were used to prime the individual-oriented self. Pictures of couples or lovers holding hands, smiling to each other or engaging in activities together were used to prime relationship- oriented self. Pictures of family gatherings, e.g., a family at the Spring Festival dinner table, clan reunion events, etc., were used to prime the familistic (group)-oriented self. Pictures such as those portraying a group of Chinese fans at a China-South Korea baseball game were used to prime the other-oriented self. Common natural landscape pictures were used as the neutral priming control group. THIS PRESENT RESEARCH To be more speciﬁc, in the four-part theory of the Chinese self, Yang (2004) provides a clearer description on self, arguing the three sub-divisions of the social-oriented self is characteristic of the Chinese self. In this study, three experiments were conducted to examine how to eﬀectively trigger diﬀerent types of self and the relative importance of the three types of self. In Experiment 1, the photo-primed method would be used to demonstrate that the three types of self can be individually triggered and indeed exist. In Experiment 2, we adopted the research model of the Yang believes that the four-part theory of the Chinese self can completely cover the Chinese self and that, being mutually distinctive, the four types of self represent the methods of interaction by which the Chinese interact with others in the four major modalities. We believe that Yang’s four-part theory of the Chinese self is a very important theory, not only having indigenous compatibility and depicting the self-construal of June 2017 \| Volume 8 \| Article 1106 Frontiers in Psychology \| www.frontiersin.org 5 The Four-Part Theory of the Chinese Self Sun Word/phrase categories In the pre test three Word/phrase categories In the pre-test, three graduate students who know the four-part theory well-worked together to select words that could reﬂect the individual-oriented self, the relationship- oriented self, the familistic (group)-oriented self and the other- oriented self. Both words of individual-oriented self and words covering the three types of social-oriented self (cross-category interpersonal words) from the “Frequency Dictionary of Written Chinese” published by the Academia Sinica (Chinese Knowledge Information Processing Group, 1994), were then included in the questionnaire. We then explained to a group of respondents about the meanings of these diﬀerent selves and asked them to judge both the category to which a word should belong (multiple choices were allowed) and a word’s degree of association with the self-category to which it was assigned (the degree was measured on a seven-point scale; the higher the number, the greater the degree of association). The self-category to which each word belongs was determined based on the respondents’ choices. The criteria of word selection were that the word was categorized into a particular self-category by more than 90% of the respondents, that the word’s degree of association with the self-category was scored at 5 or above and that the percentages at which the word was assigned to other self- categories were all below 10%, thus conﬁrming that the word was attributed to a particular self-category. The selection criteria of cross-category interpersonal words were both that a word had a percentage of over 50% [at which it was categorized into two- person relationship orientation, family (group) orientation, or other orientation] and that the word had a degree of association of 5 or above with its category. Materials limited self-regulation resource to investigate the importance of diﬀerent types of self. In Experiment 3, the relative importance of three types of self in the social-oriented self was investigated through diﬀerent scenarios of conﬂict between diﬀerent types of self and through the individuals’ choices, the relative importance of diﬀerent types of self was studied. The materials included photos that use to prime diﬀerent types of self and included diﬀerent categories of words/phrases that can reﬂect diﬀerent types of self. EXPERIMENT 1 In examining the four-part theory of the Chinese self, it is most important to conﬁrm the existence of each of the four types of self, especially the relationship-oriented self, the familistic (group)-oriented self and the other-oriented self, which are under the social-oriented self and can be individually triggered. In addition, although we believe that the individual-oriented self and the social-oriented self are distinguishable from one another, we also believe that the three types of social-oriented self are not clear-cut but instead unique, albeit with overlapping. We wish to test these ideas. In Experiment 1, the photo-primed method was employed. Sun (2007), Cheng (2006), Chen (2013), and Sun and Kao (2017) used photos to prime the individual-oriented self and social- oriented self in individuals. In this study, photos related to diﬀerent types of self were used to prime various orientations of self. We hypothesized that once the individual-oriented self has been primed, an individual’s reaction time for an individual- related word will be signiﬁcantly shorter than for other categories of words. When the relationship-oriented self is primed, an individual’s reaction time for a word related to the relationship orientation of two persons will be signiﬁcantly shorter than for other categories of words. When the familistic (group)-oriented self is primed, an individual’s reaction time for a family (group) orientation-related word is signiﬁcantly shorter than for other categories of words. When the other-oriented self is primed, an individual’s reaction time for other orientation-related words is signiﬁcantly shorter than for other categories of words. We also predicted that when the relationship-oriented self, the familistic (group)-oriented self, and the other-oriented self (all of which are under the social-oriented self) are primed, individuals will have a signiﬁcantly shorter reaction time for cross-category interpersonal words, i.e., words that reference the common aspects of basic concepts. Participants and Design The participants were 88 college students from a national university in Taiwan participated for course credit. Six participants who had extreme reaction times were excluded. There were ﬁve diﬀerent self-priming conditions, which were individual-oriented self, the relationship-oriented self, the familistic (group)-oriented self, the other-oriented self, and neutral condition. Participants in each group were only shown photos of one type according to their assigned group and asked to respond to words from all six categories. The dependent variable was the participant’s reaction time to words from various categories. After these pre-tests, a total of 358 respondents determined words from six self-categories: the individual self-category (e.g., 自主“autonomy”, 獨特“uniqueness”), the two-person relationship self-category (e.g., 甜蜜“sweet”, 卿卿我我“deeply attached to each other”), the family (group) self-category (e.g., 孝 June 2017 \| Volume 8 \| Article 1106 Frontiers in Psychology \| www.frontiersin.org 6 The Four-Part Theory of the Chinese Self Sun there was no signiﬁcant diﬀerence related to the cross-category interpersonal words (M = 1085.53, p = 0.065). 順“ﬁlial piety ”, 長幼有序“to respect for seniority”), the other- orientation self-category (e.g., 社會認可“social recognition”, 公德心“civic-minded”), neutral words (e.g., 濕潤“moist”, 彎曲“bend”), and cross-category interpersonal words (e.g., 和睦 “harmony”, 在一起“togetherness”). Each self-category included ten words or phrases. In the familistic (group)-oriented self-priming condition, a repeated measures one-way ANOVA revealed a statistically signiﬁcant eﬀect on word categories, F(5, 80) = 12.978, p < 0.001, η2 p = 0.448. Post-hoc comparisons showed that the participants had a signiﬁcantly faster average reaction time to the family (group) orientation self-category words (M = 1032.65) than to the individual-orientation self-category words (M = 1241.41, p = 0.006), the two-person relationship-orientation self-category words (M = 1519.93, p < 0.001), the other-orientation self- category words (M = 1540.85, p < 0.001), and the neutral words (M = 1214.04, p = 0.004), whereas there was no signiﬁcant diﬀerence related to the cross-category interpersonal words (M = 1131.50, p = 0.150). Procedure After arriving at the laboratory, the participant was informed that the experiment was designed to understand an individual’s ability of free association. After ﬁlling the consent form, each participant was then randomly assigned to see one of the ﬁve types of photos—one-person photos, two-person photos, family photos, social-activities photos, or natural-scenery photos. Photos were presented one at a time, and followed by a word/phrase. Participants need to decide if the word/phrase could reﬂect the thought elicited by the photo as fast as they can by pressing the “yes” or “no” key, and the reaction times were all recoded. Photos from each of the ﬁve types would paired with words/phrases from all the self-categories and neutral word category. All experimental materials were presented via computers using MediaLab software. In the other-oriented self-priming condition, a repeated measures one-way ANOVA also revealed a statistically signiﬁcant eﬀect on word categories, F(5, 85) = 2.981, p = 0.016, η2 p = 0.149. Post-hoc comparisons showed participants who were triggered the other-oriented self had a signiﬁcantly faster average reaction time to the other-orientation self-category words (M = 1038.96) than to the individual-orientation self-category words (M = 1299.79, p = 0.005), the family (group)-orientation self- category words (M =1268.91, p = 0.009), and the neutral words (M = 1227.48, p = 0.039), whereas there was no signiﬁcant diﬀerence related to the two-person relationship-orientation self- category words (M = 1098.79, p = 0.390) or the cross-category interpersonal words (M = 1124.82, p =.076). The results are presented in Table 2. Results After applying an inverse transformation on reaction time data, the assessments of the normality and sphericity assumptions were acceptable (Box and Cox, 1964). For ease of interpretation, raw reaction times will be displayed for descriptive purposes only. A two-way mixed design analysis of variance (ANOVA) with a between-subjects factor of ﬁve diﬀerent self-priming conditions and a within-subjects factor of six word categories revealed a signiﬁcant interaction between self-priming and word categories, F(20, 415) = 9.548, p < 0.001, η2 p = 0.315. To test the hypotheses, separate repeated measures one-way ANOVAs were conducted in diﬀerent self-priming conditions. Discussion h l h The results showed that when the individual-oriented self was primed, the participants had a signiﬁcantly faster reaction time to the individual-related words than to the individual- unrelated words; as expected, the three types of social-oriented self were distinguished from each other while sharing an underlying basis. Therefore, when the relationship-oriented self was primed, the participants had a signiﬁcantly faster reaction time to the words reﬂected the relationship-oriented self than to words of other self-categories, along with an accelerated reaction time to the cross-category interpersonal words, whereas they exhibited no signiﬁcant diﬀerence between the reaction time to the two-person relationship orientation self-category words. In the individual-oriented self-priming condition, a repeated measures one-way ANOVA revealed a statistically signiﬁcant eﬀect on word categories, F(5, 85) = 12.305, p < 0.001, η2 p = 0.420. Post-hoc comparisons were then performed. The participants had a signiﬁcantly faster average reaction time to the individual-orientation self-category words (M = 845.03) than to the two-person relationship-orientation self-category words (M = 919.03, p = 0.042), the family (group)-orientation self-category words (M = 994.26, p < 0.001), the other-orientation category words (M = 1117.61, p < 0.001), the neutral words (M = 1015.31, p < 0.001), and the cross-category interpersonal words (M = 952.98, p = 0.003). This result was consistent with expectations. When the familistic (group)-oriented self was primed, the participants had a signiﬁcantly faster reaction time to words reﬂected the familistic (group)-oriented self than to words of other categories, along with an accelerated reaction time to the cross-category interpersonal words; they exhibited no signiﬁcant diﬀerence between the reaction time to the cross-category interpersonal words and the reaction time to words from the familistic (group)-oriented self-category. When the other- oriented self was primed, the participants had a signiﬁcantly faster reaction time to words reﬂected the other-oriented self than to words related to the familistic (group)-oriented self; they exhibited a signiﬁcantly accelerated reaction time to cross-category interpersonal words, whereas they exhibited no signiﬁcant diﬀerence in reaction time involving words related to In the relationship-oriented self-priming condition, a repeated measures one-way ANOVA was conducted and result showed a statistically signiﬁcant eﬀect on word categories, F(5, 75) = 4.964, p < 0.001, η2 p = 0.249. June 2017 \| Volume 8 \| Article 1106 Discussion h l h Post-hoc comparisons indicated that participants who were primed relationship- oriented self had a signiﬁcantly faster average reaction time to the two-person relationship-orientation self-category words (M = 1005.93) than to the individual-orientation self-category words (M = 1338.85, p = 0.001), the family (group)-orientation self-category words (M = 1312.48, p < 0.001), the other- orientation self-category words (M = 1273.51, p < 0.001), and the neutral words (M = 1216.17, p = 0.003), whereas June 2017 \| Volume 8 \| Article 1106 June 2017 \| Volume 8 \| Article 1106 Frontiers in Psychology \| www.frontiersin.org 7 The Four-Part Theory of the Chinese Self Sun TABLE 2 \| Post-hoc comparisons of reaction time among different groups (raw data). M(SD) M(SD) SE p TRIGGERING THE INDIVIDUAL-ORIENTED SELF (N = 18) Individual-orientation self-category words 845.03 (163.8) Two-person relationship-orientation self-category words 919.03 (204.62) 31.78 0.042 Family (group)- orientation self-category words 994.26 (230.67) 37.86 0.000 Other- orientation self-category words 1117.61 (327.86) 61.42 0.000 Neutral words 1015.31 (219.25) 32.52 0.000 Cross-category interpersonal words 952.98 (225.50) 34.33 0.003 TRIGGERING THE TWO-PERSON RELATIONSHIP-ORIENTED SELF (N = 16) Two-person relationship-orientation self-category words 1005.93(284.79) Individual-orientation self-category words 1338.85 (540.04) 110.01 0.001 Family (group)- orientation self-category words 1312.48 (380.58) 76.62 0.000 Other- orientation self-category words 1273.51 (357.96) 52.95 0.000 Neutral words 1216.17 (281.62) 62.33 0.003 Cross-category interpersonal words 1085.53 (250.79) 58.03 0.065 TRIGGERING THE FAMILISTIC (GROUP)-ORIENTED SELF (N = 17) Family (group)- orientation self-category words 1032.65(302.04) Individual-orientation self-category words 1241.41 (421.66) 57.88 0.006 Two-person relationship-orientation self-category words 1519.93 (465.71) 77.15 0.000 Other- orientation self-category words 1540.85 (439.37) 83.91 0.000 Neutral words 1214.04 (324.84) 49.19 0.004 Cross-category interpersonal words 1131.50 (429.30) 57.00 0.150 TRIGGERING THE OTHER-ORIENTED SELF (N = 18) Other- orientation self-category words 1038.96(283.47) Individual-orientation self-category words 1299.79 (459.17) 98.23 0.005 Two-person relationship-orientation self-category words 1098.79 (297.95) 68.14 0.320 Family (group)- orientation self-category words 1268.91 (457.69) 95.57 0.009 Neutral words 1227.48* (424.87) 74.07 0.039 Cross-category interpersonal words 1124.82 (273.50) 67.54 0.076 TRIGGERING THE NEUTRAL CONDITION (N = 19) Neutral words 975.64(224.03) Individual-orientation self-category words 1050.59 (318.85) 64.52 0.402 Two-person relationship-orientation self-category words 982.62 (193.46) 44.34 0.766 Family (group)- orientation self-category words 1049.64 (352.17) 70.50 0.632 Other- orientation self-category words 1023.47 (261.90) 61.98 0.300 Cross-category interpersonal words 1049.16 (329.98) 71.71 0.461 All times in are in milliseconds. Asterisks indicate a signiﬁcant effect. p < 0.05, p < 0.01. TABLE 2 \| Post-hoc comparisons of reaction time among different groups (raw data). Discussion h l h Neutral words the other-oriented self. These results essentially supported the four-part theory of the Chinese self proposed by Yang that in addition to the individual-oriented self, the social-oriented self can be further divided into three diﬀerent types of self that are distinctive but do not necessarily exist independently. exaggerate, suppress or naturally express their emotions and facial expressions while watching the ﬁlm clip. The participants were then asked to perform a hand-squeezing task. The results showed that the participants who exaggerated or suppressed their emotions and facial expressions performed more poorly in the squeezing task than those from the control group, who expressed their emotions naturally. June 2017 \| Volume 8 \| Article 1106 Frontiers in Psychology \| www.frontiersin.org Procedure and Measures The experimenter informed all of the participants that the experiment purported to collect college students’ opinions. After ﬁlling the consent form, each participant was then randomly assigned an essay and asked to read it. First, each participant was assessed on his/her comprehension of the essay, the extent of his/her agreement with its content and the importance of the content. Next, each participant was informed that there were enough opinions supporting the content of the essay and asked to do his/her best to think of and write down (in the space of 10 min) 10 counterarguments. After a participant completed the rebuttal task, he/she was asked to perform a self-evaluation of his/her perceived eﬀort in completing the rebuttal task and the persuasiveness of his/her counterarguments. “I believe that to maintain a good relationship between lovers (or husband and wife), both mind and eﬀort are required; when two are getting along, in addition to enjoying the sweetness of life, they should be considerate and take care of each other. When making a major decision, the two (lovers or couple) should trust each other and have conversations. Because another half is a part of oneself and very important spiritual support, in a relationship, one should accept the other’s imperfection, try to be proud of the other and strive to contribute to making each other happy.” After a participant completed the self-evaluation, he/she was informed that before proceeding to the next step, he/she was required to perform a number-comparing task to monitor his/her attention and collect data for another experiment. Each participant was informed that he/she should try his/her best, that there was no limitation on the number of questions and that he/she could stop at any time. The number of number-comparing questions completed by the participant was set as the dependent variable. A smaller number represents the consumption of more self-resources in performing the rebuttal task. Other participants assigned to refute the familistic (group)- oriented self or other-oriented self received diﬀerent essays and were asked to formulate a rebuttal after reading the essays. Participants in the control group were assigned to read an essay on the importance of convenience stores to our lives and to formulate a rebuttal. Other participants assigned to refute the familistic (group)- oriented self or other-oriented self received diﬀerent essays and were asked to formulate a rebuttal after reading the essays. Essays related to diﬀerent types of self After the pre-test, one essay was selected for each of the three types of self; one essay on the importance of convenience stores was used as the control group. These essays showed no signiﬁcant diﬀerences in rationality and persuasiveness. The participants were asked to write a ten-argument rebuttal in 10 min after reading the essay, each of which started, “I do not agree with that view, because ...” For example, the participants assigned to refute the relationship-oriented self were given the following short essay: Participants and Design The participants of this study were college students from a national university in northern Taiwan. The valid sample size was 55 persons after excluding 4 outliners in the number- comparing tasks; 43 participants were female and 12 were male. All participants were 19 to 22 years of age (M = 20.02). This experiment was a one-factor between-subject design in which participants were randomly assigned to four scenarios to rebut the importance of diﬀerent types of self (independent variables), i.e., the relationship-oriented self, the familistic (group)-oriented self, the other-oriented self, and the “convenience store” group (the control group). Number-comparing task To understand the remaining resources of participants from diﬀerent groups after performing the rebuttal task related to the self of diﬀerent orientations, the participants were asked to perform a number-comparing task. Each question involved two numbers with an identical number of digits (4–12 digits); if the participant thought that the two numbers were identical (e.g., “553314521705” and “553314521705”), “O” was marked, whereas if the participant thought that the two numbers were diﬀerent from each other (e.g., “537608031077” and “537806031077”), “X” was marked. The measurements of manipulations task, the worse individuals will perform in a subsequent task. In Experiment 2, the same logic was adopted to test the importance of diﬀerent orientations of self. To conﬁrm that there were no signiﬁcant diﬀerences among the participants with respect to their level of comprehension of, level of agreement with and importance assessment of the descriptive essay about diﬀerent types of self, the participants were asked to make an assessment using a seven-point scale. The higher the score, the higher the level it represents. The purpose of Experiment 2 was to examine the importance of three types of social-oriented self. Participants were asked to refute after reading a short essay on the importance of a particular type of self. According to the self-regulation concept, rebuttal of the importance of diﬀerent types of self is very diﬃcult and thus consumes resources; the more importance attributed to a type of self by the individuals, the more resources they will consume in performing the rebuttal and the worse they will perform in the subsequent task. Therefore, the participants’ performance in the subsequent task can be used to test the importance of diﬀerent types of self. To conﬁrm that the participants did their best on the rebuttal tasks, the participants were also asked to perform self- evaluations on their perceived eﬀort and the persuasiveness of their completed tasks, on a seven-point scale. The higher the score, the higher the level it represents. Self-esteem scale To prevent the participant’s self-esteem level from aﬀecting his/her performance in completing the rebuttal task, the participant’s self-esteem was measured as the control variable using the “Self-Esteem Scale” (Rosenberg, 1965), which had 10 items and was measured by a seven-point scale. The higher the score, the higher the level of self-esteem. EXPERIMENT 2 y Muraven and Baumeister (2000) argued that the reason for their result is that the suppression or exaggeration task requires self-control and thus consumes self-regulation resources: after the participants perform the task, they will be aﬀected when performing other tasks that require resource-consuming self- control. Crocker et al. (2006) argue that when individuals ﬁnd something that is either important to them or relevant to their self-worth, they will invest more self-control resources in the relevant task; the more diﬃcult the task, the more resources they will consume and the worse they will perform in the subsequent task. For example, the more relevance attributed to learning for an individual’s self-worth and the more diﬃcult a learning In Experiment 2, we wanted to investigate the importance of diﬀerent types of self in the social-oriented self and adopted the research model of the limited self-regulation resource. Muraven and Baumeister (2000) argued that so-called holistic self-regulation is a ﬁnite resource and can be temporarily depleted when it is used. Each self-control task performed by an individual is bound to consume resources, and the more those resources are used, the greater the task’s eﬀect on the performance of the sequent task (Baumeister et al., 1998; Muraven et al., 1998). Muraven et al. (1998) asked participants to watch a clip of uncomfortable ﬁlm and asked them to maximally June 2017 \| Volume 8 \| Article 1106 Frontiers in Psychology \| www.frontiersin.org 8 The Four-Part Theory of the Chinese Self Sun EXPERIMENT 3 In Experiment 2, the importance to each individual of the three orientations of self was demonstrated. The question that remains is this: what is the relative importance of the three diﬀerent types of self to each individual? In other words, when conﬂict between two orientations of the self occurs in a particular situation and context, which orientation of the self is more important to the individual? This issue has yet to be clariﬁed. The perceived eﬀort in performing the rebuttal task Th i i ’ lf l i f h i ﬀ h The perceived eﬀort in performing the rebuttal task The participants’ self-evaluations of their eﬀort on the rebuttal task and the persuasiveness of their counterarguments were set as the dependent variables, and a MANOVA was conducted to determine whether there were diﬀerences in the two items among four groups of participants. The results showed that the overall eﬀect was insigniﬁcant (Wilks’ 3 = 0.92, ns), indicating that the four groups of participants showed no diﬀerence in their self-assessed eﬀort in completing the rebuttal task, with an average score of 5.89 points. No signiﬁcant diﬀerence was found in the persuasiveness of counterarguments, with an average score of 4.82 points, indicating that each group performed the rebuttal task in accordance with the instructions and that the manipulation of the rebuttal task was successful. Procedure and Measures Participants in the control group were assigned to read an essay on the importance of convenience stores to our lives and to formulate a rebuttal. June 2017 \| Volume 8 \| Article 1106 Frontiers in Psychology \| www.frontiersin.org 9 The Four-Part Theory of the Chinese Self Sun participants who refuted the familistic (group)-oriented self, both of which had results signiﬁcantly diﬀerent from the control group (p = 0.011; p = 0.001). Conversely, the group that refuted the other-oriented self showed a marginally signiﬁcant diﬀerence from the control group (p = 0.080). These results indicate that for the participants, regardless of which type of self they were refuting, the refuting process consumed a remarkable amount of their self-regulation resources, making it diﬃcult for them to concentrate on completing the subsequent task. This was especially true of participants who refuted the relationship- oriented self and family-oriented self. Participants who refuted the other-oriented self also showed a similar tendency. The results are presented in Figure 1. After each participant completed this task, he/she then was informed that the experiment was over. Finally, explanations about the rationale of the study were given and gratitude was expressed to all participants for their participation. Frontiers in Psychology \| www.frontiersin.org Comprehension and importance of the essay The assessments on a participant’s comprehension, agreement and importance on the essay were used as the dependent variables, and multivariate analysis of variance (MANOVA) was conducted on “rebuttal on scenarios of diﬀerent types of self” to understand whether there were diﬀerences in the assessments on the three items among four groups of participants. The results showed that the overall eﬀect was insigniﬁcant (Wilks’ 3 = 0.64, ns), indicating that the four groups of participants showed no diﬀerences in the assessments of their comprehension, agreement and importance evaluation related to their essays. In addition, the average score of each group on each of the indicators was above 5.7, indicating that the participants had a high level of comprehension of and agreement on the essay, while regarding its content as very important. Therefore, the manipulation of the essays was successful. Discussion I E i 2 In Experiment 2, we attempted to use the concept of limited overall self-regulation resources to test the importance of diﬀerent types of self. Based on the idea that participants would write counterarguments after reading essays on diﬀerent types of self and that the particular type of self implicated was important for a particular participant, rebuttal of a particular self orientation would cost more resources and led to poor performance in the subsequent task. It was found that the relationship-oriented self, the familistic (group)-oriented self, and the other-oriented self (all of which are under the social-oriented self) were all important self orientations, causing the participants’ poor performance in the subsequent task, especially for participants who refuted the relationship-oriented self and family-oriented self. Participants who refuted the other-oriented self showed a similar tendency. However, these results did not reveal the relative importance of diﬀerent types of self, an issue that was addressed in Experiment 3. Manipulation Checks Comprehension and importance of the essay The Number of Number-Comparing Questions Completed The number of number-comparing questions completed by the participant was set as the dependent variable, the “rebuttals on scenarios of self of diﬀerent orientations” was set as the independent variables, and the individual’s self-esteem score was set as the control variable for conducting the single-factor analysis of covariance (ANCOVA). The results showed that the eﬀect of refuting diﬀerent orientations of self was signiﬁcant [F(3, 50) = 4.32, p = 0.009, η2 p = 0.206], indicating that diﬀerent rebuttal scenarios consumed diﬀerent levels of self-regulation resources. In Experiment 3, the relative importance of various types of social-oriented self was investigated through diﬀerent scenarios of conﬂict between diﬀerent types of self. For example, Chinese revolutions have espoused the notion of “exchanging the small- love for the big-love” in which small-love refers to the love between couples, representing the relationship-oriented self, whereas big-love refers to care for a great many others, representing the other-oriented self. “Exchanging the small-love for the big-love” represents a conﬂict between the relationship- oriented self and the other-oriented self. In this experiment, situations of conﬂict between various orientations of self were set up. Through the participants’ choices, the relative importance of diﬀerent types of self was analyzed. Post-hoc comparisons were performed using LSD (least signiﬁcant diﬀerence) tests. The results showed that participants from the three groups whose rebuttals related to three orientations of self completed fewer of the number-comparing questions. This was particularly true of the group of participants who refuted the relationship-oriented self and the group of In this study, we hypothesized that when individuals are confronted by two conﬂicting orientation origins of self, they June 2017 \| Volume 8 \| Article 1106 Frontiers in Psychology \| www.frontiersin.org 10 The Four-Part Theory of the Chinese Self Sun 100.86 95.63 111.54 131 0 20 40 60 80 100 120 140 160 180 relationship-oriented self familistic(group)-oriented self other-oriented self control r e b m u n f o r e b m u n e h T - d e t e l p m o c s n o it s e u q g n ir a p m o c p = .011 p** = .001 p = .080 FIGURE 1 \| The average number of number-comparing questions completed by different groups. FIGURE 1 \| The average number of number-comparing questions completed by different groups. The Number of Number-Comparing Questions Completed conﬂict scenarios are the same (i.e., behaviors related to the self of a certain orientation are not more reasonable or more likely to occur than behaviors related to the self of another orientation), in the pre-test, each scenario that contains diﬀerent conﬂicting orientations of self was individually divided into two versions, A or B, in which behaviors related to only one orientation of self occur. Next, the two versions that contain behaviors related to a single orientation of self were compared to determine whether they demonstrate the same possibilities. are bound to choose the type of self that has a more important outcome to them and relinquish the type of self that is relatively unimportant. Therefore, we performed pairwise comparison of the three orientations of self, i.e., three groups of pairwise comparison. For each group, four scenarios were designed and the participants were asked to make a choice in each conﬂict scenario. Ultimately, by analyzing the participants’ choices, the relative importance of diﬀerent types of self was understood. The pre-test was a one-way between-subject design with 21 groups of dilemmas. Two versions (A and B) for each dilemma situation were created, each of which contained a single type of orientation of self. The participants were randomly assigned to read either the A or the B version. For example, the “dilemma of family (group) self vs. the other-oriented self” is as follows: Participants and Design The participants in Experiment 3 were 63 college students from public universities in northern Taiwan, including 48 females and 15 males. Their average age was 19.10 years. Participants were randomly assigned to scenarios of conﬂict between any two of the three orientations of self: (1) the relationship-oriented self vs. the familistic (group)-oriented self; (2) the relationship-oriented self vs. the other-oriented self; and (3) the familistic (group)-oriented self vs. the other-oriented self. “Meixiu, who is in critical condition, intends to sign an organ- donation agreement (consistent with a doctor’s recommendation) to help many patients in need. However, when her closest family members hear about her intention, they feel very sad and think she should leave her body intact. Therefore, they are opposed to her plan to donate organs.” Results For example, the actor experiences a dilemma in which he must care for either the relationship-oriented self or the other-oriented self in a situation involving concerns about other’s thoughts and social honor. The ﬁrst category involved scenarios featuring conﬂicts between the relationship-oriented self and the familistic (group)- oriented self. For example, in this scenario, the husband experiences a dilemma when he wants to take care of his wife and pursue rights for the couple while pursuing the interests of his family group. In other words, this is a conﬂict scenario of the relationship-oriented self and the familistic (group)-oriented self. The second category involved scenarios featuring conﬂicts between the relationship-oriented self and the other-oriented self. For example, the actor experiences a dilemma in which he must care for either the relationship-oriented self or the other-oriented self in a situation involving concerns about other’s thoughts and social honor. Next, ﬁtness was tested by combining answers to the scenarios featuring the three types of dilemma situations. The results showed that in the conﬂict involving “the relationship-oriented self vs. the familistic (group)-oriented self,” the participants were signiﬁcantly inclined to choose behaviors in accordance with the familistic (group)-oriented self [χ2(1, N = 311) = 88.45, p < 0.001]. In the conﬂict involving “the relationship- oriented self vs. the other-oriented self,” the participants were signiﬁcantly inclined to choose behaviors in accordance with the other-oriented self [χ2(1, N = 189) = 5.76, p = 0.02]. In the conﬂict involving “the other-oriented self vs. the familistic (group)-oriented self,” the participants were signiﬁcantly inclined to choose behaviors in accordance with the other-oriented self [χ2(1, N = 252) = 112.00, p < 0.001]. The third category involves scenarios featuring conﬂicts between the familistic (group)-oriented self and the other- oriented self in which the actor experiences a dilemma involving the need to take care of either the interests of the family group to which he belongs or the other-oriented self. Such a scenario involves concerns about other’s thoughts and social honor. Procedure and Measures After arriving at the laboratory, the participant was informed that the experiment was designed to understand an individual’s choices related to speciﬁc life situations. The experimenter might also emphasize that these situations involved actual events that befell the participant’s classmates and alumni, thus emphasizing the real-life nature of the situations so that the participant would be more engaged in the situation. The experimenter then asked the participant to imagine that he/she is the person in the situation experiencing the dilemma. The participant then read the descriptions of all of the scenarios and indicated which choice he/she would make if he/she were the person in the The B version of a “request from the familistic (group)-oriented self” is as follows: The B version of a “request from the familistic (group)-oriented self” is as follows: situation. The main dependent variables were the participant’s choices in the dilemma. The participants also needed to measure their levels of certainty about their choices using a seven-point scale. The higher the score is, the more certain the participant’s choice. After the participant completed all of the questionnaires, the experiment was completed. The experimenter then debriefed and thanked the participant for his/her participation. “Meixiu is critically ill. Her close family members are quite sad and they hope to preserve the physical integrity of her body. They are opposed to donating her organs.” Eighty six college students from universities in northern Taiwan were randomized assigned to received either the A version or B version of the questionnaire and assessed the probability that they would agree to the choice set out for the situation using a seven- point scale (the higher the score, the more likely the participant is to agree to act the same way). Frontiers in Psychology \| www.frontiersin.org Results Selection of Different Types of Self The participants’ choices were used as dependent variables to be tested on ﬁtness. The selected choices and chi-square of each scenario are shown in Table 3. The results showed that in the conﬂict situation involving “the relationship-oriented self vs. the familistic (group)-oriented self,” the participants emphasized the familistic (group)-oriented self. In the conﬂict situation involving “the relationship-oriented self vs. the other-oriented self,” the participants were more inclined to choose behaviors facilitating the other-oriented self over the relationship-oriented self; however, there was a scenario about “whether to invite singers to have a show in his/her spouse’s company or to perform for the public welfare” that was discarded in the ﬁnal analysis because many participants were confused because the scenario contained poor deﬁnitions. In the conﬂict situation involving “the familistic (group)-oriented self vs. the other- oriented self,” except for the “bread donation” scenario, in which the participants’ choices did not diﬀer signiﬁcantly, the participants strongly preferred the choices favoring other- oriented self behaviors in the other three scenarios, indicating that in the conﬂict of “the familistic (group)-oriented self vs. the other-oriented self,” the participants preferred the other-oriented self. The probability that each participant believes he/she would engage a certain behavior was then set as the dependent variable and “the behavior in a situation with a single orientation of the self” was set as the independent variable to perform the two- tailed t-test. Ultimately, among the three conﬂicting situations, two orientations of self that showed no signiﬁcantly diﬀerent probabilities of behaviors and that had a probability of above 4.5 points in each case were chosen. Four scenarios of each type of conﬂict were designed, resulting in 12 scenarios overall. Finally, based on the pre-test results, 12 scenarios of conﬂict in three categories were chosen: The ﬁrst category involved scenarios featuring conﬂicts between the relationship-oriented self and the familistic (group)- oriented self. For example, in this scenario, the husband experiences a dilemma when he wants to take care of his wife and pursue rights for the couple while pursuing the interests of his family group. In other words, this is a conﬂict scenario of the relationship-oriented self and the familistic (group)-oriented self. The second category involved scenarios featuring conﬂicts between the relationship-oriented self and the other-oriented self. Materials Scenarios of conﬂict between diﬀerent orientations of self Experiment 3 was designed to present a participant with a conﬂict situation that simultaneously triggers selves of diﬀerent orientations and to ask the participant to make a choice in the dilemma. The goal is to understand the relative importance of the three types of social-oriented self. Therefore, three types of conﬂict scenarios were needed. To further evaluate the possibility that requirements from any two orientations of self in the three In the pre-test, this dilemma is made into the A version of “the other-oriented self”: “Meixiu, who is in critical condition, intends to sign an organ- donation agreement (consistent with a doctor’s recommendation) to help many patients in need.” June 2017 \| Volume 8 \| Article 1106 Frontiers in Psychology \| www.frontiersin.org 11 The Four-Part Theory of the Chinese Self Sun The B version of a “request from the familistic (group)-oriented self” is as follows: The Level of Certainty of the Choices OTHER-ORIENTED SELF Scenario Use of underground water Site for grandpa’s birthday celebration party Organ donation Bread donation Familistic Other Familistic Other Familistic Other Familistic Other Choice 8 55 3 60 4 59 27 36 Percentage 12.7 87.3 4.8 95.2 6.3 93.7 42.9 57.1 χ2 35.06 51.57 48.02 1.29 p 0.000 0.000 0.000 0.257 FAMILISTIC (GROUP)-ORIENTED SELF VS. OTHER-ORIENTED SELF In the conﬂict of “the relationship-oriented self vs. the other- oriented self,” the participants’ level of certainty of participants about facilitating the other-oriented self (M = 5.67, SD = 0.97) was higher than the participants’ level of certainty about facilitating the relationship-oriented self (M = 5.35, SD = 0.96) [t(249) = −2.25, p = 0.03]. In the conﬂict of “the other-oriented self vs. the familistic (group)-oriented self,” the participants’ level of certainty about facilitating the other-oriented self (M = 5.81, SD = 0.92) was higher than the participants’ level of certainty about facilitating the family (group)-oriented self (M = 4.98, SD = 1.44) [t(249) = −4.82, p < 0.001]. In the conﬂict of “the relationship-oriented self vs. the other- oriented self,” the participants’ level of certainty of participants about facilitating the other-oriented self (M = 5.67, SD = 0.97) was higher than the participants’ level of certainty about facilitating the relationship-oriented self (M = 5.35, SD = 0.96) [t(249) = −2.25, p = 0.03]. In the conﬂict of “the other-oriented self vs. the familistic (group)-oriented self,” the participants’ level of certainty about facilitating the other-oriented self (M = 5.81, SD = 0.92) was higher than the participants’ level of certainty about facilitating the family (group)-oriented self (M = 4.98, SD = 1.44) [t(249) = −4.82, p < 0.001]. conﬂict between the other-oriented self and the familistic (group)-oriented self, individuals demonstrated a remarkable tendency to safeguard the other-oriented self. The results appear to conﬁrm the tendency toward “sacriﬁcing the small- self for the big-self” that is often mentioned in Chinese culture. GENERAL DISCUSSION Numerous studies have discussed the self of individuals, most frequently mentioning the division between the independent self and the dependent self (Markus and Kitayama, 1991, 2010). Yang (1993, 2004) proposed the four-part theory of the Chinese self for individuals inﬂuenced by Chinese culture, in which in addition to the individual-oriented self, the social-oriented self is subdivided into three diﬀerent types: the relationship- oriented self, the familistic (group)-oriented self and the other- oriented self. Yang believes that this multi-faceted description of self can completely cover the Chinese self. Although his theory does have deep indigenous compatibility and has attracted a great deal of attention, there remains a lack of in-depth empirical investigation. This study was intended to address that inadequacy and attempted to understand whether the three types of social-oriented self are indeed distinctive to individuals and their relative importance to each other. The Level of Certainty of the Choices The levels of certainty of the choices made by the participants were combined to serve as the dependent variable to perform the two-tailed t-test. The results indicated that in the conﬂict of “the relationship-oriented self vs. the familistic (group)-oriented self,” the participants’ level of certainty about facilitating the familistic (group)-oriented self (M = 5.50, SD = 1.05) was higher than the participants’ level of certainty about facilitating the relationship- oriented self (M = 5.12, SD = 1.09) [t(249) = −2.27, p = 0.02]. June 2017 \| Volume 8 \| Article 1106 Frontiers in Psychology \| www.frontiersin.org 12 The Four-Part Theory of the Chinese Self Sun TABLE 3 \| Choices made by individuals about behaviors involving different scenarios of conﬂicting orientations of self. RELATIONSHIP-ORIENTED SELF VS. FAMILISTIC (GROUP)-ORIENTED SELF Scenario Financial support to parents Boyfriend/family member tourism Location of rented house Marital considerations Relationship Familistic Relationship Familistic Relationship Familistic Relationship Familistic Choice 14 48 11 52 20 43 6 57 Percentage 22.2 6.2 17.5 82.5 31.7 68.3 9.5 90.5 χ2 18.65 26.68 8.40 41.29 p 0.000 0.000 0.004 0.000 RELATIONSHIP-ORIENTED SELF VS. OTHER-ORIENTED SELF Scenario Help to promote husband’s career Donation for public interests Volunteering in a hospital Relationship Other Relationship Other Relationship Other Choice 33 30 23 40 22 41 Percentage 52.4 47.6 36.5 63.5 34.9 65.1 χ2 0.14 4.59 5.73 p 0.705 0.032 0.017 FAMILISTIC (GROUP)-ORIENTED SELF VS. OTHER-ORIENTED SELF Scenario Use of underground water Site for grandpa’s birthday celebration party Organ donation Bread donation Familistic Other Familistic Other Familistic Other Familistic Other Choice 8 55 3 60 4 59 27 36 Percentage 12.7 87.3 4.8 95.2 6.3 93.7 42.9 57.1 χ2 35.06 51.57 48.02 1.29 p 0.000 0.000 0.000 0.257 TABLE 3 \| Choices made by individuals about behaviors involving different scenarios of conﬂicting orientations of self. RELATIONSHIP-ORIENTED SELF VS. OTHER-ORIENTED SELF Scenario Help to promote husband’s career Donation for public interests Volunteering in a hospital Relationship Other Relationship Other Relationship Other Choice 33 30 23 40 22 41 Percentage 52.4 47.6 36.5 63.5 34.9 65.1 χ2 0.14 4.59 5.73 p 0.705 0.032 0.017 RELATIONSHIP-ORIENTED SELF VS. OTHER-ORIENTED SELF RELATIONSHIP-ORIENTED SELF VS. OTHER-ORIENTED SELF Scenario Help to promote husband’s career FAMILISTIC (GROUP)-ORIENTED SELF VS. Discussion The results of Experiment 3 showed the participants’ level of emphasis on the three types of social-oriented self. In descending order, they were the other-oriented self, the familistic (group)-oriented self, and the relationship-oriented self. This tendency was exhibited not only in the choice preference in the conﬂict situations but also in the participants’ level of certainty about their choices, indicating that when confronted by a dilemma, individuals seemed to have a tendency to “pursue the interests of the overall situation” (顧全大局). They tended to safeguard the familistic (group)-oriented self when experiencing a conﬂict between the relationship-oriented self and the familistic (group)-oriented self, whereas in the June 2017 \| Volume 8 \| Article 1106 Frontiers in Psychology \| www.frontiersin.org 13 The Four-Part Theory of the Chinese Self Sun Although Three Types of Social-Oriented Self Share a Common Basis, They Are Distinctive experiment showed that rebuttals of arguments about the self of diﬀerent orientations were more resource-intensive than rebuttal of the essay about convenience stores, indicating that the three types of social-oriented self were very important to individuals, thus providing further support for the four-part theory of the Chinese self. In Experiment 1, the self of diﬀerent orientations was primed through photos. The results showed that as expected, selves of diﬀerent orientations could be primed individually. For example, when the familistic (group)-oriented self was primed, the participants had a signiﬁcantly shorter reaction time for words related to the familistic (group)-oriented self than for words reﬂected the relationship-oriented self or the other-oriented self. A similar phenomenon appeared when the relationship-oriented self or the other-oriented self was triggered, showing that the three types of self can be individually triggered. Because the three types of self all belong to the social-oriented self, they must have a common basis, as also conﬁrmed by the experimental results. These results suggest that regardless of which of the three types of self was triggered, the participants’ response time to cross-category interpersonal words signiﬁcantly accelerated, supporting the hypothesis of a “common basis.” This basis should combine the tendency toward high homonomy (with one’s surroundings) and low personal autonomy emphasized by Yang, which prompts Chinese individuals to attach importance to upholding harmonious interpersonal relationships and proper behavior. The characteristics, connotations, and operating principles in diﬀerent modalities determine speciﬁc interaction objects and modes, ultimately positioning and shaping the self of diﬀerent orientations. The Importance of the Other-Oriented Self Is the Most Undeniable Although the three types of social-oriented self are all important to varying degrees, in reality, there are situations in which two types of self conﬂicts, raising the following question: what is the relative importance of each type of self? In Experiment 3, pair-wise conﬂict of two diﬀerent orientations of self was designed to attempt to trigger two types of self simultaneously so that an individual can choose between the two and the relative importance of each type of self can be tested. The results showed that when the relationship-oriented self conﬂicts with the familistic (group)-oriented self, individuals clearly prefer the familistic (group)-oriented self. When the relationship-oriented self conﬂicts with the other-oriented self, individuals signiﬁcantly lean toward the other-oriented self. When the familistic (group)- oriented self conﬂicts with the other-oriented self, individuals signiﬁcantly choose the other-oriented self. Overall, the choices made by an individual experiencing conﬂicts between diﬀerent types of self reveal the levels of importance that he/she attaches to each of the three types of social-oriented self, which is, in descending order: the other-oriented self, the familistic (group)-oriented self, and the relationship-oriented self. This trend is exhibited not only in individuals’ choice preference in conﬂicting scenarios but also in the certainty of those individuals’ choices. When experiencing a conﬂict between the relationship-oriented self and the familistic (group)-oriented self, individuals exhibited a signiﬁcantly higher level of certainty when choosing behaviors facilitating the familistic (group)-oriented self. When experiencing a conﬂict between the relationship-oriented self and the other-oriented self, individuals exhibited a signiﬁcantly higher level of certainty when choosing behaviors facilitating the other-oriented self. When experiencing a conﬂict between the other-oriented self and the familistic (group)-oriented self, individuals exhibited a signiﬁcantly higher level of certainty when choosing behaviors facilitating the other-oriented self. Frontiers in Psychology \| www.frontiersin.org Each of the Three Types of Self Is Important diﬀerent types of social self, future research may considerate to trigger these selves and then investigate further how individuals with diﬀerent selves would act or perceive in the same and diﬀerent social contexts. The results also showed that when other types of self conﬂict with the other-oriented self, individuals tended to choose behaviors facilitating the other-oriented self. Why do Chinese individuals care so much about the other-oriented self? According to Yang’s four-part theory of the Chinese self, interaction objects involved in the other-oriented self are non- speciﬁc others, which can be either real or imaginary persons, e.g., “people” in the expression of “people all look down on me.” The other-oriented self displays correctly in such a context. Chinese want to be impressed and accepted by “non-speciﬁc others,” and caring for others’ opinions and norms and attending to one’s reputation are important revelations of the other-oriented self. Finally, we wonder whether this study is limited: do the results of Experiment 3 merely reﬂect social expectations? Perhaps we cannot completely rule out the possibility, but we did attempt to minimize it. In the pre-test, we assessed the possibility of the presence of each type of self’s behaviors and asked the participants to simultaneously evaluate the probability that they and others would behave in a particular way. Later, we constructed scenarios based on the resulting behaviors of two diﬀerently oriented selves. Those behaviors are highly possible and are not signiﬁcantly diﬀerent from each other in terms of their likelihood. Therefore, there should not be an issue about behaviors of one type of self being more in line with social expectations than those of another type of self. Therefore, the choice made by a participant in Experiment 3 should be his or her decision after careful consideration and can reﬂect which type of self is more important to the participant. In discussing the concept of “face” within the Chinese society, Hwang (2012) argues that Chinese people are unwilling to “lose face” in front of others—e.g., to violate public morality— because to do so would violate the Confucian self-cultivation requirements related to the individual. The other-oriented self wishes to gain others’ respect and the public’s recognition via such upbringing or behavior display. Yang (1991) argues that Chinese people regard themselves on the levels of the public self and the private self. Each of the Three Types of Self Is Important p In Experiment 2, the importance of each of the three types of social-oriented self [the relationship-oriented self, the familistic (group)-oriented self, and the other-oriented self] to each individual was investigated using the paradigm of limited self-regulation resources. After reading essays related to diﬀerent types of self, the participants were asked to write counterarguments. In general, when a certain type of self is important to an individual, that individual consumes more resources to write the rebuttal, leading to poor performance in completing the next task. In the experiment, the participants in the control group were asked to write a rebuttal to an essay claiming that convenience stores make life easier. Generally, the participants in the control group must consume substantial self-regulation resources in writing the rebuttal because in Taiwan, convenience stores are ubiquitous and facilitate daily life, allowing people to buy food and grocery, pay bills, withdraw cash, shop, and send packages. Therefore, it is very diﬃcult to refute the notion that convenience stores make life easier. These results seem to demonstrate that when confronted by such dilemmas, Chinese individuals exhibit a tendency to “pursue the interests of the big-self.” According to the four-part theory of the Chinese self, the relationship-oriented self is exhibited in the important dyadic relationship; its interaction objects are often intimate others (e.g., spouses), emphasizing interdependence and sharing. The familistic (group)-oriented self appears in the context of family or an important group. Yang believes that the basic structure and functional unit of the traditional Chinese society is the family. Therefore, Chinese people regard the family as the most important aspect, creating a familism that emphasizes the family and the pursuit of familial unity, honor and harmony. This could also be why individuals attach more importance to Besides, to eliminate the possibility that the results might be attributed to the diﬀerence in the ease of essay to counter-argue with, we pretested these essays with caution to ensure there were no signiﬁcant diﬀerences in their persuasiveness and rationality. Furthermore, in experiment 2, the manipulation check also indicated that participants comprehended and agree with the content of these essays equally well. Therefore, the results of the June 2017 \| Volume 8 \| Article 1106 14 The Four-Part Theory of the Chinese Self Sun the expression of the familistic (group)-oriented self than to the expression of the relationship-oriented self. FUNDING This work was supported by Grants from the Ministry of Science and Technology (Taiwan, R.O.C.) No. 97-2410-H-033-025, 98- 2410-H-004-193 and 99-2410-H-004-088. ETHICS STATEMENT This study was carried out in accordance with the recommendations of the guidelines of Research Ethics, the IRB of National Chengchi University, with written informed consent from all subjects. All subjects gave written informed consent in accordance with the Declaration of Helsinki. The protocol was approved by the IRB of the National Chengchi University. Overall, this study’s greatest contribution is its strong support of the four-part theory of the Chinese self. This theory of self related to Chinese people is not just conceptual argument: it is now empirically conﬁrmed. The study is also conducive to the deeper understanding of scholars who are interested in knowing more about the Chinese people. In addition to the individual- oriented self, Chinese people have the socially oriented selves of the relationship-oriented self, the familistic (group)-oriented self, and the other-oriented self. The three types of social-oriented self are mutually distinctive, can be individually triggered and are all very important to the Chinese. In addition, in this study we provided an eﬀective photo-priming procedure to activate AUTHOR CONTRIBUTIONS The author conﬁrms being the sole contributor of this work and approved it for publication. Each of the Three Types of Self Is Important The private self is the self-achieved by an individual through keeping secrets from others, whereas the public self is the self-achieved by an individual though “playing” for others. The private self is more stable, whereas the public self is prone to being inﬂuenced by or adjusted to others. The public self that “has cardinal principles in mind （識大體）” appears to be the embodiment of the other-oriented self. REFERENCES The individual- and social-oriented Chinese bicultural self: testing the theory. J. Soc. Psychol. 148, 347–374. doi: 10.3200/SOCP.148.3.347-374 Weng, C. Y., and Yang, K. S. (2003). “The development of the Chinese global self- esteem scale and multiple self-esteem scales: From individual-self oriented and social oriented-self oriented perspectives”, in Paper Presented at the Annual Meeting of the Chinese Psychological Association (Taipei). Lu, L., and Yang, K. S. (2005). Social- and individual-oriented views of self- actualization: conceptual analysis and preliminary empirical exploration. Indigenous Psychol. Res. Chinese Soc. 23, 3–69. g f y g p Yang C. F. (1991). “The self of the Chinese: theory and research orientation,” in Chinese Mind and Chinese People- from Social and Personality Perspective, eds Lu, L., Kao, S. F., Chang, T. T., Wu, H. P., and Zhang, J. (2008). The individual- and social-oriented Chinese bicultural self: a sub-cultural analysis contrasting mainland Chinese and Taiwanese. Soc. Behav. Pers. 36, 337–346. doi: 10.2224/sbp.2008.36.3.337 Yang C. F. (1991). “The self of the Chinese: theory and research orientation,” in Chinese Mind and Chinese People- from Social and Personality Perspective, eds C. F. Yang and S. R. Gao (Taipei: Yuan-Liou Publishing Co. Ltd.), 93–145. Yang, K. S. (1993). “Chinese social orientation: the social interactive perspective,” in The Mind and Behavior of the Chinese: Theory and Methodology, eds K. S. Yang and A B Yu (Taipei: Laureate Publishing Co Ltd ) 87 142 Yang, K. S. (1993). “Chinese social orientation: the social interactive perspective,” in The Mind and Behavior of the Chinese: Theory and Methodology eds K S g p p in The Mind and Behavior of the Chinese: Theory and Methodology, eds K. S. Yang and A. B. Yu (Taipei: Laureate Publishing Co. Ltd.), 87–142. Markus, H. R., and Kitayama, S. (1991). Culture and the self: implications for cognition, emotion, and motivation. Psychol. Rev. 98, 224–253. doi: 10.1037/0033-295X.98.2.224 Yang, K. S. (1995). “Chinese social orientation: an integrative analysis,” in Chinese Societies and Mental Health, eds T. Y. Lin, W. S. Tseng, and Y. K. Yeh (Hong Kong: Oxford University Press), 19–39. Markus, H. R., and Kitayama, S. (2010). Culture and self: a cycle of mutual constitution. Perspect. Psychol. Sci. 5, 420–430. doi: 10.1177/17456916103 75557 Yang, K. S. (2004). A theoretical and empirical analysis of the Chinese self from the perspective of social and individual orientation. Indigenous Psychol. Res. Chinese Soc. 22, 11–80. Muraven, M., and Baumeister, R. F. (2000). REFERENCES Crocker, J., Brook, A. T., Niiya, Y., and Villacorta, M. (2006). The pursuit of self- esteem: contingencies of self-worth and self-regulation. J. Pers. 74, 1749–1771. doi: 10.1111/j.1467-6494.2006.00427.x Baumeister, R. F., Bratslavsky, E., Muraven, M., and Tice, D. M. (1998). Ego depletion: is the active self a limited resource? J. Pers. Soc. Psychol. 74, 1252–1265. doi: 10.1037/0022-3514.74. 5.1252 Cross, S. E., Gore, J. S., and Morris, M. L. (2003). The relational-interdependent self-concept, self-concept consistency, and well-being. J. Pers. Soc. Psychol. 85, 933–944. doi: 10.1037/0022-3514.85.5.933 Box, G. E. P., and Cox, D. R. (1964). An analysis of transformations (with discussion). J. R. Stat. Soc. Ser. B Methodol. 26, 211–252. Dion, R. K., and Dion, K. L. (1993). Individualistic and collectivistic perspectives on gender and the cultural context of love and intimacy. J. Soc. Issues 49, 53–69. doi: 10.1111/j.1540-4560.1993.tb01168.x Chen, C. Y. (2013). The Authenticity of the Chinese in Diﬀerent Relational Contexts. Master’s thesis, National Chengchi University, Taipei. Chen, C. Y. (2013). The Authenticity of the Chinese in Diﬀere Greenwald, A. G., and Breckler, S. J. (1985). “To whom is the self presented?,” in The Seyand Social Life, ed B. R. Schlenker (New York, NY: McGraw-Hill), 126–145. Master’s thesis, National Chengchi University, Taipei. Cheng, S. C. (2006). Multiple Social Comparisons: The Relative Variations of the Self-Evaluation and the Evaluation of Others. Master’s thesis, Chung Yuan University, Taipei. Harter, S. (1999). The Construction of the Self. New York, NY: Guilford Press. Chinese Knowledge Information Processing Group (1994). A Frequency Dictionary of Written Chinese. Academia Sinica; Nankang, Taipei. Hwang, K. K. (2012). Foundations of Chinese Psychology: Confucian Social Relations. New York, NY: Springer. June 2017 \| Volume 8 \| Article 1106 Frontiers in Psychology \| www.frontiersin.org 15 The Four-Part Theory of the Chinese Self Sun Sun, C. R., and Wang, C. H. (2005). The self-evaluation and self-aﬃrmation process of culturally Chinese people. Indigenous Psychol. Res. Chinese Soc. 24, 139–187. Kühnen, U., and Oyserman, D. (2002). Thinking about the self inﬂuences thinking in general: cognitive consequences of salient self-concept. J. Exp. Soc. Psychol. 38, 492–499. doi: 10.1016/S0022-1031(02)00011-2 Kurman, J. (2001). Self-enhancement: is it restricted to individualistic cultures? Pers. Soc. Psychol. Bull. 27, 1705–1716. doi: 10.1177/014616720127 12013 Sun, C., and Kao, L. (2017). Coping reactions after social exclusion: the eﬀects of group related factors and individual factors. J. Educ. Psychol. 40, 61–92. Triandis, H. C. (1995). Individualism and Collectivism. Boulder, CO: Westview Press. Lu, L. (2008). Frontiers in Psychology \| www.frontiersin.org June 2017 \| Volume 8 \| Article 1106 REFERENCES Self-regulation and depletion of limited resources: does self-control resemble a muscle? Psychol. Bull. 126, 247–259. doi: 10.1037/0033-2909.126.2.247 Yang, K. S., and Lu, L. (2005). Social - vs. individual -oriented self- actualizers: conceptual analysis and empirical assessment of their psychological characteristics. Indigenous Psychol. Res. Chinese Soc. 23, 71–143. Muraven, M., Tice, D. M., and Baumeister, R. F. (1998). Self-control as limited resource: regulatory depletion patterns. J. Pers. Soc. Psychol. 74, 774–789. doi: 10.1037/0022-3514.74.3.774 Yang, K. S., Liu, Y. L., Chang, S. H., and Wang, L. (2010). Constructing a theoretical framework for the ontogenetic development of the Chinese bicultural self: a preliminary statement. Chinese J. Psychol. 52, 113–132. doi: 10.6129/CJP.2010.5202.01 Rosenberg, M. (1965). Society and the Adolescent Self-Image. Princeton, NJ: Princeton University press. Sedikides, C., and Brewer, M. B. (2001). Individual Self, Relational Self, Collective Self. Philadelphia, PA: Psychology Press. Conﬂict of Interest Statement: The author declares that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Singelis, T. M. (1994). The measurement of the independent and interdependent self-construals. Pers. Soc. Psychol. Bull. 20, 580–591. doi: 10.1177/0146167294205014 Conﬂict of Interest Statement: The author declares that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Sun, C. (2004). “‘Self-esteem is dear to me, but dearer still is interpersonal harmony.’— The individual-oriented self and social-oriented self of the Chinese”, in Paper Presented at the XVII Congress of the International Association for Cross-Cultural Psychology, (Xi’an). Copyright © 2017 Sun. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Sun, C. (2007). “Self-awareness and self-regulation of Chinese people,” in Paper Presented at the 7th Conference of Association of Asian Social Psychology (AASP), (Sabah). June 2017 \| Volume 8 \| Article 1106 Frontiers in Psychology \| www.frontiersin.org 16
https://openalex.org/W2011855429	https://www.scielo.br/j/fm/a/QnVQQQVYPqGMjrc5QX7TvZS/?lang=en&format=pdf	English	null	Inspiratory muscle strength in subjects with tetraplegia: viability of evaluation through the measurement of maximal inspiratory pressure	Fisioterapia em Movimento	2,014	cc-by	4,731	ISSN 0103-5150 Fisioter. Mov., Curitiba, v. 27, n. 3, p. 371-377, jul./set. 2014 Licenciado sob uma Licença Creative Commons DOI: http://dx.doi.org.10.1590/0103-5150.027.003.AO07 Inspiratory muscle strength in subjects with tetraplegia: viability of evaluation through the measurement of maximal inspiratory pressure Força muscular inspiratória em sujeitos com tetraplegia: viabilidade da avaliação pela medida da pressão inspiratória máxima Força muscular inspiratória em sujeitos com tetraplegia: viabilidade da avaliação pela medida da pressão inspiratória máxima [A] Marlene Aparecida Moreno[a], Juliana Viana Paris[b], Raphael do Nascimento Pereira[c], Antonio Roberto Zamunér[d], Tais Mendes de Camargo[e], Ricardo Machado Leite de Barros[f] [a] PhD, professor, Universidade Metodista de Piracicaba, Faculdade de Ciências da Saúde, Piracicaba, SP - Brazil, e-mail: ma.moreno@terra.com.br [b] MSc, Universidade Estadual de Campinas, Campinas, SP - Brazil, e-mail: julianav.paris@gmail.com [c] PhD candidate, Universidade Metodista de Piracicaba, Piracicaba, SP - Brazil, e-mail: raphaelnpfisio@yahoo.com.br [d] PhD candidate, Universidade Federal de São Carlos, São Carlos, SP - Brazil, e-mail: beto.zam@gmail.com [e] PhD candidate, Universidade Metodista de Piracicaba, Piracicaba, SP - Brazil, e-mail: taismendesdecamargo@gmail.com [f] PhD, professor, Universidade Estadual de Campinas, Campinas, SP - Brazil, e-mail: ricardomachado12@yahoo.com.br [R] Abstract Objective: To analyze the values of maximal inspiratory pressure (MIP) and sniff nasal inspiratory pres- sure (SNIP) and to verify the existence of concordance between the two evaluation methodologies, in sub- jects with tetraplegia. Materials and methods: Cross-sectional study with 17 tetraplegic men, aged 30.42 ± 7.67 years, who underwent MIP and SNIP evaluation using a respiratory pressure meter. Results: The MIP and SNIP values obtained showed no difference when compared to each other (88.42 ± 29.39 vs. 86.68 ± 25.40 cmH2O, respectively). They were, however, significantly lower compared to the predicted values (MIP = 128.92 ± 7.18; SNIP = 114.11 ± 3.19 cmH2O), with the MIP values presenting correlation (r2 = 0.94; p < 0.0001) and concordance with those of the SNIP. Conclusions: Both the MIP and SNIP values obtained were lower than the predicted values, indicating a reduction in inspiratory muscle strength (IMS). Both Marlene Aparecida Moreno[a], Juliana Viana Paris[b], Raphael do Nascimento Pereira[c], Antonio Roberto Zamunér[d], Tais Mendes de Camargo[e], Ricardo Machado Leite de Barros[f] [b] MSc, Universidade Estadual de Campinas, Campinas, SP - Brazil, e-mail: julianav.paris@gmail.com [c] PhD candidate, Universidade Metodista de Piracicaba, Piracicaba, SP - Brazil, e-mail: raphaelnpfisio@yahoo.com.br [d] PhD candidate, Universidade Federal de São Carlos, São Carlos, SP - Brazil, e-mail: beto.zam@gmail.com [e] PhD candidate, Universidade Metodista de Piracicaba, Piracicaba, SP - Brazil, e-mail: taismendesdecamargo@gmail.com [f] PhD, professor, Universidade Estadual de Campinas, Campinas, SP - Brazil, e-mail: ricardomachado12@yahoo.com.br Resumo Objetivo: Analisar os valores da pressão inspiratória máxima (PImáx) e pressão inspiratória nasal sniff (Pnsn), bem como verificar a existência de concordância entre as duas metodologias de avaliação, em sujeitos com tet- raplegia. Materiais e métodos: Estudo transversal com 17 homens tetraplégicos, idade 30,42 ± 7,67 anos, os quais foram submetidos a avaliação da PImáx e da Pnsn, por intermédio da manovacuometria. Resultados: Os valores obtidos da PImáx e da Pnsn não apresentaram diferença quando comparados entre si (88,42 ± 29,39 vs. 86,68 ± 25,40 cmH2O, respectivamente), no entanto, foram significativamente menores em relação aos valores preditos (PImáx = 128,92 ± 7,18; Pnsn = 114,11 ± 3,19 cmH2O), tendo os valores da PImáx apresentado correlação (r2 = 0,94; p < 0,0001) e concordância com os da Pnsn. Conclusões: Tanto os valores obtidos pela PImáx quanto pela Pnsn mostraram-se inferiores aos preditos, indicando redução da força muscular inspiratória (FMI). As duas técnicas apresentaram correlação e concordância, sugerindo que a PImáx pode ser utilizada como um mé- todo não invasivo para avaliação da FMI nesta população.[K] Palavras-chave: Doenças da medula espinhal. Testes de função respiratória. Músculos respiratórios. Força muscular. Abstract Objective: To analyze the values of maximal inspiratory pressure (MIP) and sniff nasal inspiratory pres- sure (SNIP) and to verify the existence of concordance between the two evaluation methodologies, in sub- jects with tetraplegia. Materials and methods: Cross-sectional study with 17 tetraplegic men, aged 30.42 ± 7.67 years, who underwent MIP and SNIP evaluation using a respiratory pressure meter. Results: The MIP and SNIP values obtained showed no difference when compared to each other (88.42 ± 29.39 vs. 86.68 ± 25.40 cmH2O, respectively). They were, however, significantly lower compared to the predicted values (MIP = 128.92 ± 7.18; SNIP = 114.11 ± 3.19 cmH2O), with the MIP values presenting correlation (r2 = 0.94; p < 0.0001) and concordance with those of the SNIP. Conclusions: Both the MIP and SNIP values obtained were lower than the predicted values, indicating a reduction in inspiratory muscle strength (IMS). Both Objective: To analyze the values of maximal inspiratory pressure (MIP) and sniff nasal inspiratory pres- sure (SNIP) and to verify the existence of concordance between the two evaluation methodologies, in sub- jects with tetraplegia. Materials and methods: Cross-sectional study with 17 tetraplegic men, aged 30.42 ± 7.67 years, who underwent MIP and SNIP evaluation using a respiratory pressure meter. Results: The MIP and SNIP values obtained showed no difference when compared to each other (88.42 ± 29.39 vs. 86.68 ± 25.40 cmH2O, respectively). They were, however, significantly lower compared to the predicted values (MIP = 128.92 ± 7.18; SNIP = 114.11 ± 3.19 cmH2O), with the MIP values presenting correlation (r2 = 0.94; p < 0.0001) and concordance with those of the SNIP. Conclusions: Both the MIP and SNIP values obtained were lower than the predicted values, indicating a reduction in inspiratory muscle strength (IMS). Both Fisioter Mov. 2014 jul/set;27(3):371-7 Moreno MA, Paris JV, Pereira RN, Zamunér AR, Camargo TM, Barros RML. 372 techniques showed correlation and concordance, suggesting that MIP can be used as a noninvasive method for IMS evaluation in this population. [P] techniques showed correlation and concordance, suggesting that MIP can be used as a noninvasive method for IMS evaluation in this population. [P] Keywords: Spinal cord diseases. Respiratory function tests. Respiratory muscles. Muscle strength Fisioter Mov. 2014 jul/set;27(3):371-7 Palavras-chave: Doenças da medula espinhal. Testes de função respiratória. Músculos respiratórios. Força muscular. Evaluation protocols MIP values were obtained using a previously cal- ibrated digital respiratory pressure meter (model MVD300, GlobalMed, Porto Alegre, Brazil). The mea- surements were made with the subjects seated in their own wheelchairs, with their feet supported, and hips and knees at 90°. To measure MIP, the nostrils were occluded using a nasal clip. The measurement was performed during maximal inspiratory effort initiated from functional residual capacity (7). The volunteer breathed through a mouthpiece contain- ing an adapter with a 2 mm opening. At least five, technically satisfactory, maximal inspiration efforts were carried out, i.e. without perioral air leakage, with inhalation maintained for at least 1s, and with values similar to each other (≤ 10%), with the highest value considered for the study (16). All the volunteers underwent a preliminary evalu- ation which consisted of obtaining personal, demo- graphic and anthropometric data. Ethical considerations ) For the performance of the measurements, the subject rested for 10 minutes prior to the test and the procedures were carefully explained to him. The examination was performed with the volun- teer sitting in his own wheelchair, with the back and seat fixed, providing a 90° angle of hip flexion, the head was held in a neutral position and a nose clip used to prevent air leakage from the nostrils. The mouthpiece was attached to the mouth of the subject, avoiding any air leakage. The forced vital capacity (FVC) maneuver was performed until three acceptable and two reproducible curves were ob- tained. During its execution, real-time graphs of the curves were provided, indicating whether they met the acceptance criteria proposed by the ATS. The criteria were adjusted for subjects with spinal cord injury due to the muscle dysfunction present, with adjustments to the expiratory time and back- extrapolated volume being required (14, 15). The FVC reference values were obtained from the spi- rometry test, forced expiratory volume in the first second (FEV1) and the ratio FEV1/FVC. The present study was approved by the human research ethics committee of the Institution, under protocol No. 55/08. The volunteers that agreed to participate signed the Terms of Free Prior Informed Consent, according to the criteria of Resolution 196/96 of the National Health Council. Design Cross-sectional study. Study group and inclusion criteria The sample was composed of 17 quadriplegic male volunteers, with a mean age of 30.42 ± 7.67 years. The following inclusion criteria were adopted: aged between 20 and 40 years, time greater than 12 months since injury, complete spinal cord injury (“A” classification in the scale of the American Spinal Injury Association - ASIA), stable clinical condition, non- smoker, and not presenting acute respiratory compli- cations or a history of cardiorespiratory diseases. The exclusion criteria considered were: spinal cord injury below C8 and inability to comprehend the protocol. Materials and methods acceptability and reproducibility, using a spirom- eter (Easy OneTM, ndd Medizintechnik AG, Zurich, Switzerland). The system was calibrated before each test, according to the manufacturer’s instructions. Spirometric variables were recorded and expressed in BTPS conditions (Body temperature and pressure saturated). Introduction pressure (MIP), new methods of non-invasive evalu- ation have been proposed, with the sniff nasal in- spiratory pressure (SNIP) test being one of these, which correlates with esophageal pressure, consid- ered the gold standard for evaluating IMS, in healthy individuals and patients with musculoskeletal and neuromuscular diseases (7, 10). In healthy subjects, SNIP can be used as the maneuver of choice for evalu- ating IMS, as it reproduces the predicted values (11). Thus, it was hypothesized that from the evaluation of SNIP it would be possible to verify the viability of IMS measurement in tetraplegic people through MIP measurement, as, in the presence of low MIP values, SNIP offers a way to differentiate inspiratory muscle weakness from the difficulty in achieving sustained continual effort (7). Subjects with spinal cord injury present serious alterations in respiratory function due to sensory and motor impairment below the level of the injury, these alterations are characterized by paralysis or respiratory muscle weakness (1, 2). This impairment can be evidenced by the considerable decrease in respiratory muscle strength, evaluated by measuring the maximal respiratory pressures (3, 4). However, in people with spinal cord injury, the values obtained from these measurements are questioned, as they are too variable to be considered as an index of inspira- tory muscle strength (IMS) (5). Alterations in ventilatory mechanics and in IMS may hinder the performance of this maneuver and thus promote inadequate results (6, 7). Since these are volitional tests, the measures depend on the comprehension and cooperation of the subject to perform maximal voluntary efforts (8, 9), and thus, underestimated values are set when faults occur in the comprehension, motivation, and coordination, or when the patient presents fatigue or discomfort while performing the maneuvers (7). Aiming to minimize the difficulties encountered in the measurement of IMS from the measurement of maximal inspiratory Considering that the functional evaluation of the inspiratory muscles is important for determining the degree of dysfunction of these muscles in people with spinal cord injury, and given the need to clar- ify the viability of using MIP for the evaluation of IMS in this population, this study aimed to evaluate MIP and SNIP and to verify the existence of correla- tion and concordance between the values obtained from the two methodologies in a sample of subjects with tetraplegia. Fisioter Mov. 2014 jul/set;27(3):371-7 Inspiratory muscle strength in subjects with tetraplegia 373 Results Table 1 presents the study sample characterization. Table 2 shows the predicted and obtained MIP and SNIP values, in which no statistically signifi- cant differences can be observed when comparing the values obtained for the two study variables. In the comparison between the predicted and ob- tained values, regarding both MIP and SNIP, the obtained values were lower than the predicted values (p < 0.05). In the relation analysis between MIP and SNIP, the results showed positive, statistically signifi- cant correlation, and concordance between the values obtained from the two evaluation methods (Figure 1). Sniff nasal inspiratory pressure distribution was conducted using the Shapiro-Wilk test. Student’s t-test was used to compare the values obtained from the two methodologies. To verify the relationship between the variables, Pearson’s cor- relation coefficient was used, and for the analysis of concordance between the methodologies, the Bland-Altman method (19) was used. A statisti- cal significance level of α=5% was adopted. The statistical procedures were performed using the GraphPad InStat version 3.05 and Medcalc version 11.5.0 applications. The SNIP was measured with the volunteers in the same position. The measurement was made with one nostril occluded with a nasal silicone plug, which re- mained connected to the respiratory pressure meter by a catheter with an approximate diameter of 1mm (17). The maneuver consisted of one maximum sniff performed through the contralateral (free) nostril, with the mouth closed, from functional residual ca- pacity. Ten sniff test maneuvers were performed (18), with an interval of 30 seconds between each one, with the selection criterion used for an acceptable sniff being, a gradual pressure increase until reaching the peak, with a duration of up to 5 seconds (7). All the values were recorded on individual forms for each subject and the highest value was used for the data analysis. The volunteers were evaluated in a random order according to the methodologies used, and, at the end of the first examination, rested for 30 minutes prior to performing the subsequent measurement. The predicted MIP (16) and SNIP (7) values were based on equations for healthy subjects. Pulmonary function In order to verify the presence of respiratory dis- turbances and thus characterize the population, the equations for the prediction of normal values, based on the equations for healthy subjects were used in accordance with the Guidelines for the pulmonary function tests (12). The pulmonary function tests were performed according to the guidelines of the American Thoracic Society - ATS (13) for technique, Fisioter Mov. 2014 jul/set;27(3):371-7 Moreno MA, Paris JV, Pereira RN, Zamunér AR, Camargo TM, Barros RML. 374 Statistical analysis Table 2 - Maximal inspiratory pressure (MIP) and Sniff nasal inspiratory pressure (SNIP) predicted and obtained values MIP (cmH2O) SNIP (cmH2O) Predicted values 128.92 ± 7.18 114.11 ± 3.19 Table 2 - Maximal inspiratory pressure (MIP) and Sniff nasal inspiratory pressure (SNIP) predicted and obtained values MIP (cmH2O) SNIP (cmH2O) Table 2 - Maximal inspiratory pressure (MIP) and Sniff nasal inspiratory pressure (SNIP) predicted and obtained values Table 2 - Maximal inspiratory pressure (MIP) and Sniff nasal inspiratory pressure (SNIP) predicted and obtained values MIP (cmH2O) SNIP (cmH2O) Predicted values 128.92 ± 7.18 114.11 ± 3.19 Obtained values 88.42 ± 29.39* 86.68 ± 25.40* Note: * = p < 0.05: Predicted values vs. Obtained values. Source: Research data. 160 140 120 100 80 60 40 40 R2 = 0.94 P < 0.0001 60 SNIP (cmH2O) MIP (cmH2O) MIP and SNIP difference Mean between MIP and SNIP (cmH2O) 80 100 120 140 40 60 80 100 120 140 160 15 20 10 5 -5 0 -10 -1.96 SD +1.96 SD 17.0 13.6 11.7 Mean -15 -20 Figure 1 - Graphical representation of the correlation analysis between Maximal inspiratory pressure (MIP) and Sniff nasal inspiratory pressure (SNIP), in cmH2O, and scatter plot for the difference and mean between and MIP and SNIP variables of the 17 tetraplegic volunteers studied Source: Research data. 160 140 120 100 80 60 40 40 R2 = 0.94 P < 0.0001 60 SNIP (cmH2O) MIP (cmH2O) 80 100 120 140 MIP and SNIP difference Mean between MIP and SNIP (cmH2O) 40 60 80 100 120 140 160 15 20 10 5 -5 0 -10 -1.96 SD +1.96 SD 17.0 13.6 11.7 Mean -15 -20 Figure 1 - Graphical representation of the correlation analysis between Maximal inspiratory pressure (MIP) and Sniff nasal inspiratory pressure (SNIP), in cmH2O, and scatter plot for the difference and mean between and MIP and SNIP variables of the 17 tetraplegic volunteers studied Source: Research data. Fisioter Mov. 2014 jul/set;27(3):371-7 Statistical analysis The values were expressed as mean and standard deviation, and the analysis of the data Table 1 - Characteristics of the volunteers studied (n = 17) Variables Values Age (years) 30.42 ± 7.67 Height (cm) 164.89 ± 38.31 Body weight (kg) 81.78 ± 34.22 BMI (kg/m2) 22.30 ± 3.63 Injury level C4 - C7 Injury duration (months) 84.36 ± 49.13 FVC (%) 51.29 ± 15.24 FEV1 (%) 52.52 ± 21.28 FEV1/FVC 100.82 ± 19.70 Note: BMI = body mass index; FVC = forced vital capacity; FEV1 = forced expiratory volume in ﬁ rst second. Source: Research data. Table 1 - Characteristics of the volunteers studied (n = 17) Variables Values Age (years) 30.42 ± 7.67 Height (cm) 164.89 ± 38.31 Body weight (kg) 81.78 ± 34.22 BMI (kg/m2) 22.30 ± 3.63 Injury level C4 - C7 Injury duration (months) 84.36 ± 49.13 FVC (%) 51.29 ± 15.24 FEV1 (%) 52.52 ± 21.28 FEV1/FVC 100.82 ± 19.70 Note: BMI = body mass index; FVC = forced vital capacity; FEV1 = forced expiratory volume in ﬁ rst second. Source: Research data. Table 1 - Characteristics of the volunteers studied (n = 17) Fisioter Mov. 2014 jul/set;27(3):371-7 Inspiratory muscle strength in subjects with tetraplegia 375 Table 2 - Maximal inspiratory pressure (MIP) and Sniff nasal inspiratory pressure (SNIP) predicted and obtained values MIP (cmH2O) SNIP (cmH2O) Predicted values 128.92 ± 7.18 114.11 ± 3.19 Obtained values 88.42 ± 29.39* 86.68 ± 25.40* Note: * = p < 0.05: Predicted values vs. Obtained values. Source: Research data. 160 140 120 100 80 60 40 40 R2 = 0.94 P < 0.0001 60 SNIP (cmH2O) MIP (cmH2O) MIP and SNIP difference Mean between MIP and SNIP (cmH2O) 80 100 120 140 40 60 80 100 120 140 160 15 20 10 5 -5 0 -10 -1.96 SD +1.96 SD 17.0 13.6 11.7 Mean -15 -20 Figure 1 - Graphical representation of the correlation analysis between Maximal inspiratory pressure (MIP) and Sniff nasal inspiratory pressure (SNIP), in cmH2O, and scatter plot for the difference and mean between and MIP and SNIP variables of the 17 tetraplegic volunteers studied Source: Research data. Discussion There are publications on the use of SNIP in healthy subjects (7), neuromuscular disease patients (18, 24, 25) and people with chronic obstructive pul- monary disease (26), however, studies with spinal cord injury patients are still scarce. Authors (27) observed that SNIP correlated positively and statis- tically significantly with the degree of spinal cord injury. They highlighted that SNIP proved to be bet- ter than MIP in discriminating the effect of the spinal cord injury level on IMS. From the preliminary re- sults, the authors also concluded that SNIP is a more accurate measure of inspiratory function than MIP, which differs from the results of the present study, in which no difference between the values obtained through the two methods were found, however, the comparison between the two studies becomes unvi- able when considering the methodological differ- ences between them. indicate viability for the use of MIP for evaluating IMS in the study population, considering that in the presence of low MIP values, SNIP offers a way to differentiate inspiratory muscle weakness from dif- ficulty in achieving sustained continual effort (7). There are publications on the use of SNIP in healthy subjects (7), neuromuscular disease patients (18, 24, 25) and people with chronic obstructive pul- monary disease (26), however, studies with spinal cord injury patients are still scarce. Authors (27) observed that SNIP correlated positively and statis- tically significantly with the degree of spinal cord injury. They highlighted that SNIP proved to be bet- ter than MIP in discriminating the effect of the spinal cord injury level on IMS. From the preliminary re- sults, the authors also concluded that SNIP is a more accurate measure of inspiratory function than MIP, which differs from the results of the present study, in which no difference between the values obtained through the two methods were found, however, the comparison between the two studies becomes unvi- able when considering the methodological differ- ences between them. healthy subjects were used, which may overestimate the difference between the predicted and obtained values. Another point to consider is the impossibil- ity of performing a gold standard test (esophageal pressure) in order to compare the MIP and SNIP values obtained. Conclusion Both the MIP and SNIP values obtained proved lower than those predicted, indicating a reduction in IMS. However, despite being reported in the literature that MIP may not be an adequate technique for the evaluation of IMS in subjects with spinal cord injury, in the present study these values presented correla- tion and concordance with SNIP, which is referred to as a more accurate measure of inspiratory function in spinal cord injury patients, suggesting that MIP can be used as a noninvasive method for the evaluation of IMS in this population. Discussion predicted can be justified by the spinal cord injury, depending on the injury level, compromising the re- spiratory muscles, considering that the main inspi- ratory muscles have a higher location in the torso and are innervated by the upper spinal segments, thus the higher the level of the spinal cord injury, the greater the compromise to the inspiratory muscles (23). Another study evaluated the influence of the level of severity (partial or complete) and the dura- tion of the spinal cord injury on the pulmonary func- tion of quadriplegics and paraplegics, demonstrating that the higher, more severe, and longer duration of the injury, the worse the results obtained in the pul- monary function test, with the severity of the injury the factor of greatest importance in the reduction of pulmonary function (2). In the present study, to better characterize the vol- unteers, the spirometric evaluation was performed at the beginning of the experimental protocol and the results indicated that the lung function of all partici- pants was reduced compared to the predicted val- ues. Changes in lung volumes and capacity are often caused by a muscle deficit (20), with the degree of compromise of these muscles dependant on the spi- nal cord injury level (21). The lack of coordination in the activation of the respiratory muscles, and the reduction in the volumes and capacities characterizes the non-parenchymal restrictive syndrome presented by this population (22). The MIP and SNIP values obtained presented no difference when compared with each other, however, they were significantly lower compared to the pre- dicted values, with the MIP values showing corre- lation and concordance with those of the SNIP. The maximal inspiratory pressure values below those Despite the difficulties mentioned in the litera- ture regarding the measurement of MIP in people with spinal cord injuries, the significant correlation and concordance between the MIP and SNIP values obtained in this study are noteworthy since they Fisioter Mov. 2014 jul/set;27(3):371-7 Moreno MA, Paris JV, Pereira RN, Zamunér AR, Camargo TM, Barros RML. 376 indicate viability for the use of MIP for evaluating IMS in the study population, considering that in the presence of low MIP values, SNIP offers a way to differentiate inspiratory muscle weakness from dif- ficulty in achieving sustained continual effort (7). Fisioter Mov. 2014 jul/set;27(3):371-7 References Souza RB. Pressões respiratórias estáticas máxi- mas. J Pneumol. 2002;28(3):155-65. 22. De Troyer A. Respiration mechanics in tetraplegia. Bull Mem Acad R Med Belg. 1997;152(1):91-7. 10. Morgan RK, McNally S, Alexander M, Conroy R, Hardiman O, Costello RW. Use of Sniff nasal-inspirato- ry force to predict survival in amyotrophic lateral scle- rosis. Am J Respir Crit Care Med. 2005;171(3):269-74. 23. Mueller G, Groot S, Woude LVD, Hopman MTE. Time- courses of lung function and respiratory muscle pressure generating capacity after spinal cord inju- ry: a prospective cohort study. J Rehabil Med. 2008; 40(4):269-76. 11. Prigent H, Lejaille M, Falaize L, Louis A, Ruquet M, Fauroux B, et al. Assessing inspiratory muscle strength by sniff nasal inspiratory pressure. Neurocritical Care. 2004;1(4):475-8. 24. Fitting JW, Paillex R, Hirt L, Aebischer P, Schluep M. Sniff nasal pressure: a sensitive respiratory test to assess progression of amyotrophic lateral sclerosis. Ann Neurol. 1999;46(6):887-93. 12. Diretrizes para testes de função pulmonar. J Pneumol. 2002;28(Suppl 3):1-82. 25. Stefanutti D, Benoist MR, Scheinmann P, Chaussain M, Fitting JW. Usefulness of sniff nasal pressure in pa- tients with neuromuscular or skeletal disorders. Am J Respir Crit Care Med. 2000;162(4 Pt 1):1507-11. 13. American Thoracic Society. Standardization of spi- rometry 1994 update. Am J Respir Crit Care Med. 1995;152(3):1107-36. 14. Ashba J, Garshick E, Tun CG, Lieberman SL, Polakoff DF, Blanchard JD, et al. Spirometry-acceptability and reproducibility in spinal cord injured subjects. J Am Paraplegia Soc. 1993;16(4):197-203. 26. Kyroussis D, Johnson L C, Hamnegard CH, Polkey MI, Moxham J. Inspiratory muscle maximum relaxation rate measured from submaximal sniff nasal pres- sure in patients with severe COPD. Thorax. 2002; 57(3):254-7. 15. Kelley A, Garshick E, Gross ER, Lieberman SL, Tun CG, Brown R. Spirometry testing standards in spinal cord injury. Chest. 2003;123(3):725-30. 27. Rocha AP, Mateus SRM, Horan TA, Beraldo PSS. De- terminação não-invasiva da pressão inspiratória em pacientes com lesão medular traumática: qual é o melhor método? J Bras Pneumol. 2009;35(3):256-60. 16. Neder JA, Andreoni S, Lerario MC, Nery LE. Reference values for lung function tests. II. Maximal respiratory pressures and voluntary ventilation. Braz J Med Biol Res. 1999;32(6):719-27. 28. Nava S, Ambrosino N, Crotti P, Fracchia C, Rampulla C. Recruitment of some respiratory muscles during three maximal inspiratory manoeuvres. Thorax. 1993; 48(7):702-7. 17. Ruppel G. Lung volume tests. In: Mottram C. Manual of Pulmonary Function Testing. St Louis: Mosby; 1994. p. 1-25. 18. References In addition to SNIP allowing the differentiation between inspiratory muscle weakness and difficulties in performing the technique, it has other advantages, especially in individuals with tetraplegia, as SNIP con- sists of a simpler technique than that performed for the MIP measurement, with a lower risk of fatigue due to being a natural, easy and short maneuver, in which peak pressure needs to be sustained for a shorter duration (11). Furthermore, the determina- tion of SNIP is considered a natural and easy proce- dure, which allows those evaluated to activate the inspiratory muscles with increased muscle fiber re- cruitment when compared to MIP, 100% and 61.1%, respectively, according to a study that evaluated the amplitude of the peak electromyography activity of the diaphragm muscle during the performance of the two methods (28). 1. Hopman MT, Van der Woude LH, Dallmeijer AJ, Snoek G, Folgering HT. Respiratory muscle strength and en- durance in individuals with tetraplegia. Spinal Cord. 1997;35(2):104-8. 2. Linn WS. Adkins RH, Gong H Jr., Waters RL. Pulmonary function in chronic spinal cord injury: a crosssectional survey of a large southern California outpatient popu- lation. Arch Phys Med Rehabil. 2000;81(6):757-63. 3. Mansel JK, Norman JR. Respiratory complications and management of spinal cord injuries. Chest. 1990; 97(6):1446-52. 4. Gounden P. Static respiratory pressures in patients with post-traumatic tetraplegia. Spinal Cord. 1997; 35(1):43-7. Considering that the procedures used to perform the MIP and SNIP measurements were not equiva- lent, it is worth highlighting that in the present in- vestigation care was taken so that the measurements started from the same lung volume, i.e., functional residual capacity, thus avoiding the promotion of differences in the activation of inspiratory muscle patterns. Regarding the limitations of this study, the interpretation of the results presented should be taken into consideration, as predictive equations for 5. 5. Goldman JM, Rose LS, Williams SJ, Silver JR, Denison DM. Effect of abdominal binders on breathing in tet- raplegic patients. Thorax. 1986;41(12):940-5. 6. Leith DE, Bradley M. Ventilatory muscle strength and endurance training. J Appl Physiol. 1976;41(4):508-16. 7. Uldry C, Fitting JW. Maximal values of sniff nasal in- spiratory pressure in healthy subjects. Thorax. 1995; 50(4):371-5. Inspiratory muscle strength in subjects with tetraplegia 377 8. Syabbalo N. Assessment of respiratory muscle function and strength. Postgrad Med J. 1998;74(870):208-15. 21. Winslow C, Rozovsky J. Effect of spinal cord injury on the respiratory system. Am J Phys Med Rehabil. 2003; 82(10):803-14. 9. Fisioter Mov. 2014 jul/set;27(3):371-7 References Lofaso F, Nicot F, Lejaille M, Falaize L, Louis A, Clement A, et al. Sniff nasal inspiratory pressure: with is the optimal number of sniffs? Eur Respir J. 2006;27(5):980-2. Received: 06/15/2013 Recebido: 15/06/2013 Approved: 12/23/2013 Aprovado: 23/12/2013 19. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measure- ment. Lancet. 1986;1(8476):307-10. 20. Roth EJ, Lu A, Primack S, Oken J, Nusshaum S, Berkowitz M, et al. Ventilatory function in cervical and high tho- racic spinal cord injury. Relationship to level of injury and tone. Am J Phys Med Rehabil. 1997; 76(4):262-7.
https://openalex.org/W2619477097	https://europepmc.org/articles/pmc5446758?pdf=render	English	null	Management of intra-operative acute pulmonary embolism during general anesthesia: a case report	BMC anesthesiology	2,017	cc-by	3,524	© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Abstract Background: Acute pulmonary embolism (APE) can be life-threatening. Early detection is even more difficult for patients under general anesthesia as common symptoms are not available and the pathophysiological course of intra-operative APE is influenced by procedures of surgery and anesthesia, which makes patients under general anesthesia a distinctive group. Case presentation: We report a case of APE during orthopedic surgery under general anesthesia. A 64-year-old female with atrial fibrillation and surgical history of varicosity underwent total right hip replacement surgery under general anesthesia. No arterial or deep vein thrombosis (DVT) was found prior to the surgery, but APE still occurred intraoperatively. The sudden decrease in PETCO2 and increase in PaCO2 combined other clues raised the suspect of APE, which is further evidenced by transesophageal echocardiogram (TEE). Multidisciplinary consultation was started immediately. After discussion with the consultation team and communication with patient’s family members, anticoagulation therapy was started and IVC filter was placed to prevent PE recurrence. The patient went through the operation and discharged uneventfully 30 days later. Conclusions: Pulmonary embolism is a rare and potentially high-risk perioperative situation, with a difficult diagnosis when occurs under anesthesia. The separation phenomenon of decrease in PETCO2 and increase in PaCO2 might be a useful and suggestive sign, enabling prompt management and therefore improving the prognosis. Keywords: Intra-operation, Acute pulmonary embolism, General anesthesia, Multidisciplinary con Keywords: Intra-operation, Acute pulmonary embolism, General anesthesia, Multidisciplinary consultation Background replacement surgery, that provides us with a meaningful phenomenon—an opposite trend in increased PaCO2 and decreased PETCO2—as an early sign of potential APE for patients undergoing general anesthesia and mechanical ventilation. Acute pulmonary embolism (APE) remains a significant medical problem, relating to more than 300,000 deaths per year in Europe [1]. Early detection and treatment are vital for a better prognosis [2]. Signs and symptoms such as dyspnea, chest pain, hemoptysis and syncope could be clues for APE detection [3, 4]. But they are not available for patients under general anesthesia, which increases the difficulty for early diagnosis and stresses the need for other indications. Moreover, the pathophysiologic course of intraoperative APE is affected by procedures of operation and anesthesia, which might be different from conscious patients with spontaneous respiration. The following report describes a case of APE during hip Management of intra-operative acute pulmonary embolism during general anesthesia: a case report Yuanyuan Mao1, Shuai Wen2, Gezi Chen1, Wei Zhang1, Yanqiu Ai1 and Jingjing Yuan1* Mao et al. BMC Anesthesiology (2017) 17:67 DOI 10.1186/s12871-017-0360-0 Mao et al. BMC Anesthesiology (2017) 17:67 DOI 10.1186/s12871-017-0360-0 * Correspondence: yjingjing_99@163.com 1Department of Anesthesiology, First Affiliated Hospital of Zhengzhou University, No.1 East-JianShe Road, Zhengzhou 450052, China Full list of author information is available at the end of the article Case presentation T2 time point was 9 min after the T1 time point p q y y Therefore she was scheduled for surgery and trans- ferred to the operating room 3 days after admission. Preoxygen and monitoring were performed, which showed BP 110/80 mmHg, HR 140 bpm and SpO2 98%. Considering her anxiety and tachycardia, 10 mg esmolol and 60 μg dexmedetomidine were given intravenously in 10 min. Soon the heart rate slowed down and stabled at 110 bpm. Arterio radialis puncture was performed for invasive blood pressure monitoring under 1% lidocaine local anesthesia. For anesthesia induction, midazolam (0.03 mg/kg), fentanyl (4 μg/kg), etomidate (0.25 mg/kg) and cisatracurium (0.2 mg/kg) were administered. After in- tubation, mechanical ventilation started with volume con- trol model (RR:12 breaths per minute, tidal volume:6-8 ml/ kg, PETCO2:35-45 mmHg). General anesthesia was main- tained with sevoflurane (1%), propofol (4-10 mg/kg/h), remifentanil (0.1–0.2 μg/kg/min), cisatracurium (0.1 mg/ kg/h) and dexmedetomidine (0.5 μg/kg/h). A central ven- ous catheter was placed in the right internal jugular vein under ultrasonic guidance in case of mass transfusion and vasoactive drugs application intraoperatively. Patient was then placed in left lateral decubitus position and the operation began. enlargement of the right atrial, abnormal echoic area in pulmonary artery (considering thrombus) and pulmon- ary hypertension (pulmonary arterial systolic pressure: 77 mmHg) (Additional file 1: Movie S1). g) ( ) Multidisciplinary consultation started simultaneously. Mechanical and manual ventilation were controlled alternately. Cedilanid (0.2 mg), phenylephrine (0.03– 0.06 mg/kg/min), 5% NaHCO3 liquid (100 ml) and methylprednisolone (500 mg) were administered intra- venously to maintain hemodynamic stability and im- prove internal environment. All fluids were limited to reduce the cardiac load. 11 min later, SpO2 increased to 95% and BP rose to 120/80 mmHg. A following systematic ultrasound examination revealed deep venous thrombosis, which was not found before the surgery and might be the source of pulmonary embolism. After discussion with sur- gical and respiratory consultation team and communica- tion with the patient’s family members, low molecular weight heparin calcium(LMWHC) (6000 IU) was given and inferior vena cava (IVC) filter was implanted in DSA operating room. After surgery, the patient was transferred to ICU intubated. Patient recovered her consciousness 3 h later and extubated the next morning. Further treatment including anticoagulation, alleviation of pulmonary hypertension and atrial fibrillation therapy was contin- ued. IVC filter was removed 20 days later. Case presentation Her condi- tion was improved and all vital signs maintained stable. The patient was discharged from hospital uneventfully one month later. Approximately 1 h later, when the surgery was almost finished and the incision was closing, a sudden pro- nounced decrease in oxygen saturation (100%–66%) was noticed, along with gradually increased airway pres- sure(22-33cmH2O) and decrease in blood pressure (lowest to 89/60 mmHg). PETCO2 decreased from 34 mmHg to 22 mmHg. ECG showed rapid atrial fibril- lation (heart rate 140-160 bpm). Auscultation revealed bilateral breath sounds. Central venous pressure (CVP) was assessed immediately with a value of 30cmH2O. In- stant arterial blood gas analysis showed hypercapnia (Table 1). The opposite trend of PaCO2 and PETCO2, combined with other signs and medical history, indi- cated the possibility of APE. Transesophageal echocar- diogram (TEE) was arranged at once, which showed Case presentation A 64-year-old, BMI 32.7 kg/m2, American Society of Anesesthesiolgists (ASA) physical status III, female was scheduled for right hip replacement because of bone fracture. She received a surgery for bilateral lower extremity varicose veins 8 years ago and recovered well. There is no past medical history of hypertension, diabetes mellitus or cerebrovascular diseases. Patient had no drug or food allergies, never smoked cigarettes or consumed alcohol. * Correspondence: yjingjing_99@163.com 1Department of Anesthesiology, First Affiliated Hospital of Zhengzhou University, No.1 East-JianShe Road, Zhengzhou 450052, China Full list of author information is available at the end of the article Chest X-ray showed heart shadow enlargement (cardiothoracic ratio: 0.65) and slight pleural effusion left Mao et al. BMC Anesthesiology (2017) 17:67 Page 2 of 4 Table 1 The arterial blood gas analysis of the patient Table 1 The arterial blood gas analysis of the patient Parameter T1 T2 pH 7.161 7.231 pCO2(mmHg) 66.9 62.7 pO2(mmHg) 40.7 76.4 SO2(%) 65.9 92.7 cLac(mmol/L) 0.6 0.6 ctO2(mmol/L) 12.6 17.8 p50(mmHg) 32.19 31.75 HCO3std(mmol/L) 18.3 21.8 HCO3−(mmol/L) 22.9 25.3 ABE(mmol/L) −6.8 −3.1 SBE(mmol/L) −4.6 -1.3 T1: the time point when respiratory and hemodynamic changes occurred. T2: the time point when the patient’s condition became stable. T2 time point was 9 min after the T1 time point side. Echocardiogram showed bilateral atrial dilatation, atrial fibrillation, slight bicuspid and tricuspid regurgita- tion, right pulmonary mild hypertension (pulmonary arterial systolic pressure: 51 mmHg) and a decreased systolic function of left ventricular. ECG showed acceler- ation of ventricular rate, atrial fibrillation and T wave non specific changes. D-dimer was 2.517 μg/ml (normal: 0–0.3 μg/ml). Other laboratory assessments were within normal limits. Despite abnormal findings in pre-procedure tests, no uncomfortable symptoms were reported. She emphasized to be in good condition before the fracture that she could do Dama square dance for 2 h every day after supper. There were no complaints except for the pain due to bone fracture. She denied further examination or treatment and insisted on an operation for the fracture as soon as possible, as it reduced her life quality severely. T1: the time point when respiratory and hemodynamic changes occurred. T2: the time point when the patient’s condition became stable. T2 time point was 9 min after the T1 time point T1: the time point when respiratory and hemodynamic changes occurred. T2: the time point when the patient’s condition became stable. Discussion Here we described a case of intra-operative APE and re- ported a meaningful sign –“separation phenomenon” during general anesthesia, which means a decrease trend in PETCO2 and increase in PaCO2. This is different from Mao et al. BMC Anesthesiology (2017) 17:67 Page 3 of 4 Page 3 of 4 conscious patients with incomplete obstruction of pul- monary artery, whose physiological dead space increases and total minute ventilation increases correspondingly. In these patients, PaCO2 is mostly lower than normal because of the compensatory mechanism. However, this mechanism does not work under general anesthesia and mechanical ventilation. Thus when APE occurred in patients with mechanical ventilation, PETCO2 does not actually reflect PaCO2 anymore. In fact, the difference of PaCO2 and PETCO2 could be a representative of dead space, which has been studied as a tool to monitor the effect of thrombolysis [5]. Beside dead space increase, low cardiac output and ventilation-perfusion mismatch also contribute to the hypoxemia and hypercapnia [6]. Though not pathognomonic, the separation phenomenon is still valuable. Just like chest pain and dyspnea, these symptoms are not specific, but in a similar context to our case, could be highly suggestive. PETCO2 is a routine mon- itoring under general anesthesia and arterial blood analysis is a quick bedside test. Thus we suggest strict monitoring of PETCO2 and arterial blood analysis especially when PETCO2 decreases without other reasons during general anesthesia. When the separation phenomenon occurred, PE should be taken into consideration. cement embolic events are common and most patients tolerate well [9]. Even if BCIS occurs, there is no specific treatment at present [8]. To confirm the cause of APE might need further test, like computed tomography. It was not conducted in this case, which was a pity, consid- ering the high cost and little effect on the treatment. Combining all the clues, the most possible reason for the hemodynamic and respiratory changes in this case was pulmonary thromboembolism. The treatment re- quires rapid and accurate risk assessment. Thrombolytic treatment restores pulmonary perfusion more rapidly than anticoagulation alone, but it also carries a risk of major bleeding, including intracranial hemorrhage. And the increased age and presence of comorbidity have been associated with higher risk of hemorrhage [10]. Surgery is also a relative contraindication to thrombolysis. Discussion After discussion with the consultation team and communica- tion with patient’s family members, anticoagulation ther- apy was started and IVC filter was placed to prevent PE recurrence. The patient finally discharged uneventfully. The good clinical outcome could be predicted by the lactate value. Plasma lactate concentration is a sensitive marker of tissue hypoxia and persistent tissue hypoxia may cause irreversible cell injury which was associated with the adverse outcome in critically ill patients [11, 12]. Studies showed that plasma lactate was prognostic for short-term PE-related complications and high plasma lac- tate (≥2 mol/L) was associated with increased in-hospital mortality [13–15]. Lactate concentration in this case remained at a low level which might due to our fast detec- tion and treatment. Thus tissue hypoperfusion was tem- porary and reversible. We suspected venous thrombus that led to APE in this patient. First, she had a surgical history of varicosity. Although no thrombus was found before the surgery, it was a risk factor of pulmonary thromboembolism. Sec- ond, intraoperative TEE showed abnormal echoic area in pulmonary artery (considering thrombus) and the fol- lowing systematic ultrasound examination revealed deep venous thrombosis. The most common cause of PE re- mains deep vein thrombosis [7]. Finally, anti-coagulation therapy was useful to improve the condition. Despite updated guidelines, PE remains an important clinical problem with a high mortality rate. Even with prophylaxis, it could still happen and sometimes ap- peared as an emergency [16]. A flow diagram specific to intra-operative APE might be a useful aid, especially for this infrequent and high-risk situation. Also, multidiscip- linary cooperation is necessary to get the most appropri- ate treatment strategy. The initial clinical symptoms in this case could also suggest a pneumothorax. But there was no further evi- dence. Procedure of central venous of catheterization might be a cause for pneumothorax. But the process was guided by ultrasound without any abnormal sign. Also, instant intraoperative auscultation confirmed bilateral breath sounds. Since it was a hip surgery, the intra-operative APE might also be caused by cement embolus. The known bone cement implantation syndrome (BCIS) is com- monly caused by PE, which is characterized by a number of clinical features such as hypoxia, hypotension, cardiac arrhythmias and increased pulmonary vascular resist- ance. It usually occurs at one of these stages of surgical procedure including femoral reaming, acetabular or femoral cement implantation, insertion of the prosthesis or joint reduction [8]. Discussion However, in this case, the hemodynamic changes and hypoxia presented at none of these stages but at the end of operation, when the wound was closing. There are studies showing that Conclusion Pulmonary embolism is a rare and potentially high-risk perioperative situation, with a difficult diagnosis when occurs under anesthesia. The separation phenomenon of decrease in PETCO2 and increase in PaCO2 might be a useful and suggestive sign, enabling prompt manage- ment and therefore improving the prognosis. Author details 1 1Department of Anesthesiology, First Affiliated Hospital of Zhengzhou University, No.1 East-JianShe Road, Zhengzhou 450052, China. 2Department 1Department of Anesthesiology, First Affiliated Hospital of Zhengzhou University, No.1 East-JianShe Road, Zhengzhou 450052, China. 2Department 1Department of Anesthesiology, First Affiliated Hospital of Zhengzhou University, No.1 East-JianShe Road, Zhengzhou 450052, China. 2Department of Modern Education Technology Center, Henan Medical College, No.8 Shuanghu Road, Longhu, Zhengzhou 451191, China. 1Department of Anesthesiology, First Affiliated Hospital of Zhengzhou University, No.1 East-JianShe Road, Zhengzhou 450052, China. 2Department of Modern Education Technology Center, Henan Medical College, No.8 Shuanghu Road, Longhu, Zhengzhou 451191, China. of Modern Education Technology Center, Henan Medical College, No.8 Shuanghu Road, Longhu, Zhengzhou 451191, China. of Modern Education Technology Center, Henan Medical College, No.8 Shuanghu Road, Longhu, Zhengzhou 451191, China. Funding 13. Vanni S, Viviani G, Baioni M, Pepe G, Nazerian P, Socci F, et al. Prognostic value of plasma lactate levels among patients with acute pulmonary embolism: the thrombo-embolism lactate outcome study. Ann Emerg Med. 2013;61(3):330–8. 13. Vanni S, Viviani G, Baioni M, Pepe G, Nazerian P, Socci F, et al. Prognostic value of plasma lactate levels among patients with acute pulmonary embolism: the thrombo-embolism lactate outcome study. Ann Emerg Med. 2013;61(3):330–8. Availability of data and materials y All data and materials described in the manuscript will be freely available to any scientist wishing to use them for non-commercial purposes. Raw data are not available as all are in the patient electronic medical record. Please consider the edited data in the case report. 14. Vanni S, Jimenez D, Nazerian P, Morello F, Parisi M, Daghini E, et al. Short- term clinical outcome of normotensive patients with acute PE and high plasma lactate. Thorax. 2015;70(4):333–8. 14. Vanni S, Jimenez D, Nazerian P, Morello F, Parisi M, Daghini E, et al. Short- term clinical outcome of normotensive patients with acute PE and high plasma lactate. Thorax. 2015;70(4):333–8. 15. Vanni S, Socci F, Pepe G, Nazerian P, Viviani G, Baioni M, et al. High plasma lactate levels are associated with increased risk of in-hospital mortality in patients with pulmonary embolism. Acad Emerg Med. 2011; 18(8):830–5. 15. Vanni S, Socci F, Pepe G, Nazerian P, Viviani G, Baioni M, et al. High plasma lactate levels are associated with increased risk of in-hospital mortality in patients with pulmonary embolism. Acad Emerg Med. 2011; 18(8):830–5. Abbreviations 9. Ereth MH, Weber JG, Abel MD, Lennon RL, Lewallen DG, Ilstrup DM, et al. Cemented versus noncemented total hip arthroplasty–embolism, hemodynamics, and intrapulmonary shunting. Mayo Clin Proc. 1992;67(11):1066–74. APE: Acute pulmonary embolism; CVP: Central venous pressure; DVT: Deep vein thrombosis; IVC: Inferior vena cava; PETCO2: End-tidal CO2 pressure; TEE: Transesophageal echocardiogram 10. Mikkola KM, Patel SR, Parker JA, Grodstein F, Goldhaber SZ. Increasing age is a major risk factor for hemorrhagic complications after pulmonary embolism thrombolysis. Am Heart J. 1997;134(1):69–72. Additional file Additional file 1: Movie S1. A real-time TEE image revealed a mobile embolus in the PA. PA, pulmonary artery; Ao, aorta. (WMV 1mb) Additional file 1: Movie S1. A real-time TEE image revealed a mobile embolus in the PA. PA, pulmonary artery; Ao, aorta. (WMV 1mb) Page 4 of 4 Page 4 of 4 Mao et al. BMC Anesthesiology (2017) 17:67 Competing interests The authors declare that they have no competing interests. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Acknowledgements y 11. Bakker J, Nijsten MW, Jansen TC. Clinical use of lactate monitoring in critically ill patients. Ann Intensive Care. 2013;3(1):12. 11. Bakker J, Nijsten MW, Jansen TC. Clinical use of lactate m critically ill patients. Ann Intensive Care. 2013;3(1):12. We would like to thank Professor Yanli Ma, Bin Li, Yong Wang and Liwei Li for their help in the rescue. critically ill patients. Ann Intensive Care. 2013;3(1):12. 12. Shoemaker WC, Appel PL, Kram HB. Tissue oxygen debt as a determinant of lethal and nonlethal postoperative organ failure. Crit Care Med. 1988; 16(11):1117–20. Authors’ contributions d b d YM and SW contributed equally to this paper: Preparation of the manuscript. GC, WZ and YA participated in patient care. JY involved in the case and drafted the manuscript. All authors read and approved the final manuscript. 16. Goldhaber SZ, Visani L, De Rosa M. Acute pulmonary embolism: clinical outcomes in the international cooperative pulmonary embolism registry (ICOPER). Lancet. 1999;353(9162):1386–9. 16. Goldhaber SZ, Visani L, De Rosa M. Acute pulmonary embolism: clinical outcomes in the international cooperative pulmonary embolism registry (ICOPER). Lancet. 1999;353(9162):1386–9. Consent for publication W i i f d Written informed consent was obtained from the patient for publication of this case report. A copy of the written consent is available for review by the Editor of this journal. Ethics approval and consent to participate Not applicable. References 1. Cohen AT, Agnelli G, Anderson FA, Arcelus JI, Bergqvist D, Brecht JG, et al. Venous thromboembolism (VTE) in Europe. The number of VTE events and associated morbidity and mortality. Thromb Haemost. 2007;98(4):756–64. 1. Cohen AT, Agnelli G, Anderson FA, Arcelus JI, Bergqvist D, Brecht JG, et al. Venous thromboembolism (VTE) in Europe. The number of VTE events and associated morbidity and mortality. Thromb Haemost. 2007;98(4):756–64. 2. He J, Cheng X, Gao M. Survey of diagnosis and treatment of acute pulmonary embolism in 21 hospitals in China. Zhonghua Yi Xue Za Zhi. 2001;81(24):1490–2. 3. Wells PS, Ginsberg JS, Anderson DR, Kearon C, Gent M, Turpie AG, et al. Use of a clinical model for safe management of patients with suspected pulmonary embolism. Ann Intern Med. 1998;129(12):997–1005. 3. Wells PS, Ginsberg JS, Anderson DR, Kearon C, Gent M, Turpie AG, et al. Use of a clinical model for safe management of patients with suspected pulmonary embolism. Ann Intern Med. 1998;129(12):997–1005. 4. Prandoni P, Lensing AW, Prins MH, Ciammaichella M, Perlati M, Mumoli N, et al. Prevalence of pulmonary embolism among patients hospitalized for syncope. N Engl J Med. 2016;375(16):1524–31. 5. Wiegand UK, Kurowski V, Giannitsis E, Katus HA, Djonlagic H. Effectiveness of end-tidal carbon dioxide tension for monitoring thrombolytic therapy in acute pulmonary embolism. Crit Care Med. 2000;28(11):3588–92. 6. Goldhaber SZ, Elliott CG. Acute pulmonary embolism: part I: epidemiology, pathophysiology, and diagnosis. Circulation. 2003;108(22):2726–9. 7. Konstantinides SV, Torbicki A, Agnelli G, Danchin N, Fitzmaurice D, Galie N, et al. 2014 ESC guidelines on the diagnosis and management of acute pulmonary embolism. Eur Heart J. 2014;35(43):3033–69. 3069a 8. Donaldson AJ, Thomson HE, Harper NJ, Kenny NW. Bone cement implantation syndrome. Br J Anaesth. 2009;102(1):12–22. Submit your next manuscript to BioMed Central and we will help you at every step: Submit your next manuscript to BioMed Central and we will help you at every step: y 4. Prandoni P, Lensing AW, Prins MH, Ciammaichella M, Perlati M, Mumoli N, et al. Prevalence of pulmonary embolism among patients hospitalized for syncope. N Engl J Med. 2016;375(16):1524–31. • We accept pre-submission inquiries • Our selector tool helps you to ﬁnd the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit and we will help you at every step: 5. Wiegand UK, Kurowski V, Giannitsis E, Katus HA, Djonlagic H. Effectiveness of end-tidal carbon dioxide tension for monitoring thrombolytic therapy in acute pulmonary embolism. Crit Care Med. 2000;28(11):3588–92.
https://openalex.org/W4381309705	https://link.springer.com/content/pdf/10.1007/s00210-023-02561-y.pdf	English	null	The treatment with sGC stimulator improves survival of hypertensive rats in response to volume-overload induced by aorto-caval fistula	Naunyn-Schmiedeberg's archives of pharmacology	2,023	cc-by	14,851	Abstract Heart failure (HF) has been declared as global pandemic and current therapies are still ineffective, especially in patients that develop concurrent cardio-renal syndrome. Considerable attention has been focused on the nitric oxide (NO)/soluble guanylyl cyclase (sGC)/cyclic guanosine monophosphate (cGMP) pathway. In the current study, we aimed to investigate the effective- ness of sGC stimulator (BAY41-8543) with the same mode of action as vericiguat, for the treatment of heart failure (HF) with cardio-renal syndrome. As a model, we chose heterozygous Ren-2 transgenic rats (TGR), with high-output heart failure, induced by aorto-caval fistula (ACF). The rats were subjected into three experimental protocols to evaluate short-term effects of the treatment, impact on blood pressure, and finally the long-term survival lasting 210 days. As control groups, we used hypertensive sham TGR and normotensive sham HanSD rats. We have shown that the sGC stimulator effectively increased the survival of rats with HF in comparison to untreated animals. After 60 days of sGC stimulator treatment, the survival was still 50% compared to 8% in the untreated rats. One-week treatment with sGC stimulator increased the excretion of cGMP in ACF TGR (109 ± 28 nnmol/12 h), but the ACE inhibitor decreased it (-63 ± 21 nnmol/12 h). Moreover, sGC stimulator caused a decrease in SBP, but this effect was only temporary (day 0: 117 ± 3; day 2: 108 ± 1; day 14: 124 ± 2 mmHg). These results support the concept that sGC stimulators might represent a valuable class of drugs to battle heart failure especially with cardio-renal syndrome, but further studies are necessary. Keywords Heart failure · ACF · sGC stimulator · BAY41-8543 · cGMP · Vericiguat Olga Gawrys1 · Zuzana Husková1 · Petra Škaroupková1 · Zuzana Honetschlägerová1 · Zdeňka Vaňourková1 · Soňa Kikerlová1 · Vojtěch Melenovský2 · Barbara Szeiffová Bačová3 · Matúš Sykora3 · Miloš Táborský4 · Luděk Červenka1,4 Olga Gawrys1 · Zuzana Husková1 · Petra Škaroupková1 · Zuzana Honetschlägerová1 · Zdeňka Vaňourková1 · Soňa Kikerlová1 · Vojtěch Melenovský2 · Barbara Szeiffová Bačová3 · Matúš Sykora3 · Miloš Táborský4 · Luděk Červenka1,4 Received: 28 March 2023 / Accepted: 30 May 2023 / Published online: 20 June 2023 © The Author(s) 2023 RESEARCH RESEARCH RESEARCH Naunyn-Schmiedeberg's Archives of Pharmacology (2023) 396:3757–3773 https://doi.org/10.1007/s00210-023-02561-y Naunyn-Schmiedeberg's Archives of Pharmacology (2023) 396:3757–3773 https://doi.org/10.1007/s00210-023-02561-y * Olga Gawrys olga.gawrys@ikem.cz 1 Experimental Medicine Centre, Institute for Clinical and Experimental Medicine, Prague, Czech Republic 2 Department of Cardiology, Institute for Clinical and Experimental Medicine, Prague, Czech Republic 3 Centre of Experimental Medicine, Slovak Academy of Sciences, Institute for Heart Research, Bratislava, Slovakia 4 Department of Internal Medicine I, Cardiology, University Hospital Olomouc and Palacký University, Olomouc, Czech Republic Animals The studies were performed in accordance with guidelines and practices established by the Animal Care and Use Com- mittee of the Institute for Clinical and Experimental Med- icine (Prague) approved by the Ministry of Health of the Czech Republic (decision number MZDR 12482/2021–5/ OVZ), which accords with the European Union Directive 63/2010 and ARRIVE guidelines (Animal Research: Report- ing of In Vivo Experiments). All animals used in the present study were bred at the Center of Experimental Medicine of this Institute (IKEM), from stock animals supplied by the Max Delbrück Center for Molecular Medicine (Berlin, Germany), which is accredited by the Czech Association for Accreditation of Laboratory Animal Care. Heterozygous TGR [transgenic rats, strain name TGR(mRen2)27] harboring the mouse Ren-2 renin gene have been recently generated as a model for the study of primary hypertension. They were generated by breeding male homozygous TGR with female homozygous Hanno- ver-Sprague Dawley (HanSD) rats. Age-matched HanSD rats served as transgene-negative normotensive controls. The animals were kept on a 12-h/12-h light/dark cycle and had free access to tap water throughout the whole observa- tion. Male TGR rats at the initial age of 8 weeks were used for experiments. At this age TGR are already in the sus- tained phase of hypertension with systolic blood pressure (SBP) comparable with hypertensive patients (SBP around 180 mmHg) and with substantial activation of endogenous renin angiotensin system (RAAS), as demonstrated in pre- vious studies including ours. HanSD and TGR rats were randomly assigned to experimental groups to make sure that the animals from a single litter does not prevail in any group. So far several drug classes targeting NO-sGC-cGMP pathway have been developed, such as NO donors (e.g. nitro- glycerine) or PDE5 inhibitors (e.g., sildenafil) (Roberto et al. 2012) and more recently sGC stimulators and sGC activators were introduced (Cordwin et al. 2021). The sGC activators (e.g. cinaciguat) activate both the oxidized and heme-free sGC, while sGC stimulators increase cGMP production independently of NO by activating the reduced heme moi- ety (Cordwin et al. 2021). sGC stimulators, such as riociguat and vericiguat, exhibit also synergistic activity with NO, because they sensitize sGC to low levels of endogenous NO by stabilizing NO–sGC binding (Stasch et al. 2001; Liu et al. 2021). Introduction (Murphy et al. 2020). HF can be categorised into three main groups based on the ejection fraction (EF), i.e., heart failure with reduced EF (HFrEF), heart failure with mid-range or mildly reduced EF (HFmrEF), and heart failure with pre- served ejection fraction (HFpEF) (Simmonds et al. 2020). Over the recent decades a significant improvement has been made in the management and treatment of HF. Cur- rent guideline-directed medical therapy (GDMT) includes combination of β-blockers, angiotensin-converting enzyme inhibitors (ACEi), angiotensin receptor–neprilysin inhibitors (ARNI), angiotensin receptor blocker (ARB) with an addi- tion of a mineralocorticoid receptor antagonists, diuretics and more recently sodium-glucose cotransporter 2 (SGLT2) inhibitors (McDonagh et al. 2021; Heidenreich et al. 2022). The prognosis and life expectancy is especially dreadful for patients that develop concurrent impairment of renal hemo- dynamic and sodium excretory function, so called cardio- renal syndrome (Rangaswami et al. 2019; McCullough et al. Heart failure (HF) represents a major health and socioeco- nomic burden with considerable morbidity and mortality. It is estimated that it affects around 23 million people globally * Olga Gawrys olga.gawrys@ikem.cz 1 Experimental Medicine Centre, Institute for Clinical and Experimental Medicine, Prague, Czech Republic 2 Department of Cardiology, Institute for Clinical and Experimental Medicine, Prague, Czech Republic 3 Centre of Experimental Medicine, Slovak Academy of Sciences, Institute for Heart Research, Bratislava, Slovakia 4 Department of Internal Medicine I, Cardiology, University Hospital Olomouc and Palacký University, Olomouc, Czech Republic 1 Experimental Medicine Centre, Institute for Clinical and Experimental Medicine, Prague, Czech Republic (0121 3456789) 3 Naunyn-Schmiedeberg's Archives of Pharmacology (2023) 396:3757–3773 3758 is routinely used in our laboratory for many years (Abassi et al. 2011; Honetschlägerová et al. 2021; Kala et al. 2021, 2023), because it represents a model of HF accompanied with development of cardio-renal syndrome. In addition, we also investigated an optimized dose of an ACE inhibi- tor (trandolapril) as the standard therapy for HF as positive control. 2022). It is now commonly acknowledged that the reciprocal interaction between cardiac and renal function plays critical role in the progression of HF (Mullens et al. 2017; Cicca- relli et al. 2021; McCullough et al. 2022). Current GDMT are still ineffective in the treatment of HF, especially with cardio-renal syndrome, and the morbidity and mortality still remain very high, with a 5-year survival rate of 25% after first hospitalization (Berliner et al. 2020; Murphy et al. 2020). Animals Vericiguat was recently approved for the treatment of heart failure with reduced ejection fraction based on posi- tive outcome of a large clinical trial in patients with HFrEF (Armstrong et al. 2020). However, the exact mode of action of sGC stimulators and how these beneficial effects in HF are mediated are not fully understood yet and particularly it is unknown if it will be effective in cardio-renal syndrome. Introduction Hence new treatment strategies are still urgently needed to improve the morbidity and mortality. 2022). It is now commonly acknowledged that the reciprocal interaction between cardiac and renal function plays critical role in the progression of HF (Mullens et al. 2017; Cicca- relli et al. 2021; McCullough et al. 2022). Current GDMT are still ineffective in the treatment of HF, especially with cardio-renal syndrome, and the morbidity and mortality still remain very high, with a 5-year survival rate of 25% after first hospitalization (Berliner et al. 2020; Murphy et al. 2020). Hence new treatment strategies are still urgently needed to improve the morbidity and mortality. Methods Lately particular attention has been focused on the nitric oxide (NO)/soluble guanylyl cyclase (sGC)/cyclic guanosine monophosphate (cGMP) pathway. In the classical cascade NO binds to sGC, which leads to a conversion of guanosine triphosphate (GTP) into cGMP (Singh et al. 2018). In turn, this secondary messenger, cGMP, acts on various effector molecules including protein kinases (PK) or phosphodiester- ases (PDE), initiating a number of downstream effects, such as vasodilation and myocardial relaxation. It was proved to exhibit anti-inflammatory, anti-proliferatory, anti-fibrosis and renal protective properties, making this pathway one of the most essential signalling cascades within the cardiovas- cular and renal systems. Its disruption can lead to develop- ment of serious disorders or progression of already existing diseases, such as heart failure, pulmonary hypertension or chronic kidney disease (Farah et al. 2018; Sandner et al. 2021a; Xia et al. 2022). 3 Series 1: Dose selection and target engagement. Effect of the short‑term treatment with sGC stimulator and ACEi Two weeks before the start of the treatment rats underwent sham-operation or ACF creation procedure. After two weeks (week labelled 0) and after exclusion of acute death cases, the rats were randomly divided into the following experimental groups and selected treatment regimens were applied for one week. On days -1, 2, 4, and 7 rats were placed in metabolic cages for 12 h urine collection. Also, blood was collected on days -5 and 8. After a week of the treatment, all rats were decapitated and organs were weighted and collected for fur- ther biochemical evaluation.cGMP was measured in the urine and in the renal tissue. Noradrenaline and angiotensin II were measured in the plasma and the kidneys (collected after decap- itation). Albuminuria, natriuresis, and daily excretion rate of nitric oxide (NO) metabolites (NOx: nitrate and nitrite; indi- rect marker of NO production) were evaluated. Additionally, to Heart failure model and exclusion criteria f In the current work, we therefore, treated hypertensive, heterozygous Ren-2 transgenic rats (TGR), with high-out- put heart failure, induced by creating an aorto-caval fistula (ACF) with the sGC stimulator (BAY41-8543), which exhibits the same mode of action as riociguat and veri- ciguat (Sandner et al. 2021b). ACF TGR model is a well- established model of volume-overload heart failure and it Eight-weeks-old male TGR rats were anesthetized with an intraperitoneal injection of ketamine/midazolam mix- ture (Calypsol, Gedeon Richter, Hungary, 160 mg/kg and Dormicum, Roche, France, 160 mg/kg). Chronic HF due to volume overload was then induced by creating an aorto- caval fistula (ACF) using a needle technique. This procedure 1 3 3 Naunyn-Schmiedeberg's Archives of Pharmacology (2023) 396:3757–3773 3759 confirm that the selected dose was effective we have measured the concentration of BAY41-8543 in plasma. confirm that the selected dose was effective we have measured the concentration of BAY41-8543 in plasma. is routinely performed in our laboratory and detailed description was reported repeatedly in our previous stud- ies (Honetschlägerová et al. 2021; Kala et al. 2021, 2023). Sham-operated rats underwent an identical procedure, but without creating ACF. The animals in which the ACF proce- dure was not successful (based on visual verification of vena cava inferior in the end of each observation) were excluded from the experiment. Detailed experimental design Detailed experimental design of all 3 series is presented on Fig. 1. After 10-day period of recovery, rats underwent either sham operation or ACF creation as described above. A week after the appropriate treatment was initiated and rats were monitored for two weeks. Series 2: Effect of the treatment with sGC stimulator and ACEi on blood pressure (BP) Ten days before the ACF creation, telemetry probes were implanted into femoral artery of TGR rats (HanSD rats were omitted in this part of the study) under ketamine/midazolam anaesthesia (as above). HD-S10 radiotelemetric probes (Data Science International, St. Paul, Minnesota, USA) were used for direct BP measurements as described previously (Sporková et al. 2014; Husková et al. 2016). Fig. 1 The experimental design of the whole study depicting the time sequence and experimental manoeuvres; MC – metabolic cage with 12 h urine collection; B – blood collection from the tail vein Analytical procedures and chemicals The sGC stimulator BAY41-8543 (2-[1-[(2-fluorophenyl) methyl]-1H-pyrazolo[3,4-b]pyridin-3-yl]-5(4-morpholinyl) -4,6-pyrimidinediamine) was kindly provided by Bayer AG, Pharmaceuticals and is a typical member of the sGC stimu- lator drug class (Stasch et al. 2002a, b). Standard pellet diet containing BAY41-8543 in the dose of 3 mg kg−1 day−1 was prepared (Albert Weber, Prague, Czech Republic). Nutrients (content in 1 kg): NL 200 g; Fiber 48 g; Fat 30 g; Vitamin A 24.000 IU; Vitamin D3 2.000 IU; Copper (Cu) 30 mg; NaCl content in the final mix 0.4%. The dose of BAY41-8543 was selected based on the previous study (Stasch et al. 2002b; Sandner et al. 2021b) and current research. SDS‑PAGE and Western blotting 5. ACF TGR treated with placebo (ACF TGR) According to our previous studies (Szeiffová Bačova et al. 2016; Sykora et al. 2023), approximately 100 mg of frozen left ventricular heart tissue was homogenized in lysis buffer [50 mmol/L Tris–HCl, 250 sucrose, 1.0 mmol/L EGTA, 1.0 mmol/L dithiothreitol, 1.0 mmol/L phenylmethylsulfo- nyl fluoride and 0.5 sodium orthovanadate (pH 7.4)] and mixed with Laemmli sample buffer. Loading samples were separated in 10% SDS-PAGE (Mini-Protean TetraCell, Bio- Rad, Hercules, CA, USA) and transferred to a nitrocellu- lose membrane (0.2 µm pore size, Advantec, Tokyo, Japan). Membranes were subsequently incubated for 4 h with 5% low-fat milk, overnight with primary antibodies and for 1 h with a horseradish peroxidase-linked secondary antibody (Table 1). Between individual steps were membranes washed in TBS-T. Protein were visualized by enhanced chemilumi- nescence method and quantitated by densitometric analysis using Carestream Molecular Imaging Software (version 5.0, Carestream Health, New Haven, CT, USA). 6. ACF TGR treated with BAY41-8543 (ACF TGR + sGC- stim) 7. ACF TGR treated with trandolapril (ACF TGR + ACEi) 8. ACF TGR treated with trandolapril and BAY41-8543 (ACF TGR + ACEi + sGCstim) The numbers of rats used in each series is described in the appropriate result section. Series 3: Effects of the long‑term treatment with sGC stimulator and ACEi on the survival rate, morphometric and histopathological scores After confirmation that the selected dose of sGC stimulator is effective, the long-term protocol was performed. All rats under- went the same ACF creation or sham operation as described above in Series 1. After 2 weeks (week labelled 0), after exclu- sion of acute death cases, the rats were randomly divided into the following experimental groups and the follow-up period of 210 days was performed. Because of the severity of the ACF procedure and high mortality (especially in TGR rats) high ini- tial n values were used in these groups (n = 30). In sham oper- ated animals and HanSD rats the initial number of animals was 10 (calculated by statistical power analysis method). 1 3 Fig. 1 The experimental design of the whole study depicting the time sequence and experimental manoeuvres; MC – metabolic cage with 12 h urine collection; B – blood collection from the tail vein 1 3 Fig. 1 The experimental design of the whole study depicting the time sequence and experimental manoeuvres; MC – metabolic cage with 12 h urine collection; B – blood collection from the tail vein 1 3 Naunyn-Schmiedeberg's Archives of Pharmacology (2023) 396:3757–3773 3760 administered in drinking water. For the best effectiveness and safety, we implemented a titration protocol, which was previously developed and validated in our laboratory. Dur- ing the first week of administration, the animals received an increasing dose (changed every two days) starting from 0.5 mg/L up to 2 mg/L, which corresponds to a final dose of 0.25 mg kg−1 day−1. In our previous studies and here we demonstrated that this titration regimen and selected doses of trandolapril, provided maximal blockade of the renin angiotensin system (RAAS) and were well tolerated both by rats with ACF-induced heart failure and by sham-operated animals. After 210 days of the observation, the surviving animals were decapitated and organs were weighted and collected for morphometric and histopathological analysis (left and right ventricle of the heart, kidney). Experimental groups 1. Sham-operated HanSD rats treated with placebo (sham HanSD) 1. Sham-operated HanSD rats treated with placebo (sham HanSD) 2. Sham-operated HanSD rats treated with BAY41-8543 (sham HanSD + sGCstim) 2. Sham-operated HanSD rats treated with BAY41-8543 (sham HanSD + sGCstim) 3. Sham-operated TGR treated with placebo (sham TGR) 3. Sham-operated TGR treated with placebo (sham TGR) 3. Sham-operated TGR treated with placebo (sham TGR) 4. Sham-operated TGR treated with BAY41-8543 (sham 4. Sham-operated TGR treated with BAY41-8543 (sham TGR + sGCstim) 4. Sham-operated TGR treated with BAY41-8543 (sham TGR + sGCstim) Immunofluorescence detection of Cx43 and quantitative image analysis Immunodetection of Cx43 distribution was performed as described previously (Benova et al. 2013). Briefly, 10 µm thick left ventricular cryosections were washed in phosphate As an ACE inhibitor (ACEi) we used trandolapril (Gopten; Abbott, Prague, Czech Republic), which was Table 1 Antibodies used for Western blot analysis and for immunofluorescence methods Antibody Dilution Host Type Supplier/# Catalogue anti-Cx43 1:5000 Rabbit Polyclonal Sigma-Aldrich, St.Louis, MO, USA, #C6219 anti-phospho-ser368-Cx43 1:1000 Rabbit Polyclonal Santa Cruz Biotechnology, Dallas, TX, USA, #sc-101660 anti-PKC-epsilon 1:2000 Rabbit Polyclonal Santa Cruz Biotechnology, Dallas, TX, USA, #sc-214 anti-GAPDH 1:1000 Rabbit Polyclonal Santa Cruz Biotechnology, Dallas, TX, USA#sc-25778 Anti-Rabbit 1:2000 - - Cell Signaling Technology, Danvers, MA, USA, #7074S anti-Cx43 1:500 Mouse Monoclonal CHEMICON International,CA, USA, #MAB 3068 Anti-Mouse, FITC 1:500 Goat Polyclonal Jackson Immuno Research Labs, West Grove, Pennsylva- nia, USA, #115–095-062 able 1 Antibodies used for Western blot analysis and for immunofluorescence methods 1 3761 Naunyn-Schmiedeberg's Archives of Pharmacology (2023) 396:3757–3773 buffer saline (PBS), fixed in ice-cold methanol, permeabi- lized in 0.3% Triton X-100 in PBS, and blocked with the solution of 1% bovine serum albumin in PBS. Tissue sec- tions were incubated overnight with primary antibody fol- lowed by 2-h incubation with secondary antibody (Table 1). Between individual steps were membranes washed in PBS. Finally, tissues sections were mounted in the Fluoromount- G™ Mounting Medium (00–4958-02, Invitrogen™, Mas- sachusetts, USA) and analysed by Zeiss Apotome 2 micro- scope (Carl Zeiss, Jena, Germany). Approximately ten randomly acquired images from every tissue were captured and analyzed. Immunofluorescence signals were analysed and defined as a number of pixels with the protein signal intensity exceeding a threshold of 30 on the 0–255 Gy scale. The total number of Cx43 positive pixels was expressed as a total integral optical density per area (IOD) (Image-Pro Plus) (Sykora et al. 2019). copper sulfate solution and finally covered by gelatine and coverslips. To detect functional arteriolar and venular capil- lary network, dipeptidyl peptidase-4 (DPP4, E.C.3.4.15.4) was examined. Cryosections were incubated in solution (1.2 mM L-Leucine 4-methoxy-β-naptylamide hydrochlo- ride; 5% dimethylformamide; 2.4 mM Fast blue BB; 0.1 M Na2 HPO4 × 2H2O; 1 M KH2PO4), poured with gelatine and covered with a coverslip. Staining areas were observed and captured by light micro- scope (Zeiss Apotome 2 microscope Carl Zeiss, Jena, Ger- many). For quantitative analysis, randomly selected areas of positive signal from every tissue were analysed. Evaluation of hydroxyproline content Measurement of hydroxyproline is a useful method to deter- mine collagen content in the samples. Briefly, left ventricle tissue was dried, hydrolyzed in 6 M HCl and oxidated by chloramine T in the acetate-citrate buffer (pH 6.0). This reaction was stopped by pipetting Ehrlich’s reagent solution. Final concentration of hydroxyproline was subsequently measured spectrophotometrically at 550 nm and expressed in mg per total weight of the left ventricle (Pelouch et al. 1993; Reddy and Enwemeka 1996). Immunofluorescence detection of Cx43 and quantitative image analysis A positive signal was expressed as a proportion of the total tissue area (Andelova et al. 2022). Evaluation of glomerulosclerosis index (GSI) and tubulointerstitial injury (TSI) The kidneys were fixed in 4% formaldehyde, dehydrated and embedded in paraffin. The sections stained with hema- toxylin–eosin and PAS (periodic acid, for Schiff reaction) were examined and evaluated in a blind-test fashion. The calculation of glomerulosclerosis index (GSI) and kidney cortical tubulointerstitial injury (TSI) was described in detail in previous studies (Nakano et al. 2008; Gawrys et al. 2018; Honetschlagerová et al. 2021; Kala et al. 2023). The maxi- mum score for GSI is 4 and for TSI is 3. Described methods are commonly employed in our laboratory for many years and standardly used for evaluating the degree of kidney dam- age in almost all our studies (Kujal et al. 2014; Sedláková et al. 2017; Honetschlagerová et al. 2021). Series 1: Dose selection and target engagement. Effect of the short‑term treatment with sGC stimulator and ACEi (Figs. 2 and 3; Table 2) The first series of experiments was aimed to evaluate the short-term effects of treatment with sGC stimulator to support dose selection and demonstrate target engagement. As a marker of target engagement for the sGC stimulator treatment, cGMP excretion in urine was measured. As depicted in Fig. 2 the excretion of cGMP increased in all groups treated with sGC stimulator (BAY41-8543). The treatment with ACEi decreased the cGMP levels in urine which could not be compensated by sGC stimulator treatment. Interestingly, before the treatment the basal values of urinary cGMP were significantly higher in all groups with heart failure after ACF creation (148.5 ± 7.6 versus 49.6 ± 4.2 nmol/12 h in sham operated TGR, # P < 0.0001 unpaired T-Test; Fig. 2A). Fig. 2 Urinary excretion of cGMP A: throughout the whole observa- tion and B: the difference (Δ) between basal values (day -1) and the end (day 7) in heterozygous Ren-2 transgenic rats (TGR) rats with aorto-caval fistula (ACF) or without (sham) and normotensive sham HanSD rats treated with sGC stimulator (BAY41-8543), or with angiotensin-converting enzyme inhibitor (ACEi), alone or combined; P > 0.05 (NS); * P ≤ 0.05; P ≤ 0.01; * P ≤ 0.001; **** P ≤ 0.0001; by 2-way ANOVA with Dunnett's (within each group) and Tukey’s (between groups) multiple comparisons tests; # P ≤ 0.0001 basal values after ACF versus sham operation (day-1) by unpaired T-test rats. The treatment with ACEi increased the excretion of NOx, but after combined therapy with sGC stimulator it was diminished. Body weight, plasma (sodium, potassium, creatinine, noradrenaline), urine parameters (sodium, potassium, nitric oxide metabolites and 8-isoprostane excretion) and renal levels of noradrenaline and nitrotyrosine collected after one week treatment are presented in Table 2. Urinary excre- tion of nitric oxide metabolites (NOx: nitrate and nitrite) was measured as an indirect marker of NO production. In all the groups which received sGC stimulator, there was a significant or tendency to decrease the NOx excretion, especially pronounced in sham TGR and sham HanSD Plasma and renal levels of ANG II are presented on Fig. 3 (panel A and B, respectively). Renal ANG II were signifi- cantly higher in both sham TGR and ACF TGR in compari- son to control sham HanSD rats (81.6 ± 6.2 and 71.0 ± 7.6 versus 38.5 ± 2.4 fmol/g, respectively; # p < 0.05 one way ANOVA; Fig. 3). Histology and enzyme histochemistry of myocardial tissue 2 Urinary excretion of cGMP A: throughout the whole observa- tion and B: the difference (Δ) between basal values (day -1) and the end (day 7) in heterozygous Ren-2 transgenic rats (TGR) rats with aorto-caval fistula (ACF) or without (sham) and normotensive sham HanSD rats treated with sGC stimulator (BAY41-8543), or with angiotensin-converting enzyme inhibitor (ACEi), alone or combined; P > 0.05 (NS); * P ≤ 0.05; P ≤ 0.01; * P ≤ 0.001; **** P ≤ 0.0001; by 2-way ANOVA with Dunnett's (within each group) and Tukey’s (between groups) multiple comparisons tests; # P ≤ 0.0001 basal values after ACF versus sham operation (day-1) by unpaired T-test 2016; Červenka et al. 2015a; Kratky et al. 2018; Gawrys et al. 2020). Data and statistical analysis All values are expressed as means ± SEM. Graph-Pad Prism software (Graph Pad Software, San Diego, California, USA) was used for statistical analysis of the data. Comparison of survival curves was performed by log-rank (Mantel-Cox) test. Multiple-group comparisons were performed by mul- tiple t test, Wilcoxon´s signed-rank test, one-way or two- way analysis of variance followed by recommended post hoc test as appropriate. Values exceeding the 95% probability limits (P < 0.05) were considered statistically significant. The significance levels are indicated on figures with aster- isks: P > 0.05 (NS); P ≤ 0.05; P ≤ 0.01; P ≤ 0.001; *P ≤ 0.0001. The data and statistical analysis comply with the recommendations on experimental design and analysis in pharmacology (Curtis et al. 2018). Fig. 2 Urinary excretion of cGMP A: throughout the whole observa- tion and B: the difference (Δ) between basal values (day -1) and the end (day 7) in heterozygous Ren-2 transgenic rats (TGR) rats with aorto-caval fistula (ACF) or without (sham) and normotensive sham HanSD rats treated with sGC stimulator (BAY41-8543), or with angiotensin-converting enzyme inhibitor (ACEi), alone or combined; P > 0.05 (NS); P ≤ 0.05; P ≤ 0.01; * P ≤ 0.001; Fig. 2 Urinary excretion of cGMP A: throughout the whole observa- tion and B: the difference (Δ) between basal values (day -1) and the end (day 7) in heterozygous Ren-2 transgenic rats (TGR) rats with aorto-caval fistula (ACF) or without (sham) and normotensive sham HanSD rats treated with sGC stimulator (BAY41-8543), or with angiotensin-converting enzyme inhibitor (ACEi), alone or combined; P > 0.05 (NS); * P ≤ 0.05; P ≤ 0.01; * P ≤ 0.001; **** P ≤ 0.0001; by 2-way ANOVA with Dunnett's (within each group) and Tukey’s (between groups) multiple comparisons tests; # P ≤ 0.0001 basal values after ACF versus sham operation (day-1) by unpaired T-test Histology and enzyme histochemistry of myocardial tissue Plasma and tissue angiotensin II (ANG II) concentrations were measured by a competitive radioimmunoassay, using the commercially available RIA kit (ED29051, IBL Int., Hamburg, Germany). Plasma creatinine was measured by FUJI DRI-CHEM analyzer using appropriate slides for cre- atinine CRE-P III (FUJIFILM Corp., Tokyo, Japan). Urine creatinine was determined using Liquick Cor-CREATININE kit that is based on modified Jaffe’s method, without depro- teinization (PZ CORMAY S.A., Poland). Nitrate/nitrite lev- els were measured by a colorimetric assay (780,001, Cay- man Chemical, Ann Arbor, MI, USA). According to Lojda (Lojda and Gutmann 1976) with modi- fications (Andelova et al. 2022), 10 µm thick left ventricular myocardial tissue cryosections were used for histological and histochemical techniques. For myocardial structural changes characterization, a conventional histological haematoxylin–eosin staining was used. Briefly, dried cryosection were fixed in 4% buffered formaldehyde, stained with hematoxylin–eosin solutions, poured with Canada balsam and covered with a coverslip. To stain collagen deposition, Van Gieson technique was performed. Heart tissue sections were also fixed within 4% buffered formaldehyde, incubated in a mixture of saturated picric acid and 1% aqueous acid fuchsine, and subsequently covered by Canada balsam and coverslips. Commercially available ELISA kits were used to measure: renal nitrotyrosine (ab113848; Abcam, Cambridge, UK); renal cGMP (ADI-900–013; Enzo, Farmingdale, NY, USA) and plasma cGMP (581,021; Cayman Chem., Ann Arbor, MI, USA); plasma and renal noradrenalin (RE59261; IBL Int., Hamburg, Germany); urine 8-isoprostane (516,351; Cayman Chem., Ann Arbor, MI, USA). Sodium and potassium in plasma and urine were measured by BWB-XP flame photometer (BWB Technologies Ltd., Berkshire, UK). Detailed protocols of plasma and tissue preparation are described in our previous studies (Husková et al. 2006, 2007, For measurement of capillary endothelium-related alka- line phosphatase AP (E.C.3.1.3.1), cryosections were incu- bated in the mixture of solution (2,7 mM Naphthol AS-MX phosphate; 4,8 mM Fast blue BB; 5% dimethylformamide; 0,1 M Tris(hydroxymethyl)aminomethane). Subsequently, the slices were rinsed in distilled water and placed in a 2% 1 3 3762 Naunyn-Schmiedeberg's Archives of Pharmacology (2023) 396:3757–3773 Fig. Series 2: Effect of the treatment with sGC stimulator and ACEi on blood pressure (Figs. 4 and 5) Blood pressure (BP) was measured by telemetry starting four days before ACF creation. The creation of ACF per se caused a significant decrease in mean arterial blood pressure (MBP, Fig. 4), which is a common phenomenon observed in this model (-52 ± 2 mmHg in all ACF operated groups) (Červenka et al. 2015b). In sham operated animals, the tem- porary decrease in blood pressure was also observed, probably as a result of anaesthesia and fluid loss during the operation. Fig. 3 Angiotensin II (ANG II) A: in plasma and B: in renal tissue in heterozygous Ren-2 transgenic rats (TGR) rats with aorto-caval fistula (ACF) or without (sham) and normotensive sham HanSD rats after 7 day treatment with sGC stimulator (BAY41-8543), or with angiotensin-converting enzyme inhibitor (ACEi), alone or com- bined. P > 0.05 (NS); * P ≤ 0.05; P ≤ 0.01; * P ≤ 0.001; **** P ≤ 0.0001 by one-way ANOVA with Tukey’s multiple comparisons tests; # p < 0.05 versus sham HanSD by one way ANOVA with Tuk- ey’s multiple comparisons tests l g p The treatment with sGC stimulator in sham operated TGR animals caused a significant MBP decrease in comparison to untreated sham TGR (day 0: 140 ± 7 versus 130 ± 2 mmHg after 14 days of treatment). In addition, all applied treat- ment regimens decreased blood pressure in comparison to untreated ACF TGR (Fig. 4) including the treatment with sGC stimulator; however, the decrease in this groups was only transient (day 0: 94 ± 4 versus 86 ± 2 mmHg after 2 days of treatment; p = 0.03 paired T-test). The course of MBP changes was different between sGC stimulator group and ACEi treated groups, but there were no significant differ- ences between both ACEi treated groups (ACF TGR + ACEi versus ACF TGR + sGCstim + ACEi, NS). Fig. 3 Angiotensin II (ANG II) A: in plasma and B: in renal tissue in heterozygous Ren-2 transgenic rats (TGR) rats with aorto-caval fistula (ACF) or without (sham) and normotensive sham HanSD rats after 7 day treatment with sGC stimulator (BAY41-8543), or with angiotensin-converting enzyme inhibitor (ACEi), alone or com- bined. Series 1: Dose selection and target engagement. Effect of the short‑term treatment with sGC stimulator and ACEi (Figs. 2 and 3; Table 2) The treatment with sGC stimulator signifi- cantly increased ANG II in sham HanSD rats, both plasma (17.7 ± 2.6 versus 11.3 ± 0.6 fmol/ml in untreated sham HanSD; p < 0.05 one way ANOVA) and kidney (81.8 ± 6.6 1 3 3 Naunyn-Schmiedeberg's Archives of Pharmacology (2023) 396:3757–3773 3763 Fig. 3 Angiotensin II (ANG II) A: in plasma and B: in renal tissue in heterozygous Ren-2 transgenic rats (TGR) rats with aorto-caval fistula (ACF) or without (sham) and normotensive sham HanSD rats after 7 day treatment with sGC stimulator (BAY41-8543), or with angiotensin-converting enzyme inhibitor (ACEi), alone or com- bined. P > 0.05 (NS); * P ≤ 0.05; P ≤ 0.01; * P ≤ 0.001; **** P ≤ 0.0001 by one-way ANOVA with Tukey’s multiple comparisons tests; # p < 0.05 versus sham HanSD by one way ANOVA with Tuk- ey’s multiple comparisons tests stimulator, we monitored the food intake and we measured the concentration of BAY41-8543 in plasma in the end of the experiment. There were no differences between groups in the food intake and the plasma levels of sGC stimulator were appropriately elevated in all treated groups (Table 2). Series 2: Effect of the treatment with sGC stimulator and ACEi on blood pressure (Figs. 4 and 5) Treatment with ACEi inhibitor decreased the renal level of ANG II, alone and also when combined with sGC stimulator.if To confirm that the selected dose is effective and the rats eat the proper amount of the food mixed with the sGC 1 1 3 Naunyn-Schmiedeberg's Archives of Pharmacology (2023) 396:3757–3773 3764 Table 2 Body weight (BW), plasma and urine parameters collected from heterozygous Ren-2 transgenic rats (TGR) rats with aorto-caval fistula (ACF) or without (sham) and normotensive sham HanSD rats treated for one week with sGC stimulator (BAY41-8543), or with angiotensin-converting enzyme inhibitor (ACEi), alone or combined (n = 8–13 in each group); BW body weight; UNa V urinary sodium excretion; UK V urinary potassium excretion; UNOx urinary excretion of nitric oxide metabolites (nitrate/nitrite); Cr creatinine; NE noradrenaline; Superscript letter given next to the value (a–g) indicate statistically significant difference versus group indicated in the second row of the table (under the group’s name) by one way ANOVA with Tukey's multiple comparisons test * p < 0.05 sham TGR + sGCstim. vs untreated sham TGR by unpaired t test; # p < 0.05 ACF TGR + sGCstim. Series 2: Effect of the treatment with sGC stimulator and ACEi on blood pressure (Figs. 4 and 5) vs untreated sham TGR by unpaired t test; # p < 0.05 ACF TGR + sGCstim. vs untreated ACF TGR by unpaired t test BW body weight; UNa V urinary sodium excretion; UK V urinary potassium excretion; UNOx urinary excretion of nitric oxide metabolites (nitrate/nitrite); Cr creatinine; NE noradrenaline; Superscript letter given next to the value (a–g) indicate statistically significant difference versus group indicated in the second row of the table (under the group’s name) by one way ANOVA with Tukey's multiple comparisons test * p < 0.05 sham TGR + sGCstim. vs untreated sham TGR by unpaired t test; # p < 0.05 ACF TGR + sGCstim. vs untreated ACF TGR by unpaired t test each group). The sham-operated groups (HanSD and TGR) were omitted for clarity from Fig. 4. the 14th day the values of BP (both systolic and diastolic) between untreated ACF TGR and ACF TGR treated with sGC stimulator were no different. The untreated TGR rats with heart failure (ACF TGR) rapidly started to die in the first few weeks of the observa- tion. This was not attenuated in the initial phase by the sGC stimulator treatment, whereas it was almost stopped from the beginning by the optimized dose of the ACEi. However, in the further progression of the disease phenotype, the stand- alone treatment with sGC stimulator significantly improved survival in comparison to untreated rats (ACF TGR). After 60 days of sGC stimulator treatment the survival was still 50% compared to 8% in the untreated rats. After day 140, the effectiveness of the treatment with the sGC stimulator was then fading out also, however being still significantly better compared to placebo. The best effectiveness on mortality in this experiment was achieved with the optimized ACE inhibi- tor administered alone (ACF TGR + ACEi); the survival in this group was 90% at the end of the experiment. Surpris- ingly, the effectiveness of the combined therapy with ACEi and sGC stimulator (ACF TGR + sGCstim) was lower than stand-alone ACEi treatment with 44% at the end of the study. Series 2: Effect of the treatment with sGC stimulator and ACEi on blood pressure (Figs. 4 and 5) P > 0.05 (NS); * P ≤ 0.05; P ≤ 0.01; * P ≤ 0.001; **** P ≤ 0.0001 by one-way ANOVA with Tukey’s multiple comparisons tests; # p < 0.05 versus sham HanSD by one way ANOVA with Tuk- ey’s multiple comparisons tests In order to better illustrate and evaluate the magnitude of antihypertensive potency of each administered drug, only ACF groups are presented on Fig. 5 (without sham- operated animals) and the values before the treatment are omitted (before and after ACF creation). The compari- son of BP changes within each treated group (compared to baseline values obtained before the implementation of treatments) revealed that the most potent in lowering systolic blood pressure was the combined treatment with ACEi and sGC stimulator (SBP on day 0: 118 ± 5 versus 98 ± 3 mmHg on day 14). In addition, ACEi given alone was clearly antihypertensive, however the statistical analy- sis did not reveal such strong effect on SBP (SBP on day 0: 119 ± 4 versus 106 ± 2 mmHg on day 14). However, when DBP values were analyzed both ACEi treated groups (alone and combined with sGC stimulator) were similarly effective.sGC stimulator caused a decrease in BP (both SBP and DBP) especially pronounced in the beginning, but this effect was only transient and less significant than for other treatment regimens (SBP day 0: 117 ± 3; day 2: 108 ± 1; day 14: 124 ± 2 mmHg). The detailed post hoc analysis of changes in each day revealed that already on versus 38.5 ± 2.4 fmol/g in untreated sham HanSD; p < 0.05 one way ANOVA); and in ACF TGR rats in the kidney (115.9 ± 18.1 versus 71.0 ± 7.6 fmol/g in untreated TGR rats; p < 0.05 one way ANOVA). Also some increasing ten- dency was observed in plasma ANG II after sGC stimulator administration to sham TGR (21.8 ± 3.9 versus 16.4 ± 3.6 fmol/ml in untreated sham TGR; NS) and to ACF TGR (21.6 ± 2.3 versus 13.3 ± 1.7 fmol/ml in untreated ACF TGR; p < 0.05 by unpaired T-test). Series 2: Effect of the treatment with sGC stimulator and ACEi on blood pressure (Figs. 4 and 5) vs untreated ACF TGR by unpaired t test sham TGR sham TGR + sGCstim ACF TGR ACF TGR + sGCstim ACF TGR + ACEi ACF TGR + sGC- stim + ACEi sham HanSD sham HanSD + sGCstim a b c d e f g h BW (g) 449 ± 13 441 ± 6 408 ± 5 429 ± 10 415 ± 9 425 ± 16 455 ± 9 470 ± 10 UNa V (mmol/12 h) 1.7 ± 0.1 2.0 ± 0.2 0.7 ± 0.1 0.6 ± 0.1 1.0 ± 0.1 1.2 ± 0.1 c,d 1.0 ± 0.1 0.6 ± 0.1 UK V (mmol/12 h) 2.4 ± 0.2 2.0 ± 0.1 1.8 ± 0.1 1.7 ± 0.1 1.9 ± 0.1 2.1 ± 0.1 1.6 ± 0.1 1.6 ± 0.1 UNOx (μmol/12 h) 5.2 ± 0.1 4.2 ± 0.4 e,g,* 5.2 ± 0.5 4.3 ± 0.4 e,g, # 7.4 ± 0.6 h 4.6 ± 0.6 e,g 7.3 ± 0.4 3.4 ± 0.4 c,g U 8-isoprostane (ng/12 h) 8.5 ± 1.2 12.7 ± 1.1 15.8 ± 1.8 a 15.3 ± 1.3 12.5 ± 1.7 12.0 ± 1.8 7.8 ± 0.6 c,d 9.9 ± 1.2 Plasma Na (mmol/L) 137 ± 4 143 ± 5 139 ± 3 137 ± 4 136 ± 2 131 ± 2 134 ± 3 132 ± 2 Plasma K (mmol/L) 4.4 ± 0.2 4.3 ± 0.1 4.8 ± 0.1 4.5 ± 0.1 5.1 ± 0.2 4.6 ± 0.1 4.4 ± 0.1 4.0 ± 0.1 Plasma Cr (µmol/L) 15 ± 1 14 ± 1 18 ± 1 16 ± 1 17 ± 1 21 ± 2 18 ± 2 13 ± 1 Plasma NE (ng/ml) 2005 ± 420 3586 ± 856 2989 ± 192 2095 ± 478 3174 ± 1110 2899 ± 740 1560 ± 579 631 ± 93 Renal NE (ng/g prot.) 211 ± 21 161 ± 12 d 278 ± 20 264 ± 39 286 ± 59 338 ± 51 293 ± 22 239 ± 17 Renal Nitrotyrosine (ng/mg prot.) 26.5 ± 2.3 18.5 ± 1.1 19.7 ± 1.6 17.1 ± 2.4 18.5 ± 2.7 15.1 ± 2.0 a 20.7 ± 3.5 16.3 ± 2.0 Plasma BAY41- 8543 (µg/L) - 29.7 ± 4.5 - 30.8 ± 5.6 - 19.7 ± 2.0 - 26.7 ± 3.4 BW body weight; UNa V urinary sodium excretion; UK V urinary potassium excretion; UNOx urinary excretion of nitric oxide metabolites (nitrate/nitrite); Cr creatinine; NE noradrenaline; Superscript letter given next to the value (a–g) indicate statistically significant difference versus group indicated in the second row of the table (under the group’s name) by one way ANOVA with Tukey's multiple comparisons test * p < 0.05 sham TGR + sGCstim. Series 3: Effects of the long‑term treatment with sGC stimulator and ACEi on survival the rate, morphometric and histopathological scores (Figs. 6‑7, Table 3, SUPPLEMENTARY FIGURES S1‑S5) * P < 0.05 versus base- line values on day 0 within each group; # values different between all other groups; $ P < 0.05 values different between BAY41-8543 and both ACEi treated groups of ACF TGR by 2-way ANOVA with Dunnet’s (within each group) and Tukey’s (between groups) multiple comparisons tests. The BP values before the treatment and sham- operated groups are omitted for clarity Fig. 4 Time course of mean arterial blood pressure (MBP) through- out the whole observation in heterozygous Ren-2 transgenic rats (TGR) rats with aorto-caval fistula (ACF) or without (sham) treated with sGC stimulator (BAY41-8543), or with angiotensin-converting enzyme inhibitor (ACEi), alone or combined. * P < 0.05 by 2-way ANOVA with Bonferroni's multiple comparisons test; # P < 0.05 ver- sus all other groups 2-way ANOVA with Tukey’s multiple compari- sons tests Fig. 5 Time course of (A) systolic (SBP) and (B) diastolic (DBP) blood pressure throughout 2-week observation in heterozygous Ren-2 transgenic rats (TGR) rats with aorto-caval fistula (ACF) treated with sGC stimulator (BAY41-8543), or with angiotensin-converting enzyme inhibitor (ACEi), alone or combined. * P < 0.05 versus base- line values on day 0 within each group; # values different between all other groups; $ P < 0.05 values different between BAY41-8543 and both ACEi treated groups of ACF TGR by 2-way ANOVA with Dunnet’s (within each group) and Tukey’s (between groups) multiple comparisons tests. The BP values before the treatment and sham- operated groups are omitted for clarity Fig. 4 Time course of mean arterial blood pressure (MBP) through- out the whole observation in heterozygous Ren-2 transgenic rats (TGR) rats with aorto-caval fistula (ACF) or without (sham) treated with sGC stimulator (BAY41-8543), or with angiotensin-converting enzyme inhibitor (ACEi), alone or combined. * P < 0.05 by 2-way ANOVA with Bonferroni's multiple comparisons test; # P < 0.05 ver- sus all other groups 2-way ANOVA with Tukey’s multiple compari- sons tests Organ weights and renal histopathology scores (Table 3) At the end of the study (after 210 days), the surviving ani- mals were decapitated and organs were weighted and col- lected for further analysis. Noteworthy the placebo treated rats with heart failure and the sGC stimulator-treated rats were excluded from the statistical analysis due to the low animal numbers; therefore, this analysis was conducted mainly to compare the effects of the treatments in sham operated ani- mals (both TGR and HanSD) and between ACEi administered alone and combined treatment with ACEi and BAY41-8543. Out of all ACF rats the lowest heart weight (expressed as a ratio to tibia length) as indicator for heart hypertrophy was in the group treated with the combined treatment with sGC- stim and ACEi (58.0 ± 3.5 mg/mm) which was not signifi- cantly different from untreated sham TGR (45.6 ± 1.2 mg/ mm). Interestingly, the heart hypertrophy in ACF-TGR treated with the combined therapy tended to be lower com- pared to the ACEi inhibitor treatment alone (58.0 ± 3.5 mg/ mm versus 65.6 ± 2.4 mg/mm, respectively; P = 0.0777 by unpaired T-test; NS). Treatment with sGC stimulator also decreased heart weight in sham TGR (41.7 ± 1.2 mg/mm versus 45.6 ± 1.2 mg/mm in untreated animals; P < 0.05 by unpaired T-test; Table 3). At the end of the study (after 210 days), the surviving ani- mals were decapitated and organs were weighted and col- lected for further analysis. Noteworthy the placebo treated rats with heart failure and the sGC stimulator-treated rats were excluded from the statistical analysis due to the low animal numbers; therefore, this analysis was conducted mainly to compare the effects of the treatments in sham operated ani- mals (both TGR and HanSD) and between ACEi administered alone and combined treatment with ACEi and BAY41-8543. Fig. 5 Time course of (A) systolic (SBP) and (B) diastolic (DBP) blood pressure throughout 2-week observation in heterozygous Ren-2 transgenic rats (TGR) rats with aorto-caval fistula (ACF) treated with sGC stimulator (BAY41-8543), or with angiotensin-converting enzyme inhibitor (ACEi), alone or combined. * P < 0.05 versus base- line values on day 0 within each group; # values different between all other groups; $ P < 0.05 values different between BAY41-8543 and both ACEi treated groups of ACF TGR by 2-way ANOVA with Dunnet’s (within each group) and Tukey’s (between groups) multiple comparisons tests. Series 3: Effects of the long‑term treatment with sGC stimulator and ACEi on survival the rate, morphometric and histopathological scores (Figs. 6‑7, Table 3, SUPPLEMENTARY FIGURES S1‑S5) The model of ACF-induced HF in TGR is characterized by high mortality rates due to very early onset of decompensation of HF as compared with the course of ACF-induced HF in HanSD rats, as reported in previous studies including ours (Melenovsky et al. 2012; Červenka et al. 2015b; Kratky et al. 2021; Kala et al. 2023). The survival rates are presented on Fig. 6. All normoten- sive HanSD rats survived until the end of 210 days observa- tion (100% survival in placebo and in the sGC stimulator treated group); in sham-operated TGR rats (both placebo and sGCstim) the survival was 90% in the end (1 rat died in 1 3 3765 Naunyn-Schmiedeberg's Archives of Pharmacology (2023) 396:3757–3773 Organ weights and renal histopathology scores (Table 3) At the end of the study (after 210 days), the surviving ani- mals were decapitated and organs were weighted and col- lected for further analysis. Noteworthy the placebo treated rats with heart failure and the sGC stimulator-treated rats were excluded from the statistical analysis due to the low animal numbers; therefore, this analysis was conducted mainly to compare the effects of the treatments in sham operated ani- mals (both TGR and HanSD) and between ACEi administered alone and combined treatment with ACEi and BAY41-8543. Out of all ACF rats the lowest heart weight (expressed as a ratio to tibia length) as indicator for heart hypertrophy was in the group treated with the combined treatment with sGC- stim and ACEi (58 0±3 5 mg/mm) which was not signifi- Fig. 4 Time course of mean arterial blood pressure (MBP) through- out the whole observation in heterozygous Ren-2 transgenic rats (TGR) rats with aorto-caval fistula (ACF) or without (sham) treated with sGC stimulator (BAY41-8543), or with angiotensin-converting enzyme inhibitor (ACEi), alone or combined. * P < 0.05 by 2-way ANOVA with Bonferroni's multiple comparisons test; # P < 0.05 ver- sus all other groups 2-way ANOVA with Tukey’s multiple compari- sons tests Fig. 5 Time course of (A) systolic (SBP) and (B) diastolic (DBP) blood pressure throughout 2-week observation in heterozygous Ren-2 transgenic rats (TGR) rats with aorto-caval fistula (ACF) treated with sGC stimulator (BAY41-8543), or with angiotensin-converting enzyme inhibitor (ACEi), alone or combined. Organ weights and renal histopathology scores (Table 3) 6 The survival rate in heterozygous Ren-2 transgenic rats (TGR) with aorto-caval fistula (ACF) treated with sGC stimulator (BAY41- 8543), or with angiotensin-converting enzyme inhibitor (ACEi), alone or combined. All groups significantly differ between each other (by log-rank Mantel-Cox test); P > 0.05 (NS); * P ≤ 0.05; P ≤ 0.01; * P ≤ 0.001; **** P ≤ 0.0001; Control normotensive and sham operated TGR groups were omitted for clarity (the survival in these groups was 90–100% in the end) value of 0.5 in any group, which is considered as a bor- derline between healthy and renal damage), irrespective of the treatment. However, some subtle changes and sig- nificant differences were observed. ACEi administered alone and combined with sGC stimulator decreased GSI (0.10 ± 0.01 and 0.07 ± 0.01, respectively) to the greatest extend and to similar level as in untreated sham HanSD rats (0.11 ± 0.02). Also ACEi (alone and combined with sGC stimulator) decreased TSI when compared to untreated sham TGR and sham TGR treated with sGC stimulator (Table 3). The treatment with sGC stimulator tended to decrease GSI in sham TGR (0.23 ± 0.01 versus 0.43 ± 0.13 in untreated sham TGR; NS), but significantly increased it in HanSD rats (0.29 ± 0.06 versus 0.11 ± 0.02 in untreated sham HanSD; P = 0.017 by unpaired T-test). These changes were not reflected in tubulointerstitial injury score, since TSI was not changed in both sham TGR and sham HanSD (Table 3). major changes in lung, liver and kidney weights between applied treatment regimens. However, in normotensive HanSD rats the treatment with sGC stimulator significantly increased almost all of the measured parameters, when compared to untreated HanSD rats (Table 3, # by T-test). Since in this group the increase in body weight (BW) was also the highest we calculated the ratio of organ weights’ to BW which was however, not statistically significant, with the exception of liver weight (sham HanSD + sGC- stim: 33.0 ± 0.06 versus 30.3 ± 0.05 mg/g in untreated sham HanSD; P = 0.003 by unpaired t-Test). To evaluate the degree of the kidney damage the glo- merulosclerosis index (GSI) and kidney cortical tubu- lointerstitial injury (TSI) were measured (Table 3). The inflammatory cell infiltration, interstitial fibrosis and tubular dilatation were evaluated. Organ weights and renal histopathology scores (Table 3) The BP values before the treatment and sham- operated groups are omitted for clarity Out of all ACF rats the lowest heart weight (expressed as a ratio to tibia length) as indicator for heart hypertrophy was in the group treated with the combined treatment with sGC- stim and ACEi (58.0 ± 3.5 mg/mm) which was not signifi- cantly different from untreated sham TGR (45.6 ± 1.2 mg/ mm). Interestingly, the heart hypertrophy in ACF-TGR treated with the combined therapy tended to be lower com- pared to the ACEi inhibitor treatment alone (58.0 ± 3.5 mg/ mm versus 65.6 ± 2.4 mg/mm, respectively; P = 0.0777 by unpaired T-test; NS). Treatment with sGC stimulator also decreased heart weight in sham TGR (41.7 ± 1.2 mg/mm versus 45.6 ± 1.2 mg/mm in untreated animals; P < 0.05 by unpaired T-test; Table 3). no difference in lung weight between applied treatments in ACF TGR groups and all values were lower than in untreated ACF TGR rats at this stage of the disease (due to lack of the data for untreated animals, we can only compare the results to our previous studies). The lung weight to tibia length ratio two weeks after the induc- tion of ACF was 78.79 ± 1.14 mg/mm (Kala et al. 2021) and after four weeks was 74.89 ± 1.43 (Kala et al. 2023), which is significantly higher than in the current study (Table 3).f The lung weight in sham TGR rats treated with sGC stimulator was slightly higher than in untreated sham TGR (53.1 ± 2.6 vs 46.8 ± 1.3, respectively; Table 3), but the difference was not statistically significant. There was Except the obvious differences between sham-operated animals versus rats with ACF and differences between healthy HanSD rats and hypertensive TGR, there were no 1 3 3766 Naunyn-Schmiedeberg's Archives of Pharmacology (2023) 396:3757–3773 Fig. 6 The survival rate in heterozygous Ren-2 transgenic rats (TGR) with aorto-caval fistula (ACF) treated with sGC stimulator (BAY41- 8543), or with angiotensin-converting enzyme inhibitor (ACEi), alone or combined. All groups significantly differ between each other (by log-rank Mantel-Cox test); P > 0.05 (NS); * P ≤ 0.05; P ≤ 0.01; * P ≤ 0.001; **** P ≤ 0.0001; Control normotensive and sham operated TGR groups were omitted for clarity (the survival in these groups was 90–100% in the end) Fig. Protein levels of Cx43 and PKCε assessed by western blot method in left ventricle (Fig. 7) The protein levels of total Cx43 were significantly decreased in sham HanSD + sGCstim rats (76 ± 4 versus 98 ± 3% in untreated sham HanSD; p = 0.02 by one way ANOVA). On the contrary, protein abundance of total Cx43 and PKCε, which directly phosphorylate Cx43, was signifi- cantly increased in hypertensive sham TGR treated with sGC stimulator (total Cx43: 133 ± 5 versus 86 ± 7 and PKCε: 195 ± 10 versus 87 ± 6% in untreated sham TGR, respec- tively; P < 0.0001 by one way ANOVA). To evaluate the intercellular communication, the pro- tein levels of Cx43 and PKCε were assessed. Structural remodelling, such as hypertrophy, excessive accumulation of extracellular matrix (ECM) and myocardial fibrosis perturb Cx43 channels mediated intercellular commu- nication in addition to down-regulation of Cx43 due to disease aetiology. The protein levels of the Cx43 phos- phorylated at serine 368 characterized by reduced chan- nel conductivity, were significantly supressed in the left ventricle of hypertensive sham TGR and normotensive sham HanSD + sGCstim (63 ± 6 and 60 ± 3 respectively versus 100 ± 5% in untreated sham HanSD; P < 0.0001 by one way ANOVA). Organ weights and renal histopathology scores (Table 3) 7 Myocardial protein levels of total Cx43 (A), phosphorylated Cx43 at serine 368 (B), protein kinase C Epsilon (C), and representa- tive western blot images (D) measured in left ventricle collected from heterozygous Ren-2 transgenic rats (TGR) rats with aorto-caval fis- tula (ACF) or without (sham) and normotensive sham HanSD rats that survived until the end of 7 months treatment with sGC stimu- lator (BAY41-8543), or with angiotensin-converting enzyme inhibi- tor (ACEi), alone or combined. Values are normalized to GAPDH and presented as mean ± SEM; Cx43 – connexin 43; PKCε – pro- tein kinase C Epsilon; GAPDH- glyceraldehyde-3-phosphate dehy- drogenase; P > 0.05 (NS); * P ≤ 0.05; P ≤ 0.01; * P ≤ 0.001; **** P ≤ 0.0001; by one-way ANOVA and Tukey’s test multiple comparison test; # versus all other groups by one-way ANOVA and Tukey’s test multiple comparison test; (n = 5 in all groups except ACF TGR + BAY41-8543 where only 2 rats survived; note that no rats sur- vived in untreated ACF TGR group) and presented as mean ± SEM; Cx43 – connexin 43; PKCε – pro- tein kinase C Epsilon; GAPDH- glyceraldehyde-3-phosphate dehy- drogenase; P > 0.05 (NS); * P ≤ 0.05; P ≤ 0.01; * P ≤ 0.001; **** P ≤ 0.0001; by one-way ANOVA and Tukey’s test multiple comparison test; # versus all other groups by one-way ANOVA and Tukey’s test multiple comparison test; (n = 5 in all groups except ACF TGR + BAY41-8543 where only 2 rats survived; note that no rats sur- vived in untreated ACF TGR group) Organ weights and renal histopathology scores (Table 3) Noteworthy all val- ues for GSI and TSI were relatively low, indicating no severe renal damage in all groups (it did not exceed the 3 3 Naunyn-Schmiedeberg's Archives of Pharmacology (2023) 396:3757–3773 3767 Protein levels of Cx43 and PKCε assessed by western blot method in left ventricle (Fig. 7) The protein levels of total Cx43 were significant decreased in sham HanSD + sGCstim rats (76 ± 4 vers 98 ± 3% in untreated sham HanSD; p 0 02 by one w Fig. 7 Myocardial protein levels of total Cx43 (A), phosphorylated Cx43 at serine 368 (B), protein kinase C Epsilon (C), and representa- tive western blot images (D) measured in left ventricle collected from heterozygous Ren-2 transgenic rats (TGR) rats with aorto-caval fis- tula (ACF) or without (sham) and normotensive sham HanSD rats that survived until the end of 7 months treatment with sGC stimu- lator (BAY41-8543), or with angiotensin-converting enzyme inhibi- tor (ACEi), alone or combined. Values are normalized to GAPDH and presented as mean ± SEM; Cx43 – connexin 43; PKCε – p tein kinase C Epsilon; GAPDH- glyceraldehyde-3-phosphate deh drogenase; P > 0.05 (NS); * P ≤ 0.05; P ≤ 0.01; * P ≤ 0.0 **** P ≤ 0.0001; by one-way ANOVA and Tukey’s test multi comparison test; # versus all other groups by one-way ANOVA a Tukey’s test multiple comparison test; (n = 5 in all groups except A TGR + BAY41-8543 where only 2 rats survived; note that no rats s vived in untreated ACF TGR group) and presented as mean ± SEM; Cx43 – connexin 43; PKCε – pro- tein kinase C Epsilon; GAPDH- glyceraldehyde-3-phosphate dehy- drogenase; P > 0.05 (NS); * P ≤ 0.05; P ≤ 0.01; * P ≤ 0.001; **** P ≤ 0.0001; by one-way ANOVA and Tukey’s test multiple comparison test; # versus all other groups by one-way ANOVA and Tukey’s test multiple comparison test; (n = 5 in all groups except ACF TGR + BAY41-8543 where only 2 rats survived; note that no rats sur- vived in untreated ACF TGR group) Fig. Myocardial capillary density (Supplementary FIGURE S3 and S4) Activity of alkaline phosphatase (AP) reflects function and myocardial density of the arterial part of capillaries. AP activity was significantly normalized in sham TGR rats after sGC stimulator treatment (73 ± 6 versus 117 ± 11% in untreated sham TGR; P < 0.001 by one way ANOVA). No other differences were detected between other groups (Supplementary FIGURE S3). Unlike AP representing the arterial part of capillaries, dipeptidyl peptidase-4 (DPP4) demonstrate a function and density of the venous part of capillaries (Supplementary FIGURE S4). Enhanced DPP4 activity is associated with pathophysiology. Image analysis revealed that the DPP4 activity was significantly increased in heterozygous TGR rats with aorto-caval fistula (ACF) or without it (sham). Treatment did not reveal any significant changes. Myocardial topology and quantification of Cx43 (Supplementary FIGURE S1) Please note that in ACF TGR+ sGCstim group only 2 rats survived until the end, hence this group was excluded from the statistical analysis, but the values are presented sham TGR (n=8) sham TGR + sGCstim (n=9) ACF TGR + sGCstim (n=2) ACF TGR + ACEi (n=20) ACF TGR + sGCstim + ACEi (n=12) sham HanSD (n=9) sham HanSD + sGCstim (n=10) a b c d e f g BW (g) 765±13 f, g 770±19 f, g 695±43 723±13 f, g 746±25 549±10 592±13 # Heart W/tibia (mg/mm) 45.6±1.2 f, g 41.7±1.2 f, * 67.4±0.3 65.6±2.4 a, b, f, g 58.0±3.5b, f, g 33.7±0.7 37.1±0.5 # Lung W /tibia (mg/mm) 46.8±1.3 53.1±2.6 56.0±2.7 57.1±2.1a, f 59.1±2.9 a, f, g 44.1±1.2 48.6±0.8 # Liver W/tibia (g/cm) 5.6±0.2 f, g 5.5±0.2 f, g 4.9±0.6 4.7±0.2 b, f 4.8±0.2 f 3.8±0.1 4.5±0.1 # Right kidney W/tibia (mg/mm) 49.8±2.1 f, g 47.6±2.0 41.0±8.6 37.5±0.6 a, g 41.9±1.0 a, b 39.1±1.1 44.4±1.7 # Left kidney W/tibia (mg/mm) 53.1±2.6f, g 50.6±2.4 40.0±3.9 37.6±0.8a, b, g 42.8±1.7 a, b 39.5±1.0 45.3±2.0 # GSI 0.43±0.13f 0.23±0.01 0.09±0.04 0.10±0.01a, g 0.07±0.01a, g 0.11±0.02 0.29±0.06 # TSI 0.26±0.06 0.23±0.06 0.20±0.00 0.06±0.01 a, b 0.08±0.02 a, b 0.13±0.03 0.14±0.02 BW – body weight; W – weight; GSI – glomerulosclerosis index; TSI – tubulointerstitial injury; Superscript letter given next to the value (a-g) indicate statistically significant difference versus group indicated in the second row of the table (under the group’s name) by one way ANOVA with Tukey's multiple comparisons test; * sham TGR + sGCstim vs sham TGR; # sham HanSD + sGCstim vs sham HanSD by unpaired t test. Please note that in ACF TGR+ sGCstim group only 2 rats survived until the end, hence this group was excluded from the statistical analysis, but the values are presented rats. sGC stimulator treatment slightly normalized this ele- vated collagen deposition in the left ventricular tissue. hypertensive sham TGR the lateral distribution of Cx43 (side-to-side pattern) was more pronounced. A quantitative image analysis of Cx43 showed a significant decrease of Cx43 in hypertensive sham TGR and normotensive sham HanSD treated with sGC stimulator, compared to normo- tensive sham HanSD rats. Haematoxylin and eosin (HE) histological staining of left ventricle, hydroxyproline assay, and collagen deposition (Supplementary FIGURE S2 and S3) To evaluate the myocardial remodelling and the level of fibro- sis HE staining of left ventricle and hydroxyproline assay was performed. HE staining of left ventricle showed that, com- pared to sham HanSD rats, the left ventricular tissue of sham TGR and ACF TGR rats exhibited enlarged cardiomyocytes and focal areas infiltrated with polymorphonuclears, mainly in sham TGR (Supplementary FIGURE S2A). Hydroxypro- line content (Supplementary FIGURE S2B) in the left ven- tricular tissue was the highest in untreated sham TGR rats (33.5 ± 2.4 mg/g versus 26.9 ± 3.3 in untreated sham HanSD; p < 0.05 by one way ANOVA). It seems that all applied treat- ments diminished the hydroxyproline content, since the val- ues were not different than in healthy normotensive HanSD rats (Supplementary FIGURE S2B). Collagen deposition demonstrated by Van Gieson staining (Supplementary FIG- URE S3) showed increased collagen incidence in untreated sham TGR rats compared to healthy normotensive HanSD Myocardial topology and quantification of Cx43 (Supplementary FIGURE S1) In healthy normotensive rats (sham HanSD), Cx43 was detected mainly at the gap junction plaques of the inter- calated discs into end-to-end pattern. However, in the 1 3 1 3 3 Naunyn-Schmiedeberg's Archives of Pharmacology (2023) 396:3757–3773 3768 were treated with sGC stimulator (BAY41-8543), or with angioten- sin-converting enzyme inhibitor (ACEi), alone or combined for 7 months Table 3 Morphometric and histological parameters measured in material collected from rats that survived until the end of the observa- tion. Heterozygous Ren-2 transgenic rats (TGR) rats with aorto-caval fistula (ACF) or without (sham) and normotensive sham HanSD rats BW – body weight; W – weight; GSI – glomerulosclerosis index; TSI – tubulointerstitial injury; Superscript letter given next to the value (a-g) indicate statistically significant difference versus group indicated in the second row of the table (under the group’s name) by one way ANOVA with Tukey's multiple comparisons test; * sham TGR + sGCstim vs sham TGR; # sham HanSD + sGCstim vs sham HanSD by unpaired t test. Discussion As already previously mentioned, the ACF creation causes a significant decrease in blood pressure (around 50 mmHg), therefore our first concept was that the combined treatment causes adverse hypotension, where the values of BP decrease below autoregulatory lev- els. Low BP is a common symptom in subjects with HFrEF (around 10–15%) and it has been repeatedly demonstrated that blood pressure lower than 90 mmHg is a marker of poor outcome in acute HF (Cautela et al. 2020). Considering the high risk of negative outcomes in patients with HFrEF, current guideline-directed medical therapy recommends the titration of ACEi up to the maximally tolerated dosage (Ouwerkerk et al. 2017; Sharma et al. 2022).sGC stimulators (including BAY41-8543, vericiguat and riociguat) may cause symptomatic hypotension, but it is modest and dose-depend- ent (Stasch et al. 2002b; Sharkovska et al. 2010; Lam et al. 2021). Moreover, it was shown that even among the popula- tion of patients predisposed to hypotension, the efficiency of vericiguat persisted regardless of baseline SBP (Lam et al. 2021). However, we cannot exclude that in this particular model of HF (ACF TGR rats), the effects of the combined treatment (ACEi + sGCstim) depend on the used sGC stimu- lator dose and/or the sensitivity of this model per se (model specific effect). This is supported by recent findings in ACF TGR model in which the ACEi effects were also blunted by adding an ETA antagonist (Kala et al. 2023). In our current study, we employed the rat model of vol- ume overload induced by creation of the aorto-caval fistula, which mimics human HF and t is recommended for pre-clin- ical research (Riehle and Bauersachs 2019; Kala et al. 2023). The ACF was created in ren-2 transgenic rats, in which the hypertension and endogenous activation of the renin angi- otensin system (RAAS) are combined, consequently pro- gressing the development of HF with cardio-renal syndrome (Sobieraj et al. 2021). The first task in our study was to determine if the selected dose of BAY41-8543 (3 mg kg−1 day−1) is effective in our model of HF, specifically if it increases the production of secondary messenger cGMP. Therefore, we measured cGMP excretion in urine which was increased after sGC stimulator administration already after one week of treatment. Since the sGC stimulator was administered in the food, we measured the food intake and the concentration of BAY41-8543 in plasma in the end of the short-term experiment. Discussion The main objective of the present study was to evalu- ate the effectiveness of sGC stimulator for the treatment of HF in ACF TGR, which is a model of high-output 3 3769 Naunyn-Schmiedeberg's Archives of Pharmacology (2023) 396:3757–3773 HF associated with development of cardio-renal syn- drome. We have shown that BAY41-8543, which exhib- its the same mode of action as the sGC stimulator veri- ciguat, effectively increased the survival of rats with HF with cardio-renal syndrome in comparison to untreated animals. comparison to placebo group. The beneficial activity of vericiguat over placebo in the clinical trial appeared after 3 months of treatment and was sustained until the end of 10.8 months follow-up period (Armstrong et al. 2020). In our study the considerable improvement was already visible after three weeks of the treatment (the mortality in untreated rats was already 75% by the 23rd day versus only 34% in sGC stimulator-treated group). The beneficial activity of sGC stimulator persisted for another 2 months (by the 59th day of treatment 48% of the rats were still alive), however after that period the effectiveness of the treatment started slowly to fade out. Currently we cannot exclude that this is related to the ACF-procedure and volume overload since preclinical studies in other chronic diseases models have shown long- term prevention (Follmann et al. 2013; Stasch et al. 2015). The discovery of orally active sGC stimulators and sGC activators was a breakthrough in the pharmacology of NO/sGC/cGMP field and seems to have an extensive therapeutic potential. Hence, there is still a great need for pre-clinical and clinical research to fully understand the mode of action and to explore their beneficial activ- ity in cardiovascular and cardio-renal diseases in special patient populations (Cordwin et al. 2021; Sandner et al. 2021b). p ( ) Interestingly, we have observed that the optimized dose of the ACEi completely prevented the mortality of the rats, but the efficacy of the ACEi in the combination with the sGC stimulator was substantially reduced. This unanticipated out- come raises some difficulties in interpretation and we cannot provide a plausible explanation. One of the possible hypoth- eses includes the excessive antihypertensive potency of both drugs in this particular model of HF. Discussion Hence, the second goal of our study was to further investi- gate the NO/sGC/cGMP pathway in pathological as well as in physiological states and to further elucidate the possible mechanisms of sGC stimulators. We therefore, have meas- ured a variety of cardiovascular and renal biomarkers and found several interesting findings. i The basal urinary levels of cGMP (before treatment) were significantly higher in all groups in which HF was induced by ACF creation (Fig. 2A). This seems in contrast to data showing impaired cGMP signalling in heart failure (Blan- ton 2020). However, it seems that NO/sGC/cGMP pathway could be initially activated to counteract the development of HF after ACF creation. Furthermore, we found that the treatment with sGC stim- ulator increased or tended to increase the ANG II levels, which also might be in compliance with increased activa- tion of NO/sGC/cGMP signalling pathway mentioned above (both by NO-independent stimulation of sGC stimulator and by endogenous NO). The reciprocal interplay between RAAS and NO/sGC/cGMP is complex and still not fully understood (Persson 2003; Krishnan et al. 2018). cGMP, a secondary messenger in NO/sGC/cGMP cascade, was shown to exhibit both stimulatory and inhibitory action on renin secretion and the results of many studies are contradic- tory (Kurtz and Wagner 1998; Persson 2003; Curnow et al. 2020). Based on the results of our studies, i.e., increased activity of RAAS after sGC stimulator treatment, we could hypothesise rather stimulatory activity on renin secretion after short term treatment. The increase in cGMP resulting in beneficial effects, such as vasodilation or anti-inflamma- tory activity, could be counterbalanced by the succeeding activation of RAAS. Noteworthy, this phenomenon was only observed in ACF TGR and sham HanSD rats treated with sGC stimulator, but not in sham TGR, in which the treat- ment with BAY41-8543 did not affect ANG II levels to such extend (only tendency). i Taken together, the number of positive effects of the treat- ment with sGC stimulator in hypertensive TGR rats (without ACF), exerted on the kidney (lower GSI) and heart (dimin- ished hypertrophy, up-regulated Cx43 and PKCε, decreased fibrosis) suggest a great potential of this class of the drug in prevention and/or treatment of persistent hypertension and related disorders. Summary and conclusions The main goal of the current study was to evaluate the effec- tiveness of sGC stimulator (BAY 41–8543) for the treatment of HF due to volume overload combined with cardio-renal syndrome (ACF TGR). We have shown that the sGC stimu- lator effectively increased the survival of ACF TGR in com- parison to untreated animals. Taken together, we believe that sGC stimulators could represent a valuable tool to treat heart failure and renal dysfunction, but more studies are necessary to elucidate the exact mechanisms of action and interactions with other classes of drugs. In fact, we have found a number of beneficial effects of sGC stimulator treatment in hypertensive sham TGR group (without ACF creation) exerted both on the renal and car- diovascular systems. The glomerulosclerosis index (GSI) tended to be lower after the sGC stimulator treatment. Fur- thermore, the analysis of the heart collected after the sur- vival experiment (after 210 days) revealed a great deal of positive activity of sGC stimulator in sham TGR in com- parison to untreated hypertensive animals. Discussion We did not record any differences in the food intake between groups and the plasma levels of sGC stimulator were in accordance to previous studies (Stasch et al. 2015). These data imply that we were able to reach sufficient exposure and target engagement with our sGC stimulator dose. In the next step we verified that our sGC stimulator does not significantly impact blood pressure. Indeed, the treatment with only sGC stimulator proved to have only transient and minor impact on blood pressure in ACF TGR rats.i After having identified the optimal dose, that lead to target engagement but not to significant BP reduction, we conducted a long-term study to evaluate potential benefi- cial effects of the sGC stimulator BAY 41–8543 in our rat HF model. We have shown that sGC stimulator significantly improved the survival of ACF TGR rats in comparison to untreated animals. This is in line with recent clinical data from the VICTORIA phase 3 pivotal clinical trial which demonstrated that the vericiguat significantly reduced the incidence of the primary outcome of death from cardio- vascular causes or first hospitalization for heart failure in The clinical importance of NO/sGC/cGMP signalling in cardiovascular and cardio-renal diseases including HF is very obvious; however, a better understanding of the under- lying mechanisms is necessary for the optimal HF treatment and prevention and selection of patients profiting best from 1 3 1 3 Naunyn-Schmiedeberg's Archives of Pharmacology (2023) 396:3757–3773 3770 contribute to the arrhythmic substrate in failing human heart promoting occurrence of life-threatening arrhythmias, hence the increase in Cx43 appears to be a mechanism to avoid lethal arrhythmia (Danik et al. 2004). We observed a number of positive effects resulted from the long-term treatment with sGC stimulator of hypertensive sham TGR, i.e., decreased heart hypertrophy, significantly up-regulated Cx43 and PKCε in the left ventricle. Moreover, the long term treat- ment with sGC stimulator tended to decrease hydroxyproline (marker of fibrosis) (Díez 2007) and significantly decreased the activity of alkaline phosphatase (marker of myocardial capillary density) (Schultz-Hector et al. 1993). Certainly, observed beneficial effects of BAY41-8543 treatment can be partially escribed to antihypertensive effectiveness of the sGC stimulator, however the magnitude of the BP decrease, which was only modest suggests also blood-pressure inde- pendent beneficial mechanisms.f sGC stimulator therapies (Numata and Takimoto 2022). 3 References Abassi Z, Goltsman I, Karram T, et al (2011) Aortocaval fistula in rat: A unique model of volume-overload congestive heart fail- ure and cardiac hypertrophy. J Biomed Biotechnol 2011:729497 https://doi.org/10.1155/2011/729497f Moreover, we are aware that the analysis of the natriuretic peptide axis is missing. Due to technical reasons, it was not possible in the current study, but it will be included in the next series of experiments in the future. Andelova K, Szeiffova Bacova B, Sykora M et al (2022) Cardiac Cx43 signaling is enhanced and TGF-β1/SMAD2/3 suppressed in response to cold acclimation and modulated by thyroid status in hairless SHRM. Biomedicines 10(7):1707. https://doi.org/10. 3390/biomedicines10071707 Supplementary information The online version contains supplemen- tary material available at https://doi.org/10.1007/s00210-023-02561-y. Armstrong PW, Pieske B, Anstrom KJ et al (2020) Vericiguat in Patients with Heart Failure and Reduced Ejection Fraction. N Engl J Med 382:1883–1893. https://doi.org/10.1056/NEJMo a1915928 Acknowledgements We would like to thank Peter Sandner, PhD from Bayer AG (Wuppertal, Germany) for providing the sGC stimulator and invaluable help with writing and reviewing the manuscript. Benova T, Viczenczova C, Radosinska J et al (2013) Melatonin attenu- ates hypertension-related proarrhythmic myocardial maladapta- tion of connexin-43 and propensity of the heart to lethalarrhyth- mias. Can J Physiol Pharmacol 91:633–639. https://doi.org/10. 1139/cjpp-2012-0393 Authors contributions The authors declare that all data were generated in-house and that no paper mill was used. LČ, VM, OG conceived and designed study; OG, ZHu, PŠ, ZHo, ZV, SK, BS and MS performed research; OG, ZHu, PŠ, ZHo, VM, BS, MS and LČ analysed data; OG and LČ wrote the paper. All authors read and approved the manuscript. Berliner D, Hänselmann A, Bauersachs J (2020) The treatment of heart failure with reduced ejection fraction. Dtsch Arztebl Int 117:376– 386. https://doi.org/10.3238/arztebl.2020.0376 Blanton RM (2020) cGMP Signaling and Modulation in Heart Failure. J Cardiovasc Pharmacol 75:385–398. https://doi.org/10.1097/FJC. 0000000000000749 Funding Open access publishing supported by the National Technical Library in Prague. This study was supported by the project National Institute for Research of Metabolic and Cardiovascular Diseases (Pro- gram EXCELES, Project No. LX22NPO5104)—funded by the Euro- pean Union—Next Generation EU. This study was also supported by the Ministry of Health of the Czech Republic (grant number 20–02- 00052) and by the Slovak Research and Development Agency (under the contract no. 21–0410). Cautela J, Tartiere JM, Cohen-Solal A et al (2020) Management of low blood pressure in ambulatory heart failure with reduced ejection fraction patients. Competing interests The authors declare no competing interests. Competing interests The authors declare no competing interests. Open Access This article is licensed under a Creative Commons Attri- bution 4.0 International License, which permits use, sharing, adapta- tion, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The second drawback regards the lack of a proper analysis of cardiac function, employing either echocardiography and/ or pressure volume analysis. However, that would require repeated usage of anaesthetics, which could impact the results of survival protocol, which was the main goal of this part of the study. Hence, we decided to conduct separate experiments aimed to explore this issue in more detail (in all experimental groups at crucial therapeutic points). This part is still ongoing, due to time-consuming and complex char- acter of the study; therefore, these results will be presented in the future separate studies. References Eur J Heart Fail 22:1357–1365. https://doi.org/ 10.1002/ejhf.1835 Červenka L, Bíbová J, Husková Z et al (2015a) Combined suppression of the intrarenal and circulating vasoconstrictor Renin-ACE-ANG II axis and augmentation of the vasodilator ACE2-ANG 1-7-Mas axis attenuates the systemic hypertension in Ren-2 transgenic rats exposed to chronic hypoxia. Physiol Res 64:11–24 Č Data availability The datasets generated during and/or analyzed dur- ing the current study are available from the corresponding author on reasonable request. Červenka L, Melenovský V, Husková Z et al (2015b) Inhibition of soluble epoxide hydrolase counteracts the development of renal dysfunction and progression of congestive heart failure in Ren-2 transgenic hypertensive rats with aorto-caval fistula. Clin Exp Pharmacol Physiol 42:795–807. https://doi.org/10.1111/1440- 1681.12419 Limitations of the study The first limitation of the study is the low numbers of ani- mals, which survived until the end of the survival protocol and the lack of control (untreated) animals for histopatho- logical analysis. The main objective of this part of the study was to evaluate the long-term effectiveness of sGC stimula- tor and ACEi on the survival of rats with heart failure and cardio-renal syndrome. The initial n numbers of animals used in this study were relatively high (n = 30) due to our calculation by statistical power analysis method. During the study, we excluded acute deaths after surgical preparations (ACF creation). Also some of the rats were excluded from Hypertension is linked with an increased occurrence of severe arrhythmias and development and progression of heart failure (Egan Benova et al. 2016). The key factors facil- itating such life-threatening events are myocardial structural remodelling, fibrosis and altered topology and disorders of connexin-43 (Cx43) channels. It has been established that down-regulation of Cx43 as well as its abnormal topology 3 3 Naunyn-Schmiedeberg's Archives of Pharmacology (2023) 396:3757–3773 3771 accords with the European Union Directive 63/2010 and ARRIVE guidelines (Animal Research: Reporting of In Vivo Experiments). accords with the European Union Directive 63/2010 and ARRIVE guidelines (Animal Research: Reporting of In Vivo Experiments). the whole analysis due to ACF atrophy, which could be veri- fied only in the end of the observation. Therefore, it was not possible to predict exactly how many samples will be avail- able in the end. Nevertheless, this analysis was conducted mainly to compare the effects of the treatments in sham operated animals (both TGR and HanSD) and between ACEi administered alone and combined treatment with ACEi and BAY41-8543. In the future, we will perform next series of experiments in which the end point will be planned more carefully to avoid this setback. https://doi.org/10.1111/bph.14153 Danik SB, Liu F, Zhang J et al (2004) Modulation of cardiac gap junction expression and arrhythmic susceptibility. Circ Res 95:1035–1041. https://doi.org/10.1161/01.RES.0000148664. 33695.2a Kala P, Miklovič M, Jíchová Š, et al (2021) Effects of epoxyeicosa- trienoic acid-enhancing therapy on the course of congestive heart failure in angiotensin ii-dependent rat hypertension: From mrna analysis towards functional in vivo evaluation. Biomedicines 9(8):1053 https://doi.org/10.3390/biomedicines9081053 Díez J (2007) Mechanisms of cardiac fibrosis in hypertension. J Clin Hypertens (greenwich) 9:546–550. https://doi.org/10.1111/j. 1524-6175.2007.06626.xf Kratky V, Kopkan L, Kikerlova S et al (2018) The role of renal vascular reactivity in the development of renal dysfunction in compensated and decompensated congestive heart failure. Kidney Blood Press Res 43:1730–1741. https://doi.org/10.1159/000495391 Egan Benova T, Szeiffova Bacova B, Viczenczova C, et al (2016) MyocardiaL connexin-43 is implicated in the prevention of malignant arrhythmia in rats suffering from essential hyper- tension. In: Update on Essential Hypertension. IntechOpen, London Kratky V, Vanourkova Z, Sykora M et al (2021) AT1 receptor blocker, but not an ACE inhibitor, prevents kidneys from hypop- erfusion during congestive heart failure in normotensive and hypertensive rats. Sci Rep 11:4271. https://doi.org/10.1038/ s41598-021-83906-6 Farah C, Michel LYM, Balligand JL (2018) Nitric oxide signalling in cardiovascular health and disease. Nat Rev Cardiol 15:292–316. https://doi.org/10.1038/nrcardio.2017.224 Krishnan SM, Kraehling JR, Eitner F et al (2018) The impact of the nitric oxide (no)/soluble guanylyl cyclase (sGC) signaling cascade on kidney health and disease: A preclinical perspective. Int J Mol Sci 19(6):1712. https://doi.org/10.3390/ijms19061712 Č Š Follmann M, Griebenow N, Hahn MG et al (2013) The chemistry and biology of soluble guanylate cyclase stimulators and activators. Angew Chemie - Int Ed 52:9442–9462. https://doi.org/10.1002/ anie.201302588 Gawrys O, Baranowska I, Gawarecka K et al (2018) Innovative lipid-based carriers containing cationic derivatives of polyiso- prenoid alcohols augment the antihypertensive effectiveness of candesartan in spontaneously hypertensive rats. Hypertens Res 41:234–245. https://doi.org/10.1038/s41440-018-0011-y Kujal P, Čertíková Chábová V, Škaroupková P et al (2014) Inhibition of soluble epoxide hydrolase is renoprotective in 5/6 nephrectomized Ren-2 transgenic hypertensive rats. Clin Exp Pharmacol Physiol 41:227–237. https://doi.org/10.1111/1440-1681.12204 41:227–237. https://doi.org/10.1111/1440-1681.12204 Kurtz A, Wagner C (1998) Role of nitric oxide in the control of renin secretion. Am J Physiol - Ren Physiol 275:F849–F862. https://doi. org/10.1152/ajprenal.1998.275.6.f849 Gawrys O, Husková Z, Baranowska I et al (2020) Combined treat- ment with epoxyeicosatrienoic acid analog and 20-hydroxyei- cosatetraenoic acid antagonist provides substantial hypotensive effect in spontaneously hypertensive rats. J Hypertens 38:1802– 1810. Declarations Ethics approval This study was performed in accordance with guide- lines and practices established by the Animal Care and Use Commit- tee of the Institute for Clinical and Experimental Medicine (Prague) approved by the Ministry of Health of the Czech Republic, which Ciccarelli M, Dawson D, Falcao-Pires I et al (2021) Reciprocal organ interactions during heart failure: A position paper from the 1 3 3772 Naunyn-Schmiedeberg's Archives of Pharmacology (2023) 396:3757–3773 ESC Working Group on Myocardial Function. Cardiovasc Res 117:2416–2433. https://doi.org/10.1093/cvr/cvab009 Husková Z, Kramer H, Vaňourková Z et al (2007) Effects of dietary salt load and salt depletion on the course of hypertension and angio- tensin II levels in male and female heterozygous Ren-2 transgenic rats. Kidney Blood Press Res 30:45–55. https://doi.org/10.1159/ 000099028 p g Cordwin DJ, Berei TJ, Pogue KT (2021) The Role of sGC Stimulators and Activators in Heart Failure With Reduced Ejection Fraction. J Cardiovasc Pharmacol Ther 26:593–600. https://doi.org/10.1177/ 10742484211042706 Husková Z, Kopkan L, Červenková L et al (2016) Intrarenal altera- tions of the angiotensin-converting enzyme type 2/angiotensin 1–7 complex of the renin-angiotensin system do not alter the course of malignant hypertension in Cyp1a1-Ren-2 transgenic rats. Clin Exp Pharmacol Physiol 43:438–449. https://doi.org/10.1111/ 1440-1681.12553 Curnow AC, Gonsalez SR, Gogulamudi VR, et al (2020) Low nitric oxide bioavailability increases renin production in the collect- ing duct. Front Physiol 11:559341 https://doi.org/10.3389/fphys. 2020.559341 Curtis MJ, Alexander S, Cirino G et al (2018) Experimental design and analysis and their reporting II: updated and simplified guid- ance for authors and peer reviewers. Br J Pharmacol 175:987–993. https://doi.org/10.1111/bph.14153 Kala P, Gawrys O, Miklovič M et al (2023) Endothelin type A recep- tor blockade attenuates aorto-caval fistula-induced heart failure in rats with angiotensin II-dependent hypertension. J Hypertens 41:99–114. https://doi.org/10.1097/HJH.0000000000003307f p y 1111/1440-1681.12310 g Persson PB (2003) Renin: Origin, secretion and synthesis. J Physiol 552:667–671. https://doi.org/10.1113/jphysiol.2003.049890 Stasch JP, Becker EM, Alonso-Alija C et al (2001) NO-independent regulatory site on soluble guanylate cyclase. Nature 410:212–215. https://doi.org/10.1038/35065611 Rangaswami J, Bhalla V, Blair JEA et al (2019) Cardiorenal Syndrome: Classification, Pathophysiology, Diagnosis, and Treatment Strate- gies: A Scientific Statement From the American Heart Associa- tion. Circulation 139:E840–E878. https://doi.org/10.1161/CIR. 0000000000000664i Stasch JP, Alonso-Alija C, Apeler H et al (2002a) Pharmacological actions of a novel NO-independent guanylyl cyclase stimulator, BAY 41–8543: In vitro studies. Br J Pharmacol 135:333–343. https://doi.org/10.1038/sj.bjp.0704484 Reddy GK, Enwemeka CS (1996) A simplified method for the analysis of hydroxyproline in biological tissues. Clin Biochem 29:225– 229. https://doi.org/10.1016/0009-9120(96)00003-6 Stasch JP, Dembowsky K, Perzborn E et al (2002b) Cardiovascular actions of a novel NO-independent guanylyl cyclase stimulator, BAY 41–8543: In vivo studies. Br J Pharmacol 135:344–355. https://doi.org/10.1038/sj.bjp.0704483f p g Riehle C, Bauersachs J (2019) Small animal models of heart failure. Cardiovasc Res 115:1838–1849. https://doi.org/10.1093/cvr/ cvz161 Stasch JP, Schlossmann J, Hocher B (2015) Renal effects of soluble guanylate cyclase stimulators and activators: A review of the pre- clinical evidence. Curr Opin Pharmacol 21:95–104. https://doi. org/10.1016/j.coph.2014.12.014 Roberto B, Evora P, M. Evora P, C. Celotto A, et al (2012) Cardio- vascular Therapeutics Targets on the NO–sGC–cGMP Signaling Pathway: A Critical Overview. Curr Drug Targets 13:1207–1214. https://doi.org/10.2174/138945012802002348 Sykora M, Kamocsaiova L, Egan Benova T et al (2019) Alterations in myocardial connexin-43 and matrix metalloproteinase-2 signaling in response to pregnancy and oxygen deprivation of wistar rats: A pilot study. Can J Physiol Pharmacol 97:829–836. https://doi.org/ 10.1139/cjpp-2018-0740 Sandner P, Follmann M, Becker-Pelster E et al (2021a) Soluble GC stimulators and activators: Past, present and future. Br J Pharma- col Oct 2:1–22. https://doi.org/10.1111/bph.15698 Sandner P, Zimmer DP, Milne GT et al (2021b) Soluble guanylate cyclase stimulators and activators. In: Schmidt HHHW, Ghezzi P, Cuadrado A (eds) Reactive Oxygen Species. Handbook of Experi- mental Pharmacology, vol. 264. Springer, Cham, pp 355–394 Sykora M, Kratky V, Kopkan L, Tribulova N (2023) Anti-Fibrotic Potential of Angiotensin (1–7) in Hemodynamically Overloaded Rat Heart. Int J Mol Sci 24:3490. https://doi.org/10.3390/ijms2 4043490f Schultz-Hector S, Balz K, Bohm M et al (1993) Cellular localization of endothelial alkaline phosphatase reaction product and enzyme protein in the myocardium. J Histochem Cytochem 41:1813–1821. https://doi.org/10.1111/bph.14153 https://doi.org/10.1097/HJH.0000000000002462 Lam CSP, Mulder H, Lopatin Y et al (2021) Blood pressure and safety events with vericiguat in the VICTORIA trial. J Am Heart Assoc 10(22):e021094 https://doi.org/10.1161/JAHA.121.021094 Liu R, Kang Y, Chen L (2021) Activation mechanism of human solu- ble guanylate cyclase by stimulators and activators. Nat Commun 12:5492. https://doi.org/10.1038/s41467-021-25617-0 Heidenreich PA, Bozkurt B, Aguilar D et al (2022) 2022 AHA/ ACC/HFSA Guideline for the Management of Heart Failure: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation 145:e895–e1032. https://doi.org/10.1161/CIR. 0000000000001063f Lojda Z, Gutmann E (1976) Histochemistry of some acid hydrolases in striated muscles of the rat. Histochemistry 49:337–342. https:// doi.org/10.1007/BF00496137 Honetschlagerová Z, Škaroupková P, Kikerlová S et al (2021) Effects of renal sympathetic denervation on the course of congestive heart failure combined with chronic kidney disease: Insight from studies with fawn-hooded hypertensive rats with volume overload induced using aorto-caval fistula. Clin Exp Hypertens 43:522–535. https:// doi.org/10.1080/10641963.2021.1907398 McCullough PA, Amin A, Pantalone KM, Ronco C (2022) Cardiorenal Nexus: A Review With Focus on Combined Chronic Heart and Kidney Failure, and Insights From Recent Clinical Trials. J Am Heart Assoc 11:1–9. https://doi.org/10.1161/JAHA.121.024139 t Assoc 11:1–9. https://doi.org/10.1161/JAHA.121.024139 McDonagh TA, Metra M, Adamo M et al (2021) 2021 ESC Guide- lines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J 42:3599–3726. https://doi.org/10.1093/eurhe artj/ehab368 g Honetschlägerová Z, Hejnová L, Novotný J et al (2021) Effects of renal denervation on the enhanced renal vascular responsiveness to angiotensin II in high-output heart failure: Angiotensin II recep- tor binding assessment and functional studies in ren-2 transgenic hypertensive rats. Biomedicines 9(12):1803. https://doi.org/10. 3390/biomedicines9121803f Melenovsky V, Skaroupkova P, Benes J et al (2012) The course of heart failure development and mortality in rats with volume overload due to aorto-caval fistula. Kidney Blood Press Res 35:167–173. https://doi.org/10.1159/000331562 Husková Z, Kramer HJ, Thumová M et al (2006) Effects of anesthesia on plasma and kidney ANG II levels in normotensive and ANG II-dependent hypertensive rats. Kidney Blood Press Res 29:74–83. https://doi.org/10.1159/000092981 Mullens W, Verbrugge FH, Nijst P, Tang WHW (2017) Renal sodium avidity in heart failure: From pathophysiology to treatment strat- egies. Eur Heart J 38:1872–1882. https://doi.org/10.1093/eurhe artj/ehx035 1 3 3773 Naunyn-Schmiedeberg's Archives of Pharmacology (2023) 396:3757–3773 Murphy SP, Ibrahim NE, Januzzi JL (2020) Heart Failure with Reduced Ejection Fraction: A Review. JAMA - J Am Med Assoc 324:488– 504. 1161/HYPERTENSIONAHA.121.17360 Pelouch V, Dixon IMC, Sethi R, Dhalla NS (1993) Alteration of col- lagenous protein profile in congestive heart failure secondary to myocardial infarction. Mol Cell Biochem 129:121–131. https:// doi.org/10.1007/BF00926360 Sporková A, Jíchová S, Husková Z et al (2014) Different mechanisms of acute versus long-term antihypertensive effects of soluble epox- ide hydrolase inhibition: studies in Cyp1a1-Ren-2 transgenic rats. Clin Exp Pharmacol Physiol 41:1003–1013. https://doi.org/10. 1111/1440 1681 12310 https://doi.org/10.1111/bph.14153 https://doi.org/10.1001/jama.2020.10262 Sharma A, Verma S, Bhatt DL et al (2022) Optimizing Foundational Therapies in Patients With HFrEF: How Do We Translate These Findings Into Clinical Care? JACC Basic to Transl Sci 7:504–517. https://doi.org/10.1016/j.jacbts.2021.10.018 p g j Nakano Y, Hirano T, Uehara K et al (2008) New rat model induced by anti-glomerular basement membrane antibody shows severe glomerular adhesion in early stage and quickly progresses to end-stage renal failure. Pathol Int 58:361–370. https://doi.org/ 10.1111/j.1440-1827.2008.02237.x Simmonds SJ, Cuijpers I, Heymans S, Jones EAV (2020) Cellular and molecular differences between HFpEF and HFrEF: a step ahead in an improved pathological understanding. Cells 9(1):242. https:// doi.org/10.3390/cells9010242 g 10.1111/j.1440-1827.2008.02237.x Numata G, Takimoto E (2022) Cyclic GMP and PKG Signaling in Heart Failure. Front Pharmacol 13:1–11. https://doi.org/10.3389/ fphar.2022.792798 Singh P, Vijayakumar S, Kalogeroupoulos A, Butler J (2018) Multiple Avenues of Modulating the Nitric Oxide Pathway in Heart Failure Clinical Trials. Curr Heart Fail Rep 15:44–52. https://doi.org/10. 1007/s11897-018-0383-y Ouwerkerk W, Voors AA, Anker SD et al (2017) Determinants and clinical outcome of uptitration of ACE-inhibitors and beta- blockers in patients with heart failure: A prospective European study. Eur Heart J 38:1883–1890. https://doi.org/10.1093/eurhe artj/ehx026 Sobieraj P, Nilsson PM, Kahan T (2021) Heart Failure Events in a Clinical Trial on Arterial Hypertension: New Insights into the SPRINT Trial. Hypertension 78:1241–1247. https://doi.org/10. 1161/HYPERTENSIONAHA.121.17360f p y 1111/1440-1681.12310 https://doi.org/10.1177/41.12.8245430 Szeiffová Bačova B, Egan Beňová T, Viczenczová C, et al (2016) Car- diac connexin-43 and PKC signaling in rats with altered thyroid status without and with omega-3 fatty acids intake. Physiol Res 65 Suppl 1:S77–90 https://doi.org/10.33549/physiolres.933413i Sedláková L, Čertíková Chábová V, Doleželová Š et al (2017) Renin– angiotensin system blockade alone or combined with ETA recep- tor blockade: effects on the course of chronic kidney disease in 5/6 nephrectomized Ren-2 transgenic hypertensive rats. Clin Exp Hyper- tens 39:183–195. https://doi.org/10.1080/10641963.2016.1235184 Xia J, Hui N, Tian L et al (2022) Development of vericiguat: The first soluble guanylate cyclase (sGC) stimulator launched for heart failure with reduced ejection fraction (HFrEF). Biomed Pharma- cother 149:1–12. https://doi.org/10.1016/j.biopha.2022.112894 Sharkovska Y, Kalk P, Lawrenz B et al (2010) Nitric oxide-independent stimulation of soluble guanylate cyclase reduces organ damage in experimental low-renin and high-renin models. J Hypertens 28:1666–1675. https://doi.org/10.1097/HJH.0b013e32833b558c Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. 1 3
https://openalex.org/W2946272789	http://fox.leuphana.de/portal/files/13926527/information_10_00182_v2.pdf	English	null	Gain Adaptation in Sliding Mode Control Using Model Predictive Control and Disturbance Compensation with Application to Actuators	Information	2,019	cc-by	9,409	Link to publication Citation for pulished version (APA): Haus, B., Mercorelli, P., & Aschemann, H. (2019). Gain Adaptation in Sliding Mode Control Using Model Predictive Control and Disturbance Compensation with Application to Actuators. Information, 10(5), Article 182. https://doi.org/10.3390/info10050182 Gain Adaptation in Sliding Mode Control Using Model Predictive Control and Disturbance Compensation with Application to Actuators Haus, Benedikt; Mercorelli, Paolo; Aschemann, Harald Published in: Document Version Publisher's PDF, also known as Version of record Citation for pulished version (APA): Haus, B., Mercorelli, P., & Aschemann, H. (2019). Gain Adaptation in Sliding Mode Control Using Model Predictive Control and Disturbance Compensation with Application to Actuators. Information, 10(5), Article 182. https://doi.org/10.3390/info10050182 Received: 12 April 2019; Accepted: 22 May 2019; Published: 25 May 2019 Abstract: In this contribution, a gain adaptation for sliding mode control (SMC) is proposed that uses both linear model predictive control (LMPC) and an estimator-based disturbance compensation. Its application is demonstrated with an electromagnetic actuator. The SMC is based on a second-order model of the electric actuator, a direct current (DC) drive, where the current dynamics and the dynamics of the motor angular velocity are addressed. The error dynamics of the SMC are stabilized by a moving horizon MPC and a Kalman ﬁlter (KF) that estimates a lumped disturbance variable. In the application under consideration, this lumped disturbance variable accounts for nonlinear friction as well as model uncertainty. Simulation results point out the beneﬁts regarding a reduction of chattering and a high control accuracy. Keywords: sliding mode control; model predictive control; adaptive control; disturbance estimation; actuators Received: 12 April 2019; Accepted: 22 May 2019; Published: 25 May 2019 1. Introduction and Literature Review 1.1. Adaptive Sliding Mode Control and Model Predictive Control Gain Adaptation in Sliding Mode Control Using Model Predictive Control and Disturbance Compensation with Application to Actuators Benedikt Haus 1,∗ , Paolo Mercorelli 1 and Harald Aschemann 2 1 Institute of Product and Process Innovation, Leuphana University of Lüneburg, Volgershall 1, D-21339 Lüneburg, Germany; mercorelli@leuphana.de 2 Chair of Mechatronics, University of Rostock, Justus-von-Liebig-Weg 6, D-18059 Rostock, Germa harald.aschemann@uni-rostock.de 1 Institute of Product and Process Innovation, Leuphana University of Lüneburg, Volgershall 1, D-21339 Lüneburg, Germany; mercorelli@leuphana.de 2 Chair of Mechatronics, University of Rostock, Justus-von-Liebig-Weg 6, D-18059 Rostock, Germany; harald.aschemann@uni-rostock.de C d h l h d * Correspondence: haus@leuphana.de General rights C i ht d General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private st • You may not further distribute the material or use it for any profit making activity or commercial gain • You may not further distribute the material or use it for any profit-making activity or • You may freely distribute the URL identifying the publication in the public portal ? y y p g y • You may freely distribute the URL identifying the publication in the public portal ? Take down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Take down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Take down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim Download date: 24. Okt.. 2024 information information 1.1. Adaptive Sliding Mode Control and Model Predictive Control While belonging to perhaps the most robust and versatile control strategies, sliding mode control (SMC) tends to suffer from high energy consumption and high-frequency oscillations in system inputs, states or even outputs, which certainly is to be avoided in tracking problems. Currently, many remedies have been proposed and successfully realized to deal with these problems. A very important one is the so-called boundary layer approach, see e.g., [1], which introduces a permissible region around the sliding surface. This layer is characterized by its thickness, inside which no switching of the control input takes place. Another frequently applied approach to chattering reduction is higher-order SMC [2], which can also be combined with a boundary layer concept [3]. Nevertheless, powerful alternatives exist towards a model-based or signal-based adaptation of the switching amplitudes. A model-based approach is shown in [4], where model reference adaptive control (MRAC) is employed to adjust the SMC gain in an application of a brushless direct current (DC) motor. The signal-based approaches discussed in the literature are generally based on integral-type scheduling rules, typically case distinctions depending on a norm or an absolute value of the sliding surface, see e.g., [5,6]. The contribution at hand looks at an innovative model-based adaptation. Thanks to countermeasures like these against chattering, SMC is suitable to be widely used in industry. In the context of drive applications, for example, it has even been used to reduce torque ripple [7]. Information 2019, 10, 182; doi:10.3390/info10050182 www.mdpi.com/journal/information www.mdpi.com/journal/information 2 of 19 Information 2019, 10, 182 Model predictive control (MPC) still constitutes a developing research ﬁeld in the context of machines and drives, though many applications already use such control strategies, e.g., in [8] for a permament magnet synchronous machine (PMSM) or in [9] and in [10] for a DC drive. The MPC approach takes into account model-based predictions and determines the control inputs by minimizing a cost function. However, compared to classical controllers like proportional-integral-derivative (PID) control, this method from the ﬁeld of optimal control requires a relatively high modeling accuracy in order to yield acceptable results. SMC, on the other hand, is known for being robust against disturbances, model mismatch and parametric uncertainties. 1.3. Actuators As important mechatronic components, electromagnetic actuators are used in many technical applications, in particular in the automotive industry and in industrial production systems. In production systems, they play a key role in motion control and precise positioning. Mechanical, pneumatic or hydraulic components tend to be replaced by electromagnetic actuators due to their high efficiency, excellent dynamic behavior and cleanliness. An important effect that needs to be considered in the mechanical part of actuators is nonlinear friction. An extended survey of friction modeling is given in [18] including a large number of literature references. Recent contributions mark progress in terms of identiﬁcation of friction phenomena and their compensation [19]. 1.2. Estimation and Kalman Filters Thanks to the high computing power of modern processors, micro-controllers, or even ﬁeld-programmable gate arrays (FPGAs), it has become possible to deploy intelligent and sophisticated control approaches, e.g., observer-based control, utilizing only a minimal number of sensors, see [11,12]. Contributions like [13–16] reﬂect the progress in theoretical studies of Kalman ﬁlters (KFs), especially concerning robustness and the ability to deal with unknown or inaccessible disturbances or model uncertainty. In many situations, time-varying disturbances like friction effects can be modeled as additional unknown inputs. In [17] for instance, a two-stage KF is implemented to estimate the pressure disturbance inside a cylinder of an internal combustion engine and its effect on the controlled output. 1.4. Main Contribution In this paper, a combination of SMC and linear MPC is proposed to create an adaptive control method. Here, LMPC adapts the switching height of the discontinuous control part and, thereby, reduces the undesired chattering effect. The combination of SMC and LMPC allows for an exploitation of the beneﬁts of both worlds, gaining both robustness and a degree of optimality with regard to the speciﬁc MPC cost function—at the cost of a more sophisticated control design as well as an increased implementation and computation effort. The introduction of a cost function and, therefore, of an optimality measure allows the intuitive balancing of the error convergence rate versus chattering amplitude penalties, in order to achieve a reasonable trade-off. Linear MPC is a straight-forward, easily implementable way of minimizing such a cost function. Furthermore, an augmented linear KF is employed with the primary goal to contribute to a fast convergence to the sliding surface, thus to unburden the switching control part of the SMC and to reduce undesired chattering. To achieve this goal, the KF estimates the disturbance (and also the ﬁrst derivative with regard to time), which is considered as an unknown input, and is subsequently used for a compensation in the SMC law. Also, since the cost function that is minimized by the MPC includes the predicted tracking error stemming from a model-based prediction scheme, an accurate model of the SMC sliding surface dynamics is necessary to allow for an optimally small SMC gain. To ensure that the underlying model assumptions of the MPC design are met, a compensation of lumped disturbances is mandatory. Thanks to the disturbance compensation using the estimates of the KF, an accurate prediction over a ﬁnite horizon becomes possible in compliance with the underlying assumptions at the KF design. The disturbance compensation may also be interpreted as a lowering of the necessary minimum SMC gain because 3 of 19 Information 2019, 10, 182 part of the disturbances are compensated for by the KF estimates—and the SMC is disburdened. In order to demonstrate the properties of the proposed control method in a practice-relevant ﬁeld, this paper considers a DC drive system that is subject to both nonlinear friction and model uncertainty. The nonlinearity is represented by the sum of the Coulomb friction model and a quadratic term depending on the relative velocity. 1.4. Main Contribution To conclude, the goal of this contribution, which represents an extension of a conference paper [20], is to conceive an SMC with optimal adaptivity that can be implemented as simply as possible. Its effectiveness is demonstrated in simulations, subject to realistic conditions regarding disturbances and model uncertainty, controlling the angular velocity of a DC drive including a nonlinear friction model. The paper is structured as follows: • Section 2 presents the physically-based model of a DC drive that is affected by a nonlinear friction torque and model uncertainty. • Section 2 presents the physically-based model of a DC drive that is affected by a nonlinear friction torque and model uncertainty. • The feedback control design is described in Section 3, where • Section 3.1 contains details on the employed SMC techniques which involve a combination of a continuous control action and a discontinuous switching part, and where • in Section 3.2, the height of the switching control action is adapted using MPC techniques to counteract undesired chattering. • Moreover, an unknown lumped disturbance—accounting for nonlinear friction and model uncertainty in the equation of motion—is estimated in Section 3.3 by a KF. This estimate is employed subsequently in the error dynamics for compensation purposes and, as a result, contributes to the reduction of the necessary switching height determined by MPC. • The beneﬁts are shown by meaningful simulation results in Section 4. • The beneﬁts are shown by meaningful simulation results in Section 4. • Finally, conclusions are given in Section 5. • Finally, conclusions are given in Section 5. 2. System Modelling It turns out that the approach can be generalized by considering a lumped disturbance torque d(t) = Tr(ω(t)) + Tu(t), where Tu(t) represents any further external loads torques, unmodelled dynamic effects and model uncertainty. The estimate can be subsequently used for a disturbance compensation. The modiﬁed system model is then given by di(t) dt = 1 L (u(t) −Ri(t) −KTω(t)) , (4) dω(t) dt = 1 J (KTi(t) −d(t))) . (5) (4) (5) In the sequel, the latter approach using a KF is followed because it promises a higher tracking accuracy. Moreover, the estimator dynamics can be speciﬁed appropriately in the KF design. In the sequel, the latter approach using a KF is followed because it promises a higher tracking accuracy. Moreover, the estimator dynamics can be speciﬁed appropriately in the KF design. 2. System Modelling The system model is based on physical considerations and involves ordinary differential equations for the armature current i(t) and the motor angular velocity ω(t) di(t) dt = 1 L (u(t) −Ri(t) −KTω(t)) , (1) dω(t) dt = 1 J (KTi(t) −Tr(ω(t))) . (2) (1) (2) Here, a nonlinear friction torque Here, a nonlinear friction torque Tr(ω(t)) = K f ω2(t) + Tr0 sign(ω(t)) (3) (3) is introduced, where K f > 0 denotes a coefﬁcient related to the quadratic term in the motor angular velocity, and Tr0 characterizes the Coulomb friction part. The nonlinear friction torque Tr(ω) = Tr(ω(t)) is depicted in Figure 1. This friction model is implemented in a regularized form and used in the simulation studies to represent nonlinear friction. 4 of 19 Information 2019, 10, 182 Figure 1. Nonlinear friction characteristic Tr(ω). Given this model description two alternatives seem to be promising to address nonlinear friction: Figure 1. Nonlinear friction characteristic Tr(ω). Given this model description, two alternatives seem to be promising to address nonlinear friction: Given this model description, two alternatives seem to be promising to address nonlinear friction 1. Feedback disturbance compensation: In this solution, the friction term (3) is assumed as known and explicitly included in the sliding mode control design. The corresponding parameters are identiﬁed beforehand by the least-squares method. In the envisaged sliding-mode control design this would involve a time differentiation of the friction model and a compensation by means of feedback. It is clear that any changes of the friction model afterwards results in an imperfect compensation. 2. Estimator-based disturbance compensation: In this approach, the detailed physical model for the nonlinear friction (3) is not employed at all in the control design. Instead, nonlinear friction is estimated by a Kalman ﬁlter. It turns out that the approach can be generalized by considering a lumped disturbance torque d(t) = Tr(ω(t)) + Tu(t), where Tu(t) represents any further external loads torques, unmodelled dynamic effects and model uncertainty. The estimate can be subsequently used for a disturbance compensation. The modiﬁed system model is then given by 2. Estimator-based disturbance compensation: In this approach, the detailed physical model for the nonlinear friction (3) is not employed at all in the control design. Instead, nonlinear friction is estimated by a Kalman ﬁlter. 3. Control Design The implementation of the overall control structure corresponds to the block diagram shown in Figure 2. The control input involves an equivalent control action, a disturbance compensation and a robustifying switching term. In this contribution, the switching height is adapted by means of a quasi-linear MPC. 5 of 19 Information 2019, 10, 182 Figure 2. Implementation of the sliding mode control (SMC) in combination with a Kalman ﬁlter (KF) for state and parameter estimation and an model predictive control (MPC) for the adaptation of the switching height. The gray block represents the direct current (DC) drive. Figure 2. Implementation of the sliding mode control (SMC) in combination with a Kalman ﬁlter (KF) for state and parameter estimation and an model predictive control (MPC) for the adaptation of the switching height. The gray block represents the direct current (DC) drive. 3.1. Feedback Control Design Using SMC 3.1. Feedback Control Design Using SMC For the derivation of the SMC law, a quadratic Lyapunov function candidate based on the integral sliding surface is considered as well using a Kalman ﬁlter with a suitable disturbance model. For the derivation of the SMC law, a quadratic Lyapunov function candidate based on the integral sliding surface is considered V(t) = 1 2s(t)2. (8) (8) The time derivative of the Lyapunov function candidate can be easily calculated. It has to fulﬁll the sliding condition, which is chosen as follows in this paper The time derivative of the Lyapunov function candidate can be easily calculated. It has to fulﬁll the sliding condition, which is chosen as follows in this paper ˙V(t) = s(t)˙s(t) ≤s(t) (−λs(t) −βsgn(s(t))) = −λs(t)2 −β\|s(t)\|. (9) (9) Now, all known terms in the time derivative ˙s(t) are compensated for by feedback, which leads to the following expression for the equivalent control ueq(t) = JL KT " ¨ωd(t) + KTR JL i(t) + K2 T JL ω(t) + α ˙ωd(t) −KT J i(t) + η (ωd(t) −ω(t)) # . (10) (10) Then, the time derivative of the sliding surface becomes Then, the time derivative of the sliding surface becomes ˙s = −KT JL [u(t) −ueq(t)] + 1 J ˙d(t) + α J d(t). (11) (11) The unknown disturbance as well as its time derivative are estimated by a Kalman ﬁlter. A disturbance compensation based on these estimates ˙ˆd and ˆd, hence, contributes to a reduction of the discontinuous switching term, i.e., to a suppression of undesired chattering. The disturbance compensation law results in The unknown disturbance as well as its time derivative are estimated by a Kalman ﬁlter. A disturbance compensation based on these estimates ˙ˆd and ˆd, hence, contributes to a reduction of the discontinuous switching term, i.e., to a suppression of undesired chattering. The disturbance compensation law results in udc(t) = L KT ˙ˆd(t) + αL KT ˆd(t). (12) (12) Finally, the switching part usw(t) can be derived from the sliding condition Finally, the switching part usw(t) can be derived from the sliding condition ˙s(t) = −KT JL usw(t) ≤−λs(t) −βsgn(s(t)). 3.1. Feedback Control Design Using SMC (13) (13) If the equality sign holds, the switching part is given by If the equality sign holds, the switching part is given by If the equality sign holds, the switching part is given by If the equality sign holds, the switching part is given by usw(t) = JL KT (λs(t) + βsgn(s(t))) . (14) (14) The overall SMC law comprises the sum of all three terms The overall SMC law comprises the sum of all three terms The overall SMC law comprises the sum of all three terms The overall SMC law comprises the sum of all three terms u(t) = ueq(t) + udc(t) + usw(t). (15) (15) Outside the boundary layer – during the convergence to the sliding surface – the time derivative ˙s(t) is governed by the nonlinear error dynamics Outside the boundary layer – during the convergence to the sliding surface – the time derivative ˙s(t) is governed by the nonlinear error dynamics ˙s(t) = −λs(t) −βsgn(s(t)). (16) (16) 3.2. Adaption of the Switching Height Using MPC 3.1. Feedback Control Design Using SMC Since the lumped disturbance d(t) is estimated by a KF and used for a subsequent disturbance compensation, the state-space representation (4) and (5) can be used in the derivation of an integral SMC. It is worth mentioning that the estimated disturbance compensates the largest part of the lumped disturbance and signiﬁcantly increases the tracking accuracy. As a result, the SMC has to cope with model imperfections that are related to the dynamics of the estimator only—leading to a signiﬁcant reduction of the necessary switching height and, thereby, reducing the undesired chattering effect. The SMC design aims at a highly accurate tracking of desired trajectories for the angular velocity with smallest possible tracking errors e(t) := ωd(t) −ω(t). For that purpose, an integral sliding surface is introduced as follows t s(t) = ˙e(t) + αe(t) + η Z t 0 e(τ) dτ −e(0), (6) (6) where e(0) represents an initial error. Its presence in s(t) could eliminate the reaching phase (s(0) = 0), see [21]. In this contribution, however, e(0) is assumed to be unknown and is set to zero in the implementation. The coefﬁcients α ∈R and η ∈R have to be positive. The time derivative of the sliding surface can be easily computed and results in ˙s(t) = ¨ωd(t) −¨ω(t) + α( ˙ωd(t) −˙ω(t)) + η (ωd(t) −ω(t)) = ¨ωd(t) − KT J di(t) dt −1 J ˙d(t) + α ˙ωd(t) − KT J i(t) −1 J d(t) + η (ωd(t) −ω(t)) = ¨ωd(t) − KT J 1 Lu(t) −R L i(t) −KT L ω(t) −1 J ˙d(t) + α ˙ωd(t) − KT J i(t) −1 J d(t) + η (ωd(t) −ω(t)) = ¨ωd(t) −KT JL u(t) + KTR JL i(t) + K2 T JL ω(t) + 1 J ˙d(t) + α ˙ωd(t) −KT J i(t) + 1 J d(t) + η (ωd(t) −ω(t)) . (7) (7) It becomes obvious that the time derivative ˙d(t) of the lumped disturbance affects the time derivative of the sliding surface ˙s(t)—and, hence, is needed in the SMC law. This explains why it is estimated mation 2019, 10, 182 6 of 1 6 of 19 Information 2019, 10, 182 as well using a Kalman ﬁlter with a suitable disturbance model. 3.2. Adaption of the Switching Height Using MPC The main idea of this paper is now to use MPC techniques to determine an optimal switching height β(k). For that purpose, the error dynamics is discretized with regard to time using the explicit Information 2019, 10, 182 7 of 19 Euler method with a sampling time of Ts = 10 µs, and the switching height β(k) is introduced as control input for the MPC Euler method with a sampling time of Ts = 10 µs, and the switching height β(k) is introduced as control input for the MPC s(k + 1) = (1 −λTs) s(k) −β(k)Tssgn(s(k)). (17) (17) 3.2.1. Converge Properties Outside the Boundary Layer Outside the boundary layer, from (17), a possible discrete-time state-space representation results Outside the boundary layer, from (17), a possible discrete-time state-space representation results s(k + 1) = ak,rs(k) + bk,rβ(k), y(k) = cks(k), (18) ⇒ak,r = 1 −λTs, bk,r = −Tssgn(s(k)), ck = c = 1, (18) where subscript r indicates the reaching phase characteristics. By repeated evaluations of the difference equation, the system behavior can be predicted as where subscript r indicates the reaching phase characteristics. By repeated evaluations of the difference equation, the system behavior can be predicted as ˆy(k + 1) = c ak,r s(k) + c bk,r β(k) (19) ˆy(k + 2) = c a2 k,r s(k) + c ak,r bk,r β(k) + c bk+1,r β(k + 1), ˆy(k + 3) = c a3 k,r s(k) + c a2 k,r bk,r β(k) + c ak,r bk+1,r β(k + 1) + c bk+2,r β(k + 2), (19) etc. It is straightforward to show that the following vector expression holds etc. It is straightforward to show that the following vector expression holds ˆy(k) = grs(k) + Fk,ruk, (20) ˆy(k) = grs(k) + Fk,ruk, (20) with with with ˆy(k) =   ˆy(k + 1) ˆy(k + 2) ... ˆy(k + p)  , uk =   β(k) β(k + 1) ... β(k + p −1)  , (21) (21) and a prediction horizon of length p. The system matrices for use in the MPC become gr =   c ak,r c a2 k,r ... c ap k,r  , Fk,r =   cbk,r 0 ... 0 cak,rbk,r cbk+1,r ... 0 ... ... ... ... cap−1 k,r bk,r cap−2 k,r bk+1,r ... 3.2.2. Convergence Properties Inside the Boundary Layer As a second measure against chattering—in addition to the disturbance compensation by means of the Kalman ﬁlter—a regularized version of the switching law is employed. This leads to the deﬁnition of a boundary layer resulting from the replacement of sgn(s(k)) by a smoothed version given by the saturation function sat s(k) Φ . Inside the boundary layer, the saturation function is linear, resulting in sat s(k) Φ = s(k) Φ : s(k + 1) = (1 −Tsλ) s(k) −Ts s(k) Φ β(k). (26) (26) This expression contains a multiplication of a state variable with the input variable and is, hence, a nonlinear term. A ﬁrst-order multivariate Taylor linearization of the function f (xT) = s(k) Φ β(k), where the vector xT = h β(k) s(k) iT denotes the independent variables, can be performed in the operating point x∗ T = h β∗(k −1) s∗(k −1) iT , which contains known values at the discrete point at time k −1. Here, the star symbol (·)∗denotes the operating point and allows for representing the corresponding values in the following equations. The Taylor series expansion up to linear terms becomes This expression contains a multiplication of a state variable with the input variable and is, hence, a nonlinear term. A ﬁrst-order multivariate Taylor linearization of the function f (xT) = s(k) Φ β(k), where the vector xT = h β(k) s(k) iT denotes the independent variables, can be performed in the operating point x∗ T = h β∗(k −1) s∗(k −1) iT , which contains known values at the discrete point at time k −1. Here, the star symbol (·)∗denotes the operating point and allows for representing the corresponding values in the following equations. The Taylor series expansion up to linear terms becomes f (xT) ≈f (x∗ T) + ∇f (x∗ T) (xT −x∗ T) , (27) (27) he operator ∇indicates the gradient of f, a row vector. A detailed description can be stated ws s(k) Φ β(k) ≈s∗(k −1) Φ β∗(k −1) + s∗(k −1) Φ (β(k) −β∗(k −1)) + β∗(k −1) Φ (s(k) −s∗(k −1)) (28) = s∗(k −1) Φ β∗(k −1) + s∗(k −1) Φ β(k) + β∗(k −1) Φ s(k) −2s∗(k −1)β∗(k −1) Φ (29) = s∗(k −1) Φ β(k) + β∗(k −1) Φ s(k) −s∗(k −1)β∗(k −1) Φ . 3.2. Adaption of the Switching Height Using MPC cbk+p−1,r  , (22) (22) where the prediction horizon p should not be chosen as too large, considering that bk,r might change unpredictably. For p = 2, these matrices simplify to gr = " 1 −λTs (1 −λTs)2 # , Fk,r = −Ts " sgn(s(k)) 0 (1 −λTs) sgn(s(k)) sgn(s(k + 1)) # . (23) (23) With given sgn(s(k)), sgn(s(k + 1)) can be obtained from the prediction step (18) using the second element of the input vector uk−1 computed in the previous step. Given the system according to (20), an optimal input β(k) has to be calculated that minimizes the following cost function J(k) = 1 2 yd(k) −ˆy(k) T Q yd(k) −ˆy(k) + 1 2uT k Ruk, (24) (24) where Q ≥0 and R > 0 are symmetric non-negative deﬁnite matrices. Moreoever, yd(k) is the sliding surface reference trajectory for the next p time steps. In this case, its elements can simply be set to zero. The corresponding solution can be stated in closed form uk = (FT k,rQFk,r + R)−1FT k,rQ yd(k) −gy(k) , (25) (25) Information 2019, 10, 182 8 of 19 where y(k) = s(k) holds and the sliding mode control switching gain β = β(k) is now chosen as the ﬁrst element of uk. The MPC with the predicted sliding surface is realized with matrices (18) that do not depend on any speciﬁc systems parameters. As a result, this approach offers an intrinsic robustness regarding the prediction. 3.2.2. Convergence Properties Inside the Boundary Layer (30) (28) (29) (30) By substituting this Taylor linearization into the difference equation, a linear ﬁrst-order discrete-time model can be derived s(k + 1) = 1 −Tsλ −Ts β∗(k −1) Φ \| {z } ak s(k) + −Tss∗(k −1) Φ \| {z } bk β(k) + w∗(k −1), (31) (31) where w∗(k −1) = Ts Φ s∗(k −1)β∗(k −1) represents a known term consisting on past information, i.e., from the previous time step. The discrete-time system matrix and the input vector become scalars and are given by ak and bk, respectively. Note that these terms are time-dependent. The output equation is scalar as well and can be stated as where w∗(k −1) = Ts Φ s∗(k −1)β∗(k −1) represents a known term consisting on past information, i.e., from the previous time step. The discrete-time system matrix and the input vector become scalars and are given by ak and bk, respectively. Note that these terms are time-dependent. The output equation is scalar as well and can be stated as y(k) = ck s(k), ck = c = 1. (32) (32) Information 2019, 10, 182 9 of 19 In analogy to the approach outside the boundary layer, the design of a quasi-linear MPC is presented that is based on the discrete-time model above. The moving prediction horizon comprises two steps in the future. With the following vectors and matrices gk = " ak ak · ak+1 # = " 1 −Tsλ −Ts β∗(k−1) Φ 1 −Tsλ −Ts β∗(k−1) Φ · 1 −Tsλ −Ts β∗(k) Φ # , (33) Fk = " c · bk 0 c · ak · bk c · bk+1 # = −Ts Φ " s∗(k −1) 0 1 −Tsλ −Ts β∗(k−1) Φ · s∗(k −1) s∗(k) # , (34) wk = " c c · ak + c # = " 1 2 −Tsλ −Ts β∗(k−1) Φ # , (35) (33) (34) 1 2 −Tsλ −Ts β∗(k−1) Φ # , (35) (35) wk = " c c · ak + c # = " 1 2 −Tsλ −Ts β∗(k−1) Φ # , (35) the solution can be determined in a closed-form expression as follows the solution can be determined in a closed-form expression as follows uk = (FT k QFk + R)−1FT k Q [yd(k) −gk · y(k) −wk · w∗(k −1)] . 3.2.2. Convergence Properties Inside the Boundary Layer (36) (36) Here, the output y(k) = s(k) is identical to the current value of the sliding surface, whereas the time-varying SMC switching gain β is chosen again as the ﬁrst element of the computed input vector uk = h ˜βk(k) ˜βk(k + 1) iT . As the input β∗(k), which shows up in gk and represents the linearization point for β, is not yet available, it is substituted by the second element of the input uk−1, which corresponds to previous time step. KF for the Estimation of a Lumped Disturbance Torque 3.3. KF for the Estimation of a Lumped Disturbance Torque 3.3. KF for the Estimation of a Lumped Disturbance Torque The model deﬁning the KF prediction step can be stated in state-space form, with the input variable u(t), xKF(t) = h [i(t) ω(t) d(t) ˙d(t) iT , (41) ˙xKF(t) = AKFxKF(t) + bKFu(t), (42) ym(t) = CKFxKF(t) (43) xKF(t) = h [i(t) ω(t) d(t) ˙d(t) iT , (41) ˙xKF(t) = AKFxKF(t) + bKFu(t), (42) ym(t) = CKFxKF(t) (43) (41) (41) (41) (42) (43) (43) AKF =   −R L −K L 0 0 K J 0 −1 J 0 0 0 0 1 0 0 0 0  , bKF =   1 L 0 0 0  , CKF = " 1 0 0 0 0 1 0 0 # . (44) (44) The discrete-time state space model can be obtained using explicit Euler discretization with step width Ts, after which the random variables wKF and vKF are introduced to represent white, uncorrelated process and measurement noise, respectively, with normal probability distributions [22]. (45) (46) (47) (48) xKF(k + 1) = AKFdxKF(k) + bKFdu(k) + wKF(k), (45) yKF(k) = CKFxKF(k) + vKF(k), (46) AKFd = I4×4 + TsAKF, (47) bKFd = TsbKF. (48) (45) (46) (47) (48) bKFd = TsbKF. (48) (48) Based on this, the a-priori estimates are calculated in the prediction step of the Kalman ﬁlter algorithm according to Based on this, the a-priori estimates are calculated in the prediction step of the Kalman ﬁlter algorithm according to ˆx− KF(k + 1) = AKFd ˆx+ KF(k) + bKFdu(k). (49) (49) For the ﬁrst step, initial values ˆx+ KF(0) can be either speciﬁed by the user or simply set to zero. The same applies to the initial uncertainty P+(0) in the following equation. The a-priori estimate of the covariance matrix is P−(k + 1) = AKFdP+(k)AT KFd + QKF, (50 (50) where QKF represents a R4×4 matrix quantifying the covariance matrix of the process noise wKF, QKF(k) = E wKF(k)wKF(k)T =   qi 0 0 0 0 qω 0 0 0 0 qd 0 0 0 0 q ˙d  . (51) (51) Here, QKF is assumed to be constant, diagonal and positive and is treated as a tuning parameter matrix. Parameter qd is set to zero since Equation (39) is a certain relationship, while q ˙d is assigned a large value since Equation (40) is not. Parameters qi and qω reﬂect the modeling uncertainty concerning Equations (38) and (37). 3.3. KF for the Estimation of a Lumped Disturbance Torque In the sequel, the combined estimation of the state variables and the external disturbance as well as its time derivative is described. The design of a corresponding KF is based on the modiﬁed system model, including a double-integrator disturbance model di(t) dt = 1 L (u(t) −Ri(t) −KTω(t)) (37) dω(t) dt = 1 J (KTi(t) −d(t))) (38) dd(t) dt = ˙d(t) (39) d ˙d(t) dt = 0 (40) di(t) dt = 1 L (u(t) −Ri(t) −KTω(t)) (37) dω(t) dt = 1 J (KTi(t) −d(t))) (38) dd(t) dt = ˙d(t) (39) d ˙d(t) dt = 0 (40) (37) (38) (39) (40) and aims at providing estimates for both state variables, an estimate ˆd for the unknown lumped disturbance and its derivative ˙ˆd. It is worth mentioning that the chain of two integrators has no input so far. Nevertheless, this integrator chain is driven by the output error feedback as well as the process noise—the stochastic part—in the framework of the KF design. As a result, the estimator states vary during the operation of the KF and highly accurate estimates are obtained for a subsequent compensation in the control structure. Whenever feedback control is applied, it is necessary to measure selected system outputs. Under realistic conditions, however, measurements are affected by errors like deterministic offsets and stochastic disturbances, e.g., white noise processes. In such cases, a KF can be advantageously employed and provides estimates with minimum covariances. The optimality conditions include an accurate system model and the knowledge about the noise characteristics. In the given case, the system model (4) and (5)—that contains a perfectly-known part and the unknown lumped disturbance—is both complete and correct. As conﬁrmed by the simulation results, the quality of the estimates is high and, hence, an accurate system model is obtained due to the estimates. The typical design of a KF addresses uncorrelated process noise and measurement noise that are assumed to be Gaussian, 10 of 19 Information 2019, 10, 182 white and with a vanishing mean value. Despite the fact that in practice the stochastic noise processes are often not perfectly known, the KF algorithm is usually still capable of providing meaningful state and disturbance estimates. The covariances can then be considered as tuning parameters like in the linear-quadratic regulator (LQR) control design. 3.3. KF for the Estimation of a Lumped Disturbance Torque The Kalman gain can now be calculated as K(k + 1) = P−(k + 1)CT KF(CKFP−(k + 1)CT KF + RKF)−1, (52) (52) 11 of 19 Information 2019, 10, 182 with the measurement matrix CKF according to the measured outputs ym(t) = CxKF(t). Here, the measurement covariance matrix RKF is related to the measurement noise vKF and has properties similar to QKF with the measurement matrix CKF according to the measured outputs ym(t) = CxKF(t). Here, the measurement covariance matrix RKF is related to the measurement noise vKF and has properties similar to QKF RKF = E vKF(k)vKF(k)T = " ri 0 0 rω # . (53) (53) In the correction step of the Kalman ﬁlter algorithm, the a-posteriori estimates for covariance and states are calculated as follows In the correction step of the Kalman ﬁlter algorithm, the a-posteriori estimates for covariance and states are calculated as follows P+(k + 1) = (I4×4 −K(k + 1)CKF) P−(k + 1), (54) ˆx+ KF(k + 1) = ˆx− KF(k + 1) + ∆xKF(k + 1), (55) (54) (55) where the correction ∆xKF(k + 1) consists of the innovation term weighted by the Kalman gain K(k + 1) ∆xKF(k + 1) = K(k + 1) ym(k + 1) −CKF ˆx− KF(k + 1) , (56) he correction ∆xKF(k + 1) consists of the innovation term weighted by the Kalman gain K(k + 1) ∆xKF(k + 1) = K(k + 1) ym(k + 1) −CKF ˆx− KF(k + 1) , (56) (56) with the measured current and velocity ym = h im ωm iT . The estimated states as well as the estimated disturbance and its time derivative are used in the SMC control law. with the measured current and velocity ym = h im ωm iT . The estimated states as well as the estimated disturbance and its time derivative are used in the SMC control law. 4. Simulations 4. Simulations In this section, simulation results are presented. They were obtained using a sampling time of Ts = 10 µs. The velocity proﬁle to be tracked was generated from step-like signals using a second-order low-pass command-shaping ﬁlter GLP(s) = 1 1 102 s2 + 2 10s + 1 . (57) (57) 4.1. Simulation Settings and Scenarios It becomes obvious that such strong discontinuities can be managed properly by an integrator disturbance model upon which the KF design is based—even though they represent the worst case for the observer part of the combined control system. Step-like changes are estimated accordingly complying with the KF estimation error dynamics. Figure 3. Friction torque Tr (left) and load torque Tl (right). Figure 3. Friction torque Tr (left) and load torque Tl (right). Two alternative load variants were tested: The red signal shows a sinusoidal signal with steps at 0.5 s and 1.5 s; all other ﬁgures were created using this load proﬁle. It must be pointed out that also with pulsed load torques, i.e., the blue signal, the results were equally good. It becomes obvious that such strong discontinuities can be managed properly by an integrator disturbance model upon which the KF design is based—even though they represent the worst case for the observer part of the combined control system. Step-like changes are estimated accordingly complying with the KF estimation error dynamics. 4.1. Simulation Settings and Scenarios 4.1. Simulation Settings and Scenarios .1. Simulation Settings and Scenarios The covariance matrices of the Kalman ﬁlter were chosen according to The covariance matrices of the Kalman ﬁlter were chosen according to RKF = diag (0.001, 500) , QKF = diag (0.001, 0.001, 0, 0.5) , (58) (58) where the third and fourth element on the diagonal of QKF relate to the double integrator disturbance model, and with the initial conditions where the third and fourth element on the diagonal of QKF relate to the double integrator disturbance model, and with the initial conditions P+(0) = diag 103, 103, 0, 103 , ˆx+ KF(0) = h 0 0 0 0 iT . (59) (59) The proposed control strategy, i.e., an adaption of the SMC switching height β by means of MPC, is compared with two other, more classical variants. These three variants are as follows: The proposed control strategy, i.e., an adaption of the SMC switching height β by means of MPC, is compared with two other, more classical variants. These three variants are as follows: 1. SMC with β = const. and switching control law (14), 2. SMC with β = const. and usw = JL KT (λs(t) + βsat(s(t)/Φ)), 3. SMC with adaptive β = β(k) and usw = JL KT (λs(t) + β(k)sat(s(t)/Φ)), i.e., the proposed strategy. The SMC design parameter λ affecting the linear term in the sliding condition was set to zero in all variants. This corresponds to the classical choice of the sliding condition. To allow for a fair comparison, the switching height β = const. = 2 × 107 and the boundary layer thickness Φ = 200 were iteratively tuned to achieve good tracking properties while maintaining robustness and only small chattering in the presence of disturbances. These disturbances were implemented as load torques, see Figure 3 (right). 12 of 19 12 of 19 Information 2019, 10, 182 Figure 3. Friction torque Tr (left) and load torque Tl (right). Two alternative load variants were tested: The red signal shows a sinusoidal signal with steps at 0.5 s and 1.5 s; all other ﬁgures were created using this load proﬁle. It must be pointed out that also with pulsed load torques, i.e., the blue signal, the results were equally good. 4.2. Results The resulting velocity tracking is demonstrated in Figure 4, where the tracking error and error energy are depicted, and Figure 5, which shows the angular velocity itself. The resulting velocity tracking is demonstrated in Figure 4, where the tracking error and error energy are depicted, and Figure 5, which shows the angular velocity itself. The resulting velocity tracking is demonstrated in Figure 4, where the tracking error and error energy are depicted, and Figure 5, which shows the angular velocity itself. Figure 4. Simulated values for the tracking errors e(t) for all the variants (left) and the integrated squared error for all the variants (right). Figure 4. Simulated values for the tracking errors e(t) for all the variants (left) and the integrated squared error for all the variants (right). 13 of 19 13 of 19 Information 2019, 10, 182 Figure 5. Tracking behaviour of the angular velocity ω: comparison of desired and simulated values for all variants (top, left); detailed views at point with the lowest angular velocity (top, right; bottom). While all the variants manage to track the desired velocity proﬁle almost perfectly (top left of Figure 5. Tracking behaviour of the angular velocity ω: comparison of desired and simulated values for all variants (top, left); detailed views at point with the lowest angular velocity (top, right; bottom). While all the variants manage to track the desired velocity proﬁle almost perfectly (top left of Figure 5), variant 1 shows a little more ripple (top right) than variants 2 and 3 (bottom). Additionally, variant 3 shows smaller deviations in the case of load torque steps (Figure 6), thanks to a momentarily larger switching height β. This effect results in a signiﬁcantly smaller error energy, see Figure 4 (right). While all the variants manage to track the desired velocity proﬁle almost perfectly (top left of Figure 5), variant 1 shows a little more ripple (top right) than variants 2 and 3 (bottom). Additionally, variant 3 shows smaller deviations in the case of load torque steps (Figure 6), thanks to a momentarily larger switching height β. This effect results in a signiﬁcantly smaller error energy, see Figure 4 (right). While all the variants manage to track the desired velocity proﬁle almost perfectly (top left of Figure 5), variant 1 shows a little more ripple (top right) than variants 2 and 3 (bottom). 4.2. Results Additionally, variant 3 shows smaller deviations in the case of load torque steps (Figure 6), thanks to a momentarily larger switching height β. This effect results in a signiﬁcantly smaller error energy, see Figure 4 (right). Figure 6. Detailed view of the velocity tracking after load torque steps. Figure 6. Detailed view of the velocity tracking after load torque steps. The value of β can be seen in Figure 7, where spikes at 0.5 s and 1.5 s are the cause for the phenomenon mentioned above. The rest of the time, β takes relatively small values but shows a high variability. 14 of 19 Information 2019, 10, 182 Figure 7. Simulated values for the switching height β (left) and a detailed view (right). Figure 7. Simulated values for the switching height β (left) and a detailed view (right). Thanks to a smaller switching height β, a signiﬁcant reduction in control input chattering is achieved. This becomes visible in Figure 8 (bottom right vs. bottom left and top right). However, the major inﬂuence stems from the equivalent control part, Equation (10), and the disturbance compensation, Equation (12), which are shown at the top left of the ﬁgure. Thanks to a smaller switching height β, a signiﬁcant reduction in control input chattering is achieved. This becomes visible in Figure 8 (bottom right vs. bottom left and top right). However, the major inﬂuence stems from the equivalent control part, Equation (10), and the disturbance compensation, Equation (12), which are shown at the top left of the ﬁgure. Figure 8. Comparison of the control inputs u: contribution of the equivalent control law and the disturbance compensation (top, left); detailed plots for all the variants (top, right; bottom). The disturbance estimates used for compensation are provided by the KF, using only measurements of current and velocity. These estimates are shown in Figure 9. The currents resulting Figure 8. Comparison of the control inputs u: contribution of the equivalent control law and the disturbance compensation (top, left); detailed plots for all the variants (top, right; bottom). The disturbance estimates used for compensation are provided by the KF, using only measurements of current and velocity. These estimates are shown in Figure 9. The currents resulting The disturbance estimates used for compensation are provided by the KF, using only measurements of current and velocity. These estimates are shown in Figure 9. 4.2. Results The currents resulting 15 of 19 Information 2019, 10, 182 from those input voltages are displayed in Figure 10 and indicate only a negligible difference between the three variants. Finally, the sliding surface s(t) is depicted in Figure 11, where the inﬂuence of the external load torque becomes obvious. The SMC variants employing a boundary layer clearly achieve superior behaviour in comparison with the classical switching variant, which has a strong chattering impact in the sliding surface. Figure 9. d(t), ˙d(t) and their estimates provided by the KF. Figure 10. Simulated currents i for all the variants. Figure 9. d(t), ˙d(t) and their estimates provided by the KF. Figure 9. d(t), ˙d(t) and their estimates provided by the KF. Figure 10. Simulated currents i for all the variants. Figure 11. Simulated values for the sliding surface s(t) for all the variants. Figure 9. d(t), ˙d(t) and their estimates provided by the KF. Figure 10. Simulated currents i for all the variants. Figure 10. Simulated currents i for all the variants. Figure 11. Simulated values for the sliding surface s(t) for all the variants. Figure 11. Simulated values for the sliding surface s(t) for all the variants. 16 of 19 Information 2019, 10, 182 16 of 19 4.3. Discussion Note that in contrast to the authors’ previous work [20], upon which this contribution is based, the sliding function s(t) remains in the close vicinity of zero. This positive effect can be attributed to the employed disturbance compensation provided by the KF. Some slight but negligible inﬂuence of the load torque in s(t) becomes visible, which is the price of a signiﬁcantly reduced chattering. In [20], due to the violation of the SMC sliding condition mentioned there, s(t) was partially non-zero in the presence of large disturbances, with no tendency to converge to zero, —despite the fact that a nearly perfect velocity tracking was achieved. This phenomenon can be explained by a model mismatch concerning the dynamics of the sliding surface in closed loop, due to a missing disturbance compensation, which can yield a mismatched MPC. This problem has been solved in the given paper by means of the disturbance compensation. To further point out the beneﬁts of the proposed approach—the combination of a disturbance estimation by a KF, ﬁnite horizon MPC and SMC gain adaptation—it is benchmarked in closed-loop simulation studies against two other widespread methods: • A disturbance observer (DOB) according to [23] and • A time-delay estimation (TDE), see [5]. • A disturbance observer (DOB) according to [23] and • A time-delay estimation (TDE), see [5]. • A time-delay estimation (TDE), see [5]. Both are implemented to provide estimates for d(t) and ˙d(t), using the same (noisy) measurements that are available to the KF as well: current i(t) and velocity ω(t). The resulting estimates can be compared in Figures 12–14. Both of these alternatives lead to similar but slightly increased error energies (the unit is omitted here) in comparison with the KF approach: 0.009024 (KF), 0.009076 (TDE) and 0.009383 (DOB). Despite a negligible difference in tracking performance, the use of either DOB or TDE in closed-loop control shows signiﬁcantly larger chattering amplitudes as compared to the KF variant: usw is in the range of ±0.04 V for DOB and TDE, where a range of ±0.02 V holds for the KF (apart from spikes of about ±1 V at 0.5 s and 1.5 s, respectively). Since TDE is based on the feedback of the acceleration ˙ω, which needs to be determined via numerical differentiation of a noisy measurement signal, the performance degradation as compared to the KF becomes evident. Figure 12. Disturbance estimates for d and ˙d using disturbance observer (DOB). Figure 12. Disturbance estimates for d and ˙d using disturbance observer (DOB). Information 2019, 10, 182 17 of 19 17 of 19 Figure 13. Disturbance estimates for d and ˙d using time-delay estimation (TDE). Figure 13. Disturbance estimates for d and ˙d using time-delay estimation (TDE). Figure 14. Disturbance estimates for d and ˙d using KF, DOB and TDE in comparison. Figure 14. Disturbance estimates for d and ˙d using KF, DOB and TDE in comparison. In fact, to achieve reasonably useful results, the necessary numerical derivative of the velocity ω(t) had to be low-pass ﬁltered using a cut-off frequency of 5000 rad/s. Lower frequencies resulted in considerably worse tracking, while higher corner frequencies led to even more noise than depicted in Figure 14. The KF offers a superior behaviour in the presence of noise because it explicitly considers noise processes. Furthermore, since the KF constitutes a disturbance estimator with an integrated de-noising state observer, its state estimates for current and velocity are used to supply the SMC with state feedback, anyway. This renders the KF a perfect solution in cases like these, where state and disturbance estimates are needed. • A time-delay estimation (TDE), see [5]. A formal proof, however, regarding the compliance with the SMC sliding condition as stated in Equation (9) using suitable choices of the MPC weighting matrices Q and R, still remains an open problem. References 1. Kachroo, P.; Tomizuka, M. Chattering reduction and error convergence in the sliding-mode control of a class of nonlinear systems. IEEE Trans. Autom. Control 1996, 41, 1063–1068. [CrossRef] 1. Kachroo, P.; Tomizuka, M. Chattering reduction and error convergence in the sliding-mode control of a class of nonlinear systems. IEEE Trans. Autom. Control 1996, 41, 1063–1068. [CrossRef] 2. Bartolini, G.; Ferrara, A.; Usai, E. Chattering avoidance by second-order sliding mode control. IEE Autom. Control 1998, 43, 241–246. [CrossRef] 3. Aschemann, H.; Haus, B.; Mercorelli, P. Second-Order Sliding Mode Control with State and Disturbance Estimation for a Permanent Magnet Linear Motor. In Proceedings of the 23rd International Conference on Methods Models in Automation Robotics (MMAR), Miedzyzdroje, Poland, 27–30 August 2018; pp. 345–350. [CrossRef] 3. Aschemann, H.; Haus, B.; Mercorelli, P. Second-Order Sliding Mode Control with State and Disturbance Estimation for a Permanent Magnet Linear Motor. In Proceedings of the 23rd International Conference on Methods Models in Automation Robotics (MMAR), Miedzyzdroje, Poland, 27–30 August 2018; pp. 345–350. [CrossRef] 4. Xia, C.; Jiang, G.; Chen, W.; Shi, T. Switching-Gain Adaptation Current Control for Brushless DC Motors. IEEE Trans. Ind. Electron. 2016, 63, 2044–2052. [CrossRef] 4. Xia, C.; Jiang, G.; Chen, W.; Shi, T. Switching-Gain Adaptation Current Control for Brushless DC Motors. IEEE Trans. Ind. Electron. 2016, 63, 2044–2052. [CrossRef] 5. Roy, S.; Kar, I.N.; Lee, J.; Jin, M. Adaptive-Robust Time-Delay Control for a Class of Uncertain Euler-Lagrange Systems. IEEE Trans. Ind. Electron. 2017, 64, 7109–7119. [CrossRef] 5. Roy, S.; Kar, I.N.; Lee, J.; Jin, M. Adaptive-Robust Time-Delay Control for a Class of Uncertain Euler-Lagrange Systems. IEEE Trans. Ind. Electron. 2017, 64, 7109–7119. [CrossRef] 6. Plestan, F.; Shtessel, Y.; Brégeault, V.; Poznyak, A. New methodologies for adaptive sliding mode control. Int. J. Control 2010, 83, 1907–1919. [CrossRef] 6. Plestan, F.; Shtessel, Y.; Brégeault, V.; Poznyak, A. New methodologies for adaptive sliding mode control. Int. J. Control 2010, 83, 1907–1919. [CrossRef] 7. Liu, J.; Li, H.; Deng, Y. Torque Ripple Minimization of PMSM Based on Robust ILC Via Adaptive Sliding Mode Control. IEEE Trans. Power Electron. 2018, 33, 3655–3671. [CrossRef] 8. Moon, H.T.; Kim, H.S.; Youn, M.J. A discrete-time predictive current control for PMSM. IEEE Trans. Power Electron. 2003, 18, 464–472. [CrossRef] 9. Mercorelli, P. Linear Generalised Model Predictive Control to Avoid Input Saturation through Matrix Conditions. WSEAS Trans. Syst. 2017, 16, 313–322. 10. Mercorelli, P. 5. Conclusions This contribution presents an adaptive tuning of the switching gain of an SMC that is achieved by means of an MPC scheme designed on the basis of the sliding surface error dynamics. To properly address disturbances and the impact of external load torque in the control approach, they are compensated for using estimates provided by a KF. The overall design is benchmarked in simulations considering a frequently used application of high practical relevance—a DC drive. The simulations clearly show that the MPC-based adaptation of the switching height represents an effective means of counteracting a drawback of classical SMC—chattering caused by a conservative choice of the switching height, which is unnecessarily high most of the time and should be reduced when permissible. 18 of 19 18 of 19 Information 2019, 10, 182 Author Contributions: Supervision, P.M. and H.A.; conceptualization, B.H. and P.M.; methodology, B.H., H.A. and P.M.; formal analysis, H.A.; software, B.H.; visualization, B.H.; writing—original draft preparation, B.H., P.M. and H.A.; writing—review and editing, B.H. and H.A. Author Contributions: Supervision, P.M. and H.A.; conceptualization, B.H. and P.M.; methodology, B.H., H.A. and P.M.; formal analysis, H.A.; software, B.H.; visualization, B.H.; writing—original draft preparation, B.H., P.M. and H.A.; writing—review and editing, B.H. and H.A. Funding: This research received no external funding. Funding: This research received no external funding. Conﬂicts of Interest: The authors declare no conﬂict of interest. Conﬂicts of Interest: The authors declare no conﬂict of interest. c⃝2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). References A Sufﬁcient Asymptotic Stability Condition in Generalised Model Predictive Control to Avoid Input Saturation. In Applied Physics, System Science and Computers II; Ntalianis, K., Croitoru, A., Eds.; Springer: Berlin, Germany, 2018. 11. Mercorelli, P. A Motion-Sensorless Control for Intake Valves in Combustion Engines. IEEE Trans. Ind. Electron. 2017, 64, 3402–3412. [CrossRef] 12. Mercorelli, P. A Two-Stage Sliding-Mode High-Gain Observer to Reduce Uncertainties and Disturbances Effects for Sensorless Control in Automotive Applications. IEEE Trans. Ind. Electron. 2015, 62, 5929–5940. [CrossRef] 13. Hsieh, C.S. Robust two-stage Kalman ﬁlters for systems with unknown inputs. IEEE Trans. Autom. 2000, 45, 2374–2378. [CrossRef] 14. Hsieh, C.S.; Chen, F.C. Optimal solution of the two-stage Kalman estimator. IEEE Trans. Autom. Control 1999, 44, 194–199. [CrossRef] 15. Hsieh, C.S. General two-stage extended Kalman ﬁlters. IEEE Trans. Autom. Control 2003, 48, 289–293. [CrossRef] 16. Haessig, D.; Friedland, B. Separate-bias estimation with reduced-order Kalman ﬁlters. IEEE Trans. Autom. Control 1998, 43, 983–987. [CrossRef] 17. Chen, L.; Mercorelli, P.; Liu, S. A Kalman estimator for detecting repetitive disturbances. In Proceedings of the 2005 American Control Conference, Portland, OR, USA, 8–10 June 2005; Volume 3, pp. 1631–1636. [CrossRef] 8. Armstrong-Hélouvry, B.; Dupont, P.; Wit, C.C.D. A survey of models, analysis tools and compensa methods for the control of machines with friction. Automatica 1994, 30, 1083–1138. [CrossRef] 19. Aschemann, H.; Haus, B.; Mercorelli, P. Sliding Mode Control and Observer-Based Disturbance Compensation for a Permanent Magnet Linear Motor. In Proceedings of the 2018 Annual American Control Conference (ACC), Milwaukee, WI, USA, 27–29 June 2018; pp. 4141–4146. [CrossRef] 19 of 19 Information 2019, 10, 182 19 of 19 20. Haus, B.; Röhl, J.H.; Mercorelli, P.; Aschemann, H. Model Predictive Control for Switching Gain Adaptation in a Sliding Mode Controller of a DC Drive with Nonlinear Friction. In Proceedings of the 22nd International Conference on System Theory, Control and Computing (ICSTCC), Sinaia, Romania, 10–12 October 2018; pp. 765–770. [CrossRef] pp 21. Slotine, J.J.E.; Li, W. Applied Nonlinear Control; Pearson: Upper Saddle River, NJ, USA, 1991. 22. Welch, G.; Bishop, G. An Introduction to the Kalman Filter. Available online: https://www.cs.unc.edu/ ~welch/media/pdf/kalman_intro.pdf (accessed on 15 May 2019). 23. Chen, W.H. Disturbance observer based control for nonlinear systems. IEEE/ASME Trans. Mechatron. 2004, 9, 706–710. [CrossRef] c⃝2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

Subsets and Splits

No community queries yet

The top public SQL queries from the community will appear here once available.